What is a GI hospitalist?

A GI hospitalist is a gastroenterologist that primarily provides inpatient care. Their main professional focus is the acute management of gastrointestinal conditions occurring in the hospital setting.

How prevalent are subspecialty hospitalists?

The rise of hospitalists has changed the landscape of medicine. The hospitalist is now the central inpatient provider responsible for patient care and day-to-day housestaff education. From 1995 to 2016, the number of hospitalists increased from 500 to over 50,000.¹ While the majority of hospitalists are generalists from the fields of internal medicine, pediatrics, and obstetrics/gynecology, some come in the form of specialists. In a recent survey, up to 10% of internal medicine subspecialists already consider themselves “hospitalists.”² However, most of these self-described hospitalists only do so part of the time. For example, many group practices have one of their members manage all the hospitalized patients for the group for certain periods of time. It is rare to find full-time subspecialist hospitalists, but there has been an emergence in this new model of GI practice. Many people are unaware of this system of care nor understand how it may influence hospital-based care.

What is the role of a GI hospitalist?

While my primary responsibility is to care for inpatients whom require GI consults, I have outpatient and administrative responsibilities. Generally speaking, I am the de facto consult attending for the year.
 

How did you decide to become a GI hospitalist?

Upon graduation from my GI fellowship, I wanted an academic job where I could work closely with fellows and manage a wide breadth of complex, high-acuity patients. During fellowship, I enjoyed all areas of gastroenterology and hepatology and did not “sub-subspecialize.” As such, I wanted a job where I would see the full spectrum of GI and liver disease. Additionally, I enjoyed seeing the sickest patients, because I felt I could make the most dramatic differences with my care.

When I was searching for jobs, I spoke with the chief of GI at the hospital where I completed my residency about how I could fill a niche. We conceived of a model that would merge my personal interests and help the division provide consistent teaching for fellows and increase inpatient billing. Prior to my arrival, attendings that staffed the consult service were expected to continue their research and outpatient clinical workload while finding time to come to the hospital. Not surprisingly, attending rounds was erratic. The fellows were left to manage patients independently, scrambled to run cases by whomever happened to be around, or waited until they could reach the attending the next day. Unsurprisingly, billing by attendings was sparse.
 

What is a typical day like in your life as a GI hospitalist?

My day starts at 7:30 a.m. either with my outpatient office hours, endoscopy session, or GI Grand Rounds. Each week, I have two morning outpatient office sessions, one morning endoscopy session, and one morning session supervising fellows’ endoscopy.

At noon, I round with a team of GI fellows, medical students, and housestaff rotators for 2 hours. After we see the new consults, the remainder of my afternoon is spent seeing the follow-up patients. For two afternoons throughout the week, I have outpatient endoscopy sessions. I typically conclude my day at 5 p.m.

For night coverage, I take emergency calls for my own patients, and share general call duties with the other members of my division. On average, I take calls for one weekday a month and five weekends per year.

Typically, GI hospitalists only cover inpatients during the daytime. All nights and weekends are covered by partners and nonemergent overnight consults are saved until the next day. They have no office work.
 

What is the most challenging part of being a GI hospitalist?

As the perpetual “GI Consult Attending,” there is the threat of burnout when confronted with a high volume of sick, complex patients. Many of the patients have multiple comorbidities and require a multidisciplinary approach. On average, we have five new consults a day and the number of active follow-up patients is 10. Nonetheless, the nature of the inpatient service makes the volume of work unpredictable. When the service is busy and the census swells, the numbers of patients requiring staffing and notes can become overwhelming.

Mike Powell/Thinkstock

How are you paid?

My compensation is based on a base salary with an incentivized system based on my RVUs and collections. For the dedicated hospitalist for a group practice, there is typically a base salary and productivity-based income. Additionally, there should be a path to partnership. Lastly, in balancing the ledger, the diminished inpatient revenue stream is offset by the lack of overhead.

What are the benefits of a GI hospitalist system?

Our system benefits the workflow for the GI fellows. Since I have started, the GI consultation rounds start at a consistent time. During these rounds, we discuss relevant GI literature and make timely plans on all patients. Oftentimes, I am able to supervise the fellows so they can fit in a scope before the end of the workday. Ultimately, the fellows know they can find me and discuss patients throughout the day. The fellows consistently have told me that the since the implementation of the hospitalist system, there has been a dramatic difference. Collectively, they feel both their education and patient care have improved.

In terms of consult efficiency, one study demonstrated that the transition to a GI hospitalist system resulted in a mean decrease in consult to urgent esophagogastroduodenoscopy (EGD) time from approximately 24 to 14 hours.³ However, this occurred in the context of a lower inpatient consult volume and only covered 2 months. Furthermore, the time from admission to EGD did not change. Nonetheless, further studies are needed to examine the impact of this model shift.

In terms of a financial benefit, at our institution the total gross inpatient charges increased more than $850,000 for the year. This was largely attributable to the 79% increase in the gross charges from follow-up notes.

For group practices, the hospitalist system makes more efficient use of physician’s time. Physicians can either focus on outpatients or inpatients without worrying about going between the office, ambulatory surgical center, and the hospital. In general, inpatients require a disproportionate amount of time relative to the revenue collected. Furthermore, by eliminating the need for group physicians to go to the hospital, they can carve out 1-2 hours of office time to increase billing.

When there is one point-person whom handles all inpatient GI, communication is facilitated among primary teams and other services. The GI hospitalist develops working relationships with surgeons, radiologists, anesthesiologists, intensivists, etc. Teams can often just text or call me directly, instead of looking for the covering attending or going through the office phone service.
 

What are drawbacks to the GI hospitalist model?

Since there is only one gastroenterologist in the hospitalist model, if that person is not doing a good job, it affects the management of GI conditions for the entire hospital.

There is a loss of continuity-of-care. When GI patients get admitted, the gastroenterologists responsible for their care will not be the person with whom they have a long-term relationship. Furthermore, when the patient gets discharged, the primary gastroenterologists will not be fully aware of the inpatient course.

Also, when outpatient and inpatient gastroenterologists become segregated based on hospital setting, they each lose out of learning the intricacies of managing patients in a different context.
 

What do you like most about being a GI hospitalist?

The GI hospitalist position creates a great opportunity for gastroenterologists to make a remarkable, immediate impact on interesting, high acuity patients. The nature of the job also has the advantage of providing reasonable hours. This may be attractive to many whom want a better work-life balance.

Dr. Wan is assistant professor of medicine, associate program director, GI Fellowship Program, New York Presbyterian/Weill Cornell Medical Center, New York, N.Y.

References

1. Wachter R.M., Goldman L. Zero to 50,000 – The 20th Anniversary of the Hospitalist. N Engl J Med. 2016 Sep 15;375[11]:1009-11.

2. Estimating the Number and Characteristics of Hospitalist Physicians in the United States and Their Possible Workforce Implications. Analysis in Brief. Available at: https://www.aamc.org/download/300620/data/aibvol12_no3-hospitalist.pdf. Accessed May 1st, 2016.

3. Mahadev S., Lebwohl B., Ramirez I., Garcia-Carrasquillo R.J., Freedberg, D.E. Transition to a GI Hospitalist System is Associated with Expedited Upper Endoscopy. Gastroenterology. 2016;150[4]:S639-40.
 

What is a GI hospitalist?

A GI hospitalist is a gastroenterologist that primarily provides inpatient care. Their main professional focus is the acute management of gastrointestinal conditions occurring in the hospital setting.

How prevalent are subspecialty hospitalists?

The rise of hospitalists has changed the landscape of medicine. The hospitalist is now the central inpatient provider responsible for patient care and day-to-day housestaff education. From 1995 to 2016, the number of hospitalists increased from 500 to over 50,000.¹ While the majority of hospitalists are generalists from the fields of internal medicine, pediatrics, and obstetrics/gynecology, some come in the form of specialists. In a recent survey, up to 10% of internal medicine subspecialists already consider themselves “hospitalists.”² However, most of these self-described hospitalists only do so part of the time. For example, many group practices have one of their members manage all the hospitalized patients for the group for certain periods of time. It is rare to find full-time subspecialist hospitalists, but there has been an emergence in this new model of GI practice. Many people are unaware of this system of care nor understand how it may influence hospital-based care.

What is the role of a GI hospitalist?

While my primary responsibility is to care for inpatients whom require GI consults, I have outpatient and administrative responsibilities. Generally speaking, I am the de facto consult attending for the year.
 

How did you decide to become a GI hospitalist?

Upon graduation from my GI fellowship, I wanted an academic job where I could work closely with fellows and manage a wide breadth of complex, high-acuity patients. During fellowship, I enjoyed all areas of gastroenterology and hepatology and did not “sub-subspecialize.” As such, I wanted a job where I would see the full spectrum of GI and liver disease. Additionally, I enjoyed seeing the sickest patients, because I felt I could make the most dramatic differences with my care.

When I was searching for jobs, I spoke with the chief of GI at the hospital where I completed my residency about how I could fill a niche. We conceived of a model that would merge my personal interests and help the division provide consistent teaching for fellows and increase inpatient billing. Prior to my arrival, attendings that staffed the consult service were expected to continue their research and outpatient clinical workload while finding time to come to the hospital. Not surprisingly, attending rounds was erratic. The fellows were left to manage patients independently, scrambled to run cases by whomever happened to be around, or waited until they could reach the attending the next day. Unsurprisingly, billing by attendings was sparse.
 

What is a typical day like in your life as a GI hospitalist?

My day starts at 7:30 a.m. either with my outpatient office hours, endoscopy session, or GI Grand Rounds. Each week, I have two morning outpatient office sessions, one morning endoscopy session, and one morning session supervising fellows’ endoscopy.

At noon, I round with a team of GI fellows, medical students, and housestaff rotators for 2 hours. After we see the new consults, the remainder of my afternoon is spent seeing the follow-up patients. For two afternoons throughout the week, I have outpatient endoscopy sessions. I typically conclude my day at 5 p.m.

For night coverage, I take emergency calls for my own patients, and share general call duties with the other members of my division. On average, I take calls for one weekday a month and five weekends per year.

Typically, GI hospitalists only cover inpatients during the daytime. All nights and weekends are covered by partners and nonemergent overnight consults are saved until the next day. They have no office work.
 

What is the most challenging part of being a GI hospitalist?

As the perpetual “GI Consult Attending,” there is the threat of burnout when confronted with a high volume of sick, complex patients. Many of the patients have multiple comorbidities and require a multidisciplinary approach. On average, we have five new consults a day and the number of active follow-up patients is 10. Nonetheless, the nature of the inpatient service makes the volume of work unpredictable. When the service is busy and the census swells, the numbers of patients requiring staffing and notes can become overwhelming.

Mike Powell/Thinkstock

How are you paid?

My compensation is based on a base salary with an incentivized system based on my RVUs and collections. For the dedicated hospitalist for a group practice, there is typically a base salary and productivity-based income. Additionally, there should be a path to partnership. Lastly, in balancing the ledger, the diminished inpatient revenue stream is offset by the lack of overhead.

What are the benefits of a GI hospitalist system?

Our system benefits the workflow for the GI fellows. Since I have started, the GI consultation rounds start at a consistent time. During these rounds, we discuss relevant GI literature and make timely plans on all patients. Oftentimes, I am able to supervise the fellows so they can fit in a scope before the end of the workday. Ultimately, the fellows know they can find me and discuss patients throughout the day. The fellows consistently have told me that the since the implementation of the hospitalist system, there has been a dramatic difference. Collectively, they feel both their education and patient care have improved.

In terms of consult efficiency, one study demonstrated that the transition to a GI hospitalist system resulted in a mean decrease in consult to urgent esophagogastroduodenoscopy (EGD) time from approximately 24 to 14 hours.³ However, this occurred in the context of a lower inpatient consult volume and only covered 2 months. Furthermore, the time from admission to EGD did not change. Nonetheless, further studies are needed to examine the impact of this model shift.

In terms of a financial benefit, at our institution the total gross inpatient charges increased more than $850,000 for the year. This was largely attributable to the 79% increase in the gross charges from follow-up notes.

For group practices, the hospitalist system makes more efficient use of physician’s time. Physicians can either focus on outpatients or inpatients without worrying about going between the office, ambulatory surgical center, and the hospital. In general, inpatients require a disproportionate amount of time relative to the revenue collected. Furthermore, by eliminating the need for group physicians to go to the hospital, they can carve out 1-2 hours of office time to increase billing.

When there is one point-person whom handles all inpatient GI, communication is facilitated among primary teams and other services. The GI hospitalist develops working relationships with surgeons, radiologists, anesthesiologists, intensivists, etc. Teams can often just text or call me directly, instead of looking for the covering attending or going through the office phone service.
 

What are drawbacks to the GI hospitalist model?

Since there is only one gastroenterologist in the hospitalist model, if that person is not doing a good job, it affects the management of GI conditions for the entire hospital.

There is a loss of continuity-of-care. When GI patients get admitted, the gastroenterologists responsible for their care will not be the person with whom they have a long-term relationship. Furthermore, when the patient gets discharged, the primary gastroenterologists will not be fully aware of the inpatient course.

Also, when outpatient and inpatient gastroenterologists become segregated based on hospital setting, they each lose out of learning the intricacies of managing patients in a different context.
 

What do you like most about being a GI hospitalist?

The GI hospitalist position creates a great opportunity for gastroenterologists to make a remarkable, immediate impact on interesting, high acuity patients. The nature of the job also has the advantage of providing reasonable hours. This may be attractive to many whom want a better work-life balance.

Dr. Wan is assistant professor of medicine, associate program director, GI Fellowship Program, New York Presbyterian/Weill Cornell Medical Center, New York, N.Y.

References

1. Wachter R.M., Goldman L. Zero to 50,000 – The 20th Anniversary of the Hospitalist. N Engl J Med. 2016 Sep 15;375[11]:1009-11.

2. Estimating the Number and Characteristics of Hospitalist Physicians in the United States and Their Possible Workforce Implications. Analysis in Brief. Available at: https://www.aamc.org/download/300620/data/aibvol12_no3-hospitalist.pdf. Accessed May 1st, 2016.

3. Mahadev S., Lebwohl B., Ramirez I., Garcia-Carrasquillo R.J., Freedberg, D.E. Transition to a GI Hospitalist System is Associated with Expedited Upper Endoscopy. Gastroenterology. 2016;150[4]:S639-40.
 

What is a GI hospitalist?

A GI hospitalist is a gastroenterologist that primarily provides inpatient care. Their main professional focus is the acute management of gastrointestinal conditions occurring in the hospital setting.

How prevalent are subspecialty hospitalists?

The rise of hospitalists has changed the landscape of medicine. The hospitalist is now the central inpatient provider responsible for patient care and day-to-day housestaff education. From 1995 to 2016, the number of hospitalists increased from 500 to over 50,000.¹ While the majority of hospitalists are generalists from the fields of internal medicine, pediatrics, and obstetrics/gynecology, some come in the form of specialists. In a recent survey, up to 10% of internal medicine subspecialists already consider themselves “hospitalists.”² However, most of these self-described hospitalists only do so part of the time. For example, many group practices have one of their members manage all the hospitalized patients for the group for certain periods of time. It is rare to find full-time subspecialist hospitalists, but there has been an emergence in this new model of GI practice. Many people are unaware of this system of care nor understand how it may influence hospital-based care.

What is the role of a GI hospitalist?

While my primary responsibility is to care for inpatients whom require GI consults, I have outpatient and administrative responsibilities. Generally speaking, I am the de facto consult attending for the year.
 

How did you decide to become a GI hospitalist?

Upon graduation from my GI fellowship, I wanted an academic job where I could work closely with fellows and manage a wide breadth of complex, high-acuity patients. During fellowship, I enjoyed all areas of gastroenterology and hepatology and did not “sub-subspecialize.” As such, I wanted a job where I would see the full spectrum of GI and liver disease. Additionally, I enjoyed seeing the sickest patients, because I felt I could make the most dramatic differences with my care.

When I was searching for jobs, I spoke with the chief of GI at the hospital where I completed my residency about how I could fill a niche. We conceived of a model that would merge my personal interests and help the division provide consistent teaching for fellows and increase inpatient billing. Prior to my arrival, attendings that staffed the consult service were expected to continue their research and outpatient clinical workload while finding time to come to the hospital. Not surprisingly, attending rounds was erratic. The fellows were left to manage patients independently, scrambled to run cases by whomever happened to be around, or waited until they could reach the attending the next day. Unsurprisingly, billing by attendings was sparse.
 

What is a typical day like in your life as a GI hospitalist?

My day starts at 7:30 a.m. either with my outpatient office hours, endoscopy session, or GI Grand Rounds. Each week, I have two morning outpatient office sessions, one morning endoscopy session, and one morning session supervising fellows’ endoscopy.

At noon, I round with a team of GI fellows, medical students, and housestaff rotators for 2 hours. After we see the new consults, the remainder of my afternoon is spent seeing the follow-up patients. For two afternoons throughout the week, I have outpatient endoscopy sessions. I typically conclude my day at 5 p.m.

For night coverage, I take emergency calls for my own patients, and share general call duties with the other members of my division. On average, I take calls for one weekday a month and five weekends per year.

Typically, GI hospitalists only cover inpatients during the daytime. All nights and weekends are covered by partners and nonemergent overnight consults are saved until the next day. They have no office work.
 

What is the most challenging part of being a GI hospitalist?

As the perpetual “GI Consult Attending,” there is the threat of burnout when confronted with a high volume of sick, complex patients. Many of the patients have multiple comorbidities and require a multidisciplinary approach. On average, we have five new consults a day and the number of active follow-up patients is 10. Nonetheless, the nature of the inpatient service makes the volume of work unpredictable. When the service is busy and the census swells, the numbers of patients requiring staffing and notes can become overwhelming.

Mike Powell/Thinkstock

How are you paid?

My compensation is based on a base salary with an incentivized system based on my RVUs and collections. For the dedicated hospitalist for a group practice, there is typically a base salary and productivity-based income. Additionally, there should be a path to partnership. Lastly, in balancing the ledger, the diminished inpatient revenue stream is offset by the lack of overhead.

What are the benefits of a GI hospitalist system?

Our system benefits the workflow for the GI fellows. Since I have started, the GI consultation rounds start at a consistent time. During these rounds, we discuss relevant GI literature and make timely plans on all patients. Oftentimes, I am able to supervise the fellows so they can fit in a scope before the end of the workday. Ultimately, the fellows know they can find me and discuss patients throughout the day. The fellows consistently have told me that the since the implementation of the hospitalist system, there has been a dramatic difference. Collectively, they feel both their education and patient care have improved.

In terms of consult efficiency, one study demonstrated that the transition to a GI hospitalist system resulted in a mean decrease in consult to urgent esophagogastroduodenoscopy (EGD) time from approximately 24 to 14 hours.³ However, this occurred in the context of a lower inpatient consult volume and only covered 2 months. Furthermore, the time from admission to EGD did not change. Nonetheless, further studies are needed to examine the impact of this model shift.

In terms of a financial benefit, at our institution the total gross inpatient charges increased more than $850,000 for the year. This was largely attributable to the 79% increase in the gross charges from follow-up notes.

For group practices, the hospitalist system makes more efficient use of physician’s time. Physicians can either focus on outpatients or inpatients without worrying about going between the office, ambulatory surgical center, and the hospital. In general, inpatients require a disproportionate amount of time relative to the revenue collected. Furthermore, by eliminating the need for group physicians to go to the hospital, they can carve out 1-2 hours of office time to increase billing.

When there is one point-person whom handles all inpatient GI, communication is facilitated among primary teams and other services. The GI hospitalist develops working relationships with surgeons, radiologists, anesthesiologists, intensivists, etc. Teams can often just text or call me directly, instead of looking for the covering attending or going through the office phone service.
 

What are drawbacks to the GI hospitalist model?

Since there is only one gastroenterologist in the hospitalist model, if that person is not doing a good job, it affects the management of GI conditions for the entire hospital.

There is a loss of continuity-of-care. When GI patients get admitted, the gastroenterologists responsible for their care will not be the person with whom they have a long-term relationship. Furthermore, when the patient gets discharged, the primary gastroenterologists will not be fully aware of the inpatient course.

Also, when outpatient and inpatient gastroenterologists become segregated based on hospital setting, they each lose out of learning the intricacies of managing patients in a different context.
 

What do you like most about being a GI hospitalist?

The GI hospitalist position creates a great opportunity for gastroenterologists to make a remarkable, immediate impact on interesting, high acuity patients. The nature of the job also has the advantage of providing reasonable hours. This may be attractive to many whom want a better work-life balance.

Dr. Wan is assistant professor of medicine, associate program director, GI Fellowship Program, New York Presbyterian/Weill Cornell Medical Center, New York, N.Y.

References

1. Wachter R.M., Goldman L. Zero to 50,000 – The 20th Anniversary of the Hospitalist. N Engl J Med. 2016 Sep 15;375[11]:1009-11.

2. Estimating the Number and Characteristics of Hospitalist Physicians in the United States and Their Possible Workforce Implications. Analysis in Brief. Available at: https://www.aamc.org/download/300620/data/aibvol12_no3-hospitalist.pdf. Accessed May 1st, 2016.

3. Mahadev S., Lebwohl B., Ramirez I., Garcia-Carrasquillo R.J., Freedberg, D.E. Transition to a GI Hospitalist System is Associated with Expedited Upper Endoscopy. Gastroenterology. 2016;150[4]:S639-40.
 

Historical perspective

The term “pancreas” derives its name from the Greek words pan (all) and kreas (flesh). Understanding pancreas physiology was first attempted in the 17th century by Regnier de Graaf^1. Giovanni Morgagni is credited with the first description of the syndrome of acute pancreatitis (AP) in 1761². Reginald Huber Fitz proposed the first classification of AP into hemorrhagic, gangrenous, and suppurative types in 1889³. The distinction of acute from chronic pancreatitis was not well described until the middle of the 20th century when Mandred W. Comfort gave a detailed account of chronic relapsing pancreatitis in 1946⁴.

Diagnosis and classification of severity

The diagnosis of AP is based on the presence of two of the three following criteria: typical abdominal pain (severe, upper abdominal pain frequently radiating to the back), serum amylase and/or lipase levels greater than 3 times the upper limit of normal, and/or characteristic imaging findings.

The original 1992 Atlanta classification provided the first blueprint to standardize how severity of AP was defined⁵. Over the years, better understanding of AP pathophysiology and its complications led to a greater focus on local and systemic determinants of severity⁶ and eventually the Revised Atlanta Classification (RAC) in 2013 (Table 1).
 

Management of acute pancreatitis

Prevention

Determination of etiology

The most common causes of AP are gallstones and alcohol, accounting for more than two-thirds of all cases¹³. Other etiologies include hypertriglyceridemia, ERCP, drugs induced, familial/hereditary, and post-traumatic. Initial work up includes a thorough history to quantify alcohol consumption and assess for recently started medications, measurement of liver injury tests¹⁴ and triglyceride levels, and performance of a transabdominal ultrasound to evaluate for biliary dilation, chole- and choledocholithiasis¹⁵.

Assessment of disease severity

Fluid resuscitation

Despite extensive research and trials using medications such as ulinastatin, octreotide, pentoxifylline, gabexate, N-acetyl cysteine, steroids, IL-10, and antibiotics²⁰, no pharmacologic agent has been shown to significantly alter the clinical course/outcomes of AP.

Adequate intravenous hydration remains the cornerstone of early management in AP²¹. Studies have demonstrated that increased intestinal permeability, secondary to reduced intestinal capillary microcirculation, leads to bacterial translocation and development of SIRS²². Intestinal microcirculation does not become as readily impaired, and there is a certain “latency” to its onset, from the insult that triggers pancreatitis. This gives rise to the concept of a “golden window” of 12-24 hours from the insult to potentially reverse such changes and prevent organ dysfunction. It has been shown that patients who are adequately resuscitated with intravenous fluids have lower risk for local and systemic complications²³.

Patients with predicted severe AP or those with persistent SIRS despite initial fluid resuscitation should be managed in a closely monitored unit, ideally an ICU. Patients with impending respiratory failure require mechanical ventilation, renal failure complicated by metabolic acidosis and/or hyperkalemia requires hemodialysis, and cardiovascular shock requires the initiation of vasopressors and continuous monitoring of blood pressure via an arterial line. A special entity that requires ICU level care is hypertriglyceridemia (HTG)-induced severe AP. HTG should be considered as the etiology of AP in certain clinical scenarios²⁸: previous history of HTG, poorly controlled diabetes mellitus, history of significant alcohol use, third trimester of pregnancy, and use of certain medications associated with HTG such as oral estrogens, tamoxifen, and propofol. Levels of triglyceride greater than 1000 mg/dL strongly point toward HTG being the etiology.

Plasmapheresis, which filters and removes triglycerides from plasma, has been reported as an efficient treatment in such patients based on case series^29,30. At this time its use may only be justified in patients with predicted severe AP from HTG, preferably within the first 24 hours of presentation.


Urgent ERCP

Nutrition

Recovery of the gut function is often delayed for several days or weeks in patients with severe AP. Studies have shown that prolonged fasting in such circumstances leads to malnutrition and worse prognosis^33,34. Enteral nutrition via a nasogastric (NG) or nasojejunal (NJ) tube is the preferred route of nutritional support, as it is associated with lower risk of infection, multi-organ failure, and mortality when compared to total parenteral nutrition³³.

The question of whether NJ feeding offers any additional advantages over NG feeding has not been clearly answered with a recent randomized trial showing NG feeds not to be inferior to NJ feeds³⁵. In regards to the timing of initiation of enteral nutrition, early nasoenteric feeding within 24 hours from presentation was found not to be superior compared to on-demand feeding in patients with predicted severe AP³⁶.


Strategies to decrease risk of recurrent attacks

Management of peripancreatic fluid collections

Patients with AP frequently develop peripancreatic fluid collections (PFCs). Based on the revised Atlanta classification, those are categorized into four types (Table 2, Figures 1-4).

The majority of acute PFCs in patients without evidence of pancreatic necrosis regress within a few weeks and thus intervention is not indicated early in the disease course. Current literature supports delaying the drainage/debridement of such collections for several weeks. The mortality from interventions decreases as the time to intervention from onset of symptoms increases⁴¹. Delaying intervention gives more time for recovery from systemic complications and allows the encapsulating wall and contents to organize further.

While surgery is still an option for patients with symptomatic mature PFCs, endoscopic ultrasound-guided drainage in expert hands has been shown to be cost effective, with shorter hospital stay and even decreased risk of cyst recurrence compared with surgical cyst-gastrostomy creation⁴⁴. Ultrasound or computed tomography-guided drainage of such collections with a percutaneous catheter is an equally efficacious option when compared to the endoscopic approach. However, patients undergoing endotherapy require fewer procedures and imaging studies and shorter length of stay⁴⁵ when compared with radiological interventions.

Although this topic has generated much debate, the majority of available evidence shows no clinical benefit from using prophylactic antibiotics to prevent infection in pancreatic necrosis⁴⁶.

Vascular complications

Vascular complications such as splanchnic vein thrombosis can occur in up to a quarter of AP patients⁴⁹. Anticoagulation is not usually indicated unless thrombosis is extensive and causes bowel ischemia. Arterial pseudoaneurysms are rare but life threatening complications of AP. They typically require interventional radiology guided coil embolization to prevent massive bleeding⁵⁰.

Abdominal compartment syndrome

Abdominal compartment syndrome is an end result of third spacing of fluid into the abdominal cavity secondary to inflammation and fluid resuscitation in severe pancreatitis. Abdominal pressure in patients can be monitored by measuring bladder pressures. Intra-abdominal hypertension is defined as a sustained pressure greater than 12 mm Hg, while abdominal compartment syndrome is defined as sustained intra-abdominal pressure greater than 20 mm Hg with new organ failure⁵¹. Intra-abdominal hypertension (IAH) is present in up to 75% of patients with severe AP. While all conservative measures to prevent development or worsening of IAH should be implemented (adequate sedation, decompression of bowel in patients with ileus, etc.), current guidelines do not recommend aggressive interventions to treat it. On the other hand, abdominal compartment syndrome is a life-threatening complication that requires urgent intervention to decrease intra-abdominal pressure, such as percutaneous drain placement or surgical fasciotomy^52,53.

Conclusion

The key principles in the management of acute pancreatitis are aggressive hydration and preventing development of end organ failure. In the last two decades there has been a paradigm shift in the guidelines for management of peripancreatic fluid collections and pancreatic necrosis. When feasible, drainage of these collections should be delayed and be performed using minimally invasive interventions. There is still an urgent need for developing and testing disease-specific treatments targeting control of the inflammatory response in the early phase of acute pancreatitis and prevention of development of severe disease with end-organ dysfunction.

Dr. Gulati is a gastroenterology and hepatology fellow at Allegheny Health Network, Pittsburgh, and Dr. Papachristou is professor of medicine, University of Pittsburgh School of Medicine, Pittsburgh.

References

1. Sleisenger and Fordtran’s Gastrointestinal and Liver Disease, Chapter 55, 923-33.

2. Morgagni G.B. [Fie Books on the Seats and Causes of Diseases as Discovered by the Anatomist]. Venice, Italy: Typographia Remondiniana;1761.

3. Fitz R.H. Boston Med Surg J. 1889;120:181-8.

4. Comfort M., Gambill E., Baggesnstoss A. Gastroenterology. 1946;6:238-76.

5. Bollen T.L., van Santvoort H.C., Besselink M.G., et al. Br J Surg. 2008;95:6–21.

6. Dellinger E.P., Forsmark C.E., Layer P., et al. Ann Surg. 2012 Dec;256[6]:875-80.

7. Kochar B., Akshintala V.S., Afghani E., et al. Gastrointest Endosc. 2015 Jan;81[1]:143-9.

8. Choudhary A., Bechtold M.L., Arif M., et al. Gastrointest Endosc. 2011 Feb;73[2]:275-82.

9. Shi Q.Q., Ning X.Y., Zhan L.L., Tang G.D., Lv X.P. World J Gastroenterol. 2014 Jun 14;20[22]:7040-8.

10. Elmunzer B.J., Waljee A.K., Elta G.H., Taylor J.R., Fehmi S.M., Higgins P.D. Gut. 2008 Sep;57[9]:1262-7.

11. Sethi S., Sethi N., Wadhwa V., Garud S., Brown A. Pancreas. 2014 Mar;43[2]:190-7. 
12. Elmunzer B.J., Serrano J., Chak A., et al. Trials. 2016 Mar 3;17[1]:120.

13. Lowenfels A.B., Maisonneuve P., Sullivan T. Curr Gastroenterol Rep. 2009;11:97-103.

14. Agarwal N., Pitchumoni C.S., Sivaprasad A.V. Am J Gastroenterol. 1990;85:356-66.

15. Tenner S., Baillie J., DeWitt J. Vege S.S. Am J Gastroenterol. 2013;108:1400-15.

16. Papachristou G.I., Muddana V., Yadav D., et al. Am J Gastroenterol. 2010;105:435-41.

17. Mounzer R., et al. Gastroenterology 2012;142:1476-82.

18. Working Group IAP/APA Acute Pancreatitis Guidelines. Pancreatology. 2013 Jul-Aug;13(4 Suppl 2):e1-15.

19. Koutroumpakis E., Wu B.U., Bakker O.J., et al. Am J Gastroenterol. 2015 Dec;110[12]:1707-16.

20. Bang U.C., Semb S., Nojgaard C., Bendtsen F. World J Gastroenterol. 2008 May 21;14[19]:2968-76.

21. Warndorf M.G., Kurtzman J.T., Bartel M.J., et al. Clin Gastroenterol Hepatol. 2011 Aug;9[8]:705-9.

22. Hotz H.G., Foitzik T., Rohweder J., et al. J Gastrointest Surg. 1998 Nov-Dec;2[6]:518-25.

23. Brown A., Baillargeon J.D., Hughes M.D., et al. Pancreatology 2002;2:104-7.

24. Wu B.U., Hwang J.Q., Gardner T.H., et al. Clin Gastroenterol Hepatol. 2011 Aug;9[8]:710-7.

25. Forsmark C.E., Baillie J., AGA Institute Clinical Practice and Economics Committee, AGA Institute Governing Board. Gastroenterology. 2007 May;132[5]:2022-44.

26. Lankisch P.G., Mahlke R., Blum T., et al. Am J Gastroenterol. 2001;96:2081-5.

27. Wu B.U., Johannes R.S., Sun X., et al. Gastroenterology 2009;137:129-35.

28. Scherer J., Singh V.P., Pitchumoni C.S., Yadav D. J Clin Gastroenterol. 2014 Mar;48[3]:195-203.

29. Gubensek J., Buturovic-Ponikvar J., Romozi K., Ponikvar R. PLoS One. 2014 Jul 21;9[7]:e102748.

30. Chen J.H., Yeh J.H., Lai H.W., Liao C.S. World J Gastroenterol. 2004 Aug 1;10[15]:2272-4.

31. Tse F., Yuan Y. Cochrane Database Syst Rev. 2012 May 16;[5]:CD009779.

32. Folsch U.R., Nitsche R., Ludtke R., et al. N Engl J Med. 1997;336:237-42.

33. Al-Omran M., Albalawi Z.H., Tashkandi M.F., Al-Ansary L.A. Cochrane Database Syst Rev. 2010 Jan 20;[1]:CD002837.

34. Li J.Y., Yu T., Chen G.C., et al. PLoS One. 2013;8[6]:e64926.

35. Singh N., Sharma B., Sharma M., et al. Pancreas. 2012 Jan;41[1]:153-9.

36. Bakker O.J., van Brunschot S., van Santvoort H.C., et al. N Engl J Med. 2014 Nov 20;371[21]:1983-93.

37. Van Baal M.C., Besselink M.G., Bakker O.J., et al. Ann Surg. 2012;255:860–6.

38. Nealon W.H., Bawduniak J., Walser E.M. Ann Surg. 2004 Jun;239[6]:741-9.

39. Sanjay P., Yeeting S., Whigham C., Judson H., Polignano F.M., Tait I.S. Surg Endosc. 2008 Aug;22[8]:1832-7.

40. Nordback I., Pelli H., Lappalainen-Lehto R., Järvinen S., Räty S., Sand J. Gastroenterology. 2009 Mar;136[3]:848-55.

41. Besselink M.G., Verwer T.J., Schoenmaeckers E.J., et al. Arch Surg. 2007;142:1194-201.

42. Besselink M., van Santvoort H., Freeman M. et al. Pancreatology. 2013 Jul-Aug;13(4 Suppl 2):e1-15.

43. Hjalmar C., van Santvoort, H., Besselink M.G., et al. N Engl J Med. 2010;362:1491-502.

44. Varadarajulu S., Bang J.Y., Sutton B.S., et al. Gastroenterology. 2013;145:583-90.e1.

45. Akshintala V.S., Saxena P., Zaheer A., et al. Gastrointest Endosc. 2014 Jun;79[6]:921-8.

46. Jiang K, Huang W, Yang XN., et al. World J Gastroenterol. 2012;18:279–84.

47. Dervenis C., Smailis D., Hatzitheoklitos E. J Hepatobiliary Pancreat Surg. 2003;10[6]:415Y418.

48. Gloor B., Muller C.A., Worni M., et al. Arch Surg. 2001;136[5]:592Y596.

49. Nadkarni N.A., Khanna S., Vege S.S. Pancreas. 2013 Aug;42[6]:924-31.

50. Marshall G.T., Howell D.A., Hansen B.L., Amberson S.M., Abourjaily G.S., Bredenberg C.E. Arch Surg. 1996 Mar;131[3]:278-83.

51. Malbrain M.L., Cheatham M.L., Kirkpatrick A., et al. Intensive Care Med. 2006 Nov;32[11]:1722-32.

52. De Waele J.J. Leppaniemi A.K. World J Surg. 2009;33:1128-33.

53. Kirkpatrick A.W., Roberts D.J., De W.J., et al. Intensive Care Med. 2013 Jul;39[7]1190-206.
 

Historical perspective

The term “pancreas” derives its name from the Greek words pan (all) and kreas (flesh). Understanding pancreas physiology was first attempted in the 17th century by Regnier de Graaf^1. Giovanni Morgagni is credited with the first description of the syndrome of acute pancreatitis (AP) in 1761². Reginald Huber Fitz proposed the first classification of AP into hemorrhagic, gangrenous, and suppurative types in 1889³. The distinction of acute from chronic pancreatitis was not well described until the middle of the 20th century when Mandred W. Comfort gave a detailed account of chronic relapsing pancreatitis in 1946⁴.

Diagnosis and classification of severity

The diagnosis of AP is based on the presence of two of the three following criteria: typical abdominal pain (severe, upper abdominal pain frequently radiating to the back), serum amylase and/or lipase levels greater than 3 times the upper limit of normal, and/or characteristic imaging findings.

The original 1992 Atlanta classification provided the first blueprint to standardize how severity of AP was defined⁵. Over the years, better understanding of AP pathophysiology and its complications led to a greater focus on local and systemic determinants of severity⁶ and eventually the Revised Atlanta Classification (RAC) in 2013 (Table 1).
 

Management of acute pancreatitis

Prevention

Determination of etiology

The most common causes of AP are gallstones and alcohol, accounting for more than two-thirds of all cases¹³. Other etiologies include hypertriglyceridemia, ERCP, drugs induced, familial/hereditary, and post-traumatic. Initial work up includes a thorough history to quantify alcohol consumption and assess for recently started medications, measurement of liver injury tests¹⁴ and triglyceride levels, and performance of a transabdominal ultrasound to evaluate for biliary dilation, chole- and choledocholithiasis¹⁵.

Assessment of disease severity

Fluid resuscitation

Despite extensive research and trials using medications such as ulinastatin, octreotide, pentoxifylline, gabexate, N-acetyl cysteine, steroids, IL-10, and antibiotics²⁰, no pharmacologic agent has been shown to significantly alter the clinical course/outcomes of AP.

Adequate intravenous hydration remains the cornerstone of early management in AP²¹. Studies have demonstrated that increased intestinal permeability, secondary to reduced intestinal capillary microcirculation, leads to bacterial translocation and development of SIRS²². Intestinal microcirculation does not become as readily impaired, and there is a certain “latency” to its onset, from the insult that triggers pancreatitis. This gives rise to the concept of a “golden window” of 12-24 hours from the insult to potentially reverse such changes and prevent organ dysfunction. It has been shown that patients who are adequately resuscitated with intravenous fluids have lower risk for local and systemic complications²³.

Patients with predicted severe AP or those with persistent SIRS despite initial fluid resuscitation should be managed in a closely monitored unit, ideally an ICU. Patients with impending respiratory failure require mechanical ventilation, renal failure complicated by metabolic acidosis and/or hyperkalemia requires hemodialysis, and cardiovascular shock requires the initiation of vasopressors and continuous monitoring of blood pressure via an arterial line. A special entity that requires ICU level care is hypertriglyceridemia (HTG)-induced severe AP. HTG should be considered as the etiology of AP in certain clinical scenarios²⁸: previous history of HTG, poorly controlled diabetes mellitus, history of significant alcohol use, third trimester of pregnancy, and use of certain medications associated with HTG such as oral estrogens, tamoxifen, and propofol. Levels of triglyceride greater than 1000 mg/dL strongly point toward HTG being the etiology.

Plasmapheresis, which filters and removes triglycerides from plasma, has been reported as an efficient treatment in such patients based on case series^29,30. At this time its use may only be justified in patients with predicted severe AP from HTG, preferably within the first 24 hours of presentation.


Urgent ERCP

Nutrition

Recovery of the gut function is often delayed for several days or weeks in patients with severe AP. Studies have shown that prolonged fasting in such circumstances leads to malnutrition and worse prognosis^33,34. Enteral nutrition via a nasogastric (NG) or nasojejunal (NJ) tube is the preferred route of nutritional support, as it is associated with lower risk of infection, multi-organ failure, and mortality when compared to total parenteral nutrition³³.

The question of whether NJ feeding offers any additional advantages over NG feeding has not been clearly answered with a recent randomized trial showing NG feeds not to be inferior to NJ feeds³⁵. In regards to the timing of initiation of enteral nutrition, early nasoenteric feeding within 24 hours from presentation was found not to be superior compared to on-demand feeding in patients with predicted severe AP³⁶.


Strategies to decrease risk of recurrent attacks

Management of peripancreatic fluid collections

Patients with AP frequently develop peripancreatic fluid collections (PFCs). Based on the revised Atlanta classification, those are categorized into four types (Table 2, Figures 1-4).

The majority of acute PFCs in patients without evidence of pancreatic necrosis regress within a few weeks and thus intervention is not indicated early in the disease course. Current literature supports delaying the drainage/debridement of such collections for several weeks. The mortality from interventions decreases as the time to intervention from onset of symptoms increases⁴¹. Delaying intervention gives more time for recovery from systemic complications and allows the encapsulating wall and contents to organize further.

While surgery is still an option for patients with symptomatic mature PFCs, endoscopic ultrasound-guided drainage in expert hands has been shown to be cost effective, with shorter hospital stay and even decreased risk of cyst recurrence compared with surgical cyst-gastrostomy creation⁴⁴. Ultrasound or computed tomography-guided drainage of such collections with a percutaneous catheter is an equally efficacious option when compared to the endoscopic approach. However, patients undergoing endotherapy require fewer procedures and imaging studies and shorter length of stay⁴⁵ when compared with radiological interventions.

Although this topic has generated much debate, the majority of available evidence shows no clinical benefit from using prophylactic antibiotics to prevent infection in pancreatic necrosis⁴⁶.

Vascular complications

Vascular complications such as splanchnic vein thrombosis can occur in up to a quarter of AP patients⁴⁹. Anticoagulation is not usually indicated unless thrombosis is extensive and causes bowel ischemia. Arterial pseudoaneurysms are rare but life threatening complications of AP. They typically require interventional radiology guided coil embolization to prevent massive bleeding⁵⁰.

Abdominal compartment syndrome

Abdominal compartment syndrome is an end result of third spacing of fluid into the abdominal cavity secondary to inflammation and fluid resuscitation in severe pancreatitis. Abdominal pressure in patients can be monitored by measuring bladder pressures. Intra-abdominal hypertension is defined as a sustained pressure greater than 12 mm Hg, while abdominal compartment syndrome is defined as sustained intra-abdominal pressure greater than 20 mm Hg with new organ failure⁵¹. Intra-abdominal hypertension (IAH) is present in up to 75% of patients with severe AP. While all conservative measures to prevent development or worsening of IAH should be implemented (adequate sedation, decompression of bowel in patients with ileus, etc.), current guidelines do not recommend aggressive interventions to treat it. On the other hand, abdominal compartment syndrome is a life-threatening complication that requires urgent intervention to decrease intra-abdominal pressure, such as percutaneous drain placement or surgical fasciotomy^52,53.

Conclusion

The key principles in the management of acute pancreatitis are aggressive hydration and preventing development of end organ failure. In the last two decades there has been a paradigm shift in the guidelines for management of peripancreatic fluid collections and pancreatic necrosis. When feasible, drainage of these collections should be delayed and be performed using minimally invasive interventions. There is still an urgent need for developing and testing disease-specific treatments targeting control of the inflammatory response in the early phase of acute pancreatitis and prevention of development of severe disease with end-organ dysfunction.

Dr. Gulati is a gastroenterology and hepatology fellow at Allegheny Health Network, Pittsburgh, and Dr. Papachristou is professor of medicine, University of Pittsburgh School of Medicine, Pittsburgh.

References

1. Sleisenger and Fordtran’s Gastrointestinal and Liver Disease, Chapter 55, 923-33.

2. Morgagni G.B. [Fie Books on the Seats and Causes of Diseases as Discovered by the Anatomist]. Venice, Italy: Typographia Remondiniana;1761.

3. Fitz R.H. Boston Med Surg J. 1889;120:181-8.

4. Comfort M., Gambill E., Baggesnstoss A. Gastroenterology. 1946;6:238-76.

5. Bollen T.L., van Santvoort H.C., Besselink M.G., et al. Br J Surg. 2008;95:6–21.

6. Dellinger E.P., Forsmark C.E., Layer P., et al. Ann Surg. 2012 Dec;256[6]:875-80.

7. Kochar B., Akshintala V.S., Afghani E., et al. Gastrointest Endosc. 2015 Jan;81[1]:143-9.

8. Choudhary A., Bechtold M.L., Arif M., et al. Gastrointest Endosc. 2011 Feb;73[2]:275-82.

9. Shi Q.Q., Ning X.Y., Zhan L.L., Tang G.D., Lv X.P. World J Gastroenterol. 2014 Jun 14;20[22]:7040-8.

10. Elmunzer B.J., Waljee A.K., Elta G.H., Taylor J.R., Fehmi S.M., Higgins P.D. Gut. 2008 Sep;57[9]:1262-7.

11. Sethi S., Sethi N., Wadhwa V., Garud S., Brown A. Pancreas. 2014 Mar;43[2]:190-7. 
12. Elmunzer B.J., Serrano J., Chak A., et al. Trials. 2016 Mar 3;17[1]:120.

13. Lowenfels A.B., Maisonneuve P., Sullivan T. Curr Gastroenterol Rep. 2009;11:97-103.

14. Agarwal N., Pitchumoni C.S., Sivaprasad A.V. Am J Gastroenterol. 1990;85:356-66.

15. Tenner S., Baillie J., DeWitt J. Vege S.S. Am J Gastroenterol. 2013;108:1400-15.

16. Papachristou G.I., Muddana V., Yadav D., et al. Am J Gastroenterol. 2010;105:435-41.

17. Mounzer R., et al. Gastroenterology 2012;142:1476-82.

18. Working Group IAP/APA Acute Pancreatitis Guidelines. Pancreatology. 2013 Jul-Aug;13(4 Suppl 2):e1-15.

19. Koutroumpakis E., Wu B.U., Bakker O.J., et al. Am J Gastroenterol. 2015 Dec;110[12]:1707-16.

20. Bang U.C., Semb S., Nojgaard C., Bendtsen F. World J Gastroenterol. 2008 May 21;14[19]:2968-76.

21. Warndorf M.G., Kurtzman J.T., Bartel M.J., et al. Clin Gastroenterol Hepatol. 2011 Aug;9[8]:705-9.

22. Hotz H.G., Foitzik T., Rohweder J., et al. J Gastrointest Surg. 1998 Nov-Dec;2[6]:518-25.

23. Brown A., Baillargeon J.D., Hughes M.D., et al. Pancreatology 2002;2:104-7.

24. Wu B.U., Hwang J.Q., Gardner T.H., et al. Clin Gastroenterol Hepatol. 2011 Aug;9[8]:710-7.

25. Forsmark C.E., Baillie J., AGA Institute Clinical Practice and Economics Committee, AGA Institute Governing Board. Gastroenterology. 2007 May;132[5]:2022-44.

26. Lankisch P.G., Mahlke R., Blum T., et al. Am J Gastroenterol. 2001;96:2081-5.

27. Wu B.U., Johannes R.S., Sun X., et al. Gastroenterology 2009;137:129-35.

28. Scherer J., Singh V.P., Pitchumoni C.S., Yadav D. J Clin Gastroenterol. 2014 Mar;48[3]:195-203.

29. Gubensek J., Buturovic-Ponikvar J., Romozi K., Ponikvar R. PLoS One. 2014 Jul 21;9[7]:e102748.

30. Chen J.H., Yeh J.H., Lai H.W., Liao C.S. World J Gastroenterol. 2004 Aug 1;10[15]:2272-4.

31. Tse F., Yuan Y. Cochrane Database Syst Rev. 2012 May 16;[5]:CD009779.

32. Folsch U.R., Nitsche R., Ludtke R., et al. N Engl J Med. 1997;336:237-42.

33. Al-Omran M., Albalawi Z.H., Tashkandi M.F., Al-Ansary L.A. Cochrane Database Syst Rev. 2010 Jan 20;[1]:CD002837.

34. Li J.Y., Yu T., Chen G.C., et al. PLoS One. 2013;8[6]:e64926.

35. Singh N., Sharma B., Sharma M., et al. Pancreas. 2012 Jan;41[1]:153-9.

36. Bakker O.J., van Brunschot S., van Santvoort H.C., et al. N Engl J Med. 2014 Nov 20;371[21]:1983-93.

37. Van Baal M.C., Besselink M.G., Bakker O.J., et al. Ann Surg. 2012;255:860–6.

38. Nealon W.H., Bawduniak J., Walser E.M. Ann Surg. 2004 Jun;239[6]:741-9.

39. Sanjay P., Yeeting S., Whigham C., Judson H., Polignano F.M., Tait I.S. Surg Endosc. 2008 Aug;22[8]:1832-7.

40. Nordback I., Pelli H., Lappalainen-Lehto R., Järvinen S., Räty S., Sand J. Gastroenterology. 2009 Mar;136[3]:848-55.

41. Besselink M.G., Verwer T.J., Schoenmaeckers E.J., et al. Arch Surg. 2007;142:1194-201.

42. Besselink M., van Santvoort H., Freeman M. et al. Pancreatology. 2013 Jul-Aug;13(4 Suppl 2):e1-15.

43. Hjalmar C., van Santvoort, H., Besselink M.G., et al. N Engl J Med. 2010;362:1491-502.

44. Varadarajulu S., Bang J.Y., Sutton B.S., et al. Gastroenterology. 2013;145:583-90.e1.

45. Akshintala V.S., Saxena P., Zaheer A., et al. Gastrointest Endosc. 2014 Jun;79[6]:921-8.

46. Jiang K, Huang W, Yang XN., et al. World J Gastroenterol. 2012;18:279–84.

47. Dervenis C., Smailis D., Hatzitheoklitos E. J Hepatobiliary Pancreat Surg. 2003;10[6]:415Y418.

48. Gloor B., Muller C.A., Worni M., et al. Arch Surg. 2001;136[5]:592Y596.

49. Nadkarni N.A., Khanna S., Vege S.S. Pancreas. 2013 Aug;42[6]:924-31.

50. Marshall G.T., Howell D.A., Hansen B.L., Amberson S.M., Abourjaily G.S., Bredenberg C.E. Arch Surg. 1996 Mar;131[3]:278-83.

51. Malbrain M.L., Cheatham M.L., Kirkpatrick A., et al. Intensive Care Med. 2006 Nov;32[11]:1722-32.

52. De Waele J.J. Leppaniemi A.K. World J Surg. 2009;33:1128-33.

53. Kirkpatrick A.W., Roberts D.J., De W.J., et al. Intensive Care Med. 2013 Jul;39[7]1190-206.
 

Historical perspective

The term “pancreas” derives its name from the Greek words pan (all) and kreas (flesh). Understanding pancreas physiology was first attempted in the 17th century by Regnier de Graaf^1. Giovanni Morgagni is credited with the first description of the syndrome of acute pancreatitis (AP) in 1761². Reginald Huber Fitz proposed the first classification of AP into hemorrhagic, gangrenous, and suppurative types in 1889³. The distinction of acute from chronic pancreatitis was not well described until the middle of the 20th century when Mandred W. Comfort gave a detailed account of chronic relapsing pancreatitis in 1946⁴.

Diagnosis and classification of severity

The diagnosis of AP is based on the presence of two of the three following criteria: typical abdominal pain (severe, upper abdominal pain frequently radiating to the back), serum amylase and/or lipase levels greater than 3 times the upper limit of normal, and/or characteristic imaging findings.

The original 1992 Atlanta classification provided the first blueprint to standardize how severity of AP was defined⁵. Over the years, better understanding of AP pathophysiology and its complications led to a greater focus on local and systemic determinants of severity⁶ and eventually the Revised Atlanta Classification (RAC) in 2013 (Table 1).
 

Management of acute pancreatitis

Prevention

Determination of etiology

The most common causes of AP are gallstones and alcohol, accounting for more than two-thirds of all cases¹³. Other etiologies include hypertriglyceridemia, ERCP, drugs induced, familial/hereditary, and post-traumatic. Initial work up includes a thorough history to quantify alcohol consumption and assess for recently started medications, measurement of liver injury tests¹⁴ and triglyceride levels, and performance of a transabdominal ultrasound to evaluate for biliary dilation, chole- and choledocholithiasis¹⁵.

Assessment of disease severity

Fluid resuscitation

Despite extensive research and trials using medications such as ulinastatin, octreotide, pentoxifylline, gabexate, N-acetyl cysteine, steroids, IL-10, and antibiotics²⁰, no pharmacologic agent has been shown to significantly alter the clinical course/outcomes of AP.

Adequate intravenous hydration remains the cornerstone of early management in AP²¹. Studies have demonstrated that increased intestinal permeability, secondary to reduced intestinal capillary microcirculation, leads to bacterial translocation and development of SIRS²². Intestinal microcirculation does not become as readily impaired, and there is a certain “latency” to its onset, from the insult that triggers pancreatitis. This gives rise to the concept of a “golden window” of 12-24 hours from the insult to potentially reverse such changes and prevent organ dysfunction. It has been shown that patients who are adequately resuscitated with intravenous fluids have lower risk for local and systemic complications²³.

Patients with predicted severe AP or those with persistent SIRS despite initial fluid resuscitation should be managed in a closely monitored unit, ideally an ICU. Patients with impending respiratory failure require mechanical ventilation, renal failure complicated by metabolic acidosis and/or hyperkalemia requires hemodialysis, and cardiovascular shock requires the initiation of vasopressors and continuous monitoring of blood pressure via an arterial line. A special entity that requires ICU level care is hypertriglyceridemia (HTG)-induced severe AP. HTG should be considered as the etiology of AP in certain clinical scenarios²⁸: previous history of HTG, poorly controlled diabetes mellitus, history of significant alcohol use, third trimester of pregnancy, and use of certain medications associated with HTG such as oral estrogens, tamoxifen, and propofol. Levels of triglyceride greater than 1000 mg/dL strongly point toward HTG being the etiology.

Plasmapheresis, which filters and removes triglycerides from plasma, has been reported as an efficient treatment in such patients based on case series^29,30. At this time its use may only be justified in patients with predicted severe AP from HTG, preferably within the first 24 hours of presentation.


Urgent ERCP

Nutrition

Recovery of the gut function is often delayed for several days or weeks in patients with severe AP. Studies have shown that prolonged fasting in such circumstances leads to malnutrition and worse prognosis^33,34. Enteral nutrition via a nasogastric (NG) or nasojejunal (NJ) tube is the preferred route of nutritional support, as it is associated with lower risk of infection, multi-organ failure, and mortality when compared to total parenteral nutrition³³.

The question of whether NJ feeding offers any additional advantages over NG feeding has not been clearly answered with a recent randomized trial showing NG feeds not to be inferior to NJ feeds³⁵. In regards to the timing of initiation of enteral nutrition, early nasoenteric feeding within 24 hours from presentation was found not to be superior compared to on-demand feeding in patients with predicted severe AP³⁶.


Strategies to decrease risk of recurrent attacks

Management of peripancreatic fluid collections

Patients with AP frequently develop peripancreatic fluid collections (PFCs). Based on the revised Atlanta classification, those are categorized into four types (Table 2, Figures 1-4).

The majority of acute PFCs in patients without evidence of pancreatic necrosis regress within a few weeks and thus intervention is not indicated early in the disease course. Current literature supports delaying the drainage/debridement of such collections for several weeks. The mortality from interventions decreases as the time to intervention from onset of symptoms increases⁴¹. Delaying intervention gives more time for recovery from systemic complications and allows the encapsulating wall and contents to organize further.

While surgery is still an option for patients with symptomatic mature PFCs, endoscopic ultrasound-guided drainage in expert hands has been shown to be cost effective, with shorter hospital stay and even decreased risk of cyst recurrence compared with surgical cyst-gastrostomy creation⁴⁴. Ultrasound or computed tomography-guided drainage of such collections with a percutaneous catheter is an equally efficacious option when compared to the endoscopic approach. However, patients undergoing endotherapy require fewer procedures and imaging studies and shorter length of stay⁴⁵ when compared with radiological interventions.

Although this topic has generated much debate, the majority of available evidence shows no clinical benefit from using prophylactic antibiotics to prevent infection in pancreatic necrosis⁴⁶.

Vascular complications

Vascular complications such as splanchnic vein thrombosis can occur in up to a quarter of AP patients⁴⁹. Anticoagulation is not usually indicated unless thrombosis is extensive and causes bowel ischemia. Arterial pseudoaneurysms are rare but life threatening complications of AP. They typically require interventional radiology guided coil embolization to prevent massive bleeding⁵⁰.

Abdominal compartment syndrome

Abdominal compartment syndrome is an end result of third spacing of fluid into the abdominal cavity secondary to inflammation and fluid resuscitation in severe pancreatitis. Abdominal pressure in patients can be monitored by measuring bladder pressures. Intra-abdominal hypertension is defined as a sustained pressure greater than 12 mm Hg, while abdominal compartment syndrome is defined as sustained intra-abdominal pressure greater than 20 mm Hg with new organ failure⁵¹. Intra-abdominal hypertension (IAH) is present in up to 75% of patients with severe AP. While all conservative measures to prevent development or worsening of IAH should be implemented (adequate sedation, decompression of bowel in patients with ileus, etc.), current guidelines do not recommend aggressive interventions to treat it. On the other hand, abdominal compartment syndrome is a life-threatening complication that requires urgent intervention to decrease intra-abdominal pressure, such as percutaneous drain placement or surgical fasciotomy^52,53.

Conclusion

The key principles in the management of acute pancreatitis are aggressive hydration and preventing development of end organ failure. In the last two decades there has been a paradigm shift in the guidelines for management of peripancreatic fluid collections and pancreatic necrosis. When feasible, drainage of these collections should be delayed and be performed using minimally invasive interventions. There is still an urgent need for developing and testing disease-specific treatments targeting control of the inflammatory response in the early phase of acute pancreatitis and prevention of development of severe disease with end-organ dysfunction.

Dr. Gulati is a gastroenterology and hepatology fellow at Allegheny Health Network, Pittsburgh, and Dr. Papachristou is professor of medicine, University of Pittsburgh School of Medicine, Pittsburgh.

References

1. Sleisenger and Fordtran’s Gastrointestinal and Liver Disease, Chapter 55, 923-33.

2. Morgagni G.B. [Fie Books on the Seats and Causes of Diseases as Discovered by the Anatomist]. Venice, Italy: Typographia Remondiniana;1761.

3. Fitz R.H. Boston Med Surg J. 1889;120:181-8.

4. Comfort M., Gambill E., Baggesnstoss A. Gastroenterology. 1946;6:238-76.

5. Bollen T.L., van Santvoort H.C., Besselink M.G., et al. Br J Surg. 2008;95:6–21.

6. Dellinger E.P., Forsmark C.E., Layer P., et al. Ann Surg. 2012 Dec;256[6]:875-80.

7. Kochar B., Akshintala V.S., Afghani E., et al. Gastrointest Endosc. 2015 Jan;81[1]:143-9.

8. Choudhary A., Bechtold M.L., Arif M., et al. Gastrointest Endosc. 2011 Feb;73[2]:275-82.

9. Shi Q.Q., Ning X.Y., Zhan L.L., Tang G.D., Lv X.P. World J Gastroenterol. 2014 Jun 14;20[22]:7040-8.

10. Elmunzer B.J., Waljee A.K., Elta G.H., Taylor J.R., Fehmi S.M., Higgins P.D. Gut. 2008 Sep;57[9]:1262-7.

11. Sethi S., Sethi N., Wadhwa V., Garud S., Brown A. Pancreas. 2014 Mar;43[2]:190-7. 
12. Elmunzer B.J., Serrano J., Chak A., et al. Trials. 2016 Mar 3;17[1]:120.

13. Lowenfels A.B., Maisonneuve P., Sullivan T. Curr Gastroenterol Rep. 2009;11:97-103.

14. Agarwal N., Pitchumoni C.S., Sivaprasad A.V. Am J Gastroenterol. 1990;85:356-66.

15. Tenner S., Baillie J., DeWitt J. Vege S.S. Am J Gastroenterol. 2013;108:1400-15.

16. Papachristou G.I., Muddana V., Yadav D., et al. Am J Gastroenterol. 2010;105:435-41.

17. Mounzer R., et al. Gastroenterology 2012;142:1476-82.

18. Working Group IAP/APA Acute Pancreatitis Guidelines. Pancreatology. 2013 Jul-Aug;13(4 Suppl 2):e1-15.

19. Koutroumpakis E., Wu B.U., Bakker O.J., et al. Am J Gastroenterol. 2015 Dec;110[12]:1707-16.

20. Bang U.C., Semb S., Nojgaard C., Bendtsen F. World J Gastroenterol. 2008 May 21;14[19]:2968-76.

21. Warndorf M.G., Kurtzman J.T., Bartel M.J., et al. Clin Gastroenterol Hepatol. 2011 Aug;9[8]:705-9.

22. Hotz H.G., Foitzik T., Rohweder J., et al. J Gastrointest Surg. 1998 Nov-Dec;2[6]:518-25.

23. Brown A., Baillargeon J.D., Hughes M.D., et al. Pancreatology 2002;2:104-7.

24. Wu B.U., Hwang J.Q., Gardner T.H., et al. Clin Gastroenterol Hepatol. 2011 Aug;9[8]:710-7.

25. Forsmark C.E., Baillie J., AGA Institute Clinical Practice and Economics Committee, AGA Institute Governing Board. Gastroenterology. 2007 May;132[5]:2022-44.

26. Lankisch P.G., Mahlke R., Blum T., et al. Am J Gastroenterol. 2001;96:2081-5.

27. Wu B.U., Johannes R.S., Sun X., et al. Gastroenterology 2009;137:129-35.

28. Scherer J., Singh V.P., Pitchumoni C.S., Yadav D. J Clin Gastroenterol. 2014 Mar;48[3]:195-203.

29. Gubensek J., Buturovic-Ponikvar J., Romozi K., Ponikvar R. PLoS One. 2014 Jul 21;9[7]:e102748.

30. Chen J.H., Yeh J.H., Lai H.W., Liao C.S. World J Gastroenterol. 2004 Aug 1;10[15]:2272-4.

31. Tse F., Yuan Y. Cochrane Database Syst Rev. 2012 May 16;[5]:CD009779.

32. Folsch U.R., Nitsche R., Ludtke R., et al. N Engl J Med. 1997;336:237-42.

33. Al-Omran M., Albalawi Z.H., Tashkandi M.F., Al-Ansary L.A. Cochrane Database Syst Rev. 2010 Jan 20;[1]:CD002837.

34. Li J.Y., Yu T., Chen G.C., et al. PLoS One. 2013;8[6]:e64926.

35. Singh N., Sharma B., Sharma M., et al. Pancreas. 2012 Jan;41[1]:153-9.

36. Bakker O.J., van Brunschot S., van Santvoort H.C., et al. N Engl J Med. 2014 Nov 20;371[21]:1983-93.

37. Van Baal M.C., Besselink M.G., Bakker O.J., et al. Ann Surg. 2012;255:860–6.

38. Nealon W.H., Bawduniak J., Walser E.M. Ann Surg. 2004 Jun;239[6]:741-9.

39. Sanjay P., Yeeting S., Whigham C., Judson H., Polignano F.M., Tait I.S. Surg Endosc. 2008 Aug;22[8]:1832-7.

40. Nordback I., Pelli H., Lappalainen-Lehto R., Järvinen S., Räty S., Sand J. Gastroenterology. 2009 Mar;136[3]:848-55.

41. Besselink M.G., Verwer T.J., Schoenmaeckers E.J., et al. Arch Surg. 2007;142:1194-201.

42. Besselink M., van Santvoort H., Freeman M. et al. Pancreatology. 2013 Jul-Aug;13(4 Suppl 2):e1-15.

43. Hjalmar C., van Santvoort, H., Besselink M.G., et al. N Engl J Med. 2010;362:1491-502.

44. Varadarajulu S., Bang J.Y., Sutton B.S., et al. Gastroenterology. 2013;145:583-90.e1.

45. Akshintala V.S., Saxena P., Zaheer A., et al. Gastrointest Endosc. 2014 Jun;79[6]:921-8.

46. Jiang K, Huang W, Yang XN., et al. World J Gastroenterol. 2012;18:279–84.

47. Dervenis C., Smailis D., Hatzitheoklitos E. J Hepatobiliary Pancreat Surg. 2003;10[6]:415Y418.

48. Gloor B., Muller C.A., Worni M., et al. Arch Surg. 2001;136[5]:592Y596.

49. Nadkarni N.A., Khanna S., Vege S.S. Pancreas. 2013 Aug;42[6]:924-31.

50. Marshall G.T., Howell D.A., Hansen B.L., Amberson S.M., Abourjaily G.S., Bredenberg C.E. Arch Surg. 1996 Mar;131[3]:278-83.

51. Malbrain M.L., Cheatham M.L., Kirkpatrick A., et al. Intensive Care Med. 2006 Nov;32[11]:1722-32.

52. De Waele J.J. Leppaniemi A.K. World J Surg. 2009;33:1128-33.

53. Kirkpatrick A.W., Roberts D.J., De W.J., et al. Intensive Care Med. 2013 Jul;39[7]1190-206.
 

AGA’s focus on young GIs

The AGA Trainee and Early Career Committee (formerly Trainee and Young GI Committee) is composed of 12 trainee and early-career AGA members and meets twice a year to develop programs and events specifically targeted to trainees and gastroenterologists (GIs) in their first five years out of fellowship training. The committee was formed by the AGA in February 2013 to address the specific needs of early-career GI professionals and to develop programs to expose younger members to all that the AGA has to offer. The new committee also became a creative space to organize efforts to increase membership among early-career GIs. Trainee and Early Career Committee members are selected for 2-year terms and represent fellowship training programs, universities, and practices from around the nation. Each committee member serves simultaneously on one other AGA committee, which gives young GIs additional opportunities for leadership roles. The committee meets regularly with AGA staff and a governing board liaison to discuss committee goals and the issues most relevant to physicians during and directly after GI fellowship training. The committee also provides feedback to other committees about how programs and initiatives might involve or impact GI fellows and recent graduates. The result is a unique focus group where young GIs from all over the country work collectively to improve the young GI experience through flagship programs like the Regional Practice Skills Workshop, the Young Delegates Program, and Trainee and Early Career events at Digestive Disease Week (DDW)®.

AGA Regional Practice Skills Workshops

The workshop agenda is similar across locations and includes sessions on career options in research and clinical practice, how to evaluate a job, contract negotiation, health care reform, financial planning, and work-life balance. The program is geared toward second- and third-year fellows, recent fellowship graduates, and those considering a job or career change. All workshops include catered meals and are free to both AGA members and non-members. Those interested in attending one of the workshops can find more information at http://www.gastro.org/trainees. The Trainee and Early Career committee is also looking to expand to additional cities in future years so that more trainees and early-career GIs can participate in these workshops.
 

The AGA Young Delegates program

The AGA highly values the efforts of our Young Delegates, and the Trainee and Early Career Committee considers them a talent pool from which we can elicit input, select committee members, and find future leaders. More importantly, we hope that the program allows young AGA members to increasingly engage with the AGA to refresh, improve, and strengthen the society. To become a Young Delegate, please visit www.gastro.org/youngdelegates to provide us with your information.
 

Trainee and early career GIs at DDW

The Trainee and Early Career Committee sponsors several events at DDW to bring together fellows and early-career GIs from all over the country. Each year, our committee hosts a DDW Trainee and Early Career symposium to provide practical advice for early-career GIs from all practice settings. Our DDW 2016 symposium was entitled “Surviving The First Years in Clinical Practice – Roundtable with the Experts,” and featured prominent leaders who shared career perspectives with attendees through formal presentations and more casual discussion. Attendees gained insider tips on how to design and run a fiscally prosperous practice, coding and documentation, and building and maintaining a clinical practice referral base from expert AGA leaders. We are now in the process of planning the DDW 2017 Trainee and Early Career symposium that will focus on “The Road to Leadership in GI.”

Come join us!

The success of the AGA depends on the 16,000 members who volunteer their time for committees, councils, and the governing board. Since its inception, the Trainee and Early Career Committee has allowed young GIs to have a role in the AGA as well as benefit from all of the resources that the AGA has to offer in leadership training, networking, and career preparation. In the past three years, participation of young GIs in the Trainee and Early Career Committee events has been on the rise, which we hope is a reflection of our efforts to address the educational needs of early GIs and the transition from fellowship to practice. We would love to see more fellows and early-career GIs involved!

For more information about the Trainee and Early Career committee, becoming a committee member, and our programs, please visit http://www.gastro.org/trainees. If you have any ideas that you think the committee should consider, please let us know at [email protected].
 

Dr. Liang is an instructor in the division of gastroenterology, New York University School of Medicine, New York, and an attending physician in the VA New York Harbor Healthcare System, New York. Dr. Kushner is a transplant hepatology fellow in the division of gastroenterology, University of California, San Francisco. Dr. May is assistant professor in the division of digestive diseases, David Geffen School of Medicine, University of California, Los Angeles, and an attending physician in the department of gastroenterology in the VA Greater Los Angeles Healthcare System, Los Angeles.

AGA’s focus on young GIs

The AGA Trainee and Early Career Committee (formerly Trainee and Young GI Committee) is composed of 12 trainee and early-career AGA members and meets twice a year to develop programs and events specifically targeted to trainees and gastroenterologists (GIs) in their first five years out of fellowship training. The committee was formed by the AGA in February 2013 to address the specific needs of early-career GI professionals and to develop programs to expose younger members to all that the AGA has to offer. The new committee also became a creative space to organize efforts to increase membership among early-career GIs. Trainee and Early Career Committee members are selected for 2-year terms and represent fellowship training programs, universities, and practices from around the nation. Each committee member serves simultaneously on one other AGA committee, which gives young GIs additional opportunities for leadership roles. The committee meets regularly with AGA staff and a governing board liaison to discuss committee goals and the issues most relevant to physicians during and directly after GI fellowship training. The committee also provides feedback to other committees about how programs and initiatives might involve or impact GI fellows and recent graduates. The result is a unique focus group where young GIs from all over the country work collectively to improve the young GI experience through flagship programs like the Regional Practice Skills Workshop, the Young Delegates Program, and Trainee and Early Career events at Digestive Disease Week (DDW)®.

AGA Regional Practice Skills Workshops

The workshop agenda is similar across locations and includes sessions on career options in research and clinical practice, how to evaluate a job, contract negotiation, health care reform, financial planning, and work-life balance. The program is geared toward second- and third-year fellows, recent fellowship graduates, and those considering a job or career change. All workshops include catered meals and are free to both AGA members and non-members. Those interested in attending one of the workshops can find more information at http://www.gastro.org/trainees. The Trainee and Early Career committee is also looking to expand to additional cities in future years so that more trainees and early-career GIs can participate in these workshops.
 

The AGA Young Delegates program

The AGA highly values the efforts of our Young Delegates, and the Trainee and Early Career Committee considers them a talent pool from which we can elicit input, select committee members, and find future leaders. More importantly, we hope that the program allows young AGA members to increasingly engage with the AGA to refresh, improve, and strengthen the society. To become a Young Delegate, please visit www.gastro.org/youngdelegates to provide us with your information.
 

Trainee and early career GIs at DDW

The Trainee and Early Career Committee sponsors several events at DDW to bring together fellows and early-career GIs from all over the country. Each year, our committee hosts a DDW Trainee and Early Career symposium to provide practical advice for early-career GIs from all practice settings. Our DDW 2016 symposium was entitled “Surviving The First Years in Clinical Practice – Roundtable with the Experts,” and featured prominent leaders who shared career perspectives with attendees through formal presentations and more casual discussion. Attendees gained insider tips on how to design and run a fiscally prosperous practice, coding and documentation, and building and maintaining a clinical practice referral base from expert AGA leaders. We are now in the process of planning the DDW 2017 Trainee and Early Career symposium that will focus on “The Road to Leadership in GI.”

Come join us!

The success of the AGA depends on the 16,000 members who volunteer their time for committees, councils, and the governing board. Since its inception, the Trainee and Early Career Committee has allowed young GIs to have a role in the AGA as well as benefit from all of the resources that the AGA has to offer in leadership training, networking, and career preparation. In the past three years, participation of young GIs in the Trainee and Early Career Committee events has been on the rise, which we hope is a reflection of our efforts to address the educational needs of early GIs and the transition from fellowship to practice. We would love to see more fellows and early-career GIs involved!

For more information about the Trainee and Early Career committee, becoming a committee member, and our programs, please visit http://www.gastro.org/trainees. If you have any ideas that you think the committee should consider, please let us know at [email protected].
 

Dr. Liang is an instructor in the division of gastroenterology, New York University School of Medicine, New York, and an attending physician in the VA New York Harbor Healthcare System, New York. Dr. Kushner is a transplant hepatology fellow in the division of gastroenterology, University of California, San Francisco. Dr. May is assistant professor in the division of digestive diseases, David Geffen School of Medicine, University of California, Los Angeles, and an attending physician in the department of gastroenterology in the VA Greater Los Angeles Healthcare System, Los Angeles.

AGA’s focus on young GIs

The AGA Trainee and Early Career Committee (formerly Trainee and Young GI Committee) is composed of 12 trainee and early-career AGA members and meets twice a year to develop programs and events specifically targeted to trainees and gastroenterologists (GIs) in their first five years out of fellowship training. The committee was formed by the AGA in February 2013 to address the specific needs of early-career GI professionals and to develop programs to expose younger members to all that the AGA has to offer. The new committee also became a creative space to organize efforts to increase membership among early-career GIs. Trainee and Early Career Committee members are selected for 2-year terms and represent fellowship training programs, universities, and practices from around the nation. Each committee member serves simultaneously on one other AGA committee, which gives young GIs additional opportunities for leadership roles. The committee meets regularly with AGA staff and a governing board liaison to discuss committee goals and the issues most relevant to physicians during and directly after GI fellowship training. The committee also provides feedback to other committees about how programs and initiatives might involve or impact GI fellows and recent graduates. The result is a unique focus group where young GIs from all over the country work collectively to improve the young GI experience through flagship programs like the Regional Practice Skills Workshop, the Young Delegates Program, and Trainee and Early Career events at Digestive Disease Week (DDW)®.

AGA Regional Practice Skills Workshops

The workshop agenda is similar across locations and includes sessions on career options in research and clinical practice, how to evaluate a job, contract negotiation, health care reform, financial planning, and work-life balance. The program is geared toward second- and third-year fellows, recent fellowship graduates, and those considering a job or career change. All workshops include catered meals and are free to both AGA members and non-members. Those interested in attending one of the workshops can find more information at http://www.gastro.org/trainees. The Trainee and Early Career committee is also looking to expand to additional cities in future years so that more trainees and early-career GIs can participate in these workshops.
 

The AGA Young Delegates program

The AGA highly values the efforts of our Young Delegates, and the Trainee and Early Career Committee considers them a talent pool from which we can elicit input, select committee members, and find future leaders. More importantly, we hope that the program allows young AGA members to increasingly engage with the AGA to refresh, improve, and strengthen the society. To become a Young Delegate, please visit www.gastro.org/youngdelegates to provide us with your information.
 

Trainee and early career GIs at DDW

The Trainee and Early Career Committee sponsors several events at DDW to bring together fellows and early-career GIs from all over the country. Each year, our committee hosts a DDW Trainee and Early Career symposium to provide practical advice for early-career GIs from all practice settings. Our DDW 2016 symposium was entitled “Surviving The First Years in Clinical Practice – Roundtable with the Experts,” and featured prominent leaders who shared career perspectives with attendees through formal presentations and more casual discussion. Attendees gained insider tips on how to design and run a fiscally prosperous practice, coding and documentation, and building and maintaining a clinical practice referral base from expert AGA leaders. We are now in the process of planning the DDW 2017 Trainee and Early Career symposium that will focus on “The Road to Leadership in GI.”

Come join us!

The success of the AGA depends on the 16,000 members who volunteer their time for committees, councils, and the governing board. Since its inception, the Trainee and Early Career Committee has allowed young GIs to have a role in the AGA as well as benefit from all of the resources that the AGA has to offer in leadership training, networking, and career preparation. In the past three years, participation of young GIs in the Trainee and Early Career Committee events has been on the rise, which we hope is a reflection of our efforts to address the educational needs of early GIs and the transition from fellowship to practice. We would love to see more fellows and early-career GIs involved!

For more information about the Trainee and Early Career committee, becoming a committee member, and our programs, please visit http://www.gastro.org/trainees. If you have any ideas that you think the committee should consider, please let us know at [email protected].
 

Dr. Liang is an instructor in the division of gastroenterology, New York University School of Medicine, New York, and an attending physician in the VA New York Harbor Healthcare System, New York. Dr. Kushner is a transplant hepatology fellow in the division of gastroenterology, University of California, San Francisco. Dr. May is assistant professor in the division of digestive diseases, David Geffen School of Medicine, University of California, Los Angeles, and an attending physician in the department of gastroenterology in the VA Greater Los Angeles Healthcare System, Los Angeles.

The correct answer is B: endoscopic suture removal. As the prevalence of bariatric surgery increases to address the obesity epidemic, endoscopists are increasingly called upon to evaluate postbariatric patients.1 In one case series of patients undergoing EGD for upper GI symptoms post-RYGB, normal postsurgical anatomy was found in 31.6%, anastomotic stricture in 52.6%, marginal ulcer in 15.8%, unraveled suture material causing functional obstruction in 4% and gastro-gastric fistula in 2.6% of cases.2 Another series reported unraveled suture material thought to be contributing to upper GI symptoms in up to 10% of cases.3 Suture material is found by a mean of 34 weeks after RYGB, and presenting symptoms include abdominal pain in 65%, nausea 52%, dysphagia 22%, and melena in 13%. Unraveled suture material may be associated with marginal ulceration, or may cause obstruction as it presents a mechanical obstruction to foodstuff as it passes through the gastrojejunal anastomosis. A series of 29 therapeutic endoscopic suture removal cases reported resolution or improvement of symptoms in 83% of patients and no complications or anastomotic leaks.3

AGA Institute

References

1. ASGE Standards of Practice Committee, Evans J.A., Muthusamy V.R., et al. The role of endoscopy in the bariatric surgery patient. Gastrointest Endosc. 2015;8:1063-72.
2. Lee J.K., Van Dam J., Morton J.M., et al. Endoscopy is accurate, safe, and effective in the assessment and management of complications following gastric bypass surgery. Am J Gastroenterol. 2009;104:575-82.
3. Yu S., Jastrow K., Clapp B., et al. Foreign material erosion after laparoscopic Roux-en-Y gastric bypass: findings and treatment. Surg Endosc. 2007;21:1216-20.

The correct answer is B: endoscopic suture removal. As the prevalence of bariatric surgery increases to address the obesity epidemic, endoscopists are increasingly called upon to evaluate postbariatric patients.1 In one case series of patients undergoing EGD for upper GI symptoms post-RYGB, normal postsurgical anatomy was found in 31.6%, anastomotic stricture in 52.6%, marginal ulcer in 15.8%, unraveled suture material causing functional obstruction in 4% and gastro-gastric fistula in 2.6% of cases.2 Another series reported unraveled suture material thought to be contributing to upper GI symptoms in up to 10% of cases.3 Suture material is found by a mean of 34 weeks after RYGB, and presenting symptoms include abdominal pain in 65%, nausea 52%, dysphagia 22%, and melena in 13%. Unraveled suture material may be associated with marginal ulceration, or may cause obstruction as it presents a mechanical obstruction to foodstuff as it passes through the gastrojejunal anastomosis. A series of 29 therapeutic endoscopic suture removal cases reported resolution or improvement of symptoms in 83% of patients and no complications or anastomotic leaks.3

AGA Institute

References

1. ASGE Standards of Practice Committee, Evans J.A., Muthusamy V.R., et al. The role of endoscopy in the bariatric surgery patient. Gastrointest Endosc. 2015;8:1063-72.
2. Lee J.K., Van Dam J., Morton J.M., et al. Endoscopy is accurate, safe, and effective in the assessment and management of complications following gastric bypass surgery. Am J Gastroenterol. 2009;104:575-82.
3. Yu S., Jastrow K., Clapp B., et al. Foreign material erosion after laparoscopic Roux-en-Y gastric bypass: findings and treatment. Surg Endosc. 2007;21:1216-20.

The correct answer is B: endoscopic suture removal. As the prevalence of bariatric surgery increases to address the obesity epidemic, endoscopists are increasingly called upon to evaluate postbariatric patients.1 In one case series of patients undergoing EGD for upper GI symptoms post-RYGB, normal postsurgical anatomy was found in 31.6%, anastomotic stricture in 52.6%, marginal ulcer in 15.8%, unraveled suture material causing functional obstruction in 4% and gastro-gastric fistula in 2.6% of cases.2 Another series reported unraveled suture material thought to be contributing to upper GI symptoms in up to 10% of cases.3 Suture material is found by a mean of 34 weeks after RYGB, and presenting symptoms include abdominal pain in 65%, nausea 52%, dysphagia 22%, and melena in 13%. Unraveled suture material may be associated with marginal ulceration, or may cause obstruction as it presents a mechanical obstruction to foodstuff as it passes through the gastrojejunal anastomosis. A series of 29 therapeutic endoscopic suture removal cases reported resolution or improvement of symptoms in 83% of patients and no complications or anastomotic leaks.3

AGA Institute

References

1. ASGE Standards of Practice Committee, Evans J.A., Muthusamy V.R., et al. The role of endoscopy in the bariatric surgery patient. Gastrointest Endosc. 2015;8:1063-72.
2. Lee J.K., Van Dam J., Morton J.M., et al. Endoscopy is accurate, safe, and effective in the assessment and management of complications following gastric bypass surgery. Am J Gastroenterol. 2009;104:575-82.
3. Yu S., Jastrow K., Clapp B., et al. Foreign material erosion after laparoscopic Roux-en-Y gastric bypass: findings and treatment. Surg Endosc. 2007;21:1216-20.

In a recent video interview, Richard Peek Jr., MD, AGAF, Editor in Chief, and Douglas Corley, MD, PhD, Deputy Editor in Chief, of Gastroenterology explained how trainees and young GIs fit into their plans for the journal. Good news: this constituency is among the editors’ top priorities.

The editors have plans to implement a year-long editorial fellowship later in their term, which will allow an individual to get hands-on experience in the editorial process.

The editors also appreciate the fresh take young investigators have on research. To encourage continued high-quality submissions from young investigators, the editors will decrease submission fees for young investigators and work to increase the visibility of young investigator research.

The editors also plan to develop new features within the Gastroenterology Mentor, Education and Training Corner that will be of interest to trainees and early career GIs.

Watch the full video interview on AGA’s YouTube Channel: https://www.youtube.com/user/AmerGastroAssn.

The discussion on young investigators begins at minute 5:24.

In a recent video interview, Richard Peek Jr., MD, AGAF, Editor in Chief, and Douglas Corley, MD, PhD, Deputy Editor in Chief, of Gastroenterology explained how trainees and young GIs fit into their plans for the journal. Good news: this constituency is among the editors’ top priorities.

The editors have plans to implement a year-long editorial fellowship later in their term, which will allow an individual to get hands-on experience in the editorial process.

The editors also appreciate the fresh take young investigators have on research. To encourage continued high-quality submissions from young investigators, the editors will decrease submission fees for young investigators and work to increase the visibility of young investigator research.

The editors also plan to develop new features within the Gastroenterology Mentor, Education and Training Corner that will be of interest to trainees and early career GIs.

Watch the full video interview on AGA’s YouTube Channel: https://www.youtube.com/user/AmerGastroAssn.

The discussion on young investigators begins at minute 5:24.

In a recent video interview, Richard Peek Jr., MD, AGAF, Editor in Chief, and Douglas Corley, MD, PhD, Deputy Editor in Chief, of Gastroenterology explained how trainees and young GIs fit into their plans for the journal. Good news: this constituency is among the editors’ top priorities.

The editors have plans to implement a year-long editorial fellowship later in their term, which will allow an individual to get hands-on experience in the editorial process.

The editors also appreciate the fresh take young investigators have on research. To encourage continued high-quality submissions from young investigators, the editors will decrease submission fees for young investigators and work to increase the visibility of young investigator research.

The editors also plan to develop new features within the Gastroenterology Mentor, Education and Training Corner that will be of interest to trainees and early career GIs.

Watch the full video interview on AGA’s YouTube Channel: https://www.youtube.com/user/AmerGastroAssn.

The discussion on young investigators begins at minute 5:24.

Dear Colleagues,

Acute pancreatitis has long been one of the “bread and butter” conditions in gastroenterology and having up-to-date knowledge on its management will serve our community well. In this issue of The New Gastroenterologist, Abhishek Gulati and Georgios Papachristou (University of Pittsburgh) provide a comprehensive review of the latest advances in the treatment of acute pancreatitis and its complications, which has direct application to GI clinical practice.

Bryson W. Katona is a instructor of medicine in the division of gasteroenterology at the University of Pennsylvania.

Dear Colleagues,

Acute pancreatitis has long been one of the “bread and butter” conditions in gastroenterology and having up-to-date knowledge on its management will serve our community well. In this issue of The New Gastroenterologist, Abhishek Gulati and Georgios Papachristou (University of Pittsburgh) provide a comprehensive review of the latest advances in the treatment of acute pancreatitis and its complications, which has direct application to GI clinical practice.

Bryson W. Katona is a instructor of medicine in the division of gasteroenterology at the University of Pennsylvania.

Dear Colleagues,

Acute pancreatitis has long been one of the “bread and butter” conditions in gastroenterology and having up-to-date knowledge on its management will serve our community well. In this issue of The New Gastroenterologist, Abhishek Gulati and Georgios Papachristou (University of Pittsburgh) provide a comprehensive review of the latest advances in the treatment of acute pancreatitis and its complications, which has direct application to GI clinical practice.

Bryson W. Katona is a instructor of medicine in the division of gasteroenterology at the University of Pennsylvania.

INTRODUCTION

Renal cell carcinoma (RCC) is the most common malignancy arising in the kidney, comprising 90% of all renal tumors.¹ Approximately 55,000 new RCC cases are diagnosed each year.² Patients with RCC are often asymptomatic, and most cases are discovered as incidental findings on abdominal imaging performed during evaluation of nonrenal complaints. Limited-stage RCC that is found early can be cured surgically, with estimated 5-year survival rates approaching 90%; however, long-term survival for metastatic disease is poor, with rates ranging from 0% to 20%.² Advanced RCC is resistant to conventional chemotherapy and radiotherapy, and outcomes for patients with metastatic or unresectable RCC remain poor. However, the recent development of new therapeutic modalities that target tumor molecular pathways has expanded the treatment options for these patients and changed the management of RCC.

EPIDEMIOLOGY AND CLASSIFICATION

Median age at diagnosis in the United States is 64 years. Men have a higher incidence of RCC than women, with the highest incidence seen in American Indian and Alaska Native men (30.1 per 100,000 population). Genetic syndromes account for 2% to 4% of all RCCs.² Risk factors for RCC include smoking, hypertension, obesity, and acquired cystic kidney disease that is associated with end-stage renal failure.³ Longer duration of tobacco use is associated with a more aggressive course.

The 2004 World Health Organization (WHO) classification of renal tumors summarizes the previous classification systems (including the Heidelberg and Mainz classification systems) to describe different categories of RCC based on histologic and molecular genetics characteristics.² Using the WHO classification criteria, RCC comprises 90% of all renal tumors, with clear cell being the most common type (80%).² Other types of renal tumors include papillary, chromophobe, oncocytoma, and collecting-duct or Bellini duct tumors. Approximately 3% to 5% of tumors are unclassified. Oncocytomas are generally considered benign, and chromophobe tumors typically have an indolent course and rarely metastasize. Sarcomatoid differentiation can be seen in any histologic type and is associated with a worse prognosis. While different types of tumors may be seen in the kidney (such as transitional cell or lymphomas), the focus of this review is the primary malignancies of the renal parenchyma.

FAMILIAL SYNDROMES

Several genetic syndromes have been identified by studying families with inherited RCC. Among these, von Hippel-Lindau (VHL) gene mutation is the most commonly found inherited genetic defect. Table 1 summarizes the incidence of gene mutations and the corresponding histologic appearance of the most common sporadic and hereditary RCCs.⁴

VHL disease is an autosomal dominant familial syndrome. Patients with this mutation are at higher risk for developing RCC (clear cell histology), retinal angiomas, pheochromocytomas, as well as hemangioblastomas of the central nervous system (CNS).⁴ Of all the genetic mutations seen in RCC, the somatic mutation in the VHL tumor-suppressor gene is by far the most common.⁵ VHL targets hypoxia–inducible factor-1 alpha (HIF-α) for ubiquitination and subsequent degradation, which has been shown to suppress the growth of clear-cell RCC in mouse models.^6–8 HIF expression under hypoxic conditions leads to activation of a number of genes important in blood vessel development, cell proliferation, and glucose metabolism, including vascular endothelial growth factor (VEGF), erythropoietin, platelet-derived growth factor beta (PDGF-β), transforming growth factor alpha (TGF-α), and glucose transporter-1 (GLUT-1). Mutation in the VHL gene prevents degradation of the HIF-α protein, thereby leading to increased expression of these downstream proteins, including MET and Axl. The upregulation of these angiogenic factors is thought to be the underlying process for increased vascularity of CNS hemangioblastomas and clear-cell renal tumors in VHL disease.^4–8

Other less common genetic syndromes seen in hereditary RCC include hereditary papillary RCC, hereditary leiomyomatosis, and Birt-Hogg-Dubé (BHD) syndrome.⁹ In hereditary papillary RCC, the MET gene is mutated. BHD syndrome is a rare, autosomal dominant syndrome characterized by hair follicle hamartomas of the face and neck. About 15% of patients have multiple renal tumors, the majority of which are of the chromophobe or mixed chromophobe-oncocytoma histology. The BHD gene encodes the protein folliculin, which is thought to be a tumor-suppressor gene.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 74-year-old man who works as an airplane mechanic repairman presents to the emergency department with sudden worsening of chronic right upper arm and shoulder pain after lifting a jug of orange juice. He does not have a significant past medical history and initially thought that his pain was due to a work-related injury. Upon initial evaluation in the emergency department he is found to have a fracture of his right humerus. Given that the fracture appears to be pathologic, further work-up is recommended.

Most patients are asymptomatic until the disease becomes advanced. The classic triad of flank pain, hematuria, and palpable abdominal mass is seen in approximately 10% of patients with RCC, partly because of earlier detection of renal masses by imaging performed for other purposes.¹⁰ Less frequently, patients present with signs or symptoms of metastatic disease such as bone pain or fracture (as seen in the case patient), painful adenopathy, and pulmonary symptoms related to mediastinal masses. Fever, weight loss, anemia, and/or varicocele often occur in young patients (≤ 46 years) and may indicate the presence of a hereditary form of the disease. Patients may present with paraneoplastic syndromes seen as abnormalities on routine blood work. These can include polycythemia or elevated liver function tests (LFTs) without the presence of liver metastases (known as Stauffer syndrome), which can be seen in localized renal tumors. Nearly half (45%) of patients present with localized disease, 25% present with locally advanced disease, and 30% present with metastatic disease.¹¹ Bone is the second most common site of distant metastatic spread (following lung) in patients with advanced RCC.

• What is the approach to initial evaluation for a patient with suspected RCC?

Initial evaluation consists of a physical exam, laboratory tests including complete blood count (CBC) and comprehensive metabolic panel (calcium, serum creatinine, LFTs, lactate dehydrogenase [LDH], and urinalysis), and imaging. Imaging studies include computed tomography (CT) scan with contrast of the abdomen and pelvis or magnetic resonance imaging (MRI) of the abdomen and chest imaging. A chest radiograph may be obtained, although a chest CT is more sensitive for the presence of pulmonary metastases. MRI can be used in patients with renal dysfunction to evaluate the renal vein and inferior vena cava (IVC) for thrombus or to determine the presence of local invasion.¹² Although bone and brain are common sites for metastases, routine imaging is not indicated unless the patient is symptomatic. The value of positron emission tomography in RCC remains undetermined at this time.

Staging is done according to the American Joint Committee on Cancer (AJCC) staging classification for RCC; the Figure summarizes the staging and 5-year survival data based on this classification scheme.^4,13

LIMITED-STAGE DISEASE

• What are the therapeutic options for limited-stage disease?

For patients with nondistant metastases, or limited-stage disease, surgical intervention with curative intent is considered. Convention suggests considering definitive surgery for patients with stage I and II disease, select patients with stage III disease with pathologically enlarged retroperitoneal lymph nodes, patients with IVC and/or cardiac atrium involvement of tumor thrombus, and patients with direct extension of the renal tumor into the ipsilateral adrenal gland if there is no evidence of distant disease. While there may be a role for aggressive surgical intervention in patients with distant metastatic disease, this topic will not be covered in this review.

SURGICAL INTERVENTION

Once patients are determined to be appropriate candidates for surgical removal of a renal tumor, the urologist will perform either a radical nephrectomy or a nephron-sparing nephrectomy, also called a partial nephrectomy. The urologist will evaluate the patient based on his or her body habitus, the location of the tumor, whether multiple tumors in one kidney or bilateral tumors are present, whether the patient has a solitary kidney or otherwise impaired kidney function, and whether the patient has a history of a hereditary syndrome involving kidney cancer as this affects the risk of future kidney tumors.

A radical nephrectomy is surgically preferred in the presence of the following factors: tumor larger than 7 cm in diameter, a more centrally located tumor, suspicion of lymph node involvement, tumor involvement with renal vein or IVC, and/or direct extension of the tumor into the ipsilateral adrenal gland. Nephrectomy involves ligation of the vascular supply (renal artery and vein) followed by removal of the kidney and surrounding Gerota’s fascia. The ipsilateral adrenal gland is removed if there is a high-risk for or presence of invasion of the adrenal gland. Removal of the adrenal gland is not standard since the literature demonstrates there is less than a 10% chance of solitary, ipsilateral adrenal gland involvement of tumor at the time of nephrectomy in the absence of high-risk features, and a recent systematic review suggests that the chance may be as low as 1.8%.¹⁴ Preoperative factors that correlated with adrenal involvement included upper pole kidney location, renal vein thrombosis, higher T stage (T3a and greater), multifocal tumors, and evidence for distant metastases or lymph node involvement. Lymphadenectomy previously had been included in radical nephrectomy but now is performed selectively. Radical nephrectomy may be performed as

Conversely, a nephron-sparing approach is preferred for tumors less than 7 cm in diameter, for patients with a solitary kidney or impaired renal function, for patients with multiple small ipsilateral tumors or with bilateral tumors, or for radical nephrectomy candidates with comorbidities for whom a limited intervention is deemed to be a lower-risk procedure. A nephron-sparing procedure may also be performed open or laparoscopically. In nephron-sparing procedures, the tumor is removed along with a small margin of normal parenchyma.¹⁵

In summary, the goal of surgical intervention is curative intent with removal of the tumor while maintaining as much residual renal function as possible to limit long-term morbidity of chronic kidney disease and associated cardiovascular events.¹⁸ Oncologic outcomes for radical nephrectomy and partial nephrectomy are similar. In one study, overall survival was slightly lower in the partial nephrectomy cohort, but only a small number of the deaths were due to RCC.¹⁹

ADJUVANT THERAPY

Adjuvant systemic therapy currently has no role following nephrectomy for RCC because no systemic therapy has been able to reduce the likelihood of relapse. Randomized trials of cytokine therapy (eg, interferon, interleukin 2) or tyrosine kinase inhibitors (TKIs; eg, sorafenib, sunitinib) with observation alone in patients with locally advanced completely resected RCC have shown no delay in time to relapse or improvement of survival with adjuvant therapy.²⁰ Similarly, adjuvant radiation therapy has not shown benefit even in patients with nodal involvement or incomplete resection.²¹ Therefore, observation remains the standard of care after nephrectomy.

RENAL TUMOR ABLATION

For patients who are deemed not to be surgical candidates due to age, comorbidities, or patient preference and who have tumors less than 4 cm in size (stage I tumors), ablative techniques may be considered. The 2 most well-studied and effective techniques at present are cryoablation and radiofrequency ablation (RFA). Microwave ablation may be an option in some facilities, but the data in RCC are limited. An emerging ablative technique under investigation is irreversible electroporation. At present, the long-term efficacy of all ablative techniques is unknown.

Patient selection is undertaken by urologists and interventional radiologists who evaluate the patient with ultrasound, CT, and/or MRI to determine the location and size of the tumor and the presence or absence of metastatic disease. A pretreatment biopsy is recommended to document the histology of the lesion to confirm a malignancy and to guide future treatment for recurrent or metastatic disease. Contraindications to the procedure include the presence of metastatic disease, a life expectancy of less than 1 year, general medical instability, or uncorrectable coagulopathy due to increased risk of bleeding complications. Tumors in close proximity to the renal hilum or collecting system are a contraindication to the procedure because of the risk for hemorrhage or damage to the collecting system. The location of the tumor in relation to the vasculature is also important to maximize efficacy because the vasculature acts as a “heat sink,” causing dissipation of the thermal energy. Occasionally, stenting of the proximal ureter due to upper tumor location is necessary to prevent thermal injury that could lead to urine leaks.

Selection of the modality to be used primarily depends on operator comfort, which translates to good patient outcomes, such as better cancer control and fewer complications. Cryoablation and RFA have both demonstrated good clinical efficacy and cancer control of 89% and 90%, respectively, with comparable complication rates.²² There have been no studies performed directly comparing the modalities.

Cryoablation

Cryoablation is performed through the insertion of a probe into the tumor, which may be done through a surgical or percutaneous approach. Once the probe is in place, a high- pressure gas (argon, nitrogen) is passed through the probe and upon entering a low pressure region the gas cools. The gas is able to cool to temperatures as low as –185°C. The tissue is then rewarmed through the use of helium, which conversely warms when entering a low pressure area. The process of freezing followed by rewarming subsequently causes cell death/tissue destruction through direct cell injury from cellular dehydration and vascular injury. Clinically, 2 freeze-thaw cycles are used to treat a tumor.^23,24

Radiofrequency ablation, or RFA, targets tumors via an electrode placed within the mass that produces intense frictional heat from medium-frequency alternating current (approximately 500 kHz) produced by a connected generator that is grounded on the patient. The thermal energy created causes coagulative necrosis. Due to the reliance on heat for tumor destruction, central lesions are less amenable to this approach because of the “heat sink” effect from the hilum.²⁴

Microwave Ablation

Microwave ablation, like RFA, relies on the generation of frictional heat to cause cell death by coagulative necrosis. In this case, the friction is created through the activation of water molecules; because of the different thermal kinetics involved with microwave ablation, the “heat sink” effect is minimized when treatment is employed near large vessels, in comparison to RFA.²⁴ The data on this mechanism of ablation are still maturing, with varied outcomes thus far. One study demonstrated outcomes comparable to RFA and cryoablation, with cancer-specific survival of 97.8% at 3 years.²⁵ However, a study by Castle and colleagues²⁶ demonstrated higher recurrence rates. The overarching impediment to widespread adoption of microwave ablation is inconclusive data gleaned from studies with small numbers of patients with limited follow up. The role of this modality will need to be revisited.

Irreversible Electroporation

Irreversible electroporation (IRE) is under investigation. IRE is a non-thermal ablative technique that employs rapid electrical pulses to create pores in cell membranes, leading to cell death. The postulated benefits of IRE include the lack of an effect from “heat sinks” and less collateral damage to the surrounding tissues, when compared with the thermal modalities. In a human phase 1 study of patients undergoing IRE prior to immediate surgical resection, the procedure appeared feasible and safe.²⁷ Significant concerns for this method of ablation possibly inducing cardiac arrhythmias, and the resultant need for sedation with neuromuscular blockade and associated electrocardiography monitoring, may impede its implementation in nonresearch settings.²⁴

ACTIVE SURVEILLANCE

Due to the more frequent use of imaging for various indications, there has been an increase in the discovery of small renal masses (SRM); 85% of RCC that present in an asymptomatic or incidental manner are tumors under 4 cm in diameter.^28,29 The role of active surveillance is evolving, but is primarily suggested for patients who are not candidates for more aggressive intervention based on comorbidities. A recent prospective, nonrandomized analysis of data from the Delayed Intervention and Surveillance for Small Renal Masses (DISSRM) registry evaluated outcomes for patients with SRM looking at primary intervention compared with active surveillance.³⁰ The primary intervention selected was at the discretion of the provider; treatments included partial nephrectomy, RFA, and cryoablation, and active surveillance patients were followed with imaging every 6 months. Progression of SRM, with recommendation for delayed intervention, was defined as a growth rate of mass greater than 0.5 cm/year, size greater than 4 cm, or hematuria. Thirty-six of 158 patients on active surveillance met criteria for progression; 21 underwent delayed intervention. Of note, even the patients who progressed but did not undergo delayed intervention did not develop metastatic disease during the follow-up interval. With a median follow-up of 2 years, cancer-specific survival was noted to be 99% and 100% at 5 years for primary intervention and active surveillance, respectively. Overall survival at 2 years for primary intervention was 98% and 96% for active surveillance; at 5 years, the survival rates were 92% and 75% (P = 0.06). Of note, 2 patients in the primary intervention arm died of RCC, while none in the active surveillance arm died. As would be expected, active surveillance patients were older, had a worse performance status, and had more comorbidities. Interestingly, 40% of patients enrolled selected active surveillance as their preferred management for SRM. The DISSRM results were consistent with data from the Renal Cell Consortium of Canada and other retrospective reviews.^31–33

• What is the approach to follow-up after treatment of localized RCC?

After a patient undergoes treatment for a localized RCC, the goal is to optimize oncologic outcomes, monitor for treatment sequelae, such as renal failure, and focus on survivorship. At this time, there is no consensus in the literature or across published national and international guidelines with regards to the appropriate schedule for surveillance to achieve these goals. In principle, the greatest risk for recurrence occurs within the first 3 years, so many guidelines focus on this timeframe. Likewise, the route of spread tends to be hematogenous, so patients present with pulmonary, bone, and brain metastases, in addition to local recurrence within the renal bed. Symptomatic recurrences often are seen

METASTATIC DISEASE

CASE CONTINUED

CT scan with contrast of the chest, abdomen, and pelvis as well as bone scan are done. CT of the abdomen and pelvis demonstrates a 7.8-cm left renal mass arising from the lower pole of the left kidney. Paraesophageal lymphadenopathy and mesenteric nodules are also noted. CT of the chest demonstrates bilateral pulmonary emboli. Bone scan is significant for increased activity related to the pathological fracture involving the right humerus. The patient undergoes surgery to stabilize the pathologic fracture of his humerus. He is diagnosed with metastatic RCC (clear cell histology) and undergoes palliative debulking nephrectomy.

• How is prognosis defined for metastatic RCC?

PROGNOSTIC MODELS

Limited-stage RCC that is found early can be cured surgically, with estimated 5-year survival rates for stage T1 and T2 disease approaching 90%; however, long-term survival for metastatic disease is poor, with rates ranging from 0% to 20%.¹³ Approximately 30% of patients have metastatic disease at diagnosis, and about one-third of patients who have undergone treatment for localized disease experience relapse.^36,37 Common sites of metastases include lung, lymph nodes, bone, liver, adrenal gland, and brain.

Prognostic scoring systems have been developed to define risk groups and assist with determining appropriate therapy in the metastatic setting. The most widely used validated prognostic factor model is that from the Memorial Sloan-Kettering Cancer Center (MSKCC), which was developed using a multivariate analysis derived from data of patients enrolled in clinical trials and treated with interferon alfa.³⁸ The factors included in the MSKCC model are Karnofsky performance status less than 80, time from diagnosis to treatment with interferon alfa less than 12 months, hemoglobin level less than lower limit of laboratory’s reference range, LDH level greater than 1.5 times the upper limit of laboratory’s reference range, and corrected serum calcium level greater than 10 mg/dL. Risk groups are categorized as favorable (0 risk factors), intermediate (1 to 2 risk factors), and poor (3 or more risk factors).³⁹ Median survival for favorable-, intermediate-, and poor-risk patients was 20, 10, and 4 months, respectively.⁴⁰

Another prognostic model, the International Metastatic RCC Database Consortium, or Heng, model was developed to evaluate prognosis in patients treated with VEGF-targeted therapy.⁴¹ This model was developed from a retrospective study of patients treated with sunitinib, sorafenib, and bevacizumab plus interferon alfa or prior immunotherapy. Prognostic factors in this model include 4 of the 5 MSKCC risk factors (hemoglobin level, corrected serum calcium level, Karnofsky performance status, and time to initial diagnosis). Additionally, this model includes both absolute neutrophil and platelet counts greater than the upper limit of normal. Risk groups are identified as favorable (0 risk factors), intermediate (1 to 2 risk factors), and poor (3 or more risk factors). Median survival for favorable-, intermediate-, and poor-risk patients was not reached, 27 months, and 8.8 months, respectively. The University of California, Los Angeles scoring algorithm to predict survival after nephrectomy and immunotherapy (SANI) in patients with metastatic RCC is another prognostic model that can be used. This simplified scoring system incorporates lymph node status, constitutional symptoms, metastases location, histology, and thyroid stimulating hormone (TSH) level.⁴²

The role of debulking or cytoreductive nephrectomy in treatment of metastatic RCC is well established. Large randomized studies have demonstrated a statistically significant median survival benefit for patients undergoing nephrectomy plus interferon alfa therapy compared with patients treated with interferon alfa alone (13.6 months versus 7.8 months, respectively).⁴³ The role of cytoreductive nephrectomy in combination with antiangiogenic agents is less clear. While a retrospective study investigating outcomes of patients with metastatic RCC receiving anti-VEGF agents showed a prolonged survival with nephrectomy, results of large randomized trials are not yet available.^44,45 Patients with lung-only metastases, good prognostic features, and a good performance status are historically the most likely to benefit from cytoreductive surgery.

CASE CONTINUED

Based on the MSKCC prognostic factor model, the patient is considered to be in the intermediate-risk group (Karnofsky performance status of 80, calcium 9.5 mg/dL, LDH 204 U/L, hemoglobin 13.6 g/dL). He is started on treatment for his bilateral pulmonary emboli and recovers well from orthopedic surgery as well as palliative debulking nephrectomy.

Several approaches to systemic therapy for advanced RCC have been taken based on the histologic type of the tumor. Clear-cell is by far the predominant histologic type in RCC. Several options are available as first-line treatment for patients with metastatic clear-cell RCC (Table 2).^46–54 These include biologic agents such as high-dose interleukin-2 (IL-2) immune therapy, as well as targeted therapies including TKIs and anti-VEGF antibodies. The mammalian target of rapamycin (mTOR) inhibitor temsirolimus is recommended as first-line therapy in patients with poor prognosis only. Second-line therapies for clear-cell RCC following antiangiogenic therapy include TKIs, mTOR inhibitors, nivolumab (PD-1 inhibitor), and the combination of the TKI lenvatinib and mTOR inhibitor everolimus.⁵⁵ In addition, after initial cytokine therapy, TKIs, temsirolimus, and the anti-VEGF antibody bevacizumab are other treatment options available to patients. Best supportive care should always be provided along with initial and subsequent therapies. Clinical trials are also an appropriate choice as first-line or subsequent therapies. All of these therapies require periodic monitoring to prevent and quickly treat adverse effects. Table 3 lists recommended monitoring parameters for each of these agents.⁵⁶

Based on several studies, TKIs seem to be less effective in patients with non–clear-cell type histology.^57,58 In these patients, risk factors can guide therapy. In the ASPEN trial, where 108 patients were randomly assigned to everolimus or sunitinib, patients in the good- and intermediate-risk groups had longer overall and median progression-free survival (PFS) on sunitinib (8.3 months versus 5.3 months, respectively). However, those in the poor-risk group had a longer median overall survival with everolimus.⁵⁹ Given that the role of targeted therapies in non–clear-cell RCCs is less well established, enrollment in clinical trials should be considered as a first-line treatment option.²¹

Sarcomatoid features can be observed in any of the histologic types of RCC, and RCC with these features has an aggressive course and a poor prognosis. Currently, there is no standard therapy for treatment of patients with metastatic or unresectable RCC with sarcomatoid features.⁶⁰ Chemotherapeutic regimens used for soft tissue sarcomas, including a trial of ifosfamide and doxorubicin, did not show any objective response.⁶¹ A small trial of 10 patients treated with doxorubicin and gemcitabine resulted in complete response in 2 patients and partial response in 1 patient.⁶²

Enrollment in a clinical trial remains a first-line treatment option for these patients. More recently, a phase 2 trial of sunitinib and gemcitabine in patients with sarcomatoid (39 patients) and/or poor-risk (33 patients) metastatic RCC showed overall response rates (ORR) of 26% and 24%, respectively. A higher clinical benefit rate (defined as ORR plus stable disease) was seen in patients with tumors containing more than 10% sarcomatoid histology, as compared with patients whose tumors contained less than 10% sarcomatoid histology. Neutropenia (n = 20), anemia (n = 10), and fatigue (n = 7) were the most common grade 3 toxicities seen in all the patients. Although this was a small study, the results showed a trend towards better efficacy of the combination therapy as compared with the single-agent regimen. Currently, another study is underway to further investigate this in a larger group of patients.⁶³

BIOLOGICS

Cytokine therapy, including high-dose IL-2 and interferon alfa, had long been the only first-line treatment option for patients with metastatic or unresectable RCC. Studies of high-dose IL-2 have shown an ORR of 25% and durable response in up to 11% of patients with clear-cell histology.⁶⁴ Toxicities were similar to those previously observed with high-dose IL-2 treatment; the most commonly observed grade 3 toxicities were hypotension and capillary leak syndrome. IL-2 requires strict monitoring (Table 3). It is important to note that retrospective studies evaluating the safety and efficacy of using IL-2 as second-line treatment in patients previously treated with TKIs demonstrated significant toxicity without achieving partial or complete response in any of the patients.⁶⁵

Prior to the advent of TKIs in the treatment of RCC, interferon alfa was a first-line treatment option for those who could not receive high-dose IL-2. It has been shown to produce response rates of approximately 20%, with maximum response seen with a higher dose range of 5 to 20 million units daily in 1 study.^66,67 However, with the introduction of TKIs, which produce a higher and more durable response, interferon alfa alone is no longer recommended as a treatment option.

Bevacizumab is a recombinant humanized monoclonal antibody that binds and neutralizes VEGF-A. Given overexpression of VEGF in RCC, the role of bevacizumab both as a single agent and in combination with interferon alfa has been investigated. In a randomized phase 2 study involving patients with cytokine-refractory disease, bevacizumab produced a 10% response rate and PFS of 4.8 months compared to patients treated with placebo.⁶⁸ In the AVOREN trial, the addition of bevacizumab (10 mg/kg intravenously [IV] every 2 weeks) to interferon alfa (9 million units subcutaneously [SC] 3 times weekly) was shown to significantly increase PFS compared with interferon alfa alone (10.2 months versus 5.4 months; P = 0.0001).^47,48 Adverse effects of this combination therapy include fatigue and asthenia. Additionally, hypertension, proteinuria, and bleeding occurred.

TYROSINE KINASE INHIBITORS

TKIs have largely replaced IL-2 as first-line therapy for metastatic RCC. Axitinib, pazopanib, sorafenib, and sunitinib and can be used as first-line therapy. All of the TKIs can be used as subsequent therapy.

Sunitinib

Sunitinib is an orally administered TKI that inhibits VEGF receptor (VEGFR) types 1 and 2, PDGF receptors (PDGFR) α and β, stem cell factor receptor (c-Kit), and FLT-3 and RET kinases. Motzer and colleagues^52,53 compared sunitinib 50 mg daily orally for 4 weeks with 2 weeks off to the then standard of care, interferon alfa 9 million units SC 3 times weekly. Sunitinib significantly increased the overall objective response rate (47% versus 12%; P < 0.001), PFS (11 versus 5 months; P < 0.001), and overall survival (26.4 versus 21.8 months; hazard ratio [HR], 0.821). The most common side effects are diarrhea, fatigue, nausea/vomiting, anorexia, hypertension, stomatitis, and hand-foot syndrome, occurring in more than 30% of patients. Often patients will require dose reductions or temporary discontinuations to tolerate therapy. Alternative dosing strategies (eg, 50 mg dose orally daily for 2 weeks alternating with 1-week free interval) have been attempted but not prospectively evaluated for efficacy.^69–71

Pazopanib

Pazopanib is an oral multi-kinase inhibitor of VEGFR types 1 and 2, PDGFR, and c-KIT. Results of a phase 3 trial comparing pazopanib (800 mg orally daily) to placebo favored the TKI, with a PFS of 9.2 months versus 4.2 months. A subset of treatment-naïve patients had a longer PFS of 11.1 versus 2.8 months and a response rate of 32% versus 4%.⁷² This led to a noninferiority phase 3 trial comparing pazopanib with sunitinib as first-line therapy.⁵⁰ In this study, PFS was similar (8.4 versus 9.5 months; HR 1.05), and overall safety and quality-of-life endpoints favored pazopanib. Much less fatigue, stomatitis, hand-foot syndrome, and thrombocytopenia occurred with pazopanib, whereas hair color changes, weight loss, alopecia, and elevations of LFT enzymes occurred more frequently with pazopanib. Hypertension is common with the administration of pazopanib as well.

Sorafenib

Sorafenib is an orally administered inhibitor of Raf, serine/threonine kinase, VEGFR, PDGFR, FLT-3, c-Kit, and RET. The pivotal phase 3 Treatment Approaches in Renal Cancer Global Evaluation Trial (TARGET) compared sorafenib (400 mg orally twice daily) with placebo in patients who had progressed on prior cytokine-based therapy.⁷³ A final analysis, which excluded patients who were allowed to cross over therapies, found improved overall survival times (14.3 versus 1.8 months, P = 0.029).⁵¹ Sorafenib is associated with lower rates of diarrhea, rash, fatigue, hand-foot syndrome, alopecia, hypertension, and nausea than sunitinib, although these agents have not been compared to one another.

Axitinib

Axitinib is an oral inhibitor of VEGFRs 1, 2, and 3. Results of the phase 3 AXIS trial comparing axitinib (5 mg orally twice daily) with sorafenib (400 mg orally twice daily) in patients receiving 1 prior systemic therapy showed axitinib was more active than sorafenib in improving ORR (19% versus 9%; P = 0.001) and PFS (6.7 versus 4.7 months; P < 0.001), although no difference in overall survival times was noted.⁷⁴ In a subsequent phase 3 trial comparing these drugs in the first-line setting, axitinib showed a nonsignificantly higher response rate and PFS. Despite this, the National Comprehensive Cancer Network guidelines consider axitinib an acceptable first-line therapy because activity with acceptable toxicity was demonstrated (Table 2).⁴⁶ The most common adverse effects of axitinib are diarrhea, hypertension, fatigue, decreased appetite, dysphonia, hypothyroidism, and upper abdominal pain.

CABOZANTINIB

Given that resistance eventually develops in most patients treated with standard treatments, including bevacizumab and TKIs, the need to evaluate the safety and efficacy of novel agents targeting VEGFR and overcoming this resistance is of vital importance. Cabozantinib is an oral small-molecule inhibitor of VEGFR, Met, and Axl, all tyrosine kinases implicated in metastatic RCC. Overexpression of Met and Axl, which occurs as a result of inactivation of the VHL gene, is associated with a poor prognosis in patients with RCC. In a

MTOR INHIBITORS

The mTOR inhibitors, temsirolimus and everolimus, are also approved for the treatment of metastatic or advanced RCC. These drugs block mTOR’s phosphorylation and subsequent translation of mRNA to inhibit cell proliferation, cell growth, and angiogenesis.⁷⁷ Temsirolimus can be used as first-line therapy for patients with a poor prognosis, and everolimus is appropriate as a subsequent therapy.

Temsirolimus is an intravenous prodrug of rapamycin. It was the first of the class to be approved for metastatic RCC for treatment-naïve patients with a poor prognosis (ie, at least 3 of 6 predictors of poor survival based on MSKCC model).⁵⁴ The pivotal ARCC trial compared temsirolimus (25 mg IV weekly) alone, interferon alfa (3 million units SC 3 times weekly) alone, or the combination (temsirolimus 15 mg IV weekly plus interferon alfa 6 million units SC 3 times weekly). In this trial, temsirolimus monotherapy produced a significantly longer overall survival time than interferon alfa alone (10.9 versus 7.3 months; P = 0.008) and improved PFS time when administered alone or in combination with interferon alfa (3.8 and 3.7 months, respectively, versus 1.9 months). Because no real efficacy advantage of the combination was demonstrated, temsirolimus is administered alone. The most common adverse effects of temsirolimus are asthenia, rash, anemia, nausea, anorexia, pain, and dyspnea. Additionally, hyperglycemia, hyper-cholesterolemia, and hyperlipidemia occur with these agents. Noninfectious pneumonitis is a rare but often fatal complication.

Everolimus is also an orally administered derivative of rapamycin that is approved for use after failure of VEGF-targeted therapies. The results of the landmark trial RECORD-1 demonstrated that everolimus (10 mg orally daily) is effective at prolonging PFS (4 versus 1.9 months; P < 0.001) when compared with best supportive care, a viable treatment option at the time of approval.⁷⁸ The most common adverse effects of everolimus are stomatitis, rash, fatigue, asthenia, and diarrhea. As with temsirolimus, elevations in glucose, lipids, and triglycerides and noninfectious pneumonitis can occur.

TKI + MTOR INHIBITOR

Lenvatinib is also a small molecule targeting multiple tyrosine kinases, primarily VEGF2. Combined with the mTOR inhibitor everolimus, it has been shown to be an effective regimen in patients with metastatic RCC who have failed other therapies. In a randomized phase 2 study involving patients with advanced or metastatic clear-cell RCC, patients were randomly assigned to receive either lenvatinib (24 mg/day), everolimus (10 mg/day), or lenvatinib plus everolimus (18 mg/day and 5 mg/day, respectively). Patients received the treatment continuously on a 28-day cycle until progression or inability to tolerate toxicity. Patients in the lenvatinib plus everolimus arm had median PFS of 14.6 months (95% CI 5.9 to 20.1) versus 5.5 months (95% CI 3.5 to 7.1) with everlolimus alone (HR 0.40 [95% CI 0.24 to 0.68]; P = 0.0005). PFS with levantinib alone was 7.4 months (95% CI 5.6 to 10.20; HR 0.66 [95% CI 0.30 to 1.10]; P = 0.12). In addition, PFS with levantinib alone was significantly prolonged in comparison with everolimus alone (HR 0.61 [95% CI 0.38 to 0.98]; P = 0.048). Grade 3 or 4 toxicity were less frequent in the everolimus only arm and the most common grade 3 or 4 toxicity in the lenvatinib plus everolimus arm was diarrhea. The results of this study show that the combination of lenvatinib plus everolimus is an acceptable second-line option for treatment of patients with advanced or metastatic RCC.⁵⁵

The patient is initially started on pazopanib and tolerates the medication well, with partial response to the treatment. However, on restaging scans he is noted to have small bowel perforation. Pazopanib is discontinued until the patient has a full recovery. He is then started on everolimus. Restaging scans done 3 months after starting everolimus demonstrate disease progression.

• What is the appropriate next step in treatment?

PD1 BLOCKADE

Programmed death 1 (PD-1) protein is a T-cell inhibitory receptor with 2 ligands, PD-L1 and PD-L2. PD-L1 is expressed on many tumors. Blocking the interaction between PD-1 and PD-L1 by anti-PD-1 humanized antibodies potentiates a robust immune response and has been a breakthrough in the field of cancer immunotherapy.⁷⁹ Previous studies have demonstrated that overexpression of PD-L1 leads to worse outcomes and poor prognosis in patients with RCC.⁸⁰ Nivolumab, a fully human IgG4 PD-1 immune checkpoint inhibitor, blocks the interaction between PD-1 and its ligands, PD-L1 and PD-L2. In a randomized, open-label, phase 3 study comparing nivolumab with everolimus in patients with RCC who had previously undergone treatment with other standard therapies, Motzer and colleagues⁸¹ demonstrated a longer overall survival time and fewer adverse effects with nivolumab. In this study, 821 patients with clear-cell RCC were randomly assigned to receive nivolumab (3 mg/kg of body weight IV every 2 weeks) or everolimus (10 mg orally once daily). The median overall survival time with nivolumab was 25 months versus 19.6 months with everolimus (P < 0.0148). Nineteen percent of patients receiving nivolumab experienced grade 3 or 4 toxicities, with fatigue being the most common adverse effect. Grade 3 or 4 toxicities were observed in 37% of patients treated with everolimus, with anemia being the most common. Based on the results of this trial, on November 23, 2015, the U.S. Food and Drug Administration approved nivolumab to treat patients with metastatic RCC who have received a prior antiangiogenic therapy.

CASE CONCLUSION

Both TKI and mTOR inhibitor therapy fail, and the patient is eligible for third-line therapy. Because of his previous GI perforation, other TKIs are not an option. The patient opts for enrollment in hospice due to declining performance status. For other patients in this situation with a good performance status, nivolumab would be a reasonable option.

FUTURE DIRECTIONS

With the approval of nivolumab, multiple treatment options are now available for patients with metastatic or unresectable RCC. Development of other PD-1 inhibitors and immunotherapies as well as multi-targeted TKIs will only serve to expand treatment options for these patients. Given the aggressive course and poor prognosis of non-clear cell renal cell tumors and those with sarcomatoid features, evaluation of systemic and targeted therapies for these subtypes should remain active areas of research and investigation.

INTRODUCTION

Renal cell carcinoma (RCC) is the most common malignancy arising in the kidney, comprising 90% of all renal tumors.¹ Approximately 55,000 new RCC cases are diagnosed each year.² Patients with RCC are often asymptomatic, and most cases are discovered as incidental findings on abdominal imaging performed during evaluation of nonrenal complaints. Limited-stage RCC that is found early can be cured surgically, with estimated 5-year survival rates approaching 90%; however, long-term survival for metastatic disease is poor, with rates ranging from 0% to 20%.² Advanced RCC is resistant to conventional chemotherapy and radiotherapy, and outcomes for patients with metastatic or unresectable RCC remain poor. However, the recent development of new therapeutic modalities that target tumor molecular pathways has expanded the treatment options for these patients and changed the management of RCC.

EPIDEMIOLOGY AND CLASSIFICATION

Median age at diagnosis in the United States is 64 years. Men have a higher incidence of RCC than women, with the highest incidence seen in American Indian and Alaska Native men (30.1 per 100,000 population). Genetic syndromes account for 2% to 4% of all RCCs.² Risk factors for RCC include smoking, hypertension, obesity, and acquired cystic kidney disease that is associated with end-stage renal failure.³ Longer duration of tobacco use is associated with a more aggressive course.

The 2004 World Health Organization (WHO) classification of renal tumors summarizes the previous classification systems (including the Heidelberg and Mainz classification systems) to describe different categories of RCC based on histologic and molecular genetics characteristics.² Using the WHO classification criteria, RCC comprises 90% of all renal tumors, with clear cell being the most common type (80%).² Other types of renal tumors include papillary, chromophobe, oncocytoma, and collecting-duct or Bellini duct tumors. Approximately 3% to 5% of tumors are unclassified. Oncocytomas are generally considered benign, and chromophobe tumors typically have an indolent course and rarely metastasize. Sarcomatoid differentiation can be seen in any histologic type and is associated with a worse prognosis. While different types of tumors may be seen in the kidney (such as transitional cell or lymphomas), the focus of this review is the primary malignancies of the renal parenchyma.

FAMILIAL SYNDROMES

Several genetic syndromes have been identified by studying families with inherited RCC. Among these, von Hippel-Lindau (VHL) gene mutation is the most commonly found inherited genetic defect. Table 1 summarizes the incidence of gene mutations and the corresponding histologic appearance of the most common sporadic and hereditary RCCs.⁴

VHL disease is an autosomal dominant familial syndrome. Patients with this mutation are at higher risk for developing RCC (clear cell histology), retinal angiomas, pheochromocytomas, as well as hemangioblastomas of the central nervous system (CNS).⁴ Of all the genetic mutations seen in RCC, the somatic mutation in the VHL tumor-suppressor gene is by far the most common.⁵ VHL targets hypoxia–inducible factor-1 alpha (HIF-α) for ubiquitination and subsequent degradation, which has been shown to suppress the growth of clear-cell RCC in mouse models.^6–8 HIF expression under hypoxic conditions leads to activation of a number of genes important in blood vessel development, cell proliferation, and glucose metabolism, including vascular endothelial growth factor (VEGF), erythropoietin, platelet-derived growth factor beta (PDGF-β), transforming growth factor alpha (TGF-α), and glucose transporter-1 (GLUT-1). Mutation in the VHL gene prevents degradation of the HIF-α protein, thereby leading to increased expression of these downstream proteins, including MET and Axl. The upregulation of these angiogenic factors is thought to be the underlying process for increased vascularity of CNS hemangioblastomas and clear-cell renal tumors in VHL disease.^4–8

Other less common genetic syndromes seen in hereditary RCC include hereditary papillary RCC, hereditary leiomyomatosis, and Birt-Hogg-Dubé (BHD) syndrome.⁹ In hereditary papillary RCC, the MET gene is mutated. BHD syndrome is a rare, autosomal dominant syndrome characterized by hair follicle hamartomas of the face and neck. About 15% of patients have multiple renal tumors, the majority of which are of the chromophobe or mixed chromophobe-oncocytoma histology. The BHD gene encodes the protein folliculin, which is thought to be a tumor-suppressor gene.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 74-year-old man who works as an airplane mechanic repairman presents to the emergency department with sudden worsening of chronic right upper arm and shoulder pain after lifting a jug of orange juice. He does not have a significant past medical history and initially thought that his pain was due to a work-related injury. Upon initial evaluation in the emergency department he is found to have a fracture of his right humerus. Given that the fracture appears to be pathologic, further work-up is recommended.

Most patients are asymptomatic until the disease becomes advanced. The classic triad of flank pain, hematuria, and palpable abdominal mass is seen in approximately 10% of patients with RCC, partly because of earlier detection of renal masses by imaging performed for other purposes.¹⁰ Less frequently, patients present with signs or symptoms of metastatic disease such as bone pain or fracture (as seen in the case patient), painful adenopathy, and pulmonary symptoms related to mediastinal masses. Fever, weight loss, anemia, and/or varicocele often occur in young patients (≤ 46 years) and may indicate the presence of a hereditary form of the disease. Patients may present with paraneoplastic syndromes seen as abnormalities on routine blood work. These can include polycythemia or elevated liver function tests (LFTs) without the presence of liver metastases (known as Stauffer syndrome), which can be seen in localized renal tumors. Nearly half (45%) of patients present with localized disease, 25% present with locally advanced disease, and 30% present with metastatic disease.¹¹ Bone is the second most common site of distant metastatic spread (following lung) in patients with advanced RCC.

• What is the approach to initial evaluation for a patient with suspected RCC?

Initial evaluation consists of a physical exam, laboratory tests including complete blood count (CBC) and comprehensive metabolic panel (calcium, serum creatinine, LFTs, lactate dehydrogenase [LDH], and urinalysis), and imaging. Imaging studies include computed tomography (CT) scan with contrast of the abdomen and pelvis or magnetic resonance imaging (MRI) of the abdomen and chest imaging. A chest radiograph may be obtained, although a chest CT is more sensitive for the presence of pulmonary metastases. MRI can be used in patients with renal dysfunction to evaluate the renal vein and inferior vena cava (IVC) for thrombus or to determine the presence of local invasion.¹² Although bone and brain are common sites for metastases, routine imaging is not indicated unless the patient is symptomatic. The value of positron emission tomography in RCC remains undetermined at this time.

Staging is done according to the American Joint Committee on Cancer (AJCC) staging classification for RCC; the Figure summarizes the staging and 5-year survival data based on this classification scheme.^4,13

LIMITED-STAGE DISEASE

• What are the therapeutic options for limited-stage disease?

For patients with nondistant metastases, or limited-stage disease, surgical intervention with curative intent is considered. Convention suggests considering definitive surgery for patients with stage I and II disease, select patients with stage III disease with pathologically enlarged retroperitoneal lymph nodes, patients with IVC and/or cardiac atrium involvement of tumor thrombus, and patients with direct extension of the renal tumor into the ipsilateral adrenal gland if there is no evidence of distant disease. While there may be a role for aggressive surgical intervention in patients with distant metastatic disease, this topic will not be covered in this review.

SURGICAL INTERVENTION

Once patients are determined to be appropriate candidates for surgical removal of a renal tumor, the urologist will perform either a radical nephrectomy or a nephron-sparing nephrectomy, also called a partial nephrectomy. The urologist will evaluate the patient based on his or her body habitus, the location of the tumor, whether multiple tumors in one kidney or bilateral tumors are present, whether the patient has a solitary kidney or otherwise impaired kidney function, and whether the patient has a history of a hereditary syndrome involving kidney cancer as this affects the risk of future kidney tumors.

A radical nephrectomy is surgically preferred in the presence of the following factors: tumor larger than 7 cm in diameter, a more centrally located tumor, suspicion of lymph node involvement, tumor involvement with renal vein or IVC, and/or direct extension of the tumor into the ipsilateral adrenal gland. Nephrectomy involves ligation of the vascular supply (renal artery and vein) followed by removal of the kidney and surrounding Gerota’s fascia. The ipsilateral adrenal gland is removed if there is a high-risk for or presence of invasion of the adrenal gland. Removal of the adrenal gland is not standard since the literature demonstrates there is less than a 10% chance of solitary, ipsilateral adrenal gland involvement of tumor at the time of nephrectomy in the absence of high-risk features, and a recent systematic review suggests that the chance may be as low as 1.8%.¹⁴ Preoperative factors that correlated with adrenal involvement included upper pole kidney location, renal vein thrombosis, higher T stage (T3a and greater), multifocal tumors, and evidence for distant metastases or lymph node involvement. Lymphadenectomy previously had been included in radical nephrectomy but now is performed selectively. Radical nephrectomy may be performed as

Conversely, a nephron-sparing approach is preferred for tumors less than 7 cm in diameter, for patients with a solitary kidney or impaired renal function, for patients with multiple small ipsilateral tumors or with bilateral tumors, or for radical nephrectomy candidates with comorbidities for whom a limited intervention is deemed to be a lower-risk procedure. A nephron-sparing procedure may also be performed open or laparoscopically. In nephron-sparing procedures, the tumor is removed along with a small margin of normal parenchyma.¹⁵

In summary, the goal of surgical intervention is curative intent with removal of the tumor while maintaining as much residual renal function as possible to limit long-term morbidity of chronic kidney disease and associated cardiovascular events.¹⁸ Oncologic outcomes for radical nephrectomy and partial nephrectomy are similar. In one study, overall survival was slightly lower in the partial nephrectomy cohort, but only a small number of the deaths were due to RCC.¹⁹

ADJUVANT THERAPY

Adjuvant systemic therapy currently has no role following nephrectomy for RCC because no systemic therapy has been able to reduce the likelihood of relapse. Randomized trials of cytokine therapy (eg, interferon, interleukin 2) or tyrosine kinase inhibitors (TKIs; eg, sorafenib, sunitinib) with observation alone in patients with locally advanced completely resected RCC have shown no delay in time to relapse or improvement of survival with adjuvant therapy.²⁰ Similarly, adjuvant radiation therapy has not shown benefit even in patients with nodal involvement or incomplete resection.²¹ Therefore, observation remains the standard of care after nephrectomy.

RENAL TUMOR ABLATION

For patients who are deemed not to be surgical candidates due to age, comorbidities, or patient preference and who have tumors less than 4 cm in size (stage I tumors), ablative techniques may be considered. The 2 most well-studied and effective techniques at present are cryoablation and radiofrequency ablation (RFA). Microwave ablation may be an option in some facilities, but the data in RCC are limited. An emerging ablative technique under investigation is irreversible electroporation. At present, the long-term efficacy of all ablative techniques is unknown.

Patient selection is undertaken by urologists and interventional radiologists who evaluate the patient with ultrasound, CT, and/or MRI to determine the location and size of the tumor and the presence or absence of metastatic disease. A pretreatment biopsy is recommended to document the histology of the lesion to confirm a malignancy and to guide future treatment for recurrent or metastatic disease. Contraindications to the procedure include the presence of metastatic disease, a life expectancy of less than 1 year, general medical instability, or uncorrectable coagulopathy due to increased risk of bleeding complications. Tumors in close proximity to the renal hilum or collecting system are a contraindication to the procedure because of the risk for hemorrhage or damage to the collecting system. The location of the tumor in relation to the vasculature is also important to maximize efficacy because the vasculature acts as a “heat sink,” causing dissipation of the thermal energy. Occasionally, stenting of the proximal ureter due to upper tumor location is necessary to prevent thermal injury that could lead to urine leaks.

Selection of the modality to be used primarily depends on operator comfort, which translates to good patient outcomes, such as better cancer control and fewer complications. Cryoablation and RFA have both demonstrated good clinical efficacy and cancer control of 89% and 90%, respectively, with comparable complication rates.²² There have been no studies performed directly comparing the modalities.

Cryoablation

Cryoablation is performed through the insertion of a probe into the tumor, which may be done through a surgical or percutaneous approach. Once the probe is in place, a high- pressure gas (argon, nitrogen) is passed through the probe and upon entering a low pressure region the gas cools. The gas is able to cool to temperatures as low as –185°C. The tissue is then rewarmed through the use of helium, which conversely warms when entering a low pressure area. The process of freezing followed by rewarming subsequently causes cell death/tissue destruction through direct cell injury from cellular dehydration and vascular injury. Clinically, 2 freeze-thaw cycles are used to treat a tumor.^23,24

Radiofrequency ablation, or RFA, targets tumors via an electrode placed within the mass that produces intense frictional heat from medium-frequency alternating current (approximately 500 kHz) produced by a connected generator that is grounded on the patient. The thermal energy created causes coagulative necrosis. Due to the reliance on heat for tumor destruction, central lesions are less amenable to this approach because of the “heat sink” effect from the hilum.²⁴

Microwave Ablation

Microwave ablation, like RFA, relies on the generation of frictional heat to cause cell death by coagulative necrosis. In this case, the friction is created through the activation of water molecules; because of the different thermal kinetics involved with microwave ablation, the “heat sink” effect is minimized when treatment is employed near large vessels, in comparison to RFA.²⁴ The data on this mechanism of ablation are still maturing, with varied outcomes thus far. One study demonstrated outcomes comparable to RFA and cryoablation, with cancer-specific survival of 97.8% at 3 years.²⁵ However, a study by Castle and colleagues²⁶ demonstrated higher recurrence rates. The overarching impediment to widespread adoption of microwave ablation is inconclusive data gleaned from studies with small numbers of patients with limited follow up. The role of this modality will need to be revisited.

Irreversible Electroporation

Irreversible electroporation (IRE) is under investigation. IRE is a non-thermal ablative technique that employs rapid electrical pulses to create pores in cell membranes, leading to cell death. The postulated benefits of IRE include the lack of an effect from “heat sinks” and less collateral damage to the surrounding tissues, when compared with the thermal modalities. In a human phase 1 study of patients undergoing IRE prior to immediate surgical resection, the procedure appeared feasible and safe.²⁷ Significant concerns for this method of ablation possibly inducing cardiac arrhythmias, and the resultant need for sedation with neuromuscular blockade and associated electrocardiography monitoring, may impede its implementation in nonresearch settings.²⁴

ACTIVE SURVEILLANCE

Due to the more frequent use of imaging for various indications, there has been an increase in the discovery of small renal masses (SRM); 85% of RCC that present in an asymptomatic or incidental manner are tumors under 4 cm in diameter.^28,29 The role of active surveillance is evolving, but is primarily suggested for patients who are not candidates for more aggressive intervention based on comorbidities. A recent prospective, nonrandomized analysis of data from the Delayed Intervention and Surveillance for Small Renal Masses (DISSRM) registry evaluated outcomes for patients with SRM looking at primary intervention compared with active surveillance.³⁰ The primary intervention selected was at the discretion of the provider; treatments included partial nephrectomy, RFA, and cryoablation, and active surveillance patients were followed with imaging every 6 months. Progression of SRM, with recommendation for delayed intervention, was defined as a growth rate of mass greater than 0.5 cm/year, size greater than 4 cm, or hematuria. Thirty-six of 158 patients on active surveillance met criteria for progression; 21 underwent delayed intervention. Of note, even the patients who progressed but did not undergo delayed intervention did not develop metastatic disease during the follow-up interval. With a median follow-up of 2 years, cancer-specific survival was noted to be 99% and 100% at 5 years for primary intervention and active surveillance, respectively. Overall survival at 2 years for primary intervention was 98% and 96% for active surveillance; at 5 years, the survival rates were 92% and 75% (P = 0.06). Of note, 2 patients in the primary intervention arm died of RCC, while none in the active surveillance arm died. As would be expected, active surveillance patients were older, had a worse performance status, and had more comorbidities. Interestingly, 40% of patients enrolled selected active surveillance as their preferred management for SRM. The DISSRM results were consistent with data from the Renal Cell Consortium of Canada and other retrospective reviews.^31–33

• What is the approach to follow-up after treatment of localized RCC?

After a patient undergoes treatment for a localized RCC, the goal is to optimize oncologic outcomes, monitor for treatment sequelae, such as renal failure, and focus on survivorship. At this time, there is no consensus in the literature or across published national and international guidelines with regards to the appropriate schedule for surveillance to achieve these goals. In principle, the greatest risk for recurrence occurs within the first 3 years, so many guidelines focus on this timeframe. Likewise, the route of spread tends to be hematogenous, so patients present with pulmonary, bone, and brain metastases, in addition to local recurrence within the renal bed. Symptomatic recurrences often are seen

METASTATIC DISEASE

CASE CONTINUED

CT scan with contrast of the chest, abdomen, and pelvis as well as bone scan are done. CT of the abdomen and pelvis demonstrates a 7.8-cm left renal mass arising from the lower pole of the left kidney. Paraesophageal lymphadenopathy and mesenteric nodules are also noted. CT of the chest demonstrates bilateral pulmonary emboli. Bone scan is significant for increased activity related to the pathological fracture involving the right humerus. The patient undergoes surgery to stabilize the pathologic fracture of his humerus. He is diagnosed with metastatic RCC (clear cell histology) and undergoes palliative debulking nephrectomy.

• How is prognosis defined for metastatic RCC?

PROGNOSTIC MODELS

Limited-stage RCC that is found early can be cured surgically, with estimated 5-year survival rates for stage T1 and T2 disease approaching 90%; however, long-term survival for metastatic disease is poor, with rates ranging from 0% to 20%.¹³ Approximately 30% of patients have metastatic disease at diagnosis, and about one-third of patients who have undergone treatment for localized disease experience relapse.^36,37 Common sites of metastases include lung, lymph nodes, bone, liver, adrenal gland, and brain.

Prognostic scoring systems have been developed to define risk groups and assist with determining appropriate therapy in the metastatic setting. The most widely used validated prognostic factor model is that from the Memorial Sloan-Kettering Cancer Center (MSKCC), which was developed using a multivariate analysis derived from data of patients enrolled in clinical trials and treated with interferon alfa.³⁸ The factors included in the MSKCC model are Karnofsky performance status less than 80, time from diagnosis to treatment with interferon alfa less than 12 months, hemoglobin level less than lower limit of laboratory’s reference range, LDH level greater than 1.5 times the upper limit of laboratory’s reference range, and corrected serum calcium level greater than 10 mg/dL. Risk groups are categorized as favorable (0 risk factors), intermediate (1 to 2 risk factors), and poor (3 or more risk factors).³⁹ Median survival for favorable-, intermediate-, and poor-risk patients was 20, 10, and 4 months, respectively.⁴⁰

Another prognostic model, the International Metastatic RCC Database Consortium, or Heng, model was developed to evaluate prognosis in patients treated with VEGF-targeted therapy.⁴¹ This model was developed from a retrospective study of patients treated with sunitinib, sorafenib, and bevacizumab plus interferon alfa or prior immunotherapy. Prognostic factors in this model include 4 of the 5 MSKCC risk factors (hemoglobin level, corrected serum calcium level, Karnofsky performance status, and time to initial diagnosis). Additionally, this model includes both absolute neutrophil and platelet counts greater than the upper limit of normal. Risk groups are identified as favorable (0 risk factors), intermediate (1 to 2 risk factors), and poor (3 or more risk factors). Median survival for favorable-, intermediate-, and poor-risk patients was not reached, 27 months, and 8.8 months, respectively. The University of California, Los Angeles scoring algorithm to predict survival after nephrectomy and immunotherapy (SANI) in patients with metastatic RCC is another prognostic model that can be used. This simplified scoring system incorporates lymph node status, constitutional symptoms, metastases location, histology, and thyroid stimulating hormone (TSH) level.⁴²

The role of debulking or cytoreductive nephrectomy in treatment of metastatic RCC is well established. Large randomized studies have demonstrated a statistically significant median survival benefit for patients undergoing nephrectomy plus interferon alfa therapy compared with patients treated with interferon alfa alone (13.6 months versus 7.8 months, respectively).⁴³ The role of cytoreductive nephrectomy in combination with antiangiogenic agents is less clear. While a retrospective study investigating outcomes of patients with metastatic RCC receiving anti-VEGF agents showed a prolonged survival with nephrectomy, results of large randomized trials are not yet available.^44,45 Patients with lung-only metastases, good prognostic features, and a good performance status are historically the most likely to benefit from cytoreductive surgery.

CASE CONTINUED

Based on the MSKCC prognostic factor model, the patient is considered to be in the intermediate-risk group (Karnofsky performance status of 80, calcium 9.5 mg/dL, LDH 204 U/L, hemoglobin 13.6 g/dL). He is started on treatment for his bilateral pulmonary emboli and recovers well from orthopedic surgery as well as palliative debulking nephrectomy.

Several approaches to systemic therapy for advanced RCC have been taken based on the histologic type of the tumor. Clear-cell is by far the predominant histologic type in RCC. Several options are available as first-line treatment for patients with metastatic clear-cell RCC (Table 2).^46–54 These include biologic agents such as high-dose interleukin-2 (IL-2) immune therapy, as well as targeted therapies including TKIs and anti-VEGF antibodies. The mammalian target of rapamycin (mTOR) inhibitor temsirolimus is recommended as first-line therapy in patients with poor prognosis only. Second-line therapies for clear-cell RCC following antiangiogenic therapy include TKIs, mTOR inhibitors, nivolumab (PD-1 inhibitor), and the combination of the TKI lenvatinib and mTOR inhibitor everolimus.⁵⁵ In addition, after initial cytokine therapy, TKIs, temsirolimus, and the anti-VEGF antibody bevacizumab are other treatment options available to patients. Best supportive care should always be provided along with initial and subsequent therapies. Clinical trials are also an appropriate choice as first-line or subsequent therapies. All of these therapies require periodic monitoring to prevent and quickly treat adverse effects. Table 3 lists recommended monitoring parameters for each of these agents.⁵⁶

Based on several studies, TKIs seem to be less effective in patients with non–clear-cell type histology.^57,58 In these patients, risk factors can guide therapy. In the ASPEN trial, where 108 patients were randomly assigned to everolimus or sunitinib, patients in the good- and intermediate-risk groups had longer overall and median progression-free survival (PFS) on sunitinib (8.3 months versus 5.3 months, respectively). However, those in the poor-risk group had a longer median overall survival with everolimus.⁵⁹ Given that the role of targeted therapies in non–clear-cell RCCs is less well established, enrollment in clinical trials should be considered as a first-line treatment option.²¹

Sarcomatoid features can be observed in any of the histologic types of RCC, and RCC with these features has an aggressive course and a poor prognosis. Currently, there is no standard therapy for treatment of patients with metastatic or unresectable RCC with sarcomatoid features.⁶⁰ Chemotherapeutic regimens used for soft tissue sarcomas, including a trial of ifosfamide and doxorubicin, did not show any objective response.⁶¹ A small trial of 10 patients treated with doxorubicin and gemcitabine resulted in complete response in 2 patients and partial response in 1 patient.⁶²

Enrollment in a clinical trial remains a first-line treatment option for these patients. More recently, a phase 2 trial of sunitinib and gemcitabine in patients with sarcomatoid (39 patients) and/or poor-risk (33 patients) metastatic RCC showed overall response rates (ORR) of 26% and 24%, respectively. A higher clinical benefit rate (defined as ORR plus stable disease) was seen in patients with tumors containing more than 10% sarcomatoid histology, as compared with patients whose tumors contained less than 10% sarcomatoid histology. Neutropenia (n = 20), anemia (n = 10), and fatigue (n = 7) were the most common grade 3 toxicities seen in all the patients. Although this was a small study, the results showed a trend towards better efficacy of the combination therapy as compared with the single-agent regimen. Currently, another study is underway to further investigate this in a larger group of patients.⁶³

BIOLOGICS

Cytokine therapy, including high-dose IL-2 and interferon alfa, had long been the only first-line treatment option for patients with metastatic or unresectable RCC. Studies of high-dose IL-2 have shown an ORR of 25% and durable response in up to 11% of patients with clear-cell histology.⁶⁴ Toxicities were similar to those previously observed with high-dose IL-2 treatment; the most commonly observed grade 3 toxicities were hypotension and capillary leak syndrome. IL-2 requires strict monitoring (Table 3). It is important to note that retrospective studies evaluating the safety and efficacy of using IL-2 as second-line treatment in patients previously treated with TKIs demonstrated significant toxicity without achieving partial or complete response in any of the patients.⁶⁵

Prior to the advent of TKIs in the treatment of RCC, interferon alfa was a first-line treatment option for those who could not receive high-dose IL-2. It has been shown to produce response rates of approximately 20%, with maximum response seen with a higher dose range of 5 to 20 million units daily in 1 study.^66,67 However, with the introduction of TKIs, which produce a higher and more durable response, interferon alfa alone is no longer recommended as a treatment option.

Bevacizumab is a recombinant humanized monoclonal antibody that binds and neutralizes VEGF-A. Given overexpression of VEGF in RCC, the role of bevacizumab both as a single agent and in combination with interferon alfa has been investigated. In a randomized phase 2 study involving patients with cytokine-refractory disease, bevacizumab produced a 10% response rate and PFS of 4.8 months compared to patients treated with placebo.⁶⁸ In the AVOREN trial, the addition of bevacizumab (10 mg/kg intravenously [IV] every 2 weeks) to interferon alfa (9 million units subcutaneously [SC] 3 times weekly) was shown to significantly increase PFS compared with interferon alfa alone (10.2 months versus 5.4 months; P = 0.0001).^47,48 Adverse effects of this combination therapy include fatigue and asthenia. Additionally, hypertension, proteinuria, and bleeding occurred.

TYROSINE KINASE INHIBITORS

TKIs have largely replaced IL-2 as first-line therapy for metastatic RCC. Axitinib, pazopanib, sorafenib, and sunitinib and can be used as first-line therapy. All of the TKIs can be used as subsequent therapy.

Sunitinib

Sunitinib is an orally administered TKI that inhibits VEGF receptor (VEGFR) types 1 and 2, PDGF receptors (PDGFR) α and β, stem cell factor receptor (c-Kit), and FLT-3 and RET kinases. Motzer and colleagues^52,53 compared sunitinib 50 mg daily orally for 4 weeks with 2 weeks off to the then standard of care, interferon alfa 9 million units SC 3 times weekly. Sunitinib significantly increased the overall objective response rate (47% versus 12%; P < 0.001), PFS (11 versus 5 months; P < 0.001), and overall survival (26.4 versus 21.8 months; hazard ratio [HR], 0.821). The most common side effects are diarrhea, fatigue, nausea/vomiting, anorexia, hypertension, stomatitis, and hand-foot syndrome, occurring in more than 30% of patients. Often patients will require dose reductions or temporary discontinuations to tolerate therapy. Alternative dosing strategies (eg, 50 mg dose orally daily for 2 weeks alternating with 1-week free interval) have been attempted but not prospectively evaluated for efficacy.^69–71

Pazopanib

Pazopanib is an oral multi-kinase inhibitor of VEGFR types 1 and 2, PDGFR, and c-KIT. Results of a phase 3 trial comparing pazopanib (800 mg orally daily) to placebo favored the TKI, with a PFS of 9.2 months versus 4.2 months. A subset of treatment-naïve patients had a longer PFS of 11.1 versus 2.8 months and a response rate of 32% versus 4%.⁷² This led to a noninferiority phase 3 trial comparing pazopanib with sunitinib as first-line therapy.⁵⁰ In this study, PFS was similar (8.4 versus 9.5 months; HR 1.05), and overall safety and quality-of-life endpoints favored pazopanib. Much less fatigue, stomatitis, hand-foot syndrome, and thrombocytopenia occurred with pazopanib, whereas hair color changes, weight loss, alopecia, and elevations of LFT enzymes occurred more frequently with pazopanib. Hypertension is common with the administration of pazopanib as well.

Sorafenib

Sorafenib is an orally administered inhibitor of Raf, serine/threonine kinase, VEGFR, PDGFR, FLT-3, c-Kit, and RET. The pivotal phase 3 Treatment Approaches in Renal Cancer Global Evaluation Trial (TARGET) compared sorafenib (400 mg orally twice daily) with placebo in patients who had progressed on prior cytokine-based therapy.⁷³ A final analysis, which excluded patients who were allowed to cross over therapies, found improved overall survival times (14.3 versus 1.8 months, P = 0.029).⁵¹ Sorafenib is associated with lower rates of diarrhea, rash, fatigue, hand-foot syndrome, alopecia, hypertension, and nausea than sunitinib, although these agents have not been compared to one another.

Axitinib

Axitinib is an oral inhibitor of VEGFRs 1, 2, and 3. Results of the phase 3 AXIS trial comparing axitinib (5 mg orally twice daily) with sorafenib (400 mg orally twice daily) in patients receiving 1 prior systemic therapy showed axitinib was more active than sorafenib in improving ORR (19% versus 9%; P = 0.001) and PFS (6.7 versus 4.7 months; P < 0.001), although no difference in overall survival times was noted.⁷⁴ In a subsequent phase 3 trial comparing these drugs in the first-line setting, axitinib showed a nonsignificantly higher response rate and PFS. Despite this, the National Comprehensive Cancer Network guidelines consider axitinib an acceptable first-line therapy because activity with acceptable toxicity was demonstrated (Table 2).⁴⁶ The most common adverse effects of axitinib are diarrhea, hypertension, fatigue, decreased appetite, dysphonia, hypothyroidism, and upper abdominal pain.

CABOZANTINIB

Given that resistance eventually develops in most patients treated with standard treatments, including bevacizumab and TKIs, the need to evaluate the safety and efficacy of novel agents targeting VEGFR and overcoming this resistance is of vital importance. Cabozantinib is an oral small-molecule inhibitor of VEGFR, Met, and Axl, all tyrosine kinases implicated in metastatic RCC. Overexpression of Met and Axl, which occurs as a result of inactivation of the VHL gene, is associated with a poor prognosis in patients with RCC. In a

MTOR INHIBITORS

The mTOR inhibitors, temsirolimus and everolimus, are also approved for the treatment of metastatic or advanced RCC. These drugs block mTOR’s phosphorylation and subsequent translation of mRNA to inhibit cell proliferation, cell growth, and angiogenesis.⁷⁷ Temsirolimus can be used as first-line therapy for patients with a poor prognosis, and everolimus is appropriate as a subsequent therapy.

Temsirolimus is an intravenous prodrug of rapamycin. It was the first of the class to be approved for metastatic RCC for treatment-naïve patients with a poor prognosis (ie, at least 3 of 6 predictors of poor survival based on MSKCC model).⁵⁴ The pivotal ARCC trial compared temsirolimus (25 mg IV weekly) alone, interferon alfa (3 million units SC 3 times weekly) alone, or the combination (temsirolimus 15 mg IV weekly plus interferon alfa 6 million units SC 3 times weekly). In this trial, temsirolimus monotherapy produced a significantly longer overall survival time than interferon alfa alone (10.9 versus 7.3 months; P = 0.008) and improved PFS time when administered alone or in combination with interferon alfa (3.8 and 3.7 months, respectively, versus 1.9 months). Because no real efficacy advantage of the combination was demonstrated, temsirolimus is administered alone. The most common adverse effects of temsirolimus are asthenia, rash, anemia, nausea, anorexia, pain, and dyspnea. Additionally, hyperglycemia, hyper-cholesterolemia, and hyperlipidemia occur with these agents. Noninfectious pneumonitis is a rare but often fatal complication.

Everolimus is also an orally administered derivative of rapamycin that is approved for use after failure of VEGF-targeted therapies. The results of the landmark trial RECORD-1 demonstrated that everolimus (10 mg orally daily) is effective at prolonging PFS (4 versus 1.9 months; P < 0.001) when compared with best supportive care, a viable treatment option at the time of approval.⁷⁸ The most common adverse effects of everolimus are stomatitis, rash, fatigue, asthenia, and diarrhea. As with temsirolimus, elevations in glucose, lipids, and triglycerides and noninfectious pneumonitis can occur.

TKI + MTOR INHIBITOR

Lenvatinib is also a small molecule targeting multiple tyrosine kinases, primarily VEGF2. Combined with the mTOR inhibitor everolimus, it has been shown to be an effective regimen in patients with metastatic RCC who have failed other therapies. In a randomized phase 2 study involving patients with advanced or metastatic clear-cell RCC, patients were randomly assigned to receive either lenvatinib (24 mg/day), everolimus (10 mg/day), or lenvatinib plus everolimus (18 mg/day and 5 mg/day, respectively). Patients received the treatment continuously on a 28-day cycle until progression or inability to tolerate toxicity. Patients in the lenvatinib plus everolimus arm had median PFS of 14.6 months (95% CI 5.9 to 20.1) versus 5.5 months (95% CI 3.5 to 7.1) with everlolimus alone (HR 0.40 [95% CI 0.24 to 0.68]; P = 0.0005). PFS with levantinib alone was 7.4 months (95% CI 5.6 to 10.20; HR 0.66 [95% CI 0.30 to 1.10]; P = 0.12). In addition, PFS with levantinib alone was significantly prolonged in comparison with everolimus alone (HR 0.61 [95% CI 0.38 to 0.98]; P = 0.048). Grade 3 or 4 toxicity were less frequent in the everolimus only arm and the most common grade 3 or 4 toxicity in the lenvatinib plus everolimus arm was diarrhea. The results of this study show that the combination of lenvatinib plus everolimus is an acceptable second-line option for treatment of patients with advanced or metastatic RCC.⁵⁵

The patient is initially started on pazopanib and tolerates the medication well, with partial response to the treatment. However, on restaging scans he is noted to have small bowel perforation. Pazopanib is discontinued until the patient has a full recovery. He is then started on everolimus. Restaging scans done 3 months after starting everolimus demonstrate disease progression.

• What is the appropriate next step in treatment?

PD1 BLOCKADE

Programmed death 1 (PD-1) protein is a T-cell inhibitory receptor with 2 ligands, PD-L1 and PD-L2. PD-L1 is expressed on many tumors. Blocking the interaction between PD-1 and PD-L1 by anti-PD-1 humanized antibodies potentiates a robust immune response and has been a breakthrough in the field of cancer immunotherapy.⁷⁹ Previous studies have demonstrated that overexpression of PD-L1 leads to worse outcomes and poor prognosis in patients with RCC.⁸⁰ Nivolumab, a fully human IgG4 PD-1 immune checkpoint inhibitor, blocks the interaction between PD-1 and its ligands, PD-L1 and PD-L2. In a randomized, open-label, phase 3 study comparing nivolumab with everolimus in patients with RCC who had previously undergone treatment with other standard therapies, Motzer and colleagues⁸¹ demonstrated a longer overall survival time and fewer adverse effects with nivolumab. In this study, 821 patients with clear-cell RCC were randomly assigned to receive nivolumab (3 mg/kg of body weight IV every 2 weeks) or everolimus (10 mg orally once daily). The median overall survival time with nivolumab was 25 months versus 19.6 months with everolimus (P < 0.0148). Nineteen percent of patients receiving nivolumab experienced grade 3 or 4 toxicities, with fatigue being the most common adverse effect. Grade 3 or 4 toxicities were observed in 37% of patients treated with everolimus, with anemia being the most common. Based on the results of this trial, on November 23, 2015, the U.S. Food and Drug Administration approved nivolumab to treat patients with metastatic RCC who have received a prior antiangiogenic therapy.

CASE CONCLUSION

Both TKI and mTOR inhibitor therapy fail, and the patient is eligible for third-line therapy. Because of his previous GI perforation, other TKIs are not an option. The patient opts for enrollment in hospice due to declining performance status. For other patients in this situation with a good performance status, nivolumab would be a reasonable option.

FUTURE DIRECTIONS

With the approval of nivolumab, multiple treatment options are now available for patients with metastatic or unresectable RCC. Development of other PD-1 inhibitors and immunotherapies as well as multi-targeted TKIs will only serve to expand treatment options for these patients. Given the aggressive course and poor prognosis of non-clear cell renal cell tumors and those with sarcomatoid features, evaluation of systemic and targeted therapies for these subtypes should remain active areas of research and investigation.

INTRODUCTION

Renal cell carcinoma (RCC) is the most common malignancy arising in the kidney, comprising 90% of all renal tumors.¹ Approximately 55,000 new RCC cases are diagnosed each year.² Patients with RCC are often asymptomatic, and most cases are discovered as incidental findings on abdominal imaging performed during evaluation of nonrenal complaints. Limited-stage RCC that is found early can be cured surgically, with estimated 5-year survival rates approaching 90%; however, long-term survival for metastatic disease is poor, with rates ranging from 0% to 20%.² Advanced RCC is resistant to conventional chemotherapy and radiotherapy, and outcomes for patients with metastatic or unresectable RCC remain poor. However, the recent development of new therapeutic modalities that target tumor molecular pathways has expanded the treatment options for these patients and changed the management of RCC.

EPIDEMIOLOGY AND CLASSIFICATION

Median age at diagnosis in the United States is 64 years. Men have a higher incidence of RCC than women, with the highest incidence seen in American Indian and Alaska Native men (30.1 per 100,000 population). Genetic syndromes account for 2% to 4% of all RCCs.² Risk factors for RCC include smoking, hypertension, obesity, and acquired cystic kidney disease that is associated with end-stage renal failure.³ Longer duration of tobacco use is associated with a more aggressive course.

The 2004 World Health Organization (WHO) classification of renal tumors summarizes the previous classification systems (including the Heidelberg and Mainz classification systems) to describe different categories of RCC based on histologic and molecular genetics characteristics.² Using the WHO classification criteria, RCC comprises 90% of all renal tumors, with clear cell being the most common type (80%).² Other types of renal tumors include papillary, chromophobe, oncocytoma, and collecting-duct or Bellini duct tumors. Approximately 3% to 5% of tumors are unclassified. Oncocytomas are generally considered benign, and chromophobe tumors typically have an indolent course and rarely metastasize. Sarcomatoid differentiation can be seen in any histologic type and is associated with a worse prognosis. While different types of tumors may be seen in the kidney (such as transitional cell or lymphomas), the focus of this review is the primary malignancies of the renal parenchyma.

FAMILIAL SYNDROMES

Several genetic syndromes have been identified by studying families with inherited RCC. Among these, von Hippel-Lindau (VHL) gene mutation is the most commonly found inherited genetic defect. Table 1 summarizes the incidence of gene mutations and the corresponding histologic appearance of the most common sporadic and hereditary RCCs.⁴

VHL disease is an autosomal dominant familial syndrome. Patients with this mutation are at higher risk for developing RCC (clear cell histology), retinal angiomas, pheochromocytomas, as well as hemangioblastomas of the central nervous system (CNS).⁴ Of all the genetic mutations seen in RCC, the somatic mutation in the VHL tumor-suppressor gene is by far the most common.⁵ VHL targets hypoxia–inducible factor-1 alpha (HIF-α) for ubiquitination and subsequent degradation, which has been shown to suppress the growth of clear-cell RCC in mouse models.^6–8 HIF expression under hypoxic conditions leads to activation of a number of genes important in blood vessel development, cell proliferation, and glucose metabolism, including vascular endothelial growth factor (VEGF), erythropoietin, platelet-derived growth factor beta (PDGF-β), transforming growth factor alpha (TGF-α), and glucose transporter-1 (GLUT-1). Mutation in the VHL gene prevents degradation of the HIF-α protein, thereby leading to increased expression of these downstream proteins, including MET and Axl. The upregulation of these angiogenic factors is thought to be the underlying process for increased vascularity of CNS hemangioblastomas and clear-cell renal tumors in VHL disease.^4–8

Other less common genetic syndromes seen in hereditary RCC include hereditary papillary RCC, hereditary leiomyomatosis, and Birt-Hogg-Dubé (BHD) syndrome.⁹ In hereditary papillary RCC, the MET gene is mutated. BHD syndrome is a rare, autosomal dominant syndrome characterized by hair follicle hamartomas of the face and neck. About 15% of patients have multiple renal tumors, the majority of which are of the chromophobe or mixed chromophobe-oncocytoma histology. The BHD gene encodes the protein folliculin, which is thought to be a tumor-suppressor gene.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 74-year-old man who works as an airplane mechanic repairman presents to the emergency department with sudden worsening of chronic right upper arm and shoulder pain after lifting a jug of orange juice. He does not have a significant past medical history and initially thought that his pain was due to a work-related injury. Upon initial evaluation in the emergency department he is found to have a fracture of his right humerus. Given that the fracture appears to be pathologic, further work-up is recommended.

Most patients are asymptomatic until the disease becomes advanced. The classic triad of flank pain, hematuria, and palpable abdominal mass is seen in approximately 10% of patients with RCC, partly because of earlier detection of renal masses by imaging performed for other purposes.¹⁰ Less frequently, patients present with signs or symptoms of metastatic disease such as bone pain or fracture (as seen in the case patient), painful adenopathy, and pulmonary symptoms related to mediastinal masses. Fever, weight loss, anemia, and/or varicocele often occur in young patients (≤ 46 years) and may indicate the presence of a hereditary form of the disease. Patients may present with paraneoplastic syndromes seen as abnormalities on routine blood work. These can include polycythemia or elevated liver function tests (LFTs) without the presence of liver metastases (known as Stauffer syndrome), which can be seen in localized renal tumors. Nearly half (45%) of patients present with localized disease, 25% present with locally advanced disease, and 30% present with metastatic disease.¹¹ Bone is the second most common site of distant metastatic spread (following lung) in patients with advanced RCC.

• What is the approach to initial evaluation for a patient with suspected RCC?

Initial evaluation consists of a physical exam, laboratory tests including complete blood count (CBC) and comprehensive metabolic panel (calcium, serum creatinine, LFTs, lactate dehydrogenase [LDH], and urinalysis), and imaging. Imaging studies include computed tomography (CT) scan with contrast of the abdomen and pelvis or magnetic resonance imaging (MRI) of the abdomen and chest imaging. A chest radiograph may be obtained, although a chest CT is more sensitive for the presence of pulmonary metastases. MRI can be used in patients with renal dysfunction to evaluate the renal vein and inferior vena cava (IVC) for thrombus or to determine the presence of local invasion.¹² Although bone and brain are common sites for metastases, routine imaging is not indicated unless the patient is symptomatic. The value of positron emission tomography in RCC remains undetermined at this time.

Staging is done according to the American Joint Committee on Cancer (AJCC) staging classification for RCC; the Figure summarizes the staging and 5-year survival data based on this classification scheme.^4,13

LIMITED-STAGE DISEASE

• What are the therapeutic options for limited-stage disease?

For patients with nondistant metastases, or limited-stage disease, surgical intervention with curative intent is considered. Convention suggests considering definitive surgery for patients with stage I and II disease, select patients with stage III disease with pathologically enlarged retroperitoneal lymph nodes, patients with IVC and/or cardiac atrium involvement of tumor thrombus, and patients with direct extension of the renal tumor into the ipsilateral adrenal gland if there is no evidence of distant disease. While there may be a role for aggressive surgical intervention in patients with distant metastatic disease, this topic will not be covered in this review.

SURGICAL INTERVENTION

Once patients are determined to be appropriate candidates for surgical removal of a renal tumor, the urologist will perform either a radical nephrectomy or a nephron-sparing nephrectomy, also called a partial nephrectomy. The urologist will evaluate the patient based on his or her body habitus, the location of the tumor, whether multiple tumors in one kidney or bilateral tumors are present, whether the patient has a solitary kidney or otherwise impaired kidney function, and whether the patient has a history of a hereditary syndrome involving kidney cancer as this affects the risk of future kidney tumors.

A radical nephrectomy is surgically preferred in the presence of the following factors: tumor larger than 7 cm in diameter, a more centrally located tumor, suspicion of lymph node involvement, tumor involvement with renal vein or IVC, and/or direct extension of the tumor into the ipsilateral adrenal gland. Nephrectomy involves ligation of the vascular supply (renal artery and vein) followed by removal of the kidney and surrounding Gerota’s fascia. The ipsilateral adrenal gland is removed if there is a high-risk for or presence of invasion of the adrenal gland. Removal of the adrenal gland is not standard since the literature demonstrates there is less than a 10% chance of solitary, ipsilateral adrenal gland involvement of tumor at the time of nephrectomy in the absence of high-risk features, and a recent systematic review suggests that the chance may be as low as 1.8%.¹⁴ Preoperative factors that correlated with adrenal involvement included upper pole kidney location, renal vein thrombosis, higher T stage (T3a and greater), multifocal tumors, and evidence for distant metastases or lymph node involvement. Lymphadenectomy previously had been included in radical nephrectomy but now is performed selectively. Radical nephrectomy may be performed as

Conversely, a nephron-sparing approach is preferred for tumors less than 7 cm in diameter, for patients with a solitary kidney or impaired renal function, for patients with multiple small ipsilateral tumors or with bilateral tumors, or for radical nephrectomy candidates with comorbidities for whom a limited intervention is deemed to be a lower-risk procedure. A nephron-sparing procedure may also be performed open or laparoscopically. In nephron-sparing procedures, the tumor is removed along with a small margin of normal parenchyma.¹⁵

In summary, the goal of surgical intervention is curative intent with removal of the tumor while maintaining as much residual renal function as possible to limit long-term morbidity of chronic kidney disease and associated cardiovascular events.¹⁸ Oncologic outcomes for radical nephrectomy and partial nephrectomy are similar. In one study, overall survival was slightly lower in the partial nephrectomy cohort, but only a small number of the deaths were due to RCC.¹⁹

ADJUVANT THERAPY

Adjuvant systemic therapy currently has no role following nephrectomy for RCC because no systemic therapy has been able to reduce the likelihood of relapse. Randomized trials of cytokine therapy (eg, interferon, interleukin 2) or tyrosine kinase inhibitors (TKIs; eg, sorafenib, sunitinib) with observation alone in patients with locally advanced completely resected RCC have shown no delay in time to relapse or improvement of survival with adjuvant therapy.²⁰ Similarly, adjuvant radiation therapy has not shown benefit even in patients with nodal involvement or incomplete resection.²¹ Therefore, observation remains the standard of care after nephrectomy.

RENAL TUMOR ABLATION

For patients who are deemed not to be surgical candidates due to age, comorbidities, or patient preference and who have tumors less than 4 cm in size (stage I tumors), ablative techniques may be considered. The 2 most well-studied and effective techniques at present are cryoablation and radiofrequency ablation (RFA). Microwave ablation may be an option in some facilities, but the data in RCC are limited. An emerging ablative technique under investigation is irreversible electroporation. At present, the long-term efficacy of all ablative techniques is unknown.

Patient selection is undertaken by urologists and interventional radiologists who evaluate the patient with ultrasound, CT, and/or MRI to determine the location and size of the tumor and the presence or absence of metastatic disease. A pretreatment biopsy is recommended to document the histology of the lesion to confirm a malignancy and to guide future treatment for recurrent or metastatic disease. Contraindications to the procedure include the presence of metastatic disease, a life expectancy of less than 1 year, general medical instability, or uncorrectable coagulopathy due to increased risk of bleeding complications. Tumors in close proximity to the renal hilum or collecting system are a contraindication to the procedure because of the risk for hemorrhage or damage to the collecting system. The location of the tumor in relation to the vasculature is also important to maximize efficacy because the vasculature acts as a “heat sink,” causing dissipation of the thermal energy. Occasionally, stenting of the proximal ureter due to upper tumor location is necessary to prevent thermal injury that could lead to urine leaks.

Selection of the modality to be used primarily depends on operator comfort, which translates to good patient outcomes, such as better cancer control and fewer complications. Cryoablation and RFA have both demonstrated good clinical efficacy and cancer control of 89% and 90%, respectively, with comparable complication rates.²² There have been no studies performed directly comparing the modalities.

Cryoablation

Cryoablation is performed through the insertion of a probe into the tumor, which may be done through a surgical or percutaneous approach. Once the probe is in place, a high- pressure gas (argon, nitrogen) is passed through the probe and upon entering a low pressure region the gas cools. The gas is able to cool to temperatures as low as –185°C. The tissue is then rewarmed through the use of helium, which conversely warms when entering a low pressure area. The process of freezing followed by rewarming subsequently causes cell death/tissue destruction through direct cell injury from cellular dehydration and vascular injury. Clinically, 2 freeze-thaw cycles are used to treat a tumor.^23,24

Radiofrequency ablation, or RFA, targets tumors via an electrode placed within the mass that produces intense frictional heat from medium-frequency alternating current (approximately 500 kHz) produced by a connected generator that is grounded on the patient. The thermal energy created causes coagulative necrosis. Due to the reliance on heat for tumor destruction, central lesions are less amenable to this approach because of the “heat sink” effect from the hilum.²⁴

Microwave Ablation

Microwave ablation, like RFA, relies on the generation of frictional heat to cause cell death by coagulative necrosis. In this case, the friction is created through the activation of water molecules; because of the different thermal kinetics involved with microwave ablation, the “heat sink” effect is minimized when treatment is employed near large vessels, in comparison to RFA.²⁴ The data on this mechanism of ablation are still maturing, with varied outcomes thus far. One study demonstrated outcomes comparable to RFA and cryoablation, with cancer-specific survival of 97.8% at 3 years.²⁵ However, a study by Castle and colleagues²⁶ demonstrated higher recurrence rates. The overarching impediment to widespread adoption of microwave ablation is inconclusive data gleaned from studies with small numbers of patients with limited follow up. The role of this modality will need to be revisited.

Irreversible Electroporation

Irreversible electroporation (IRE) is under investigation. IRE is a non-thermal ablative technique that employs rapid electrical pulses to create pores in cell membranes, leading to cell death. The postulated benefits of IRE include the lack of an effect from “heat sinks” and less collateral damage to the surrounding tissues, when compared with the thermal modalities. In a human phase 1 study of patients undergoing IRE prior to immediate surgical resection, the procedure appeared feasible and safe.²⁷ Significant concerns for this method of ablation possibly inducing cardiac arrhythmias, and the resultant need for sedation with neuromuscular blockade and associated electrocardiography monitoring, may impede its implementation in nonresearch settings.²⁴

ACTIVE SURVEILLANCE

Due to the more frequent use of imaging for various indications, there has been an increase in the discovery of small renal masses (SRM); 85% of RCC that present in an asymptomatic or incidental manner are tumors under 4 cm in diameter.^28,29 The role of active surveillance is evolving, but is primarily suggested for patients who are not candidates for more aggressive intervention based on comorbidities. A recent prospective, nonrandomized analysis of data from the Delayed Intervention and Surveillance for Small Renal Masses (DISSRM) registry evaluated outcomes for patients with SRM looking at primary intervention compared with active surveillance.³⁰ The primary intervention selected was at the discretion of the provider; treatments included partial nephrectomy, RFA, and cryoablation, and active surveillance patients were followed with imaging every 6 months. Progression of SRM, with recommendation for delayed intervention, was defined as a growth rate of mass greater than 0.5 cm/year, size greater than 4 cm, or hematuria. Thirty-six of 158 patients on active surveillance met criteria for progression; 21 underwent delayed intervention. Of note, even the patients who progressed but did not undergo delayed intervention did not develop metastatic disease during the follow-up interval. With a median follow-up of 2 years, cancer-specific survival was noted to be 99% and 100% at 5 years for primary intervention and active surveillance, respectively. Overall survival at 2 years for primary intervention was 98% and 96% for active surveillance; at 5 years, the survival rates were 92% and 75% (P = 0.06). Of note, 2 patients in the primary intervention arm died of RCC, while none in the active surveillance arm died. As would be expected, active surveillance patients were older, had a worse performance status, and had more comorbidities. Interestingly, 40% of patients enrolled selected active surveillance as their preferred management for SRM. The DISSRM results were consistent with data from the Renal Cell Consortium of Canada and other retrospective reviews.^31–33

• What is the approach to follow-up after treatment of localized RCC?

After a patient undergoes treatment for a localized RCC, the goal is to optimize oncologic outcomes, monitor for treatment sequelae, such as renal failure, and focus on survivorship. At this time, there is no consensus in the literature or across published national and international guidelines with regards to the appropriate schedule for surveillance to achieve these goals. In principle, the greatest risk for recurrence occurs within the first 3 years, so many guidelines focus on this timeframe. Likewise, the route of spread tends to be hematogenous, so patients present with pulmonary, bone, and brain metastases, in addition to local recurrence within the renal bed. Symptomatic recurrences often are seen

METASTATIC DISEASE

CASE CONTINUED

CT scan with contrast of the chest, abdomen, and pelvis as well as bone scan are done. CT of the abdomen and pelvis demonstrates a 7.8-cm left renal mass arising from the lower pole of the left kidney. Paraesophageal lymphadenopathy and mesenteric nodules are also noted. CT of the chest demonstrates bilateral pulmonary emboli. Bone scan is significant for increased activity related to the pathological fracture involving the right humerus. The patient undergoes surgery to stabilize the pathologic fracture of his humerus. He is diagnosed with metastatic RCC (clear cell histology) and undergoes palliative debulking nephrectomy.

• How is prognosis defined for metastatic RCC?

PROGNOSTIC MODELS

Limited-stage RCC that is found early can be cured surgically, with estimated 5-year survival rates for stage T1 and T2 disease approaching 90%; however, long-term survival for metastatic disease is poor, with rates ranging from 0% to 20%.¹³ Approximately 30% of patients have metastatic disease at diagnosis, and about one-third of patients who have undergone treatment for localized disease experience relapse.^36,37 Common sites of metastases include lung, lymph nodes, bone, liver, adrenal gland, and brain.

Prognostic scoring systems have been developed to define risk groups and assist with determining appropriate therapy in the metastatic setting. The most widely used validated prognostic factor model is that from the Memorial Sloan-Kettering Cancer Center (MSKCC), which was developed using a multivariate analysis derived from data of patients enrolled in clinical trials and treated with interferon alfa.³⁸ The factors included in the MSKCC model are Karnofsky performance status less than 80, time from diagnosis to treatment with interferon alfa less than 12 months, hemoglobin level less than lower limit of laboratory’s reference range, LDH level greater than 1.5 times the upper limit of laboratory’s reference range, and corrected serum calcium level greater than 10 mg/dL. Risk groups are categorized as favorable (0 risk factors), intermediate (1 to 2 risk factors), and poor (3 or more risk factors).³⁹ Median survival for favorable-, intermediate-, and poor-risk patients was 20, 10, and 4 months, respectively.⁴⁰

Another prognostic model, the International Metastatic RCC Database Consortium, or Heng, model was developed to evaluate prognosis in patients treated with VEGF-targeted therapy.⁴¹ This model was developed from a retrospective study of patients treated with sunitinib, sorafenib, and bevacizumab plus interferon alfa or prior immunotherapy. Prognostic factors in this model include 4 of the 5 MSKCC risk factors (hemoglobin level, corrected serum calcium level, Karnofsky performance status, and time to initial diagnosis). Additionally, this model includes both absolute neutrophil and platelet counts greater than the upper limit of normal. Risk groups are identified as favorable (0 risk factors), intermediate (1 to 2 risk factors), and poor (3 or more risk factors). Median survival for favorable-, intermediate-, and poor-risk patients was not reached, 27 months, and 8.8 months, respectively. The University of California, Los Angeles scoring algorithm to predict survival after nephrectomy and immunotherapy (SANI) in patients with metastatic RCC is another prognostic model that can be used. This simplified scoring system incorporates lymph node status, constitutional symptoms, metastases location, histology, and thyroid stimulating hormone (TSH) level.⁴²

The role of debulking or cytoreductive nephrectomy in treatment of metastatic RCC is well established. Large randomized studies have demonstrated a statistically significant median survival benefit for patients undergoing nephrectomy plus interferon alfa therapy compared with patients treated with interferon alfa alone (13.6 months versus 7.8 months, respectively).⁴³ The role of cytoreductive nephrectomy in combination with antiangiogenic agents is less clear. While a retrospective study investigating outcomes of patients with metastatic RCC receiving anti-VEGF agents showed a prolonged survival with nephrectomy, results of large randomized trials are not yet available.^44,45 Patients with lung-only metastases, good prognostic features, and a good performance status are historically the most likely to benefit from cytoreductive surgery.

CASE CONTINUED

Based on the MSKCC prognostic factor model, the patient is considered to be in the intermediate-risk group (Karnofsky performance status of 80, calcium 9.5 mg/dL, LDH 204 U/L, hemoglobin 13.6 g/dL). He is started on treatment for his bilateral pulmonary emboli and recovers well from orthopedic surgery as well as palliative debulking nephrectomy.

Several approaches to systemic therapy for advanced RCC have been taken based on the histologic type of the tumor. Clear-cell is by far the predominant histologic type in RCC. Several options are available as first-line treatment for patients with metastatic clear-cell RCC (Table 2).^46–54 These include biologic agents such as high-dose interleukin-2 (IL-2) immune therapy, as well as targeted therapies including TKIs and anti-VEGF antibodies. The mammalian target of rapamycin (mTOR) inhibitor temsirolimus is recommended as first-line therapy in patients with poor prognosis only. Second-line therapies for clear-cell RCC following antiangiogenic therapy include TKIs, mTOR inhibitors, nivolumab (PD-1 inhibitor), and the combination of the TKI lenvatinib and mTOR inhibitor everolimus.⁵⁵ In addition, after initial cytokine therapy, TKIs, temsirolimus, and the anti-VEGF antibody bevacizumab are other treatment options available to patients. Best supportive care should always be provided along with initial and subsequent therapies. Clinical trials are also an appropriate choice as first-line or subsequent therapies. All of these therapies require periodic monitoring to prevent and quickly treat adverse effects. Table 3 lists recommended monitoring parameters for each of these agents.⁵⁶

Based on several studies, TKIs seem to be less effective in patients with non–clear-cell type histology.^57,58 In these patients, risk factors can guide therapy. In the ASPEN trial, where 108 patients were randomly assigned to everolimus or sunitinib, patients in the good- and intermediate-risk groups had longer overall and median progression-free survival (PFS) on sunitinib (8.3 months versus 5.3 months, respectively). However, those in the poor-risk group had a longer median overall survival with everolimus.⁵⁹ Given that the role of targeted therapies in non–clear-cell RCCs is less well established, enrollment in clinical trials should be considered as a first-line treatment option.²¹

Sarcomatoid features can be observed in any of the histologic types of RCC, and RCC with these features has an aggressive course and a poor prognosis. Currently, there is no standard therapy for treatment of patients with metastatic or unresectable RCC with sarcomatoid features.⁶⁰ Chemotherapeutic regimens used for soft tissue sarcomas, including a trial of ifosfamide and doxorubicin, did not show any objective response.⁶¹ A small trial of 10 patients treated with doxorubicin and gemcitabine resulted in complete response in 2 patients and partial response in 1 patient.⁶²

Enrollment in a clinical trial remains a first-line treatment option for these patients. More recently, a phase 2 trial of sunitinib and gemcitabine in patients with sarcomatoid (39 patients) and/or poor-risk (33 patients) metastatic RCC showed overall response rates (ORR) of 26% and 24%, respectively. A higher clinical benefit rate (defined as ORR plus stable disease) was seen in patients with tumors containing more than 10% sarcomatoid histology, as compared with patients whose tumors contained less than 10% sarcomatoid histology. Neutropenia (n = 20), anemia (n = 10), and fatigue (n = 7) were the most common grade 3 toxicities seen in all the patients. Although this was a small study, the results showed a trend towards better efficacy of the combination therapy as compared with the single-agent regimen. Currently, another study is underway to further investigate this in a larger group of patients.⁶³

BIOLOGICS

Cytokine therapy, including high-dose IL-2 and interferon alfa, had long been the only first-line treatment option for patients with metastatic or unresectable RCC. Studies of high-dose IL-2 have shown an ORR of 25% and durable response in up to 11% of patients with clear-cell histology.⁶⁴ Toxicities were similar to those previously observed with high-dose IL-2 treatment; the most commonly observed grade 3 toxicities were hypotension and capillary leak syndrome. IL-2 requires strict monitoring (Table 3). It is important to note that retrospective studies evaluating the safety and efficacy of using IL-2 as second-line treatment in patients previously treated with TKIs demonstrated significant toxicity without achieving partial or complete response in any of the patients.⁶⁵

Prior to the advent of TKIs in the treatment of RCC, interferon alfa was a first-line treatment option for those who could not receive high-dose IL-2. It has been shown to produce response rates of approximately 20%, with maximum response seen with a higher dose range of 5 to 20 million units daily in 1 study.^66,67 However, with the introduction of TKIs, which produce a higher and more durable response, interferon alfa alone is no longer recommended as a treatment option.

Bevacizumab is a recombinant humanized monoclonal antibody that binds and neutralizes VEGF-A. Given overexpression of VEGF in RCC, the role of bevacizumab both as a single agent and in combination with interferon alfa has been investigated. In a randomized phase 2 study involving patients with cytokine-refractory disease, bevacizumab produced a 10% response rate and PFS of 4.8 months compared to patients treated with placebo.⁶⁸ In the AVOREN trial, the addition of bevacizumab (10 mg/kg intravenously [IV] every 2 weeks) to interferon alfa (9 million units subcutaneously [SC] 3 times weekly) was shown to significantly increase PFS compared with interferon alfa alone (10.2 months versus 5.4 months; P = 0.0001).^47,48 Adverse effects of this combination therapy include fatigue and asthenia. Additionally, hypertension, proteinuria, and bleeding occurred.

TYROSINE KINASE INHIBITORS

TKIs have largely replaced IL-2 as first-line therapy for metastatic RCC. Axitinib, pazopanib, sorafenib, and sunitinib and can be used as first-line therapy. All of the TKIs can be used as subsequent therapy.

Sunitinib

Sunitinib is an orally administered TKI that inhibits VEGF receptor (VEGFR) types 1 and 2, PDGF receptors (PDGFR) α and β, stem cell factor receptor (c-Kit), and FLT-3 and RET kinases. Motzer and colleagues^52,53 compared sunitinib 50 mg daily orally for 4 weeks with 2 weeks off to the then standard of care, interferon alfa 9 million units SC 3 times weekly. Sunitinib significantly increased the overall objective response rate (47% versus 12%; P < 0.001), PFS (11 versus 5 months; P < 0.001), and overall survival (26.4 versus 21.8 months; hazard ratio [HR], 0.821). The most common side effects are diarrhea, fatigue, nausea/vomiting, anorexia, hypertension, stomatitis, and hand-foot syndrome, occurring in more than 30% of patients. Often patients will require dose reductions or temporary discontinuations to tolerate therapy. Alternative dosing strategies (eg, 50 mg dose orally daily for 2 weeks alternating with 1-week free interval) have been attempted but not prospectively evaluated for efficacy.^69–71

Pazopanib

Pazopanib is an oral multi-kinase inhibitor of VEGFR types 1 and 2, PDGFR, and c-KIT. Results of a phase 3 trial comparing pazopanib (800 mg orally daily) to placebo favored the TKI, with a PFS of 9.2 months versus 4.2 months. A subset of treatment-naïve patients had a longer PFS of 11.1 versus 2.8 months and a response rate of 32% versus 4%.⁷² This led to a noninferiority phase 3 trial comparing pazopanib with sunitinib as first-line therapy.⁵⁰ In this study, PFS was similar (8.4 versus 9.5 months; HR 1.05), and overall safety and quality-of-life endpoints favored pazopanib. Much less fatigue, stomatitis, hand-foot syndrome, and thrombocytopenia occurred with pazopanib, whereas hair color changes, weight loss, alopecia, and elevations of LFT enzymes occurred more frequently with pazopanib. Hypertension is common with the administration of pazopanib as well.

Sorafenib

Sorafenib is an orally administered inhibitor of Raf, serine/threonine kinase, VEGFR, PDGFR, FLT-3, c-Kit, and RET. The pivotal phase 3 Treatment Approaches in Renal Cancer Global Evaluation Trial (TARGET) compared sorafenib (400 mg orally twice daily) with placebo in patients who had progressed on prior cytokine-based therapy.⁷³ A final analysis, which excluded patients who were allowed to cross over therapies, found improved overall survival times (14.3 versus 1.8 months, P = 0.029).⁵¹ Sorafenib is associated with lower rates of diarrhea, rash, fatigue, hand-foot syndrome, alopecia, hypertension, and nausea than sunitinib, although these agents have not been compared to one another.

Axitinib

Axitinib is an oral inhibitor of VEGFRs 1, 2, and 3. Results of the phase 3 AXIS trial comparing axitinib (5 mg orally twice daily) with sorafenib (400 mg orally twice daily) in patients receiving 1 prior systemic therapy showed axitinib was more active than sorafenib in improving ORR (19% versus 9%; P = 0.001) and PFS (6.7 versus 4.7 months; P < 0.001), although no difference in overall survival times was noted.⁷⁴ In a subsequent phase 3 trial comparing these drugs in the first-line setting, axitinib showed a nonsignificantly higher response rate and PFS. Despite this, the National Comprehensive Cancer Network guidelines consider axitinib an acceptable first-line therapy because activity with acceptable toxicity was demonstrated (Table 2).⁴⁶ The most common adverse effects of axitinib are diarrhea, hypertension, fatigue, decreased appetite, dysphonia, hypothyroidism, and upper abdominal pain.

CABOZANTINIB

Given that resistance eventually develops in most patients treated with standard treatments, including bevacizumab and TKIs, the need to evaluate the safety and efficacy of novel agents targeting VEGFR and overcoming this resistance is of vital importance. Cabozantinib is an oral small-molecule inhibitor of VEGFR, Met, and Axl, all tyrosine kinases implicated in metastatic RCC. Overexpression of Met and Axl, which occurs as a result of inactivation of the VHL gene, is associated with a poor prognosis in patients with RCC. In a

MTOR INHIBITORS

The mTOR inhibitors, temsirolimus and everolimus, are also approved for the treatment of metastatic or advanced RCC. These drugs block mTOR’s phosphorylation and subsequent translation of mRNA to inhibit cell proliferation, cell growth, and angiogenesis.⁷⁷ Temsirolimus can be used as first-line therapy for patients with a poor prognosis, and everolimus is appropriate as a subsequent therapy.

Temsirolimus is an intravenous prodrug of rapamycin. It was the first of the class to be approved for metastatic RCC for treatment-naïve patients with a poor prognosis (ie, at least 3 of 6 predictors of poor survival based on MSKCC model).⁵⁴ The pivotal ARCC trial compared temsirolimus (25 mg IV weekly) alone, interferon alfa (3 million units SC 3 times weekly) alone, or the combination (temsirolimus 15 mg IV weekly plus interferon alfa 6 million units SC 3 times weekly). In this trial, temsirolimus monotherapy produced a significantly longer overall survival time than interferon alfa alone (10.9 versus 7.3 months; P = 0.008) and improved PFS time when administered alone or in combination with interferon alfa (3.8 and 3.7 months, respectively, versus 1.9 months). Because no real efficacy advantage of the combination was demonstrated, temsirolimus is administered alone. The most common adverse effects of temsirolimus are asthenia, rash, anemia, nausea, anorexia, pain, and dyspnea. Additionally, hyperglycemia, hyper-cholesterolemia, and hyperlipidemia occur with these agents. Noninfectious pneumonitis is a rare but often fatal complication.

Everolimus is also an orally administered derivative of rapamycin that is approved for use after failure of VEGF-targeted therapies. The results of the landmark trial RECORD-1 demonstrated that everolimus (10 mg orally daily) is effective at prolonging PFS (4 versus 1.9 months; P < 0.001) when compared with best supportive care, a viable treatment option at the time of approval.⁷⁸ The most common adverse effects of everolimus are stomatitis, rash, fatigue, asthenia, and diarrhea. As with temsirolimus, elevations in glucose, lipids, and triglycerides and noninfectious pneumonitis can occur.

TKI + MTOR INHIBITOR

Lenvatinib is also a small molecule targeting multiple tyrosine kinases, primarily VEGF2. Combined with the mTOR inhibitor everolimus, it has been shown to be an effective regimen in patients with metastatic RCC who have failed other therapies. In a randomized phase 2 study involving patients with advanced or metastatic clear-cell RCC, patients were randomly assigned to receive either lenvatinib (24 mg/day), everolimus (10 mg/day), or lenvatinib plus everolimus (18 mg/day and 5 mg/day, respectively). Patients received the treatment continuously on a 28-day cycle until progression or inability to tolerate toxicity. Patients in the lenvatinib plus everolimus arm had median PFS of 14.6 months (95% CI 5.9 to 20.1) versus 5.5 months (95% CI 3.5 to 7.1) with everlolimus alone (HR 0.40 [95% CI 0.24 to 0.68]; P = 0.0005). PFS with levantinib alone was 7.4 months (95% CI 5.6 to 10.20; HR 0.66 [95% CI 0.30 to 1.10]; P = 0.12). In addition, PFS with levantinib alone was significantly prolonged in comparison with everolimus alone (HR 0.61 [95% CI 0.38 to 0.98]; P = 0.048). Grade 3 or 4 toxicity were less frequent in the everolimus only arm and the most common grade 3 or 4 toxicity in the lenvatinib plus everolimus arm was diarrhea. The results of this study show that the combination of lenvatinib plus everolimus is an acceptable second-line option for treatment of patients with advanced or metastatic RCC.⁵⁵

The patient is initially started on pazopanib and tolerates the medication well, with partial response to the treatment. However, on restaging scans he is noted to have small bowel perforation. Pazopanib is discontinued until the patient has a full recovery. He is then started on everolimus. Restaging scans done 3 months after starting everolimus demonstrate disease progression.

• What is the appropriate next step in treatment?

PD1 BLOCKADE

Programmed death 1 (PD-1) protein is a T-cell inhibitory receptor with 2 ligands, PD-L1 and PD-L2. PD-L1 is expressed on many tumors. Blocking the interaction between PD-1 and PD-L1 by anti-PD-1 humanized antibodies potentiates a robust immune response and has been a breakthrough in the field of cancer immunotherapy.⁷⁹ Previous studies have demonstrated that overexpression of PD-L1 leads to worse outcomes and poor prognosis in patients with RCC.⁸⁰ Nivolumab, a fully human IgG4 PD-1 immune checkpoint inhibitor, blocks the interaction between PD-1 and its ligands, PD-L1 and PD-L2. In a randomized, open-label, phase 3 study comparing nivolumab with everolimus in patients with RCC who had previously undergone treatment with other standard therapies, Motzer and colleagues⁸¹ demonstrated a longer overall survival time and fewer adverse effects with nivolumab. In this study, 821 patients with clear-cell RCC were randomly assigned to receive nivolumab (3 mg/kg of body weight IV every 2 weeks) or everolimus (10 mg orally once daily). The median overall survival time with nivolumab was 25 months versus 19.6 months with everolimus (P < 0.0148). Nineteen percent of patients receiving nivolumab experienced grade 3 or 4 toxicities, with fatigue being the most common adverse effect. Grade 3 or 4 toxicities were observed in 37% of patients treated with everolimus, with anemia being the most common. Based on the results of this trial, on November 23, 2015, the U.S. Food and Drug Administration approved nivolumab to treat patients with metastatic RCC who have received a prior antiangiogenic therapy.

CASE CONCLUSION

Both TKI and mTOR inhibitor therapy fail, and the patient is eligible for third-line therapy. Because of his previous GI perforation, other TKIs are not an option. The patient opts for enrollment in hospice due to declining performance status. For other patients in this situation with a good performance status, nivolumab would be a reasonable option.

FUTURE DIRECTIONS

With the approval of nivolumab, multiple treatment options are now available for patients with metastatic or unresectable RCC. Development of other PD-1 inhibitors and immunotherapies as well as multi-targeted TKIs will only serve to expand treatment options for these patients. Given the aggressive course and poor prognosis of non-clear cell renal cell tumors and those with sarcomatoid features, evaluation of systemic and targeted therapies for these subtypes should remain active areas of research and investigation.

INTRODUCTION

Soft tissue sarcomas (STSs) are rare adult tumors, with 3.4 new cases per 100,000 persons or 12,310 expected new cases in 2016.¹ Sarcomas are a heterogeneous collection of tumors that affect fat, muscle, nerve, nerve sheath, vascular, and connective tissues. There are more than 50 histological subtypes that comprise this diverse category of tumors. Treatment varies by stage, with limb-sparing surgery representing the mainstay of curative-intent treatment. Radiation and chemotherapy may also be considered depending on the size, grade, and location of the tumor. Survival rates have been stagnant until recently, with a disease-specific survival hovering around 65%.¹ Given the complexity of these cases, all patients ideally should be evaluated and treated by a multidisciplinary team at an institution with extensive experience treating STS.²

EPIDEMIOLOGY AND CLASSIFICATION

The most common STS subtypes are gastrointestinal stromal tumor (GIST), undifferentiate pleomorphic sarcoma (previously referred to as malignant fibrous histiocytoma), liposarcoma, leiomyosarcoma, synovial sarcoma, malignant peripheral nerve sheath tumor, rhabdomyosarcoma, and unclassified sarcoma.³ Liposarcoma is one of the most common subtypes, comprising 20% of all STSs; it is subdivided into well-differentiated/dedifferentiated liposarcomas, myxoid/round cell liposarcomas, and pleomorphic liposarcomas. Well-differentiated liposarcomas tend to occur in the retroperitoneum and limbs, while both myxoid and round cell as well as pleomorphic liposarcomas more commonly originate on the limbs. Histology varies based on subtype and ranges from mature-appearing adipocytes and fibroblasts to undifferentiated cells with minimal lipogenic differentiation.⁴

Leiomyosarcomas are smooth muscle tumors and are usually located in the retroperitoneum, but have also been associated with peripheral soft tissue and vasculature. Typical histology ranges from well-defined areas of spindle-shaped cells to poorly differentiated anaplastic spindle cells.^5,6 Synovial sarcomas are a distinct type of STS that can show epithelial differentiation and account for 5% of adult STSs. The extremities are the most common presenting location (90%).⁷

Rhabdomyosarcomas are skeletal muscle tumors and are further subdivided into embryonal, alveolar, and pleomorphic subtypes. Embryonal histology ranges from primitive mesenchymal-appearing cells to highly differentiated muscle cells. Alveolar rhabdomyosarcoma has the worst prognosis of the subtypes and consists of round cells with high nuclear-to-chromatin ratios that form “glandular-like” or “alveolar” spaces.⁸ Pleomorphic rhabdomyosarcomas are composed of rhabdomyoblasts that can affect many different locations, but most commonly present on the lower extremities.⁹

Malignant peripheral nerve sheath tumor (MPNST) comprises 5% to 10% of all STSs. These tumors are associated with neurofibromatosis type 1 (NF-1), with 25% to 50% of tumors occurring in NF-1 patients. Additionally, most patients have a truncating lesion in the NF1 gene on chromosome 17.¹⁰ Anghileri et al in their single institution analysis of 205 patients with MPNSTs found the 2 most common presenting sites were the trunk and extremities. Histologically, these tumors have dense fascicles of spindle cells.¹⁰

GISTs are the most common STS of the gastrointestinal (GI) tract. Previously, GISTs were classified as smooth muscle tumors and were not accounted for in the literature as a separate entity distinct from leiomyomas, leiomyoblastomas, and leiomyosarcomas.¹¹ GISTs are found throughout the GI tract: the most common sites are the stomach (60%) and small intestine (30%). Less common sites include duodenum (4%–5%), esophagus (1%), rectum (1%–2%), and appendix (< 0.2%).¹² GISTs can be spindle cell, epithelioid, or mesenchymal tumors. Immunohistochemically, GISTs are KIT (CD117) positive. Other cell markers that are also commonly positive include CD34 (60%–70%) and smooth muscle actin (SMA) (25%).¹¹ The majority of GISTs (80%) have an activating c-KIT gene mutation. The most common mutation site is exon 11, with less common c-KIT gene mutations also occurring at exon 9 or 13. Not all GISTs have KIT mutations. The second most common mutation is the PDGFRA mutation (5%–10% of GISTs).² A minority of GISTs are negative for both KIT and PDGFRA mutations. These tumors were previously called wild-type, but as the majority have either a succinate dehydrogenase (SDH) loss of function or loss of SDHB protein expression, they are now referred to as SDH-deficient GISTs.² GISTs vary in aggressiveness from incidental to aggressive. Typically, small intestine and rectal GISTs are more aggressive than gastric GISTs. Both size and mitotic rate help to predict the metastatic potential of the tumor. Tumors less than 2 cm in size and having a mitotic rate of less than 5 per 50 high-power fields (hpf) have the lowest risk of metastases, while tumors greater than 5 cm and with more than 5 mitoses per 50 hpf have the highest rates of metastases.¹²

Undifferentiated sarcomas have no specific features and typically consist of primitive mesenchymal cells.

CLINICAL EVALUATION

CASE PRESENTATION

Initial Presentation and History

A 55-year-old man presents to his primary care physician with a painless mass in his anterior thigh. The mass has been present for the past 3 months and he believes that it is enlarging. The patient has a history of well-controlled hypertension and hyperlipidemia. His medications include atorvastatin and hydrochlorothiazide. He has no known drug allergies. Family history is notable for diabetes and hypertension. He drinks 4 to 5 alcoholic drinks a week and he is a former smoker. He quit smoking in his 30s and only smoked intermittently prior to quitting. He denies any illicit drug use. He works as a high school principal. Currently, he feels well. His review of systems is otherwise noncontributory.

Physical Examination

On physical exam, he is afebrile with a blood pressure of 132/75 mm Hg, respiratory rate of 10 breaths/min, and oxygen saturation of 99% on room air. He is a well appearing, overweight male. His head and neck exam is unremarkable. Lung exam reveals clear breath sounds, and cardiac exam reveals a regular rate and rhythm. His abdomen is obese, soft, and without hepatosplenomegaly. There is a large, fixed mass on the anterior lateral aspect of his right thigh. He has no appreciable lymphadenopathy. His neurological exam is unremarkable.

• What are risk factors for sarcoma?

There are few known risk factors for sarcoma. Established risks factors include prior radiation therapy, chronic lymphedema, viruses, and genetic cancer syndromes including Li-Fraumeni syndrome, hereditary retinoblastoma, and NF-1. Other environmental exposures include phenoxyacetic acids and chlorophenols.¹⁴ The majority of cases are sporadic, with only a minority of patients having one of these known risk factors.¹⁵ Up to one third of sarcomas have a specific translocation and are driven by fusion oncogenes (Table 1).

A painless mass is the most typical presenting symptom. Size at presentation varies based on location, with extremity and head and neck locations typically presenting at smaller sizes than retroperitoneal tumors.¹⁴ Patients may experience pain and numbness as the mass enlarges and impinges on surrounding structures including nerves and vasculature. The vast majority of patients are without systemic symptoms.

• How is sarcoma staged?

The American Joint Committee on Cancer (AJCC) staging system is the most widely used staging system in the United States. The latest AJCC manual was updated in 2010 to include a 3-tiered grading system where the tumor is classified according to tumor size, lymph node involvement, metastases, and grade at time of diagnosis (Table 2 and Table 3). Additionally, tumor depth in relation to deep fascia is also taken into account, with superficial tumors being assigned a designation of “a” and deep tumors a designation of “b.”

Previously, 2 of the most widely used grading systems were the National Cancer Institute (NCI) and French Federation of Cancer Centers Sarcoma Group (FNCLCC) systems, both 3-tier grading systems. The main components that determine the NCI grade are the tumor’s histologic type and location and the amount of tumor necrosis. The FNCLCC system evaluation focuses on tumor differentiation, mitotic rate, and amount of tumor necrosis. A study that compared the NCI and FNCLCC grading systems found that FNCLCC was a better predictor of mortality and distant metastasis.¹⁶ Previously, the AJCC was a 4-tier grading system, but the 2010 version was updated to the 3-tier FNCLCC grading system. Additionally, the AJCC system has reclassified single lymph node disease as stage III as it confers better survival than metastatic disease.¹⁷ It is important that pathology be evaluated by a sarcoma specialist as disagreements with regard to histologic subtype and grade are common.^18,19

• What are the most important prognostic factors?

Prognostic factors include grade, size, and presence of metastases at presentation. Best survival is associated with low-grade, small tumors with no metastases at time of diagnosis.¹⁴

Imaging should be undertaken to help differentiate between benign and malignant lesions. Ideally, it should be undertaken before a biopsy is planned as the imaging can be used to plan biopsy as well as provide invaluable prognostic information. There are several imaging modalities that should be considered during the preliminary work-up and staging of STSs. Conventional imaging includes magnetic resonance imaging (MRI) of the original tumor site; computed tomography (CT) to evaluate for pulmonary metastases and, depending on location, liver metastases; and in the case of small, low-grade tumors, chest radiography. MRI is considered the test of choice for soft tissue masses and can help delineate benign masses such as hematomas, lipomas, and hemangiomas from sarcomas.²⁰ It is difficult to compare the accuracy of positron emission tomography (PET)/CT to CT and MRI because most studies have evaluated PET/CT in parallel with CT and MRI.²¹ Tateishi et al compared the accuracy of conventional imaging, PET/CT, and PET/CT combined with conventional imaging at determining the TNM staging for 117 patients. They found that conventional imaging correctly classified 77% of patients, PET alone correctly classified 70%, PET/CT correctly classified 83%, and PET/CT combined with conventional imaging correctly staged 87%.²²

• Which subtypes are most likely to metastasize?

Although the vast majority of sarcomas spread hematogenously, 3 have a propensity to spread lymphogenously: epithelioid sarcoma, rhabdomyosarcoma, and clear-cell sarcoma. Additionally, certain subtypes are more likely to metastasize: leiomyosarcomas, synovial sarcomas, neurogenic sarcomas, rhabdomyosarcomas, and epithelioid sarcomas.²³ Sarcomas metastasize to the lungs more frequently than to the liver. The metastatic pattern is defined primarily by sarcoma subtype and site of primary tumor. Sarcomas rarely metastasize to the brain (~1%).

MANAGEMENT

CASE CONTINUED

The patient undergoes an ultrasound to better visualize the mass. Given the heterogeneous character of the mass, he is referred for an MRI to evaluate the mass and a CT scan of the chest, abdomen, and pelvis to evaluate for distant metastases. MRI reveals a 5.1 cm × 4.6 cm heterogeneous mass invading the superficial fascia of the rectus femoris muscle. No suspicious lymph nodes or other masses are identified on imaging. The patient next undergoes an image-guided core needle biopsy. Pathology from that procedure is consistent with a stage III, T2bNxMx, grade 3, dedifferentiated liposarcoma.

• What is the best management approach for this patient?

SURGERY

Surgery is the mainstay of treatment for STS. Patients with the best prognosis are those who undergo complete resection with negative surgical margins.^24,25 Goal tumor-free margin is 1 to 3 cm.²⁶ Complete resection confers the best long-term survival. Both local and metastatic recurrence is higher in patients with incomplete resection and positive margins.^24,25 In a study that analyzed 2084 localized primary STSs, patients with negative margins had a local recurrence rate of 15% versus a rate of 28% in patients with positive margins. This translated into higher 5-year local recurrence-free survival for patients with negative surgical margins (82%) compared to patients with positive margins (65%).²⁷ Another study similarly found that patients with negative margins at referral to their institution who underwent postoperative radiation had high local control rates of 93% (95% confidence interval [CI] 87% to 97%) at 5, 10, and 15 years.²⁶ Although radiation improves local control, neither preoperative or postoperative radiation has been shown to improve progression-free or overall survival.²⁸ Other factors that are associated with risk of recurrence are tumor location, history of previous recurrence, age of patient, histopathology, tumor grade, and tumor size. Approximately 40% to 50% of patients with high-grade tumors (defined as size > 5 cm, deep location, and high grade) will develop distant metastases.²⁹

Zagars et al found that positive or uncertain resection margin had a relative risk of local recurrence of 2.0 (95% CI 1.3 to 3.1; P = 0.002), and presentation with locally recurrent disease (vs new tumor) had a relative risk of local recurrence of 2.0 (95% CI 1.2 to 3.4; P = 0.013).²⁶ Patients with STS of head and neck and deep trunk have higher recurrence rates than those with superficial trunk and extremity STS. A single-institution retrospective review demonstrated that patients with completely resectable retroperitoneal sarcomas have longer median survival (103 months) compared to patients with incompletely resected abdominal sarcomas (18 months).²⁵

CASE CONTINUED

The patient is referred for limb-sparing surgery after presentation at a multidisciplinary tumor board. Prior to undergoing resection of the tumor, he is also referred to radiation-oncology to discuss the risks and benefits of combination radiotherapy and surgery as opposed to surgical resection alone.

• What is the evidence for radiation therapy?

RADIATION THERAPY

Radiation therapy is used in the preoperative, intraoperative, and postoperative settings to reduce the risk of local recurrence. There are several options for radiation, including external beam radiation therapy (EBRT), intraoperative radiation, and brachytherapy. A newer strategy, intensity-modulated radiation therapy (IMRT), utilizes 3-dimensional modeling to reduce radiation dosages. Overall there are no differences in overall survival or local recurrence rates between preoperative and postoperative radiation in STS.²⁸

The rationale behind preoperative radiation is that it reduces seeding of tumor cells, especially at the time of surgery.³⁰ Additionally, for EBRT, preoperative radiation has smaller field sizes and lower radiation doses. It can also help to reduce the size of the tumor prior to resection. Intraoperative radiation is often paired with preoperative radiation as a boost dose given only to the area of residual tumor.

Suit et al reviewed patients treated at a single institution with limb-sparing surgery and different radiation strategies. Local control rates between preoperative and postoperative radiation groups were not statistically significant. Local recurrence was linked to grade and size of the tumor in both groups. The authors did note, however, that the preoperative radiation group tended to have larger tumor sizes at baseline compared to the patients who received postoperative radiation.³⁰ A study that compared 190 patients who received preoperative and postoperative EBRT or brachytherapy (primary end point was wound complications, and local control was a secondary end point) showed a trend towards greater local control with preoperative radiation; however, the preoperative radiation group had significantly more wound complications compared to the postoperative radiation group.³¹

Yang et al found that postoperative EBRT decreases rates of local recurrence compared to surgery alone in high-grade extremity sarcomas.³² However, there were no differences in rates of distant metastases and overall survival between the 2 treatment groups. Similarly, in patients with low-grade sarcoma, there were fewer local recurrences in those who received EBRT and surgery as compared to surgery alone.³² Another study that evaluated 164 patients who received either adjuvant brachytherapy or no further therapy after complete resection found that brachytherapy reduced local recurrence in high-grade sarcomas. No difference in local recurrence rates was found in patients with low-grade sarcomas, nor was a significant difference found in the rates of distant metastases and overall survival between the 2 treatment groups.³³ With regards to IMRT, a single institution cohort experience with 41 patients who received IMRT following limb-sparing surgery had similar local control rates when compared to historical controls.³⁴

CASE CONTINUED

After discussion of the risks and benefits of radiation therapy, the patient opts for preoperative radiation prior to resection of his liposarcoma. He receives 50 Gy of EBRT prior to undergoing resection. Resection results in R1 margin consistent with microscopic disease. He receives 16 Gy of EBRT as a boost after recovery from his resection.²

• What is the evidence for neoadjuvant and adjuvant chemotherapy for stage I tumors?

CHEMOTHERAPY

Localized Sarcoma

For localized sarcoma, limb-sparing resection with or without radiation forms the backbone of treatment. Studies have evaluated chemotherapy in both the neoadjuvant and adjuvant settings, with the vast majority of studies evaluating doxorubicin-based chemotherapy regimens in the adjuvant settings. Due to the rare nature of sarcomas, most studies are not sufficiently powered to detect significant benefit from chemotherapy. Several trials evaluating chemotherapy regimens in the neoadjuvant and adjuvant settings needed to be terminated prematurely due to inadequate enrollment into the study. ^35,36

For stage IA (T1a-Tb, N0, M0, low grade) tumors, no additional therapy is recommended after limb-sparing surgery with appropriate surgical margins. For stage IB (T2a-2b, N0, M0, low grade) tumors with insufficient margins, re-resection and radiation therapy should be considered, while for stage IIA (T1a-1b, N0, M0, G2-3) tumors preoperative or postoperative radiation therapy is recommended.² Studies have not found benefit of adjuvant chemotherapy in these low-grade, stage I tumors in terms of progression-free survival and overall survival.³⁷

• At what stage should chemotherapy be considered?

For stage IIb and stage III tumors, surgery and radiation therapy again form the backbone of therapy; however, neoadjuvant and adjuvant chemotherapy are also recommended as considerations. Anthracycline-based chemotherapy with either single-agent doxorubicin or doxorubicin and ifosfamide in combination are considered first-line chemotherapy agents in locally advanced STS.^2,29,37

Evidence regarding the efficacy of both neoadjuvant and adjuvant chemotherapy regimens in the setting of locally advanced high-grade STS has been mixed. The Sarcoma Meta-analysis Collaboration evaluated 14 trials of doxorubicin-based adjuvant chemotherapy and found a trend towards overall survival in the treatment groups that received chemotherapy.³⁷ All trials included in the meta-analysis compared patients with localized resectable soft-tissue sarcomas who were randomized to either adjuvant chemotherapy or no adjuvant chemotherapy after limb-sparing surgery with or without radiation therapy. None of the individual trials showed a significant benefit, and all trials had large confidence intervals; however, the meta-analysis showed significant benefit in the chemotherapy treatment groups with regard to local recurrence, distant recurrence, and progression-free survival. No significant difference in overall survival was found.³⁷ Pervais et al updated the Sarcoma Meta-analysis Collaboration’s 1997 meta-analysis with the inclusion of 4 new trials that evaluated doxorubicin combined with ifosfamide and found that both patients who received doxorubicin-based regimens or doxorubicin with ifosfamide had significant decreases in distant and overall recurrences. Only the trials that utilized doxorubicin and ifosfamide had an improved overall survival that was statistically significant (hazard ratio 0.56 [95% CI 0.36 to 0.85]; P = 0.01).²⁹ Although no significant heterogeneity was found among the trials included in either meta-analysis, a variety of sarcomas were included in each clinical trial evaluated. Given the extremely small number of each sarcoma subtype present in each trial, subgroup analysis is difficult and prone to inaccuracies. As a result, it is not known if certain histological subtypes are more or less responsive to chemotherapy.^37–39

One randomized controlled trial evaluated neoadjuvant chemotherapy in high-risk sarcomas defined as tumors greater than 8 cm or grade II/III tumors. This study evaluated doxorubicin and ifosfamide and found no significant difference in disease-free and overall survival in the neoadjuvant therapy group compared to the control group.³⁵ There remains controversy in the literature with regards to adjuvant chemotherapy. Many oncologists offer adjuvant chemotherapy to patients with certain stage III subtypes. Examples of subtypes that may be offered adjuvant therapy include myxoid liposarcomas, synovial sarcomas, and leiomyosarcomas.² With regards to how many cycles of chemotherapy should be considered, a noninferiority study compared 3 cycles of epirubicin and ifosfamide to 5 cycles of epirubicin and ifosfamide in patients with high-risk locally advanced adult STSs. Three cycles of preoperative epirubicin and ifosfamide was found to be noninferior to 5 cycles with regards to overall survival.³⁸

• What is this patient’s risk for recurrence?

The patient is at intermediate risk for recurrence. Numerous studies have demonstrated that tumor size, grade, and location are the most important factors to determine risk of recurrence, with larger size, higher grades, and deeper locations being associated with higher risk of recurrence. In an analysis of 1041 patients with STS of the extremities, high grade was the most important risk factor for distant metastases.³⁹ The highest risk of recurrence is within the first 2 years. Given that the patient’s initial tumor was located in the extremity, he is more likely to have a distant metastasis as his site of recurrence; individuals with retroperitoneal tumors and visceral tumors are more likely to recur locally.⁴⁰ For STSs of the extremity, distant metastases determine overall survival, whereas patients with retroperitoneal sarcomas can die from complications of local metastases.⁴¹ Once a patient develops distant metastases, the most important prognostic factor is the size of the tumor, with tumors larger than 10 cm having a relative risk of 1.5 (95% CI 1.0 to 2.0).³⁹

• What are the recommendations for surveillance?

Surveillance recommendations are based on the stage of the sarcoma. Stage I tumors are the least likely to recur either locally or distally. As a result, it is recommended that stage I tumors be followed with history and physical exam every 3 to 6 months for the first 2 to 3 years, and then annually after the first 2 to 3 years. Chest x-rays should be considered every 6 to 12 months.² For stage II–IV tumors, history and physical exam is recommended every 3 to 6 months for the first 2 to 3 years. Chest and distant metastases imaging should also be performed every 3 to 6 months during this time frame. For the next 2 years, history and physical exam and imaging are recommended every 6 months. After the first 4 to 5 years, annual follow-up is recommended.²

CASE CONTINUED

The patient does well for 1 year. With physical therapy, he regains most of the strength and coordination of the lower extremity. He is followed every 3 months with chest x-rays and a MRI of the thigh for the first year. On his fourth follow-up clinic visit, he describes increased dyspnea on exertion over the previous few weeks and is found to have multiple lung metastases in both lungs on chest x-ray. He undergoes further evaluation for metastases and is not found to have any other metastatic lesions. Bronchoscopy and biopsy of 1 of the lung nodules confirms recurrent dedifferentiated liposarcoma.

• Should this patient undergo metastectomy?

An analysis of 3149 patients with STS treated at Memorial Sloan-Kettering who developed lung metastases found that patients with pulmonary metastases have survival rates of 25%. The most important prognostic factor for survival was complete resection of all metastases.⁴² For stage IV disease, surgery is used only in certain instances. In instances where tumor is more localized or limited, removal of metastases or metastectomy can play a role in management.²

CASE CONTINUED

Because the patient’s metastases are limited to the lungs, he is referred for metastectomy. He undergoes wedge resection for definitive diagnosis but it is not possible to completely resect all of the metastases. He is thus referred to a medical oncologist to discuss his treatment options.

• What are treatment options for unresectable or metastatic disease?

Metastatic Disease

Unlike local and locally advanced disease, chemotherapy forms the backbone of treatment in stage IV disease. Doxorubicin and olaratumab or doxorubicin and ifosfamide in combination are considered first line in metastatic disease. Response rates for single-agent doxorubicin range from 16% to 27%, while phase 2 and phase 3 studies of doxorubicin and ifosfamide have found response rates ranging from 18% to 36%.⁴³ In addition, the effectiveness of doxorubicin and ifosfamide phase 2 and 3 trials varied. Edmonson et al found a tumor regression rate of 34% for doxorubicin and ifosfamide as compared to 20% for doxorubicin alone.⁴⁴ In comparison, Santoro et al found a response rate of 21.3% for doxorubicin alone and 25.2% for doxorubicin and ifosfamide.⁴⁵ Neither study found increased survival benefit for doxorubicin and ifosfamide when compared to doxorubicin alone. In a Cochrane review evaluating randomized trials that compared doxorubicin and combination chemotherapy regimens, response rates varied from 14% for doxorubicin in combination with streptomycin to 34% for doxorubicin and ifosfamide. Most trials did not show a significant benefit for combination therapies when compared to doxorubicin alone.⁴³ Mean survival with doxorubicin or doxorubicin and ifosfamide is 12 months. High rates of recurrence highlight the need for additional chemotherapy regimens.

The newest approved agent is olaratumab, a monoclonal antibody that binds platelet-derived growth factor receptor alpha and prevents receptor activation. A phase 1-b and phase 2 trial evaluated patients with locally advanced and metastatic STS and randomly assigned them to either olaratumab and doxorubicin or doxorubicin alone.⁴⁶ Progression-free survival for olaratumab/doxorubicin was 6.6 months (95% CI 4.1 to 8.3) compared to 4.1 months (95% CI 2.8 to 5.4) for doxorubicin alone. The objective response rate was 18.2% (95% CI 9.8 to 29.6) for olaratumab/doxorubicin compared to 7.5% (95% CI 2.5 to 6.6) for doxorubicin alone. Furthermore, the median overall survival for olaratumab plus doxorubicin was 26.5 months (95% CI 20.9 to 31.7) compared to 14.7 months for doxorubicin alone (95% CI 5.5 to 26.0). Impressively, this improved response was notable across histological types. Furthermore, patients who had previously been treated with more than 1 regimen and those who were treatment naïve had similar response rates.⁴⁶

• What are second-line treatment options?

Doxorubicin has been used in combination with several other agents including dacarbazine (DTIC) as well as DTIC and ifosfamide (MAID). Borden et al evaluated patients with metastatic STS and randomly assigned the patients to either doxorubicin or doxorubicin and DTIC. Combination therapy demonstrated better tumor response than doxorubicin alone: 30% complete or partial response for combination therapy and 18% for doxorubicin alone.⁴⁷ However, Omura et al

Several additional regimens have undergone evaluation in metastatic and recurrent STSs. Gemcitabine has been used both as a single agent and as part of combination therapy in many studies. Studies with gemcitabine in combination with either docetaxel or DTIC have been the most efficacious. In a phase 2 trial, patients with metastatic STS were randomly assigned to either gemcitabine alone or gemcitabine and docetaxel. Combination therapy had a higher response rate (16% versus 8%) and longer overall survival (17.9 months versus 11.5 months) than gemcitabine alone.⁵⁰ Furthermore, a phase 2 trial of gemcitabine and docetaxel in patients with unresectable leiomyosarcoma showed an overall response rate of 56%, with 3 complete and 15 partial responses among the 34 patients enrolled in the study.⁵¹

A phase 2 trial randomly assigned patients with unresectable or metastatic STS to either DTIC or combination gemcitabine and DTIC.⁵² Gemcitabine-DTIC had a superior progression-free survival at 3 months (56% [95% CI 43% to 69%]) as compared to DTIC alone (37% [95% CI 23.5% to 50%]). Furthermore, mean progression-free survival and overall survival were improved in the gemcitabine-DTIC group (4.2 months and 16.8 months) as compared to the DTIC group (2.0 months and 8.2 months).⁵² DTIC has a single-agent response rate of 16%, but has been shown to be particularly effective in the setting of leiomyosarcomas.⁴⁹

• Does response to treatment regimens differ by histologic subtype?

The majority of STS trials include many different histologic subtypes. Given the rarity of sarcomas as a whole, many trials have had difficulty recruiting adequate numbers of patients to have sufficient power to definitely determine if the treatment under investigation has clinical benefit. Furthermore, the patients recruited have been heterogeneous with regard to subtype. Many older studies hypothesized that the efficacy of chemotherapeutic agents vary based on histologic subtype; however, for most subtypes the number of individuals included in those trials was too low to evaluate efficacy based on subtype.

Some exceptions exist, however. For example, both gemcitabine-DTIC and gemcitabine-docetaxel have been found to be particularly effective in the treatment of leiomyosarcomas.^50,52 Additionally, a retrospective study found a 51% overall response rate for patients with myxoid liposarcomas treated with trabectedin.⁵³ Studies of patients with angiosarcoma treated with paclitaxel have demonstrated response rates of 43% and 53%.^54,55

• What are the newest approved and investigational agents?

A recently approved agent is trabectedin, a tris tetrahydroisoquinoline alkaloid isolated from ascidians that binds to the minor groove of DNA and causes disruptions in the cell cycle. Samuels et al reported data from a single-arm, open-label expanded access trial that evaluated patients with advanced metastatic sarcomas.⁵⁶ In this study, patients with liposarcomas and leiomyosarcomas had an objective response rate of 6.9% (95% CI 4.8 to 9.6) as compared to a rate of 5.9% (95% CI 4.4 to 7.8) for all assessable patients. Median survival was 11.9 months for all patients, with improved median survivals for liposarcoma and leiomyosarcomas of 16.2 months (95% CI 14.1 to 19.5) compared to 8.4 months (95% CI 7.1 to 10.7 months) for other subtypes.⁵⁶

Schöffski et al evaluated eribulin, a chemotherapeutic agent that affects microtubule dynamics, in a phase 2 trial of patients with progressive or high-grade STS with progression on previous chemotherapy. They found a median progression-free survival of 2.6 months (95% CI 1.7 to 6.2) for adipocytic sarcoma, 2.9 months (95% CI 2.4 to 4.6) for leiomyosarcoma, 2.6 months (95% CI 2.3 to 4.3) for synovial sarcoma, and 2.1 months (95% CI 1.4 to 2.9) for other sarcomas.⁵⁷

Van der Graaf and colleagues randomly assigned patients with metastatic nonadipocytic STS to pazopanib or placebo in a phase 3 trial. Pazopanib is a small-molecule endothelial growth factor inhibitor with activity against vascular endothelial growth factors 1, 2, and 3 as well as platelet-derived growth factors. Median progression-free survival was 4.6 months (95% CI 3.7 to 4.8) with pazopanib compared to 1.6 months (95% CI 0.9 to 1.8) with placebo.⁵⁸ Adipocytic sarcomas (liposarcomas) were excluded from the trial because phase 2 trials had found a lower rate of progression-free survival (26%) for them compared to other subtypes.

Toxicities were seen with each of the regimens studied and were common in the randomized trials, with higher rates of toxicities in the combination chemotherapy regimens. The most common toxicities are myelosuppression, nausea, and vomiting. In the doxorubicin trials, the most common toxicities were myelosuppression, nausea, and vomiting.⁴⁴

Ifosfamide both as an individual agent and in combination with doxorubicin has higher rates and higher grades of toxicity than doxorubicin alone. Myelosuppression is the most common toxicity associated with ifosfamide, and the most commonly affected cell line is leukocytes.⁴⁴ Combination doxorubicin and ifosfamide also had high rates of nausea and vomiting (95%) and alopecia (100%).³⁵

Neutropenia is the most common toxicity associated with gemcitabine and dacarbazine, while their most common nonhematologic toxicities are fatigue and nausea.^52,59 Trabectedin’s most common toxicities are nausea (29%), neutropenia (24%), and fatigue (23%). It has also been shown to cause increased alkaline phosphatase (20%) and alanine aminotransferase (19%) levels.⁵⁶ In a phase 2 study of eribulin, 50% of patients had neutropenia, and other toxicities included fatigue, alopecia, nausea, sensory neuropathy, and thrombocytopenia.⁵⁷ Pazopanib is generally well tolerated; the most common toxicities are fatigue (65%), diarrhea (58%), nausea (54%), and hypertension (41%).⁵⁸ Higher rates of neutropenia, mucositis, nausea, vomiting, diarrhea, and transfusion reactions were seen with olaratumab and doxorubicin compared to doxorubicin alone in phase 1b and 2 studies.⁴⁶

CASE CONCLUSION

Given his poor prognosis with unresectable metastatic undifferentiated liposarcoma, the patient considers a clinical trial prior to undergoing combined therapy with doxorubicin and ifosfamide. He tolerates therapy well with stable disease at 6 months.

CONCLUSION

STSs are a heterogeneous collection of rare tumors. Low-grade, localized tumors have the best prognosis, and patients who undergo complete resection have the best long-term survival. Due to the rarity of STSs, trials often have limited enrollment, and little progress has been made with regards to treatment and survival rates for metastatic and unresectable disease. All patients should be evaluated and treated at specialized sarcoma centers. This case highlights the need for continued research and clinical trials to improve overall survival of patients with sarcoma.

INTRODUCTION

Soft tissue sarcomas (STSs) are rare adult tumors, with 3.4 new cases per 100,000 persons or 12,310 expected new cases in 2016.¹ Sarcomas are a heterogeneous collection of tumors that affect fat, muscle, nerve, nerve sheath, vascular, and connective tissues. There are more than 50 histological subtypes that comprise this diverse category of tumors. Treatment varies by stage, with limb-sparing surgery representing the mainstay of curative-intent treatment. Radiation and chemotherapy may also be considered depending on the size, grade, and location of the tumor. Survival rates have been stagnant until recently, with a disease-specific survival hovering around 65%.¹ Given the complexity of these cases, all patients ideally should be evaluated and treated by a multidisciplinary team at an institution with extensive experience treating STS.²

EPIDEMIOLOGY AND CLASSIFICATION

The most common STS subtypes are gastrointestinal stromal tumor (GIST), undifferentiate pleomorphic sarcoma (previously referred to as malignant fibrous histiocytoma), liposarcoma, leiomyosarcoma, synovial sarcoma, malignant peripheral nerve sheath tumor, rhabdomyosarcoma, and unclassified sarcoma.³ Liposarcoma is one of the most common subtypes, comprising 20% of all STSs; it is subdivided into well-differentiated/dedifferentiated liposarcomas, myxoid/round cell liposarcomas, and pleomorphic liposarcomas. Well-differentiated liposarcomas tend to occur in the retroperitoneum and limbs, while both myxoid and round cell as well as pleomorphic liposarcomas more commonly originate on the limbs. Histology varies based on subtype and ranges from mature-appearing adipocytes and fibroblasts to undifferentiated cells with minimal lipogenic differentiation.⁴

Leiomyosarcomas are smooth muscle tumors and are usually located in the retroperitoneum, but have also been associated with peripheral soft tissue and vasculature. Typical histology ranges from well-defined areas of spindle-shaped cells to poorly differentiated anaplastic spindle cells.^5,6 Synovial sarcomas are a distinct type of STS that can show epithelial differentiation and account for 5% of adult STSs. The extremities are the most common presenting location (90%).⁷

Rhabdomyosarcomas are skeletal muscle tumors and are further subdivided into embryonal, alveolar, and pleomorphic subtypes. Embryonal histology ranges from primitive mesenchymal-appearing cells to highly differentiated muscle cells. Alveolar rhabdomyosarcoma has the worst prognosis of the subtypes and consists of round cells with high nuclear-to-chromatin ratios that form “glandular-like” or “alveolar” spaces.⁸ Pleomorphic rhabdomyosarcomas are composed of rhabdomyoblasts that can affect many different locations, but most commonly present on the lower extremities.⁹

Malignant peripheral nerve sheath tumor (MPNST) comprises 5% to 10% of all STSs. These tumors are associated with neurofibromatosis type 1 (NF-1), with 25% to 50% of tumors occurring in NF-1 patients. Additionally, most patients have a truncating lesion in the NF1 gene on chromosome 17.¹⁰ Anghileri et al in their single institution analysis of 205 patients with MPNSTs found the 2 most common presenting sites were the trunk and extremities. Histologically, these tumors have dense fascicles of spindle cells.¹⁰

GISTs are the most common STS of the gastrointestinal (GI) tract. Previously, GISTs were classified as smooth muscle tumors and were not accounted for in the literature as a separate entity distinct from leiomyomas, leiomyoblastomas, and leiomyosarcomas.¹¹ GISTs are found throughout the GI tract: the most common sites are the stomach (60%) and small intestine (30%). Less common sites include duodenum (4%–5%), esophagus (1%), rectum (1%–2%), and appendix (< 0.2%).¹² GISTs can be spindle cell, epithelioid, or mesenchymal tumors. Immunohistochemically, GISTs are KIT (CD117) positive. Other cell markers that are also commonly positive include CD34 (60%–70%) and smooth muscle actin (SMA) (25%).¹¹ The majority of GISTs (80%) have an activating c-KIT gene mutation. The most common mutation site is exon 11, with less common c-KIT gene mutations also occurring at exon 9 or 13. Not all GISTs have KIT mutations. The second most common mutation is the PDGFRA mutation (5%–10% of GISTs).² A minority of GISTs are negative for both KIT and PDGFRA mutations. These tumors were previously called wild-type, but as the majority have either a succinate dehydrogenase (SDH) loss of function or loss of SDHB protein expression, they are now referred to as SDH-deficient GISTs.² GISTs vary in aggressiveness from incidental to aggressive. Typically, small intestine and rectal GISTs are more aggressive than gastric GISTs. Both size and mitotic rate help to predict the metastatic potential of the tumor. Tumors less than 2 cm in size and having a mitotic rate of less than 5 per 50 high-power fields (hpf) have the lowest risk of metastases, while tumors greater than 5 cm and with more than 5 mitoses per 50 hpf have the highest rates of metastases.¹²

Undifferentiated sarcomas have no specific features and typically consist of primitive mesenchymal cells.

CLINICAL EVALUATION

CASE PRESENTATION

Initial Presentation and History

A 55-year-old man presents to his primary care physician with a painless mass in his anterior thigh. The mass has been present for the past 3 months and he believes that it is enlarging. The patient has a history of well-controlled hypertension and hyperlipidemia. His medications include atorvastatin and hydrochlorothiazide. He has no known drug allergies. Family history is notable for diabetes and hypertension. He drinks 4 to 5 alcoholic drinks a week and he is a former smoker. He quit smoking in his 30s and only smoked intermittently prior to quitting. He denies any illicit drug use. He works as a high school principal. Currently, he feels well. His review of systems is otherwise noncontributory.

Physical Examination

On physical exam, he is afebrile with a blood pressure of 132/75 mm Hg, respiratory rate of 10 breaths/min, and oxygen saturation of 99% on room air. He is a well appearing, overweight male. His head and neck exam is unremarkable. Lung exam reveals clear breath sounds, and cardiac exam reveals a regular rate and rhythm. His abdomen is obese, soft, and without hepatosplenomegaly. There is a large, fixed mass on the anterior lateral aspect of his right thigh. He has no appreciable lymphadenopathy. His neurological exam is unremarkable.

• What are risk factors for sarcoma?

There are few known risk factors for sarcoma. Established risks factors include prior radiation therapy, chronic lymphedema, viruses, and genetic cancer syndromes including Li-Fraumeni syndrome, hereditary retinoblastoma, and NF-1. Other environmental exposures include phenoxyacetic acids and chlorophenols.¹⁴ The majority of cases are sporadic, with only a minority of patients having one of these known risk factors.¹⁵ Up to one third of sarcomas have a specific translocation and are driven by fusion oncogenes (Table 1).

A painless mass is the most typical presenting symptom. Size at presentation varies based on location, with extremity and head and neck locations typically presenting at smaller sizes than retroperitoneal tumors.¹⁴ Patients may experience pain and numbness as the mass enlarges and impinges on surrounding structures including nerves and vasculature. The vast majority of patients are without systemic symptoms.

• How is sarcoma staged?

The American Joint Committee on Cancer (AJCC) staging system is the most widely used staging system in the United States. The latest AJCC manual was updated in 2010 to include a 3-tiered grading system where the tumor is classified according to tumor size, lymph node involvement, metastases, and grade at time of diagnosis (Table 2 and Table 3). Additionally, tumor depth in relation to deep fascia is also taken into account, with superficial tumors being assigned a designation of “a” and deep tumors a designation of “b.”

Previously, 2 of the most widely used grading systems were the National Cancer Institute (NCI) and French Federation of Cancer Centers Sarcoma Group (FNCLCC) systems, both 3-tier grading systems. The main components that determine the NCI grade are the tumor’s histologic type and location and the amount of tumor necrosis. The FNCLCC system evaluation focuses on tumor differentiation, mitotic rate, and amount of tumor necrosis. A study that compared the NCI and FNCLCC grading systems found that FNCLCC was a better predictor of mortality and distant metastasis.¹⁶ Previously, the AJCC was a 4-tier grading system, but the 2010 version was updated to the 3-tier FNCLCC grading system. Additionally, the AJCC system has reclassified single lymph node disease as stage III as it confers better survival than metastatic disease.¹⁷ It is important that pathology be evaluated by a sarcoma specialist as disagreements with regard to histologic subtype and grade are common.^18,19

• What are the most important prognostic factors?

Prognostic factors include grade, size, and presence of metastases at presentation. Best survival is associated with low-grade, small tumors with no metastases at time of diagnosis.¹⁴

Imaging should be undertaken to help differentiate between benign and malignant lesions. Ideally, it should be undertaken before a biopsy is planned as the imaging can be used to plan biopsy as well as provide invaluable prognostic information. There are several imaging modalities that should be considered during the preliminary work-up and staging of STSs. Conventional imaging includes magnetic resonance imaging (MRI) of the original tumor site; computed tomography (CT) to evaluate for pulmonary metastases and, depending on location, liver metastases; and in the case of small, low-grade tumors, chest radiography. MRI is considered the test of choice for soft tissue masses and can help delineate benign masses such as hematomas, lipomas, and hemangiomas from sarcomas.²⁰ It is difficult to compare the accuracy of positron emission tomography (PET)/CT to CT and MRI because most studies have evaluated PET/CT in parallel with CT and MRI.²¹ Tateishi et al compared the accuracy of conventional imaging, PET/CT, and PET/CT combined with conventional imaging at determining the TNM staging for 117 patients. They found that conventional imaging correctly classified 77% of patients, PET alone correctly classified 70%, PET/CT correctly classified 83%, and PET/CT combined with conventional imaging correctly staged 87%.²²

• Which subtypes are most likely to metastasize?

Although the vast majority of sarcomas spread hematogenously, 3 have a propensity to spread lymphogenously: epithelioid sarcoma, rhabdomyosarcoma, and clear-cell sarcoma. Additionally, certain subtypes are more likely to metastasize: leiomyosarcomas, synovial sarcomas, neurogenic sarcomas, rhabdomyosarcomas, and epithelioid sarcomas.²³ Sarcomas metastasize to the lungs more frequently than to the liver. The metastatic pattern is defined primarily by sarcoma subtype and site of primary tumor. Sarcomas rarely metastasize to the brain (~1%).

MANAGEMENT

CASE CONTINUED

The patient undergoes an ultrasound to better visualize the mass. Given the heterogeneous character of the mass, he is referred for an MRI to evaluate the mass and a CT scan of the chest, abdomen, and pelvis to evaluate for distant metastases. MRI reveals a 5.1 cm × 4.6 cm heterogeneous mass invading the superficial fascia of the rectus femoris muscle. No suspicious lymph nodes or other masses are identified on imaging. The patient next undergoes an image-guided core needle biopsy. Pathology from that procedure is consistent with a stage III, T2bNxMx, grade 3, dedifferentiated liposarcoma.

• What is the best management approach for this patient?

SURGERY

Surgery is the mainstay of treatment for STS. Patients with the best prognosis are those who undergo complete resection with negative surgical margins.^24,25 Goal tumor-free margin is 1 to 3 cm.²⁶ Complete resection confers the best long-term survival. Both local and metastatic recurrence is higher in patients with incomplete resection and positive margins.^24,25 In a study that analyzed 2084 localized primary STSs, patients with negative margins had a local recurrence rate of 15% versus a rate of 28% in patients with positive margins. This translated into higher 5-year local recurrence-free survival for patients with negative surgical margins (82%) compared to patients with positive margins (65%).²⁷ Another study similarly found that patients with negative margins at referral to their institution who underwent postoperative radiation had high local control rates of 93% (95% confidence interval [CI] 87% to 97%) at 5, 10, and 15 years.²⁶ Although radiation improves local control, neither preoperative or postoperative radiation has been shown to improve progression-free or overall survival.²⁸ Other factors that are associated with risk of recurrence are tumor location, history of previous recurrence, age of patient, histopathology, tumor grade, and tumor size. Approximately 40% to 50% of patients with high-grade tumors (defined as size > 5 cm, deep location, and high grade) will develop distant metastases.²⁹

Zagars et al found that positive or uncertain resection margin had a relative risk of local recurrence of 2.0 (95% CI 1.3 to 3.1; P = 0.002), and presentation with locally recurrent disease (vs new tumor) had a relative risk of local recurrence of 2.0 (95% CI 1.2 to 3.4; P = 0.013).²⁶ Patients with STS of head and neck and deep trunk have higher recurrence rates than those with superficial trunk and extremity STS. A single-institution retrospective review demonstrated that patients with completely resectable retroperitoneal sarcomas have longer median survival (103 months) compared to patients with incompletely resected abdominal sarcomas (18 months).²⁵

CASE CONTINUED

The patient is referred for limb-sparing surgery after presentation at a multidisciplinary tumor board. Prior to undergoing resection of the tumor, he is also referred to radiation-oncology to discuss the risks and benefits of combination radiotherapy and surgery as opposed to surgical resection alone.

• What is the evidence for radiation therapy?

RADIATION THERAPY

Radiation therapy is used in the preoperative, intraoperative, and postoperative settings to reduce the risk of local recurrence. There are several options for radiation, including external beam radiation therapy (EBRT), intraoperative radiation, and brachytherapy. A newer strategy, intensity-modulated radiation therapy (IMRT), utilizes 3-dimensional modeling to reduce radiation dosages. Overall there are no differences in overall survival or local recurrence rates between preoperative and postoperative radiation in STS.²⁸

The rationale behind preoperative radiation is that it reduces seeding of tumor cells, especially at the time of surgery.³⁰ Additionally, for EBRT, preoperative radiation has smaller field sizes and lower radiation doses. It can also help to reduce the size of the tumor prior to resection. Intraoperative radiation is often paired with preoperative radiation as a boost dose given only to the area of residual tumor.

Suit et al reviewed patients treated at a single institution with limb-sparing surgery and different radiation strategies. Local control rates between preoperative and postoperative radiation groups were not statistically significant. Local recurrence was linked to grade and size of the tumor in both groups. The authors did note, however, that the preoperative radiation group tended to have larger tumor sizes at baseline compared to the patients who received postoperative radiation.³⁰ A study that compared 190 patients who received preoperative and postoperative EBRT or brachytherapy (primary end point was wound complications, and local control was a secondary end point) showed a trend towards greater local control with preoperative radiation; however, the preoperative radiation group had significantly more wound complications compared to the postoperative radiation group.³¹

Yang et al found that postoperative EBRT decreases rates of local recurrence compared to surgery alone in high-grade extremity sarcomas.³² However, there were no differences in rates of distant metastases and overall survival between the 2 treatment groups. Similarly, in patients with low-grade sarcoma, there were fewer local recurrences in those who received EBRT and surgery as compared to surgery alone.³² Another study that evaluated 164 patients who received either adjuvant brachytherapy or no further therapy after complete resection found that brachytherapy reduced local recurrence in high-grade sarcomas. No difference in local recurrence rates was found in patients with low-grade sarcomas, nor was a significant difference found in the rates of distant metastases and overall survival between the 2 treatment groups.³³ With regards to IMRT, a single institution cohort experience with 41 patients who received IMRT following limb-sparing surgery had similar local control rates when compared to historical controls.³⁴

CASE CONTINUED

After discussion of the risks and benefits of radiation therapy, the patient opts for preoperative radiation prior to resection of his liposarcoma. He receives 50 Gy of EBRT prior to undergoing resection. Resection results in R1 margin consistent with microscopic disease. He receives 16 Gy of EBRT as a boost after recovery from his resection.²

• What is the evidence for neoadjuvant and adjuvant chemotherapy for stage I tumors?

CHEMOTHERAPY

Localized Sarcoma

For localized sarcoma, limb-sparing resection with or without radiation forms the backbone of treatment. Studies have evaluated chemotherapy in both the neoadjuvant and adjuvant settings, with the vast majority of studies evaluating doxorubicin-based chemotherapy regimens in the adjuvant settings. Due to the rare nature of sarcomas, most studies are not sufficiently powered to detect significant benefit from chemotherapy. Several trials evaluating chemotherapy regimens in the neoadjuvant and adjuvant settings needed to be terminated prematurely due to inadequate enrollment into the study. ^35,36

For stage IA (T1a-Tb, N0, M0, low grade) tumors, no additional therapy is recommended after limb-sparing surgery with appropriate surgical margins. For stage IB (T2a-2b, N0, M0, low grade) tumors with insufficient margins, re-resection and radiation therapy should be considered, while for stage IIA (T1a-1b, N0, M0, G2-3) tumors preoperative or postoperative radiation therapy is recommended.² Studies have not found benefit of adjuvant chemotherapy in these low-grade, stage I tumors in terms of progression-free survival and overall survival.³⁷

• At what stage should chemotherapy be considered?

For stage IIb and stage III tumors, surgery and radiation therapy again form the backbone of therapy; however, neoadjuvant and adjuvant chemotherapy are also recommended as considerations. Anthracycline-based chemotherapy with either single-agent doxorubicin or doxorubicin and ifosfamide in combination are considered first-line chemotherapy agents in locally advanced STS.^2,29,37

Evidence regarding the efficacy of both neoadjuvant and adjuvant chemotherapy regimens in the setting of locally advanced high-grade STS has been mixed. The Sarcoma Meta-analysis Collaboration evaluated 14 trials of doxorubicin-based adjuvant chemotherapy and found a trend towards overall survival in the treatment groups that received chemotherapy.³⁷ All trials included in the meta-analysis compared patients with localized resectable soft-tissue sarcomas who were randomized to either adjuvant chemotherapy or no adjuvant chemotherapy after limb-sparing surgery with or without radiation therapy. None of the individual trials showed a significant benefit, and all trials had large confidence intervals; however, the meta-analysis showed significant benefit in the chemotherapy treatment groups with regard to local recurrence, distant recurrence, and progression-free survival. No significant difference in overall survival was found.³⁷ Pervais et al updated the Sarcoma Meta-analysis Collaboration’s 1997 meta-analysis with the inclusion of 4 new trials that evaluated doxorubicin combined with ifosfamide and found that both patients who received doxorubicin-based regimens or doxorubicin with ifosfamide had significant decreases in distant and overall recurrences. Only the trials that utilized doxorubicin and ifosfamide had an improved overall survival that was statistically significant (hazard ratio 0.56 [95% CI 0.36 to 0.85]; P = 0.01).²⁹ Although no significant heterogeneity was found among the trials included in either meta-analysis, a variety of sarcomas were included in each clinical trial evaluated. Given the extremely small number of each sarcoma subtype present in each trial, subgroup analysis is difficult and prone to inaccuracies. As a result, it is not known if certain histological subtypes are more or less responsive to chemotherapy.^37–39

One randomized controlled trial evaluated neoadjuvant chemotherapy in high-risk sarcomas defined as tumors greater than 8 cm or grade II/III tumors. This study evaluated doxorubicin and ifosfamide and found no significant difference in disease-free and overall survival in the neoadjuvant therapy group compared to the control group.³⁵ There remains controversy in the literature with regards to adjuvant chemotherapy. Many oncologists offer adjuvant chemotherapy to patients with certain stage III subtypes. Examples of subtypes that may be offered adjuvant therapy include myxoid liposarcomas, synovial sarcomas, and leiomyosarcomas.² With regards to how many cycles of chemotherapy should be considered, a noninferiority study compared 3 cycles of epirubicin and ifosfamide to 5 cycles of epirubicin and ifosfamide in patients with high-risk locally advanced adult STSs. Three cycles of preoperative epirubicin and ifosfamide was found to be noninferior to 5 cycles with regards to overall survival.³⁸

• What is this patient’s risk for recurrence?

The patient is at intermediate risk for recurrence. Numerous studies have demonstrated that tumor size, grade, and location are the most important factors to determine risk of recurrence, with larger size, higher grades, and deeper locations being associated with higher risk of recurrence. In an analysis of 1041 patients with STS of the extremities, high grade was the most important risk factor for distant metastases.³⁹ The highest risk of recurrence is within the first 2 years. Given that the patient’s initial tumor was located in the extremity, he is more likely to have a distant metastasis as his site of recurrence; individuals with retroperitoneal tumors and visceral tumors are more likely to recur locally.⁴⁰ For STSs of the extremity, distant metastases determine overall survival, whereas patients with retroperitoneal sarcomas can die from complications of local metastases.⁴¹ Once a patient develops distant metastases, the most important prognostic factor is the size of the tumor, with tumors larger than 10 cm having a relative risk of 1.5 (95% CI 1.0 to 2.0).³⁹

• What are the recommendations for surveillance?

Surveillance recommendations are based on the stage of the sarcoma. Stage I tumors are the least likely to recur either locally or distally. As a result, it is recommended that stage I tumors be followed with history and physical exam every 3 to 6 months for the first 2 to 3 years, and then annually after the first 2 to 3 years. Chest x-rays should be considered every 6 to 12 months.² For stage II–IV tumors, history and physical exam is recommended every 3 to 6 months for the first 2 to 3 years. Chest and distant metastases imaging should also be performed every 3 to 6 months during this time frame. For the next 2 years, history and physical exam and imaging are recommended every 6 months. After the first 4 to 5 years, annual follow-up is recommended.²

CASE CONTINUED

The patient does well for 1 year. With physical therapy, he regains most of the strength and coordination of the lower extremity. He is followed every 3 months with chest x-rays and a MRI of the thigh for the first year. On his fourth follow-up clinic visit, he describes increased dyspnea on exertion over the previous few weeks and is found to have multiple lung metastases in both lungs on chest x-ray. He undergoes further evaluation for metastases and is not found to have any other metastatic lesions. Bronchoscopy and biopsy of 1 of the lung nodules confirms recurrent dedifferentiated liposarcoma.

• Should this patient undergo metastectomy?

An analysis of 3149 patients with STS treated at Memorial Sloan-Kettering who developed lung metastases found that patients with pulmonary metastases have survival rates of 25%. The most important prognostic factor for survival was complete resection of all metastases.⁴² For stage IV disease, surgery is used only in certain instances. In instances where tumor is more localized or limited, removal of metastases or metastectomy can play a role in management.²

CASE CONTINUED

Because the patient’s metastases are limited to the lungs, he is referred for metastectomy. He undergoes wedge resection for definitive diagnosis but it is not possible to completely resect all of the metastases. He is thus referred to a medical oncologist to discuss his treatment options.

• What are treatment options for unresectable or metastatic disease?

Metastatic Disease

Unlike local and locally advanced disease, chemotherapy forms the backbone of treatment in stage IV disease. Doxorubicin and olaratumab or doxorubicin and ifosfamide in combination are considered first line in metastatic disease. Response rates for single-agent doxorubicin range from 16% to 27%, while phase 2 and phase 3 studies of doxorubicin and ifosfamide have found response rates ranging from 18% to 36%.⁴³ In addition, the effectiveness of doxorubicin and ifosfamide phase 2 and 3 trials varied. Edmonson et al found a tumor regression rate of 34% for doxorubicin and ifosfamide as compared to 20% for doxorubicin alone.⁴⁴ In comparison, Santoro et al found a response rate of 21.3% for doxorubicin alone and 25.2% for doxorubicin and ifosfamide.⁴⁵ Neither study found increased survival benefit for doxorubicin and ifosfamide when compared to doxorubicin alone. In a Cochrane review evaluating randomized trials that compared doxorubicin and combination chemotherapy regimens, response rates varied from 14% for doxorubicin in combination with streptomycin to 34% for doxorubicin and ifosfamide. Most trials did not show a significant benefit for combination therapies when compared to doxorubicin alone.⁴³ Mean survival with doxorubicin or doxorubicin and ifosfamide is 12 months. High rates of recurrence highlight the need for additional chemotherapy regimens.

The newest approved agent is olaratumab, a monoclonal antibody that binds platelet-derived growth factor receptor alpha and prevents receptor activation. A phase 1-b and phase 2 trial evaluated patients with locally advanced and metastatic STS and randomly assigned them to either olaratumab and doxorubicin or doxorubicin alone.⁴⁶ Progression-free survival for olaratumab/doxorubicin was 6.6 months (95% CI 4.1 to 8.3) compared to 4.1 months (95% CI 2.8 to 5.4) for doxorubicin alone. The objective response rate was 18.2% (95% CI 9.8 to 29.6) for olaratumab/doxorubicin compared to 7.5% (95% CI 2.5 to 6.6) for doxorubicin alone. Furthermore, the median overall survival for olaratumab plus doxorubicin was 26.5 months (95% CI 20.9 to 31.7) compared to 14.7 months for doxorubicin alone (95% CI 5.5 to 26.0). Impressively, this improved response was notable across histological types. Furthermore, patients who had previously been treated with more than 1 regimen and those who were treatment naïve had similar response rates.⁴⁶

• What are second-line treatment options?

Doxorubicin has been used in combination with several other agents including dacarbazine (DTIC) as well as DTIC and ifosfamide (MAID). Borden et al evaluated patients with metastatic STS and randomly assigned the patients to either doxorubicin or doxorubicin and DTIC. Combination therapy demonstrated better tumor response than doxorubicin alone: 30% complete or partial response for combination therapy and 18% for doxorubicin alone.⁴⁷ However, Omura et al

Several additional regimens have undergone evaluation in metastatic and recurrent STSs. Gemcitabine has been used both as a single agent and as part of combination therapy in many studies. Studies with gemcitabine in combination with either docetaxel or DTIC have been the most efficacious. In a phase 2 trial, patients with metastatic STS were randomly assigned to either gemcitabine alone or gemcitabine and docetaxel. Combination therapy had a higher response rate (16% versus 8%) and longer overall survival (17.9 months versus 11.5 months) than gemcitabine alone.⁵⁰ Furthermore, a phase 2 trial of gemcitabine and docetaxel in patients with unresectable leiomyosarcoma showed an overall response rate of 56%, with 3 complete and 15 partial responses among the 34 patients enrolled in the study.⁵¹

A phase 2 trial randomly assigned patients with unresectable or metastatic STS to either DTIC or combination gemcitabine and DTIC.⁵² Gemcitabine-DTIC had a superior progression-free survival at 3 months (56% [95% CI 43% to 69%]) as compared to DTIC alone (37% [95% CI 23.5% to 50%]). Furthermore, mean progression-free survival and overall survival were improved in the gemcitabine-DTIC group (4.2 months and 16.8 months) as compared to the DTIC group (2.0 months and 8.2 months).⁵² DTIC has a single-agent response rate of 16%, but has been shown to be particularly effective in the setting of leiomyosarcomas.⁴⁹

• Does response to treatment regimens differ by histologic subtype?

The majority of STS trials include many different histologic subtypes. Given the rarity of sarcomas as a whole, many trials have had difficulty recruiting adequate numbers of patients to have sufficient power to definitely determine if the treatment under investigation has clinical benefit. Furthermore, the patients recruited have been heterogeneous with regard to subtype. Many older studies hypothesized that the efficacy of chemotherapeutic agents vary based on histologic subtype; however, for most subtypes the number of individuals included in those trials was too low to evaluate efficacy based on subtype.

Some exceptions exist, however. For example, both gemcitabine-DTIC and gemcitabine-docetaxel have been found to be particularly effective in the treatment of leiomyosarcomas.^50,52 Additionally, a retrospective study found a 51% overall response rate for patients with myxoid liposarcomas treated with trabectedin.⁵³ Studies of patients with angiosarcoma treated with paclitaxel have demonstrated response rates of 43% and 53%.^54,55

• What are the newest approved and investigational agents?

A recently approved agent is trabectedin, a tris tetrahydroisoquinoline alkaloid isolated from ascidians that binds to the minor groove of DNA and causes disruptions in the cell cycle. Samuels et al reported data from a single-arm, open-label expanded access trial that evaluated patients with advanced metastatic sarcomas.⁵⁶ In this study, patients with liposarcomas and leiomyosarcomas had an objective response rate of 6.9% (95% CI 4.8 to 9.6) as compared to a rate of 5.9% (95% CI 4.4 to 7.8) for all assessable patients. Median survival was 11.9 months for all patients, with improved median survivals for liposarcoma and leiomyosarcomas of 16.2 months (95% CI 14.1 to 19.5) compared to 8.4 months (95% CI 7.1 to 10.7 months) for other subtypes.⁵⁶

Schöffski et al evaluated eribulin, a chemotherapeutic agent that affects microtubule dynamics, in a phase 2 trial of patients with progressive or high-grade STS with progression on previous chemotherapy. They found a median progression-free survival of 2.6 months (95% CI 1.7 to 6.2) for adipocytic sarcoma, 2.9 months (95% CI 2.4 to 4.6) for leiomyosarcoma, 2.6 months (95% CI 2.3 to 4.3) for synovial sarcoma, and 2.1 months (95% CI 1.4 to 2.9) for other sarcomas.⁵⁷

Van der Graaf and colleagues randomly assigned patients with metastatic nonadipocytic STS to pazopanib or placebo in a phase 3 trial. Pazopanib is a small-molecule endothelial growth factor inhibitor with activity against vascular endothelial growth factors 1, 2, and 3 as well as platelet-derived growth factors. Median progression-free survival was 4.6 months (95% CI 3.7 to 4.8) with pazopanib compared to 1.6 months (95% CI 0.9 to 1.8) with placebo.⁵⁸ Adipocytic sarcomas (liposarcomas) were excluded from the trial because phase 2 trials had found a lower rate of progression-free survival (26%) for them compared to other subtypes.

Toxicities were seen with each of the regimens studied and were common in the randomized trials, with higher rates of toxicities in the combination chemotherapy regimens. The most common toxicities are myelosuppression, nausea, and vomiting. In the doxorubicin trials, the most common toxicities were myelosuppression, nausea, and vomiting.⁴⁴

Ifosfamide both as an individual agent and in combination with doxorubicin has higher rates and higher grades of toxicity than doxorubicin alone. Myelosuppression is the most common toxicity associated with ifosfamide, and the most commonly affected cell line is leukocytes.⁴⁴ Combination doxorubicin and ifosfamide also had high rates of nausea and vomiting (95%) and alopecia (100%).³⁵

Neutropenia is the most common toxicity associated with gemcitabine and dacarbazine, while their most common nonhematologic toxicities are fatigue and nausea.^52,59 Trabectedin’s most common toxicities are nausea (29%), neutropenia (24%), and fatigue (23%). It has also been shown to cause increased alkaline phosphatase (20%) and alanine aminotransferase (19%) levels.⁵⁶ In a phase 2 study of eribulin, 50% of patients had neutropenia, and other toxicities included fatigue, alopecia, nausea, sensory neuropathy, and thrombocytopenia.⁵⁷ Pazopanib is generally well tolerated; the most common toxicities are fatigue (65%), diarrhea (58%), nausea (54%), and hypertension (41%).⁵⁸ Higher rates of neutropenia, mucositis, nausea, vomiting, diarrhea, and transfusion reactions were seen with olaratumab and doxorubicin compared to doxorubicin alone in phase 1b and 2 studies.⁴⁶

CASE CONCLUSION

Given his poor prognosis with unresectable metastatic undifferentiated liposarcoma, the patient considers a clinical trial prior to undergoing combined therapy with doxorubicin and ifosfamide. He tolerates therapy well with stable disease at 6 months.

CONCLUSION

STSs are a heterogeneous collection of rare tumors. Low-grade, localized tumors have the best prognosis, and patients who undergo complete resection have the best long-term survival. Due to the rarity of STSs, trials often have limited enrollment, and little progress has been made with regards to treatment and survival rates for metastatic and unresectable disease. All patients should be evaluated and treated at specialized sarcoma centers. This case highlights the need for continued research and clinical trials to improve overall survival of patients with sarcoma.

INTRODUCTION

Soft tissue sarcomas (STSs) are rare adult tumors, with 3.4 new cases per 100,000 persons or 12,310 expected new cases in 2016.¹ Sarcomas are a heterogeneous collection of tumors that affect fat, muscle, nerve, nerve sheath, vascular, and connective tissues. There are more than 50 histological subtypes that comprise this diverse category of tumors. Treatment varies by stage, with limb-sparing surgery representing the mainstay of curative-intent treatment. Radiation and chemotherapy may also be considered depending on the size, grade, and location of the tumor. Survival rates have been stagnant until recently, with a disease-specific survival hovering around 65%.¹ Given the complexity of these cases, all patients ideally should be evaluated and treated by a multidisciplinary team at an institution with extensive experience treating STS.²

EPIDEMIOLOGY AND CLASSIFICATION

The most common STS subtypes are gastrointestinal stromal tumor (GIST), undifferentiate pleomorphic sarcoma (previously referred to as malignant fibrous histiocytoma), liposarcoma, leiomyosarcoma, synovial sarcoma, malignant peripheral nerve sheath tumor, rhabdomyosarcoma, and unclassified sarcoma.³ Liposarcoma is one of the most common subtypes, comprising 20% of all STSs; it is subdivided into well-differentiated/dedifferentiated liposarcomas, myxoid/round cell liposarcomas, and pleomorphic liposarcomas. Well-differentiated liposarcomas tend to occur in the retroperitoneum and limbs, while both myxoid and round cell as well as pleomorphic liposarcomas more commonly originate on the limbs. Histology varies based on subtype and ranges from mature-appearing adipocytes and fibroblasts to undifferentiated cells with minimal lipogenic differentiation.⁴

Leiomyosarcomas are smooth muscle tumors and are usually located in the retroperitoneum, but have also been associated with peripheral soft tissue and vasculature. Typical histology ranges from well-defined areas of spindle-shaped cells to poorly differentiated anaplastic spindle cells.^5,6 Synovial sarcomas are a distinct type of STS that can show epithelial differentiation and account for 5% of adult STSs. The extremities are the most common presenting location (90%).⁷

Rhabdomyosarcomas are skeletal muscle tumors and are further subdivided into embryonal, alveolar, and pleomorphic subtypes. Embryonal histology ranges from primitive mesenchymal-appearing cells to highly differentiated muscle cells. Alveolar rhabdomyosarcoma has the worst prognosis of the subtypes and consists of round cells with high nuclear-to-chromatin ratios that form “glandular-like” or “alveolar” spaces.⁸ Pleomorphic rhabdomyosarcomas are composed of rhabdomyoblasts that can affect many different locations, but most commonly present on the lower extremities.⁹

Malignant peripheral nerve sheath tumor (MPNST) comprises 5% to 10% of all STSs. These tumors are associated with neurofibromatosis type 1 (NF-1), with 25% to 50% of tumors occurring in NF-1 patients. Additionally, most patients have a truncating lesion in the NF1 gene on chromosome 17.¹⁰ Anghileri et al in their single institution analysis of 205 patients with MPNSTs found the 2 most common presenting sites were the trunk and extremities. Histologically, these tumors have dense fascicles of spindle cells.¹⁰

GISTs are the most common STS of the gastrointestinal (GI) tract. Previously, GISTs were classified as smooth muscle tumors and were not accounted for in the literature as a separate entity distinct from leiomyomas, leiomyoblastomas, and leiomyosarcomas.¹¹ GISTs are found throughout the GI tract: the most common sites are the stomach (60%) and small intestine (30%). Less common sites include duodenum (4%–5%), esophagus (1%), rectum (1%–2%), and appendix (< 0.2%).¹² GISTs can be spindle cell, epithelioid, or mesenchymal tumors. Immunohistochemically, GISTs are KIT (CD117) positive. Other cell markers that are also commonly positive include CD34 (60%–70%) and smooth muscle actin (SMA) (25%).¹¹ The majority of GISTs (80%) have an activating c-KIT gene mutation. The most common mutation site is exon 11, with less common c-KIT gene mutations also occurring at exon 9 or 13. Not all GISTs have KIT mutations. The second most common mutation is the PDGFRA mutation (5%–10% of GISTs).² A minority of GISTs are negative for both KIT and PDGFRA mutations. These tumors were previously called wild-type, but as the majority have either a succinate dehydrogenase (SDH) loss of function or loss of SDHB protein expression, they are now referred to as SDH-deficient GISTs.² GISTs vary in aggressiveness from incidental to aggressive. Typically, small intestine and rectal GISTs are more aggressive than gastric GISTs. Both size and mitotic rate help to predict the metastatic potential of the tumor. Tumors less than 2 cm in size and having a mitotic rate of less than 5 per 50 high-power fields (hpf) have the lowest risk of metastases, while tumors greater than 5 cm and with more than 5 mitoses per 50 hpf have the highest rates of metastases.¹²

Undifferentiated sarcomas have no specific features and typically consist of primitive mesenchymal cells.

CLINICAL EVALUATION

CASE PRESENTATION

Initial Presentation and History

A 55-year-old man presents to his primary care physician with a painless mass in his anterior thigh. The mass has been present for the past 3 months and he believes that it is enlarging. The patient has a history of well-controlled hypertension and hyperlipidemia. His medications include atorvastatin and hydrochlorothiazide. He has no known drug allergies. Family history is notable for diabetes and hypertension. He drinks 4 to 5 alcoholic drinks a week and he is a former smoker. He quit smoking in his 30s and only smoked intermittently prior to quitting. He denies any illicit drug use. He works as a high school principal. Currently, he feels well. His review of systems is otherwise noncontributory.

Physical Examination

On physical exam, he is afebrile with a blood pressure of 132/75 mm Hg, respiratory rate of 10 breaths/min, and oxygen saturation of 99% on room air. He is a well appearing, overweight male. His head and neck exam is unremarkable. Lung exam reveals clear breath sounds, and cardiac exam reveals a regular rate and rhythm. His abdomen is obese, soft, and without hepatosplenomegaly. There is a large, fixed mass on the anterior lateral aspect of his right thigh. He has no appreciable lymphadenopathy. His neurological exam is unremarkable.

• What are risk factors for sarcoma?

There are few known risk factors for sarcoma. Established risks factors include prior radiation therapy, chronic lymphedema, viruses, and genetic cancer syndromes including Li-Fraumeni syndrome, hereditary retinoblastoma, and NF-1. Other environmental exposures include phenoxyacetic acids and chlorophenols.¹⁴ The majority of cases are sporadic, with only a minority of patients having one of these known risk factors.¹⁵ Up to one third of sarcomas have a specific translocation and are driven by fusion oncogenes (Table 1).

A painless mass is the most typical presenting symptom. Size at presentation varies based on location, with extremity and head and neck locations typically presenting at smaller sizes than retroperitoneal tumors.¹⁴ Patients may experience pain and numbness as the mass enlarges and impinges on surrounding structures including nerves and vasculature. The vast majority of patients are without systemic symptoms.

• How is sarcoma staged?

The American Joint Committee on Cancer (AJCC) staging system is the most widely used staging system in the United States. The latest AJCC manual was updated in 2010 to include a 3-tiered grading system where the tumor is classified according to tumor size, lymph node involvement, metastases, and grade at time of diagnosis (Table 2 and Table 3). Additionally, tumor depth in relation to deep fascia is also taken into account, with superficial tumors being assigned a designation of “a” and deep tumors a designation of “b.”

Previously, 2 of the most widely used grading systems were the National Cancer Institute (NCI) and French Federation of Cancer Centers Sarcoma Group (FNCLCC) systems, both 3-tier grading systems. The main components that determine the NCI grade are the tumor’s histologic type and location and the amount of tumor necrosis. The FNCLCC system evaluation focuses on tumor differentiation, mitotic rate, and amount of tumor necrosis. A study that compared the NCI and FNCLCC grading systems found that FNCLCC was a better predictor of mortality and distant metastasis.¹⁶ Previously, the AJCC was a 4-tier grading system, but the 2010 version was updated to the 3-tier FNCLCC grading system. Additionally, the AJCC system has reclassified single lymph node disease as stage III as it confers better survival than metastatic disease.¹⁷ It is important that pathology be evaluated by a sarcoma specialist as disagreements with regard to histologic subtype and grade are common.^18,19

• What are the most important prognostic factors?

Prognostic factors include grade, size, and presence of metastases at presentation. Best survival is associated with low-grade, small tumors with no metastases at time of diagnosis.¹⁴

Imaging should be undertaken to help differentiate between benign and malignant lesions. Ideally, it should be undertaken before a biopsy is planned as the imaging can be used to plan biopsy as well as provide invaluable prognostic information. There are several imaging modalities that should be considered during the preliminary work-up and staging of STSs. Conventional imaging includes magnetic resonance imaging (MRI) of the original tumor site; computed tomography (CT) to evaluate for pulmonary metastases and, depending on location, liver metastases; and in the case of small, low-grade tumors, chest radiography. MRI is considered the test of choice for soft tissue masses and can help delineate benign masses such as hematomas, lipomas, and hemangiomas from sarcomas.²⁰ It is difficult to compare the accuracy of positron emission tomography (PET)/CT to CT and MRI because most studies have evaluated PET/CT in parallel with CT and MRI.²¹ Tateishi et al compared the accuracy of conventional imaging, PET/CT, and PET/CT combined with conventional imaging at determining the TNM staging for 117 patients. They found that conventional imaging correctly classified 77% of patients, PET alone correctly classified 70%, PET/CT correctly classified 83%, and PET/CT combined with conventional imaging correctly staged 87%.²²

• Which subtypes are most likely to metastasize?

Although the vast majority of sarcomas spread hematogenously, 3 have a propensity to spread lymphogenously: epithelioid sarcoma, rhabdomyosarcoma, and clear-cell sarcoma. Additionally, certain subtypes are more likely to metastasize: leiomyosarcomas, synovial sarcomas, neurogenic sarcomas, rhabdomyosarcomas, and epithelioid sarcomas.²³ Sarcomas metastasize to the lungs more frequently than to the liver. The metastatic pattern is defined primarily by sarcoma subtype and site of primary tumor. Sarcomas rarely metastasize to the brain (~1%).

MANAGEMENT

CASE CONTINUED

The patient undergoes an ultrasound to better visualize the mass. Given the heterogeneous character of the mass, he is referred for an MRI to evaluate the mass and a CT scan of the chest, abdomen, and pelvis to evaluate for distant metastases. MRI reveals a 5.1 cm × 4.6 cm heterogeneous mass invading the superficial fascia of the rectus femoris muscle. No suspicious lymph nodes or other masses are identified on imaging. The patient next undergoes an image-guided core needle biopsy. Pathology from that procedure is consistent with a stage III, T2bNxMx, grade 3, dedifferentiated liposarcoma.

• What is the best management approach for this patient?

SURGERY

Surgery is the mainstay of treatment for STS. Patients with the best prognosis are those who undergo complete resection with negative surgical margins.^24,25 Goal tumor-free margin is 1 to 3 cm.²⁶ Complete resection confers the best long-term survival. Both local and metastatic recurrence is higher in patients with incomplete resection and positive margins.^24,25 In a study that analyzed 2084 localized primary STSs, patients with negative margins had a local recurrence rate of 15% versus a rate of 28% in patients with positive margins. This translated into higher 5-year local recurrence-free survival for patients with negative surgical margins (82%) compared to patients with positive margins (65%).²⁷ Another study similarly found that patients with negative margins at referral to their institution who underwent postoperative radiation had high local control rates of 93% (95% confidence interval [CI] 87% to 97%) at 5, 10, and 15 years.²⁶ Although radiation improves local control, neither preoperative or postoperative radiation has been shown to improve progression-free or overall survival.²⁸ Other factors that are associated with risk of recurrence are tumor location, history of previous recurrence, age of patient, histopathology, tumor grade, and tumor size. Approximately 40% to 50% of patients with high-grade tumors (defined as size > 5 cm, deep location, and high grade) will develop distant metastases.²⁹

Zagars et al found that positive or uncertain resection margin had a relative risk of local recurrence of 2.0 (95% CI 1.3 to 3.1; P = 0.002), and presentation with locally recurrent disease (vs new tumor) had a relative risk of local recurrence of 2.0 (95% CI 1.2 to 3.4; P = 0.013).²⁶ Patients with STS of head and neck and deep trunk have higher recurrence rates than those with superficial trunk and extremity STS. A single-institution retrospective review demonstrated that patients with completely resectable retroperitoneal sarcomas have longer median survival (103 months) compared to patients with incompletely resected abdominal sarcomas (18 months).²⁵

CASE CONTINUED

The patient is referred for limb-sparing surgery after presentation at a multidisciplinary tumor board. Prior to undergoing resection of the tumor, he is also referred to radiation-oncology to discuss the risks and benefits of combination radiotherapy and surgery as opposed to surgical resection alone.

• What is the evidence for radiation therapy?

RADIATION THERAPY

Radiation therapy is used in the preoperative, intraoperative, and postoperative settings to reduce the risk of local recurrence. There are several options for radiation, including external beam radiation therapy (EBRT), intraoperative radiation, and brachytherapy. A newer strategy, intensity-modulated radiation therapy (IMRT), utilizes 3-dimensional modeling to reduce radiation dosages. Overall there are no differences in overall survival or local recurrence rates between preoperative and postoperative radiation in STS.²⁸

The rationale behind preoperative radiation is that it reduces seeding of tumor cells, especially at the time of surgery.³⁰ Additionally, for EBRT, preoperative radiation has smaller field sizes and lower radiation doses. It can also help to reduce the size of the tumor prior to resection. Intraoperative radiation is often paired with preoperative radiation as a boost dose given only to the area of residual tumor.

Suit et al reviewed patients treated at a single institution with limb-sparing surgery and different radiation strategies. Local control rates between preoperative and postoperative radiation groups were not statistically significant. Local recurrence was linked to grade and size of the tumor in both groups. The authors did note, however, that the preoperative radiation group tended to have larger tumor sizes at baseline compared to the patients who received postoperative radiation.³⁰ A study that compared 190 patients who received preoperative and postoperative EBRT or brachytherapy (primary end point was wound complications, and local control was a secondary end point) showed a trend towards greater local control with preoperative radiation; however, the preoperative radiation group had significantly more wound complications compared to the postoperative radiation group.³¹

Yang et al found that postoperative EBRT decreases rates of local recurrence compared to surgery alone in high-grade extremity sarcomas.³² However, there were no differences in rates of distant metastases and overall survival between the 2 treatment groups. Similarly, in patients with low-grade sarcoma, there were fewer local recurrences in those who received EBRT and surgery as compared to surgery alone.³² Another study that evaluated 164 patients who received either adjuvant brachytherapy or no further therapy after complete resection found that brachytherapy reduced local recurrence in high-grade sarcomas. No difference in local recurrence rates was found in patients with low-grade sarcomas, nor was a significant difference found in the rates of distant metastases and overall survival between the 2 treatment groups.³³ With regards to IMRT, a single institution cohort experience with 41 patients who received IMRT following limb-sparing surgery had similar local control rates when compared to historical controls.³⁴

CASE CONTINUED

After discussion of the risks and benefits of radiation therapy, the patient opts for preoperative radiation prior to resection of his liposarcoma. He receives 50 Gy of EBRT prior to undergoing resection. Resection results in R1 margin consistent with microscopic disease. He receives 16 Gy of EBRT as a boost after recovery from his resection.²

• What is the evidence for neoadjuvant and adjuvant chemotherapy for stage I tumors?

CHEMOTHERAPY

Localized Sarcoma

For localized sarcoma, limb-sparing resection with or without radiation forms the backbone of treatment. Studies have evaluated chemotherapy in both the neoadjuvant and adjuvant settings, with the vast majority of studies evaluating doxorubicin-based chemotherapy regimens in the adjuvant settings. Due to the rare nature of sarcomas, most studies are not sufficiently powered to detect significant benefit from chemotherapy. Several trials evaluating chemotherapy regimens in the neoadjuvant and adjuvant settings needed to be terminated prematurely due to inadequate enrollment into the study. ^35,36

For stage IA (T1a-Tb, N0, M0, low grade) tumors, no additional therapy is recommended after limb-sparing surgery with appropriate surgical margins. For stage IB (T2a-2b, N0, M0, low grade) tumors with insufficient margins, re-resection and radiation therapy should be considered, while for stage IIA (T1a-1b, N0, M0, G2-3) tumors preoperative or postoperative radiation therapy is recommended.² Studies have not found benefit of adjuvant chemotherapy in these low-grade, stage I tumors in terms of progression-free survival and overall survival.³⁷

• At what stage should chemotherapy be considered?

For stage IIb and stage III tumors, surgery and radiation therapy again form the backbone of therapy; however, neoadjuvant and adjuvant chemotherapy are also recommended as considerations. Anthracycline-based chemotherapy with either single-agent doxorubicin or doxorubicin and ifosfamide in combination are considered first-line chemotherapy agents in locally advanced STS.^2,29,37

Evidence regarding the efficacy of both neoadjuvant and adjuvant chemotherapy regimens in the setting of locally advanced high-grade STS has been mixed. The Sarcoma Meta-analysis Collaboration evaluated 14 trials of doxorubicin-based adjuvant chemotherapy and found a trend towards overall survival in the treatment groups that received chemotherapy.³⁷ All trials included in the meta-analysis compared patients with localized resectable soft-tissue sarcomas who were randomized to either adjuvant chemotherapy or no adjuvant chemotherapy after limb-sparing surgery with or without radiation therapy. None of the individual trials showed a significant benefit, and all trials had large confidence intervals; however, the meta-analysis showed significant benefit in the chemotherapy treatment groups with regard to local recurrence, distant recurrence, and progression-free survival. No significant difference in overall survival was found.³⁷ Pervais et al updated the Sarcoma Meta-analysis Collaboration’s 1997 meta-analysis with the inclusion of 4 new trials that evaluated doxorubicin combined with ifosfamide and found that both patients who received doxorubicin-based regimens or doxorubicin with ifosfamide had significant decreases in distant and overall recurrences. Only the trials that utilized doxorubicin and ifosfamide had an improved overall survival that was statistically significant (hazard ratio 0.56 [95% CI 0.36 to 0.85]; P = 0.01).²⁹ Although no significant heterogeneity was found among the trials included in either meta-analysis, a variety of sarcomas were included in each clinical trial evaluated. Given the extremely small number of each sarcoma subtype present in each trial, subgroup analysis is difficult and prone to inaccuracies. As a result, it is not known if certain histological subtypes are more or less responsive to chemotherapy.^37–39

One randomized controlled trial evaluated neoadjuvant chemotherapy in high-risk sarcomas defined as tumors greater than 8 cm or grade II/III tumors. This study evaluated doxorubicin and ifosfamide and found no significant difference in disease-free and overall survival in the neoadjuvant therapy group compared to the control group.³⁵ There remains controversy in the literature with regards to adjuvant chemotherapy. Many oncologists offer adjuvant chemotherapy to patients with certain stage III subtypes. Examples of subtypes that may be offered adjuvant therapy include myxoid liposarcomas, synovial sarcomas, and leiomyosarcomas.² With regards to how many cycles of chemotherapy should be considered, a noninferiority study compared 3 cycles of epirubicin and ifosfamide to 5 cycles of epirubicin and ifosfamide in patients with high-risk locally advanced adult STSs. Three cycles of preoperative epirubicin and ifosfamide was found to be noninferior to 5 cycles with regards to overall survival.³⁸

• What is this patient’s risk for recurrence?

The patient is at intermediate risk for recurrence. Numerous studies have demonstrated that tumor size, grade, and location are the most important factors to determine risk of recurrence, with larger size, higher grades, and deeper locations being associated with higher risk of recurrence. In an analysis of 1041 patients with STS of the extremities, high grade was the most important risk factor for distant metastases.³⁹ The highest risk of recurrence is within the first 2 years. Given that the patient’s initial tumor was located in the extremity, he is more likely to have a distant metastasis as his site of recurrence; individuals with retroperitoneal tumors and visceral tumors are more likely to recur locally.⁴⁰ For STSs of the extremity, distant metastases determine overall survival, whereas patients with retroperitoneal sarcomas can die from complications of local metastases.⁴¹ Once a patient develops distant metastases, the most important prognostic factor is the size of the tumor, with tumors larger than 10 cm having a relative risk of 1.5 (95% CI 1.0 to 2.0).³⁹

• What are the recommendations for surveillance?

Surveillance recommendations are based on the stage of the sarcoma. Stage I tumors are the least likely to recur either locally or distally. As a result, it is recommended that stage I tumors be followed with history and physical exam every 3 to 6 months for the first 2 to 3 years, and then annually after the first 2 to 3 years. Chest x-rays should be considered every 6 to 12 months.² For stage II–IV tumors, history and physical exam is recommended every 3 to 6 months for the first 2 to 3 years. Chest and distant metastases imaging should also be performed every 3 to 6 months during this time frame. For the next 2 years, history and physical exam and imaging are recommended every 6 months. After the first 4 to 5 years, annual follow-up is recommended.²

CASE CONTINUED

The patient does well for 1 year. With physical therapy, he regains most of the strength and coordination of the lower extremity. He is followed every 3 months with chest x-rays and a MRI of the thigh for the first year. On his fourth follow-up clinic visit, he describes increased dyspnea on exertion over the previous few weeks and is found to have multiple lung metastases in both lungs on chest x-ray. He undergoes further evaluation for metastases and is not found to have any other metastatic lesions. Bronchoscopy and biopsy of 1 of the lung nodules confirms recurrent dedifferentiated liposarcoma.

• Should this patient undergo metastectomy?

An analysis of 3149 patients with STS treated at Memorial Sloan-Kettering who developed lung metastases found that patients with pulmonary metastases have survival rates of 25%. The most important prognostic factor for survival was complete resection of all metastases.⁴² For stage IV disease, surgery is used only in certain instances. In instances where tumor is more localized or limited, removal of metastases or metastectomy can play a role in management.²

CASE CONTINUED

Because the patient’s metastases are limited to the lungs, he is referred for metastectomy. He undergoes wedge resection for definitive diagnosis but it is not possible to completely resect all of the metastases. He is thus referred to a medical oncologist to discuss his treatment options.

• What are treatment options for unresectable or metastatic disease?

Metastatic Disease

Unlike local and locally advanced disease, chemotherapy forms the backbone of treatment in stage IV disease. Doxorubicin and olaratumab or doxorubicin and ifosfamide in combination are considered first line in metastatic disease. Response rates for single-agent doxorubicin range from 16% to 27%, while phase 2 and phase 3 studies of doxorubicin and ifosfamide have found response rates ranging from 18% to 36%.⁴³ In addition, the effectiveness of doxorubicin and ifosfamide phase 2 and 3 trials varied. Edmonson et al found a tumor regression rate of 34% for doxorubicin and ifosfamide as compared to 20% for doxorubicin alone.⁴⁴ In comparison, Santoro et al found a response rate of 21.3% for doxorubicin alone and 25.2% for doxorubicin and ifosfamide.⁴⁵ Neither study found increased survival benefit for doxorubicin and ifosfamide when compared to doxorubicin alone. In a Cochrane review evaluating randomized trials that compared doxorubicin and combination chemotherapy regimens, response rates varied from 14% for doxorubicin in combination with streptomycin to 34% for doxorubicin and ifosfamide. Most trials did not show a significant benefit for combination therapies when compared to doxorubicin alone.⁴³ Mean survival with doxorubicin or doxorubicin and ifosfamide is 12 months. High rates of recurrence highlight the need for additional chemotherapy regimens.

The newest approved agent is olaratumab, a monoclonal antibody that binds platelet-derived growth factor receptor alpha and prevents receptor activation. A phase 1-b and phase 2 trial evaluated patients with locally advanced and metastatic STS and randomly assigned them to either olaratumab and doxorubicin or doxorubicin alone.⁴⁶ Progression-free survival for olaratumab/doxorubicin was 6.6 months (95% CI 4.1 to 8.3) compared to 4.1 months (95% CI 2.8 to 5.4) for doxorubicin alone. The objective response rate was 18.2% (95% CI 9.8 to 29.6) for olaratumab/doxorubicin compared to 7.5% (95% CI 2.5 to 6.6) for doxorubicin alone. Furthermore, the median overall survival for olaratumab plus doxorubicin was 26.5 months (95% CI 20.9 to 31.7) compared to 14.7 months for doxorubicin alone (95% CI 5.5 to 26.0). Impressively, this improved response was notable across histological types. Furthermore, patients who had previously been treated with more than 1 regimen and those who were treatment naïve had similar response rates.⁴⁶

• What are second-line treatment options?

Doxorubicin has been used in combination with several other agents including dacarbazine (DTIC) as well as DTIC and ifosfamide (MAID). Borden et al evaluated patients with metastatic STS and randomly assigned the patients to either doxorubicin or doxorubicin and DTIC. Combination therapy demonstrated better tumor response than doxorubicin alone: 30% complete or partial response for combination therapy and 18% for doxorubicin alone.⁴⁷ However, Omura et al

Several additional regimens have undergone evaluation in metastatic and recurrent STSs. Gemcitabine has been used both as a single agent and as part of combination therapy in many studies. Studies with gemcitabine in combination with either docetaxel or DTIC have been the most efficacious. In a phase 2 trial, patients with metastatic STS were randomly assigned to either gemcitabine alone or gemcitabine and docetaxel. Combination therapy had a higher response rate (16% versus 8%) and longer overall survival (17.9 months versus 11.5 months) than gemcitabine alone.⁵⁰ Furthermore, a phase 2 trial of gemcitabine and docetaxel in patients with unresectable leiomyosarcoma showed an overall response rate of 56%, with 3 complete and 15 partial responses among the 34 patients enrolled in the study.⁵¹

A phase 2 trial randomly assigned patients with unresectable or metastatic STS to either DTIC or combination gemcitabine and DTIC.⁵² Gemcitabine-DTIC had a superior progression-free survival at 3 months (56% [95% CI 43% to 69%]) as compared to DTIC alone (37% [95% CI 23.5% to 50%]). Furthermore, mean progression-free survival and overall survival were improved in the gemcitabine-DTIC group (4.2 months and 16.8 months) as compared to the DTIC group (2.0 months and 8.2 months).⁵² DTIC has a single-agent response rate of 16%, but has been shown to be particularly effective in the setting of leiomyosarcomas.⁴⁹

• Does response to treatment regimens differ by histologic subtype?

The majority of STS trials include many different histologic subtypes. Given the rarity of sarcomas as a whole, many trials have had difficulty recruiting adequate numbers of patients to have sufficient power to definitely determine if the treatment under investigation has clinical benefit. Furthermore, the patients recruited have been heterogeneous with regard to subtype. Many older studies hypothesized that the efficacy of chemotherapeutic agents vary based on histologic subtype; however, for most subtypes the number of individuals included in those trials was too low to evaluate efficacy based on subtype.

Some exceptions exist, however. For example, both gemcitabine-DTIC and gemcitabine-docetaxel have been found to be particularly effective in the treatment of leiomyosarcomas.^50,52 Additionally, a retrospective study found a 51% overall response rate for patients with myxoid liposarcomas treated with trabectedin.⁵³ Studies of patients with angiosarcoma treated with paclitaxel have demonstrated response rates of 43% and 53%.^54,55

• What are the newest approved and investigational agents?

A recently approved agent is trabectedin, a tris tetrahydroisoquinoline alkaloid isolated from ascidians that binds to the minor groove of DNA and causes disruptions in the cell cycle. Samuels et al reported data from a single-arm, open-label expanded access trial that evaluated patients with advanced metastatic sarcomas.⁵⁶ In this study, patients with liposarcomas and leiomyosarcomas had an objective response rate of 6.9% (95% CI 4.8 to 9.6) as compared to a rate of 5.9% (95% CI 4.4 to 7.8) for all assessable patients. Median survival was 11.9 months for all patients, with improved median survivals for liposarcoma and leiomyosarcomas of 16.2 months (95% CI 14.1 to 19.5) compared to 8.4 months (95% CI 7.1 to 10.7 months) for other subtypes.⁵⁶

Schöffski et al evaluated eribulin, a chemotherapeutic agent that affects microtubule dynamics, in a phase 2 trial of patients with progressive or high-grade STS with progression on previous chemotherapy. They found a median progression-free survival of 2.6 months (95% CI 1.7 to 6.2) for adipocytic sarcoma, 2.9 months (95% CI 2.4 to 4.6) for leiomyosarcoma, 2.6 months (95% CI 2.3 to 4.3) for synovial sarcoma, and 2.1 months (95% CI 1.4 to 2.9) for other sarcomas.⁵⁷

Van der Graaf and colleagues randomly assigned patients with metastatic nonadipocytic STS to pazopanib or placebo in a phase 3 trial. Pazopanib is a small-molecule endothelial growth factor inhibitor with activity against vascular endothelial growth factors 1, 2, and 3 as well as platelet-derived growth factors. Median progression-free survival was 4.6 months (95% CI 3.7 to 4.8) with pazopanib compared to 1.6 months (95% CI 0.9 to 1.8) with placebo.⁵⁸ Adipocytic sarcomas (liposarcomas) were excluded from the trial because phase 2 trials had found a lower rate of progression-free survival (26%) for them compared to other subtypes.

Toxicities were seen with each of the regimens studied and were common in the randomized trials, with higher rates of toxicities in the combination chemotherapy regimens. The most common toxicities are myelosuppression, nausea, and vomiting. In the doxorubicin trials, the most common toxicities were myelosuppression, nausea, and vomiting.⁴⁴

Ifosfamide both as an individual agent and in combination with doxorubicin has higher rates and higher grades of toxicity than doxorubicin alone. Myelosuppression is the most common toxicity associated with ifosfamide, and the most commonly affected cell line is leukocytes.⁴⁴ Combination doxorubicin and ifosfamide also had high rates of nausea and vomiting (95%) and alopecia (100%).³⁵

Neutropenia is the most common toxicity associated with gemcitabine and dacarbazine, while their most common nonhematologic toxicities are fatigue and nausea.^52,59 Trabectedin’s most common toxicities are nausea (29%), neutropenia (24%), and fatigue (23%). It has also been shown to cause increased alkaline phosphatase (20%) and alanine aminotransferase (19%) levels.⁵⁶ In a phase 2 study of eribulin, 50% of patients had neutropenia, and other toxicities included fatigue, alopecia, nausea, sensory neuropathy, and thrombocytopenia.⁵⁷ Pazopanib is generally well tolerated; the most common toxicities are fatigue (65%), diarrhea (58%), nausea (54%), and hypertension (41%).⁵⁸ Higher rates of neutropenia, mucositis, nausea, vomiting, diarrhea, and transfusion reactions were seen with olaratumab and doxorubicin compared to doxorubicin alone in phase 1b and 2 studies.⁴⁶

CASE CONCLUSION

Given his poor prognosis with unresectable metastatic undifferentiated liposarcoma, the patient considers a clinical trial prior to undergoing combined therapy with doxorubicin and ifosfamide. He tolerates therapy well with stable disease at 6 months.

CONCLUSION

STSs are a heterogeneous collection of rare tumors. Low-grade, localized tumors have the best prognosis, and patients who undergo complete resection have the best long-term survival. Due to the rarity of STSs, trials often have limited enrollment, and little progress has been made with regards to treatment and survival rates for metastatic and unresectable disease. All patients should be evaluated and treated at specialized sarcoma centers. This case highlights the need for continued research and clinical trials to improve overall survival of patients with sarcoma.

INTRODUCTION

Sickle cell disease is the most common inherited blood disorder in the world. It affects more than 100,000 individuals in the United States, and millions more worldwide.¹ Sickle cell disease is most commonly found in individuals of African heritage, but the disease also occurs in Hispanics and people of Middle Eastern and subcontinent Indian heritage.² The distribution of the sickle hemoglobin (hemoglobin S [HbS]) allele overlaps with the distribution of malaria; HbS carriers, or individuals with sickle cell trait, have protection against malaria,³ and are not considered to have sickle cell disease.

Sickle cell disease is a severe monogenic disorder marked by significant morbidity and mortality, affecting every organ in the body.⁴ The term sickle cell disease refers to all genotypes that cause sickling; the most common are the homozygous hemoglobin SS (HbSS) and compound heterozygotes hemoglobin SC (HbSC), hemoglobin S–β⁰-thalassemia (HbSβ⁰),and hemoglobin S–β⁺-thalassemia(HbSβ⁺), although HbS and several rarer hemoglobin variants such as HbSO(Arab) and HbSD(Punjab) can also cause sickle cell disease.The term sickle cell anemia refers exclusively to the most severe genotypes, HbSS and HbSβ⁰.⁵ Common sickling genotypes along with their relative clinical severity are shown in Table 1.^6–11

This article reviews the pathophysiology of sickle cell disease, common clinical complications, and available therapies. A complex case which illustrates diagnostic and management challenges is presented as well.

PATHOPHYSIOLOGY

HbS is the result of a substitution of valine for glutamic acid in the sixth amino acid of the β-globin chain.¹² The change from a hydrophilic to a hydrophobic amino acid causes the hemoglobin molecules to stack, or polymerize, when deoxygenated. This rigid rod of hemoglobin distorts the cell, producing the characteristic crescent or sickle shape that gives the disease its name.¹³ Polymerization of hemoglobin within the cell is promoted by dehydration, which increases the concentration of HbS.^13,14 Polymerization occurs when hemoglobin is in the deoxygenated state.¹³

The sickle red blood cell is abnormal; it is rigid and dense, and lacks the deformability needed to navigate the microvasculature.¹⁵ Blockages of blood flow result in painful vaso-occlusion that is the hallmark of the disease, and that also can cause damage to the spleen, kidneys, and liver.¹⁶ The sickle red cell is also fragile, with a lifespan of only 20 days compared to the 120-day lifespan of a normal red blood cell.¹³ Frequent hemolysis results in anemia and the release of free hemoglobin, which both scavenges nitric oxide and impairs the production of more nitric oxide, which is essential for vasodilatation.¹⁷ This contributes to vascular dysfunction and an increased risk for stroke.¹⁸ If untreated, the natural course of sickle cell anemia is mortality in early childhood in most cases.¹⁹ Common chronic and acute sickle cell disease–related complications and recommended therapies, based on 2014 National Institutes of Health guidelines, are shown in Table 2 and Table 3.²⁰

One of the most challenging aspects of sickle cell disease is its clinical variability. While in general, HbSS and HbSβ⁰ are the most severe genotypes, there are patients with HbSC and HbSb⁺ who have significant sickle-cell–related complications, and may have a more severe clinical course than a HbSS patient.²¹ A great deal of this clinical variability cannot be explained, but some can be attributed to endogenous fetal hemoglobin (HbF) levels.^22–24 The importance of HbF levels in sickle cell disease was first noted by a pediatrician in the 1940s.²⁵ She observed that sickle cell disease complications in children under the age of 1 were rare, and attributed it to the presence of HbF.²⁵ HbF levels decline more slowly in individuals with hemoglobinopathies, reaching their nadir after the age of 5 rather than within 6 months of birth in individuals without hemoglobinopathies.²⁶ HbF levels remain elevated lifelong in most sickle cell disease patients, especially those with the HbSS and HbSβ⁰ genotypes. Levels of HbF vary widely between individuals, from zero to 20% to 30%, with a median of 10%.^26–28 Individuals who produce more HbF have a milder course, in general.²⁴ An association between the 4 β-globin haplotypes and HbF levels has been reported in the past,^27,29 but more sophisticated next-generation sequencing has revealed causal variants in BCL11A and HBS1L-MYB that contribute approximately 50% of the observed variability in HbF levels.^30–33

Co-inheritance of α-thalassemia also modifies disease course; less available α-globin chains results in a lower hemoglobin concentration within the cell. Paradoxically, this results in a higher overall hemoglobin level, as there is a reduction in polymerization, and therefore sickling due to lower HbS concentrations in the cell. Patients therefore are less anemic, reducing the risk of stroke in childhood,^34,35 but blood viscosity may be higher, resulting in more frequent pain crises and increased risk³⁶ of avascular necrosis.^34,35,37 It is often helpful to think of sickle cell patients as falling into 1 of 2 groups: high hemolysis/low hemoglobin and high viscosity/high hemoglobin. Individuals with high rates of hemolysis are at greater risk for stroke, pulmonary hypertension, and acute chest syndrome (ACS). Higher rates of hemolysis result in higher levels of free hemoglobin, which scavenges nitric oxide. This leads to the vascular damage and dysfunction that contributes to the associated clinical complications. This phenotype is most commonly seen in HbSS and HbSβ⁰.³⁸ High hemoglobin/high viscosity phenotypes are most often found in HbSC patients and in sickle cell anemia with α-thalassemia coinheritance.^39–42

TREATMENT OPTIONS

In high-resource countries with newborn screening, the initiation of penicillin prophylaxis has dramatically altered the natural history of the disease, allowing the majority of patients to reach adulthood.⁴³ Penicillin prophylaxis is usually discontinued at age 5 years; however, individuals who have undergone surgical splenectomy or have had pneumococcal sepsis on penicillin prophylaxis may remain on penicillin to age 18 or beyond.²⁰

Another advance in sickle cell care is screening for stroke risk through transcranial Doppler ultrasound (TCD).^44–47 This screening tool has reduced the incidence of childhood stroke from 10% by age 11 to 1%. TCDs typically cannot be performed after the age of 16 due to changes in the skull. Individuals found to have abnormal (elevated) TCD velocities are placed on chronic transfusion therapy for primary stroke prevention. They may remain on monthly chronic transfusions, with the goal of suppressing the percentage of HbS to 30% to 50% indefinitely. A clinical trial (STOPII) designed to determine if pediatric sickle cell disease patients on chronic transfusion therapy for primary stroke prevention could be safely taken off transfusion therapy was discontinued early due to an excess of strokes and conversion to abnormal TCD velocities in the untransfused arm.⁴⁴ Individuals who have experienced an ischemic stroke have a 70% risk of another stroke, and must remain on chronic transfusion therapy indefinitely. Chronic transfusion reduces their stroke risk to 13%.

The only widely used pharmacologic therapy for sickle cell disease is hydroxyurea.^12,48–50 A significant portion of the benefit of hydroxyurea stems from its induction of HbF.⁵¹ HbF does not sickle, and it interrupts the polymerization of HbS in the cell, if present in high enough concentrations.⁵⁰ The level of HbF needed to achieve clinical improvement is not known, but in vitro assays suggest 20% HbF is needed to prevent sickling.^52,53 However, endogenous and hydroxyurea-induced HbF is not distributed evenly through the red cells, so sickling is possible regardless of the level of HbF induced.^54,55 Hydroxyurea likely has other disease-modifying effects as well, including reduction of white blood cell count and reticulocyte count and reduction of red cell adhesion to the endothelium.^56–58 Clinical criteria for initiation of hydroxyurea in adult sickle cell disease patients are shown in Table 4.²⁰ Hydroxyurea is given daily and is dosed to maximum tolerated dose for the individual by following the absolute neutrophil count (ANC). The goal ANC is between 2000 and 4000/µL. At times, absolute reticulocyte count (ARC) can be dose-limiting; goal ARC is greater than 70,000/µL.⁵⁹ Platelet counts may be reduced as well, especially in HbSC patients.^60,61

The only curative therapy for sickle cell disease is hematopoietic stem cell transplant.⁶² Transplant use is limited by availability of matched sibling donors,⁶² and even at experienced centers transplant carries a small risk for mortality, graft rejection, and graft-versus-host disease. Furthermore, consensus on disease complications for which transplant is recommended is also lacking.^63–65 Clinical trials of gene therapy for sickle cell disease and thalassemia are ongoing.⁶⁶

COMPLICATIONS AND DISEASE-SPECIFIC THERAPIES

CASE PRESENTATION

A 26-year-old African-American man who works as a school bus driver presents to an academic center’s emergency department complaining of pain in his left leg, similar to prior pain events. He is described as having sickle cell trait, although no hemoglobin profile is available in his chart. He describes the pain as dull and aching, 10/10 in intensity. A complete blood count (CBC) is obtained; it reveals a hemoglobin of 14.5 g/dL, white blood cell (WBC) count of 5600/µL, and platelet count of 344,000/µL. His CBC is also notable for a mean corpuscular volume (MCV) of 72 fL, a mean corpuscular hemoglobin concentration (MCHC) of 37 g/dL, and a red blood cell distribution width (RDW) of 12. Slide review of a peripheral blood smear shows 2+ target cells (Figure).

The patient is given 6 mg of morphine, which provides some relief of his pain, and is discharged with a prescription for hydrocodone bitartrate/acetaminophen 5/325 mg. The diagnosis given is musculoskeletal pain, and he is instructed to follow-up with a primary care physician. His past medical history is significant for 4 or 5 visits to the emergency department per year in the past 4 years. Prior to 4 years ago, he rarely required medical attention.

• What laboratory and clinical features might lead you to question the diagnosis of sickle cell trait in this patient?

The patient’s hemoglobin is within normal range, which is consistent with sickle cell trait; however, he is microcytic, with a normal RDW. It is possible to be mildly microcytic in the early stages of iron deficiency, prior to the development of anemia, but the RDW would typically be elevated, demonstrating the presence of newer, smaller cells produced under conditions of iron deficiency.⁶⁷ It is also possible that his microcytosis with a normal RDW could represent sickle cell trait with co-inheritance of β-thalassemia. Up to 30% of African Americans have β-thalassemia,² and 1 in 10 have sickle cell trait.⁶⁸ However, a high MCHC, indicating the presence of dense cells, and target cells noted on slide review are most consistent with HbSC.⁹ HbSC patients, especially males, can have hemoglobin levels in the normal range.⁴ The biggest inconsistency with the diagnosis of sickle cell trait is his history of frequent pain events. Individuals with sickle cell trait rarely present with pain crises, except under extreme conditions of dehydration or high altitude.⁶⁸ Sickle cell trait is generally regarded as a benign condition, although a study of U.S. military recruits found a 30-fold higher risk of sudden death during basic training in persons with sickle cell trait.⁶⁹ Additional sickle cell trait–related complications include hematuria, risk of splenic sequestration or infarct under extreme conditions and high altitude, and a rare and usually fatal renal malignancy, renal medullary carcinoma, which is vanishingly rare in individuals without sickle cell trait.^70,71 Although the patient reported having sickle cell trait, this diagnosis should have been verified with a hemoglobin panel, given his atypical presentation.²⁰

In sickle cell disease, vaso-occlusive pain events can be common, often beginning in early childhood.¹⁷ This disease complication accounts for 95% of all adult sickle cell disease hospitalizations.⁷² There is a great deal of variability in pain symptoms between individuals, and within individuals at various times in their lives:⁷³ 30% have no pain events, 50% have occasional events, and 20% have monthly or more frequent events that require hospitalization.⁷⁴ The frequency and severity of pain events are modulated by HbF levels, β-thalassemia status, genotypes, therapies like hydroxyurea, or in rare cases, chronic transfusion therapy.²³ Personal factors, such as psychosocial stressors, also contribute to the frequency of pain events.⁷⁵ Pain event triggers include exposure to cold water, windy or cold weather, temperature changes, and extreme temperatures.^76–79 Patient age also contributes to pain event frequency. Many patients see an increase in pain event frequency in their late 20s, and a marked decrease in their 40s.^23,73 More than 3 pain events per year is associated with reduced life expectancy.²³

Acute management of pain episodes involves nonsteroidal anti-inflammatory drugs, oral opioids, and when hospitalization is required, intravenous opioids, often delivered via patient-controlled analgesia (PCA) pumps.⁷⁹ As sickle cell disease patients become teenagers and young adults, some experience an increased frequency of pain episodes, with fewer pain-free days, or a failure to return to baseline before the next pain crisis occurs.^80,81 This is characteristic of emerging chronic pain.⁸² Chronic pain is a significant problem in adult patients with sickle cell disease, with up to 85% reporting pain on most days.^72,80 The development of chronic pain may be reduced by early and aggressive treatment of acute pain events, as well as use of hydroxyurea to reduce the number of pain events. Many adult sickle cell patients with chronic pain are treated with daily opioids.²⁰ Given the significant side effects of chronic opioid use—sedation, respiratory depression, itching, nausea, and impairment of function and quality of life—non-opioid therapies are under investigation.⁸³ Many chronic pain patients have symptoms of neuropathic pain, and may benefit from neuropathic agents like gabapentin, both to reduce opioid use and to more effectively treat chronic neuropathic pain, which is known to respond poorly to opioids.^84–86

• Is the patient’s peripheral blood smear consistent with a diagnosis of sickle cell trait?

Several target cells are visible, which is not typical of sickle cell trait, but may be seen in HbSC or thalassemia. The finding of an intracellular crystal is pathognomonic for HbSC or HbCC. HbC polymerizes in high oxygen conditions, opposite of HbS, which polymerizes in low oxygen conditions.⁹

CASE CONTINUED

The patient’s family history is significant for a sister who died at age 3 from sickle cell–related complications, and a sister with sickle cell trait who had a cholecystectomy for gallstones at age 22. His father died at age 38 due to unknown causes. The sickle cell trait status of his parents is unknown. His mother is alive, and has hypertension.

• Is the medical history of this patient’s family members consistent with sickle cell trait?

It is unlikely that sickle cell trait would result in early death in childhood, or in gallstones at age 22. Gallstones in early adulthood is a common presentation for HbSC patients not diagnosed by newborn screening.⁸⁷ Any hemolytic condition can lead to the formation of hemoglobin-containing pigmented gallstones, biliary sludge, and obstruction of the gallbladder. In the presence of right-sided abdominal pain, a serum bilirubin level of more than 4 mg/dL should lead to measurement of direct bilirubin; if greater than 10% of total, imaging of the gallbladder should be obtained. In sickle cell disease, 30% of patients will have gallstones by 18 years of age. The low hemolysis/high viscosity phenotype patients are typically older at diagnosis. Co-inheritance of Gilbert syndrome and sickle cell disease is not uncommon, and can result in formation of gallstones at a young age; Gilbert syndrome alone typically results in gallstones in mid-life.⁸⁸

Two months later, the patient presents again to the emergency department with the same complaint of leg pain, as well as abdominal pain. His hemoglobin is 12.5 g/dL, and his platelet count is 134,000/µL. His pain is not improved with 3 doses of morphine 6 mg intravenously, and he is admitted to the medicine service. A hemoglobin profile is obtained, revealing 52% HbS, 45% HbC, and 1.5% HbF, consistent with HbSC. In sickle cell trait, the hemoglobin profile is 60% HbA and 40% HbS (available α-globin prefers to pair with a normal β-globin, so the ratio of HbA to HbS is 60:40, not 50:50).

On the second hospital day, the patient’s hemoglobin drops to 7.2 g/dL and his platelet count decreases to 44,000/µL. His abdomen is distended and diffusely tender. The internist transfuses him with 2 units of packed red blood cells (PRBC), after which his hemoglobin increases to 11 g/dL, while his platelet count increases to 112,000/µL. Following the transfusion, his abdominal pain resolves, as does his anemia and thrombocytopenia.

• What caused this patient’s anemia and thrombocytopenia?

High on the differential diagnosis is a splenic sequestration. Acute splenic sequestration occurs when red cells are trapped in the splenic sinuses. Massive splenic enlargement may occur over several hours.^89,90 Unrecognized splenic sequestration has a high mortality rate from severe anemia and splenic rupture.⁹⁰ Splenic sequestration must be ruled out in a sickle cell patient with abdominal pain accompanied by dropping platelet and red cell counts, especially in milder subtypes that often have splenic function preserved into adolescence and adulthood. Sickle cell anemia patients usually become functionally asplenic in early childhood.^89,91,92 The rise in hemoglobin, more than would be expected from 2 units of PRBC, plus the improvement in platelet count without a platelet transfusion observed in the case patient strongly supports the diagnosis of splenic sequestration.

Splenic sequestration can occur in any sickle cell patient whose spleen has not fibrosed. Splenic sequestration in adulthood is not uncommon in HbSC patients, who often have preserved splenic function into adulthood.^93–95

Clinical signs of splenic sequestration include a rapid drop in hemoglobin, rise in reticulocyte count, a tender, enlarged spleen, and, in severe cases, hypovolemia.^89,93 It is treated with prompt blood transfusion, but care must be taken not to overtransfuse the patient, as the spleen can trap several grams of hemoglobin, which may be released upon transfusion, potentially causing life-threatening hyperviscosity.⁸⁹ Hemoglobin levels must be checked following transfusion in suspected splenic sequestration, and “mini transfusions” of 5 mL/kg are recommended in sickle cell disease patients who are hemodynamically stable.²⁰

Hepatic sequestration may also occur, but it is much less common than splenic sequestration.⁹⁶ Other conditions on the differential diagnosis include thrombotic thrombocytopenic purpura, which would be unlikely to respond to a transfusion. ACS can cause a drop in hemoglobin, and is treated with simple or exchange transfusions.⁹⁷ ACS is less likely without respiratory symptoms or oxygen requirement, and usually is not associated with thrombocytopenia. Sepsis may also cause anemia and thrombocytopenia, but again would not likely respond to a simple transfusion. The patient’s response to transfusion is consistent with a sequestering event, not a destructive event as in the case of sepsis.

CASE CONTINUED

Imaging reveals a grossly enlarged spleen, which is having a mass effect on the left kidney. The patient is started on hydroxyurea therapy at 500 mg 3 times daily. Discharge instructions include following up with his primary care physician, continuing hydroxyurea therapy, and receiving yearly dilated eye exams to evaluate for proliferative sickle retinopathy.

• Are these discharge instructions complete?

Splenic sequestration has a 50% recurrence rate.⁹⁸ In very young children, watchful waiting or chronic transfusion may be implemented to preserve the immunologic function of the spleen and reduce the risk of sepsis.⁸⁹ Splenectomy after a single episode of sequestration in adults is a matter of debate, with experts advising both watchful waiting⁹⁹ and splenectomy after recovery from the first sequestering event.¹⁰⁰ The patient should have been informed of the risk for recurrence, and the signs and symptoms of splenic sequestration as well as the need for emergency medical attention should have been discussed. Splenic sequestration may be milder in adults than in children, but fatal sequestrations have been reported.^95,101–103

Proliferative sickle cell retinopathy is a high viscosity/high hemoglobin complication that may occur more frequently in HbSC than HbSS, with an incidence of 33% in HbSC.^42,104 Spontaneous regression of retinopathy occurs in approximately 32% of eyes, and laser or scatter photocoagulation is an effective intervention.¹⁰⁵

• Would the patient need to be transfused prior to splenectomy?

Preoperative transfusion therapy is standard of care for HbSS patients undergoing general anesthesia. The TRAP study found that simple “top off” transfusion to a hemoglobin of 10 g/dL was as effective at preventing postoperative sickle cell–related complications as exchange transfusion to HbS of 30% or less, and had fewer transfusion-related complications like alloimmunization.¹⁰⁶ There is little data regarding preoperative transfusions in HbSC disease. A retrospective study suggests that HbSC patients undergoing abdominal surgeries should be transfused.¹⁰⁷ The higher hemoglobin level of the typical HbSC patient necessitates exchange transfusion to avoid hyperviscosity.

• Is hydroxyurea therapy indicated in this patient?

• Has it been dosed appropriately?

If the patient had the HbSS subtype, hydroxyurea would be clearly indicated, given his frequent pain events.²⁰ HbSC patients may be placed on hydroxyurea on a case-by-case basis, but evidence for its efficacy in this sickle cell subtype is lacking.¹⁰⁸ Large clinical trials like the Multi-Center Study of Hydroxyurea (MSH) that established the safety and efficacy of hydroxyurea in sickle cell anemia excluded HbSC and HbSβ⁺ patients.¹⁰⁹ These mild to moderate subtypes produce less HbF at baseline, and typically have a minimal to modest rise in HbF on hydroxyurea.¹¹⁰ In sickle cell anemia, hydroxyurea is titrated to maximum tolerated dose, defined as an ANC of 2000 to 4000/µL and an ARC of 70,000/µL or higher.⁵³ Because of their lower levels of chronic inflammation and lower reticulocyte counts due to higher hemoglobin levels, many HbSC and HbSβ⁺ patients have values in that range before initiating hydroxyurea therapy.⁹ Cytopenias, particularly of platelets in HbSC, occur at low doses of hydroxyurea.¹¹¹

Of note, although the half-life of hydroxyurea would suggest that 3 times daily dosing is indicated, daily dosing has been found to have equal response and is preferred. Another concern is the monitoring of this myelosuppressive medication. This patient has repeatedly failed to obtain a primary care physician or a hematologist, and hydroxyurea requires laboratory monitoring at least every 2 months, especially in a HbSC patient with a very large spleen who is at significant risk for thrombocytopenia and neutropenia.⁹

CASE CONTINUED

A week after discharge from his admission for abdominal pain diagnosed as splenic sequestration, the patient presents again to the emergency department with abdominal pain which he reports is his typical sickle cell pain. Hemoglobin is 13.8 g/dL, platelet count is 388,000/µL, and alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels are both 10 times their prior value. Creatinine is 1.2 mg/dL (0.75 mg/dL on his prior admission), and total bilirubin is 3 mg/dL, with 0.3 mg/dL direct bilirubin. He undergoes an ultrasound exam of his gallbladder, which reveals sludge and a possible gallstone. There is no evidence of cholecystitis. General surgery performs a laparoscopic cholecystectomy.

• Was this cholecystectomy necessary?

In patients with sickle cell disease, symptomatic gallstones and gallbladder sludge should be observed; recurrent abdominal pain without a significant change in bilirubin may not be due to gallstones or sludge, and therefore may not be relieved by cholecystectomy.^112,113 In sickle cell disease, 40% of patients with gallbladder sludge do not develop gallstones.⁸⁷ The patient’s bilirubin level was at baseline, and there was no increase in the direct (conjugated) fraction. Watchful waiting would have been appropriate, with cholecystectomy being performed if he experienced recurrent symptoms associated with fatty foods accompanied by an elevation in direct bilirubin.

More concerning and deserving of investigation was his elevated liver enzymes. Patients with sickle cell disease may experience recurrent ischemia and reperfusion injuries in the liver, which is called right upper quadrant syndrome. On autopsy of 70 sickle cell patients, 91% had hepatomegaly and 34% had focal necrosis.¹¹⁴ AST is often elevated in sickle cell disease, as it is affected by hemolysis. In this patient, both AST and ALT are elevated, consistent with a hepatocellular disorder. His abdominal pain and ALT rise may be a sign of a hepatic crisis.¹¹⁵ Rapid resolution of ALT elevation in a matter of days suggests a vaso-occlusive, inflammatory event that is self- limiting. Prolonged AST elevation requires further investigation, with consideration of autoimmune hepatitis, viral hepatitis, or iron overload. Iron overload is unlikely in this patient given his lifetime history of only 1 transfusion. Hepatic iron overload typically occurs in sickle cell disease after a minimum of 10 transfusions.¹¹⁵

CASE CONTINUED

The patient is discharged on the day after the procedure, with instructions to continue his hydroxyurea.

• Should the patient resume hydroxyurea therapy?

Hydroxyurea is hepatically cleared and thus it should be held until his liver function tests normalize.¹⁰⁶

CASE CONTINUED

Two months later, the patient presents to the emergency department with abdominal pain that moves to his left leg. A CBC is obtained, showing a hemoglobin of 11.8 g/dL and a platelet count of 144,000/µL. He is given 2 doses of morphine 6 mg intravenously, and reports that his leg pain is now a 4/10. He is discharged home with a prescription for hydrocodone/acetaminophen.

• Is the emergency department evaluation sufficient?

This patient remains at high risk for splenic sequestration,⁹³ with a hemoglobin 2 g lower than it was 2 months ago and platelets less than half. This decline could be consistent with early splenic sequestration.²⁰ Additionally, he had elevated liver function tests on a recent admission, as well as rising creatinine, without evidence of resolution. It is not appropriate to discharge him without checking a chemistry and liver panel, and abdominal imaging should be considered. The best plan would be to admit him for observation, given his risk for splenic sequestration, and consult surgery for an elective splenectomy if he has a second episode of splenic sequestration 2 months after the first.¹⁰⁰ His abdominal pain that migrates to his left leg could be due to his massive splenomegaly compressing his left kidney, as noted on imaging during his recent admission for splenic sequestration

CASE CONTINUED

An hour after discharge from the emergency department, EMS is called to his home for intractable pain. He is found lying on the floor, and reports excruciating left leg pain. He is brought to the closest hospital, a community hospital that he has not visited previously. There, he is admitted for hydration and pain control and placed on hydromorphone 2 mg every 4 hours as needed for pain. His hemoglobin is 10.8 g/dL, and platelets are 121,000/µL. A chemistry panel is remarkable for a creatinine level of 1.5 mg/dL and a potassium level of 3.2 mEq/L. Liver function tests are not obtained. After 3 doses of hydromorphone, he falls asleep. He is not in a monitored bed, and intravenous fluids, while ordered, are not started. At 6:30 AM the day after admission, he cannot be aroused on a routine vital sign check; he has an SpO₂ of 60%, a blood pressure of 80/60 mm Hg, and heart rate of 148 beats/min. A rapid response is called, and naloxone is administered along with oxygen by face mask and several fluid boluses. His systolic blood pressure increases to 100 mm Hg from a low of 70 mm Hg. His SpO₂ increases to 92%, and he is arousable and alert, although he reports 10/10 leg pain. His abdomen is noted to be distended and tender.

• What may have contributed to his clinical condition?

The patient is opioid tolerant and has received equivalent doses of opioids in the past without excess sedation. He may have liver dysfunction making him unable to metabolize opioids effectively. His hemoglobin and platelets continue to decline, raising concern for splenic sequestration versus sepsis. Failure to place him on a monitor allowed his hypoxia to continue for an unknown amount of time, placing him at high risk for developing ACS. Lack of intravenous hydration while he was too sedated to drink likely exacerbated his sickling.

At 9:20 AM, a CBC is obtained and reveals a hemoglobin of 4.8 g/dL and a platelet count of 44,000/µL. Two units of stat O negative blood are administered, and preparations are made to administer an exchange transfusion. A liver panel is obtained 3 hours later, which reveals an AST level of 1200 U/L and an ALT level of 1050 U/L. His bilirubin is 10 mg/dL, and his lactate dehydrogenase level is 1800 U/L. His urine is dark and is positive for bilirubin and ketones. He is transferred to the intensive care unit. A chest X-ray shows pulmonary congestion. Hematology/oncology is consulted.

He receives a 7-unit red blood cell exchange, which reduces his HbS to 11%. He continues to be hypotensive, and requires norepinephrine to support his blood pressure. Antibiotic therapy is started. His creatinine concentration rises to 2.3 mg/dL, potassium is 7.8 mEq/L, and bicarbonate is 12 mEq/L. He is placed on hemodialysis.

Computed tomography of the chest and abdomen reveals lower posterior lung infiltrates and a grossly enlarged spleen. He requires intubation. He is given a diagnosis of ACS in addition to kidney failure, liver failure, and “sickle crisis.” He continues to require daily to twice daily transfusions to maintain a hemoglobin of 7 to 9 g/dL, and his abdominal distension increases. As his condition worsens, surgery is consulted to discuss a liver transplant. He is deemed to not be a surgical candidate, and he passes away 6 days after entering the hospital. The immediate cause of death is listed as vaso-occlusive crisis, with ACS and sickle crisis listed as contributors.

• Are the causes of death accurate and complete?

If vaso-occlusive crisis is used to indicate a pain event, it is not an accurate cause of death. Pain is one of the most distressing complications of sickle cell disease, and frequent pain events are associated with early mortality,^4,80 but they are not in themselves fatal. ACS is the number one cause of death in sickle cell disease,⁴ and it likely contributed to this patient’s death. Sickle crisis is a vague term that should not be used in this context. Causes of death should include splenic sequestration and multisystem organ failure. Multisystem organ failure in sickle cell disease often responds to aggressive transfusion therapy, which this patient received.^116–118

CONCLUSION

Sickle cell disease is a complex chronic disease that impacts almost every organ system in the body. Clinicians may be inclined to attribute most pain in a patient with sickle cell disease to a simple vaso-occlusive crisis, treat them for this, and not investigate further. As the case presented here demonstrates, failure to identify the actual life-threatening process occurring in a patient with sickle cell disease presenting with pain can result in preventable early mortality. Clinicians must approach a sickle cell patient reporting pain in a thoughtful manner, and consider a complete differential diagnosis, including both sickle cell disease complications and those unrelated to sickle cell disease. Knowledge of the disease courses of the different sickle cell genotypes is essential, and must go beyond a superficial hierarchy of severity, but rather include an understanding of the complications each genotype is most prone to, and at what ages. Complete laboratory assessment, including a comprehensive metabolic panel, should be performed on all admitted patients, not just a complete blood count. Treating pain with high-dose opioids, while appropriate in an uncomplicated pain crisis, can lead to ACS or even respiratory failure in a patient with uninvestigated liver and kidney dysfunction. The most important lesson to remember is that even the sickle cell disease patient who has been given the unfortunate and pejorative label of “frequent flyer” by some providers has the potential for rapid deterioration into multisystem organ failure and death.

INTRODUCTION

Sickle cell disease is the most common inherited blood disorder in the world. It affects more than 100,000 individuals in the United States, and millions more worldwide.¹ Sickle cell disease is most commonly found in individuals of African heritage, but the disease also occurs in Hispanics and people of Middle Eastern and subcontinent Indian heritage.² The distribution of the sickle hemoglobin (hemoglobin S [HbS]) allele overlaps with the distribution of malaria; HbS carriers, or individuals with sickle cell trait, have protection against malaria,³ and are not considered to have sickle cell disease.

Sickle cell disease is a severe monogenic disorder marked by significant morbidity and mortality, affecting every organ in the body.⁴ The term sickle cell disease refers to all genotypes that cause sickling; the most common are the homozygous hemoglobin SS (HbSS) and compound heterozygotes hemoglobin SC (HbSC), hemoglobin S–β⁰-thalassemia (HbSβ⁰),and hemoglobin S–β⁺-thalassemia(HbSβ⁺), although HbS and several rarer hemoglobin variants such as HbSO(Arab) and HbSD(Punjab) can also cause sickle cell disease.The term sickle cell anemia refers exclusively to the most severe genotypes, HbSS and HbSβ⁰.⁵ Common sickling genotypes along with their relative clinical severity are shown in Table 1.^6–11

This article reviews the pathophysiology of sickle cell disease, common clinical complications, and available therapies. A complex case which illustrates diagnostic and management challenges is presented as well.

PATHOPHYSIOLOGY

HbS is the result of a substitution of valine for glutamic acid in the sixth amino acid of the β-globin chain.¹² The change from a hydrophilic to a hydrophobic amino acid causes the hemoglobin molecules to stack, or polymerize, when deoxygenated. This rigid rod of hemoglobin distorts the cell, producing the characteristic crescent or sickle shape that gives the disease its name.¹³ Polymerization of hemoglobin within the cell is promoted by dehydration, which increases the concentration of HbS.^13,14 Polymerization occurs when hemoglobin is in the deoxygenated state.¹³

The sickle red blood cell is abnormal; it is rigid and dense, and lacks the deformability needed to navigate the microvasculature.¹⁵ Blockages of blood flow result in painful vaso-occlusion that is the hallmark of the disease, and that also can cause damage to the spleen, kidneys, and liver.¹⁶ The sickle red cell is also fragile, with a lifespan of only 20 days compared to the 120-day lifespan of a normal red blood cell.¹³ Frequent hemolysis results in anemia and the release of free hemoglobin, which both scavenges nitric oxide and impairs the production of more nitric oxide, which is essential for vasodilatation.¹⁷ This contributes to vascular dysfunction and an increased risk for stroke.¹⁸ If untreated, the natural course of sickle cell anemia is mortality in early childhood in most cases.¹⁹ Common chronic and acute sickle cell disease–related complications and recommended therapies, based on 2014 National Institutes of Health guidelines, are shown in Table 2 and Table 3.²⁰

One of the most challenging aspects of sickle cell disease is its clinical variability. While in general, HbSS and HbSβ⁰ are the most severe genotypes, there are patients with HbSC and HbSb⁺ who have significant sickle-cell–related complications, and may have a more severe clinical course than a HbSS patient.²¹ A great deal of this clinical variability cannot be explained, but some can be attributed to endogenous fetal hemoglobin (HbF) levels.^22–24 The importance of HbF levels in sickle cell disease was first noted by a pediatrician in the 1940s.²⁵ She observed that sickle cell disease complications in children under the age of 1 were rare, and attributed it to the presence of HbF.²⁵ HbF levels decline more slowly in individuals with hemoglobinopathies, reaching their nadir after the age of 5 rather than within 6 months of birth in individuals without hemoglobinopathies.²⁶ HbF levels remain elevated lifelong in most sickle cell disease patients, especially those with the HbSS and HbSβ⁰ genotypes. Levels of HbF vary widely between individuals, from zero to 20% to 30%, with a median of 10%.^26–28 Individuals who produce more HbF have a milder course, in general.²⁴ An association between the 4 β-globin haplotypes and HbF levels has been reported in the past,^27,29 but more sophisticated next-generation sequencing has revealed causal variants in BCL11A and HBS1L-MYB that contribute approximately 50% of the observed variability in HbF levels.^30–33

Co-inheritance of α-thalassemia also modifies disease course; less available α-globin chains results in a lower hemoglobin concentration within the cell. Paradoxically, this results in a higher overall hemoglobin level, as there is a reduction in polymerization, and therefore sickling due to lower HbS concentrations in the cell. Patients therefore are less anemic, reducing the risk of stroke in childhood,^34,35 but blood viscosity may be higher, resulting in more frequent pain crises and increased risk³⁶ of avascular necrosis.^34,35,37 It is often helpful to think of sickle cell patients as falling into 1 of 2 groups: high hemolysis/low hemoglobin and high viscosity/high hemoglobin. Individuals with high rates of hemolysis are at greater risk for stroke, pulmonary hypertension, and acute chest syndrome (ACS). Higher rates of hemolysis result in higher levels of free hemoglobin, which scavenges nitric oxide. This leads to the vascular damage and dysfunction that contributes to the associated clinical complications. This phenotype is most commonly seen in HbSS and HbSβ⁰.³⁸ High hemoglobin/high viscosity phenotypes are most often found in HbSC patients and in sickle cell anemia with α-thalassemia coinheritance.^39–42

TREATMENT OPTIONS

In high-resource countries with newborn screening, the initiation of penicillin prophylaxis has dramatically altered the natural history of the disease, allowing the majority of patients to reach adulthood.⁴³ Penicillin prophylaxis is usually discontinued at age 5 years; however, individuals who have undergone surgical splenectomy or have had pneumococcal sepsis on penicillin prophylaxis may remain on penicillin to age 18 or beyond.²⁰

Another advance in sickle cell care is screening for stroke risk through transcranial Doppler ultrasound (TCD).^44–47 This screening tool has reduced the incidence of childhood stroke from 10% by age 11 to 1%. TCDs typically cannot be performed after the age of 16 due to changes in the skull. Individuals found to have abnormal (elevated) TCD velocities are placed on chronic transfusion therapy for primary stroke prevention. They may remain on monthly chronic transfusions, with the goal of suppressing the percentage of HbS to 30% to 50% indefinitely. A clinical trial (STOPII) designed to determine if pediatric sickle cell disease patients on chronic transfusion therapy for primary stroke prevention could be safely taken off transfusion therapy was discontinued early due to an excess of strokes and conversion to abnormal TCD velocities in the untransfused arm.⁴⁴ Individuals who have experienced an ischemic stroke have a 70% risk of another stroke, and must remain on chronic transfusion therapy indefinitely. Chronic transfusion reduces their stroke risk to 13%.

The only widely used pharmacologic therapy for sickle cell disease is hydroxyurea.^12,48–50 A significant portion of the benefit of hydroxyurea stems from its induction of HbF.⁵¹ HbF does not sickle, and it interrupts the polymerization of HbS in the cell, if present in high enough concentrations.⁵⁰ The level of HbF needed to achieve clinical improvement is not known, but in vitro assays suggest 20% HbF is needed to prevent sickling.^52,53 However, endogenous and hydroxyurea-induced HbF is not distributed evenly through the red cells, so sickling is possible regardless of the level of HbF induced.^54,55 Hydroxyurea likely has other disease-modifying effects as well, including reduction of white blood cell count and reticulocyte count and reduction of red cell adhesion to the endothelium.^56–58 Clinical criteria for initiation of hydroxyurea in adult sickle cell disease patients are shown in Table 4.²⁰ Hydroxyurea is given daily and is dosed to maximum tolerated dose for the individual by following the absolute neutrophil count (ANC). The goal ANC is between 2000 and 4000/µL. At times, absolute reticulocyte count (ARC) can be dose-limiting; goal ARC is greater than 70,000/µL.⁵⁹ Platelet counts may be reduced as well, especially in HbSC patients.^60,61

The only curative therapy for sickle cell disease is hematopoietic stem cell transplant.⁶² Transplant use is limited by availability of matched sibling donors,⁶² and even at experienced centers transplant carries a small risk for mortality, graft rejection, and graft-versus-host disease. Furthermore, consensus on disease complications for which transplant is recommended is also lacking.^63–65 Clinical trials of gene therapy for sickle cell disease and thalassemia are ongoing.⁶⁶

COMPLICATIONS AND DISEASE-SPECIFIC THERAPIES

CASE PRESENTATION

A 26-year-old African-American man who works as a school bus driver presents to an academic center’s emergency department complaining of pain in his left leg, similar to prior pain events. He is described as having sickle cell trait, although no hemoglobin profile is available in his chart. He describes the pain as dull and aching, 10/10 in intensity. A complete blood count (CBC) is obtained; it reveals a hemoglobin of 14.5 g/dL, white blood cell (WBC) count of 5600/µL, and platelet count of 344,000/µL. His CBC is also notable for a mean corpuscular volume (MCV) of 72 fL, a mean corpuscular hemoglobin concentration (MCHC) of 37 g/dL, and a red blood cell distribution width (RDW) of 12. Slide review of a peripheral blood smear shows 2+ target cells (Figure).

The patient is given 6 mg of morphine, which provides some relief of his pain, and is discharged with a prescription for hydrocodone bitartrate/acetaminophen 5/325 mg. The diagnosis given is musculoskeletal pain, and he is instructed to follow-up with a primary care physician. His past medical history is significant for 4 or 5 visits to the emergency department per year in the past 4 years. Prior to 4 years ago, he rarely required medical attention.

• What laboratory and clinical features might lead you to question the diagnosis of sickle cell trait in this patient?

The patient’s hemoglobin is within normal range, which is consistent with sickle cell trait; however, he is microcytic, with a normal RDW. It is possible to be mildly microcytic in the early stages of iron deficiency, prior to the development of anemia, but the RDW would typically be elevated, demonstrating the presence of newer, smaller cells produced under conditions of iron deficiency.⁶⁷ It is also possible that his microcytosis with a normal RDW could represent sickle cell trait with co-inheritance of β-thalassemia. Up to 30% of African Americans have β-thalassemia,² and 1 in 10 have sickle cell trait.⁶⁸ However, a high MCHC, indicating the presence of dense cells, and target cells noted on slide review are most consistent with HbSC.⁹ HbSC patients, especially males, can have hemoglobin levels in the normal range.⁴ The biggest inconsistency with the diagnosis of sickle cell trait is his history of frequent pain events. Individuals with sickle cell trait rarely present with pain crises, except under extreme conditions of dehydration or high altitude.⁶⁸ Sickle cell trait is generally regarded as a benign condition, although a study of U.S. military recruits found a 30-fold higher risk of sudden death during basic training in persons with sickle cell trait.⁶⁹ Additional sickle cell trait–related complications include hematuria, risk of splenic sequestration or infarct under extreme conditions and high altitude, and a rare and usually fatal renal malignancy, renal medullary carcinoma, which is vanishingly rare in individuals without sickle cell trait.^70,71 Although the patient reported having sickle cell trait, this diagnosis should have been verified with a hemoglobin panel, given his atypical presentation.²⁰

In sickle cell disease, vaso-occlusive pain events can be common, often beginning in early childhood.¹⁷ This disease complication accounts for 95% of all adult sickle cell disease hospitalizations.⁷² There is a great deal of variability in pain symptoms between individuals, and within individuals at various times in their lives:⁷³ 30% have no pain events, 50% have occasional events, and 20% have monthly or more frequent events that require hospitalization.⁷⁴ The frequency and severity of pain events are modulated by HbF levels, β-thalassemia status, genotypes, therapies like hydroxyurea, or in rare cases, chronic transfusion therapy.²³ Personal factors, such as psychosocial stressors, also contribute to the frequency of pain events.⁷⁵ Pain event triggers include exposure to cold water, windy or cold weather, temperature changes, and extreme temperatures.^76–79 Patient age also contributes to pain event frequency. Many patients see an increase in pain event frequency in their late 20s, and a marked decrease in their 40s.^23,73 More than 3 pain events per year is associated with reduced life expectancy.²³

Acute management of pain episodes involves nonsteroidal anti-inflammatory drugs, oral opioids, and when hospitalization is required, intravenous opioids, often delivered via patient-controlled analgesia (PCA) pumps.⁷⁹ As sickle cell disease patients become teenagers and young adults, some experience an increased frequency of pain episodes, with fewer pain-free days, or a failure to return to baseline before the next pain crisis occurs.^80,81 This is characteristic of emerging chronic pain.⁸² Chronic pain is a significant problem in adult patients with sickle cell disease, with up to 85% reporting pain on most days.^72,80 The development of chronic pain may be reduced by early and aggressive treatment of acute pain events, as well as use of hydroxyurea to reduce the number of pain events. Many adult sickle cell patients with chronic pain are treated with daily opioids.²⁰ Given the significant side effects of chronic opioid use—sedation, respiratory depression, itching, nausea, and impairment of function and quality of life—non-opioid therapies are under investigation.⁸³ Many chronic pain patients have symptoms of neuropathic pain, and may benefit from neuropathic agents like gabapentin, both to reduce opioid use and to more effectively treat chronic neuropathic pain, which is known to respond poorly to opioids.^84–86

• Is the patient’s peripheral blood smear consistent with a diagnosis of sickle cell trait?

Several target cells are visible, which is not typical of sickle cell trait, but may be seen in HbSC or thalassemia. The finding of an intracellular crystal is pathognomonic for HbSC or HbCC. HbC polymerizes in high oxygen conditions, opposite of HbS, which polymerizes in low oxygen conditions.⁹

CASE CONTINUED

The patient’s family history is significant for a sister who died at age 3 from sickle cell–related complications, and a sister with sickle cell trait who had a cholecystectomy for gallstones at age 22. His father died at age 38 due to unknown causes. The sickle cell trait status of his parents is unknown. His mother is alive, and has hypertension.

• Is the medical history of this patient’s family members consistent with sickle cell trait?

It is unlikely that sickle cell trait would result in early death in childhood, or in gallstones at age 22. Gallstones in early adulthood is a common presentation for HbSC patients not diagnosed by newborn screening.⁸⁷ Any hemolytic condition can lead to the formation of hemoglobin-containing pigmented gallstones, biliary sludge, and obstruction of the gallbladder. In the presence of right-sided abdominal pain, a serum bilirubin level of more than 4 mg/dL should lead to measurement of direct bilirubin; if greater than 10% of total, imaging of the gallbladder should be obtained. In sickle cell disease, 30% of patients will have gallstones by 18 years of age. The low hemolysis/high viscosity phenotype patients are typically older at diagnosis. Co-inheritance of Gilbert syndrome and sickle cell disease is not uncommon, and can result in formation of gallstones at a young age; Gilbert syndrome alone typically results in gallstones in mid-life.⁸⁸

Two months later, the patient presents again to the emergency department with the same complaint of leg pain, as well as abdominal pain. His hemoglobin is 12.5 g/dL, and his platelet count is 134,000/µL. His pain is not improved with 3 doses of morphine 6 mg intravenously, and he is admitted to the medicine service. A hemoglobin profile is obtained, revealing 52% HbS, 45% HbC, and 1.5% HbF, consistent with HbSC. In sickle cell trait, the hemoglobin profile is 60% HbA and 40% HbS (available α-globin prefers to pair with a normal β-globin, so the ratio of HbA to HbS is 60:40, not 50:50).

On the second hospital day, the patient’s hemoglobin drops to 7.2 g/dL and his platelet count decreases to 44,000/µL. His abdomen is distended and diffusely tender. The internist transfuses him with 2 units of packed red blood cells (PRBC), after which his hemoglobin increases to 11 g/dL, while his platelet count increases to 112,000/µL. Following the transfusion, his abdominal pain resolves, as does his anemia and thrombocytopenia.

• What caused this patient’s anemia and thrombocytopenia?

High on the differential diagnosis is a splenic sequestration. Acute splenic sequestration occurs when red cells are trapped in the splenic sinuses. Massive splenic enlargement may occur over several hours.^89,90 Unrecognized splenic sequestration has a high mortality rate from severe anemia and splenic rupture.⁹⁰ Splenic sequestration must be ruled out in a sickle cell patient with abdominal pain accompanied by dropping platelet and red cell counts, especially in milder subtypes that often have splenic function preserved into adolescence and adulthood. Sickle cell anemia patients usually become functionally asplenic in early childhood.^89,91,92 The rise in hemoglobin, more than would be expected from 2 units of PRBC, plus the improvement in platelet count without a platelet transfusion observed in the case patient strongly supports the diagnosis of splenic sequestration.

Splenic sequestration can occur in any sickle cell patient whose spleen has not fibrosed. Splenic sequestration in adulthood is not uncommon in HbSC patients, who often have preserved splenic function into adulthood.^93–95

Clinical signs of splenic sequestration include a rapid drop in hemoglobin, rise in reticulocyte count, a tender, enlarged spleen, and, in severe cases, hypovolemia.^89,93 It is treated with prompt blood transfusion, but care must be taken not to overtransfuse the patient, as the spleen can trap several grams of hemoglobin, which may be released upon transfusion, potentially causing life-threatening hyperviscosity.⁸⁹ Hemoglobin levels must be checked following transfusion in suspected splenic sequestration, and “mini transfusions” of 5 mL/kg are recommended in sickle cell disease patients who are hemodynamically stable.²⁰

Hepatic sequestration may also occur, but it is much less common than splenic sequestration.⁹⁶ Other conditions on the differential diagnosis include thrombotic thrombocytopenic purpura, which would be unlikely to respond to a transfusion. ACS can cause a drop in hemoglobin, and is treated with simple or exchange transfusions.⁹⁷ ACS is less likely without respiratory symptoms or oxygen requirement, and usually is not associated with thrombocytopenia. Sepsis may also cause anemia and thrombocytopenia, but again would not likely respond to a simple transfusion. The patient’s response to transfusion is consistent with a sequestering event, not a destructive event as in the case of sepsis.

CASE CONTINUED

Imaging reveals a grossly enlarged spleen, which is having a mass effect on the left kidney. The patient is started on hydroxyurea therapy at 500 mg 3 times daily. Discharge instructions include following up with his primary care physician, continuing hydroxyurea therapy, and receiving yearly dilated eye exams to evaluate for proliferative sickle retinopathy.

• Are these discharge instructions complete?

Splenic sequestration has a 50% recurrence rate.⁹⁸ In very young children, watchful waiting or chronic transfusion may be implemented to preserve the immunologic function of the spleen and reduce the risk of sepsis.⁸⁹ Splenectomy after a single episode of sequestration in adults is a matter of debate, with experts advising both watchful waiting⁹⁹ and splenectomy after recovery from the first sequestering event.¹⁰⁰ The patient should have been informed of the risk for recurrence, and the signs and symptoms of splenic sequestration as well as the need for emergency medical attention should have been discussed. Splenic sequestration may be milder in adults than in children, but fatal sequestrations have been reported.^95,101–103

Proliferative sickle cell retinopathy is a high viscosity/high hemoglobin complication that may occur more frequently in HbSC than HbSS, with an incidence of 33% in HbSC.^42,104 Spontaneous regression of retinopathy occurs in approximately 32% of eyes, and laser or scatter photocoagulation is an effective intervention.¹⁰⁵

• Would the patient need to be transfused prior to splenectomy?

Preoperative transfusion therapy is standard of care for HbSS patients undergoing general anesthesia. The TRAP study found that simple “top off” transfusion to a hemoglobin of 10 g/dL was as effective at preventing postoperative sickle cell–related complications as exchange transfusion to HbS of 30% or less, and had fewer transfusion-related complications like alloimmunization.¹⁰⁶ There is little data regarding preoperative transfusions in HbSC disease. A retrospective study suggests that HbSC patients undergoing abdominal surgeries should be transfused.¹⁰⁷ The higher hemoglobin level of the typical HbSC patient necessitates exchange transfusion to avoid hyperviscosity.

• Is hydroxyurea therapy indicated in this patient?

• Has it been dosed appropriately?

If the patient had the HbSS subtype, hydroxyurea would be clearly indicated, given his frequent pain events.²⁰ HbSC patients may be placed on hydroxyurea on a case-by-case basis, but evidence for its efficacy in this sickle cell subtype is lacking.¹⁰⁸ Large clinical trials like the Multi-Center Study of Hydroxyurea (MSH) that established the safety and efficacy of hydroxyurea in sickle cell anemia excluded HbSC and HbSβ⁺ patients.¹⁰⁹ These mild to moderate subtypes produce less HbF at baseline, and typically have a minimal to modest rise in HbF on hydroxyurea.¹¹⁰ In sickle cell anemia, hydroxyurea is titrated to maximum tolerated dose, defined as an ANC of 2000 to 4000/µL and an ARC of 70,000/µL or higher.⁵³ Because of their lower levels of chronic inflammation and lower reticulocyte counts due to higher hemoglobin levels, many HbSC and HbSβ⁺ patients have values in that range before initiating hydroxyurea therapy.⁹ Cytopenias, particularly of platelets in HbSC, occur at low doses of hydroxyurea.¹¹¹

Of note, although the half-life of hydroxyurea would suggest that 3 times daily dosing is indicated, daily dosing has been found to have equal response and is preferred. Another concern is the monitoring of this myelosuppressive medication. This patient has repeatedly failed to obtain a primary care physician or a hematologist, and hydroxyurea requires laboratory monitoring at least every 2 months, especially in a HbSC patient with a very large spleen who is at significant risk for thrombocytopenia and neutropenia.⁹

CASE CONTINUED

A week after discharge from his admission for abdominal pain diagnosed as splenic sequestration, the patient presents again to the emergency department with abdominal pain which he reports is his typical sickle cell pain. Hemoglobin is 13.8 g/dL, platelet count is 388,000/µL, and alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels are both 10 times their prior value. Creatinine is 1.2 mg/dL (0.75 mg/dL on his prior admission), and total bilirubin is 3 mg/dL, with 0.3 mg/dL direct bilirubin. He undergoes an ultrasound exam of his gallbladder, which reveals sludge and a possible gallstone. There is no evidence of cholecystitis. General surgery performs a laparoscopic cholecystectomy.

• Was this cholecystectomy necessary?

In patients with sickle cell disease, symptomatic gallstones and gallbladder sludge should be observed; recurrent abdominal pain without a significant change in bilirubin may not be due to gallstones or sludge, and therefore may not be relieved by cholecystectomy.^112,113 In sickle cell disease, 40% of patients with gallbladder sludge do not develop gallstones.⁸⁷ The patient’s bilirubin level was at baseline, and there was no increase in the direct (conjugated) fraction. Watchful waiting would have been appropriate, with cholecystectomy being performed if he experienced recurrent symptoms associated with fatty foods accompanied by an elevation in direct bilirubin.

More concerning and deserving of investigation was his elevated liver enzymes. Patients with sickle cell disease may experience recurrent ischemia and reperfusion injuries in the liver, which is called right upper quadrant syndrome. On autopsy of 70 sickle cell patients, 91% had hepatomegaly and 34% had focal necrosis.¹¹⁴ AST is often elevated in sickle cell disease, as it is affected by hemolysis. In this patient, both AST and ALT are elevated, consistent with a hepatocellular disorder. His abdominal pain and ALT rise may be a sign of a hepatic crisis.¹¹⁵ Rapid resolution of ALT elevation in a matter of days suggests a vaso-occlusive, inflammatory event that is self- limiting. Prolonged AST elevation requires further investigation, with consideration of autoimmune hepatitis, viral hepatitis, or iron overload. Iron overload is unlikely in this patient given his lifetime history of only 1 transfusion. Hepatic iron overload typically occurs in sickle cell disease after a minimum of 10 transfusions.¹¹⁵

CASE CONTINUED

The patient is discharged on the day after the procedure, with instructions to continue his hydroxyurea.

• Should the patient resume hydroxyurea therapy?

Hydroxyurea is hepatically cleared and thus it should be held until his liver function tests normalize.¹⁰⁶

CASE CONTINUED

Two months later, the patient presents to the emergency department with abdominal pain that moves to his left leg. A CBC is obtained, showing a hemoglobin of 11.8 g/dL and a platelet count of 144,000/µL. He is given 2 doses of morphine 6 mg intravenously, and reports that his leg pain is now a 4/10. He is discharged home with a prescription for hydrocodone/acetaminophen.

• Is the emergency department evaluation sufficient?

This patient remains at high risk for splenic sequestration,⁹³ with a hemoglobin 2 g lower than it was 2 months ago and platelets less than half. This decline could be consistent with early splenic sequestration.²⁰ Additionally, he had elevated liver function tests on a recent admission, as well as rising creatinine, without evidence of resolution. It is not appropriate to discharge him without checking a chemistry and liver panel, and abdominal imaging should be considered. The best plan would be to admit him for observation, given his risk for splenic sequestration, and consult surgery for an elective splenectomy if he has a second episode of splenic sequestration 2 months after the first.¹⁰⁰ His abdominal pain that migrates to his left leg could be due to his massive splenomegaly compressing his left kidney, as noted on imaging during his recent admission for splenic sequestration

CASE CONTINUED

An hour after discharge from the emergency department, EMS is called to his home for intractable pain. He is found lying on the floor, and reports excruciating left leg pain. He is brought to the closest hospital, a community hospital that he has not visited previously. There, he is admitted for hydration and pain control and placed on hydromorphone 2 mg every 4 hours as needed for pain. His hemoglobin is 10.8 g/dL, and platelets are 121,000/µL. A chemistry panel is remarkable for a creatinine level of 1.5 mg/dL and a potassium level of 3.2 mEq/L. Liver function tests are not obtained. After 3 doses of hydromorphone, he falls asleep. He is not in a monitored bed, and intravenous fluids, while ordered, are not started. At 6:30 AM the day after admission, he cannot be aroused on a routine vital sign check; he has an SpO₂ of 60%, a blood pressure of 80/60 mm Hg, and heart rate of 148 beats/min. A rapid response is called, and naloxone is administered along with oxygen by face mask and several fluid boluses. His systolic blood pressure increases to 100 mm Hg from a low of 70 mm Hg. His SpO₂ increases to 92%, and he is arousable and alert, although he reports 10/10 leg pain. His abdomen is noted to be distended and tender.

• What may have contributed to his clinical condition?

The patient is opioid tolerant and has received equivalent doses of opioids in the past without excess sedation. He may have liver dysfunction making him unable to metabolize opioids effectively. His hemoglobin and platelets continue to decline, raising concern for splenic sequestration versus sepsis. Failure to place him on a monitor allowed his hypoxia to continue for an unknown amount of time, placing him at high risk for developing ACS. Lack of intravenous hydration while he was too sedated to drink likely exacerbated his sickling.

At 9:20 AM, a CBC is obtained and reveals a hemoglobin of 4.8 g/dL and a platelet count of 44,000/µL. Two units of stat O negative blood are administered, and preparations are made to administer an exchange transfusion. A liver panel is obtained 3 hours later, which reveals an AST level of 1200 U/L and an ALT level of 1050 U/L. His bilirubin is 10 mg/dL, and his lactate dehydrogenase level is 1800 U/L. His urine is dark and is positive for bilirubin and ketones. He is transferred to the intensive care unit. A chest X-ray shows pulmonary congestion. Hematology/oncology is consulted.

He receives a 7-unit red blood cell exchange, which reduces his HbS to 11%. He continues to be hypotensive, and requires norepinephrine to support his blood pressure. Antibiotic therapy is started. His creatinine concentration rises to 2.3 mg/dL, potassium is 7.8 mEq/L, and bicarbonate is 12 mEq/L. He is placed on hemodialysis.

Computed tomography of the chest and abdomen reveals lower posterior lung infiltrates and a grossly enlarged spleen. He requires intubation. He is given a diagnosis of ACS in addition to kidney failure, liver failure, and “sickle crisis.” He continues to require daily to twice daily transfusions to maintain a hemoglobin of 7 to 9 g/dL, and his abdominal distension increases. As his condition worsens, surgery is consulted to discuss a liver transplant. He is deemed to not be a surgical candidate, and he passes away 6 days after entering the hospital. The immediate cause of death is listed as vaso-occlusive crisis, with ACS and sickle crisis listed as contributors.

• Are the causes of death accurate and complete?

If vaso-occlusive crisis is used to indicate a pain event, it is not an accurate cause of death. Pain is one of the most distressing complications of sickle cell disease, and frequent pain events are associated with early mortality,^4,80 but they are not in themselves fatal. ACS is the number one cause of death in sickle cell disease,⁴ and it likely contributed to this patient’s death. Sickle crisis is a vague term that should not be used in this context. Causes of death should include splenic sequestration and multisystem organ failure. Multisystem organ failure in sickle cell disease often responds to aggressive transfusion therapy, which this patient received.^116–118

CONCLUSION

Sickle cell disease is a complex chronic disease that impacts almost every organ system in the body. Clinicians may be inclined to attribute most pain in a patient with sickle cell disease to a simple vaso-occlusive crisis, treat them for this, and not investigate further. As the case presented here demonstrates, failure to identify the actual life-threatening process occurring in a patient with sickle cell disease presenting with pain can result in preventable early mortality. Clinicians must approach a sickle cell patient reporting pain in a thoughtful manner, and consider a complete differential diagnosis, including both sickle cell disease complications and those unrelated to sickle cell disease. Knowledge of the disease courses of the different sickle cell genotypes is essential, and must go beyond a superficial hierarchy of severity, but rather include an understanding of the complications each genotype is most prone to, and at what ages. Complete laboratory assessment, including a comprehensive metabolic panel, should be performed on all admitted patients, not just a complete blood count. Treating pain with high-dose opioids, while appropriate in an uncomplicated pain crisis, can lead to ACS or even respiratory failure in a patient with uninvestigated liver and kidney dysfunction. The most important lesson to remember is that even the sickle cell disease patient who has been given the unfortunate and pejorative label of “frequent flyer” by some providers has the potential for rapid deterioration into multisystem organ failure and death.

INTRODUCTION

Sickle cell disease is the most common inherited blood disorder in the world. It affects more than 100,000 individuals in the United States, and millions more worldwide.¹ Sickle cell disease is most commonly found in individuals of African heritage, but the disease also occurs in Hispanics and people of Middle Eastern and subcontinent Indian heritage.² The distribution of the sickle hemoglobin (hemoglobin S [HbS]) allele overlaps with the distribution of malaria; HbS carriers, or individuals with sickle cell trait, have protection against malaria,³ and are not considered to have sickle cell disease.

Sickle cell disease is a severe monogenic disorder marked by significant morbidity and mortality, affecting every organ in the body.⁴ The term sickle cell disease refers to all genotypes that cause sickling; the most common are the homozygous hemoglobin SS (HbSS) and compound heterozygotes hemoglobin SC (HbSC), hemoglobin S–β⁰-thalassemia (HbSβ⁰),and hemoglobin S–β⁺-thalassemia(HbSβ⁺), although HbS and several rarer hemoglobin variants such as HbSO(Arab) and HbSD(Punjab) can also cause sickle cell disease.The term sickle cell anemia refers exclusively to the most severe genotypes, HbSS and HbSβ⁰.⁵ Common sickling genotypes along with their relative clinical severity are shown in Table 1.^6–11

This article reviews the pathophysiology of sickle cell disease, common clinical complications, and available therapies. A complex case which illustrates diagnostic and management challenges is presented as well.

PATHOPHYSIOLOGY

HbS is the result of a substitution of valine for glutamic acid in the sixth amino acid of the β-globin chain.¹² The change from a hydrophilic to a hydrophobic amino acid causes the hemoglobin molecules to stack, or polymerize, when deoxygenated. This rigid rod of hemoglobin distorts the cell, producing the characteristic crescent or sickle shape that gives the disease its name.¹³ Polymerization of hemoglobin within the cell is promoted by dehydration, which increases the concentration of HbS.^13,14 Polymerization occurs when hemoglobin is in the deoxygenated state.¹³

The sickle red blood cell is abnormal; it is rigid and dense, and lacks the deformability needed to navigate the microvasculature.¹⁵ Blockages of blood flow result in painful vaso-occlusion that is the hallmark of the disease, and that also can cause damage to the spleen, kidneys, and liver.¹⁶ The sickle red cell is also fragile, with a lifespan of only 20 days compared to the 120-day lifespan of a normal red blood cell.¹³ Frequent hemolysis results in anemia and the release of free hemoglobin, which both scavenges nitric oxide and impairs the production of more nitric oxide, which is essential for vasodilatation.¹⁷ This contributes to vascular dysfunction and an increased risk for stroke.¹⁸ If untreated, the natural course of sickle cell anemia is mortality in early childhood in most cases.¹⁹ Common chronic and acute sickle cell disease–related complications and recommended therapies, based on 2014 National Institutes of Health guidelines, are shown in Table 2 and Table 3.²⁰

One of the most challenging aspects of sickle cell disease is its clinical variability. While in general, HbSS and HbSβ⁰ are the most severe genotypes, there are patients with HbSC and HbSb⁺ who have significant sickle-cell–related complications, and may have a more severe clinical course than a HbSS patient.²¹ A great deal of this clinical variability cannot be explained, but some can be attributed to endogenous fetal hemoglobin (HbF) levels.^22–24 The importance of HbF levels in sickle cell disease was first noted by a pediatrician in the 1940s.²⁵ She observed that sickle cell disease complications in children under the age of 1 were rare, and attributed it to the presence of HbF.²⁵ HbF levels decline more slowly in individuals with hemoglobinopathies, reaching their nadir after the age of 5 rather than within 6 months of birth in individuals without hemoglobinopathies.²⁶ HbF levels remain elevated lifelong in most sickle cell disease patients, especially those with the HbSS and HbSβ⁰ genotypes. Levels of HbF vary widely between individuals, from zero to 20% to 30%, with a median of 10%.^26–28 Individuals who produce more HbF have a milder course, in general.²⁴ An association between the 4 β-globin haplotypes and HbF levels has been reported in the past,^27,29 but more sophisticated next-generation sequencing has revealed causal variants in BCL11A and HBS1L-MYB that contribute approximately 50% of the observed variability in HbF levels.^30–33

Co-inheritance of α-thalassemia also modifies disease course; less available α-globin chains results in a lower hemoglobin concentration within the cell. Paradoxically, this results in a higher overall hemoglobin level, as there is a reduction in polymerization, and therefore sickling due to lower HbS concentrations in the cell. Patients therefore are less anemic, reducing the risk of stroke in childhood,^34,35 but blood viscosity may be higher, resulting in more frequent pain crises and increased risk³⁶ of avascular necrosis.^34,35,37 It is often helpful to think of sickle cell patients as falling into 1 of 2 groups: high hemolysis/low hemoglobin and high viscosity/high hemoglobin. Individuals with high rates of hemolysis are at greater risk for stroke, pulmonary hypertension, and acute chest syndrome (ACS). Higher rates of hemolysis result in higher levels of free hemoglobin, which scavenges nitric oxide. This leads to the vascular damage and dysfunction that contributes to the associated clinical complications. This phenotype is most commonly seen in HbSS and HbSβ⁰.³⁸ High hemoglobin/high viscosity phenotypes are most often found in HbSC patients and in sickle cell anemia with α-thalassemia coinheritance.^39–42

TREATMENT OPTIONS

In high-resource countries with newborn screening, the initiation of penicillin prophylaxis has dramatically altered the natural history of the disease, allowing the majority of patients to reach adulthood.⁴³ Penicillin prophylaxis is usually discontinued at age 5 years; however, individuals who have undergone surgical splenectomy or have had pneumococcal sepsis on penicillin prophylaxis may remain on penicillin to age 18 or beyond.²⁰

Another advance in sickle cell care is screening for stroke risk through transcranial Doppler ultrasound (TCD).^44–47 This screening tool has reduced the incidence of childhood stroke from 10% by age 11 to 1%. TCDs typically cannot be performed after the age of 16 due to changes in the skull. Individuals found to have abnormal (elevated) TCD velocities are placed on chronic transfusion therapy for primary stroke prevention. They may remain on monthly chronic transfusions, with the goal of suppressing the percentage of HbS to 30% to 50% indefinitely. A clinical trial (STOPII) designed to determine if pediatric sickle cell disease patients on chronic transfusion therapy for primary stroke prevention could be safely taken off transfusion therapy was discontinued early due to an excess of strokes and conversion to abnormal TCD velocities in the untransfused arm.⁴⁴ Individuals who have experienced an ischemic stroke have a 70% risk of another stroke, and must remain on chronic transfusion therapy indefinitely. Chronic transfusion reduces their stroke risk to 13%.

The only widely used pharmacologic therapy for sickle cell disease is hydroxyurea.^12,48–50 A significant portion of the benefit of hydroxyurea stems from its induction of HbF.⁵¹ HbF does not sickle, and it interrupts the polymerization of HbS in the cell, if present in high enough concentrations.⁵⁰ The level of HbF needed to achieve clinical improvement is not known, but in vitro assays suggest 20% HbF is needed to prevent sickling.^52,53 However, endogenous and hydroxyurea-induced HbF is not distributed evenly through the red cells, so sickling is possible regardless of the level of HbF induced.^54,55 Hydroxyurea likely has other disease-modifying effects as well, including reduction of white blood cell count and reticulocyte count and reduction of red cell adhesion to the endothelium.^56–58 Clinical criteria for initiation of hydroxyurea in adult sickle cell disease patients are shown in Table 4.²⁰ Hydroxyurea is given daily and is dosed to maximum tolerated dose for the individual by following the absolute neutrophil count (ANC). The goal ANC is between 2000 and 4000/µL. At times, absolute reticulocyte count (ARC) can be dose-limiting; goal ARC is greater than 70,000/µL.⁵⁹ Platelet counts may be reduced as well, especially in HbSC patients.^60,61

The only curative therapy for sickle cell disease is hematopoietic stem cell transplant.⁶² Transplant use is limited by availability of matched sibling donors,⁶² and even at experienced centers transplant carries a small risk for mortality, graft rejection, and graft-versus-host disease. Furthermore, consensus on disease complications for which transplant is recommended is also lacking.^63–65 Clinical trials of gene therapy for sickle cell disease and thalassemia are ongoing.⁶⁶

COMPLICATIONS AND DISEASE-SPECIFIC THERAPIES

CASE PRESENTATION

A 26-year-old African-American man who works as a school bus driver presents to an academic center’s emergency department complaining of pain in his left leg, similar to prior pain events. He is described as having sickle cell trait, although no hemoglobin profile is available in his chart. He describes the pain as dull and aching, 10/10 in intensity. A complete blood count (CBC) is obtained; it reveals a hemoglobin of 14.5 g/dL, white blood cell (WBC) count of 5600/µL, and platelet count of 344,000/µL. His CBC is also notable for a mean corpuscular volume (MCV) of 72 fL, a mean corpuscular hemoglobin concentration (MCHC) of 37 g/dL, and a red blood cell distribution width (RDW) of 12. Slide review of a peripheral blood smear shows 2+ target cells (Figure).

The patient is given 6 mg of morphine, which provides some relief of his pain, and is discharged with a prescription for hydrocodone bitartrate/acetaminophen 5/325 mg. The diagnosis given is musculoskeletal pain, and he is instructed to follow-up with a primary care physician. His past medical history is significant for 4 or 5 visits to the emergency department per year in the past 4 years. Prior to 4 years ago, he rarely required medical attention.

• What laboratory and clinical features might lead you to question the diagnosis of sickle cell trait in this patient?

The patient’s hemoglobin is within normal range, which is consistent with sickle cell trait; however, he is microcytic, with a normal RDW. It is possible to be mildly microcytic in the early stages of iron deficiency, prior to the development of anemia, but the RDW would typically be elevated, demonstrating the presence of newer, smaller cells produced under conditions of iron deficiency.⁶⁷ It is also possible that his microcytosis with a normal RDW could represent sickle cell trait with co-inheritance of β-thalassemia. Up to 30% of African Americans have β-thalassemia,² and 1 in 10 have sickle cell trait.⁶⁸ However, a high MCHC, indicating the presence of dense cells, and target cells noted on slide review are most consistent with HbSC.⁹ HbSC patients, especially males, can have hemoglobin levels in the normal range.⁴ The biggest inconsistency with the diagnosis of sickle cell trait is his history of frequent pain events. Individuals with sickle cell trait rarely present with pain crises, except under extreme conditions of dehydration or high altitude.⁶⁸ Sickle cell trait is generally regarded as a benign condition, although a study of U.S. military recruits found a 30-fold higher risk of sudden death during basic training in persons with sickle cell trait.⁶⁹ Additional sickle cell trait–related complications include hematuria, risk of splenic sequestration or infarct under extreme conditions and high altitude, and a rare and usually fatal renal malignancy, renal medullary carcinoma, which is vanishingly rare in individuals without sickle cell trait.^70,71 Although the patient reported having sickle cell trait, this diagnosis should have been verified with a hemoglobin panel, given his atypical presentation.²⁰

In sickle cell disease, vaso-occlusive pain events can be common, often beginning in early childhood.¹⁷ This disease complication accounts for 95% of all adult sickle cell disease hospitalizations.⁷² There is a great deal of variability in pain symptoms between individuals, and within individuals at various times in their lives:⁷³ 30% have no pain events, 50% have occasional events, and 20% have monthly or more frequent events that require hospitalization.⁷⁴ The frequency and severity of pain events are modulated by HbF levels, β-thalassemia status, genotypes, therapies like hydroxyurea, or in rare cases, chronic transfusion therapy.²³ Personal factors, such as psychosocial stressors, also contribute to the frequency of pain events.⁷⁵ Pain event triggers include exposure to cold water, windy or cold weather, temperature changes, and extreme temperatures.^76–79 Patient age also contributes to pain event frequency. Many patients see an increase in pain event frequency in their late 20s, and a marked decrease in their 40s.^23,73 More than 3 pain events per year is associated with reduced life expectancy.²³

Acute management of pain episodes involves nonsteroidal anti-inflammatory drugs, oral opioids, and when hospitalization is required, intravenous opioids, often delivered via patient-controlled analgesia (PCA) pumps.⁷⁹ As sickle cell disease patients become teenagers and young adults, some experience an increased frequency of pain episodes, with fewer pain-free days, or a failure to return to baseline before the next pain crisis occurs.^80,81 This is characteristic of emerging chronic pain.⁸² Chronic pain is a significant problem in adult patients with sickle cell disease, with up to 85% reporting pain on most days.^72,80 The development of chronic pain may be reduced by early and aggressive treatment of acute pain events, as well as use of hydroxyurea to reduce the number of pain events. Many adult sickle cell patients with chronic pain are treated with daily opioids.²⁰ Given the significant side effects of chronic opioid use—sedation, respiratory depression, itching, nausea, and impairment of function and quality of life—non-opioid therapies are under investigation.⁸³ Many chronic pain patients have symptoms of neuropathic pain, and may benefit from neuropathic agents like gabapentin, both to reduce opioid use and to more effectively treat chronic neuropathic pain, which is known to respond poorly to opioids.^84–86

• Is the patient’s peripheral blood smear consistent with a diagnosis of sickle cell trait?

Several target cells are visible, which is not typical of sickle cell trait, but may be seen in HbSC or thalassemia. The finding of an intracellular crystal is pathognomonic for HbSC or HbCC. HbC polymerizes in high oxygen conditions, opposite of HbS, which polymerizes in low oxygen conditions.⁹

CASE CONTINUED

The patient’s family history is significant for a sister who died at age 3 from sickle cell–related complications, and a sister with sickle cell trait who had a cholecystectomy for gallstones at age 22. His father died at age 38 due to unknown causes. The sickle cell trait status of his parents is unknown. His mother is alive, and has hypertension.

• Is the medical history of this patient’s family members consistent with sickle cell trait?

It is unlikely that sickle cell trait would result in early death in childhood, or in gallstones at age 22. Gallstones in early adulthood is a common presentation for HbSC patients not diagnosed by newborn screening.⁸⁷ Any hemolytic condition can lead to the formation of hemoglobin-containing pigmented gallstones, biliary sludge, and obstruction of the gallbladder. In the presence of right-sided abdominal pain, a serum bilirubin level of more than 4 mg/dL should lead to measurement of direct bilirubin; if greater than 10% of total, imaging of the gallbladder should be obtained. In sickle cell disease, 30% of patients will have gallstones by 18 years of age. The low hemolysis/high viscosity phenotype patients are typically older at diagnosis. Co-inheritance of Gilbert syndrome and sickle cell disease is not uncommon, and can result in formation of gallstones at a young age; Gilbert syndrome alone typically results in gallstones in mid-life.⁸⁸

Two months later, the patient presents again to the emergency department with the same complaint of leg pain, as well as abdominal pain. His hemoglobin is 12.5 g/dL, and his platelet count is 134,000/µL. His pain is not improved with 3 doses of morphine 6 mg intravenously, and he is admitted to the medicine service. A hemoglobin profile is obtained, revealing 52% HbS, 45% HbC, and 1.5% HbF, consistent with HbSC. In sickle cell trait, the hemoglobin profile is 60% HbA and 40% HbS (available α-globin prefers to pair with a normal β-globin, so the ratio of HbA to HbS is 60:40, not 50:50).

On the second hospital day, the patient’s hemoglobin drops to 7.2 g/dL and his platelet count decreases to 44,000/µL. His abdomen is distended and diffusely tender. The internist transfuses him with 2 units of packed red blood cells (PRBC), after which his hemoglobin increases to 11 g/dL, while his platelet count increases to 112,000/µL. Following the transfusion, his abdominal pain resolves, as does his anemia and thrombocytopenia.

• What caused this patient’s anemia and thrombocytopenia?

High on the differential diagnosis is a splenic sequestration. Acute splenic sequestration occurs when red cells are trapped in the splenic sinuses. Massive splenic enlargement may occur over several hours.^89,90 Unrecognized splenic sequestration has a high mortality rate from severe anemia and splenic rupture.⁹⁰ Splenic sequestration must be ruled out in a sickle cell patient with abdominal pain accompanied by dropping platelet and red cell counts, especially in milder subtypes that often have splenic function preserved into adolescence and adulthood. Sickle cell anemia patients usually become functionally asplenic in early childhood.^89,91,92 The rise in hemoglobin, more than would be expected from 2 units of PRBC, plus the improvement in platelet count without a platelet transfusion observed in the case patient strongly supports the diagnosis of splenic sequestration.

Splenic sequestration can occur in any sickle cell patient whose spleen has not fibrosed. Splenic sequestration in adulthood is not uncommon in HbSC patients, who often have preserved splenic function into adulthood.^93–95

Clinical signs of splenic sequestration include a rapid drop in hemoglobin, rise in reticulocyte count, a tender, enlarged spleen, and, in severe cases, hypovolemia.^89,93 It is treated with prompt blood transfusion, but care must be taken not to overtransfuse the patient, as the spleen can trap several grams of hemoglobin, which may be released upon transfusion, potentially causing life-threatening hyperviscosity.⁸⁹ Hemoglobin levels must be checked following transfusion in suspected splenic sequestration, and “mini transfusions” of 5 mL/kg are recommended in sickle cell disease patients who are hemodynamically stable.²⁰

Hepatic sequestration may also occur, but it is much less common than splenic sequestration.⁹⁶ Other conditions on the differential diagnosis include thrombotic thrombocytopenic purpura, which would be unlikely to respond to a transfusion. ACS can cause a drop in hemoglobin, and is treated with simple or exchange transfusions.⁹⁷ ACS is less likely without respiratory symptoms or oxygen requirement, and usually is not associated with thrombocytopenia. Sepsis may also cause anemia and thrombocytopenia, but again would not likely respond to a simple transfusion. The patient’s response to transfusion is consistent with a sequestering event, not a destructive event as in the case of sepsis.

CASE CONTINUED

Imaging reveals a grossly enlarged spleen, which is having a mass effect on the left kidney. The patient is started on hydroxyurea therapy at 500 mg 3 times daily. Discharge instructions include following up with his primary care physician, continuing hydroxyurea therapy, and receiving yearly dilated eye exams to evaluate for proliferative sickle retinopathy.

• Are these discharge instructions complete?

Splenic sequestration has a 50% recurrence rate.⁹⁸ In very young children, watchful waiting or chronic transfusion may be implemented to preserve the immunologic function of the spleen and reduce the risk of sepsis.⁸⁹ Splenectomy after a single episode of sequestration in adults is a matter of debate, with experts advising both watchful waiting⁹⁹ and splenectomy after recovery from the first sequestering event.¹⁰⁰ The patient should have been informed of the risk for recurrence, and the signs and symptoms of splenic sequestration as well as the need for emergency medical attention should have been discussed. Splenic sequestration may be milder in adults than in children, but fatal sequestrations have been reported.^95,101–103

Proliferative sickle cell retinopathy is a high viscosity/high hemoglobin complication that may occur more frequently in HbSC than HbSS, with an incidence of 33% in HbSC.^42,104 Spontaneous regression of retinopathy occurs in approximately 32% of eyes, and laser or scatter photocoagulation is an effective intervention.¹⁰⁵

• Would the patient need to be transfused prior to splenectomy?

Preoperative transfusion therapy is standard of care for HbSS patients undergoing general anesthesia. The TRAP study found that simple “top off” transfusion to a hemoglobin of 10 g/dL was as effective at preventing postoperative sickle cell–related complications as exchange transfusion to HbS of 30% or less, and had fewer transfusion-related complications like alloimmunization.¹⁰⁶ There is little data regarding preoperative transfusions in HbSC disease. A retrospective study suggests that HbSC patients undergoing abdominal surgeries should be transfused.¹⁰⁷ The higher hemoglobin level of the typical HbSC patient necessitates exchange transfusion to avoid hyperviscosity.

• Is hydroxyurea therapy indicated in this patient?

• Has it been dosed appropriately?

If the patient had the HbSS subtype, hydroxyurea would be clearly indicated, given his frequent pain events.²⁰ HbSC patients may be placed on hydroxyurea on a case-by-case basis, but evidence for its efficacy in this sickle cell subtype is lacking.¹⁰⁸ Large clinical trials like the Multi-Center Study of Hydroxyurea (MSH) that established the safety and efficacy of hydroxyurea in sickle cell anemia excluded HbSC and HbSβ⁺ patients.¹⁰⁹ These mild to moderate subtypes produce less HbF at baseline, and typically have a minimal to modest rise in HbF on hydroxyurea.¹¹⁰ In sickle cell anemia, hydroxyurea is titrated to maximum tolerated dose, defined as an ANC of 2000 to 4000/µL and an ARC of 70,000/µL or higher.⁵³ Because of their lower levels of chronic inflammation and lower reticulocyte counts due to higher hemoglobin levels, many HbSC and HbSβ⁺ patients have values in that range before initiating hydroxyurea therapy.⁹ Cytopenias, particularly of platelets in HbSC, occur at low doses of hydroxyurea.¹¹¹

Of note, although the half-life of hydroxyurea would suggest that 3 times daily dosing is indicated, daily dosing has been found to have equal response and is preferred. Another concern is the monitoring of this myelosuppressive medication. This patient has repeatedly failed to obtain a primary care physician or a hematologist, and hydroxyurea requires laboratory monitoring at least every 2 months, especially in a HbSC patient with a very large spleen who is at significant risk for thrombocytopenia and neutropenia.⁹

CASE CONTINUED

A week after discharge from his admission for abdominal pain diagnosed as splenic sequestration, the patient presents again to the emergency department with abdominal pain which he reports is his typical sickle cell pain. Hemoglobin is 13.8 g/dL, platelet count is 388,000/µL, and alanine aminotransferase (ALT) and aspartate aminotransferase (AST) levels are both 10 times their prior value. Creatinine is 1.2 mg/dL (0.75 mg/dL on his prior admission), and total bilirubin is 3 mg/dL, with 0.3 mg/dL direct bilirubin. He undergoes an ultrasound exam of his gallbladder, which reveals sludge and a possible gallstone. There is no evidence of cholecystitis. General surgery performs a laparoscopic cholecystectomy.

• Was this cholecystectomy necessary?

In patients with sickle cell disease, symptomatic gallstones and gallbladder sludge should be observed; recurrent abdominal pain without a significant change in bilirubin may not be due to gallstones or sludge, and therefore may not be relieved by cholecystectomy.^112,113 In sickle cell disease, 40% of patients with gallbladder sludge do not develop gallstones.⁸⁷ The patient’s bilirubin level was at baseline, and there was no increase in the direct (conjugated) fraction. Watchful waiting would have been appropriate, with cholecystectomy being performed if he experienced recurrent symptoms associated with fatty foods accompanied by an elevation in direct bilirubin.

More concerning and deserving of investigation was his elevated liver enzymes. Patients with sickle cell disease may experience recurrent ischemia and reperfusion injuries in the liver, which is called right upper quadrant syndrome. On autopsy of 70 sickle cell patients, 91% had hepatomegaly and 34% had focal necrosis.¹¹⁴ AST is often elevated in sickle cell disease, as it is affected by hemolysis. In this patient, both AST and ALT are elevated, consistent with a hepatocellular disorder. His abdominal pain and ALT rise may be a sign of a hepatic crisis.¹¹⁵ Rapid resolution of ALT elevation in a matter of days suggests a vaso-occlusive, inflammatory event that is self- limiting. Prolonged AST elevation requires further investigation, with consideration of autoimmune hepatitis, viral hepatitis, or iron overload. Iron overload is unlikely in this patient given his lifetime history of only 1 transfusion. Hepatic iron overload typically occurs in sickle cell disease after a minimum of 10 transfusions.¹¹⁵

CASE CONTINUED

The patient is discharged on the day after the procedure, with instructions to continue his hydroxyurea.

• Should the patient resume hydroxyurea therapy?

Hydroxyurea is hepatically cleared and thus it should be held until his liver function tests normalize.¹⁰⁶

CASE CONTINUED

Two months later, the patient presents to the emergency department with abdominal pain that moves to his left leg. A CBC is obtained, showing a hemoglobin of 11.8 g/dL and a platelet count of 144,000/µL. He is given 2 doses of morphine 6 mg intravenously, and reports that his leg pain is now a 4/10. He is discharged home with a prescription for hydrocodone/acetaminophen.

• Is the emergency department evaluation sufficient?

This patient remains at high risk for splenic sequestration,⁹³ with a hemoglobin 2 g lower than it was 2 months ago and platelets less than half. This decline could be consistent with early splenic sequestration.²⁰ Additionally, he had elevated liver function tests on a recent admission, as well as rising creatinine, without evidence of resolution. It is not appropriate to discharge him without checking a chemistry and liver panel, and abdominal imaging should be considered. The best plan would be to admit him for observation, given his risk for splenic sequestration, and consult surgery for an elective splenectomy if he has a second episode of splenic sequestration 2 months after the first.¹⁰⁰ His abdominal pain that migrates to his left leg could be due to his massive splenomegaly compressing his left kidney, as noted on imaging during his recent admission for splenic sequestration

CASE CONTINUED

An hour after discharge from the emergency department, EMS is called to his home for intractable pain. He is found lying on the floor, and reports excruciating left leg pain. He is brought to the closest hospital, a community hospital that he has not visited previously. There, he is admitted for hydration and pain control and placed on hydromorphone 2 mg every 4 hours as needed for pain. His hemoglobin is 10.8 g/dL, and platelets are 121,000/µL. A chemistry panel is remarkable for a creatinine level of 1.5 mg/dL and a potassium level of 3.2 mEq/L. Liver function tests are not obtained. After 3 doses of hydromorphone, he falls asleep. He is not in a monitored bed, and intravenous fluids, while ordered, are not started. At 6:30 AM the day after admission, he cannot be aroused on a routine vital sign check; he has an SpO₂ of 60%, a blood pressure of 80/60 mm Hg, and heart rate of 148 beats/min. A rapid response is called, and naloxone is administered along with oxygen by face mask and several fluid boluses. His systolic blood pressure increases to 100 mm Hg from a low of 70 mm Hg. His SpO₂ increases to 92%, and he is arousable and alert, although he reports 10/10 leg pain. His abdomen is noted to be distended and tender.

• What may have contributed to his clinical condition?

The patient is opioid tolerant and has received equivalent doses of opioids in the past without excess sedation. He may have liver dysfunction making him unable to metabolize opioids effectively. His hemoglobin and platelets continue to decline, raising concern for splenic sequestration versus sepsis. Failure to place him on a monitor allowed his hypoxia to continue for an unknown amount of time, placing him at high risk for developing ACS. Lack of intravenous hydration while he was too sedated to drink likely exacerbated his sickling.

At 9:20 AM, a CBC is obtained and reveals a hemoglobin of 4.8 g/dL and a platelet count of 44,000/µL. Two units of stat O negative blood are administered, and preparations are made to administer an exchange transfusion. A liver panel is obtained 3 hours later, which reveals an AST level of 1200 U/L and an ALT level of 1050 U/L. His bilirubin is 10 mg/dL, and his lactate dehydrogenase level is 1800 U/L. His urine is dark and is positive for bilirubin and ketones. He is transferred to the intensive care unit. A chest X-ray shows pulmonary congestion. Hematology/oncology is consulted.

He receives a 7-unit red blood cell exchange, which reduces his HbS to 11%. He continues to be hypotensive, and requires norepinephrine to support his blood pressure. Antibiotic therapy is started. His creatinine concentration rises to 2.3 mg/dL, potassium is 7.8 mEq/L, and bicarbonate is 12 mEq/L. He is placed on hemodialysis.

Computed tomography of the chest and abdomen reveals lower posterior lung infiltrates and a grossly enlarged spleen. He requires intubation. He is given a diagnosis of ACS in addition to kidney failure, liver failure, and “sickle crisis.” He continues to require daily to twice daily transfusions to maintain a hemoglobin of 7 to 9 g/dL, and his abdominal distension increases. As his condition worsens, surgery is consulted to discuss a liver transplant. He is deemed to not be a surgical candidate, and he passes away 6 days after entering the hospital. The immediate cause of death is listed as vaso-occlusive crisis, with ACS and sickle crisis listed as contributors.

• Are the causes of death accurate and complete?

If vaso-occlusive crisis is used to indicate a pain event, it is not an accurate cause of death. Pain is one of the most distressing complications of sickle cell disease, and frequent pain events are associated with early mortality,^4,80 but they are not in themselves fatal. ACS is the number one cause of death in sickle cell disease,⁴ and it likely contributed to this patient’s death. Sickle crisis is a vague term that should not be used in this context. Causes of death should include splenic sequestration and multisystem organ failure. Multisystem organ failure in sickle cell disease often responds to aggressive transfusion therapy, which this patient received.^116–118

CONCLUSION

Sickle cell disease is a complex chronic disease that impacts almost every organ system in the body. Clinicians may be inclined to attribute most pain in a patient with sickle cell disease to a simple vaso-occlusive crisis, treat them for this, and not investigate further. As the case presented here demonstrates, failure to identify the actual life-threatening process occurring in a patient with sickle cell disease presenting with pain can result in preventable early mortality. Clinicians must approach a sickle cell patient reporting pain in a thoughtful manner, and consider a complete differential diagnosis, including both sickle cell disease complications and those unrelated to sickle cell disease. Knowledge of the disease courses of the different sickle cell genotypes is essential, and must go beyond a superficial hierarchy of severity, but rather include an understanding of the complications each genotype is most prone to, and at what ages. Complete laboratory assessment, including a comprehensive metabolic panel, should be performed on all admitted patients, not just a complete blood count. Treating pain with high-dose opioids, while appropriate in an uncomplicated pain crisis, can lead to ACS or even respiratory failure in a patient with uninvestigated liver and kidney dysfunction. The most important lesson to remember is that even the sickle cell disease patient who has been given the unfortunate and pejorative label of “frequent flyer” by some providers has the potential for rapid deterioration into multisystem organ failure and death.

Over 20 million endoscopic procedures are performed in the United States annually. A small but significant number of patients undergoing these procedures will have valvular heart disease. Bacteremia can occur after endoscopic procedures, and this may carry an increased risk for the development of infective endocarditis in patients with valvular heart disease.

Throughout the 1980s to 2000s, based on expert opinion and the related American Heart Association (AHA) guidelines in 1997, many patients with valvular abnormalities would routinely receive antibiotics before endoscopy. There would be much anxiety among patients and providers to choose the right antibiotic and not to “miss” a patient with cardiac issues, especially those scheduled as open access. Antibiotics were stocked in GI labs, and questionnaires or intake forms inquired about cardiac history.

There are also significant implications for savings in costs and time as well as for reducing adverse events associated with antibiotics. Theoretically, the new guidelines also improved access to endoscopy as those patients who might have avoided procedures due to the perceived risk of endocarditis or the need for prophylactic antibiotics no longer had that concern.

After decades of pre-endoscopy prophylactic antibiotic use, clinical acceptance of the new guidelines took time. During training, I remember concerned looks from patients who had previously routinely received antibiotics. In addition, some consulting cardiologists and referring physicians were either not aware of the new guidelines or were not yet comfortable with the new recommendations, despite the fact that “brushing your teeth leads to more instances of bacteremia than endoscopy” was well known.

Despite the slow start, inappropriate use of prophylactic antibiotics was phased out, and postendoscopy endocarditis did not appear as a new problem, thereby confirming the wisdom of the new approach. This example shows how one article that appeared in the June 2007* issue of GI & Hepatology News introduced an important recommendation that resulted in a major shift in the practice of endoscopy.

Gyanprakash A. Ketwaroo, MD, MSc is an assistant professor in the division of gastroenterology and hepatology at Bayor College of Medicine, Houston, and an Associate Editor of GI & Hepatology News.

*This story was corrected on 1/10/16.

Over 20 million endoscopic procedures are performed in the United States annually. A small but significant number of patients undergoing these procedures will have valvular heart disease. Bacteremia can occur after endoscopic procedures, and this may carry an increased risk for the development of infective endocarditis in patients with valvular heart disease.

Throughout the 1980s to 2000s, based on expert opinion and the related American Heart Association (AHA) guidelines in 1997, many patients with valvular abnormalities would routinely receive antibiotics before endoscopy. There would be much anxiety among patients and providers to choose the right antibiotic and not to “miss” a patient with cardiac issues, especially those scheduled as open access. Antibiotics were stocked in GI labs, and questionnaires or intake forms inquired about cardiac history.

There are also significant implications for savings in costs and time as well as for reducing adverse events associated with antibiotics. Theoretically, the new guidelines also improved access to endoscopy as those patients who might have avoided procedures due to the perceived risk of endocarditis or the need for prophylactic antibiotics no longer had that concern.

After decades of pre-endoscopy prophylactic antibiotic use, clinical acceptance of the new guidelines took time. During training, I remember concerned looks from patients who had previously routinely received antibiotics. In addition, some consulting cardiologists and referring physicians were either not aware of the new guidelines or were not yet comfortable with the new recommendations, despite the fact that “brushing your teeth leads to more instances of bacteremia than endoscopy” was well known.

Despite the slow start, inappropriate use of prophylactic antibiotics was phased out, and postendoscopy endocarditis did not appear as a new problem, thereby confirming the wisdom of the new approach. This example shows how one article that appeared in the June 2007* issue of GI & Hepatology News introduced an important recommendation that resulted in a major shift in the practice of endoscopy.

Gyanprakash A. Ketwaroo, MD, MSc is an assistant professor in the division of gastroenterology and hepatology at Bayor College of Medicine, Houston, and an Associate Editor of GI & Hepatology News.

*This story was corrected on 1/10/16.

Over 20 million endoscopic procedures are performed in the United States annually. A small but significant number of patients undergoing these procedures will have valvular heart disease. Bacteremia can occur after endoscopic procedures, and this may carry an increased risk for the development of infective endocarditis in patients with valvular heart disease.

Throughout the 1980s to 2000s, based on expert opinion and the related American Heart Association (AHA) guidelines in 1997, many patients with valvular abnormalities would routinely receive antibiotics before endoscopy. There would be much anxiety among patients and providers to choose the right antibiotic and not to “miss” a patient with cardiac issues, especially those scheduled as open access. Antibiotics were stocked in GI labs, and questionnaires or intake forms inquired about cardiac history.

There are also significant implications for savings in costs and time as well as for reducing adverse events associated with antibiotics. Theoretically, the new guidelines also improved access to endoscopy as those patients who might have avoided procedures due to the perceived risk of endocarditis or the need for prophylactic antibiotics no longer had that concern.

After decades of pre-endoscopy prophylactic antibiotic use, clinical acceptance of the new guidelines took time. During training, I remember concerned looks from patients who had previously routinely received antibiotics. In addition, some consulting cardiologists and referring physicians were either not aware of the new guidelines or were not yet comfortable with the new recommendations, despite the fact that “brushing your teeth leads to more instances of bacteremia than endoscopy” was well known.

Despite the slow start, inappropriate use of prophylactic antibiotics was phased out, and postendoscopy endocarditis did not appear as a new problem, thereby confirming the wisdom of the new approach. This example shows how one article that appeared in the June 2007* issue of GI & Hepatology News introduced an important recommendation that resulted in a major shift in the practice of endoscopy.

Gyanprakash A. Ketwaroo, MD, MSc is an assistant professor in the division of gastroenterology and hepatology at Bayor College of Medicine, Houston, and an Associate Editor of GI & Hepatology News.

*This story was corrected on 1/10/16.