INTRODUCTION

Gastric cancer is the fifth most common cancer worldwide and the third leading cause of cancer death in both females and males.¹ More than 70% of gastric cancer cases occur in the developing world, with approximately 50% occurring in East Asia.² Gastric cancer is less common in the United States, with an incidence of 12.3 cases in males and 6.0 cases in females per 100,000 per year and a disproportionately higher incidence in Asians.³ According to the Surveillance, Epidemiology, and End Results Program, approximately 26,370 new cases of stomach cancer were diagnosed in the United States in 2016, and an estimated 10,730 people died of this disease.⁴ Since the 1970s, the 5-year relative survival rate for gastric cancer in the United States has improved from 15% in 1975 to 29% in 2009.⁵ In contrast, in Japan and Korea, where screening programs have been implemented, the 5-year survival rate approaches 70%.⁶

RISK FACTORS AND CLASSIFICATION

A variety of risk factors have been linked to gastric cancer. Diets high in salt, salt-preserved foods, and/or processed meats have been associated with an increased risk for developing gastric cancer.^7,8 Obesity and smoking have also been implicated in gastric cancer.^9,10 Several studies have demonstrated a strong association between Helicobacter pylori and the development of gastric cancer.^11–13 It is believed that H. pylori infection leads to chronic active gastritis, atrophic gastritis, and intestinal metaplasia. Interestingly, mass eradication of H. pylori has not been shown to reduce the risk for gastric cancer.¹⁴ Therefore, treatment of H. pylori should only be considered in patients with active peptic ulcer disease.¹⁵ Other risk factors include Epstein-Barr virus (EBV), prior gastric surgery, and radiation exposure.^16–18 Family history of gastric cancer, hereditary nonpolyposis colon cancer, Li-Fraumeni syndrome, and hereditary diffuse gastric cancer caused by mutations in the E-cadherin gene increase the risk.¹⁷

The anatomic distinction between gastric cancer and cancer of the gastroesophageal junction (GEJ) has been a topic of debate. The Siewert classification is the most widely used system and divides GEJ adenocarcinoma into 3 categories:²⁰ type I tumor: adenocarcinoma of distal esophagus, located 1 cm to 5 cm above the GEJ; type II tumor: true carcinoma of gastric cardia, located within 1 cm above and 2 cm below the GEJ; type III tumor: subcardial gastric carcinoma, located 2 cm to 5 cm below the GEJ, and infiltrates esophagus from below.

The American Joint Committee on Cancer (AJCC) has updated the latest (7th) edition of TMN staging for stomach cancer to include tumors arising more than 5 cm distally of the GEJ or within 5 cm of the GEJ but without extension to the esophagus or GEJ.²¹

In the following sections, neoadjuvant and adjuvant therapy in gastric cancer are discussed using a case presentation to illustrate important concepts.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 43-year old male with no significant past medical history presents with epigastric abdominal pain and heart burn for the past few weeks. He denies nausea, vomiting, melena, or hematochezia. His primary care physician (PCP) diagnoses him with gastroesophageal reflux disease (GERD) and initiates a trial of pantoprazole. Over the next 2 to 3 months, his symptoms do not improve and he has an associated 40-lb weight loss. Both social history and family history are noncontributory. Physical exam reveals epigastric tenderness without rebound or guarding. Laboratory evaluation reveals a hemoglobin of 12.6 g/dL with a mean corpuscular volume of 72 fL. A comprehensive chemistry profile is within normal limits. Given the constellation of presenting symptoms, especially the unintentional weight loss and the presence of microcytic anemia, his PCP suspects a malignant process and refers the patient to a gastroenterologist.

• What are the next appropriate steps for diagnosis?

The most common presenting symptoms of gastric cancer are weight loss and abdominal pain.²² Less commonly, patients exhibit nausea, anorexia, and dysphagia with proximal tumors. Melena is seen in only about 20% of patients. In Japan, where gastric cancer is more prevalent, mass screening programs allow for detection at an earlier stage, which partially accounts for the better survival rates seen in Asia as compared to the United States. Diagnostic work-up includes esophagogastroduodenoscopy (EGD) to assess Siewert category and to obtain a tissue sample for diagnosis. Full staging requires a complete blood count (CBC) with differential; comprehensive chemistry profile; computed tomography (CT) of chest/abdomen/pelvis with oral and intravenous contrast; endoscopic ultrasound (EUS) if no M1 disease is identified; positron emission tomography (PET)-CT if there is no evidence of M1 disease and if clinically indicated; and laparoscopy with cytology for clinical stage T1b or higher.²³ Patients should be staged according to the TMN staging system (Table 1).

CASE CONTINUED

The patient undergoes EGD, which reveals a large ulcerated, partially circumferential mass measuring approximately 4 cm. The mass extends from the gastric body to the cardia. Biopsy of the mass reveals poorly differentiated adenocarcinoma as well as H. pylori–associated gastritis. He is given antibiotic therapy and undergoes complete work-up of his newly diagnosed gastric adenocarcinoma. CT of the chest/abdomen/pelvis demonstrates a large gastric mass with gastrohepatic and distal perigastric adenopathy, compatible with locally advance primary gastric cancer. There is no evidence of distant metastasis. PET scan shows a large hypermetabolic mass in the stomach body and increased FDG activity in 3 small nodes along the lesser gastric curvature and in 1 node in the gastrohepatic region. EUS reveals a malignant gastric tumor in the body of the stomach, which is staged as T3, and a few malignant-appearing lymph nodes in the perigastric region. Fine-needle aspiration of the perigastric lymph node is performed and the sample obtained is positive for malignant cells. Diagnostic laparoscopy with peritoneal washings is performed and cytology is negative for malignant cells. The patient is staged as clinical stage IIB (T3N1M0).

• How should this patient with newly diagnosed, locally advanced, resectable gastric cancer be managed?

SURGERY

Surgical resection for localized gastric cancer is the mainstay of treatment with curative intent. Only very early stage (Tis or T1a) tumors can be considered for endoscopic mucosal resection. Regarding surgical resection, distal gastric cancers are typically treated with subtotal gastrectomy because there is no survival difference between subtotal and total gastrectomy.^24,25 Moreover, subtotal gastrectomy is associated with better nutritional status and quality of life. For proximal tumors, total gastrectomy is preferred as subtotal gastrectomy has been associated with a higher incidence of reflux esophagitis and anastomotic stenosis.²⁶ In terms of surgical approach, multiple studies have shown that a laparoscopic approach has a lower complication rate and similar outcomes in terms of cancer recurrence and long-term survival when compared to open gastrectomy.^27–29 Thus, a laparoscopic approach is often used in academic centers with highly experienced surgeons.

The extent of lymph node dissection remains a topic of debate. A D1 dissection involves the removal of perigastric lymph nodes. A D2 dissection is a D1 dissection plus the removal of lymph nodes along the left gastric artery, common hepatic artery, celiac artery, splenic hilum, and splenic artery. D2 lymphadenectomy has become the standard of care in Eastern countries where gastric cancer is more prevalent, such as Japan and Korea.³⁰ In Western countries, including the United States, less extensive lymphadenectomies are performed. Both randomized clinical trials and meta-analyses have failed to demonstrate an overall survival advantage of D2 dissection over D1 dissection.^31,32 A Dutch trial by Bonenkamp et al involving 711 patients, one of the largest randomized trials of D1 and D2 lymphadenectomy, showed that D2 patients had a higher operative mortality rate than D1 patients (10% versus 4%, P = 0.004) and experienced more complications (43% versus 25%, P < 0.001).³³ In a 15-year follow-up of this study, patients who had a D2 resection had lower locoregional recurrence and gastric-cancer–related death rates compared to those who had a D1 resection; however, D2 resection was associated with a significantly higher operative mortality and complication rate compared to D1.³⁴ In addition, a 2015 Cochrane meta-analysis has demonstrated improved disease-specific survival (DSS) with D2 dissection (hazard ratio [HR] 0.81 [95% confidence interval {CI} 0.71 to 0.92]).³⁵ Currently, the National Comprehensive Cancer Network (NCCN) recommends a D1 or a modified D2 gastrectomy with at least 15 lymph nodes removed for examination, with D2 lymphadenectomies only to be performed at experienced centers.²³

SYSTEMIC CHEMOTHERAPY

Locally advanced gastric cancer (T3-T4 or node positive) requires systemic chemotherapy in addition to surgery, as this intervention improves the 5-year overall survival by 10% to 15%.³⁶ Systemic therapy should also be considered in patients with T2N0 disease with high-risk features: poorly differentiated or high-grade cancer; lymphovascular invasion; neural invasion; age younger than 50 years; and patients who did not undergo D2 dissection.²³ Currently, there is no global consensus on the best treatment approach. In the United States, where a less aggressive lymph-node dissection is performed, adjuvant chemoradiotherapy after surgery is more commonly seen. In Europe, perioperative (preoperative and postoperative) chemotherapy is the standard treatment. In Japan, adjuvant chemotherapy after D2 lymphadenectomy is the standard of care.³⁷ These regional preferences are largely due to randomized clinical trials that have shown benefit for each approach. The landmark trials are discussed in the following sections and are summarized in Table 2.

Neoadjuvant chemotherapy has the benefit of “downstaging” locally advanced tumors to allow for curative resection. Phase 2 clinical trials have also demonstrated good pathologic response rates and high R0 resection rates following neoadjuvant chemotherapy.^38,39 However, phase 3 trials to support this treatment approach are lacking. In the European Organisation for Research and Treatment of Cancer (EORTC) 40954 trial, patients with stage III or IV gastric or GEJ cancer were randomly assigned to surgery with or without preoperative cisplatin, leucovorin, and infusional fluorouracil (5-FU).⁴⁰ The trial was stopped early due to poor accrual after 144 patients were randomized. The neoadjuvant chemotherapy arm had a higher R0 resection rate compared to the surgery alone arm (82% versus 67%, respectively, P = 0.036) but a higher postoperative complication rate (27% versus 16%, respectively, P = 0.09). More important, after a median follow-up of 4.4 years, a survival benefit could not be shown, with 2-year survival rates of 72.7% and 69.9% in the neoadjuvant and surgery-only arms, respectively (HR 0.84 [95% CI 0.52 to 1.35], P = 0.466). Due to the lack of large trials, a meta-analysis assessing the effectiveness of neoadjuvant chemotherapy combined with surgery versus surgery alone in advanced gastric and gastroesophageal cancer was performed.⁴¹ The analysis included 12 randomized controlled trials with a total of 1820 patients. Neoadjuvant chemotherapy was shown to slightly improve the survival rate (odds ratio [OR] 1.32 [95% CI 1.07 to 1.64], P = 0.01). It significantly improved the 3-year progression-free survival (PFS; OR 1.85 [95% CI 1.39 to 2.46], P < 0.0001), tumor down-staging rate (OR 1.71 [95% CI 1.26 to 2.33], P = 0.0006), and R0 resection rate (OR 1.38 [95% CI 1.08 to 1.78], P = 0.01). There were no differences between the 2 arms in terms of relapse rates, operative complications, perioperative mortality, and grade 3/4 adverse effects. While these results are encouraging, further randomized clinical trials are needed to clarify the role of neoadjuvant chemotherapy and its impact on overall survival.

Perioperative Chemotherapy

The results of the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial published in 2006 established perioperative chemotherapy as standard of care in patients with resectable gastric and gastroesophageal adenocarcinoma.⁴² A total of 503 patients with potentially resectable gastric and lower esophageal adenocarcinoma were randomly assigned to perioperative chemotherapy plus surgery or surgery alone. Perioperative chemotherapy consisted of 3 preoperative and postoperative cycles of epirubicin, cisplatin, and infusional 5-FU (ECF). At a median follow-up of 4 years, the perioperative-chemotherapy group had a significantly better PFS (HR 0.66 [95% CI 0.53 to 0.81], P < 0.001) as well as overall survival (HR 0.75 [95% CI 0.60 to 0.93], P = 0.009). The 5-year overall survival rate was 36.3% in the perioperative chemotherapy group and 23% in the surgery group. Of note, there was a greater proportion of stage T1/T2 tumors (52% versus 37%, P = 0.002) and N0/N1 disease (84% versus 71%) in the perioperative-chemotherapy group compared to the surgery alone group. In addition, only 42% of patients in the perioperative chemotherapy group completed all 6 cycles of chemotherapy.

The administration of ECF is often difficult since the 5-FU component requires a central venous access device and an ambulatory infusion pump and the cisplatin component is associated with nephrotoxicity and ototoxicity. The REAL-2 trial was a randomized phase 3 clinical trial that assessed whether 5-FU could be replaced by capecitabine and cisplatin by oxaliplatin in the ECF regimen.⁴³ Between June 2000 and May 2005, a total of 1002 patients with locally advanced esophageal/GEJ/gastric cancer were enrolled. Patients were randomly assigned to 1 of 4 triplet therapies: epirubicin and cisplatin plus either 5-FU (ECF) or capecitabine (ECX) or epirubicin and oxaliplatin plus either 5-FU (EOF) or capecitabine (EOX). After a median follow-up of approximately 18 months, the overall survival in the capecitabine groups did not differ significantly from that in the 5-FU groups (HR 0.88 [95% CI 0.77 to 1.00], P = 0.06), nor did overall survival in the oxaliplatin groups differ significantly from that in the cisplatin groups (HR 0.91 [95% CI 0.79 to 1.04], P = 0.16). Interestingly, the 1-year survival rate was longer in the EOX group than in the ECF group (46.8% versus 37.7%, respectively; HR 0.80 [95% CI 0.66 to 0.97], P = 0.02). This translated into an overall survival of 11.2 months for the EOX group and 9.9 months for the ECF group. Therefore, EOX is preferred over ECF in clinical practice.

The French FNLCC/FFCD trial published in 2011 provided further support for perioperative chemotherapy.⁴⁴ A total of 224 patients with adenocarcinoma of the lower esophagus, GEJ, or stomach were randomly assigned to perioperative chemotherapy plus surgery or surgery alone. The perioperative-chemotherapy group received 2 to 3 cycles of preoperative chemotherapy and 3 to 4 cycles of postoperative chemotherapy, consisting of infusional 5-FU (800 mg/m² daily for days 1 to 5) and cisplatin (100 mg/m² on day 1). In patients receiving preoperative chemotherapy, 38% experienced at least grade 3 to 4 toxicity. Among the 109 patients who received at least 1 cycle of preoperative chemotherapy, only 54 patients (50%) received postoperative chemotherapy. Despite this, the perioperative-chemotherapy group had a statistically significant higher R0 resection rate (84% versus 74%, P = 0.04) compared to the surgery alone group. At a median follow-up of 5.7 years, the perioperative chemotherapy group had an improved overall survival (HR for death 0.69 [95% CI 0.50 to 0.95], P = 0.02) and disease-free survival (DFS; HR for recurrence or death 0.65 [95% CI 0.48 to 0.89], P = 0.003). This translated into 5-year overall survival rates of 38% versus 24% and 5-year DFS rates of 34% versus 19%. One caveat to this study is that the majority of patients (64%) had GEJ cancer and only 25% had gastric cancer. In the multivariate analysis, the 2 significant prognostic factors for overall survival were the administration of preoperative chemotherapy (P = 0.01) and tumor site at the GEJ (P < 0.01).

Adjuvant Chemoradiotherapy

The INT-0116 (Intergroup 0116) study published in 2001 established adjuvant chemoradiotherapy as the standard approach for resectable gastric cancer in the United States. In this study, a total of 556 patients with resected gastric or GEJ cancer were randomly assigned to surgery alone or surgery followed by adjuvant 5-FU/leucovorin bolus chemotherapy, sandwiched with 5-FU–based chemoradiation (45 Gy).⁴⁵ In the chemoradiotherapy group, 64% of patients completed treatment and grade 3 and 4 toxicity occurred in 41% and 32%, respectively. However, only 3 patients (1%) died from treatment-related toxicity. At a median follow-up of 5 years, the median overall survival was 36 months in the chemoradiation group and 27 months in the surgery group. Overall survival rate was 50% in the combined modality group and 41% in the surgery-alone group, with a HR of 1.35 (95% CI 1.09 to 1.66, P = 0.005). The 3-year DFS was 48% in the chemoradiotherapy group and 31% in the surgery-alone group, corresponding to a DFS of 30 months and 19 months, respectively. Even after a median follow-up of 10 years, these positive results were maintained, with a HR for survival of 1.32 (95% CI 1.10 to 1.60, P = 0.0046) and HR for DFS of 1.51 (95% CI 1.25 to 1.83, P < 0.001).⁴⁶ A criticism of the INT-0116 study is that 54% of patients had less than a D1 lymph node dissection, suggesting that adjuvant chemoradiation may have compensated for suboptimal surgery.

CALGB 80101, a United States Intergroup study, compared the INT-0116 protocol regimen (bolus 5-FU/leucovorin with 5-FU plus concurrent radiotherapy) to postoperative ECF sandwiched with 5-FU plus concurrent radiotherapy.⁴⁷ The study included patients with resected gastric or GEJ adenocarcinoma that extended beyond the muscularis propria or was node positive. The percentage of patients with gastric versus GEJ cancer was not reported. A total of 546 patients were randomized. Preliminary results were presented at the 2011 American Society of Clinical Oncology meeting. The ECF arm had lower rates of grade 3/4 toxicities, including neutropenia, diarrhea, and mucositis. However, there was no difference in overall survival (3-year overall survival of 52% versus 50% for ECF and 5-FU/leucovorin, respectively) or DFS (3-year DFS of 47% versus 46% for ECF and 5-FU/leucovorin, respectively). The trial was not adequately powered to assess noninferiority. The location of the primary tumor (GEJ versus proximal versus distal stomach) did not have any effect on treatment outcome.

The Adjuvant Chemoradiation Therapy in Stomach Cancer (ARTIST) trial was the first study to compare adjuvant chemoradiotherapy with adjuvant chemotherapy in patients with D2-resected gastric cancer.⁴⁸ A total of 458 patients were randomly assigned to 6 cycles of XP chemotherapy (capecitabine 2000 mg/m² per day on days 1–14 and cisplatin 60 mg/m² on day 1, every 3 weeks) or XP/radiotherapy/XP (2 cycles of XP followed by 45 Gy radiotherapy with concurrent daily capecitabine [825 mg/m² twice daily] and 2 cycles of XP). After a median follow-up of 84 months, there was no difference in DFS or overall survival between treatment arms (HR for progression 0.74 [95% CI 0.52 to 1.05], P = 0.09; HR for death 1.13 [95% CI 0.78 to 1.65], P = 0.53).⁴⁹ However, subgroup analysis showed that chemoradiotherapy significantly improved DFS in patients with node-positive disease (3-year DFS 76% versus 72%, P = 0.004).

Adjuvant Chemotherapy

Data supporting the use of adjuvant chemotherapy alone is largely derived from trials done in Asia, typically after a D2 lymph node dissection, and thus adjuvant chemotherapy has become the standard of care in that region. In the Japanese Adjuvant Chemotherapy Trial of S-1 for Gastric Cancer (ACTS-GC), a total of 1059 patients with stage II or III gastric cancer who had undergone surgery with a D2 lymphadenectomy were randomly assigned to 1 year of S-1 (an oral fluoropyrimidine) or surgery alone.⁵⁰ The 5-year overall survival rate was 72% in the S-1 group and 61% in the surgery-only group (HR 0.669 [95% CI 0.54 to 0.83]).⁵¹ The 5-year relapse-free survival was 65% in the S-1 group and 53% in the surgery-only group (HR 0.65 [95% CI 0.537 to 0.793]). Of note, both arms of the ACTS-GC trial had significantly higher 5-year overall survival rates compared to the INT-0116 and MAGIC trials: 43% versus 28% and 36% versus 23% for the treatment and control groups, respectively.^42,45 Consequently, it is unclear if the benefit of adjuvant chemotherapy can be translated to Western countries.

The Korean Capecitabine and Oxaliplatin Adjuvant Study in Stomach Cancer (CLASSIC) trial published in 2012 also established the role of adjuvant chemotherapy after D2 gastrectomy.⁵² A total of 1035 patients with stage II-IIIB gastric cancer who had curative D2 gastrectomy were randomly assigned to 8 cycles of adjuvant XELOX (capecitabine 1000 mg/m² twice daily on days 1–14 plus oxaliplatin 130 mg/m² on day 1, 21-day cycle) or surgery alone. Median follow-up was 34 months in both arms and 67% of patients in the chemotherapy arm completed all 8 cycles of planned chemotherapy. The 3-year DFS was 74% in the chemotherapy group and 59% in the surgery-only group (HR 0.56 [95% CI 0.44 to 0.72], P < 0.0001). There was a trend toward improvement in overall survival (83% versus 78%, HR 0.72 [95% CI 0.52 to 1.00]). After 5 years of follow-up, the improvement in overall survival became statistically significant (78% versus 69%, HR 0.66 [95% CI 0.51 to 0.85]).⁵³

The benefit of adjuvant chemotherapy was reinforced by a 2010 meta-analysis comparing adjuvant chemotherapy to surgery alone in patients with resected gastric cancer.⁵⁴ A total of 17 randomized controlled trials were included. Adjuvant fluorouracil-based chemotherapy was associated with a statistically significant improved overall survival (HR 0.82 [95% CI 0.76 to 0.90], P < 0.001) and DFS (HR 0.82 [95% CI 0.75 to 0.90], P < 0.001). Five-year overall survival increased from 49.6% to 55.3% with chemotherapy.

Since data exists for all 3 approaches (perioperative chemotherapy, adjuvant chemoradiotherapy, and adjuvant chemotherapy), various meta-analyses have been done to clarify which approach is the best. In a recent meta-analysis of 6 randomized controlled trials reported between 2010 and 2012, which involved 1171 patients with resected gastric cancer, adjuvant chemotherapy was compared to adjuvant chemoradiotherapy.⁵⁵ Five of the studies were from East Asia, while one was from a Western country. Adjuvant chemoradiation was associated with a lower local-regional recurrence rate (OR 0.46 [95% CI 0.32 to 0.67]) and better 5-year DFS rate (OR 1.56 [95% CI 1.09 to 2.24]). However, there was no statistical difference in 5-year overall survival rate (OR 1.32 [95% CI 0.92 to 1.88]). Similar results were reported by Zhou et al in 2016.⁵⁶ This meta-analysis included 4 randomized controlled trials reported between 2010 and 2015, with a total of 960 patients who had undergone a D2 resection for gastric cancer. Compared to adjuvant chemotherapy, adjuvant chemoradiotherapy significantly reduced the loco-regional recurrence rate (LRRR; relative risk [RR] 0.50 [95% CI 0.34 to 0.74], P = 0.0005) and improved DFS (HR 0.73 [95% CI 0.60 to 0.89], P = 0.002). Again, no difference in overall survival was seen (HR 0.91 [95% CI 0.74 to 1.11], P = 0.34).

Adjuvant chemotherapy and perioperative chemotherapy have also been compared. In a recent meta-analysis of 14 randomized controlled trials (8 Asian, 6 European) involving 2093 patients with resected gastric or GEJ cancer, perioperative chemotherapy was associated with improved overall survival when compared to adjuvant chemotherapy (HR 0.48 [95% CI 0.35 to 0.67], P < 0.001).⁵⁷ The benefit of perioperative chemotherapy over adjuvant chemotherapy has also been reported in a 2016 meta-analysis by Zhao et al.⁵⁸ A total of 1240 patients were included from 5 randomized controlled trials and 6 clinical controlled trials, all from Asian countries. The 5-year overall survival rate was significantly better in the perioperative chemotherapy group compared to the adjuvant chemotherapy group (RR 0.77 [95% CI 0.64 to 0.92], P < 0.01). Furthermore, the 2 groups showed no significant differences in the postoperative complication rates (RR 0.98 [95% CI 0.63 to 1.51], P = 0.91) or adverse effects of chemotherapy (P > 0.05 for all adverse effects).

While these meta-analyses may offer some insight on the best treatment approach, they should be interpreted with caution. Most studies included in these meta-analyses were from Asian countries, and their findings may not be applicable to Western countries. Furthermore, the heterogeneity of trials and inclusion of nonrandomized trials make it difficult to draw conclusions. There are several ongoing trials that will help to define the optimal treatment approach.

CASE CONTINUED

The patient is presented at tumor board and the consensus is to proceed with the perioperative chemotherapy approach. The patient undergoes echocardiography, which reveals a normal ejection fraction. He receives 3 cycles of neoadjuvant EOX (epirubicin, oxaliplatin, and capecitabine). After 3 cycles of neoadjuvant EOX, the patient has a repeat CT that shows marked interval reduction in the size of the primary gastric neoplasm and interval decrease in the size of the small perigastric lymph nodes. He subsequently undergoes a total gastrectomy with J-tube placement. Pathology shows ypT3N0 disease with 0 out of 47 lymph nodes involved and negative margins. He then receives 3 cycles of adjuvant EOX.

• What are the recommendations for surveillance?

According to the current NCCN guidelines, a history and physical exam should be performed every 3 to 6 months for 1 to 2 years, then every 6 to 12 months for 3 to 5 years, and then annually.²³ Labs, CT chest/abdomen, and EGD should be done as clinically indicated. Patients who have undergone surgical resection should be monitored for nutritional deficiencies (vitamin B12 and iron).

GASTROESOPHAGEAL JUNCTION TUMORS

Tumors arising in the GEJ or gastric cardia within 5 cm of the GEJ that extend into the GEJ or distal esophagus are staged and treated as esophageal cancers.²¹ The primary treatment for T1/T2N0 tumors is surgical resection. In patients with T3 or higher or node-positive adenocarcinoma of the GEJ, a combined modality approach is preferred, with preoperative chemoradiotherapy followed by surgical resection.⁵⁹ The CROSS trial demonstrated a significant survival benefit with preoperative chemoradiation using carboplatin/paclitaxel compared to surgery alone in patients with esophageal or GEJ cancer (49 months versus 24 months, HR 0.66, P = 0.003).⁶⁰

ONGOING TRIALS

As mentioned previously, several randomized clinical trials are in progress to clarify the optimal treatment approach. The MAGIC-B/MRC-ST03 is a randomized phase 2/3 trial looking at perioperative epirubicin, cisplatin, and capecitabine (ECX) with or without bevacizumab in patients with resectable lower esophageal, GEJ, or gastric cancer.⁶¹ The TOPGEAR trial, a randomized phase 2/3 study being conducted in Canada and Europe, is comparing perioperative ECF chemotherapy with preoperative chemoradiation plus perioperative ECF chemotherapy.⁶² In Asia, the PRODIGY trial is a phase 3, open-label, randomized study comparing neoadjuvant docetaxel, oxaliplatin, and S-1 followed by surgery and adjuvant S-1 versus surgery plus adjuvant S-1 in patients with locally advanced gastric cancer (T2-T4 or node positive).⁶³ Primary endpoint is PFS and secondary endpoints are overall survival, R0 resection rate, and safety.

Trials comparing adjuvant chemotherapy to adjuvant chemoradiotherapy are also being conducted. In the Dutch CRITICS study, a randomized phase 3 trial, patients with stage Ib-Iva resectable gastric cancer were given 3 cycles of epirubicin, cisplatin/oxaliplatin, and capecitabine (ECC/EOC), followed by D2 resection and either 3 cycles of ECC/EOC or chemoradiation with weekly cisplatin and daily capecitabine.⁶⁴ Between January 2007 and April 2015, a total of 788 patients were enrolled. In a preliminary report presented at ASCO in 2016, the 5-year survival rate was similar between the 2 arms (41.3% for chemotherapy arm and 40.9% for chemoradiotherapy arm, P = 0.99). The Korean ARTIST II trial is comparing adjuvant S-1 and S-1/oxaliplatin with or without radiotherapy in patients with D2-resected gastric cancer.⁶⁵ Similarly, the NCT01711242 trial is comparing adjuvant XELOX alone versus XELOX with concurrent capecitabine/radiotherapy in patients with resected D2 gastric cancer.⁶⁶

The ToGA trial established a survival benefit of trastuzumab in combination with chemotherapy in HER2-positive metastatic gastric cancer.⁶⁷ Consequently, there are ongoing clinical trials to assess the role of trastuzumab in nonmetastatic gastric cancer. The TOXAG study is a phase 2 trial looking at the safety and tolerability of adjuvant oxaliplatin, capecitabine, and trastuzumab with radiation in patients with resected HER2-positive gastric or GEJ adenocarcinoma.⁶⁸ The NCT01130337 clinical trial is evaluating perioperative XELOX/trastuzumab in patients with resectable gastric or GEJ adeno-carcinoma.⁶⁹

Gastric cancer is the fifth most common cancer worldwide, with the greatest incidence in East Asia. Survival outcomes are better in Asian countries when compared to the United States. This difference in survival may be related to the presence of mass screening programs in Asia, which allows for detection at an earlier stage and the use of a more extensive surgical approach (ie, D2 resection). Risk factors for developing gastric cancer include: diets high in salt/salt-preserved foods or processed meats, obesity, smoking, H. pylori infection, EBV, prior gastric surgery, radiation exposure, and positive family history.

According to the latest edition of TMN staging, gastric cancer includes tumors arising more than 5 cm distally of the GEJ or within 5 cm of the GEJ but without extension to the esophagus or GEJ. Diagnostic work-up includes: EGD with biopsy; basic labs; CT chest/abdomen/pelvis with oral and intravenous contrast; EUS if no M1 disease is identified; PET-CT if there is no M1 disease and if clinically indicated; and diagnostic laparoscopy with cytology for clinical stage T1b or higher.

The mainstay of treatment is surgical resection. Laparoscopic approach is preferred over open gastrectomy due to lower complication rates and similar survival outcomes. Current NCCN guidelines recommend a D1 or a modified D2 lymph node dissection with at least 15 lymph nodes removed for examination. Systemic chemotherapy is required in locally advanced gastric cancer (T3-T4 or node positive) and should be considered in T2N0 disease with high-risk features. Currently, there is no global consensus on the optimal treatment approach. Data from various trials have shown benefit for each approach. Regional preferences are: perioperative chemotherapy in Europe; adjuvant chemoradiotherapy in the United States; and adjuvant chemotherapy in Asia. In an effort to better define the optimal treatment approach, several randomized clinical trials are being conducted. According to the current NCCN guidelines, the following treatment approaches are acceptable and are supported by data in the trial listed in parentheses:

• Perioperative chemotherapy

° 5-FU/cisplatin (French FNLCC/FCCD trial)⁴⁴ or

° ECF (MAGIC trial)⁴² or

° ECF modifications: EOX, EOF, ECX (REAL-2 trial)⁴³

• Adjuvant chemoradiotherapy

° 5-FU/leucovorin sandwiched with 5-FU-based chemoradiation (INT-0116 trial)⁴⁵

• Adjuvant chemotherapy (after D2 resection)

° Capecitabine/oxaliplatin (CLASSIC trial)⁵² or

° Capecitabine/cisplatin (ARTIST trial)^48,49

INTRODUCTION

Gastric cancer is the fifth most common cancer worldwide and the third leading cause of cancer death in both females and males.¹ More than 70% of gastric cancer cases occur in the developing world, with approximately 50% occurring in East Asia.² Gastric cancer is less common in the United States, with an incidence of 12.3 cases in males and 6.0 cases in females per 100,000 per year and a disproportionately higher incidence in Asians.³ According to the Surveillance, Epidemiology, and End Results Program, approximately 26,370 new cases of stomach cancer were diagnosed in the United States in 2016, and an estimated 10,730 people died of this disease.⁴ Since the 1970s, the 5-year relative survival rate for gastric cancer in the United States has improved from 15% in 1975 to 29% in 2009.⁵ In contrast, in Japan and Korea, where screening programs have been implemented, the 5-year survival rate approaches 70%.⁶

RISK FACTORS AND CLASSIFICATION

A variety of risk factors have been linked to gastric cancer. Diets high in salt, salt-preserved foods, and/or processed meats have been associated with an increased risk for developing gastric cancer.^7,8 Obesity and smoking have also been implicated in gastric cancer.^9,10 Several studies have demonstrated a strong association between Helicobacter pylori and the development of gastric cancer.^11–13 It is believed that H. pylori infection leads to chronic active gastritis, atrophic gastritis, and intestinal metaplasia. Interestingly, mass eradication of H. pylori has not been shown to reduce the risk for gastric cancer.¹⁴ Therefore, treatment of H. pylori should only be considered in patients with active peptic ulcer disease.¹⁵ Other risk factors include Epstein-Barr virus (EBV), prior gastric surgery, and radiation exposure.^16–18 Family history of gastric cancer, hereditary nonpolyposis colon cancer, Li-Fraumeni syndrome, and hereditary diffuse gastric cancer caused by mutations in the E-cadherin gene increase the risk.¹⁷

The anatomic distinction between gastric cancer and cancer of the gastroesophageal junction (GEJ) has been a topic of debate. The Siewert classification is the most widely used system and divides GEJ adenocarcinoma into 3 categories:²⁰ type I tumor: adenocarcinoma of distal esophagus, located 1 cm to 5 cm above the GEJ; type II tumor: true carcinoma of gastric cardia, located within 1 cm above and 2 cm below the GEJ; type III tumor: subcardial gastric carcinoma, located 2 cm to 5 cm below the GEJ, and infiltrates esophagus from below.

The American Joint Committee on Cancer (AJCC) has updated the latest (7th) edition of TMN staging for stomach cancer to include tumors arising more than 5 cm distally of the GEJ or within 5 cm of the GEJ but without extension to the esophagus or GEJ.²¹

In the following sections, neoadjuvant and adjuvant therapy in gastric cancer are discussed using a case presentation to illustrate important concepts.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 43-year old male with no significant past medical history presents with epigastric abdominal pain and heart burn for the past few weeks. He denies nausea, vomiting, melena, or hematochezia. His primary care physician (PCP) diagnoses him with gastroesophageal reflux disease (GERD) and initiates a trial of pantoprazole. Over the next 2 to 3 months, his symptoms do not improve and he has an associated 40-lb weight loss. Both social history and family history are noncontributory. Physical exam reveals epigastric tenderness without rebound or guarding. Laboratory evaluation reveals a hemoglobin of 12.6 g/dL with a mean corpuscular volume of 72 fL. A comprehensive chemistry profile is within normal limits. Given the constellation of presenting symptoms, especially the unintentional weight loss and the presence of microcytic anemia, his PCP suspects a malignant process and refers the patient to a gastroenterologist.

• What are the next appropriate steps for diagnosis?

The most common presenting symptoms of gastric cancer are weight loss and abdominal pain.²² Less commonly, patients exhibit nausea, anorexia, and dysphagia with proximal tumors. Melena is seen in only about 20% of patients. In Japan, where gastric cancer is more prevalent, mass screening programs allow for detection at an earlier stage, which partially accounts for the better survival rates seen in Asia as compared to the United States. Diagnostic work-up includes esophagogastroduodenoscopy (EGD) to assess Siewert category and to obtain a tissue sample for diagnosis. Full staging requires a complete blood count (CBC) with differential; comprehensive chemistry profile; computed tomography (CT) of chest/abdomen/pelvis with oral and intravenous contrast; endoscopic ultrasound (EUS) if no M1 disease is identified; positron emission tomography (PET)-CT if there is no evidence of M1 disease and if clinically indicated; and laparoscopy with cytology for clinical stage T1b or higher.²³ Patients should be staged according to the TMN staging system (Table 1).

CASE CONTINUED

The patient undergoes EGD, which reveals a large ulcerated, partially circumferential mass measuring approximately 4 cm. The mass extends from the gastric body to the cardia. Biopsy of the mass reveals poorly differentiated adenocarcinoma as well as H. pylori–associated gastritis. He is given antibiotic therapy and undergoes complete work-up of his newly diagnosed gastric adenocarcinoma. CT of the chest/abdomen/pelvis demonstrates a large gastric mass with gastrohepatic and distal perigastric adenopathy, compatible with locally advance primary gastric cancer. There is no evidence of distant metastasis. PET scan shows a large hypermetabolic mass in the stomach body and increased FDG activity in 3 small nodes along the lesser gastric curvature and in 1 node in the gastrohepatic region. EUS reveals a malignant gastric tumor in the body of the stomach, which is staged as T3, and a few malignant-appearing lymph nodes in the perigastric region. Fine-needle aspiration of the perigastric lymph node is performed and the sample obtained is positive for malignant cells. Diagnostic laparoscopy with peritoneal washings is performed and cytology is negative for malignant cells. The patient is staged as clinical stage IIB (T3N1M0).

• How should this patient with newly diagnosed, locally advanced, resectable gastric cancer be managed?

SURGERY

Surgical resection for localized gastric cancer is the mainstay of treatment with curative intent. Only very early stage (Tis or T1a) tumors can be considered for endoscopic mucosal resection. Regarding surgical resection, distal gastric cancers are typically treated with subtotal gastrectomy because there is no survival difference between subtotal and total gastrectomy.^24,25 Moreover, subtotal gastrectomy is associated with better nutritional status and quality of life. For proximal tumors, total gastrectomy is preferred as subtotal gastrectomy has been associated with a higher incidence of reflux esophagitis and anastomotic stenosis.²⁶ In terms of surgical approach, multiple studies have shown that a laparoscopic approach has a lower complication rate and similar outcomes in terms of cancer recurrence and long-term survival when compared to open gastrectomy.^27–29 Thus, a laparoscopic approach is often used in academic centers with highly experienced surgeons.

The extent of lymph node dissection remains a topic of debate. A D1 dissection involves the removal of perigastric lymph nodes. A D2 dissection is a D1 dissection plus the removal of lymph nodes along the left gastric artery, common hepatic artery, celiac artery, splenic hilum, and splenic artery. D2 lymphadenectomy has become the standard of care in Eastern countries where gastric cancer is more prevalent, such as Japan and Korea.³⁰ In Western countries, including the United States, less extensive lymphadenectomies are performed. Both randomized clinical trials and meta-analyses have failed to demonstrate an overall survival advantage of D2 dissection over D1 dissection.^31,32 A Dutch trial by Bonenkamp et al involving 711 patients, one of the largest randomized trials of D1 and D2 lymphadenectomy, showed that D2 patients had a higher operative mortality rate than D1 patients (10% versus 4%, P = 0.004) and experienced more complications (43% versus 25%, P < 0.001).³³ In a 15-year follow-up of this study, patients who had a D2 resection had lower locoregional recurrence and gastric-cancer–related death rates compared to those who had a D1 resection; however, D2 resection was associated with a significantly higher operative mortality and complication rate compared to D1.³⁴ In addition, a 2015 Cochrane meta-analysis has demonstrated improved disease-specific survival (DSS) with D2 dissection (hazard ratio [HR] 0.81 [95% confidence interval {CI} 0.71 to 0.92]).³⁵ Currently, the National Comprehensive Cancer Network (NCCN) recommends a D1 or a modified D2 gastrectomy with at least 15 lymph nodes removed for examination, with D2 lymphadenectomies only to be performed at experienced centers.²³

SYSTEMIC CHEMOTHERAPY

Locally advanced gastric cancer (T3-T4 or node positive) requires systemic chemotherapy in addition to surgery, as this intervention improves the 5-year overall survival by 10% to 15%.³⁶ Systemic therapy should also be considered in patients with T2N0 disease with high-risk features: poorly differentiated or high-grade cancer; lymphovascular invasion; neural invasion; age younger than 50 years; and patients who did not undergo D2 dissection.²³ Currently, there is no global consensus on the best treatment approach. In the United States, where a less aggressive lymph-node dissection is performed, adjuvant chemoradiotherapy after surgery is more commonly seen. In Europe, perioperative (preoperative and postoperative) chemotherapy is the standard treatment. In Japan, adjuvant chemotherapy after D2 lymphadenectomy is the standard of care.³⁷ These regional preferences are largely due to randomized clinical trials that have shown benefit for each approach. The landmark trials are discussed in the following sections and are summarized in Table 2.

Neoadjuvant chemotherapy has the benefit of “downstaging” locally advanced tumors to allow for curative resection. Phase 2 clinical trials have also demonstrated good pathologic response rates and high R0 resection rates following neoadjuvant chemotherapy.^38,39 However, phase 3 trials to support this treatment approach are lacking. In the European Organisation for Research and Treatment of Cancer (EORTC) 40954 trial, patients with stage III or IV gastric or GEJ cancer were randomly assigned to surgery with or without preoperative cisplatin, leucovorin, and infusional fluorouracil (5-FU).⁴⁰ The trial was stopped early due to poor accrual after 144 patients were randomized. The neoadjuvant chemotherapy arm had a higher R0 resection rate compared to the surgery alone arm (82% versus 67%, respectively, P = 0.036) but a higher postoperative complication rate (27% versus 16%, respectively, P = 0.09). More important, after a median follow-up of 4.4 years, a survival benefit could not be shown, with 2-year survival rates of 72.7% and 69.9% in the neoadjuvant and surgery-only arms, respectively (HR 0.84 [95% CI 0.52 to 1.35], P = 0.466). Due to the lack of large trials, a meta-analysis assessing the effectiveness of neoadjuvant chemotherapy combined with surgery versus surgery alone in advanced gastric and gastroesophageal cancer was performed.⁴¹ The analysis included 12 randomized controlled trials with a total of 1820 patients. Neoadjuvant chemotherapy was shown to slightly improve the survival rate (odds ratio [OR] 1.32 [95% CI 1.07 to 1.64], P = 0.01). It significantly improved the 3-year progression-free survival (PFS; OR 1.85 [95% CI 1.39 to 2.46], P < 0.0001), tumor down-staging rate (OR 1.71 [95% CI 1.26 to 2.33], P = 0.0006), and R0 resection rate (OR 1.38 [95% CI 1.08 to 1.78], P = 0.01). There were no differences between the 2 arms in terms of relapse rates, operative complications, perioperative mortality, and grade 3/4 adverse effects. While these results are encouraging, further randomized clinical trials are needed to clarify the role of neoadjuvant chemotherapy and its impact on overall survival.

Perioperative Chemotherapy

The results of the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial published in 2006 established perioperative chemotherapy as standard of care in patients with resectable gastric and gastroesophageal adenocarcinoma.⁴² A total of 503 patients with potentially resectable gastric and lower esophageal adenocarcinoma were randomly assigned to perioperative chemotherapy plus surgery or surgery alone. Perioperative chemotherapy consisted of 3 preoperative and postoperative cycles of epirubicin, cisplatin, and infusional 5-FU (ECF). At a median follow-up of 4 years, the perioperative-chemotherapy group had a significantly better PFS (HR 0.66 [95% CI 0.53 to 0.81], P < 0.001) as well as overall survival (HR 0.75 [95% CI 0.60 to 0.93], P = 0.009). The 5-year overall survival rate was 36.3% in the perioperative chemotherapy group and 23% in the surgery group. Of note, there was a greater proportion of stage T1/T2 tumors (52% versus 37%, P = 0.002) and N0/N1 disease (84% versus 71%) in the perioperative-chemotherapy group compared to the surgery alone group. In addition, only 42% of patients in the perioperative chemotherapy group completed all 6 cycles of chemotherapy.

The administration of ECF is often difficult since the 5-FU component requires a central venous access device and an ambulatory infusion pump and the cisplatin component is associated with nephrotoxicity and ototoxicity. The REAL-2 trial was a randomized phase 3 clinical trial that assessed whether 5-FU could be replaced by capecitabine and cisplatin by oxaliplatin in the ECF regimen.⁴³ Between June 2000 and May 2005, a total of 1002 patients with locally advanced esophageal/GEJ/gastric cancer were enrolled. Patients were randomly assigned to 1 of 4 triplet therapies: epirubicin and cisplatin plus either 5-FU (ECF) or capecitabine (ECX) or epirubicin and oxaliplatin plus either 5-FU (EOF) or capecitabine (EOX). After a median follow-up of approximately 18 months, the overall survival in the capecitabine groups did not differ significantly from that in the 5-FU groups (HR 0.88 [95% CI 0.77 to 1.00], P = 0.06), nor did overall survival in the oxaliplatin groups differ significantly from that in the cisplatin groups (HR 0.91 [95% CI 0.79 to 1.04], P = 0.16). Interestingly, the 1-year survival rate was longer in the EOX group than in the ECF group (46.8% versus 37.7%, respectively; HR 0.80 [95% CI 0.66 to 0.97], P = 0.02). This translated into an overall survival of 11.2 months for the EOX group and 9.9 months for the ECF group. Therefore, EOX is preferred over ECF in clinical practice.

The French FNLCC/FFCD trial published in 2011 provided further support for perioperative chemotherapy.⁴⁴ A total of 224 patients with adenocarcinoma of the lower esophagus, GEJ, or stomach were randomly assigned to perioperative chemotherapy plus surgery or surgery alone. The perioperative-chemotherapy group received 2 to 3 cycles of preoperative chemotherapy and 3 to 4 cycles of postoperative chemotherapy, consisting of infusional 5-FU (800 mg/m² daily for days 1 to 5) and cisplatin (100 mg/m² on day 1). In patients receiving preoperative chemotherapy, 38% experienced at least grade 3 to 4 toxicity. Among the 109 patients who received at least 1 cycle of preoperative chemotherapy, only 54 patients (50%) received postoperative chemotherapy. Despite this, the perioperative-chemotherapy group had a statistically significant higher R0 resection rate (84% versus 74%, P = 0.04) compared to the surgery alone group. At a median follow-up of 5.7 years, the perioperative chemotherapy group had an improved overall survival (HR for death 0.69 [95% CI 0.50 to 0.95], P = 0.02) and disease-free survival (DFS; HR for recurrence or death 0.65 [95% CI 0.48 to 0.89], P = 0.003). This translated into 5-year overall survival rates of 38% versus 24% and 5-year DFS rates of 34% versus 19%. One caveat to this study is that the majority of patients (64%) had GEJ cancer and only 25% had gastric cancer. In the multivariate analysis, the 2 significant prognostic factors for overall survival were the administration of preoperative chemotherapy (P = 0.01) and tumor site at the GEJ (P < 0.01).

Adjuvant Chemoradiotherapy

The INT-0116 (Intergroup 0116) study published in 2001 established adjuvant chemoradiotherapy as the standard approach for resectable gastric cancer in the United States. In this study, a total of 556 patients with resected gastric or GEJ cancer were randomly assigned to surgery alone or surgery followed by adjuvant 5-FU/leucovorin bolus chemotherapy, sandwiched with 5-FU–based chemoradiation (45 Gy).⁴⁵ In the chemoradiotherapy group, 64% of patients completed treatment and grade 3 and 4 toxicity occurred in 41% and 32%, respectively. However, only 3 patients (1%) died from treatment-related toxicity. At a median follow-up of 5 years, the median overall survival was 36 months in the chemoradiation group and 27 months in the surgery group. Overall survival rate was 50% in the combined modality group and 41% in the surgery-alone group, with a HR of 1.35 (95% CI 1.09 to 1.66, P = 0.005). The 3-year DFS was 48% in the chemoradiotherapy group and 31% in the surgery-alone group, corresponding to a DFS of 30 months and 19 months, respectively. Even after a median follow-up of 10 years, these positive results were maintained, with a HR for survival of 1.32 (95% CI 1.10 to 1.60, P = 0.0046) and HR for DFS of 1.51 (95% CI 1.25 to 1.83, P < 0.001).⁴⁶ A criticism of the INT-0116 study is that 54% of patients had less than a D1 lymph node dissection, suggesting that adjuvant chemoradiation may have compensated for suboptimal surgery.

CALGB 80101, a United States Intergroup study, compared the INT-0116 protocol regimen (bolus 5-FU/leucovorin with 5-FU plus concurrent radiotherapy) to postoperative ECF sandwiched with 5-FU plus concurrent radiotherapy.⁴⁷ The study included patients with resected gastric or GEJ adenocarcinoma that extended beyond the muscularis propria or was node positive. The percentage of patients with gastric versus GEJ cancer was not reported. A total of 546 patients were randomized. Preliminary results were presented at the 2011 American Society of Clinical Oncology meeting. The ECF arm had lower rates of grade 3/4 toxicities, including neutropenia, diarrhea, and mucositis. However, there was no difference in overall survival (3-year overall survival of 52% versus 50% for ECF and 5-FU/leucovorin, respectively) or DFS (3-year DFS of 47% versus 46% for ECF and 5-FU/leucovorin, respectively). The trial was not adequately powered to assess noninferiority. The location of the primary tumor (GEJ versus proximal versus distal stomach) did not have any effect on treatment outcome.

The Adjuvant Chemoradiation Therapy in Stomach Cancer (ARTIST) trial was the first study to compare adjuvant chemoradiotherapy with adjuvant chemotherapy in patients with D2-resected gastric cancer.⁴⁸ A total of 458 patients were randomly assigned to 6 cycles of XP chemotherapy (capecitabine 2000 mg/m² per day on days 1–14 and cisplatin 60 mg/m² on day 1, every 3 weeks) or XP/radiotherapy/XP (2 cycles of XP followed by 45 Gy radiotherapy with concurrent daily capecitabine [825 mg/m² twice daily] and 2 cycles of XP). After a median follow-up of 84 months, there was no difference in DFS or overall survival between treatment arms (HR for progression 0.74 [95% CI 0.52 to 1.05], P = 0.09; HR for death 1.13 [95% CI 0.78 to 1.65], P = 0.53).⁴⁹ However, subgroup analysis showed that chemoradiotherapy significantly improved DFS in patients with node-positive disease (3-year DFS 76% versus 72%, P = 0.004).

Adjuvant Chemotherapy

Data supporting the use of adjuvant chemotherapy alone is largely derived from trials done in Asia, typically after a D2 lymph node dissection, and thus adjuvant chemotherapy has become the standard of care in that region. In the Japanese Adjuvant Chemotherapy Trial of S-1 for Gastric Cancer (ACTS-GC), a total of 1059 patients with stage II or III gastric cancer who had undergone surgery with a D2 lymphadenectomy were randomly assigned to 1 year of S-1 (an oral fluoropyrimidine) or surgery alone.⁵⁰ The 5-year overall survival rate was 72% in the S-1 group and 61% in the surgery-only group (HR 0.669 [95% CI 0.54 to 0.83]).⁵¹ The 5-year relapse-free survival was 65% in the S-1 group and 53% in the surgery-only group (HR 0.65 [95% CI 0.537 to 0.793]). Of note, both arms of the ACTS-GC trial had significantly higher 5-year overall survival rates compared to the INT-0116 and MAGIC trials: 43% versus 28% and 36% versus 23% for the treatment and control groups, respectively.^42,45 Consequently, it is unclear if the benefit of adjuvant chemotherapy can be translated to Western countries.

The Korean Capecitabine and Oxaliplatin Adjuvant Study in Stomach Cancer (CLASSIC) trial published in 2012 also established the role of adjuvant chemotherapy after D2 gastrectomy.⁵² A total of 1035 patients with stage II-IIIB gastric cancer who had curative D2 gastrectomy were randomly assigned to 8 cycles of adjuvant XELOX (capecitabine 1000 mg/m² twice daily on days 1–14 plus oxaliplatin 130 mg/m² on day 1, 21-day cycle) or surgery alone. Median follow-up was 34 months in both arms and 67% of patients in the chemotherapy arm completed all 8 cycles of planned chemotherapy. The 3-year DFS was 74% in the chemotherapy group and 59% in the surgery-only group (HR 0.56 [95% CI 0.44 to 0.72], P < 0.0001). There was a trend toward improvement in overall survival (83% versus 78%, HR 0.72 [95% CI 0.52 to 1.00]). After 5 years of follow-up, the improvement in overall survival became statistically significant (78% versus 69%, HR 0.66 [95% CI 0.51 to 0.85]).⁵³

The benefit of adjuvant chemotherapy was reinforced by a 2010 meta-analysis comparing adjuvant chemotherapy to surgery alone in patients with resected gastric cancer.⁵⁴ A total of 17 randomized controlled trials were included. Adjuvant fluorouracil-based chemotherapy was associated with a statistically significant improved overall survival (HR 0.82 [95% CI 0.76 to 0.90], P < 0.001) and DFS (HR 0.82 [95% CI 0.75 to 0.90], P < 0.001). Five-year overall survival increased from 49.6% to 55.3% with chemotherapy.

Since data exists for all 3 approaches (perioperative chemotherapy, adjuvant chemoradiotherapy, and adjuvant chemotherapy), various meta-analyses have been done to clarify which approach is the best. In a recent meta-analysis of 6 randomized controlled trials reported between 2010 and 2012, which involved 1171 patients with resected gastric cancer, adjuvant chemotherapy was compared to adjuvant chemoradiotherapy.⁵⁵ Five of the studies were from East Asia, while one was from a Western country. Adjuvant chemoradiation was associated with a lower local-regional recurrence rate (OR 0.46 [95% CI 0.32 to 0.67]) and better 5-year DFS rate (OR 1.56 [95% CI 1.09 to 2.24]). However, there was no statistical difference in 5-year overall survival rate (OR 1.32 [95% CI 0.92 to 1.88]). Similar results were reported by Zhou et al in 2016.⁵⁶ This meta-analysis included 4 randomized controlled trials reported between 2010 and 2015, with a total of 960 patients who had undergone a D2 resection for gastric cancer. Compared to adjuvant chemotherapy, adjuvant chemoradiotherapy significantly reduced the loco-regional recurrence rate (LRRR; relative risk [RR] 0.50 [95% CI 0.34 to 0.74], P = 0.0005) and improved DFS (HR 0.73 [95% CI 0.60 to 0.89], P = 0.002). Again, no difference in overall survival was seen (HR 0.91 [95% CI 0.74 to 1.11], P = 0.34).

Adjuvant chemotherapy and perioperative chemotherapy have also been compared. In a recent meta-analysis of 14 randomized controlled trials (8 Asian, 6 European) involving 2093 patients with resected gastric or GEJ cancer, perioperative chemotherapy was associated with improved overall survival when compared to adjuvant chemotherapy (HR 0.48 [95% CI 0.35 to 0.67], P < 0.001).⁵⁷ The benefit of perioperative chemotherapy over adjuvant chemotherapy has also been reported in a 2016 meta-analysis by Zhao et al.⁵⁸ A total of 1240 patients were included from 5 randomized controlled trials and 6 clinical controlled trials, all from Asian countries. The 5-year overall survival rate was significantly better in the perioperative chemotherapy group compared to the adjuvant chemotherapy group (RR 0.77 [95% CI 0.64 to 0.92], P < 0.01). Furthermore, the 2 groups showed no significant differences in the postoperative complication rates (RR 0.98 [95% CI 0.63 to 1.51], P = 0.91) or adverse effects of chemotherapy (P > 0.05 for all adverse effects).

While these meta-analyses may offer some insight on the best treatment approach, they should be interpreted with caution. Most studies included in these meta-analyses were from Asian countries, and their findings may not be applicable to Western countries. Furthermore, the heterogeneity of trials and inclusion of nonrandomized trials make it difficult to draw conclusions. There are several ongoing trials that will help to define the optimal treatment approach.

CASE CONTINUED

The patient is presented at tumor board and the consensus is to proceed with the perioperative chemotherapy approach. The patient undergoes echocardiography, which reveals a normal ejection fraction. He receives 3 cycles of neoadjuvant EOX (epirubicin, oxaliplatin, and capecitabine). After 3 cycles of neoadjuvant EOX, the patient has a repeat CT that shows marked interval reduction in the size of the primary gastric neoplasm and interval decrease in the size of the small perigastric lymph nodes. He subsequently undergoes a total gastrectomy with J-tube placement. Pathology shows ypT3N0 disease with 0 out of 47 lymph nodes involved and negative margins. He then receives 3 cycles of adjuvant EOX.

• What are the recommendations for surveillance?

According to the current NCCN guidelines, a history and physical exam should be performed every 3 to 6 months for 1 to 2 years, then every 6 to 12 months for 3 to 5 years, and then annually.²³ Labs, CT chest/abdomen, and EGD should be done as clinically indicated. Patients who have undergone surgical resection should be monitored for nutritional deficiencies (vitamin B12 and iron).

GASTROESOPHAGEAL JUNCTION TUMORS

Tumors arising in the GEJ or gastric cardia within 5 cm of the GEJ that extend into the GEJ or distal esophagus are staged and treated as esophageal cancers.²¹ The primary treatment for T1/T2N0 tumors is surgical resection. In patients with T3 or higher or node-positive adenocarcinoma of the GEJ, a combined modality approach is preferred, with preoperative chemoradiotherapy followed by surgical resection.⁵⁹ The CROSS trial demonstrated a significant survival benefit with preoperative chemoradiation using carboplatin/paclitaxel compared to surgery alone in patients with esophageal or GEJ cancer (49 months versus 24 months, HR 0.66, P = 0.003).⁶⁰

ONGOING TRIALS

As mentioned previously, several randomized clinical trials are in progress to clarify the optimal treatment approach. The MAGIC-B/MRC-ST03 is a randomized phase 2/3 trial looking at perioperative epirubicin, cisplatin, and capecitabine (ECX) with or without bevacizumab in patients with resectable lower esophageal, GEJ, or gastric cancer.⁶¹ The TOPGEAR trial, a randomized phase 2/3 study being conducted in Canada and Europe, is comparing perioperative ECF chemotherapy with preoperative chemoradiation plus perioperative ECF chemotherapy.⁶² In Asia, the PRODIGY trial is a phase 3, open-label, randomized study comparing neoadjuvant docetaxel, oxaliplatin, and S-1 followed by surgery and adjuvant S-1 versus surgery plus adjuvant S-1 in patients with locally advanced gastric cancer (T2-T4 or node positive).⁶³ Primary endpoint is PFS and secondary endpoints are overall survival, R0 resection rate, and safety.

Trials comparing adjuvant chemotherapy to adjuvant chemoradiotherapy are also being conducted. In the Dutch CRITICS study, a randomized phase 3 trial, patients with stage Ib-Iva resectable gastric cancer were given 3 cycles of epirubicin, cisplatin/oxaliplatin, and capecitabine (ECC/EOC), followed by D2 resection and either 3 cycles of ECC/EOC or chemoradiation with weekly cisplatin and daily capecitabine.⁶⁴ Between January 2007 and April 2015, a total of 788 patients were enrolled. In a preliminary report presented at ASCO in 2016, the 5-year survival rate was similar between the 2 arms (41.3% for chemotherapy arm and 40.9% for chemoradiotherapy arm, P = 0.99). The Korean ARTIST II trial is comparing adjuvant S-1 and S-1/oxaliplatin with or without radiotherapy in patients with D2-resected gastric cancer.⁶⁵ Similarly, the NCT01711242 trial is comparing adjuvant XELOX alone versus XELOX with concurrent capecitabine/radiotherapy in patients with resected D2 gastric cancer.⁶⁶

The ToGA trial established a survival benefit of trastuzumab in combination with chemotherapy in HER2-positive metastatic gastric cancer.⁶⁷ Consequently, there are ongoing clinical trials to assess the role of trastuzumab in nonmetastatic gastric cancer. The TOXAG study is a phase 2 trial looking at the safety and tolerability of adjuvant oxaliplatin, capecitabine, and trastuzumab with radiation in patients with resected HER2-positive gastric or GEJ adenocarcinoma.⁶⁸ The NCT01130337 clinical trial is evaluating perioperative XELOX/trastuzumab in patients with resectable gastric or GEJ adeno-carcinoma.⁶⁹

Gastric cancer is the fifth most common cancer worldwide, with the greatest incidence in East Asia. Survival outcomes are better in Asian countries when compared to the United States. This difference in survival may be related to the presence of mass screening programs in Asia, which allows for detection at an earlier stage and the use of a more extensive surgical approach (ie, D2 resection). Risk factors for developing gastric cancer include: diets high in salt/salt-preserved foods or processed meats, obesity, smoking, H. pylori infection, EBV, prior gastric surgery, radiation exposure, and positive family history.

According to the latest edition of TMN staging, gastric cancer includes tumors arising more than 5 cm distally of the GEJ or within 5 cm of the GEJ but without extension to the esophagus or GEJ. Diagnostic work-up includes: EGD with biopsy; basic labs; CT chest/abdomen/pelvis with oral and intravenous contrast; EUS if no M1 disease is identified; PET-CT if there is no M1 disease and if clinically indicated; and diagnostic laparoscopy with cytology for clinical stage T1b or higher.

The mainstay of treatment is surgical resection. Laparoscopic approach is preferred over open gastrectomy due to lower complication rates and similar survival outcomes. Current NCCN guidelines recommend a D1 or a modified D2 lymph node dissection with at least 15 lymph nodes removed for examination. Systemic chemotherapy is required in locally advanced gastric cancer (T3-T4 or node positive) and should be considered in T2N0 disease with high-risk features. Currently, there is no global consensus on the optimal treatment approach. Data from various trials have shown benefit for each approach. Regional preferences are: perioperative chemotherapy in Europe; adjuvant chemoradiotherapy in the United States; and adjuvant chemotherapy in Asia. In an effort to better define the optimal treatment approach, several randomized clinical trials are being conducted. According to the current NCCN guidelines, the following treatment approaches are acceptable and are supported by data in the trial listed in parentheses:

• Perioperative chemotherapy

° 5-FU/cisplatin (French FNLCC/FCCD trial)⁴⁴ or

° ECF (MAGIC trial)⁴² or

° ECF modifications: EOX, EOF, ECX (REAL-2 trial)⁴³

• Adjuvant chemoradiotherapy

° 5-FU/leucovorin sandwiched with 5-FU-based chemoradiation (INT-0116 trial)⁴⁵

• Adjuvant chemotherapy (after D2 resection)

° Capecitabine/oxaliplatin (CLASSIC trial)⁵² or

° Capecitabine/cisplatin (ARTIST trial)^48,49

INTRODUCTION

Gastric cancer is the fifth most common cancer worldwide and the third leading cause of cancer death in both females and males.¹ More than 70% of gastric cancer cases occur in the developing world, with approximately 50% occurring in East Asia.² Gastric cancer is less common in the United States, with an incidence of 12.3 cases in males and 6.0 cases in females per 100,000 per year and a disproportionately higher incidence in Asians.³ According to the Surveillance, Epidemiology, and End Results Program, approximately 26,370 new cases of stomach cancer were diagnosed in the United States in 2016, and an estimated 10,730 people died of this disease.⁴ Since the 1970s, the 5-year relative survival rate for gastric cancer in the United States has improved from 15% in 1975 to 29% in 2009.⁵ In contrast, in Japan and Korea, where screening programs have been implemented, the 5-year survival rate approaches 70%.⁶

RISK FACTORS AND CLASSIFICATION

A variety of risk factors have been linked to gastric cancer. Diets high in salt, salt-preserved foods, and/or processed meats have been associated with an increased risk for developing gastric cancer.^7,8 Obesity and smoking have also been implicated in gastric cancer.^9,10 Several studies have demonstrated a strong association between Helicobacter pylori and the development of gastric cancer.^11–13 It is believed that H. pylori infection leads to chronic active gastritis, atrophic gastritis, and intestinal metaplasia. Interestingly, mass eradication of H. pylori has not been shown to reduce the risk for gastric cancer.¹⁴ Therefore, treatment of H. pylori should only be considered in patients with active peptic ulcer disease.¹⁵ Other risk factors include Epstein-Barr virus (EBV), prior gastric surgery, and radiation exposure.^16–18 Family history of gastric cancer, hereditary nonpolyposis colon cancer, Li-Fraumeni syndrome, and hereditary diffuse gastric cancer caused by mutations in the E-cadherin gene increase the risk.¹⁷

The anatomic distinction between gastric cancer and cancer of the gastroesophageal junction (GEJ) has been a topic of debate. The Siewert classification is the most widely used system and divides GEJ adenocarcinoma into 3 categories:²⁰ type I tumor: adenocarcinoma of distal esophagus, located 1 cm to 5 cm above the GEJ; type II tumor: true carcinoma of gastric cardia, located within 1 cm above and 2 cm below the GEJ; type III tumor: subcardial gastric carcinoma, located 2 cm to 5 cm below the GEJ, and infiltrates esophagus from below.

The American Joint Committee on Cancer (AJCC) has updated the latest (7th) edition of TMN staging for stomach cancer to include tumors arising more than 5 cm distally of the GEJ or within 5 cm of the GEJ but without extension to the esophagus or GEJ.²¹

In the following sections, neoadjuvant and adjuvant therapy in gastric cancer are discussed using a case presentation to illustrate important concepts.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 43-year old male with no significant past medical history presents with epigastric abdominal pain and heart burn for the past few weeks. He denies nausea, vomiting, melena, or hematochezia. His primary care physician (PCP) diagnoses him with gastroesophageal reflux disease (GERD) and initiates a trial of pantoprazole. Over the next 2 to 3 months, his symptoms do not improve and he has an associated 40-lb weight loss. Both social history and family history are noncontributory. Physical exam reveals epigastric tenderness without rebound or guarding. Laboratory evaluation reveals a hemoglobin of 12.6 g/dL with a mean corpuscular volume of 72 fL. A comprehensive chemistry profile is within normal limits. Given the constellation of presenting symptoms, especially the unintentional weight loss and the presence of microcytic anemia, his PCP suspects a malignant process and refers the patient to a gastroenterologist.

• What are the next appropriate steps for diagnosis?

The most common presenting symptoms of gastric cancer are weight loss and abdominal pain.²² Less commonly, patients exhibit nausea, anorexia, and dysphagia with proximal tumors. Melena is seen in only about 20% of patients. In Japan, where gastric cancer is more prevalent, mass screening programs allow for detection at an earlier stage, which partially accounts for the better survival rates seen in Asia as compared to the United States. Diagnostic work-up includes esophagogastroduodenoscopy (EGD) to assess Siewert category and to obtain a tissue sample for diagnosis. Full staging requires a complete blood count (CBC) with differential; comprehensive chemistry profile; computed tomography (CT) of chest/abdomen/pelvis with oral and intravenous contrast; endoscopic ultrasound (EUS) if no M1 disease is identified; positron emission tomography (PET)-CT if there is no evidence of M1 disease and if clinically indicated; and laparoscopy with cytology for clinical stage T1b or higher.²³ Patients should be staged according to the TMN staging system (Table 1).

CASE CONTINUED

The patient undergoes EGD, which reveals a large ulcerated, partially circumferential mass measuring approximately 4 cm. The mass extends from the gastric body to the cardia. Biopsy of the mass reveals poorly differentiated adenocarcinoma as well as H. pylori–associated gastritis. He is given antibiotic therapy and undergoes complete work-up of his newly diagnosed gastric adenocarcinoma. CT of the chest/abdomen/pelvis demonstrates a large gastric mass with gastrohepatic and distal perigastric adenopathy, compatible with locally advance primary gastric cancer. There is no evidence of distant metastasis. PET scan shows a large hypermetabolic mass in the stomach body and increased FDG activity in 3 small nodes along the lesser gastric curvature and in 1 node in the gastrohepatic region. EUS reveals a malignant gastric tumor in the body of the stomach, which is staged as T3, and a few malignant-appearing lymph nodes in the perigastric region. Fine-needle aspiration of the perigastric lymph node is performed and the sample obtained is positive for malignant cells. Diagnostic laparoscopy with peritoneal washings is performed and cytology is negative for malignant cells. The patient is staged as clinical stage IIB (T3N1M0).

• How should this patient with newly diagnosed, locally advanced, resectable gastric cancer be managed?

SURGERY

Surgical resection for localized gastric cancer is the mainstay of treatment with curative intent. Only very early stage (Tis or T1a) tumors can be considered for endoscopic mucosal resection. Regarding surgical resection, distal gastric cancers are typically treated with subtotal gastrectomy because there is no survival difference between subtotal and total gastrectomy.^24,25 Moreover, subtotal gastrectomy is associated with better nutritional status and quality of life. For proximal tumors, total gastrectomy is preferred as subtotal gastrectomy has been associated with a higher incidence of reflux esophagitis and anastomotic stenosis.²⁶ In terms of surgical approach, multiple studies have shown that a laparoscopic approach has a lower complication rate and similar outcomes in terms of cancer recurrence and long-term survival when compared to open gastrectomy.^27–29 Thus, a laparoscopic approach is often used in academic centers with highly experienced surgeons.

The extent of lymph node dissection remains a topic of debate. A D1 dissection involves the removal of perigastric lymph nodes. A D2 dissection is a D1 dissection plus the removal of lymph nodes along the left gastric artery, common hepatic artery, celiac artery, splenic hilum, and splenic artery. D2 lymphadenectomy has become the standard of care in Eastern countries where gastric cancer is more prevalent, such as Japan and Korea.³⁰ In Western countries, including the United States, less extensive lymphadenectomies are performed. Both randomized clinical trials and meta-analyses have failed to demonstrate an overall survival advantage of D2 dissection over D1 dissection.^31,32 A Dutch trial by Bonenkamp et al involving 711 patients, one of the largest randomized trials of D1 and D2 lymphadenectomy, showed that D2 patients had a higher operative mortality rate than D1 patients (10% versus 4%, P = 0.004) and experienced more complications (43% versus 25%, P < 0.001).³³ In a 15-year follow-up of this study, patients who had a D2 resection had lower locoregional recurrence and gastric-cancer–related death rates compared to those who had a D1 resection; however, D2 resection was associated with a significantly higher operative mortality and complication rate compared to D1.³⁴ In addition, a 2015 Cochrane meta-analysis has demonstrated improved disease-specific survival (DSS) with D2 dissection (hazard ratio [HR] 0.81 [95% confidence interval {CI} 0.71 to 0.92]).³⁵ Currently, the National Comprehensive Cancer Network (NCCN) recommends a D1 or a modified D2 gastrectomy with at least 15 lymph nodes removed for examination, with D2 lymphadenectomies only to be performed at experienced centers.²³

SYSTEMIC CHEMOTHERAPY

Locally advanced gastric cancer (T3-T4 or node positive) requires systemic chemotherapy in addition to surgery, as this intervention improves the 5-year overall survival by 10% to 15%.³⁶ Systemic therapy should also be considered in patients with T2N0 disease with high-risk features: poorly differentiated or high-grade cancer; lymphovascular invasion; neural invasion; age younger than 50 years; and patients who did not undergo D2 dissection.²³ Currently, there is no global consensus on the best treatment approach. In the United States, where a less aggressive lymph-node dissection is performed, adjuvant chemoradiotherapy after surgery is more commonly seen. In Europe, perioperative (preoperative and postoperative) chemotherapy is the standard treatment. In Japan, adjuvant chemotherapy after D2 lymphadenectomy is the standard of care.³⁷ These regional preferences are largely due to randomized clinical trials that have shown benefit for each approach. The landmark trials are discussed in the following sections and are summarized in Table 2.

Neoadjuvant chemotherapy has the benefit of “downstaging” locally advanced tumors to allow for curative resection. Phase 2 clinical trials have also demonstrated good pathologic response rates and high R0 resection rates following neoadjuvant chemotherapy.^38,39 However, phase 3 trials to support this treatment approach are lacking. In the European Organisation for Research and Treatment of Cancer (EORTC) 40954 trial, patients with stage III or IV gastric or GEJ cancer were randomly assigned to surgery with or without preoperative cisplatin, leucovorin, and infusional fluorouracil (5-FU).⁴⁰ The trial was stopped early due to poor accrual after 144 patients were randomized. The neoadjuvant chemotherapy arm had a higher R0 resection rate compared to the surgery alone arm (82% versus 67%, respectively, P = 0.036) but a higher postoperative complication rate (27% versus 16%, respectively, P = 0.09). More important, after a median follow-up of 4.4 years, a survival benefit could not be shown, with 2-year survival rates of 72.7% and 69.9% in the neoadjuvant and surgery-only arms, respectively (HR 0.84 [95% CI 0.52 to 1.35], P = 0.466). Due to the lack of large trials, a meta-analysis assessing the effectiveness of neoadjuvant chemotherapy combined with surgery versus surgery alone in advanced gastric and gastroesophageal cancer was performed.⁴¹ The analysis included 12 randomized controlled trials with a total of 1820 patients. Neoadjuvant chemotherapy was shown to slightly improve the survival rate (odds ratio [OR] 1.32 [95% CI 1.07 to 1.64], P = 0.01). It significantly improved the 3-year progression-free survival (PFS; OR 1.85 [95% CI 1.39 to 2.46], P < 0.0001), tumor down-staging rate (OR 1.71 [95% CI 1.26 to 2.33], P = 0.0006), and R0 resection rate (OR 1.38 [95% CI 1.08 to 1.78], P = 0.01). There were no differences between the 2 arms in terms of relapse rates, operative complications, perioperative mortality, and grade 3/4 adverse effects. While these results are encouraging, further randomized clinical trials are needed to clarify the role of neoadjuvant chemotherapy and its impact on overall survival.

Perioperative Chemotherapy

The results of the Medical Research Council Adjuvant Gastric Infusional Chemotherapy (MAGIC) trial published in 2006 established perioperative chemotherapy as standard of care in patients with resectable gastric and gastroesophageal adenocarcinoma.⁴² A total of 503 patients with potentially resectable gastric and lower esophageal adenocarcinoma were randomly assigned to perioperative chemotherapy plus surgery or surgery alone. Perioperative chemotherapy consisted of 3 preoperative and postoperative cycles of epirubicin, cisplatin, and infusional 5-FU (ECF). At a median follow-up of 4 years, the perioperative-chemotherapy group had a significantly better PFS (HR 0.66 [95% CI 0.53 to 0.81], P < 0.001) as well as overall survival (HR 0.75 [95% CI 0.60 to 0.93], P = 0.009). The 5-year overall survival rate was 36.3% in the perioperative chemotherapy group and 23% in the surgery group. Of note, there was a greater proportion of stage T1/T2 tumors (52% versus 37%, P = 0.002) and N0/N1 disease (84% versus 71%) in the perioperative-chemotherapy group compared to the surgery alone group. In addition, only 42% of patients in the perioperative chemotherapy group completed all 6 cycles of chemotherapy.

The administration of ECF is often difficult since the 5-FU component requires a central venous access device and an ambulatory infusion pump and the cisplatin component is associated with nephrotoxicity and ototoxicity. The REAL-2 trial was a randomized phase 3 clinical trial that assessed whether 5-FU could be replaced by capecitabine and cisplatin by oxaliplatin in the ECF regimen.⁴³ Between June 2000 and May 2005, a total of 1002 patients with locally advanced esophageal/GEJ/gastric cancer were enrolled. Patients were randomly assigned to 1 of 4 triplet therapies: epirubicin and cisplatin plus either 5-FU (ECF) or capecitabine (ECX) or epirubicin and oxaliplatin plus either 5-FU (EOF) or capecitabine (EOX). After a median follow-up of approximately 18 months, the overall survival in the capecitabine groups did not differ significantly from that in the 5-FU groups (HR 0.88 [95% CI 0.77 to 1.00], P = 0.06), nor did overall survival in the oxaliplatin groups differ significantly from that in the cisplatin groups (HR 0.91 [95% CI 0.79 to 1.04], P = 0.16). Interestingly, the 1-year survival rate was longer in the EOX group than in the ECF group (46.8% versus 37.7%, respectively; HR 0.80 [95% CI 0.66 to 0.97], P = 0.02). This translated into an overall survival of 11.2 months for the EOX group and 9.9 months for the ECF group. Therefore, EOX is preferred over ECF in clinical practice.

The French FNLCC/FFCD trial published in 2011 provided further support for perioperative chemotherapy.⁴⁴ A total of 224 patients with adenocarcinoma of the lower esophagus, GEJ, or stomach were randomly assigned to perioperative chemotherapy plus surgery or surgery alone. The perioperative-chemotherapy group received 2 to 3 cycles of preoperative chemotherapy and 3 to 4 cycles of postoperative chemotherapy, consisting of infusional 5-FU (800 mg/m² daily for days 1 to 5) and cisplatin (100 mg/m² on day 1). In patients receiving preoperative chemotherapy, 38% experienced at least grade 3 to 4 toxicity. Among the 109 patients who received at least 1 cycle of preoperative chemotherapy, only 54 patients (50%) received postoperative chemotherapy. Despite this, the perioperative-chemotherapy group had a statistically significant higher R0 resection rate (84% versus 74%, P = 0.04) compared to the surgery alone group. At a median follow-up of 5.7 years, the perioperative chemotherapy group had an improved overall survival (HR for death 0.69 [95% CI 0.50 to 0.95], P = 0.02) and disease-free survival (DFS; HR for recurrence or death 0.65 [95% CI 0.48 to 0.89], P = 0.003). This translated into 5-year overall survival rates of 38% versus 24% and 5-year DFS rates of 34% versus 19%. One caveat to this study is that the majority of patients (64%) had GEJ cancer and only 25% had gastric cancer. In the multivariate analysis, the 2 significant prognostic factors for overall survival were the administration of preoperative chemotherapy (P = 0.01) and tumor site at the GEJ (P < 0.01).

Adjuvant Chemoradiotherapy

The INT-0116 (Intergroup 0116) study published in 2001 established adjuvant chemoradiotherapy as the standard approach for resectable gastric cancer in the United States. In this study, a total of 556 patients with resected gastric or GEJ cancer were randomly assigned to surgery alone or surgery followed by adjuvant 5-FU/leucovorin bolus chemotherapy, sandwiched with 5-FU–based chemoradiation (45 Gy).⁴⁵ In the chemoradiotherapy group, 64% of patients completed treatment and grade 3 and 4 toxicity occurred in 41% and 32%, respectively. However, only 3 patients (1%) died from treatment-related toxicity. At a median follow-up of 5 years, the median overall survival was 36 months in the chemoradiation group and 27 months in the surgery group. Overall survival rate was 50% in the combined modality group and 41% in the surgery-alone group, with a HR of 1.35 (95% CI 1.09 to 1.66, P = 0.005). The 3-year DFS was 48% in the chemoradiotherapy group and 31% in the surgery-alone group, corresponding to a DFS of 30 months and 19 months, respectively. Even after a median follow-up of 10 years, these positive results were maintained, with a HR for survival of 1.32 (95% CI 1.10 to 1.60, P = 0.0046) and HR for DFS of 1.51 (95% CI 1.25 to 1.83, P < 0.001).⁴⁶ A criticism of the INT-0116 study is that 54% of patients had less than a D1 lymph node dissection, suggesting that adjuvant chemoradiation may have compensated for suboptimal surgery.

CALGB 80101, a United States Intergroup study, compared the INT-0116 protocol regimen (bolus 5-FU/leucovorin with 5-FU plus concurrent radiotherapy) to postoperative ECF sandwiched with 5-FU plus concurrent radiotherapy.⁴⁷ The study included patients with resected gastric or GEJ adenocarcinoma that extended beyond the muscularis propria or was node positive. The percentage of patients with gastric versus GEJ cancer was not reported. A total of 546 patients were randomized. Preliminary results were presented at the 2011 American Society of Clinical Oncology meeting. The ECF arm had lower rates of grade 3/4 toxicities, including neutropenia, diarrhea, and mucositis. However, there was no difference in overall survival (3-year overall survival of 52% versus 50% for ECF and 5-FU/leucovorin, respectively) or DFS (3-year DFS of 47% versus 46% for ECF and 5-FU/leucovorin, respectively). The trial was not adequately powered to assess noninferiority. The location of the primary tumor (GEJ versus proximal versus distal stomach) did not have any effect on treatment outcome.

The Adjuvant Chemoradiation Therapy in Stomach Cancer (ARTIST) trial was the first study to compare adjuvant chemoradiotherapy with adjuvant chemotherapy in patients with D2-resected gastric cancer.⁴⁸ A total of 458 patients were randomly assigned to 6 cycles of XP chemotherapy (capecitabine 2000 mg/m² per day on days 1–14 and cisplatin 60 mg/m² on day 1, every 3 weeks) or XP/radiotherapy/XP (2 cycles of XP followed by 45 Gy radiotherapy with concurrent daily capecitabine [825 mg/m² twice daily] and 2 cycles of XP). After a median follow-up of 84 months, there was no difference in DFS or overall survival between treatment arms (HR for progression 0.74 [95% CI 0.52 to 1.05], P = 0.09; HR for death 1.13 [95% CI 0.78 to 1.65], P = 0.53).⁴⁹ However, subgroup analysis showed that chemoradiotherapy significantly improved DFS in patients with node-positive disease (3-year DFS 76% versus 72%, P = 0.004).

Adjuvant Chemotherapy

Data supporting the use of adjuvant chemotherapy alone is largely derived from trials done in Asia, typically after a D2 lymph node dissection, and thus adjuvant chemotherapy has become the standard of care in that region. In the Japanese Adjuvant Chemotherapy Trial of S-1 for Gastric Cancer (ACTS-GC), a total of 1059 patients with stage II or III gastric cancer who had undergone surgery with a D2 lymphadenectomy were randomly assigned to 1 year of S-1 (an oral fluoropyrimidine) or surgery alone.⁵⁰ The 5-year overall survival rate was 72% in the S-1 group and 61% in the surgery-only group (HR 0.669 [95% CI 0.54 to 0.83]).⁵¹ The 5-year relapse-free survival was 65% in the S-1 group and 53% in the surgery-only group (HR 0.65 [95% CI 0.537 to 0.793]). Of note, both arms of the ACTS-GC trial had significantly higher 5-year overall survival rates compared to the INT-0116 and MAGIC trials: 43% versus 28% and 36% versus 23% for the treatment and control groups, respectively.^42,45 Consequently, it is unclear if the benefit of adjuvant chemotherapy can be translated to Western countries.

The Korean Capecitabine and Oxaliplatin Adjuvant Study in Stomach Cancer (CLASSIC) trial published in 2012 also established the role of adjuvant chemotherapy after D2 gastrectomy.⁵² A total of 1035 patients with stage II-IIIB gastric cancer who had curative D2 gastrectomy were randomly assigned to 8 cycles of adjuvant XELOX (capecitabine 1000 mg/m² twice daily on days 1–14 plus oxaliplatin 130 mg/m² on day 1, 21-day cycle) or surgery alone. Median follow-up was 34 months in both arms and 67% of patients in the chemotherapy arm completed all 8 cycles of planned chemotherapy. The 3-year DFS was 74% in the chemotherapy group and 59% in the surgery-only group (HR 0.56 [95% CI 0.44 to 0.72], P < 0.0001). There was a trend toward improvement in overall survival (83% versus 78%, HR 0.72 [95% CI 0.52 to 1.00]). After 5 years of follow-up, the improvement in overall survival became statistically significant (78% versus 69%, HR 0.66 [95% CI 0.51 to 0.85]).⁵³

The benefit of adjuvant chemotherapy was reinforced by a 2010 meta-analysis comparing adjuvant chemotherapy to surgery alone in patients with resected gastric cancer.⁵⁴ A total of 17 randomized controlled trials were included. Adjuvant fluorouracil-based chemotherapy was associated with a statistically significant improved overall survival (HR 0.82 [95% CI 0.76 to 0.90], P < 0.001) and DFS (HR 0.82 [95% CI 0.75 to 0.90], P < 0.001). Five-year overall survival increased from 49.6% to 55.3% with chemotherapy.

Since data exists for all 3 approaches (perioperative chemotherapy, adjuvant chemoradiotherapy, and adjuvant chemotherapy), various meta-analyses have been done to clarify which approach is the best. In a recent meta-analysis of 6 randomized controlled trials reported between 2010 and 2012, which involved 1171 patients with resected gastric cancer, adjuvant chemotherapy was compared to adjuvant chemoradiotherapy.⁵⁵ Five of the studies were from East Asia, while one was from a Western country. Adjuvant chemoradiation was associated with a lower local-regional recurrence rate (OR 0.46 [95% CI 0.32 to 0.67]) and better 5-year DFS rate (OR 1.56 [95% CI 1.09 to 2.24]). However, there was no statistical difference in 5-year overall survival rate (OR 1.32 [95% CI 0.92 to 1.88]). Similar results were reported by Zhou et al in 2016.⁵⁶ This meta-analysis included 4 randomized controlled trials reported between 2010 and 2015, with a total of 960 patients who had undergone a D2 resection for gastric cancer. Compared to adjuvant chemotherapy, adjuvant chemoradiotherapy significantly reduced the loco-regional recurrence rate (LRRR; relative risk [RR] 0.50 [95% CI 0.34 to 0.74], P = 0.0005) and improved DFS (HR 0.73 [95% CI 0.60 to 0.89], P = 0.002). Again, no difference in overall survival was seen (HR 0.91 [95% CI 0.74 to 1.11], P = 0.34).

Adjuvant chemotherapy and perioperative chemotherapy have also been compared. In a recent meta-analysis of 14 randomized controlled trials (8 Asian, 6 European) involving 2093 patients with resected gastric or GEJ cancer, perioperative chemotherapy was associated with improved overall survival when compared to adjuvant chemotherapy (HR 0.48 [95% CI 0.35 to 0.67], P < 0.001).⁵⁷ The benefit of perioperative chemotherapy over adjuvant chemotherapy has also been reported in a 2016 meta-analysis by Zhao et al.⁵⁸ A total of 1240 patients were included from 5 randomized controlled trials and 6 clinical controlled trials, all from Asian countries. The 5-year overall survival rate was significantly better in the perioperative chemotherapy group compared to the adjuvant chemotherapy group (RR 0.77 [95% CI 0.64 to 0.92], P < 0.01). Furthermore, the 2 groups showed no significant differences in the postoperative complication rates (RR 0.98 [95% CI 0.63 to 1.51], P = 0.91) or adverse effects of chemotherapy (P > 0.05 for all adverse effects).

While these meta-analyses may offer some insight on the best treatment approach, they should be interpreted with caution. Most studies included in these meta-analyses were from Asian countries, and their findings may not be applicable to Western countries. Furthermore, the heterogeneity of trials and inclusion of nonrandomized trials make it difficult to draw conclusions. There are several ongoing trials that will help to define the optimal treatment approach.

CASE CONTINUED

The patient is presented at tumor board and the consensus is to proceed with the perioperative chemotherapy approach. The patient undergoes echocardiography, which reveals a normal ejection fraction. He receives 3 cycles of neoadjuvant EOX (epirubicin, oxaliplatin, and capecitabine). After 3 cycles of neoadjuvant EOX, the patient has a repeat CT that shows marked interval reduction in the size of the primary gastric neoplasm and interval decrease in the size of the small perigastric lymph nodes. He subsequently undergoes a total gastrectomy with J-tube placement. Pathology shows ypT3N0 disease with 0 out of 47 lymph nodes involved and negative margins. He then receives 3 cycles of adjuvant EOX.

• What are the recommendations for surveillance?

According to the current NCCN guidelines, a history and physical exam should be performed every 3 to 6 months for 1 to 2 years, then every 6 to 12 months for 3 to 5 years, and then annually.²³ Labs, CT chest/abdomen, and EGD should be done as clinically indicated. Patients who have undergone surgical resection should be monitored for nutritional deficiencies (vitamin B12 and iron).

GASTROESOPHAGEAL JUNCTION TUMORS

Tumors arising in the GEJ or gastric cardia within 5 cm of the GEJ that extend into the GEJ or distal esophagus are staged and treated as esophageal cancers.²¹ The primary treatment for T1/T2N0 tumors is surgical resection. In patients with T3 or higher or node-positive adenocarcinoma of the GEJ, a combined modality approach is preferred, with preoperative chemoradiotherapy followed by surgical resection.⁵⁹ The CROSS trial demonstrated a significant survival benefit with preoperative chemoradiation using carboplatin/paclitaxel compared to surgery alone in patients with esophageal or GEJ cancer (49 months versus 24 months, HR 0.66, P = 0.003).⁶⁰

ONGOING TRIALS

As mentioned previously, several randomized clinical trials are in progress to clarify the optimal treatment approach. The MAGIC-B/MRC-ST03 is a randomized phase 2/3 trial looking at perioperative epirubicin, cisplatin, and capecitabine (ECX) with or without bevacizumab in patients with resectable lower esophageal, GEJ, or gastric cancer.⁶¹ The TOPGEAR trial, a randomized phase 2/3 study being conducted in Canada and Europe, is comparing perioperative ECF chemotherapy with preoperative chemoradiation plus perioperative ECF chemotherapy.⁶² In Asia, the PRODIGY trial is a phase 3, open-label, randomized study comparing neoadjuvant docetaxel, oxaliplatin, and S-1 followed by surgery and adjuvant S-1 versus surgery plus adjuvant S-1 in patients with locally advanced gastric cancer (T2-T4 or node positive).⁶³ Primary endpoint is PFS and secondary endpoints are overall survival, R0 resection rate, and safety.

Trials comparing adjuvant chemotherapy to adjuvant chemoradiotherapy are also being conducted. In the Dutch CRITICS study, a randomized phase 3 trial, patients with stage Ib-Iva resectable gastric cancer were given 3 cycles of epirubicin, cisplatin/oxaliplatin, and capecitabine (ECC/EOC), followed by D2 resection and either 3 cycles of ECC/EOC or chemoradiation with weekly cisplatin and daily capecitabine.⁶⁴ Between January 2007 and April 2015, a total of 788 patients were enrolled. In a preliminary report presented at ASCO in 2016, the 5-year survival rate was similar between the 2 arms (41.3% for chemotherapy arm and 40.9% for chemoradiotherapy arm, P = 0.99). The Korean ARTIST II trial is comparing adjuvant S-1 and S-1/oxaliplatin with or without radiotherapy in patients with D2-resected gastric cancer.⁶⁵ Similarly, the NCT01711242 trial is comparing adjuvant XELOX alone versus XELOX with concurrent capecitabine/radiotherapy in patients with resected D2 gastric cancer.⁶⁶

The ToGA trial established a survival benefit of trastuzumab in combination with chemotherapy in HER2-positive metastatic gastric cancer.⁶⁷ Consequently, there are ongoing clinical trials to assess the role of trastuzumab in nonmetastatic gastric cancer. The TOXAG study is a phase 2 trial looking at the safety and tolerability of adjuvant oxaliplatin, capecitabine, and trastuzumab with radiation in patients with resected HER2-positive gastric or GEJ adenocarcinoma.⁶⁸ The NCT01130337 clinical trial is evaluating perioperative XELOX/trastuzumab in patients with resectable gastric or GEJ adeno-carcinoma.⁶⁹

Gastric cancer is the fifth most common cancer worldwide, with the greatest incidence in East Asia. Survival outcomes are better in Asian countries when compared to the United States. This difference in survival may be related to the presence of mass screening programs in Asia, which allows for detection at an earlier stage and the use of a more extensive surgical approach (ie, D2 resection). Risk factors for developing gastric cancer include: diets high in salt/salt-preserved foods or processed meats, obesity, smoking, H. pylori infection, EBV, prior gastric surgery, radiation exposure, and positive family history.

According to the latest edition of TMN staging, gastric cancer includes tumors arising more than 5 cm distally of the GEJ or within 5 cm of the GEJ but without extension to the esophagus or GEJ. Diagnostic work-up includes: EGD with biopsy; basic labs; CT chest/abdomen/pelvis with oral and intravenous contrast; EUS if no M1 disease is identified; PET-CT if there is no M1 disease and if clinically indicated; and diagnostic laparoscopy with cytology for clinical stage T1b or higher.

The mainstay of treatment is surgical resection. Laparoscopic approach is preferred over open gastrectomy due to lower complication rates and similar survival outcomes. Current NCCN guidelines recommend a D1 or a modified D2 lymph node dissection with at least 15 lymph nodes removed for examination. Systemic chemotherapy is required in locally advanced gastric cancer (T3-T4 or node positive) and should be considered in T2N0 disease with high-risk features. Currently, there is no global consensus on the optimal treatment approach. Data from various trials have shown benefit for each approach. Regional preferences are: perioperative chemotherapy in Europe; adjuvant chemoradiotherapy in the United States; and adjuvant chemotherapy in Asia. In an effort to better define the optimal treatment approach, several randomized clinical trials are being conducted. According to the current NCCN guidelines, the following treatment approaches are acceptable and are supported by data in the trial listed in parentheses:

• Perioperative chemotherapy

° 5-FU/cisplatin (French FNLCC/FCCD trial)⁴⁴ or

° ECF (MAGIC trial)⁴² or

° ECF modifications: EOX, EOF, ECX (REAL-2 trial)⁴³

• Adjuvant chemoradiotherapy

° 5-FU/leucovorin sandwiched with 5-FU-based chemoradiation (INT-0116 trial)⁴⁵

• Adjuvant chemotherapy (after D2 resection)

° Capecitabine/oxaliplatin (CLASSIC trial)⁵² or

° Capecitabine/cisplatin (ARTIST trial)^48,49

INTRODUCTION

Hemophilia A and B are the most common severe inherited bleeding disorders. The incidence of hemophilia is 1 in 5000 live male births, with hemophilia A occurring 4 times more commonly than hemophilia B. The associated decrease in factor VIII in hemophilia A was initially identified in 1947, and the decrease in factor IX associated with hemophilia B was identified 5 years later.^1,2 Both conditions are inherited as X-linked recessive traits. Queen Victoria of Britain, who reigned from 1837 to 1901, was a carrier of hemophilia and had 2 carrier daughters, Alice and Beatrice, and a son with hemophilia, Leopold.³ In 1984 and 1985, the genes for factor VIII and factor IX were cloned, and in 1989 recombinant factor VIII was first used clinically.^4–7

PATHOPHYSIOLOGY

Both factors VIII and IX are crucial for normal thrombin generation in the coagulation pathway. After any injury, the initial hemostatic event is the formation of a platelet plug. Once the platelet plug is formed, subsequent generation of fibrin prevents continued oozing from the affected site. In hemophilia A and B, the propagation phase of coagulation is impaired, and as a result, the formation of clot is delayed and is not robust. Due to the delayed formation of an abnormal clot, patients with hemophilia do not bleed rapidly but rather ooze continuously. Rebleeding is a common occurrence in inadequately treated patients.⁸

GENETICS

The gene for factor VIII (F8) is located in the most distal band (Xq28) of the long arm of the X chromosome. Spanning more than 186 kb, it is one of the largest genes known.^9,10 The gene for factor IX (F9) is located at Xq27.1 and spans 33 kb.⁷ Defects in the F8 gene associated with hemophilia A may be divided into several categories: gross gene rearrangements; insertions or deletions of genetic sequence of a size varying from 1 base pair up to the entire gene; or single DNA base substitutions resulting in either amino acid replacement (missense), premature peptide chain termination (nonsense, or stop mutations), or mRNA splicing defects. All classes of defects can result in severe disease. However, the single most clinically important defect is a gene rearrangement (an inversion) involving F8 intron 22, which results in approximately 50% of all severe hemophilia A cases worldwide.^11,12 In hemophilia B, point mutations are by far the most common type of abnormality. Generally, they are caused by DNA polymerases adding the wrong nucleotide during replication.¹³

HEMOPHILIA IN FEMALES

X-Inactivation (also called Lyonization) is a process that occurs early in embryonic development in female mammals where 1 of the 2 copies of the X chromosome present is inactivated; it is the reason why some female carriers of hemophilia can become symptomatic. Approximately one third of carriers have clotting factor levels of less than 60% of normal and may experience abnormal bleeding.^14,15 In most cases, carriers experience symptoms similar to those seen in men with mild hemophilia, as well as some that are specific to women. Symptomatic carriers and women with hemophilia may bruise more easily; may experience prolonged bleeding after surgery; may experience serious bleeding after trauma; often have heavier and more prolonged bleeding during their periods (menorrhagia) and are more likely to require an iron supplement or to undergo hysterectomy; and are more likely to have postpartum bleeding following childbirth.^14,15

CLINICAL MANIFESTATIONS

Hemorrhage in patients with hemophilia may occur with minimal or unknown trauma. Patients with severe hemophilia (factor level of < 1 IU/dL or < 1% of normal) often experience spontaneous bleeding into joints or muscles. Those with moderate hemophilia (factor level of 1–5 IU/dL or 1%–5% of normal) seldom experience spontaneous hemorrhage and usually have prolonged bleeding with minor trauma or surgery. Patients with mild hemophilia (factor level > 5 IU/dL but less than 40 IU/dL or > 5% but < 40% of normal) experience severe hemorrhage only following moderate to severe trauma or surgery, and rarely experience spontaneous bleeding. Depending on the site, bleeding can be serious (joints; muscles, especially deep compartments [iliopsoas, calf, and forearm]; mucous membranes in the mouth, gums, nose, and genitourinary tract) or life-threatening (intracranial, neck/throat, gastrointestinal). The joints and muscles are the most common sites of bleeding (Table 1).

MUSCULOSKELETAL BLEEDING

The hallmark of hemophilia is deep bleeding into the joints and muscles. Without prophylactic factor treatment, patients with severe hemophilia A or B may have a bleeding episode as often as once or twice a week. Hemarthrosis episodes typically begin when the child reaches the toddler age. One of the first signs of hemarthrosis is a tingling sensation and feeling of warmth which is soon followed by pain and decreased range of motion of the joint as a result of distension of the joint capsule. Prompt, aggressive treatment with factor replacement therapy is the key to prevent further bleeding and minimize potential long-term complications. Severe chronic arthropathy may develop in older children and adults who have not received aggressive treatment (Figure).

Bleeding into the muscle can manifest as a vague feeling of pain on motion. Swelling may not be obvious and the mass may be difficult to palpate, although the circumference of the affected limb will be increased. Among the muscle bleeds, iliopsoas bleed deserves a special mention because of its potential to cause life-threatening hypovolemic shock as large volumes of blood can be lost into the retroperitoneal space. These patients present with vague abdominal pain or upper thigh discomfort. The hip is flexed and outwardly rotated. The diagnosis is confirmed by computed tomography (CT) or ultrasound.

LIFE-THREATENING HEMORRHAGE

Central Nervous System Bleeding

Most central nervous system (CNS) events, which involve bleeding inside the skull or spinal canal, are caused by trauma. CNS hemorrhage is the most common form of severe hemophilic trauma. However, since patients with hemophilia can experience bleeding even weeks after a minor head injury, a history of head trauma may be hard to determine, particularly in children. Spontaneous CNS bleeding in individuals with hemophilia is rare except when there has been a recent antecedent CNS hemorrhage (ie, a recurrent bleed at a previously injured site) or when there is an associated anatomic lesion that predisposes to acute hemorrhage (eg, aneurysm or arteriovenous malformation). Data from the Universal Data Collection Project of the U.S. Centers for Disease Control and Prevention indicates that predisposing risk factors for intracranial hemorrhage include HIV infection, presence of inhibitory antibodies, and age younger than 5 years or older than 51 years.¹⁶ Neonatal intracranial hemorrhage is most commonly due to birth trauma. Difficult vaginal deliveries (often requiring the application of forceps or vacuum extraction) are predisposing factors for intracranial hemorrhage in hemophilic newborns.

The site of intracranial CNS bleeding can be subdural, epidural, or intraparenchymal. Bleeding at any of these sites can cause rapidly deteriorating CNS brain function, associated brain swelling, and, in the most extreme circumstances, herniation of the brainstem and rapid death. If the bleeding is stopped with rapid clotting factor replacement, adverse clinical effects can be avoided. However, with intraparenchymal hemorrhage, even small hemorrhages can induce permanent structural and/or neurologic sequelae (in particular, if the anatomic site of the bleed is essential for routine brain function).¹⁷

Throat and Neck Hemorrhage

An acute neck injury or a retropharyngeal hemorrhage induced by dental or oral surgical instrumentation can lead to a dissecting facial plane hematoma. This in turn can sometimes lead to compression and acute airway compromise. Bleeding from these injuries that is compressing or compromising the airway may require a rapid clinical response.¹⁸ The time from the injury until the trachea is compressed may be long, sometimes many hours. However, once the compression is sufficient to cause difficulty breathing, there may be a short amount of time to stop the bleeding and prevent complete respiratory obstruction.

MUCOCUTANEOUS BLEEDING

One of the common manifestations of hemophilia is oral bleeding. Tooth extraction poses a specific problem, and bleeding following extraction can be the first symptom that leads to the diagnosis of hemophilia. Bleeding after circumcision may also suggest the diagnosis. In 1 study cohort looking at sites of initial bleeding episodes in babies with hemophilia diagnosed before the age of 2 years, bleeding from circumcision and other iatrogenic causes tended to be most common in the neonatal period. Circumcision bleeding events occurred more often in infants with no family history (43%) as compared to those born to known maternal carriers (9.2%) or to mothers with some other family history of hemophilia (14.3%).¹⁹

Gastrointestinal (GI) bleeding occurs occasionally in hemophilia, and a wide spectrum of esophageal and GI bleeding may occur. A review of 41 episodes of GI bleeding in hemophilia patients who presented to 1 institution over 10 years implicated duodenal ulcer (22%), unknown site (22%), and gastritis (14%) as the most common sources.²⁰ Mallory-Weiss syndrome has also been cited as a cause for upper GI bleeding in hemophilia patients.²¹

PRINCIPLES OF TREATMENT

Understanding the pathophysiology of hemophilia as well as the type and severity of hemophilia and the inhibitor status in an individual patient are paramount in the management of a patient with hemophilia. In the past, management mainly focused on the treatment of acute bleeding episodes (Table 2). With data showing the benefit of bleed prevention, the management of hemophilia now focuses on prophylaxis of bleeding episodes, which prevents chronic arthropathy and improves quality of life.

ACUTE BLEEDING EPISODES

Dosing of Factor VIII Products

Dosing for factor VIII concentrate is as follows: 1 IU of factor VIII concentrate per kg will increase the circulating factor VIII level by 2% (ie, patient weight in kg × 50 IU/kg = 100% correction). For example, a 30-kg patient requiring 100% correction of factor VIII needs an infusion of 1500 IU of factor VIII (30 kg × 50 IU/kg).

Dosing of Factor IX Products

Dosing for factor IX concentrate is as follows: 1 IU of factor IX concentrate per kg will increase the circulating factor IX level by 1% (ie, patient weight in kg × 100 IU/kg = 100% correction). For example, a 30-kg patient requiring 100% correction of factor IX needs an infusion of 3000 IU of factor IX (30 kg × 100 IU/kg). Higher doses (120 to 130 IU/kg) of the recombinant factor IX product BeneFIX (Pfizer) may be needed to reach the 100% circulating factor IX level.

ADJUVANT THERAPY

Desmopressin

Desmopressin is a synthetic vasopressin analogue that increases plasma factor VIII and von Willebrand factor (VWF) levels; it is used to prevent and treat bleeding episodes associated with dental and surgical procedures in patients with mild and moderate hemophilia A and von Willebrand disease.²² Desmopressin causes the release of VWF and factor VIII from storage in the Weibel–Palade bodies of the endothelial cells that line the blood vessels. Individual response to desmopressin varies, with factor VIII level increasing between 2 and 15 times baseline level in patients with mild or moderate hemophilia A.²³ It is therefore recommended that patients undergo a therapeutic trial of desmopressin with laboratory measurement of response to factor VIII before it is used for treatment of bleeding episodes or as prophylactic therapy before dental and other surgical procedures. A similar response is generally seen in an individual patient with subsequent doses, and thus the factor VIII level attained after a trial dose can be used to predict the response to future therapy.²⁴

The recommended intravenous dosage of desmopressin is 0.3 µg/kg, administered in 25 to 50 mL of normal saline, over a period of 20 to 30 minutes.²⁵ A concentrated form of desmopressin is available for intranasal administration to treat bleeding disorders. The appropriate dose of concentrated intranasal desmopressin is 150 µg (1 puff) for persons weighing less than 50 kg, and 300 µg (1 puff in each nostril) for persons weighing more than 50 kg.²⁶

Antifibrinolytic Therapy

Antifibrinolytics (both epsilon-aminocaproic acid [EACA] and tranexamic acid) reversibly block the lysine binding sites of plasminogen, preventing its activation to plasmin and thus inhibiting the lysis of polymerized fibrin. EACA is also believed to stabilize the active form of thrombin activatable fibrinolysis inhibitor (TAFIa). It is believed that inactivation of TAFIa is due to conformational rearrangements in the TAFIa molecule; EACA has been shown to slow down spontaneous inactivation of TAFIa, thus curtailing fibrinolysis.²⁷ Although hemostasis is generally achieved with either factor VIII replacement or desmopressin, the risk of recurrent bleeding from oral mucosal surfaces is dramatically reduced with the use of antifibrinolytic agents. These agents are typically contraindicated in patients with hematuria because they can cause a clot to form in the urinary bladder or ureters, leading to obstruction.

EACA is available in intravenous, oral tablet, and elixir formulations; the oral dose is 100 to 200 mg/kg initially (maximum dose, 10 g), followed by 50 to 100 mg/kg per dose every 6 hours (maximum dose, 5 g). Tranexamic acid is available in 650-mg capsules; the dose is 25 mg/kg every 6 to 8 hours.^28,29 To treat spontaneous oral hemorrhage or to prevent bleeding from dental procedures in patients with hemophilia, either drug is usually begun in conjunction with desmopressin or factor replacement therapy immediately prior to the procedure and continued for up to 7 days or until mucosal healing is complete. Nonsteroidal anti-inflammatory drugs and aspirin affect platelet function and hence are contraindicated in affected individuals.³⁰

PROPHYLAXIS

Patients with mild to moderate hemophilia typically bleed only after trauma, although the trauma needed to induce bleeding may be more minor than that which would cause bleeding in a normal individual. They usually do not suffer from significant morbidities, whereas patients with severe hemophilia often have spontaneous severe muscle and joint bleeds and can develop early crippling hemophilic arthropathy. Hence, routine prophylaxis has now become the standard of care in the United States and other developed countries in the management of patients with severe hemophilia. Prophylactic replacement therapy with cryoprecipitate in boys with severe hemophilia was first used nearly 50 years ago in Sweden³¹ and the Netherlands,³² and was shown to reduce the number and the severity of bleeds.³² Moreover, it was observed that early prophylaxis was more effective in preventing arthropathy compared to starting later in life, and that radiologic joint damage could not be reversed by prophylaxis. Subsequently, primary prophylaxis, defined as the start of regular, continuous treatment before the age of 2 years or after the occurrence of first joint bleed,³³ was recommended and eventually became the standard treatment; it is currently recommended by the World Health Organization/World Federation of Hemophilia (WFH).³⁴

The timing to begin prophylaxis is somewhat controversial, but many authors suggest starting prophylaxis before the first hemarthrosis occurs. Several studies have reported a wide variation in the age at first joint bleed, ranging from 0.2 to 5.8 years, with medians of 1.6 to 1.7 years.^35,36 It has been suggested that arthropathy is best prevented if prophylaxis is started before the second or third joint bleed, but the benefits of starting before the occurrence of first bleed have not been established.^37,38 The Swedish experience provides strong support for early prophylaxis.³⁹ In an analysis of 121 patients with severe hemophilia, age at initiation of prophylaxis was an independent predictor of the development of arthropathy, but dose and interval of prophylaxis at the start of prophylactic treatment were not.³⁹

In the Italian ESPRIT study, it was shown that children randomly assigned to prophylaxis had significantly fewer total bleeding episodes and joint bleeding episodes compared with those assigned to episodic therapy. Eleven of 21 patients (52%) in the prophylaxis group had on average less than 1 hemarthrosis per year, whereas only 4 of 19 patients in the episodic therapy group (21%) had the same low frequency of bleeding (P < 0.05).⁴⁰ In a study of long-term prophylaxis versus on-demand treatment comparing age-matched Danish and Russian patients, the median annual number of joint bleeds in patients on prophylaxis was 1, while patients managed with on-demand treatment experienced a median of 37 joint bleeds. Patients taking prophylaxis also had a statistically significantly better quality of life estimate (P < 0.001) and better functional independence.⁴¹ In another trial, prophylaxis was initiated between the ages of 6 and 30 months based on a history of joint hemorrhage rather than age. Radiologic evidence of preserved joint architecture was found in 93% of participants in the prophylaxis group at 6 years of age. In this group, 18 of 32 (56%) children had 1 or 2 bleeds into one or more index joints before prophylaxis, and 17 (53%) had 1 to 5 hemorrhages into 1 or more index joints during prophylaxis. Prophylaxis was efficacious in decreasing bleeding and joint damage after up to 5 hemarthroses.⁴²

Optimal Prophylactic Regimen

Although the benefits of prophylactic replacement therapy are firmly established, the optimal dose and frequency remain unclear. The half-life of clotting factor concentrates is short: about 8 hours for factor VIII in children, and about 12 hours for factor IX. As a result, prophylactic therapy is most effective when given frequently. The most common factor VIII concentrate dosing regimen for prophylaxis in hemophilia A is 25 to 40 IU/kg 3 times per week; for hemophilia B, a dose of 80 to 100 IU/kg is given twice weekly. This is aimed at a pre-infusion level > 1% to mimic the clinical phenotype of moderate hemophilia.

Recently, the US Food and Drug Administration (FDA) approved the first long-lasting antihemophilic factor (recombinant) Fc fusion protein for use in adults and children with hemophilia A. This medication contains the Fc region of human immunoglobulin G1 (IgG1), which binds to the neonatal Fc receptor (FcRn). FcRn is part of a naturally occurring pathway that delays lysosomal degradation of immunoglobulins by cycling them back into circulation and prolonging their plasma half-life. Dosing for routine prophylaxis is 50 IU/kg every 4 days; it may be adjusted based on patient response, with dosing in the range of 25 to 65 IU/kg at 3- to 5-day intervals. More frequent or higher doses up to 80 IU/kg may be required in children younger than 6 years.⁴³

DEVELOPMENT OF INHIBITORS

FACTOR VIII INHIBITORS

Despite the success in the clinical management of hemophilia A, treated patients remain at risk for developing neutralizing antibodies that inhibit factor VIII activity. An inhibitor is a polyclonal high-affinity IgG that is directed against the factor VIII protein and renders exogenous factor ineffective. IgG4 antibodies are predominant and do not fix complement.

Risk Factors

The pathophysiology underlying the development of factor VIII inhibitors is a T-helper (Th)–cell dependent event that involves antigen-presenting cells and B lymphocytes; why only a fraction of patients experience this adverse effect of factor therapy is not known. Patients with mild/moderate hemophilia have a lower risk for inhibitor development than those with severe hemophilia A. The estimated prevalence of inhibitors ranges from 3% to 13% in mild to moderate disease,^44–46 and up to 36% in severe hemophilia A.^47,48 Usually the presence of an inhibitor in patients with mild/moderate hemophilia is suggested by a change in bleeding pattern: patients who previously used to bleed only after trauma or surgery suddenly start to experience severe spontaneous bleeding. This change in bleeding pattern is explained by cross-reactivity of the inhibitor with the mutated factor VIII of the patient, resulting in a residual factor level of < 0.01 IU/dL.^49–51 Occasionally, there is no change in the residual factor VIII level but an inhibitor is detected in the Bethesda assay and/or there is lack of efficacy of factor VIII trans-fusions.^51–53

Genetic factors. Data indicate that the risk of developing neutralizing antibodies is to a large extent determined by patient-related genetic factors.^54,55 The immune response to factor VIII is similar in up to 80% of family members, significantly higher than expected compared with data from unrelated subjects. In a meta-analysis of patients with severe hemophilia A, the inhibitor incidence was twice as high in African American patients as compared with white patients.⁵⁶ One study showed that patients of Hispanic ancestry with severe hemophilia A have a higher prevalence of neutralizing inhibitors than non-Hispanic white patients.⁵⁷

Type of causative mutation. In severe hemophilia A, the risk of inhibitor formation is associated with the type of mutation. More disruptive mutations in the factor VIII gene, such as the intron 22 inversion, large gene deletions, and stop codons are associated with an approximately 35% risk of inhibitor formation, compared with only about 5% in those with missense mutation and small deletions.⁵⁸ Persons with mutations involving large gene deletions, nonsense mutations, and intrachromosomal aberrations are usually at higher risk for the development of inhibitors than persons with missense mutations, small deletions/insertions, and splice site mutations.^59,60 A relatively high risk is also encountered in patients with splicing errors and frame-shift mutations.⁶¹

Major histocompatibility complex. The HLA class I alleles A3, B7, and C7, as well as the class II alleles DQA0102, DQB0602, and DR15 have all been associated with a slightly higher risk for inhibitor development in unrelated patients, whereas the HLA C2, DQA0103, DQB0603, and DR13 alleles might be protective.^62,63

Immune-regulatory molecules. In the Malmö International Brother Study, polymorphic sites in the genes coding for interleukin 10 (IL-10), tumor necrosis factor-α, and cytotoxic T lymphocyte–associated protein 4 were all associated with the risk of developing inhibitors.^64–66 In this study, a 134 bp–long variant of a CAA microsatellite in the promoter region (IL-10.G) was identified in 26.8% of patients with hemophilia A. Thirty-two of these patients (72.7%) developed inhibitors as compared with 37.5% of those without the allele.⁶⁵

Intensive exposure to factor VIII. Inhibitors in mild/moderate hemophilia seem to occur more commonly later in life, and an episode of intensive treatment with factor VIII concentrate has been reported to precede detection in most reported cases. In the series reported by Hay et al,⁶⁷ 16 out of 26 inhibitors were detected after such intensive replacement therapy, and no particular concentrate was implicated.

INHIBITORS TO FACTOR IX

Factor IX inhibitors are relatively uncommon, occurring in only 1% to 3% of persons with hemophilia B. This is in striking contrast to hemophilia A, where approximately 30% of patients develop inhibitors. The majority of patients with hemophilia B who develop inhibitors have severe hemophilia B.

Risk Factors

Certain mutations in the factor IX gene are associated with an increased incidence of inhibitor development. Large deletions and frame-shift mutations leading to the loss of coding information are much more likely to be associated with inhibitor development. Large deletions account for only 1% to 3% of all hemophilia B patients, but account for 50% of inhibitor patients.⁶⁸ Patients with hemophilia B who develop inhibitors are at risk for developing anaphylactic reactions to factor IX–containing products. Anaphylaxis occurred more frequently in families with null mutations (large deletions, frame-shift mutations, or nonsense mutations) than in those with missense mutations.⁶⁹ With hemophilia A, approximately 40% to 50% of black individuals develop inhibitors, but no such association has been found in hemophilia B. Individuals who develop an inhibitor to factor IX do so relatively early in life (within the first 4 to 5 years), after a median of 9 to 11 exposure days to any factor IX–containing products. Because of the severity of a potential anaphylactic reaction occurring early in life after very few exposures to factor IX, all infants and small children with severe hemophilia B should be closely followed over their first 10 infusions with any factor IX–containing products in a facility equipped to treat anaphylactic shock.^70–72 A comparison of inhibitors in hemophilia A and B is shown in Table 3.

TREATMENT OF ACUTE BLEEDS IN PATIENTS WITH FACTOR VIII INHIBITORS

The available therapeutic agents for treatment of acute hemorrhage in children with hemophilia A with an inhibitor include high-dose recombinant or plasma-derived factor VIII concentrate, activated prothrombin complex concentrates (aPCCs), and recombinant activated factor VII (rFVIIa). In addition, antifibrinolytics may be used as an adjunct therapy.

Patient response to each treatment varies widely, with some patients responding well to one treatment and less well to another. Neither the patient's history nor standard lab tests can assist in making the best choice for the patient. A personalized approach to factor selection is used, and the dosing of that particular agent is often determined primarily by clinical assessment. Inhibitors are quantitated using the Bethesda inhibitor assay and clinically are classified as low- and high-responding inhibitors (Table 4). Inhibitor screening should be done prior to invasive procedures and periodically during the first 50 days of treatment since the risk for inhibitor development is highest during this period.

Low-Responding Inhibitors

A low-responding inhibitor is one in which inhibitor titers are < 5 Bethesda units (BU)/mL; patients with low-responding inhibitors can generally be treated with factor VIII concentrates at higher doses.⁷³ Because the effect of factor VIII inhibitor is usually delayed, the Bethesda titer in plasma is determined after a 2-hour incubation period. As a result of this time delay, continuous administration of factor VIII is usually found to be effective.⁷⁴ For a serious limb- or life-threatening bleeding episode, a bolus infusion of 100 IU of factor VIII per kg of body weight is administered, and the level is maintained by treatment at a rate of 20 IU/kg/hr. An assay for factor VIII should be performed 1 hour after the bolus infusion and at least daily thereafter. As the antibody titer drops, the daily level of factor VIII may rise and thus downward adjustment of the continuous infusion rate may be required. For routine joint and muscle hemorrhage, patients can usually be managed with infusions at twice the usual dosage. Routine inhibitor assays should be performed after exposure to factor VIII to determine whether an anamnestic response has occurred.

High-Responding Inhibitors

Most clinicians caring for patients with limb- or life-threatening bleeding episodes prefer to use products for which therapeutic levels can be monitored. As described earlier, continuous admin-istration of factor VIII is often effective because of the time delay in inhibition by the antibody. An initial dose of 100 to 200 IU/kg can be administered, and factor VIII levels can be determined 1 hour after initiation of continuous infusion at a rate of 20 to 40 IU/kg/hr. If a factor VIII level cannot be obtained (ie, patients with inhibitor titers > 5 to 10 BU/mL), alternative approaches include the bypassing agents aPCC and rFVIIa.

First used in the 1970s, aPCCs represented a significant improvement in the management in patients with hemophilia with inhibitors. They contain multiple activated serine protease molecules; activated factor X and prothrombin are the main active components in FEIBA (factor eight inhibitor bypassing activity), the most commonly used aPCC in the United States. FEIBA is a pooled plasma product that contains activated factors II, VII, IX and X, and has a duration of action of about 6 to 12 hours. For treatment of acute bleeds, the recommended dose of FEIBA is 50 to 100 IU/kg infused every 8 to 12 hours (maximum daily dose of 200 IU/kg). There is a risk of thrombosis/disseminated intravascular coagulation (DIC) with very large doses given frequently (> 200 IU/kg/day).

rFVIIa directly activates factor X and increases thrombin production on the surface of activated platelets in the absence of factor VIII or factor IX. Standard dosing of rFVIIa is 90 to 120 µg/kg, and many hemophilia treatment centers use higher doses (270 µg/kg/dose), especially in children and young adults. The half-life is about 1.5 to 3 hours, and therefore frequent administration (every 2–6 hours) is required. In one study that assessed the safety and efficacy of fixed-dose rFVIIa in the home setting, hemostasis was achieved in 566 (92%) of evaluable bleeding episodes, and following administration of the additional maintenance dose, hemostasis was maintained in 95% of successfully treated cases.⁷⁵ As with aPCCs, there is no standardized quantitative laboratory test for measuring the effectiveness of rFVIIa therapy.

All currently used bypassing agents are associated with a risk of thrombotic complications including thromboembolism, DIC, and myocardial infarction. These complications are very rare in patients with hemophilia, however. In general, bypassing agents work for most bleeds and for most patients, but are not as predictable as factor replacement therapy and cannot be monitored by laboratory assays.

TREATMENT OF ACUTE BLEEDS IN PATIENTS WITH FACTOR IX INHIBITORS

rFVIIa and FEIBA are the mainstays of treatment of bleeding episodes in individuals with hemophilia B complicated by an inhibitor to factor IX. Treatment of hemorrhagic episodes in these patients depends on the type of bleeding episode experienced, the inhibitor classification (high- versus low-responding [Table 4]), and the history and severity of infusion reactions. Patients with low-responding inhibitors who have not experienced infusion reactions may be treated with doses of factor IX concentrate calculated to overcome the inhibitor titer and achieve a hemostatic level. In patients with high-responding inhibitors, the use of factor IX concentrates is impractical because of the inhibitor titer or the anamnestic response. Regardless of inhibitor titer, in patients with a history of an anaphylactic event, factor IX usage is contraindicated.

The most commonly used therapy for hemostatic control in patients with high-responding inhibitors with factor IX deficiency and a history of infusion reaction is rFVIIa; the standard dosing regimen is 90 to 120 µg/kg/dose administered every 2 to 3 hours, with a maximum dose of 270 µg/kg/dose. aPCCs, which contain factor IX, can be utilized if the patient has not experienced prior infusion reactions. Repeated exposures to products containing factor IX may stimulate the inhibitor titer and prevent its natural decline over time. This can pose a problem in cases of life- or limb-threatening hemorrhage unresponsive to rFVIIa as these patients will not have factor IX available as an effective mode of therapy. The dosing of FEIBA ranges from 50 to 100 IU/kg every 12 hours, with daily dosing not to exceed 200 IU/kg.

IMMUNE TOLERANCE INDUCTION

Because of the associated inhibitor-related morbidity resulting from limited treatment options, antibody eradication is the ultimate goal in inhibitor management. The only proven strategy for achieving antigen-specific tolerance to factor VIII or factor IX is immune tolerance induction (ITI) therapy. Successful ITI in hemophilia A is currently defined as both an undetectable inhibitor titer (< 0.6 BU), and normalized factor VIII pharmacokinetics, which in turn is defined as plasma factor VIII recovery > 66% of expected and a half-life > 6 hours, determined following a 72-hour factor VIII exposure-free period (Consensus Proceedings from the Second International Conference on Immune Tolerance Therapy, Bonn, Germany, 1997 [unpublished]). Once successful immune tolerance is achieved, long-term prophylaxis is commonly instituted. Using conclusions drawn from international consensus criteria and analysis of the International Immune Tolerance Registry, the I-ITI study has defined ITI failure by the presence of either of 2 criteria:

1. Failure to attain the definition of success within 33 months of uninterrupted ITI;

2. Failure to demonstrate a progressive 20% reduction in inhibitor titer over each 6-month period of uninterrupted ITI, beginning 3 months after initiation to allow for expected anamnesis.^76–78

This definition implies a minimum ITI trial period of 9 months before failure is declared.

The European Hemophilia Standardization Board (EHSB), the International Consensus Panel (ICP), and the United Kingdom Hemophilia Center Doctors’ Organization (UKHCDO) have agreed that it is preferable to initiate ITI at a titer of < 10 BU/mL, unless, per the ICP, the titer does not decline over a period of 1 to 2 years and/or inhibitor development is associated with severe or life-threatening bleeding. The ICP noted that for “poor-risk” ITI patients (defined by a historical titer of > 200 BU/mL and/or a pre-ITI inhibitor titer of > 10 BU/mL and/or an interval of > 5 years since inhibitor diagnosis), published efficacy data are limited to dosing regimens > 200 IU/kg/day. The groups all independently concluded that ITI has been successfully performed using recombinant and plasma-derived factor VIII replacement therapy (usually the product on which they developed the inhibitor), and that there are no data to support the superiority of any single product type.^79–81 However, both EHSB and ICP have suggested that VWF-containing concentrates be considered for patients who fail ITI using high-purity factor VIII.^79,80

The recommendations from US guidelines for ITI in patients with hemophilia A and inhibitors are listed in Table 5.⁸²

Before the advent of factor products for the treatment of hemophilia, hemarthrosis was one of the leading causes of morbidity. Today, the routine use of prophylactic treatment has resulted in a significant improvement in the lifestyle, quality of life, and life expectancy of these patients. However, despite best efforts, some patients will have severe joint destruction as a result of repeated articular bleeding episodes during their early years. This leads to pain and significant functional disability, thus impairing the quality of life. The basic pathology behind hemophilic arthropathy is chronic synovitis.

It is common to observe a pattern of repeated bleeding (chronic hemarthrosis), especially in patients with severe hemophilia, that can lead to chronic synovitis, inflammatory arthritis, and progressive arthropathy. Therefore, the key to preventing hemophilic arthropathy is aggressive management of the initial hemarthrosis. This is generally accomplished with the use of clotting factor replacement, restorative physiotherapy, and close clinical follow-up. If chronic synovitis develops, synovectomy may be considered in order to slow the progression of the hemophilic arthropathy and to prevent the development of major articular surface erosions that can lead to end-stage arthropathy.⁸³ Primary prophylaxis is discussed earlier and is the mainstay of prevention of chronic hemophilic arthropathy.

SYNOVECTOMY

The emergence of chronic hemophilic hemarthrosis is incited by a hypertrophic and highly vascular synovium. Removal of the synovium prevents further joint damage,⁸⁴ and can be accomplished through surgical and nonsurgical procedures.

Surgical excision of the hypertrophic synovium can be performed through open or arthroscopic procedures. The open approach has largely been replaced by arthroscopic synovectomies. Regardless of the approach, these patients need prolonged hospitalization, extensive factor replacement, and exhaustive physiotherapy. Moreover, patients with inhibitors are usually not considered candidates for surgical synovectomy.

Chemical and radioactive agents injected intra-articularly can decrease the volume and activity of the synovial tissue. Due to the minimally invasive nature of these procedures, nonsurgical synovectomies are of special importance for hemophilic patients with inhibitors to clotting factors.

Chemical Synovectomy

Chemical synovectomies, using thiotepa, osmic acid, D-penicillamine and other agents, have been used in the distant past. Rifampicin, which is used an antibiotic, is now the most commonly used chemical for the purpose of synovectomy, and the one that has shown better results in terms of decreasing hemarthrosis.⁸⁵ Each one of the injections should be accompanied by prophylactic administration of clotting factor concentrate. Excellent results (no synovitis and restoration of previous function) have been reported in up to 83% of patients at an average of 2.4 years after the intra-articular injection of rifampicin. As the pathology of the joint becomes more severe, however, the number of injections required to achieve improvement increases. Younger patients and smaller joints benefit more from this procedure.

Radiation Synovectomy

Radiosynovectomy (RS) and radiosynoviorthesis are common terms used to describe the synovial ablation accomplished by intra-articular injection of radioisotopes. Isotopes of gold, yttrium, rhenium, and dysprosium have been used to perform radiation synovectomies in patients with hemophilia. Yttrium-90, a pure beta emitter with adequate particle size and depth penetration, has been used successfully for the treatment of hemophilic synovitis.

The local (growth plate and articular cartilage) and remote effects of radiation are a concern. There have been no reported cases of growth plate disturbance after radiosynovectomy, even after the use of beta emitters such as gold-198.⁸⁶ Articular cartilage is highly resistant to radiation, and although damage is theoretically possible, none has been reported. Progressive degeneration of treated joints does occur, but the rate is slower than that expected without radiosynovectomy. The principal concern is the potential for late, radiation-induced neoplasia. However, the safety of intra-articular radioisotopes is supported by a long-term follow-up study of more than 5000 RS procedures performed for rheumatoid arthritis, which found no reported radiation-induced malignancies.⁸⁷

One review analyzed the safety of RS in pediatric patients with hemophilia to provide a risk-benefit assessment. During knee RS, patients receive a radiation dose of approximately 0.74 mSv, and during elbow and ankle RS, a dose of approximately 0.32 mSv. The radiation dose from natural sources is approximately 2 mSv per year and the recommended limit for patients (apart from natural sources) is 1 mSv per year. The lifetime cancer risk increases about 0.5% per 100 mSv per year. Considering the risks and benefits of RS, the authors recommend that clinicians consider this procedure in children with inhibitors or in patients without inhibitors when bleeding is recurrent and persistent despite aggressive factor replacement.⁸⁸ External-beam radiation has been extensively studied and carries a small risk of osteosarcoma induction.

ACQUIRED INHIBITORS TO FACTOR VIII

Acquired hemophilia (AH) has an estimated prevalence of 1.48 cases per million per year, and a reported mortality between 9% and 22%.^89,90 AH is uncommon in children younger than 16 years (prevalence estimated at 0.045/million/year), and may be underdiagnosed in persons older than age 85 (prevalence estimated at 14.7/million/year).⁸⁹ In the largest published population series, 50% to 60% of diagnosed individuals were previously healthy with no identified underlying disease state.^90–91 Underlying conditions consistently associated with AH include pregnancy, evolving or pre-existing autoimmune or malignant disorders, and rarely medications. Primary among the autoimmune disorders are collagen vascular disorders, including systemic lupus erythematosus, rheumatoid arthritis, myasthenia gravis, multiple sclerosis, and autoimmune hemolytic anemia. Most antibodies are mixtures of polyclonal IgG1 and IgG4 immunoglobulins, with the IgG4 molecules mainly responsible for inhibiting clotting activity. The clinical picture of AH is characterized by acute onset of severe bleeding in individuals who previously had no history of bleeding diathesis. Patients generally present with mucocutaneous bleeding (eg, epistaxis and gastrointestinal bleeding), as well as soft tissue bleeding (eg, extensive ecchymoses and hematomas).

The 2 major goals of treatment of AH are the immediate control of acute and chronic bleeding and the long-term suppression/eradication of the autoantibody inhibitor. For patients with an inhibitor titer < 5 BU/mL, administration of desmopressin and concentrates of human recombinant factor VIII may raise the factor VIII activity levels in plasma. If the inhibitor titer is > 5 BU/mL, or if bleeding persists despite infusions of factor VIII concentrates, then factor VIII bypassing agents, such as aPCCs or rFVIIa, are indicated. Local measures for treatment of mucosal hemorrhage, such as antifibrinolytic agents or topical fibrin glues, are helpful.

The primary aim in long-term management of AH is to eradicate the factor VIII autoantibodies so that further bleeding can be averted. Although in some clinical situations (postpartum women and drug-related AH) factor VIII antibodies may remit spontaneously, most published guidelines and algorithms recommend early initiation of eradication therapy. This is usually achieved through immunosuppressive medications or immunomodulation. Successful immunosuppression regimens in AH have most frequently used corticosteroids as the cornerstone, either as a single agent or in combination with cyclophosphamide. In a prospective randomized trial involving 31 participants treated with prednisone 1 mg/kg/day for 3 weeks, 32% achieved complete remission. In participants with antibody persistence after 3 weeks, switching to oral cyclophosphamide 2 mg/kg/day as second-line therapy appeared more effective than continuing prednisone (complete remission rate 50% versus 42%).⁹²

Other immunosuppressive medications have been employed for eradication of refractory autoantibody inhibitors, including azathioprine, cyclosporine, tacrolimus, mycophenolate motefil, and sirolimus. Controlled studies have not been performed to confirm their comparative safety and efficacy in sufficiently large populations. Anti-CD20 antibody has been used to treat inhibitors in patients with both congenital and acquired hemophilia.^93,94 Other less frequently used treatment options include administration of intravenous immunoglobulins (IVIG) in large doses. IVIG by itself rarely is able to induce a complete remission, but may be useful adjunctive therapy along with immunosuppressants, as part of an ITI regimen, or with extracorporeal plasmapheresis.

INTRODUCTION

Hemophilia A and B are the most common severe inherited bleeding disorders. The incidence of hemophilia is 1 in 5000 live male births, with hemophilia A occurring 4 times more commonly than hemophilia B. The associated decrease in factor VIII in hemophilia A was initially identified in 1947, and the decrease in factor IX associated with hemophilia B was identified 5 years later.^1,2 Both conditions are inherited as X-linked recessive traits. Queen Victoria of Britain, who reigned from 1837 to 1901, was a carrier of hemophilia and had 2 carrier daughters, Alice and Beatrice, and a son with hemophilia, Leopold.³ In 1984 and 1985, the genes for factor VIII and factor IX were cloned, and in 1989 recombinant factor VIII was first used clinically.^4–7

PATHOPHYSIOLOGY

Both factors VIII and IX are crucial for normal thrombin generation in the coagulation pathway. After any injury, the initial hemostatic event is the formation of a platelet plug. Once the platelet plug is formed, subsequent generation of fibrin prevents continued oozing from the affected site. In hemophilia A and B, the propagation phase of coagulation is impaired, and as a result, the formation of clot is delayed and is not robust. Due to the delayed formation of an abnormal clot, patients with hemophilia do not bleed rapidly but rather ooze continuously. Rebleeding is a common occurrence in inadequately treated patients.⁸

GENETICS

The gene for factor VIII (F8) is located in the most distal band (Xq28) of the long arm of the X chromosome. Spanning more than 186 kb, it is one of the largest genes known.^9,10 The gene for factor IX (F9) is located at Xq27.1 and spans 33 kb.⁷ Defects in the F8 gene associated with hemophilia A may be divided into several categories: gross gene rearrangements; insertions or deletions of genetic sequence of a size varying from 1 base pair up to the entire gene; or single DNA base substitutions resulting in either amino acid replacement (missense), premature peptide chain termination (nonsense, or stop mutations), or mRNA splicing defects. All classes of defects can result in severe disease. However, the single most clinically important defect is a gene rearrangement (an inversion) involving F8 intron 22, which results in approximately 50% of all severe hemophilia A cases worldwide.^11,12 In hemophilia B, point mutations are by far the most common type of abnormality. Generally, they are caused by DNA polymerases adding the wrong nucleotide during replication.¹³

HEMOPHILIA IN FEMALES

X-Inactivation (also called Lyonization) is a process that occurs early in embryonic development in female mammals where 1 of the 2 copies of the X chromosome present is inactivated; it is the reason why some female carriers of hemophilia can become symptomatic. Approximately one third of carriers have clotting factor levels of less than 60% of normal and may experience abnormal bleeding.^14,15 In most cases, carriers experience symptoms similar to those seen in men with mild hemophilia, as well as some that are specific to women. Symptomatic carriers and women with hemophilia may bruise more easily; may experience prolonged bleeding after surgery; may experience serious bleeding after trauma; often have heavier and more prolonged bleeding during their periods (menorrhagia) and are more likely to require an iron supplement or to undergo hysterectomy; and are more likely to have postpartum bleeding following childbirth.^14,15

CLINICAL MANIFESTATIONS

Hemorrhage in patients with hemophilia may occur with minimal or unknown trauma. Patients with severe hemophilia (factor level of < 1 IU/dL or < 1% of normal) often experience spontaneous bleeding into joints or muscles. Those with moderate hemophilia (factor level of 1–5 IU/dL or 1%–5% of normal) seldom experience spontaneous hemorrhage and usually have prolonged bleeding with minor trauma or surgery. Patients with mild hemophilia (factor level > 5 IU/dL but less than 40 IU/dL or > 5% but < 40% of normal) experience severe hemorrhage only following moderate to severe trauma or surgery, and rarely experience spontaneous bleeding. Depending on the site, bleeding can be serious (joints; muscles, especially deep compartments [iliopsoas, calf, and forearm]; mucous membranes in the mouth, gums, nose, and genitourinary tract) or life-threatening (intracranial, neck/throat, gastrointestinal). The joints and muscles are the most common sites of bleeding (Table 1).

MUSCULOSKELETAL BLEEDING

The hallmark of hemophilia is deep bleeding into the joints and muscles. Without prophylactic factor treatment, patients with severe hemophilia A or B may have a bleeding episode as often as once or twice a week. Hemarthrosis episodes typically begin when the child reaches the toddler age. One of the first signs of hemarthrosis is a tingling sensation and feeling of warmth which is soon followed by pain and decreased range of motion of the joint as a result of distension of the joint capsule. Prompt, aggressive treatment with factor replacement therapy is the key to prevent further bleeding and minimize potential long-term complications. Severe chronic arthropathy may develop in older children and adults who have not received aggressive treatment (Figure).

Bleeding into the muscle can manifest as a vague feeling of pain on motion. Swelling may not be obvious and the mass may be difficult to palpate, although the circumference of the affected limb will be increased. Among the muscle bleeds, iliopsoas bleed deserves a special mention because of its potential to cause life-threatening hypovolemic shock as large volumes of blood can be lost into the retroperitoneal space. These patients present with vague abdominal pain or upper thigh discomfort. The hip is flexed and outwardly rotated. The diagnosis is confirmed by computed tomography (CT) or ultrasound.

LIFE-THREATENING HEMORRHAGE

Central Nervous System Bleeding

Most central nervous system (CNS) events, which involve bleeding inside the skull or spinal canal, are caused by trauma. CNS hemorrhage is the most common form of severe hemophilic trauma. However, since patients with hemophilia can experience bleeding even weeks after a minor head injury, a history of head trauma may be hard to determine, particularly in children. Spontaneous CNS bleeding in individuals with hemophilia is rare except when there has been a recent antecedent CNS hemorrhage (ie, a recurrent bleed at a previously injured site) or when there is an associated anatomic lesion that predisposes to acute hemorrhage (eg, aneurysm or arteriovenous malformation). Data from the Universal Data Collection Project of the U.S. Centers for Disease Control and Prevention indicates that predisposing risk factors for intracranial hemorrhage include HIV infection, presence of inhibitory antibodies, and age younger than 5 years or older than 51 years.¹⁶ Neonatal intracranial hemorrhage is most commonly due to birth trauma. Difficult vaginal deliveries (often requiring the application of forceps or vacuum extraction) are predisposing factors for intracranial hemorrhage in hemophilic newborns.

The site of intracranial CNS bleeding can be subdural, epidural, or intraparenchymal. Bleeding at any of these sites can cause rapidly deteriorating CNS brain function, associated brain swelling, and, in the most extreme circumstances, herniation of the brainstem and rapid death. If the bleeding is stopped with rapid clotting factor replacement, adverse clinical effects can be avoided. However, with intraparenchymal hemorrhage, even small hemorrhages can induce permanent structural and/or neurologic sequelae (in particular, if the anatomic site of the bleed is essential for routine brain function).¹⁷

Throat and Neck Hemorrhage

An acute neck injury or a retropharyngeal hemorrhage induced by dental or oral surgical instrumentation can lead to a dissecting facial plane hematoma. This in turn can sometimes lead to compression and acute airway compromise. Bleeding from these injuries that is compressing or compromising the airway may require a rapid clinical response.¹⁸ The time from the injury until the trachea is compressed may be long, sometimes many hours. However, once the compression is sufficient to cause difficulty breathing, there may be a short amount of time to stop the bleeding and prevent complete respiratory obstruction.

MUCOCUTANEOUS BLEEDING

One of the common manifestations of hemophilia is oral bleeding. Tooth extraction poses a specific problem, and bleeding following extraction can be the first symptom that leads to the diagnosis of hemophilia. Bleeding after circumcision may also suggest the diagnosis. In 1 study cohort looking at sites of initial bleeding episodes in babies with hemophilia diagnosed before the age of 2 years, bleeding from circumcision and other iatrogenic causes tended to be most common in the neonatal period. Circumcision bleeding events occurred more often in infants with no family history (43%) as compared to those born to known maternal carriers (9.2%) or to mothers with some other family history of hemophilia (14.3%).¹⁹

Gastrointestinal (GI) bleeding occurs occasionally in hemophilia, and a wide spectrum of esophageal and GI bleeding may occur. A review of 41 episodes of GI bleeding in hemophilia patients who presented to 1 institution over 10 years implicated duodenal ulcer (22%), unknown site (22%), and gastritis (14%) as the most common sources.²⁰ Mallory-Weiss syndrome has also been cited as a cause for upper GI bleeding in hemophilia patients.²¹

PRINCIPLES OF TREATMENT

Understanding the pathophysiology of hemophilia as well as the type and severity of hemophilia and the inhibitor status in an individual patient are paramount in the management of a patient with hemophilia. In the past, management mainly focused on the treatment of acute bleeding episodes (Table 2). With data showing the benefit of bleed prevention, the management of hemophilia now focuses on prophylaxis of bleeding episodes, which prevents chronic arthropathy and improves quality of life.

ACUTE BLEEDING EPISODES

Dosing of Factor VIII Products

Dosing for factor VIII concentrate is as follows: 1 IU of factor VIII concentrate per kg will increase the circulating factor VIII level by 2% (ie, patient weight in kg × 50 IU/kg = 100% correction). For example, a 30-kg patient requiring 100% correction of factor VIII needs an infusion of 1500 IU of factor VIII (30 kg × 50 IU/kg).

Dosing of Factor IX Products

Dosing for factor IX concentrate is as follows: 1 IU of factor IX concentrate per kg will increase the circulating factor IX level by 1% (ie, patient weight in kg × 100 IU/kg = 100% correction). For example, a 30-kg patient requiring 100% correction of factor IX needs an infusion of 3000 IU of factor IX (30 kg × 100 IU/kg). Higher doses (120 to 130 IU/kg) of the recombinant factor IX product BeneFIX (Pfizer) may be needed to reach the 100% circulating factor IX level.

ADJUVANT THERAPY

Desmopressin

Desmopressin is a synthetic vasopressin analogue that increases plasma factor VIII and von Willebrand factor (VWF) levels; it is used to prevent and treat bleeding episodes associated with dental and surgical procedures in patients with mild and moderate hemophilia A and von Willebrand disease.²² Desmopressin causes the release of VWF and factor VIII from storage in the Weibel–Palade bodies of the endothelial cells that line the blood vessels. Individual response to desmopressin varies, with factor VIII level increasing between 2 and 15 times baseline level in patients with mild or moderate hemophilia A.²³ It is therefore recommended that patients undergo a therapeutic trial of desmopressin with laboratory measurement of response to factor VIII before it is used for treatment of bleeding episodes or as prophylactic therapy before dental and other surgical procedures. A similar response is generally seen in an individual patient with subsequent doses, and thus the factor VIII level attained after a trial dose can be used to predict the response to future therapy.²⁴

The recommended intravenous dosage of desmopressin is 0.3 µg/kg, administered in 25 to 50 mL of normal saline, over a period of 20 to 30 minutes.²⁵ A concentrated form of desmopressin is available for intranasal administration to treat bleeding disorders. The appropriate dose of concentrated intranasal desmopressin is 150 µg (1 puff) for persons weighing less than 50 kg, and 300 µg (1 puff in each nostril) for persons weighing more than 50 kg.²⁶

Antifibrinolytic Therapy

Antifibrinolytics (both epsilon-aminocaproic acid [EACA] and tranexamic acid) reversibly block the lysine binding sites of plasminogen, preventing its activation to plasmin and thus inhibiting the lysis of polymerized fibrin. EACA is also believed to stabilize the active form of thrombin activatable fibrinolysis inhibitor (TAFIa). It is believed that inactivation of TAFIa is due to conformational rearrangements in the TAFIa molecule; EACA has been shown to slow down spontaneous inactivation of TAFIa, thus curtailing fibrinolysis.²⁷ Although hemostasis is generally achieved with either factor VIII replacement or desmopressin, the risk of recurrent bleeding from oral mucosal surfaces is dramatically reduced with the use of antifibrinolytic agents. These agents are typically contraindicated in patients with hematuria because they can cause a clot to form in the urinary bladder or ureters, leading to obstruction.

EACA is available in intravenous, oral tablet, and elixir formulations; the oral dose is 100 to 200 mg/kg initially (maximum dose, 10 g), followed by 50 to 100 mg/kg per dose every 6 hours (maximum dose, 5 g). Tranexamic acid is available in 650-mg capsules; the dose is 25 mg/kg every 6 to 8 hours.^28,29 To treat spontaneous oral hemorrhage or to prevent bleeding from dental procedures in patients with hemophilia, either drug is usually begun in conjunction with desmopressin or factor replacement therapy immediately prior to the procedure and continued for up to 7 days or until mucosal healing is complete. Nonsteroidal anti-inflammatory drugs and aspirin affect platelet function and hence are contraindicated in affected individuals.³⁰

PROPHYLAXIS

Patients with mild to moderate hemophilia typically bleed only after trauma, although the trauma needed to induce bleeding may be more minor than that which would cause bleeding in a normal individual. They usually do not suffer from significant morbidities, whereas patients with severe hemophilia often have spontaneous severe muscle and joint bleeds and can develop early crippling hemophilic arthropathy. Hence, routine prophylaxis has now become the standard of care in the United States and other developed countries in the management of patients with severe hemophilia. Prophylactic replacement therapy with cryoprecipitate in boys with severe hemophilia was first used nearly 50 years ago in Sweden³¹ and the Netherlands,³² and was shown to reduce the number and the severity of bleeds.³² Moreover, it was observed that early prophylaxis was more effective in preventing arthropathy compared to starting later in life, and that radiologic joint damage could not be reversed by prophylaxis. Subsequently, primary prophylaxis, defined as the start of regular, continuous treatment before the age of 2 years or after the occurrence of first joint bleed,³³ was recommended and eventually became the standard treatment; it is currently recommended by the World Health Organization/World Federation of Hemophilia (WFH).³⁴

The timing to begin prophylaxis is somewhat controversial, but many authors suggest starting prophylaxis before the first hemarthrosis occurs. Several studies have reported a wide variation in the age at first joint bleed, ranging from 0.2 to 5.8 years, with medians of 1.6 to 1.7 years.^35,36 It has been suggested that arthropathy is best prevented if prophylaxis is started before the second or third joint bleed, but the benefits of starting before the occurrence of first bleed have not been established.^37,38 The Swedish experience provides strong support for early prophylaxis.³⁹ In an analysis of 121 patients with severe hemophilia, age at initiation of prophylaxis was an independent predictor of the development of arthropathy, but dose and interval of prophylaxis at the start of prophylactic treatment were not.³⁹

In the Italian ESPRIT study, it was shown that children randomly assigned to prophylaxis had significantly fewer total bleeding episodes and joint bleeding episodes compared with those assigned to episodic therapy. Eleven of 21 patients (52%) in the prophylaxis group had on average less than 1 hemarthrosis per year, whereas only 4 of 19 patients in the episodic therapy group (21%) had the same low frequency of bleeding (P < 0.05).⁴⁰ In a study of long-term prophylaxis versus on-demand treatment comparing age-matched Danish and Russian patients, the median annual number of joint bleeds in patients on prophylaxis was 1, while patients managed with on-demand treatment experienced a median of 37 joint bleeds. Patients taking prophylaxis also had a statistically significantly better quality of life estimate (P < 0.001) and better functional independence.⁴¹ In another trial, prophylaxis was initiated between the ages of 6 and 30 months based on a history of joint hemorrhage rather than age. Radiologic evidence of preserved joint architecture was found in 93% of participants in the prophylaxis group at 6 years of age. In this group, 18 of 32 (56%) children had 1 or 2 bleeds into one or more index joints before prophylaxis, and 17 (53%) had 1 to 5 hemorrhages into 1 or more index joints during prophylaxis. Prophylaxis was efficacious in decreasing bleeding and joint damage after up to 5 hemarthroses.⁴²

Optimal Prophylactic Regimen

Although the benefits of prophylactic replacement therapy are firmly established, the optimal dose and frequency remain unclear. The half-life of clotting factor concentrates is short: about 8 hours for factor VIII in children, and about 12 hours for factor IX. As a result, prophylactic therapy is most effective when given frequently. The most common factor VIII concentrate dosing regimen for prophylaxis in hemophilia A is 25 to 40 IU/kg 3 times per week; for hemophilia B, a dose of 80 to 100 IU/kg is given twice weekly. This is aimed at a pre-infusion level > 1% to mimic the clinical phenotype of moderate hemophilia.

Recently, the US Food and Drug Administration (FDA) approved the first long-lasting antihemophilic factor (recombinant) Fc fusion protein for use in adults and children with hemophilia A. This medication contains the Fc region of human immunoglobulin G1 (IgG1), which binds to the neonatal Fc receptor (FcRn). FcRn is part of a naturally occurring pathway that delays lysosomal degradation of immunoglobulins by cycling them back into circulation and prolonging their plasma half-life. Dosing for routine prophylaxis is 50 IU/kg every 4 days; it may be adjusted based on patient response, with dosing in the range of 25 to 65 IU/kg at 3- to 5-day intervals. More frequent or higher doses up to 80 IU/kg may be required in children younger than 6 years.⁴³

DEVELOPMENT OF INHIBITORS

FACTOR VIII INHIBITORS

Despite the success in the clinical management of hemophilia A, treated patients remain at risk for developing neutralizing antibodies that inhibit factor VIII activity. An inhibitor is a polyclonal high-affinity IgG that is directed against the factor VIII protein and renders exogenous factor ineffective. IgG4 antibodies are predominant and do not fix complement.

Risk Factors

The pathophysiology underlying the development of factor VIII inhibitors is a T-helper (Th)–cell dependent event that involves antigen-presenting cells and B lymphocytes; why only a fraction of patients experience this adverse effect of factor therapy is not known. Patients with mild/moderate hemophilia have a lower risk for inhibitor development than those with severe hemophilia A. The estimated prevalence of inhibitors ranges from 3% to 13% in mild to moderate disease,^44–46 and up to 36% in severe hemophilia A.^47,48 Usually the presence of an inhibitor in patients with mild/moderate hemophilia is suggested by a change in bleeding pattern: patients who previously used to bleed only after trauma or surgery suddenly start to experience severe spontaneous bleeding. This change in bleeding pattern is explained by cross-reactivity of the inhibitor with the mutated factor VIII of the patient, resulting in a residual factor level of < 0.01 IU/dL.^49–51 Occasionally, there is no change in the residual factor VIII level but an inhibitor is detected in the Bethesda assay and/or there is lack of efficacy of factor VIII trans-fusions.^51–53

Genetic factors. Data indicate that the risk of developing neutralizing antibodies is to a large extent determined by patient-related genetic factors.^54,55 The immune response to factor VIII is similar in up to 80% of family members, significantly higher than expected compared with data from unrelated subjects. In a meta-analysis of patients with severe hemophilia A, the inhibitor incidence was twice as high in African American patients as compared with white patients.⁵⁶ One study showed that patients of Hispanic ancestry with severe hemophilia A have a higher prevalence of neutralizing inhibitors than non-Hispanic white patients.⁵⁷

Type of causative mutation. In severe hemophilia A, the risk of inhibitor formation is associated with the type of mutation. More disruptive mutations in the factor VIII gene, such as the intron 22 inversion, large gene deletions, and stop codons are associated with an approximately 35% risk of inhibitor formation, compared with only about 5% in those with missense mutation and small deletions.⁵⁸ Persons with mutations involving large gene deletions, nonsense mutations, and intrachromosomal aberrations are usually at higher risk for the development of inhibitors than persons with missense mutations, small deletions/insertions, and splice site mutations.^59,60 A relatively high risk is also encountered in patients with splicing errors and frame-shift mutations.⁶¹

Major histocompatibility complex. The HLA class I alleles A3, B7, and C7, as well as the class II alleles DQA0102, DQB0602, and DR15 have all been associated with a slightly higher risk for inhibitor development in unrelated patients, whereas the HLA C2, DQA0103, DQB0603, and DR13 alleles might be protective.^62,63

Immune-regulatory molecules. In the Malmö International Brother Study, polymorphic sites in the genes coding for interleukin 10 (IL-10), tumor necrosis factor-α, and cytotoxic T lymphocyte–associated protein 4 were all associated with the risk of developing inhibitors.^64–66 In this study, a 134 bp–long variant of a CAA microsatellite in the promoter region (IL-10.G) was identified in 26.8% of patients with hemophilia A. Thirty-two of these patients (72.7%) developed inhibitors as compared with 37.5% of those without the allele.⁶⁵

Intensive exposure to factor VIII. Inhibitors in mild/moderate hemophilia seem to occur more commonly later in life, and an episode of intensive treatment with factor VIII concentrate has been reported to precede detection in most reported cases. In the series reported by Hay et al,⁶⁷ 16 out of 26 inhibitors were detected after such intensive replacement therapy, and no particular concentrate was implicated.

INHIBITORS TO FACTOR IX

Factor IX inhibitors are relatively uncommon, occurring in only 1% to 3% of persons with hemophilia B. This is in striking contrast to hemophilia A, where approximately 30% of patients develop inhibitors. The majority of patients with hemophilia B who develop inhibitors have severe hemophilia B.

Risk Factors

Certain mutations in the factor IX gene are associated with an increased incidence of inhibitor development. Large deletions and frame-shift mutations leading to the loss of coding information are much more likely to be associated with inhibitor development. Large deletions account for only 1% to 3% of all hemophilia B patients, but account for 50% of inhibitor patients.⁶⁸ Patients with hemophilia B who develop inhibitors are at risk for developing anaphylactic reactions to factor IX–containing products. Anaphylaxis occurred more frequently in families with null mutations (large deletions, frame-shift mutations, or nonsense mutations) than in those with missense mutations.⁶⁹ With hemophilia A, approximately 40% to 50% of black individuals develop inhibitors, but no such association has been found in hemophilia B. Individuals who develop an inhibitor to factor IX do so relatively early in life (within the first 4 to 5 years), after a median of 9 to 11 exposure days to any factor IX–containing products. Because of the severity of a potential anaphylactic reaction occurring early in life after very few exposures to factor IX, all infants and small children with severe hemophilia B should be closely followed over their first 10 infusions with any factor IX–containing products in a facility equipped to treat anaphylactic shock.^70–72 A comparison of inhibitors in hemophilia A and B is shown in Table 3.

TREATMENT OF ACUTE BLEEDS IN PATIENTS WITH FACTOR VIII INHIBITORS

The available therapeutic agents for treatment of acute hemorrhage in children with hemophilia A with an inhibitor include high-dose recombinant or plasma-derived factor VIII concentrate, activated prothrombin complex concentrates (aPCCs), and recombinant activated factor VII (rFVIIa). In addition, antifibrinolytics may be used as an adjunct therapy.

Patient response to each treatment varies widely, with some patients responding well to one treatment and less well to another. Neither the patient's history nor standard lab tests can assist in making the best choice for the patient. A personalized approach to factor selection is used, and the dosing of that particular agent is often determined primarily by clinical assessment. Inhibitors are quantitated using the Bethesda inhibitor assay and clinically are classified as low- and high-responding inhibitors (Table 4). Inhibitor screening should be done prior to invasive procedures and periodically during the first 50 days of treatment since the risk for inhibitor development is highest during this period.

Low-Responding Inhibitors

A low-responding inhibitor is one in which inhibitor titers are < 5 Bethesda units (BU)/mL; patients with low-responding inhibitors can generally be treated with factor VIII concentrates at higher doses.⁷³ Because the effect of factor VIII inhibitor is usually delayed, the Bethesda titer in plasma is determined after a 2-hour incubation period. As a result of this time delay, continuous administration of factor VIII is usually found to be effective.⁷⁴ For a serious limb- or life-threatening bleeding episode, a bolus infusion of 100 IU of factor VIII per kg of body weight is administered, and the level is maintained by treatment at a rate of 20 IU/kg/hr. An assay for factor VIII should be performed 1 hour after the bolus infusion and at least daily thereafter. As the antibody titer drops, the daily level of factor VIII may rise and thus downward adjustment of the continuous infusion rate may be required. For routine joint and muscle hemorrhage, patients can usually be managed with infusions at twice the usual dosage. Routine inhibitor assays should be performed after exposure to factor VIII to determine whether an anamnestic response has occurred.

High-Responding Inhibitors

Most clinicians caring for patients with limb- or life-threatening bleeding episodes prefer to use products for which therapeutic levels can be monitored. As described earlier, continuous admin-istration of factor VIII is often effective because of the time delay in inhibition by the antibody. An initial dose of 100 to 200 IU/kg can be administered, and factor VIII levels can be determined 1 hour after initiation of continuous infusion at a rate of 20 to 40 IU/kg/hr. If a factor VIII level cannot be obtained (ie, patients with inhibitor titers > 5 to 10 BU/mL), alternative approaches include the bypassing agents aPCC and rFVIIa.

First used in the 1970s, aPCCs represented a significant improvement in the management in patients with hemophilia with inhibitors. They contain multiple activated serine protease molecules; activated factor X and prothrombin are the main active components in FEIBA (factor eight inhibitor bypassing activity), the most commonly used aPCC in the United States. FEIBA is a pooled plasma product that contains activated factors II, VII, IX and X, and has a duration of action of about 6 to 12 hours. For treatment of acute bleeds, the recommended dose of FEIBA is 50 to 100 IU/kg infused every 8 to 12 hours (maximum daily dose of 200 IU/kg). There is a risk of thrombosis/disseminated intravascular coagulation (DIC) with very large doses given frequently (> 200 IU/kg/day).

rFVIIa directly activates factor X and increases thrombin production on the surface of activated platelets in the absence of factor VIII or factor IX. Standard dosing of rFVIIa is 90 to 120 µg/kg, and many hemophilia treatment centers use higher doses (270 µg/kg/dose), especially in children and young adults. The half-life is about 1.5 to 3 hours, and therefore frequent administration (every 2–6 hours) is required. In one study that assessed the safety and efficacy of fixed-dose rFVIIa in the home setting, hemostasis was achieved in 566 (92%) of evaluable bleeding episodes, and following administration of the additional maintenance dose, hemostasis was maintained in 95% of successfully treated cases.⁷⁵ As with aPCCs, there is no standardized quantitative laboratory test for measuring the effectiveness of rFVIIa therapy.

All currently used bypassing agents are associated with a risk of thrombotic complications including thromboembolism, DIC, and myocardial infarction. These complications are very rare in patients with hemophilia, however. In general, bypassing agents work for most bleeds and for most patients, but are not as predictable as factor replacement therapy and cannot be monitored by laboratory assays.

TREATMENT OF ACUTE BLEEDS IN PATIENTS WITH FACTOR IX INHIBITORS

rFVIIa and FEIBA are the mainstays of treatment of bleeding episodes in individuals with hemophilia B complicated by an inhibitor to factor IX. Treatment of hemorrhagic episodes in these patients depends on the type of bleeding episode experienced, the inhibitor classification (high- versus low-responding [Table 4]), and the history and severity of infusion reactions. Patients with low-responding inhibitors who have not experienced infusion reactions may be treated with doses of factor IX concentrate calculated to overcome the inhibitor titer and achieve a hemostatic level. In patients with high-responding inhibitors, the use of factor IX concentrates is impractical because of the inhibitor titer or the anamnestic response. Regardless of inhibitor titer, in patients with a history of an anaphylactic event, factor IX usage is contraindicated.

The most commonly used therapy for hemostatic control in patients with high-responding inhibitors with factor IX deficiency and a history of infusion reaction is rFVIIa; the standard dosing regimen is 90 to 120 µg/kg/dose administered every 2 to 3 hours, with a maximum dose of 270 µg/kg/dose. aPCCs, which contain factor IX, can be utilized if the patient has not experienced prior infusion reactions. Repeated exposures to products containing factor IX may stimulate the inhibitor titer and prevent its natural decline over time. This can pose a problem in cases of life- or limb-threatening hemorrhage unresponsive to rFVIIa as these patients will not have factor IX available as an effective mode of therapy. The dosing of FEIBA ranges from 50 to 100 IU/kg every 12 hours, with daily dosing not to exceed 200 IU/kg.

IMMUNE TOLERANCE INDUCTION

Because of the associated inhibitor-related morbidity resulting from limited treatment options, antibody eradication is the ultimate goal in inhibitor management. The only proven strategy for achieving antigen-specific tolerance to factor VIII or factor IX is immune tolerance induction (ITI) therapy. Successful ITI in hemophilia A is currently defined as both an undetectable inhibitor titer (< 0.6 BU), and normalized factor VIII pharmacokinetics, which in turn is defined as plasma factor VIII recovery > 66% of expected and a half-life > 6 hours, determined following a 72-hour factor VIII exposure-free period (Consensus Proceedings from the Second International Conference on Immune Tolerance Therapy, Bonn, Germany, 1997 [unpublished]). Once successful immune tolerance is achieved, long-term prophylaxis is commonly instituted. Using conclusions drawn from international consensus criteria and analysis of the International Immune Tolerance Registry, the I-ITI study has defined ITI failure by the presence of either of 2 criteria:

1. Failure to attain the definition of success within 33 months of uninterrupted ITI;

2. Failure to demonstrate a progressive 20% reduction in inhibitor titer over each 6-month period of uninterrupted ITI, beginning 3 months after initiation to allow for expected anamnesis.^76–78

This definition implies a minimum ITI trial period of 9 months before failure is declared.

The European Hemophilia Standardization Board (EHSB), the International Consensus Panel (ICP), and the United Kingdom Hemophilia Center Doctors’ Organization (UKHCDO) have agreed that it is preferable to initiate ITI at a titer of < 10 BU/mL, unless, per the ICP, the titer does not decline over a period of 1 to 2 years and/or inhibitor development is associated with severe or life-threatening bleeding. The ICP noted that for “poor-risk” ITI patients (defined by a historical titer of > 200 BU/mL and/or a pre-ITI inhibitor titer of > 10 BU/mL and/or an interval of > 5 years since inhibitor diagnosis), published efficacy data are limited to dosing regimens > 200 IU/kg/day. The groups all independently concluded that ITI has been successfully performed using recombinant and plasma-derived factor VIII replacement therapy (usually the product on which they developed the inhibitor), and that there are no data to support the superiority of any single product type.^79–81 However, both EHSB and ICP have suggested that VWF-containing concentrates be considered for patients who fail ITI using high-purity factor VIII.^79,80

The recommendations from US guidelines for ITI in patients with hemophilia A and inhibitors are listed in Table 5.⁸²

Before the advent of factor products for the treatment of hemophilia, hemarthrosis was one of the leading causes of morbidity. Today, the routine use of prophylactic treatment has resulted in a significant improvement in the lifestyle, quality of life, and life expectancy of these patients. However, despite best efforts, some patients will have severe joint destruction as a result of repeated articular bleeding episodes during their early years. This leads to pain and significant functional disability, thus impairing the quality of life. The basic pathology behind hemophilic arthropathy is chronic synovitis.

It is common to observe a pattern of repeated bleeding (chronic hemarthrosis), especially in patients with severe hemophilia, that can lead to chronic synovitis, inflammatory arthritis, and progressive arthropathy. Therefore, the key to preventing hemophilic arthropathy is aggressive management of the initial hemarthrosis. This is generally accomplished with the use of clotting factor replacement, restorative physiotherapy, and close clinical follow-up. If chronic synovitis develops, synovectomy may be considered in order to slow the progression of the hemophilic arthropathy and to prevent the development of major articular surface erosions that can lead to end-stage arthropathy.⁸³ Primary prophylaxis is discussed earlier and is the mainstay of prevention of chronic hemophilic arthropathy.

SYNOVECTOMY

The emergence of chronic hemophilic hemarthrosis is incited by a hypertrophic and highly vascular synovium. Removal of the synovium prevents further joint damage,⁸⁴ and can be accomplished through surgical and nonsurgical procedures.

Surgical excision of the hypertrophic synovium can be performed through open or arthroscopic procedures. The open approach has largely been replaced by arthroscopic synovectomies. Regardless of the approach, these patients need prolonged hospitalization, extensive factor replacement, and exhaustive physiotherapy. Moreover, patients with inhibitors are usually not considered candidates for surgical synovectomy.

Chemical and radioactive agents injected intra-articularly can decrease the volume and activity of the synovial tissue. Due to the minimally invasive nature of these procedures, nonsurgical synovectomies are of special importance for hemophilic patients with inhibitors to clotting factors.

Chemical Synovectomy

Chemical synovectomies, using thiotepa, osmic acid, D-penicillamine and other agents, have been used in the distant past. Rifampicin, which is used an antibiotic, is now the most commonly used chemical for the purpose of synovectomy, and the one that has shown better results in terms of decreasing hemarthrosis.⁸⁵ Each one of the injections should be accompanied by prophylactic administration of clotting factor concentrate. Excellent results (no synovitis and restoration of previous function) have been reported in up to 83% of patients at an average of 2.4 years after the intra-articular injection of rifampicin. As the pathology of the joint becomes more severe, however, the number of injections required to achieve improvement increases. Younger patients and smaller joints benefit more from this procedure.

Radiation Synovectomy

Radiosynovectomy (RS) and radiosynoviorthesis are common terms used to describe the synovial ablation accomplished by intra-articular injection of radioisotopes. Isotopes of gold, yttrium, rhenium, and dysprosium have been used to perform radiation synovectomies in patients with hemophilia. Yttrium-90, a pure beta emitter with adequate particle size and depth penetration, has been used successfully for the treatment of hemophilic synovitis.

The local (growth plate and articular cartilage) and remote effects of radiation are a concern. There have been no reported cases of growth plate disturbance after radiosynovectomy, even after the use of beta emitters such as gold-198.⁸⁶ Articular cartilage is highly resistant to radiation, and although damage is theoretically possible, none has been reported. Progressive degeneration of treated joints does occur, but the rate is slower than that expected without radiosynovectomy. The principal concern is the potential for late, radiation-induced neoplasia. However, the safety of intra-articular radioisotopes is supported by a long-term follow-up study of more than 5000 RS procedures performed for rheumatoid arthritis, which found no reported radiation-induced malignancies.⁸⁷

One review analyzed the safety of RS in pediatric patients with hemophilia to provide a risk-benefit assessment. During knee RS, patients receive a radiation dose of approximately 0.74 mSv, and during elbow and ankle RS, a dose of approximately 0.32 mSv. The radiation dose from natural sources is approximately 2 mSv per year and the recommended limit for patients (apart from natural sources) is 1 mSv per year. The lifetime cancer risk increases about 0.5% per 100 mSv per year. Considering the risks and benefits of RS, the authors recommend that clinicians consider this procedure in children with inhibitors or in patients without inhibitors when bleeding is recurrent and persistent despite aggressive factor replacement.⁸⁸ External-beam radiation has been extensively studied and carries a small risk of osteosarcoma induction.

ACQUIRED INHIBITORS TO FACTOR VIII

Acquired hemophilia (AH) has an estimated prevalence of 1.48 cases per million per year, and a reported mortality between 9% and 22%.^89,90 AH is uncommon in children younger than 16 years (prevalence estimated at 0.045/million/year), and may be underdiagnosed in persons older than age 85 (prevalence estimated at 14.7/million/year).⁸⁹ In the largest published population series, 50% to 60% of diagnosed individuals were previously healthy with no identified underlying disease state.^90–91 Underlying conditions consistently associated with AH include pregnancy, evolving or pre-existing autoimmune or malignant disorders, and rarely medications. Primary among the autoimmune disorders are collagen vascular disorders, including systemic lupus erythematosus, rheumatoid arthritis, myasthenia gravis, multiple sclerosis, and autoimmune hemolytic anemia. Most antibodies are mixtures of polyclonal IgG1 and IgG4 immunoglobulins, with the IgG4 molecules mainly responsible for inhibiting clotting activity. The clinical picture of AH is characterized by acute onset of severe bleeding in individuals who previously had no history of bleeding diathesis. Patients generally present with mucocutaneous bleeding (eg, epistaxis and gastrointestinal bleeding), as well as soft tissue bleeding (eg, extensive ecchymoses and hematomas).

The 2 major goals of treatment of AH are the immediate control of acute and chronic bleeding and the long-term suppression/eradication of the autoantibody inhibitor. For patients with an inhibitor titer < 5 BU/mL, administration of desmopressin and concentrates of human recombinant factor VIII may raise the factor VIII activity levels in plasma. If the inhibitor titer is > 5 BU/mL, or if bleeding persists despite infusions of factor VIII concentrates, then factor VIII bypassing agents, such as aPCCs or rFVIIa, are indicated. Local measures for treatment of mucosal hemorrhage, such as antifibrinolytic agents or topical fibrin glues, are helpful.

The primary aim in long-term management of AH is to eradicate the factor VIII autoantibodies so that further bleeding can be averted. Although in some clinical situations (postpartum women and drug-related AH) factor VIII antibodies may remit spontaneously, most published guidelines and algorithms recommend early initiation of eradication therapy. This is usually achieved through immunosuppressive medications or immunomodulation. Successful immunosuppression regimens in AH have most frequently used corticosteroids as the cornerstone, either as a single agent or in combination with cyclophosphamide. In a prospective randomized trial involving 31 participants treated with prednisone 1 mg/kg/day for 3 weeks, 32% achieved complete remission. In participants with antibody persistence after 3 weeks, switching to oral cyclophosphamide 2 mg/kg/day as second-line therapy appeared more effective than continuing prednisone (complete remission rate 50% versus 42%).⁹²

Other immunosuppressive medications have been employed for eradication of refractory autoantibody inhibitors, including azathioprine, cyclosporine, tacrolimus, mycophenolate motefil, and sirolimus. Controlled studies have not been performed to confirm their comparative safety and efficacy in sufficiently large populations. Anti-CD20 antibody has been used to treat inhibitors in patients with both congenital and acquired hemophilia.^93,94 Other less frequently used treatment options include administration of intravenous immunoglobulins (IVIG) in large doses. IVIG by itself rarely is able to induce a complete remission, but may be useful adjunctive therapy along with immunosuppressants, as part of an ITI regimen, or with extracorporeal plasmapheresis.

INTRODUCTION

Hemophilia A and B are the most common severe inherited bleeding disorders. The incidence of hemophilia is 1 in 5000 live male births, with hemophilia A occurring 4 times more commonly than hemophilia B. The associated decrease in factor VIII in hemophilia A was initially identified in 1947, and the decrease in factor IX associated with hemophilia B was identified 5 years later.^1,2 Both conditions are inherited as X-linked recessive traits. Queen Victoria of Britain, who reigned from 1837 to 1901, was a carrier of hemophilia and had 2 carrier daughters, Alice and Beatrice, and a son with hemophilia, Leopold.³ In 1984 and 1985, the genes for factor VIII and factor IX were cloned, and in 1989 recombinant factor VIII was first used clinically.^4–7

PATHOPHYSIOLOGY

Both factors VIII and IX are crucial for normal thrombin generation in the coagulation pathway. After any injury, the initial hemostatic event is the formation of a platelet plug. Once the platelet plug is formed, subsequent generation of fibrin prevents continued oozing from the affected site. In hemophilia A and B, the propagation phase of coagulation is impaired, and as a result, the formation of clot is delayed and is not robust. Due to the delayed formation of an abnormal clot, patients with hemophilia do not bleed rapidly but rather ooze continuously. Rebleeding is a common occurrence in inadequately treated patients.⁸

GENETICS

The gene for factor VIII (F8) is located in the most distal band (Xq28) of the long arm of the X chromosome. Spanning more than 186 kb, it is one of the largest genes known.^9,10 The gene for factor IX (F9) is located at Xq27.1 and spans 33 kb.⁷ Defects in the F8 gene associated with hemophilia A may be divided into several categories: gross gene rearrangements; insertions or deletions of genetic sequence of a size varying from 1 base pair up to the entire gene; or single DNA base substitutions resulting in either amino acid replacement (missense), premature peptide chain termination (nonsense, or stop mutations), or mRNA splicing defects. All classes of defects can result in severe disease. However, the single most clinically important defect is a gene rearrangement (an inversion) involving F8 intron 22, which results in approximately 50% of all severe hemophilia A cases worldwide.^11,12 In hemophilia B, point mutations are by far the most common type of abnormality. Generally, they are caused by DNA polymerases adding the wrong nucleotide during replication.¹³

HEMOPHILIA IN FEMALES

X-Inactivation (also called Lyonization) is a process that occurs early in embryonic development in female mammals where 1 of the 2 copies of the X chromosome present is inactivated; it is the reason why some female carriers of hemophilia can become symptomatic. Approximately one third of carriers have clotting factor levels of less than 60% of normal and may experience abnormal bleeding.^14,15 In most cases, carriers experience symptoms similar to those seen in men with mild hemophilia, as well as some that are specific to women. Symptomatic carriers and women with hemophilia may bruise more easily; may experience prolonged bleeding after surgery; may experience serious bleeding after trauma; often have heavier and more prolonged bleeding during their periods (menorrhagia) and are more likely to require an iron supplement or to undergo hysterectomy; and are more likely to have postpartum bleeding following childbirth.^14,15

CLINICAL MANIFESTATIONS

Hemorrhage in patients with hemophilia may occur with minimal or unknown trauma. Patients with severe hemophilia (factor level of < 1 IU/dL or < 1% of normal) often experience spontaneous bleeding into joints or muscles. Those with moderate hemophilia (factor level of 1–5 IU/dL or 1%–5% of normal) seldom experience spontaneous hemorrhage and usually have prolonged bleeding with minor trauma or surgery. Patients with mild hemophilia (factor level > 5 IU/dL but less than 40 IU/dL or > 5% but < 40% of normal) experience severe hemorrhage only following moderate to severe trauma or surgery, and rarely experience spontaneous bleeding. Depending on the site, bleeding can be serious (joints; muscles, especially deep compartments [iliopsoas, calf, and forearm]; mucous membranes in the mouth, gums, nose, and genitourinary tract) or life-threatening (intracranial, neck/throat, gastrointestinal). The joints and muscles are the most common sites of bleeding (Table 1).

MUSCULOSKELETAL BLEEDING

The hallmark of hemophilia is deep bleeding into the joints and muscles. Without prophylactic factor treatment, patients with severe hemophilia A or B may have a bleeding episode as often as once or twice a week. Hemarthrosis episodes typically begin when the child reaches the toddler age. One of the first signs of hemarthrosis is a tingling sensation and feeling of warmth which is soon followed by pain and decreased range of motion of the joint as a result of distension of the joint capsule. Prompt, aggressive treatment with factor replacement therapy is the key to prevent further bleeding and minimize potential long-term complications. Severe chronic arthropathy may develop in older children and adults who have not received aggressive treatment (Figure).

Bleeding into the muscle can manifest as a vague feeling of pain on motion. Swelling may not be obvious and the mass may be difficult to palpate, although the circumference of the affected limb will be increased. Among the muscle bleeds, iliopsoas bleed deserves a special mention because of its potential to cause life-threatening hypovolemic shock as large volumes of blood can be lost into the retroperitoneal space. These patients present with vague abdominal pain or upper thigh discomfort. The hip is flexed and outwardly rotated. The diagnosis is confirmed by computed tomography (CT) or ultrasound.

LIFE-THREATENING HEMORRHAGE

Central Nervous System Bleeding

Most central nervous system (CNS) events, which involve bleeding inside the skull or spinal canal, are caused by trauma. CNS hemorrhage is the most common form of severe hemophilic trauma. However, since patients with hemophilia can experience bleeding even weeks after a minor head injury, a history of head trauma may be hard to determine, particularly in children. Spontaneous CNS bleeding in individuals with hemophilia is rare except when there has been a recent antecedent CNS hemorrhage (ie, a recurrent bleed at a previously injured site) or when there is an associated anatomic lesion that predisposes to acute hemorrhage (eg, aneurysm or arteriovenous malformation). Data from the Universal Data Collection Project of the U.S. Centers for Disease Control and Prevention indicates that predisposing risk factors for intracranial hemorrhage include HIV infection, presence of inhibitory antibodies, and age younger than 5 years or older than 51 years.¹⁶ Neonatal intracranial hemorrhage is most commonly due to birth trauma. Difficult vaginal deliveries (often requiring the application of forceps or vacuum extraction) are predisposing factors for intracranial hemorrhage in hemophilic newborns.

The site of intracranial CNS bleeding can be subdural, epidural, or intraparenchymal. Bleeding at any of these sites can cause rapidly deteriorating CNS brain function, associated brain swelling, and, in the most extreme circumstances, herniation of the brainstem and rapid death. If the bleeding is stopped with rapid clotting factor replacement, adverse clinical effects can be avoided. However, with intraparenchymal hemorrhage, even small hemorrhages can induce permanent structural and/or neurologic sequelae (in particular, if the anatomic site of the bleed is essential for routine brain function).¹⁷

Throat and Neck Hemorrhage

An acute neck injury or a retropharyngeal hemorrhage induced by dental or oral surgical instrumentation can lead to a dissecting facial plane hematoma. This in turn can sometimes lead to compression and acute airway compromise. Bleeding from these injuries that is compressing or compromising the airway may require a rapid clinical response.¹⁸ The time from the injury until the trachea is compressed may be long, sometimes many hours. However, once the compression is sufficient to cause difficulty breathing, there may be a short amount of time to stop the bleeding and prevent complete respiratory obstruction.

MUCOCUTANEOUS BLEEDING

One of the common manifestations of hemophilia is oral bleeding. Tooth extraction poses a specific problem, and bleeding following extraction can be the first symptom that leads to the diagnosis of hemophilia. Bleeding after circumcision may also suggest the diagnosis. In 1 study cohort looking at sites of initial bleeding episodes in babies with hemophilia diagnosed before the age of 2 years, bleeding from circumcision and other iatrogenic causes tended to be most common in the neonatal period. Circumcision bleeding events occurred more often in infants with no family history (43%) as compared to those born to known maternal carriers (9.2%) or to mothers with some other family history of hemophilia (14.3%).¹⁹

Gastrointestinal (GI) bleeding occurs occasionally in hemophilia, and a wide spectrum of esophageal and GI bleeding may occur. A review of 41 episodes of GI bleeding in hemophilia patients who presented to 1 institution over 10 years implicated duodenal ulcer (22%), unknown site (22%), and gastritis (14%) as the most common sources.²⁰ Mallory-Weiss syndrome has also been cited as a cause for upper GI bleeding in hemophilia patients.²¹

PRINCIPLES OF TREATMENT

Understanding the pathophysiology of hemophilia as well as the type and severity of hemophilia and the inhibitor status in an individual patient are paramount in the management of a patient with hemophilia. In the past, management mainly focused on the treatment of acute bleeding episodes (Table 2). With data showing the benefit of bleed prevention, the management of hemophilia now focuses on prophylaxis of bleeding episodes, which prevents chronic arthropathy and improves quality of life.

ACUTE BLEEDING EPISODES

Dosing of Factor VIII Products

Dosing for factor VIII concentrate is as follows: 1 IU of factor VIII concentrate per kg will increase the circulating factor VIII level by 2% (ie, patient weight in kg × 50 IU/kg = 100% correction). For example, a 30-kg patient requiring 100% correction of factor VIII needs an infusion of 1500 IU of factor VIII (30 kg × 50 IU/kg).

Dosing of Factor IX Products

Dosing for factor IX concentrate is as follows: 1 IU of factor IX concentrate per kg will increase the circulating factor IX level by 1% (ie, patient weight in kg × 100 IU/kg = 100% correction). For example, a 30-kg patient requiring 100% correction of factor IX needs an infusion of 3000 IU of factor IX (30 kg × 100 IU/kg). Higher doses (120 to 130 IU/kg) of the recombinant factor IX product BeneFIX (Pfizer) may be needed to reach the 100% circulating factor IX level.

ADJUVANT THERAPY

Desmopressin

Desmopressin is a synthetic vasopressin analogue that increases plasma factor VIII and von Willebrand factor (VWF) levels; it is used to prevent and treat bleeding episodes associated with dental and surgical procedures in patients with mild and moderate hemophilia A and von Willebrand disease.²² Desmopressin causes the release of VWF and factor VIII from storage in the Weibel–Palade bodies of the endothelial cells that line the blood vessels. Individual response to desmopressin varies, with factor VIII level increasing between 2 and 15 times baseline level in patients with mild or moderate hemophilia A.²³ It is therefore recommended that patients undergo a therapeutic trial of desmopressin with laboratory measurement of response to factor VIII before it is used for treatment of bleeding episodes or as prophylactic therapy before dental and other surgical procedures. A similar response is generally seen in an individual patient with subsequent doses, and thus the factor VIII level attained after a trial dose can be used to predict the response to future therapy.²⁴

The recommended intravenous dosage of desmopressin is 0.3 µg/kg, administered in 25 to 50 mL of normal saline, over a period of 20 to 30 minutes.²⁵ A concentrated form of desmopressin is available for intranasal administration to treat bleeding disorders. The appropriate dose of concentrated intranasal desmopressin is 150 µg (1 puff) for persons weighing less than 50 kg, and 300 µg (1 puff in each nostril) for persons weighing more than 50 kg.²⁶

Antifibrinolytic Therapy

Antifibrinolytics (both epsilon-aminocaproic acid [EACA] and tranexamic acid) reversibly block the lysine binding sites of plasminogen, preventing its activation to plasmin and thus inhibiting the lysis of polymerized fibrin. EACA is also believed to stabilize the active form of thrombin activatable fibrinolysis inhibitor (TAFIa). It is believed that inactivation of TAFIa is due to conformational rearrangements in the TAFIa molecule; EACA has been shown to slow down spontaneous inactivation of TAFIa, thus curtailing fibrinolysis.²⁷ Although hemostasis is generally achieved with either factor VIII replacement or desmopressin, the risk of recurrent bleeding from oral mucosal surfaces is dramatically reduced with the use of antifibrinolytic agents. These agents are typically contraindicated in patients with hematuria because they can cause a clot to form in the urinary bladder or ureters, leading to obstruction.

EACA is available in intravenous, oral tablet, and elixir formulations; the oral dose is 100 to 200 mg/kg initially (maximum dose, 10 g), followed by 50 to 100 mg/kg per dose every 6 hours (maximum dose, 5 g). Tranexamic acid is available in 650-mg capsules; the dose is 25 mg/kg every 6 to 8 hours.^28,29 To treat spontaneous oral hemorrhage or to prevent bleeding from dental procedures in patients with hemophilia, either drug is usually begun in conjunction with desmopressin or factor replacement therapy immediately prior to the procedure and continued for up to 7 days or until mucosal healing is complete. Nonsteroidal anti-inflammatory drugs and aspirin affect platelet function and hence are contraindicated in affected individuals.³⁰

PROPHYLAXIS

Patients with mild to moderate hemophilia typically bleed only after trauma, although the trauma needed to induce bleeding may be more minor than that which would cause bleeding in a normal individual. They usually do not suffer from significant morbidities, whereas patients with severe hemophilia often have spontaneous severe muscle and joint bleeds and can develop early crippling hemophilic arthropathy. Hence, routine prophylaxis has now become the standard of care in the United States and other developed countries in the management of patients with severe hemophilia. Prophylactic replacement therapy with cryoprecipitate in boys with severe hemophilia was first used nearly 50 years ago in Sweden³¹ and the Netherlands,³² and was shown to reduce the number and the severity of bleeds.³² Moreover, it was observed that early prophylaxis was more effective in preventing arthropathy compared to starting later in life, and that radiologic joint damage could not be reversed by prophylaxis. Subsequently, primary prophylaxis, defined as the start of regular, continuous treatment before the age of 2 years or after the occurrence of first joint bleed,³³ was recommended and eventually became the standard treatment; it is currently recommended by the World Health Organization/World Federation of Hemophilia (WFH).³⁴

The timing to begin prophylaxis is somewhat controversial, but many authors suggest starting prophylaxis before the first hemarthrosis occurs. Several studies have reported a wide variation in the age at first joint bleed, ranging from 0.2 to 5.8 years, with medians of 1.6 to 1.7 years.^35,36 It has been suggested that arthropathy is best prevented if prophylaxis is started before the second or third joint bleed, but the benefits of starting before the occurrence of first bleed have not been established.^37,38 The Swedish experience provides strong support for early prophylaxis.³⁹ In an analysis of 121 patients with severe hemophilia, age at initiation of prophylaxis was an independent predictor of the development of arthropathy, but dose and interval of prophylaxis at the start of prophylactic treatment were not.³⁹

In the Italian ESPRIT study, it was shown that children randomly assigned to prophylaxis had significantly fewer total bleeding episodes and joint bleeding episodes compared with those assigned to episodic therapy. Eleven of 21 patients (52%) in the prophylaxis group had on average less than 1 hemarthrosis per year, whereas only 4 of 19 patients in the episodic therapy group (21%) had the same low frequency of bleeding (P < 0.05).⁴⁰ In a study of long-term prophylaxis versus on-demand treatment comparing age-matched Danish and Russian patients, the median annual number of joint bleeds in patients on prophylaxis was 1, while patients managed with on-demand treatment experienced a median of 37 joint bleeds. Patients taking prophylaxis also had a statistically significantly better quality of life estimate (P < 0.001) and better functional independence.⁴¹ In another trial, prophylaxis was initiated between the ages of 6 and 30 months based on a history of joint hemorrhage rather than age. Radiologic evidence of preserved joint architecture was found in 93% of participants in the prophylaxis group at 6 years of age. In this group, 18 of 32 (56%) children had 1 or 2 bleeds into one or more index joints before prophylaxis, and 17 (53%) had 1 to 5 hemorrhages into 1 or more index joints during prophylaxis. Prophylaxis was efficacious in decreasing bleeding and joint damage after up to 5 hemarthroses.⁴²

Optimal Prophylactic Regimen

Although the benefits of prophylactic replacement therapy are firmly established, the optimal dose and frequency remain unclear. The half-life of clotting factor concentrates is short: about 8 hours for factor VIII in children, and about 12 hours for factor IX. As a result, prophylactic therapy is most effective when given frequently. The most common factor VIII concentrate dosing regimen for prophylaxis in hemophilia A is 25 to 40 IU/kg 3 times per week; for hemophilia B, a dose of 80 to 100 IU/kg is given twice weekly. This is aimed at a pre-infusion level > 1% to mimic the clinical phenotype of moderate hemophilia.

Recently, the US Food and Drug Administration (FDA) approved the first long-lasting antihemophilic factor (recombinant) Fc fusion protein for use in adults and children with hemophilia A. This medication contains the Fc region of human immunoglobulin G1 (IgG1), which binds to the neonatal Fc receptor (FcRn). FcRn is part of a naturally occurring pathway that delays lysosomal degradation of immunoglobulins by cycling them back into circulation and prolonging their plasma half-life. Dosing for routine prophylaxis is 50 IU/kg every 4 days; it may be adjusted based on patient response, with dosing in the range of 25 to 65 IU/kg at 3- to 5-day intervals. More frequent or higher doses up to 80 IU/kg may be required in children younger than 6 years.⁴³

DEVELOPMENT OF INHIBITORS

FACTOR VIII INHIBITORS

Despite the success in the clinical management of hemophilia A, treated patients remain at risk for developing neutralizing antibodies that inhibit factor VIII activity. An inhibitor is a polyclonal high-affinity IgG that is directed against the factor VIII protein and renders exogenous factor ineffective. IgG4 antibodies are predominant and do not fix complement.

Risk Factors

The pathophysiology underlying the development of factor VIII inhibitors is a T-helper (Th)–cell dependent event that involves antigen-presenting cells and B lymphocytes; why only a fraction of patients experience this adverse effect of factor therapy is not known. Patients with mild/moderate hemophilia have a lower risk for inhibitor development than those with severe hemophilia A. The estimated prevalence of inhibitors ranges from 3% to 13% in mild to moderate disease,^44–46 and up to 36% in severe hemophilia A.^47,48 Usually the presence of an inhibitor in patients with mild/moderate hemophilia is suggested by a change in bleeding pattern: patients who previously used to bleed only after trauma or surgery suddenly start to experience severe spontaneous bleeding. This change in bleeding pattern is explained by cross-reactivity of the inhibitor with the mutated factor VIII of the patient, resulting in a residual factor level of < 0.01 IU/dL.^49–51 Occasionally, there is no change in the residual factor VIII level but an inhibitor is detected in the Bethesda assay and/or there is lack of efficacy of factor VIII trans-fusions.^51–53

Genetic factors. Data indicate that the risk of developing neutralizing antibodies is to a large extent determined by patient-related genetic factors.^54,55 The immune response to factor VIII is similar in up to 80% of family members, significantly higher than expected compared with data from unrelated subjects. In a meta-analysis of patients with severe hemophilia A, the inhibitor incidence was twice as high in African American patients as compared with white patients.⁵⁶ One study showed that patients of Hispanic ancestry with severe hemophilia A have a higher prevalence of neutralizing inhibitors than non-Hispanic white patients.⁵⁷

Type of causative mutation. In severe hemophilia A, the risk of inhibitor formation is associated with the type of mutation. More disruptive mutations in the factor VIII gene, such as the intron 22 inversion, large gene deletions, and stop codons are associated with an approximately 35% risk of inhibitor formation, compared with only about 5% in those with missense mutation and small deletions.⁵⁸ Persons with mutations involving large gene deletions, nonsense mutations, and intrachromosomal aberrations are usually at higher risk for the development of inhibitors than persons with missense mutations, small deletions/insertions, and splice site mutations.^59,60 A relatively high risk is also encountered in patients with splicing errors and frame-shift mutations.⁶¹

Major histocompatibility complex. The HLA class I alleles A3, B7, and C7, as well as the class II alleles DQA0102, DQB0602, and DR15 have all been associated with a slightly higher risk for inhibitor development in unrelated patients, whereas the HLA C2, DQA0103, DQB0603, and DR13 alleles might be protective.^62,63

Immune-regulatory molecules. In the Malmö International Brother Study, polymorphic sites in the genes coding for interleukin 10 (IL-10), tumor necrosis factor-α, and cytotoxic T lymphocyte–associated protein 4 were all associated with the risk of developing inhibitors.^64–66 In this study, a 134 bp–long variant of a CAA microsatellite in the promoter region (IL-10.G) was identified in 26.8% of patients with hemophilia A. Thirty-two of these patients (72.7%) developed inhibitors as compared with 37.5% of those without the allele.⁶⁵

Intensive exposure to factor VIII. Inhibitors in mild/moderate hemophilia seem to occur more commonly later in life, and an episode of intensive treatment with factor VIII concentrate has been reported to precede detection in most reported cases. In the series reported by Hay et al,⁶⁷ 16 out of 26 inhibitors were detected after such intensive replacement therapy, and no particular concentrate was implicated.

INHIBITORS TO FACTOR IX

Factor IX inhibitors are relatively uncommon, occurring in only 1% to 3% of persons with hemophilia B. This is in striking contrast to hemophilia A, where approximately 30% of patients develop inhibitors. The majority of patients with hemophilia B who develop inhibitors have severe hemophilia B.

Risk Factors

Certain mutations in the factor IX gene are associated with an increased incidence of inhibitor development. Large deletions and frame-shift mutations leading to the loss of coding information are much more likely to be associated with inhibitor development. Large deletions account for only 1% to 3% of all hemophilia B patients, but account for 50% of inhibitor patients.⁶⁸ Patients with hemophilia B who develop inhibitors are at risk for developing anaphylactic reactions to factor IX–containing products. Anaphylaxis occurred more frequently in families with null mutations (large deletions, frame-shift mutations, or nonsense mutations) than in those with missense mutations.⁶⁹ With hemophilia A, approximately 40% to 50% of black individuals develop inhibitors, but no such association has been found in hemophilia B. Individuals who develop an inhibitor to factor IX do so relatively early in life (within the first 4 to 5 years), after a median of 9 to 11 exposure days to any factor IX–containing products. Because of the severity of a potential anaphylactic reaction occurring early in life after very few exposures to factor IX, all infants and small children with severe hemophilia B should be closely followed over their first 10 infusions with any factor IX–containing products in a facility equipped to treat anaphylactic shock.^70–72 A comparison of inhibitors in hemophilia A and B is shown in Table 3.

TREATMENT OF ACUTE BLEEDS IN PATIENTS WITH FACTOR VIII INHIBITORS

The available therapeutic agents for treatment of acute hemorrhage in children with hemophilia A with an inhibitor include high-dose recombinant or plasma-derived factor VIII concentrate, activated prothrombin complex concentrates (aPCCs), and recombinant activated factor VII (rFVIIa). In addition, antifibrinolytics may be used as an adjunct therapy.

Patient response to each treatment varies widely, with some patients responding well to one treatment and less well to another. Neither the patient's history nor standard lab tests can assist in making the best choice for the patient. A personalized approach to factor selection is used, and the dosing of that particular agent is often determined primarily by clinical assessment. Inhibitors are quantitated using the Bethesda inhibitor assay and clinically are classified as low- and high-responding inhibitors (Table 4). Inhibitor screening should be done prior to invasive procedures and periodically during the first 50 days of treatment since the risk for inhibitor development is highest during this period.

Low-Responding Inhibitors

A low-responding inhibitor is one in which inhibitor titers are < 5 Bethesda units (BU)/mL; patients with low-responding inhibitors can generally be treated with factor VIII concentrates at higher doses.⁷³ Because the effect of factor VIII inhibitor is usually delayed, the Bethesda titer in plasma is determined after a 2-hour incubation period. As a result of this time delay, continuous administration of factor VIII is usually found to be effective.⁷⁴ For a serious limb- or life-threatening bleeding episode, a bolus infusion of 100 IU of factor VIII per kg of body weight is administered, and the level is maintained by treatment at a rate of 20 IU/kg/hr. An assay for factor VIII should be performed 1 hour after the bolus infusion and at least daily thereafter. As the antibody titer drops, the daily level of factor VIII may rise and thus downward adjustment of the continuous infusion rate may be required. For routine joint and muscle hemorrhage, patients can usually be managed with infusions at twice the usual dosage. Routine inhibitor assays should be performed after exposure to factor VIII to determine whether an anamnestic response has occurred.

High-Responding Inhibitors

Most clinicians caring for patients with limb- or life-threatening bleeding episodes prefer to use products for which therapeutic levels can be monitored. As described earlier, continuous admin-istration of factor VIII is often effective because of the time delay in inhibition by the antibody. An initial dose of 100 to 200 IU/kg can be administered, and factor VIII levels can be determined 1 hour after initiation of continuous infusion at a rate of 20 to 40 IU/kg/hr. If a factor VIII level cannot be obtained (ie, patients with inhibitor titers > 5 to 10 BU/mL), alternative approaches include the bypassing agents aPCC and rFVIIa.

First used in the 1970s, aPCCs represented a significant improvement in the management in patients with hemophilia with inhibitors. They contain multiple activated serine protease molecules; activated factor X and prothrombin are the main active components in FEIBA (factor eight inhibitor bypassing activity), the most commonly used aPCC in the United States. FEIBA is a pooled plasma product that contains activated factors II, VII, IX and X, and has a duration of action of about 6 to 12 hours. For treatment of acute bleeds, the recommended dose of FEIBA is 50 to 100 IU/kg infused every 8 to 12 hours (maximum daily dose of 200 IU/kg). There is a risk of thrombosis/disseminated intravascular coagulation (DIC) with very large doses given frequently (> 200 IU/kg/day).

rFVIIa directly activates factor X and increases thrombin production on the surface of activated platelets in the absence of factor VIII or factor IX. Standard dosing of rFVIIa is 90 to 120 µg/kg, and many hemophilia treatment centers use higher doses (270 µg/kg/dose), especially in children and young adults. The half-life is about 1.5 to 3 hours, and therefore frequent administration (every 2–6 hours) is required. In one study that assessed the safety and efficacy of fixed-dose rFVIIa in the home setting, hemostasis was achieved in 566 (92%) of evaluable bleeding episodes, and following administration of the additional maintenance dose, hemostasis was maintained in 95% of successfully treated cases.⁷⁵ As with aPCCs, there is no standardized quantitative laboratory test for measuring the effectiveness of rFVIIa therapy.

All currently used bypassing agents are associated with a risk of thrombotic complications including thromboembolism, DIC, and myocardial infarction. These complications are very rare in patients with hemophilia, however. In general, bypassing agents work for most bleeds and for most patients, but are not as predictable as factor replacement therapy and cannot be monitored by laboratory assays.

TREATMENT OF ACUTE BLEEDS IN PATIENTS WITH FACTOR IX INHIBITORS

rFVIIa and FEIBA are the mainstays of treatment of bleeding episodes in individuals with hemophilia B complicated by an inhibitor to factor IX. Treatment of hemorrhagic episodes in these patients depends on the type of bleeding episode experienced, the inhibitor classification (high- versus low-responding [Table 4]), and the history and severity of infusion reactions. Patients with low-responding inhibitors who have not experienced infusion reactions may be treated with doses of factor IX concentrate calculated to overcome the inhibitor titer and achieve a hemostatic level. In patients with high-responding inhibitors, the use of factor IX concentrates is impractical because of the inhibitor titer or the anamnestic response. Regardless of inhibitor titer, in patients with a history of an anaphylactic event, factor IX usage is contraindicated.

The most commonly used therapy for hemostatic control in patients with high-responding inhibitors with factor IX deficiency and a history of infusion reaction is rFVIIa; the standard dosing regimen is 90 to 120 µg/kg/dose administered every 2 to 3 hours, with a maximum dose of 270 µg/kg/dose. aPCCs, which contain factor IX, can be utilized if the patient has not experienced prior infusion reactions. Repeated exposures to products containing factor IX may stimulate the inhibitor titer and prevent its natural decline over time. This can pose a problem in cases of life- or limb-threatening hemorrhage unresponsive to rFVIIa as these patients will not have factor IX available as an effective mode of therapy. The dosing of FEIBA ranges from 50 to 100 IU/kg every 12 hours, with daily dosing not to exceed 200 IU/kg.

IMMUNE TOLERANCE INDUCTION

Because of the associated inhibitor-related morbidity resulting from limited treatment options, antibody eradication is the ultimate goal in inhibitor management. The only proven strategy for achieving antigen-specific tolerance to factor VIII or factor IX is immune tolerance induction (ITI) therapy. Successful ITI in hemophilia A is currently defined as both an undetectable inhibitor titer (< 0.6 BU), and normalized factor VIII pharmacokinetics, which in turn is defined as plasma factor VIII recovery > 66% of expected and a half-life > 6 hours, determined following a 72-hour factor VIII exposure-free period (Consensus Proceedings from the Second International Conference on Immune Tolerance Therapy, Bonn, Germany, 1997 [unpublished]). Once successful immune tolerance is achieved, long-term prophylaxis is commonly instituted. Using conclusions drawn from international consensus criteria and analysis of the International Immune Tolerance Registry, the I-ITI study has defined ITI failure by the presence of either of 2 criteria:

1. Failure to attain the definition of success within 33 months of uninterrupted ITI;

2. Failure to demonstrate a progressive 20% reduction in inhibitor titer over each 6-month period of uninterrupted ITI, beginning 3 months after initiation to allow for expected anamnesis.^76–78

This definition implies a minimum ITI trial period of 9 months before failure is declared.

The European Hemophilia Standardization Board (EHSB), the International Consensus Panel (ICP), and the United Kingdom Hemophilia Center Doctors’ Organization (UKHCDO) have agreed that it is preferable to initiate ITI at a titer of < 10 BU/mL, unless, per the ICP, the titer does not decline over a period of 1 to 2 years and/or inhibitor development is associated with severe or life-threatening bleeding. The ICP noted that for “poor-risk” ITI patients (defined by a historical titer of > 200 BU/mL and/or a pre-ITI inhibitor titer of > 10 BU/mL and/or an interval of > 5 years since inhibitor diagnosis), published efficacy data are limited to dosing regimens > 200 IU/kg/day. The groups all independently concluded that ITI has been successfully performed using recombinant and plasma-derived factor VIII replacement therapy (usually the product on which they developed the inhibitor), and that there are no data to support the superiority of any single product type.^79–81 However, both EHSB and ICP have suggested that VWF-containing concentrates be considered for patients who fail ITI using high-purity factor VIII.^79,80

The recommendations from US guidelines for ITI in patients with hemophilia A and inhibitors are listed in Table 5.⁸²

Before the advent of factor products for the treatment of hemophilia, hemarthrosis was one of the leading causes of morbidity. Today, the routine use of prophylactic treatment has resulted in a significant improvement in the lifestyle, quality of life, and life expectancy of these patients. However, despite best efforts, some patients will have severe joint destruction as a result of repeated articular bleeding episodes during their early years. This leads to pain and significant functional disability, thus impairing the quality of life. The basic pathology behind hemophilic arthropathy is chronic synovitis.

It is common to observe a pattern of repeated bleeding (chronic hemarthrosis), especially in patients with severe hemophilia, that can lead to chronic synovitis, inflammatory arthritis, and progressive arthropathy. Therefore, the key to preventing hemophilic arthropathy is aggressive management of the initial hemarthrosis. This is generally accomplished with the use of clotting factor replacement, restorative physiotherapy, and close clinical follow-up. If chronic synovitis develops, synovectomy may be considered in order to slow the progression of the hemophilic arthropathy and to prevent the development of major articular surface erosions that can lead to end-stage arthropathy.⁸³ Primary prophylaxis is discussed earlier and is the mainstay of prevention of chronic hemophilic arthropathy.

SYNOVECTOMY

The emergence of chronic hemophilic hemarthrosis is incited by a hypertrophic and highly vascular synovium. Removal of the synovium prevents further joint damage,⁸⁴ and can be accomplished through surgical and nonsurgical procedures.

Surgical excision of the hypertrophic synovium can be performed through open or arthroscopic procedures. The open approach has largely been replaced by arthroscopic synovectomies. Regardless of the approach, these patients need prolonged hospitalization, extensive factor replacement, and exhaustive physiotherapy. Moreover, patients with inhibitors are usually not considered candidates for surgical synovectomy.

Chemical and radioactive agents injected intra-articularly can decrease the volume and activity of the synovial tissue. Due to the minimally invasive nature of these procedures, nonsurgical synovectomies are of special importance for hemophilic patients with inhibitors to clotting factors.

Chemical Synovectomy

Chemical synovectomies, using thiotepa, osmic acid, D-penicillamine and other agents, have been used in the distant past. Rifampicin, which is used an antibiotic, is now the most commonly used chemical for the purpose of synovectomy, and the one that has shown better results in terms of decreasing hemarthrosis.⁸⁵ Each one of the injections should be accompanied by prophylactic administration of clotting factor concentrate. Excellent results (no synovitis and restoration of previous function) have been reported in up to 83% of patients at an average of 2.4 years after the intra-articular injection of rifampicin. As the pathology of the joint becomes more severe, however, the number of injections required to achieve improvement increases. Younger patients and smaller joints benefit more from this procedure.

Radiation Synovectomy

Radiosynovectomy (RS) and radiosynoviorthesis are common terms used to describe the synovial ablation accomplished by intra-articular injection of radioisotopes. Isotopes of gold, yttrium, rhenium, and dysprosium have been used to perform radiation synovectomies in patients with hemophilia. Yttrium-90, a pure beta emitter with adequate particle size and depth penetration, has been used successfully for the treatment of hemophilic synovitis.

The local (growth plate and articular cartilage) and remote effects of radiation are a concern. There have been no reported cases of growth plate disturbance after radiosynovectomy, even after the use of beta emitters such as gold-198.⁸⁶ Articular cartilage is highly resistant to radiation, and although damage is theoretically possible, none has been reported. Progressive degeneration of treated joints does occur, but the rate is slower than that expected without radiosynovectomy. The principal concern is the potential for late, radiation-induced neoplasia. However, the safety of intra-articular radioisotopes is supported by a long-term follow-up study of more than 5000 RS procedures performed for rheumatoid arthritis, which found no reported radiation-induced malignancies.⁸⁷

One review analyzed the safety of RS in pediatric patients with hemophilia to provide a risk-benefit assessment. During knee RS, patients receive a radiation dose of approximately 0.74 mSv, and during elbow and ankle RS, a dose of approximately 0.32 mSv. The radiation dose from natural sources is approximately 2 mSv per year and the recommended limit for patients (apart from natural sources) is 1 mSv per year. The lifetime cancer risk increases about 0.5% per 100 mSv per year. Considering the risks and benefits of RS, the authors recommend that clinicians consider this procedure in children with inhibitors or in patients without inhibitors when bleeding is recurrent and persistent despite aggressive factor replacement.⁸⁸ External-beam radiation has been extensively studied and carries a small risk of osteosarcoma induction.

ACQUIRED INHIBITORS TO FACTOR VIII

Acquired hemophilia (AH) has an estimated prevalence of 1.48 cases per million per year, and a reported mortality between 9% and 22%.^89,90 AH is uncommon in children younger than 16 years (prevalence estimated at 0.045/million/year), and may be underdiagnosed in persons older than age 85 (prevalence estimated at 14.7/million/year).⁸⁹ In the largest published population series, 50% to 60% of diagnosed individuals were previously healthy with no identified underlying disease state.^90–91 Underlying conditions consistently associated with AH include pregnancy, evolving or pre-existing autoimmune or malignant disorders, and rarely medications. Primary among the autoimmune disorders are collagen vascular disorders, including systemic lupus erythematosus, rheumatoid arthritis, myasthenia gravis, multiple sclerosis, and autoimmune hemolytic anemia. Most antibodies are mixtures of polyclonal IgG1 and IgG4 immunoglobulins, with the IgG4 molecules mainly responsible for inhibiting clotting activity. The clinical picture of AH is characterized by acute onset of severe bleeding in individuals who previously had no history of bleeding diathesis. Patients generally present with mucocutaneous bleeding (eg, epistaxis and gastrointestinal bleeding), as well as soft tissue bleeding (eg, extensive ecchymoses and hematomas).

The 2 major goals of treatment of AH are the immediate control of acute and chronic bleeding and the long-term suppression/eradication of the autoantibody inhibitor. For patients with an inhibitor titer < 5 BU/mL, administration of desmopressin and concentrates of human recombinant factor VIII may raise the factor VIII activity levels in plasma. If the inhibitor titer is > 5 BU/mL, or if bleeding persists despite infusions of factor VIII concentrates, then factor VIII bypassing agents, such as aPCCs or rFVIIa, are indicated. Local measures for treatment of mucosal hemorrhage, such as antifibrinolytic agents or topical fibrin glues, are helpful.

The primary aim in long-term management of AH is to eradicate the factor VIII autoantibodies so that further bleeding can be averted. Although in some clinical situations (postpartum women and drug-related AH) factor VIII antibodies may remit spontaneously, most published guidelines and algorithms recommend early initiation of eradication therapy. This is usually achieved through immunosuppressive medications or immunomodulation. Successful immunosuppression regimens in AH have most frequently used corticosteroids as the cornerstone, either as a single agent or in combination with cyclophosphamide. In a prospective randomized trial involving 31 participants treated with prednisone 1 mg/kg/day for 3 weeks, 32% achieved complete remission. In participants with antibody persistence after 3 weeks, switching to oral cyclophosphamide 2 mg/kg/day as second-line therapy appeared more effective than continuing prednisone (complete remission rate 50% versus 42%).⁹²

Other immunosuppressive medications have been employed for eradication of refractory autoantibody inhibitors, including azathioprine, cyclosporine, tacrolimus, mycophenolate motefil, and sirolimus. Controlled studies have not been performed to confirm their comparative safety and efficacy in sufficiently large populations. Anti-CD20 antibody has been used to treat inhibitors in patients with both congenital and acquired hemophilia.^93,94 Other less frequently used treatment options include administration of intravenous immunoglobulins (IVIG) in large doses. IVIG by itself rarely is able to induce a complete remission, but may be useful adjunctive therapy along with immunosuppressants, as part of an ITI regimen, or with extracorporeal plasmapheresis.

INTRODUCTION

Mantle cell lymphoma (MCL) is an uncommon, distinct clinical subtype of non-Hodgkin lymphoma (NHL) that comprises approximately 8% of all lymphoma diagnoses in the United States and Europe.^1,2 Considered incurable, MCL often presents in advanced stages, particularly with involvement of the lymph nodes, spleen, bone marrow, and gastrointestinal tract in the form of lymphomatous polyps. MCL disproportionately affects males, and incidence rises with age, with a median age at diagnosis of 68 years.² Historically, the prognosis of patients with MCL has been among the poorest among B-cell lymphoma patients, with a median overall survival (OS) of 3 to 5 years, and time to treatment failure (TTF) of 18 to 24 months, although this is improving in the modern era.³ Less frequently, patients with MCL display isolated bone marrow, peripheral blood, and splenic involvement. These cases tend to behave more indolently with longer survival.^4,5 Recent advances in therapy have dramatically impacted treatment alternatives and outcomes for MCL. As such, the therapeutic and prognostic landscape of MCL is evolving rapidly.

PATHOGENESIS

The histologic diagnosis of MCL by morphology alone is often challenging. Accurate diagnosis relies on immunohistochemical staining for the purposes of immunophenotyping.⁶ MCL typically expresses B-cell markers CD5 and CD20, and lacks both CD10 and CD23. The genetic hallmark of MCL is the t(11;14) (q13;q32) chromosomal translocation leading to upregulation of the cyclin D1 protein, a critical regulator of the G1 phase of the cell cycle. Specifically, the t(11;14) translocation, present in virtually all cases of MCL, juxtaposes the proto-oncogene CCND1 to the immunoglobulin heavy chain gene.⁷ Consequently, cyclin D1, normally not expressed in B lymphocytes, becomes constitutively overexpressed. This alteration is thought to facilitate the deregulation of the cell cycle at the G1-S phase transition.⁸

Gene expression profiling studies have underscored the importance of cell cycle deregulation in MCL, and high proliferation is associated with a worse prognosis.⁹ More than 50% of the genes associated with poor outcomes were derived from the “proliferation signature” that was more highly expressed in dividing cells. In the seminal Rosenwald study, a gene expression–based outcome model was constructed in which the proliferation signature average represents a linear variable that assigns a discrete probability of survival to an individual patient.⁹ The proliferative index, or proliferative signature, of MCL can be estimated by the percentage of Ki-67–positive cells present in the tumor through immunohistochemistry. This is often used as a marker of poor outcomes, and as a surrogate for the proliferative signature in MCL that can be incorporated into clinical practice (as opposed to gene expression profiling). Statistically significant differences in OS have emerged between groups of MCL patients with Ki-67–positive cells comprising less than 30% of their tumor sample (favorable) and those with Ki-67–positive cells comprising 30% or greater (unfavorable).¹⁰

Recent data has also identified the importance of the transcription factor SOX 11 (SRY-related HMG-box), which regulates multiple cellular transcriptional events, including cell proliferation and differentiation, apoptosis, and angiogenesis.¹¹ MCL expressing SOX 11 behaves more aggressively than MCL variants lacking SOX 11 expression, and tends to accumulate more genetic alterations.¹² Moreover, lack of SOX 11 expression characterizes a subset of MCL that does not carry the t(11;14) translocation.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 62-year-old man with a history of diabetes mellitus and hypertension presents with cervical lymphadenopathy, fatigue, and early satiety over the past several months. He is otherwise in good health. His Eastern Cooperative Oncology Group (ECOG) performance status is 1. On physical examination, 3-cm lymphadenopathy in the bilateral cervical chain is noted. Bilateral axillary lymph nodes measure 2 to 4 cm. His spleen is enlarged and is palpable at approximately 5 cm below the costal margin. A complete blood count reveals a total white blood cell (WBC) count of 14,000 cells/μL, with 68% lymphocytes and a normal distribution of neutrophils. Hemoglobin is 11 g/dL, and platelet count is 112,000/μL. The lactate dehydrogenase (LDH) level is 322 U/L (upper limit of normal: 225 U/L).

• How is MCL diagnosed?

Diagnosis of MCL requires review by expert hematopathologists.¹³ Whenever possible, an excisional biopsy should be performed for the adequate characterization of lymph node architecture and evaluation by immunohistochemistry. Aside from the characteristic expression of CD5 and CD20 and absence of CD23, MCL should express cyclin D1, which reflects t(11;14). If cyclin D1 is inconclusive or unavailable, fluorescent in situ hybridization (FISH) for t(11;14) should be performed.⁸ Patients often have circulating malignant lymphocytes, or leukemic phase MCL. Flow cytometry of the peripheral blood can detect traditional surface markers, and FISH can also be performed on circulating abnormal lymphocytes.

RISK STRATIFICATION

At diagnosis, patients should undergo risk stratification in order to understand prognosis and possibly guide treatment. In MCL, the MCL international prognostic index (MIPI) is used. The MIPI is a prognostic tool developed exclusively for patients with MCL using data from 455 patients with advanced-stage MCL treated within 3 European clinical trials.¹⁶ The MIPI classified patients into risk groups based on age, ECOG performance status, LDH level, and WBC count. Patients were categorized into low-risk (44% of patients, median OS not reached), intermediate-risk (35%, median OS 51 months), and high-risk groups (21%, median OS 29 months). This is done through a logarithmic calculation, which can be accessed through online calculators (a prototype example can be found at www.qxmd.com/calculate-online/hematology/prognosis-mantle-cell-lymphoma-mipi). Cell proliferation using the Ki-67 index was evaluated in an exploratory analysis (the biologic [“B”] MIPI), and also demonstrated strong prognostic relevance.¹⁶ Currently, treatment of MCL patients is not stratified by MIPI outside of a clinical trial, but this useful tool assists in assessing patient prognosis and has been validated for use with both conventional chemoimmunotherapy and in the setting of autologous stem cell transplant (autoSCT).^16,17 At this point in time, the MIPI score is not used to stratify treatment, although some clinical trials are incorporating the use of the MIPI score at diagnosis. Nonetheless, given its prognostic importance, the MIPI should be performed for all MCL patients undergoing staging and evaluation for treatment to establish disease risk.

As noted, the proliferative signature, represented by the Ki-67 protein, is also highly prognostic in MCL. Ki-67 is expressed in the late G1, S, G2, and M phases of the cell cycle. The Ki-67 index is defined by the hematopathologist as the percentage of lymphoma cells staining positive for Ki-67 protein, based on the number of cells per high-power field. There is significant interobserver variability in this process, which can be minimized by assessing Ki-67 quantitatively using computer software. The prognostic significance of Ki-67 at diagnosis was established in large studies of MCL patient cohorts, with survival differing by up to 3 years.^18,19 Determann et al demonstrated the utility of the proliferative index in patients with MCL treated with standard chemoimmunotherapy.¹⁰ In this study, 249 patients with advanced-stage MCL treated within randomized trials conducted by the European MCL Network were analyzed. The Ki-67 index was found to be extremely prognostic of OS, independent of other clinical risk factors, including the MIPI score. As a continuous variable, Ki-67 indices of greater than 10% correlated with poor outcomes. The Ki-67 index has also been confirmed as prognostic in relapsed MCL.²⁰ It is important to note that, as a unique feature, the Ki-67 index has remained an independent prognostic factor, even when incorporated into the “B” MIPI.

TREATMENT

CASE CONTINUED

The patient undergoes an excisional biopsy of a cervical lymph node, which demonstrates an abnormal proliferation of small-medium–sized lymphocytes with slightly irregular nuclear contours. Immunohistochemistry shows that the abnormal lymphocytes are positive for CD20 and CD5, negative for CD10 and CD23, and diffusely positive for cyclin D1, consistent with a diagnosis of MCL. The proliferative index, as measured by the Ki-67 immunostain, is 40%. A bone marrow aspirate and biopsy are then obtained, which show a clonal population of B lymphocytes expressing the same immunophenotype as the lymph node (positive for CD20 and CD5, negative for CD10 and CD23, cyclin D1 positive). A CT scan of the neck, chest, abdomen, and pelvis with contrast is obtained, along with a PET scan. These studies identify extensive hypermetabolic lymphadenopathy in the bilateral cervical chains, supraclavicular areas, mediastinum, and hilum. Mesenteric lymph nodes are also enlarged and hypermetabolic, as are retroperitoneal lymph nodes. The spleen is noted to be enlarged with multiple hypermetabolic lesions. Based on the presence of extensive lymphadenopathy as well as bone marrow involvement, the patient is diagnosed with stage IV MCL. He undergoes risk-stratification with the MIPI. His MIPI score is 6.3, high risk.

FRONTLINE THERAPY

Role of Watchful Waiting

A small proportion of MCL patients have indolent disease that can be observed. This population is more likely to have leukemic-phase MCL with circulating lymphocytes, splenomegaly, and bone marrow involvement and absent or minimal lymphadenopathy.^4,5 A retrospective study of 97 patients established that deferment of initial therapy in MCL is acceptable in some patients.⁵ In this study, approximately one third of patients with MCL were observed for more than 3 months before initiating systemic therapy, and the median time to treatment for the observation group was 12 months. Most patients undergoing observation had a low-risk MIPI. Patients were not harmed by observation, as no OS differences were observed among groups. This study underscores that deferred treatment can be an acceptable alternative in selected MCL patients for a short period of time. In practice, the type of patient who would be appropriate for this approach is someone who is frail, elderly, and with multiple comorbidities. Additionally, expectant observation could be considered for patients with limited-stage or low-volume MCL, low Ki-67 index, and low-risk MIPI scores.

Approach to Therapy

Treatment of MCL is generally approached by evaluating patient age and fitness for treatment. While there is no accepted standard, for younger patients healthy enough to tolerate aggressive approaches, treatment often involves an intensive cytarabine-containing regimen, which is consolidated with an autoSCT. This approach results in the longest remission duration, with some series suggesting a plateau in survival after 5 years, with no relapses.²¹ Nonintensive conventional chemotherapy alone is often reserved for the frailer or older patient. Given that remission durations with chemotherapy alone in MCL are short, goals of treatment focus on maximizing benefit and remission duration and minimizing risk of toxicity.

Standard Chemotherapy: Elderly and/or Frail Patients

Conventional chemotherapy alone for the treatment of MCL results in a 70% to 85% overall response rate (ORR) and 7% to 30% complete response (CR) rate.²² Rituximab, a mouse humanized monoclonal IgG1 anti-CD20 antibody, is used as standard of care in combination with chemotherapy, since its addition has been found to increase response rates and extend both progression-free survival (PFS) and OS compared to chemotherapy alone.^23,24 However, chemoimmunotherapy approaches do not provide long-term control of MCL and are considered noncurative. Various regimens have been studied and include anthracycline-containing regimens such as R-CHOP (rituximab with cyclophosphamide, doxorubicin, vincristine, prednisone),²² combination chemotherapy with antimetabolites such as R-hyper-CVAD (hyper-fractionated rituximab with cyclophosphamide, vincristine, doxorubicin, dexamethasone, alternating with methotrexate and cytarabine),²⁵ purine analogue–based regimens such as R-FC (rituximab with fludarabine and cyclophosphamide),²⁶ bortezomib-containing regimens,²⁷ and alkylator-based treatment with BR (bendamustine and rituximab) (Table 1).^28,29 Among these, the most commonly used are R-CHOP and BR.

Two large randomized studies compared R-CHOP for 6 cycles to BR for 6 cycles in patients with indolent NHL and MCL. Among MCL patients, BR resulted in superior PFS compared to R-CHOP (69 months versus 26 months) but no benefit in OS.^28,29 The ORR to R-CHOP was approximately 90%, with a PFS of 21 months in the Rummel et al study.²⁹ This study included more than 80 centers in Germany and enrolled 549 patients with MCL, follicular lymphoma, small lymphocytic lymphoma, marginal zone lymphoma, and Waldenström macroglobulinemia. Patients were randomized in a 1:1 fashion. Among these, 46 patients received BR and 48 received R-CHOP (18% for both, respectively). It should be noted that patients in the BR group had significantly less toxicity and experienced fewer side effects than did those in the R-CHOP group. Similarly, BR-treated patients had a lower frequency of hematologic side effects and infections of any grade. However, drug-associated skin reactions and allergies were more common with BR compared to R-CHOP. The study by Flinn and colleagues was an international randomized, noninferiority phase 3 study designed to evaluate the efficacy and safety of BR compared with R-CHOP or R-CVP (rituximab plus cyclophosphamide, vincristine, and prednisone) for treatment-naive patients with MCL or other indolent NHL. The primary endpoint was CR. In this study, BR was found to be noninferior to R-CHOP and R-CVP based on CR rate (31% versus 25%, respectively; P = 0.0225). Response rates in general were high: 97% for BR and 91% for R-CHOP/R-CVP (P = 0.0102). Here, BR-treated patients experienced more nausea, emesis, and drug-induced hypersensitivity compared to the R-CHOP and R-CVP groups.

Bendamustine in combination with other agents has also been studied in the frontline setting. Visco and colleagues evaluated the combination of bendamustine with rituximab and cytarabine (R-BAC) in older patients with MCL (age 65 or older).⁶³ This phase 2, two-stage study enrolled 40 patients and had a dose-finding arm for cytarabine in combination with BR. It permitted relapsed/refractory patients, but 50% had newly diagnosed, previously untreated MCL. The regimen had an impressive ORR of 100%, with CR rates of 95% for previously untreated patients. PFS at 2 years was 95%. R-BAC was well tolerated, with the primary toxicity being reversible myelosuppression.

BR was combined with the proteasome inhibitor bortezomib and dexamethasone in a phase 2 study.⁶⁴ This Lymphoma Study Association (LYSA) study evaluated 76 patients with newly diagnosed MCL older than age 65 years. BR was administered in standard doses (bendamustine 90 mg/m² on days 1 and 2 and rituximab 375 mg/m² IV on day 1) and bortezomib was administered subcutaneously on days 1, 4, 8, and 11, with acyclovir for viral prophylaxis. Patients received 6 cycles. The ORR was 87% and the CR was 60%. Patients experienced toxicity, and not all bortezomib doses were administered due to neurotoxic or hematologic side effects.

A randomized phase 3 study compared R-CHOP to the VR-CAP regimen (R-CHOP regimen but bortezomib replaces vincristine on days 1, 4, 8, 11, at 1.3 mg/m²) in 487 newly diagnosed MCL patients.²⁷ Median PFS was superior in the VR-CAP group compared with R-CHOP (14.4 months versus 24.7 months, respectively). Additionally, rates of CR were superior in the VR-CAP group (53% compared to 42% with R-CHOP). However, there was more hematologic toxicity with VR-CAP. On the basis of these findings, the U.S. Food and Drug Administration approved bortezomib for the frontline treatment of MCL.

Other chemoimmunotherapy combinations containing bortezomib have been studied in frontline MCL treatment, with promising results. These include bortezomib in combination with R-CHOP or modified R-hyper-CVAD, as well as bortezomib in combination with CHOP-like treatments and purine analogues.^27,30–32 The ongoing ECOG 1411 study is currently evaluating bortezomib added to BR for induction therapy of newly diagnosed MCL in a 4-arm randomized trial. Patients receive BR with or without bortezomib during induction and are then randomly assigned to maintenance with either rituximab alone or rituximab with lenalidomide. Other novel combination agents are actively being studied in frontline MCL treatment, including lenalidomide and rituximab and BR with lenalidomide.

Intensification of Therapy and AutoSCT: Fitter and/or Younger Patients

Short response duration has created the need for post-remission therapy in MCL. One approach to improve remission duration in MCL is to intensify induction through the use of cytarabine-containing regimens and/or consolidation with high-dose chemotherapy, typically using BEAM (carmustine, etoposide, cytarabine, melphalan) and autoSCT (Table 2). The cytarabine-containing R-hyper-CVAD regimen, developed at the MD Anderson Cancer Center, resulted in a 97% ORR and an 87% CR rate, with TTF of nearly 5 years. However, nearly one third of patients were unable to complete treatment due to toxicity, and 5 patients developed secondary myelodysplastic syndrome or acute myeloid leukemia.³³ The feasibility of this R-hyper-CVAD regimen was tested in a multicenter cooperative group setting, but similar results were not seen; in this study, nearly 40% of patients were unable to complete the full scheduled course of treatment due to toxicity.³⁴

​Other ways to intensify therapy in MCL involve adding a second non-cross-resistant cytarabine-containing regimen to R-CHOP after remission, such as DHAP (dexamethasone, high-dose cytarabine, cisplatin), followed by consolidation with an autoSCT. A retrospective registry from the National Comprehensive Cancer Network sought to compare the efficacy of different treatment approaches in the frontline setting. They studied 167 patients with MCL and compared 4 groups: treatment with R-hyper-CVAD, either with or without autoSCT, and treatment with R-CHOP, either with or without autoSCT. This study found that in patients younger than 65, R-CHOP followed by autoSCT or R-hyper-CVAD without autoSCT resulted in similar PF and OS, but was superior to R-CHOP alone for newly diagnosed MCL patients.³⁵ These data support more intensive regimens in younger and fitter patients. Several other prospective and randomized studies have demonstrated clinical benefit for patients with MCL undergoing autoSCT in first remission. Of particular importance is the seminal phase 3 study of the European MCL Network, which established the role of autoSCT in this setting.⁶¹ In this prospective randomized trial involving 122 newly diagnosed MCL patients who responded to CHOP-like induction, patients in CR derived a greater benefit from autoSCT.

More recent studies have demonstrated similar benefits using cytarabine-based autoSCT. The Nordic MCL2 study evaluated 160 patients using R-CHOP, alternating with rituximab and high-dose cytarabine, followed by autoSCT. This study used “maxi-CHOP,” an augmented CHOP regimen (cyclophosphamide 1200 mg/m², doxorubicin 75 mg/m², but standard doses of vincristine [2 mg] and prednisone [100 mg days 1–5]), alternating with 4 infusions of cytarabine at 2 g/m² and standard doses of rituximab (375 mg/m²). Patients then received conditioning with BEAM and autoSCT. Patients were evaluated for the presence of minimal residual disease (MRD) and for the t(11;14) or clonal immunoglobulin heavy chain gene rearrangement with polymerase chain reaction (PCR). Patients with MRD were offered therapy with rituximab at 375 mg/m² weekly for 4 doses. This combination resulted in 10-year OS rates of 58%.³⁶ In a multicenter study involving 78 patients from the Cancer and Leukemia Group B (CALGB), R-CHOP followed by high-dose cytarabine and BEAM-based autoSCT resulted in a 5-year OS of 64%.³⁷ A single-arm phase 2 study from the Netherlands also tested R-CHOP followed by high-dose cytarabine and BEAM-based autoSCT. Nonhematologic toxicities were 22% after high-dose cytarabine, and 55% after BEAM. The ORR was 70%, with a 64% CR rate and 66% OS at 4 years.³⁸ The French GELA group used 3 cycles of R-CHOP and 3 cycles of R-DHAP in a phase 2 study of young (under age 66) MCL patients. Following R-CHOP, the ORR was 93%, and following R-DHAP the ORR was 95%. Five-year OSA was 75%.³⁹ A large randomized phase 3 study by Hermine and colleagues of the EMCLN confirmed the benefit of this approach in 497 patients with newly diagnosed MCL. R-CHOP for 6 cycles followed by autoSCT was compared to R-CHOP for 3 cycles alternating with R-DHAP for 3 cycles and autoSCT with a cytarabine-based conditioning regimen. The addition of cytarabine significantly increased rates of CR, TTF, and OS, without increasing toxicity.⁴⁰

CASE CONTINUED

The patient is treated with R-CHOP chemotherapy for 3 cycles followed by R-DHAP. His course is complicated by mild tinnitus and acute kidney injury from cisplatin that promptly resolves. Three weeks following treatment, a restaging PET/CT scan shows resolution of all lymphadenopathy, with no hypermetabolic uptake, consistent with a complete remission. A repeat bone marrow biopsy shows no involvement with MCL. He subsequently undergoes an autoSCT, and restaging CT/PET 3 months following autoSCT shows continued remission. He is monitored every 3 to 6 months over the next several years.

He has a 4.5-year disease remission, after which he develops growing palpable lymphadenopathy on exam and progressive anemia and thrombocytopenia. A bone marrow biopsy is repeated, which shows recurrent MCL. Restaging diagnostic imaging with a CT scan reveals lymphadenopathy above and below the diaphragm. An axillary lymph node biopsy also demonstrates recurrent MCL. At this time the patient is otherwise in fairly good health, except for feeling fatigued. His ECOG performance status is 1. He begins therapy with bortezomib at a dose of 1.3 mg/m² intravenously on days 1, 4, 8, and 11 for 6 cycles. His treatment course is complicated by painful sensory peripheral neuropathy of the bilateral lower extremities. Restaging studies at the completion of therapy demonstrate that he has achieved a partial response, with a 50% reduction in the size of involved lymphadenopathy and some residual areas of hypermetabolic uptake. His peripheral cytopenias improve moderately.

TREATMENT OF RELAPSED MCL

Single-Agent and Combination Chemotherapy

Whenever possible, and since there is no standard, patients with relapsed MCL should be offered a clinical trial. Outside of a clinical study, many of the treatment regimens used at diagnosis can also be applied in the relapsed setting. In relapsed MCL, Rummel et al showed that BR for 4 cycles resulted in an ORR of 90%, with a CR of 60%. The median PFS was 24 months.⁴¹ Bortezomib, an inhibitor of the proteasome-ubiquitin pathway, leads to apoptosis and cell cycle arrest in MCL.⁴² Multiple studies have evaluated bortezomib both as a single agent and in combination for patients with relapsed MCL. In 2006, bortezomib became the first agent approved by the FDA in relapsed or refractory MCL, based on the phase 2 PINNACLE study. This prospective multicenter study involving 155 patients demonstrated an ORR of 33%, CR rate of 8%, and median treatment duration of 9 months. The median time to progression was 6 months.⁴³ Subsequently, bortezomib-containing combinations evolved. In a multicenter study of relapsed and refractory indolent NHL and MCL, Friedberg and colleagues evaluated bortezomib in combination with BR.⁴⁴ In the MCL cohort, the ORR was 71%. These promising results led to the study of this combination in the frontline setting. The ongoing ECOG 1411 study is using BR for the frontline treatment of MCL with or without bortezomib as induction. This study also includes rituximab maintenance, and randomizes patients to undergo maintenance with or without the immunomodulator lenalidomide. Bortezomib has been associated with herpes simplex and herpes zoster reactivation. Neuropathy has also been observed with bortezomib, which can be attenuated by administering it subcutaneously.

Lenalidomide is an immunomodulatory agent derived from thalidomide. It has significant activity and is a mainstay of treatment in multiple myeloma. Lenalidomide acts by enhancing cellular immunity, has antiproliferative effects, and inhibits T-cell function leading to growth inhibitory effects in the tumor microenvironment.⁴⁵ In MCL, lenalidomide has demonstrated clinical activity both as a single agent and in combination, as well as in preclinical studies establishing its pro-apoptotic effects.⁴⁶ The pivotal EMERGE study evaluated monotherapy with lenalidomide in heavily pretreated relapsed and refractory MCL. This multicenter international study of 134 patents reported an ORR of 28% with a 7.5% CR rate and median PFS of 4 months. All patients had relapsed or progressed following bortezomib. This led to the approval of lenalidomide by the FDA in 2013 for the treatment of patients with MCL whose disease relapsed or progressed following 2 prior therapies, one of which included bortezomib.⁴⁷ Lenalidomide has been associated with neutropenia, secondary cancers, and deep venous thrombosis.

In combination with other agents in the relapsed setting, lenalidomide shows broader activity. A phase 1/2 study by Wang and colleagues demonstrated an ORR of 57%; the median response duration was 19 months when lenalidomide was combined with rituximab for relapsed/refractory MCL.⁴⁸

Novel Therapies

More recently, novel treatment approaches have been tested in MCL based on an increased understanding of aberrant signaling pathways in this disease (Table 3). Constitutive activation of B-cell receptor signaling is critical for the survival and proliferation of lymphomas, and has led to the development of targeted agents inhibiting B-cell receptor–associated protein kinases. Bruton’s tyrosine kinase (BTK) is one essential component of the B-cell receptor.⁴⁹ In particular, proteins upstream of the BTK pathway have been implicated in growth and proliferation of MCL, suggesting that inhibition of BTK may impede lymphomagenesis.⁵⁰ Ibrutinib is an oral inhibitor of BTK, and demonstrates activity in multiple lymphoma subtypes. In a phase 1 study of ibrutinib in relapsed and refractory hematologic malignancies, an ORR of 60% was observed in 50 evaluable patients, with 16% CR. Median PFS was 13 months. Among these, 7 of 9 patients with MCL responded, including 3 CRs.⁵¹ Given these promising results, a phase 2 multicenter study evaluating ibrutinib in relapsed and refractory MCL was completed.⁵² At a dose of 560 mg daily, the response rate was 68%, with CR of 21%. The most common observed treatment-related side effects included diarrhea, fatigue, and nausea. Neutropenia and thrombocytopenia were also observed. Of importance, 5% of patients had grade 3 or higher bleeding events, including subdural hematoma, gastrointestinal bleeding, and hematuria. The estimated OS rate was 58% at 18 months. On the basis of this study, the FDA approved ibrutinib for relapsed and refractory MCL in November 2013.

The PI3K pathway is another survival pathway that is dysfunctional in several hematologic disorders, including MCL. Overexpression of PI3K and its downstream targets contributes to MCL pathogenesis.⁵³ Idelalisib is an oral small molecule inhibitor of the delta isoform of PI3K that is dosed daily; it was approved by the FDA for the treatment of relapsed and refractory follicular lymphoma, small lymphocytic lymphoma, and chronic lymphocytic leukemia. It is being further evaluated in MCL. A dose-escalation phase 1 study in heavily pre-treated MCL patients established safety and tolerability.⁵⁴ Efficacy analysis showed an ORR of 40%, CR of 5%, and 1-year OS of 22%. Further phase 2 studies testing idelalisib as a single agent and in combination for MCL are ongoing. Side effects of idelalisib include elevated liver enzymes, pneumonitis, and diarrhea.

The BCL family of proteins is involved in both pro-and anti-apoptotic functions. BCL2 is an intracellular protein that blocks apoptosis. ABT-199 is an oral BCL2 inhibitor that in early clinical trials has shown very promising activity in MCL. In a phase 1 study of 31 relapsed and refractory NHL patients, all 8 MCL patients (100% ORR) responded to ABT-199 therapy.⁵⁵ Given these promising initial results, multiple studies evaluating ABT-199 are ongoing in MCL as part of first-line treatment as well as for relapsed disease. ABT-199 has been implicated in tumor lysis syndrome, and in early studies of chronic lymphocytic leukemia, fatal tumor lysis was observed.

The mammalian target of rapamycin (mTOR) inhibitor temsirolimus has been evaluated in relapsed MCL. It is given weekly at 250 mg intravenously. Response rates to single-agent temsirolimus are approximately 20% to 35%, and are higher when combined with rituximab.^56,57 The phase 2 study evaluating temsirolimus as a single agent enrolled 35 heavily pre-treated patients. ORR was 38% with only 1 CR. The duration of response was 7 months. Temsirolimus is approved for relapsed MCL in Europe but not in the United States. Similar to the other targeted agents, temsirolimus is actively being studied in combination with other active agents in MCL. Adverse effects noted with temsirolimus include diarrhea, stomatitis, and rash. Thrombocytopenia requiring dose reductions is another frequently observed complication.

Radioimmunotherapy

Radioimmunotherapy (RIT) has been studied extensively in MCL. RIT consists of anti-CD 20 antibodies coupled to radioactive particles that deliver radiation to targeted cells, minimizing toxicity to surrounding tissues. RIT is not used as frequently in the modern era as it had been in the past. At this time, only yttrium-90-ibritumomab tiuxetan is available.

RIT has been evaluated in MCL both at the time of relapse⁵⁸ and more recently, as part of a conditioning regimen prior to autoSCT, with good tolerability.^65–67 Averse events noted with RIT include hematologic toxicity (can be prolonged), hypothyroidism, and in rare cases, myelodysplastic syndrome and acute leukemia. The bone marrow must have less than 25% involvement with disease prior to administration. Wang and colleagues evaluated yttrium-90-ibritumomab tiuxetan in 34 heavily pretreated patients with MCL.⁵⁸ They observed an ORR of 31%. The median event-free survival (EFS) was 6 months, but in patients achieving either CR or PR, EFS was 28 months. A 21-month OS was noted.

In the upfront setting, RIT has been added as a mechanism of intensification. A recent Nordic group study of RIT with autoSCT did not find benefit with the addition of RIT.⁵⁹ An ECOG study recently added yttrium-90-ibritumomab tiuxetan after CHOP chemotherapy in the upfront treatment of MCL, with good tolerability.⁵⁵ However, when added to R-hyper-CVAD, the combination had unexpected high rates of hematologic toxicity, including grade 3/4 cytopenias and an unacceptably high rate of secondary malignancies.⁶⁸

AutoSCT or Allogeneic Transplant

While many studies noted above have established the beneficial role of autoSCT in MCL in first remission, the role of allogeneic transplant (alloSCT) in MCL remains controversial. A recent large retrospective study conducted by the Center for International Blood and Marrow Transplant Research (CIBMTR) evaluated 519 patients with MCL who underwent both autoSCT and alloSCT.⁶⁰ Patients were grouped into an early cohort (transplant in first PR or CR, and 2 or fewer treatments) and late cohort (all other patients). The analysis had mature follow up. A multivariate analysis demonstrated that early autoSCT was associated with superior outcomes compared to autoSCT performed later. While it was not possible to demonstrate a survival benefit favoring autoSCT over reduced intensity (RIC) alloSCT, patients transplanted later in their disease course had shorter OS. For patients receiving autoSCT in CR 1 following only 1 prior line of therapy, OS at 5 years was 75% and PFS was 70%. Patients undergoing RIC followed by alloSCT had fewer relapses, but this was negated by higher nonrelapse mortality (25%), resulting in a PFS similar to autoSCT.

CASE CONCLUSION

After treatment with bortezomib the patient is well for 9 months. Subsequently, however, he develops increasing lymphadenopathy and progressive fatigue. He is then started on lenalidomide 25 mg orally daily for 21 out of 28 days. He experiences significant fatigue with lenalidomide and prolonged neutropenia requiring dose delays, despite dose modification to 10 mg orally daily. He requires discontinuation of lenalidomide. Given persistent disease, the patient then begins treatment with ibrutinib. Within a few days of starting ibrutinib therapy, he experiences a marked but transient leukocytosis. Two months later, the patient’s palpable lymphadenopathy has decreased, and his anemia and thrombocytopenia related to MCL are improving. He has tolerated treatment well. His course has been complicated only by a mild, pruritic maculopapular eruption on his chest, back, and arms, that was responsive to topical low-dose steroids. He remains on ibrutinib 1 year later.

CONCLUSION

Advances in our understanding of MCL treatment are revolutionizing the approach to this once deadly disease. Over the next several years, these gains will weave themselves into the current treatment paradigm and likely alter the treatment landscape for MCL as we know it.

INTRODUCTION

Mantle cell lymphoma (MCL) is an uncommon, distinct clinical subtype of non-Hodgkin lymphoma (NHL) that comprises approximately 8% of all lymphoma diagnoses in the United States and Europe.^1,2 Considered incurable, MCL often presents in advanced stages, particularly with involvement of the lymph nodes, spleen, bone marrow, and gastrointestinal tract in the form of lymphomatous polyps. MCL disproportionately affects males, and incidence rises with age, with a median age at diagnosis of 68 years.² Historically, the prognosis of patients with MCL has been among the poorest among B-cell lymphoma patients, with a median overall survival (OS) of 3 to 5 years, and time to treatment failure (TTF) of 18 to 24 months, although this is improving in the modern era.³ Less frequently, patients with MCL display isolated bone marrow, peripheral blood, and splenic involvement. These cases tend to behave more indolently with longer survival.^4,5 Recent advances in therapy have dramatically impacted treatment alternatives and outcomes for MCL. As such, the therapeutic and prognostic landscape of MCL is evolving rapidly.

PATHOGENESIS

The histologic diagnosis of MCL by morphology alone is often challenging. Accurate diagnosis relies on immunohistochemical staining for the purposes of immunophenotyping.⁶ MCL typically expresses B-cell markers CD5 and CD20, and lacks both CD10 and CD23. The genetic hallmark of MCL is the t(11;14) (q13;q32) chromosomal translocation leading to upregulation of the cyclin D1 protein, a critical regulator of the G1 phase of the cell cycle. Specifically, the t(11;14) translocation, present in virtually all cases of MCL, juxtaposes the proto-oncogene CCND1 to the immunoglobulin heavy chain gene.⁷ Consequently, cyclin D1, normally not expressed in B lymphocytes, becomes constitutively overexpressed. This alteration is thought to facilitate the deregulation of the cell cycle at the G1-S phase transition.⁸

Gene expression profiling studies have underscored the importance of cell cycle deregulation in MCL, and high proliferation is associated with a worse prognosis.⁹ More than 50% of the genes associated with poor outcomes were derived from the “proliferation signature” that was more highly expressed in dividing cells. In the seminal Rosenwald study, a gene expression–based outcome model was constructed in which the proliferation signature average represents a linear variable that assigns a discrete probability of survival to an individual patient.⁹ The proliferative index, or proliferative signature, of MCL can be estimated by the percentage of Ki-67–positive cells present in the tumor through immunohistochemistry. This is often used as a marker of poor outcomes, and as a surrogate for the proliferative signature in MCL that can be incorporated into clinical practice (as opposed to gene expression profiling). Statistically significant differences in OS have emerged between groups of MCL patients with Ki-67–positive cells comprising less than 30% of their tumor sample (favorable) and those with Ki-67–positive cells comprising 30% or greater (unfavorable).¹⁰

Recent data has also identified the importance of the transcription factor SOX 11 (SRY-related HMG-box), which regulates multiple cellular transcriptional events, including cell proliferation and differentiation, apoptosis, and angiogenesis.¹¹ MCL expressing SOX 11 behaves more aggressively than MCL variants lacking SOX 11 expression, and tends to accumulate more genetic alterations.¹² Moreover, lack of SOX 11 expression characterizes a subset of MCL that does not carry the t(11;14) translocation.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 62-year-old man with a history of diabetes mellitus and hypertension presents with cervical lymphadenopathy, fatigue, and early satiety over the past several months. He is otherwise in good health. His Eastern Cooperative Oncology Group (ECOG) performance status is 1. On physical examination, 3-cm lymphadenopathy in the bilateral cervical chain is noted. Bilateral axillary lymph nodes measure 2 to 4 cm. His spleen is enlarged and is palpable at approximately 5 cm below the costal margin. A complete blood count reveals a total white blood cell (WBC) count of 14,000 cells/μL, with 68% lymphocytes and a normal distribution of neutrophils. Hemoglobin is 11 g/dL, and platelet count is 112,000/μL. The lactate dehydrogenase (LDH) level is 322 U/L (upper limit of normal: 225 U/L).

• How is MCL diagnosed?

Diagnosis of MCL requires review by expert hematopathologists.¹³ Whenever possible, an excisional biopsy should be performed for the adequate characterization of lymph node architecture and evaluation by immunohistochemistry. Aside from the characteristic expression of CD5 and CD20 and absence of CD23, MCL should express cyclin D1, which reflects t(11;14). If cyclin D1 is inconclusive or unavailable, fluorescent in situ hybridization (FISH) for t(11;14) should be performed.⁸ Patients often have circulating malignant lymphocytes, or leukemic phase MCL. Flow cytometry of the peripheral blood can detect traditional surface markers, and FISH can also be performed on circulating abnormal lymphocytes.

RISK STRATIFICATION

At diagnosis, patients should undergo risk stratification in order to understand prognosis and possibly guide treatment. In MCL, the MCL international prognostic index (MIPI) is used. The MIPI is a prognostic tool developed exclusively for patients with MCL using data from 455 patients with advanced-stage MCL treated within 3 European clinical trials.¹⁶ The MIPI classified patients into risk groups based on age, ECOG performance status, LDH level, and WBC count. Patients were categorized into low-risk (44% of patients, median OS not reached), intermediate-risk (35%, median OS 51 months), and high-risk groups (21%, median OS 29 months). This is done through a logarithmic calculation, which can be accessed through online calculators (a prototype example can be found at www.qxmd.com/calculate-online/hematology/prognosis-mantle-cell-lymphoma-mipi). Cell proliferation using the Ki-67 index was evaluated in an exploratory analysis (the biologic [“B”] MIPI), and also demonstrated strong prognostic relevance.¹⁶ Currently, treatment of MCL patients is not stratified by MIPI outside of a clinical trial, but this useful tool assists in assessing patient prognosis and has been validated for use with both conventional chemoimmunotherapy and in the setting of autologous stem cell transplant (autoSCT).^16,17 At this point in time, the MIPI score is not used to stratify treatment, although some clinical trials are incorporating the use of the MIPI score at diagnosis. Nonetheless, given its prognostic importance, the MIPI should be performed for all MCL patients undergoing staging and evaluation for treatment to establish disease risk.

As noted, the proliferative signature, represented by the Ki-67 protein, is also highly prognostic in MCL. Ki-67 is expressed in the late G1, S, G2, and M phases of the cell cycle. The Ki-67 index is defined by the hematopathologist as the percentage of lymphoma cells staining positive for Ki-67 protein, based on the number of cells per high-power field. There is significant interobserver variability in this process, which can be minimized by assessing Ki-67 quantitatively using computer software. The prognostic significance of Ki-67 at diagnosis was established in large studies of MCL patient cohorts, with survival differing by up to 3 years.^18,19 Determann et al demonstrated the utility of the proliferative index in patients with MCL treated with standard chemoimmunotherapy.¹⁰ In this study, 249 patients with advanced-stage MCL treated within randomized trials conducted by the European MCL Network were analyzed. The Ki-67 index was found to be extremely prognostic of OS, independent of other clinical risk factors, including the MIPI score. As a continuous variable, Ki-67 indices of greater than 10% correlated with poor outcomes. The Ki-67 index has also been confirmed as prognostic in relapsed MCL.²⁰ It is important to note that, as a unique feature, the Ki-67 index has remained an independent prognostic factor, even when incorporated into the “B” MIPI.

TREATMENT

CASE CONTINUED

The patient undergoes an excisional biopsy of a cervical lymph node, which demonstrates an abnormal proliferation of small-medium–sized lymphocytes with slightly irregular nuclear contours. Immunohistochemistry shows that the abnormal lymphocytes are positive for CD20 and CD5, negative for CD10 and CD23, and diffusely positive for cyclin D1, consistent with a diagnosis of MCL. The proliferative index, as measured by the Ki-67 immunostain, is 40%. A bone marrow aspirate and biopsy are then obtained, which show a clonal population of B lymphocytes expressing the same immunophenotype as the lymph node (positive for CD20 and CD5, negative for CD10 and CD23, cyclin D1 positive). A CT scan of the neck, chest, abdomen, and pelvis with contrast is obtained, along with a PET scan. These studies identify extensive hypermetabolic lymphadenopathy in the bilateral cervical chains, supraclavicular areas, mediastinum, and hilum. Mesenteric lymph nodes are also enlarged and hypermetabolic, as are retroperitoneal lymph nodes. The spleen is noted to be enlarged with multiple hypermetabolic lesions. Based on the presence of extensive lymphadenopathy as well as bone marrow involvement, the patient is diagnosed with stage IV MCL. He undergoes risk-stratification with the MIPI. His MIPI score is 6.3, high risk.

FRONTLINE THERAPY

Role of Watchful Waiting

A small proportion of MCL patients have indolent disease that can be observed. This population is more likely to have leukemic-phase MCL with circulating lymphocytes, splenomegaly, and bone marrow involvement and absent or minimal lymphadenopathy.^4,5 A retrospective study of 97 patients established that deferment of initial therapy in MCL is acceptable in some patients.⁵ In this study, approximately one third of patients with MCL were observed for more than 3 months before initiating systemic therapy, and the median time to treatment for the observation group was 12 months. Most patients undergoing observation had a low-risk MIPI. Patients were not harmed by observation, as no OS differences were observed among groups. This study underscores that deferred treatment can be an acceptable alternative in selected MCL patients for a short period of time. In practice, the type of patient who would be appropriate for this approach is someone who is frail, elderly, and with multiple comorbidities. Additionally, expectant observation could be considered for patients with limited-stage or low-volume MCL, low Ki-67 index, and low-risk MIPI scores.

Approach to Therapy

Treatment of MCL is generally approached by evaluating patient age and fitness for treatment. While there is no accepted standard, for younger patients healthy enough to tolerate aggressive approaches, treatment often involves an intensive cytarabine-containing regimen, which is consolidated with an autoSCT. This approach results in the longest remission duration, with some series suggesting a plateau in survival after 5 years, with no relapses.²¹ Nonintensive conventional chemotherapy alone is often reserved for the frailer or older patient. Given that remission durations with chemotherapy alone in MCL are short, goals of treatment focus on maximizing benefit and remission duration and minimizing risk of toxicity.

Standard Chemotherapy: Elderly and/or Frail Patients

Conventional chemotherapy alone for the treatment of MCL results in a 70% to 85% overall response rate (ORR) and 7% to 30% complete response (CR) rate.²² Rituximab, a mouse humanized monoclonal IgG1 anti-CD20 antibody, is used as standard of care in combination with chemotherapy, since its addition has been found to increase response rates and extend both progression-free survival (PFS) and OS compared to chemotherapy alone.^23,24 However, chemoimmunotherapy approaches do not provide long-term control of MCL and are considered noncurative. Various regimens have been studied and include anthracycline-containing regimens such as R-CHOP (rituximab with cyclophosphamide, doxorubicin, vincristine, prednisone),²² combination chemotherapy with antimetabolites such as R-hyper-CVAD (hyper-fractionated rituximab with cyclophosphamide, vincristine, doxorubicin, dexamethasone, alternating with methotrexate and cytarabine),²⁵ purine analogue–based regimens such as R-FC (rituximab with fludarabine and cyclophosphamide),²⁶ bortezomib-containing regimens,²⁷ and alkylator-based treatment with BR (bendamustine and rituximab) (Table 1).^28,29 Among these, the most commonly used are R-CHOP and BR.

Two large randomized studies compared R-CHOP for 6 cycles to BR for 6 cycles in patients with indolent NHL and MCL. Among MCL patients, BR resulted in superior PFS compared to R-CHOP (69 months versus 26 months) but no benefit in OS.^28,29 The ORR to R-CHOP was approximately 90%, with a PFS of 21 months in the Rummel et al study.²⁹ This study included more than 80 centers in Germany and enrolled 549 patients with MCL, follicular lymphoma, small lymphocytic lymphoma, marginal zone lymphoma, and Waldenström macroglobulinemia. Patients were randomized in a 1:1 fashion. Among these, 46 patients received BR and 48 received R-CHOP (18% for both, respectively). It should be noted that patients in the BR group had significantly less toxicity and experienced fewer side effects than did those in the R-CHOP group. Similarly, BR-treated patients had a lower frequency of hematologic side effects and infections of any grade. However, drug-associated skin reactions and allergies were more common with BR compared to R-CHOP. The study by Flinn and colleagues was an international randomized, noninferiority phase 3 study designed to evaluate the efficacy and safety of BR compared with R-CHOP or R-CVP (rituximab plus cyclophosphamide, vincristine, and prednisone) for treatment-naive patients with MCL or other indolent NHL. The primary endpoint was CR. In this study, BR was found to be noninferior to R-CHOP and R-CVP based on CR rate (31% versus 25%, respectively; P = 0.0225). Response rates in general were high: 97% for BR and 91% for R-CHOP/R-CVP (P = 0.0102). Here, BR-treated patients experienced more nausea, emesis, and drug-induced hypersensitivity compared to the R-CHOP and R-CVP groups.

Bendamustine in combination with other agents has also been studied in the frontline setting. Visco and colleagues evaluated the combination of bendamustine with rituximab and cytarabine (R-BAC) in older patients with MCL (age 65 or older).⁶³ This phase 2, two-stage study enrolled 40 patients and had a dose-finding arm for cytarabine in combination with BR. It permitted relapsed/refractory patients, but 50% had newly diagnosed, previously untreated MCL. The regimen had an impressive ORR of 100%, with CR rates of 95% for previously untreated patients. PFS at 2 years was 95%. R-BAC was well tolerated, with the primary toxicity being reversible myelosuppression.

BR was combined with the proteasome inhibitor bortezomib and dexamethasone in a phase 2 study.⁶⁴ This Lymphoma Study Association (LYSA) study evaluated 76 patients with newly diagnosed MCL older than age 65 years. BR was administered in standard doses (bendamustine 90 mg/m² on days 1 and 2 and rituximab 375 mg/m² IV on day 1) and bortezomib was administered subcutaneously on days 1, 4, 8, and 11, with acyclovir for viral prophylaxis. Patients received 6 cycles. The ORR was 87% and the CR was 60%. Patients experienced toxicity, and not all bortezomib doses were administered due to neurotoxic or hematologic side effects.

A randomized phase 3 study compared R-CHOP to the VR-CAP regimen (R-CHOP regimen but bortezomib replaces vincristine on days 1, 4, 8, 11, at 1.3 mg/m²) in 487 newly diagnosed MCL patients.²⁷ Median PFS was superior in the VR-CAP group compared with R-CHOP (14.4 months versus 24.7 months, respectively). Additionally, rates of CR were superior in the VR-CAP group (53% compared to 42% with R-CHOP). However, there was more hematologic toxicity with VR-CAP. On the basis of these findings, the U.S. Food and Drug Administration approved bortezomib for the frontline treatment of MCL.

Other chemoimmunotherapy combinations containing bortezomib have been studied in frontline MCL treatment, with promising results. These include bortezomib in combination with R-CHOP or modified R-hyper-CVAD, as well as bortezomib in combination with CHOP-like treatments and purine analogues.^27,30–32 The ongoing ECOG 1411 study is currently evaluating bortezomib added to BR for induction therapy of newly diagnosed MCL in a 4-arm randomized trial. Patients receive BR with or without bortezomib during induction and are then randomly assigned to maintenance with either rituximab alone or rituximab with lenalidomide. Other novel combination agents are actively being studied in frontline MCL treatment, including lenalidomide and rituximab and BR with lenalidomide.

Intensification of Therapy and AutoSCT: Fitter and/or Younger Patients

Short response duration has created the need for post-remission therapy in MCL. One approach to improve remission duration in MCL is to intensify induction through the use of cytarabine-containing regimens and/or consolidation with high-dose chemotherapy, typically using BEAM (carmustine, etoposide, cytarabine, melphalan) and autoSCT (Table 2). The cytarabine-containing R-hyper-CVAD regimen, developed at the MD Anderson Cancer Center, resulted in a 97% ORR and an 87% CR rate, with TTF of nearly 5 years. However, nearly one third of patients were unable to complete treatment due to toxicity, and 5 patients developed secondary myelodysplastic syndrome or acute myeloid leukemia.³³ The feasibility of this R-hyper-CVAD regimen was tested in a multicenter cooperative group setting, but similar results were not seen; in this study, nearly 40% of patients were unable to complete the full scheduled course of treatment due to toxicity.³⁴

​Other ways to intensify therapy in MCL involve adding a second non-cross-resistant cytarabine-containing regimen to R-CHOP after remission, such as DHAP (dexamethasone, high-dose cytarabine, cisplatin), followed by consolidation with an autoSCT. A retrospective registry from the National Comprehensive Cancer Network sought to compare the efficacy of different treatment approaches in the frontline setting. They studied 167 patients with MCL and compared 4 groups: treatment with R-hyper-CVAD, either with or without autoSCT, and treatment with R-CHOP, either with or without autoSCT. This study found that in patients younger than 65, R-CHOP followed by autoSCT or R-hyper-CVAD without autoSCT resulted in similar PF and OS, but was superior to R-CHOP alone for newly diagnosed MCL patients.³⁵ These data support more intensive regimens in younger and fitter patients. Several other prospective and randomized studies have demonstrated clinical benefit for patients with MCL undergoing autoSCT in first remission. Of particular importance is the seminal phase 3 study of the European MCL Network, which established the role of autoSCT in this setting.⁶¹ In this prospective randomized trial involving 122 newly diagnosed MCL patients who responded to CHOP-like induction, patients in CR derived a greater benefit from autoSCT.

More recent studies have demonstrated similar benefits using cytarabine-based autoSCT. The Nordic MCL2 study evaluated 160 patients using R-CHOP, alternating with rituximab and high-dose cytarabine, followed by autoSCT. This study used “maxi-CHOP,” an augmented CHOP regimen (cyclophosphamide 1200 mg/m², doxorubicin 75 mg/m², but standard doses of vincristine [2 mg] and prednisone [100 mg days 1–5]), alternating with 4 infusions of cytarabine at 2 g/m² and standard doses of rituximab (375 mg/m²). Patients then received conditioning with BEAM and autoSCT. Patients were evaluated for the presence of minimal residual disease (MRD) and for the t(11;14) or clonal immunoglobulin heavy chain gene rearrangement with polymerase chain reaction (PCR). Patients with MRD were offered therapy with rituximab at 375 mg/m² weekly for 4 doses. This combination resulted in 10-year OS rates of 58%.³⁶ In a multicenter study involving 78 patients from the Cancer and Leukemia Group B (CALGB), R-CHOP followed by high-dose cytarabine and BEAM-based autoSCT resulted in a 5-year OS of 64%.³⁷ A single-arm phase 2 study from the Netherlands also tested R-CHOP followed by high-dose cytarabine and BEAM-based autoSCT. Nonhematologic toxicities were 22% after high-dose cytarabine, and 55% after BEAM. The ORR was 70%, with a 64% CR rate and 66% OS at 4 years.³⁸ The French GELA group used 3 cycles of R-CHOP and 3 cycles of R-DHAP in a phase 2 study of young (under age 66) MCL patients. Following R-CHOP, the ORR was 93%, and following R-DHAP the ORR was 95%. Five-year OSA was 75%.³⁹ A large randomized phase 3 study by Hermine and colleagues of the EMCLN confirmed the benefit of this approach in 497 patients with newly diagnosed MCL. R-CHOP for 6 cycles followed by autoSCT was compared to R-CHOP for 3 cycles alternating with R-DHAP for 3 cycles and autoSCT with a cytarabine-based conditioning regimen. The addition of cytarabine significantly increased rates of CR, TTF, and OS, without increasing toxicity.⁴⁰

CASE CONTINUED

The patient is treated with R-CHOP chemotherapy for 3 cycles followed by R-DHAP. His course is complicated by mild tinnitus and acute kidney injury from cisplatin that promptly resolves. Three weeks following treatment, a restaging PET/CT scan shows resolution of all lymphadenopathy, with no hypermetabolic uptake, consistent with a complete remission. A repeat bone marrow biopsy shows no involvement with MCL. He subsequently undergoes an autoSCT, and restaging CT/PET 3 months following autoSCT shows continued remission. He is monitored every 3 to 6 months over the next several years.

He has a 4.5-year disease remission, after which he develops growing palpable lymphadenopathy on exam and progressive anemia and thrombocytopenia. A bone marrow biopsy is repeated, which shows recurrent MCL. Restaging diagnostic imaging with a CT scan reveals lymphadenopathy above and below the diaphragm. An axillary lymph node biopsy also demonstrates recurrent MCL. At this time the patient is otherwise in fairly good health, except for feeling fatigued. His ECOG performance status is 1. He begins therapy with bortezomib at a dose of 1.3 mg/m² intravenously on days 1, 4, 8, and 11 for 6 cycles. His treatment course is complicated by painful sensory peripheral neuropathy of the bilateral lower extremities. Restaging studies at the completion of therapy demonstrate that he has achieved a partial response, with a 50% reduction in the size of involved lymphadenopathy and some residual areas of hypermetabolic uptake. His peripheral cytopenias improve moderately.

TREATMENT OF RELAPSED MCL

Single-Agent and Combination Chemotherapy

Whenever possible, and since there is no standard, patients with relapsed MCL should be offered a clinical trial. Outside of a clinical study, many of the treatment regimens used at diagnosis can also be applied in the relapsed setting. In relapsed MCL, Rummel et al showed that BR for 4 cycles resulted in an ORR of 90%, with a CR of 60%. The median PFS was 24 months.⁴¹ Bortezomib, an inhibitor of the proteasome-ubiquitin pathway, leads to apoptosis and cell cycle arrest in MCL.⁴² Multiple studies have evaluated bortezomib both as a single agent and in combination for patients with relapsed MCL. In 2006, bortezomib became the first agent approved by the FDA in relapsed or refractory MCL, based on the phase 2 PINNACLE study. This prospective multicenter study involving 155 patients demonstrated an ORR of 33%, CR rate of 8%, and median treatment duration of 9 months. The median time to progression was 6 months.⁴³ Subsequently, bortezomib-containing combinations evolved. In a multicenter study of relapsed and refractory indolent NHL and MCL, Friedberg and colleagues evaluated bortezomib in combination with BR.⁴⁴ In the MCL cohort, the ORR was 71%. These promising results led to the study of this combination in the frontline setting. The ongoing ECOG 1411 study is using BR for the frontline treatment of MCL with or without bortezomib as induction. This study also includes rituximab maintenance, and randomizes patients to undergo maintenance with or without the immunomodulator lenalidomide. Bortezomib has been associated with herpes simplex and herpes zoster reactivation. Neuropathy has also been observed with bortezomib, which can be attenuated by administering it subcutaneously.

Lenalidomide is an immunomodulatory agent derived from thalidomide. It has significant activity and is a mainstay of treatment in multiple myeloma. Lenalidomide acts by enhancing cellular immunity, has antiproliferative effects, and inhibits T-cell function leading to growth inhibitory effects in the tumor microenvironment.⁴⁵ In MCL, lenalidomide has demonstrated clinical activity both as a single agent and in combination, as well as in preclinical studies establishing its pro-apoptotic effects.⁴⁶ The pivotal EMERGE study evaluated monotherapy with lenalidomide in heavily pretreated relapsed and refractory MCL. This multicenter international study of 134 patents reported an ORR of 28% with a 7.5% CR rate and median PFS of 4 months. All patients had relapsed or progressed following bortezomib. This led to the approval of lenalidomide by the FDA in 2013 for the treatment of patients with MCL whose disease relapsed or progressed following 2 prior therapies, one of which included bortezomib.⁴⁷ Lenalidomide has been associated with neutropenia, secondary cancers, and deep venous thrombosis.

In combination with other agents in the relapsed setting, lenalidomide shows broader activity. A phase 1/2 study by Wang and colleagues demonstrated an ORR of 57%; the median response duration was 19 months when lenalidomide was combined with rituximab for relapsed/refractory MCL.⁴⁸

Novel Therapies

More recently, novel treatment approaches have been tested in MCL based on an increased understanding of aberrant signaling pathways in this disease (Table 3). Constitutive activation of B-cell receptor signaling is critical for the survival and proliferation of lymphomas, and has led to the development of targeted agents inhibiting B-cell receptor–associated protein kinases. Bruton’s tyrosine kinase (BTK) is one essential component of the B-cell receptor.⁴⁹ In particular, proteins upstream of the BTK pathway have been implicated in growth and proliferation of MCL, suggesting that inhibition of BTK may impede lymphomagenesis.⁵⁰ Ibrutinib is an oral inhibitor of BTK, and demonstrates activity in multiple lymphoma subtypes. In a phase 1 study of ibrutinib in relapsed and refractory hematologic malignancies, an ORR of 60% was observed in 50 evaluable patients, with 16% CR. Median PFS was 13 months. Among these, 7 of 9 patients with MCL responded, including 3 CRs.⁵¹ Given these promising results, a phase 2 multicenter study evaluating ibrutinib in relapsed and refractory MCL was completed.⁵² At a dose of 560 mg daily, the response rate was 68%, with CR of 21%. The most common observed treatment-related side effects included diarrhea, fatigue, and nausea. Neutropenia and thrombocytopenia were also observed. Of importance, 5% of patients had grade 3 or higher bleeding events, including subdural hematoma, gastrointestinal bleeding, and hematuria. The estimated OS rate was 58% at 18 months. On the basis of this study, the FDA approved ibrutinib for relapsed and refractory MCL in November 2013.

The PI3K pathway is another survival pathway that is dysfunctional in several hematologic disorders, including MCL. Overexpression of PI3K and its downstream targets contributes to MCL pathogenesis.⁵³ Idelalisib is an oral small molecule inhibitor of the delta isoform of PI3K that is dosed daily; it was approved by the FDA for the treatment of relapsed and refractory follicular lymphoma, small lymphocytic lymphoma, and chronic lymphocytic leukemia. It is being further evaluated in MCL. A dose-escalation phase 1 study in heavily pre-treated MCL patients established safety and tolerability.⁵⁴ Efficacy analysis showed an ORR of 40%, CR of 5%, and 1-year OS of 22%. Further phase 2 studies testing idelalisib as a single agent and in combination for MCL are ongoing. Side effects of idelalisib include elevated liver enzymes, pneumonitis, and diarrhea.

The BCL family of proteins is involved in both pro-and anti-apoptotic functions. BCL2 is an intracellular protein that blocks apoptosis. ABT-199 is an oral BCL2 inhibitor that in early clinical trials has shown very promising activity in MCL. In a phase 1 study of 31 relapsed and refractory NHL patients, all 8 MCL patients (100% ORR) responded to ABT-199 therapy.⁵⁵ Given these promising initial results, multiple studies evaluating ABT-199 are ongoing in MCL as part of first-line treatment as well as for relapsed disease. ABT-199 has been implicated in tumor lysis syndrome, and in early studies of chronic lymphocytic leukemia, fatal tumor lysis was observed.

The mammalian target of rapamycin (mTOR) inhibitor temsirolimus has been evaluated in relapsed MCL. It is given weekly at 250 mg intravenously. Response rates to single-agent temsirolimus are approximately 20% to 35%, and are higher when combined with rituximab.^56,57 The phase 2 study evaluating temsirolimus as a single agent enrolled 35 heavily pre-treated patients. ORR was 38% with only 1 CR. The duration of response was 7 months. Temsirolimus is approved for relapsed MCL in Europe but not in the United States. Similar to the other targeted agents, temsirolimus is actively being studied in combination with other active agents in MCL. Adverse effects noted with temsirolimus include diarrhea, stomatitis, and rash. Thrombocytopenia requiring dose reductions is another frequently observed complication.

Radioimmunotherapy

Radioimmunotherapy (RIT) has been studied extensively in MCL. RIT consists of anti-CD 20 antibodies coupled to radioactive particles that deliver radiation to targeted cells, minimizing toxicity to surrounding tissues. RIT is not used as frequently in the modern era as it had been in the past. At this time, only yttrium-90-ibritumomab tiuxetan is available.

RIT has been evaluated in MCL both at the time of relapse⁵⁸ and more recently, as part of a conditioning regimen prior to autoSCT, with good tolerability.^65–67 Averse events noted with RIT include hematologic toxicity (can be prolonged), hypothyroidism, and in rare cases, myelodysplastic syndrome and acute leukemia. The bone marrow must have less than 25% involvement with disease prior to administration. Wang and colleagues evaluated yttrium-90-ibritumomab tiuxetan in 34 heavily pretreated patients with MCL.⁵⁸ They observed an ORR of 31%. The median event-free survival (EFS) was 6 months, but in patients achieving either CR or PR, EFS was 28 months. A 21-month OS was noted.

In the upfront setting, RIT has been added as a mechanism of intensification. A recent Nordic group study of RIT with autoSCT did not find benefit with the addition of RIT.⁵⁹ An ECOG study recently added yttrium-90-ibritumomab tiuxetan after CHOP chemotherapy in the upfront treatment of MCL, with good tolerability.⁵⁵ However, when added to R-hyper-CVAD, the combination had unexpected high rates of hematologic toxicity, including grade 3/4 cytopenias and an unacceptably high rate of secondary malignancies.⁶⁸

AutoSCT or Allogeneic Transplant

While many studies noted above have established the beneficial role of autoSCT in MCL in first remission, the role of allogeneic transplant (alloSCT) in MCL remains controversial. A recent large retrospective study conducted by the Center for International Blood and Marrow Transplant Research (CIBMTR) evaluated 519 patients with MCL who underwent both autoSCT and alloSCT.⁶⁰ Patients were grouped into an early cohort (transplant in first PR or CR, and 2 or fewer treatments) and late cohort (all other patients). The analysis had mature follow up. A multivariate analysis demonstrated that early autoSCT was associated with superior outcomes compared to autoSCT performed later. While it was not possible to demonstrate a survival benefit favoring autoSCT over reduced intensity (RIC) alloSCT, patients transplanted later in their disease course had shorter OS. For patients receiving autoSCT in CR 1 following only 1 prior line of therapy, OS at 5 years was 75% and PFS was 70%. Patients undergoing RIC followed by alloSCT had fewer relapses, but this was negated by higher nonrelapse mortality (25%), resulting in a PFS similar to autoSCT.

CASE CONCLUSION

After treatment with bortezomib the patient is well for 9 months. Subsequently, however, he develops increasing lymphadenopathy and progressive fatigue. He is then started on lenalidomide 25 mg orally daily for 21 out of 28 days. He experiences significant fatigue with lenalidomide and prolonged neutropenia requiring dose delays, despite dose modification to 10 mg orally daily. He requires discontinuation of lenalidomide. Given persistent disease, the patient then begins treatment with ibrutinib. Within a few days of starting ibrutinib therapy, he experiences a marked but transient leukocytosis. Two months later, the patient’s palpable lymphadenopathy has decreased, and his anemia and thrombocytopenia related to MCL are improving. He has tolerated treatment well. His course has been complicated only by a mild, pruritic maculopapular eruption on his chest, back, and arms, that was responsive to topical low-dose steroids. He remains on ibrutinib 1 year later.

CONCLUSION

Advances in our understanding of MCL treatment are revolutionizing the approach to this once deadly disease. Over the next several years, these gains will weave themselves into the current treatment paradigm and likely alter the treatment landscape for MCL as we know it.

INTRODUCTION

Mantle cell lymphoma (MCL) is an uncommon, distinct clinical subtype of non-Hodgkin lymphoma (NHL) that comprises approximately 8% of all lymphoma diagnoses in the United States and Europe.^1,2 Considered incurable, MCL often presents in advanced stages, particularly with involvement of the lymph nodes, spleen, bone marrow, and gastrointestinal tract in the form of lymphomatous polyps. MCL disproportionately affects males, and incidence rises with age, with a median age at diagnosis of 68 years.² Historically, the prognosis of patients with MCL has been among the poorest among B-cell lymphoma patients, with a median overall survival (OS) of 3 to 5 years, and time to treatment failure (TTF) of 18 to 24 months, although this is improving in the modern era.³ Less frequently, patients with MCL display isolated bone marrow, peripheral blood, and splenic involvement. These cases tend to behave more indolently with longer survival.^4,5 Recent advances in therapy have dramatically impacted treatment alternatives and outcomes for MCL. As such, the therapeutic and prognostic landscape of MCL is evolving rapidly.

PATHOGENESIS

The histologic diagnosis of MCL by morphology alone is often challenging. Accurate diagnosis relies on immunohistochemical staining for the purposes of immunophenotyping.⁶ MCL typically expresses B-cell markers CD5 and CD20, and lacks both CD10 and CD23. The genetic hallmark of MCL is the t(11;14) (q13;q32) chromosomal translocation leading to upregulation of the cyclin D1 protein, a critical regulator of the G1 phase of the cell cycle. Specifically, the t(11;14) translocation, present in virtually all cases of MCL, juxtaposes the proto-oncogene CCND1 to the immunoglobulin heavy chain gene.⁷ Consequently, cyclin D1, normally not expressed in B lymphocytes, becomes constitutively overexpressed. This alteration is thought to facilitate the deregulation of the cell cycle at the G1-S phase transition.⁸

Gene expression profiling studies have underscored the importance of cell cycle deregulation in MCL, and high proliferation is associated with a worse prognosis.⁹ More than 50% of the genes associated with poor outcomes were derived from the “proliferation signature” that was more highly expressed in dividing cells. In the seminal Rosenwald study, a gene expression–based outcome model was constructed in which the proliferation signature average represents a linear variable that assigns a discrete probability of survival to an individual patient.⁹ The proliferative index, or proliferative signature, of MCL can be estimated by the percentage of Ki-67–positive cells present in the tumor through immunohistochemistry. This is often used as a marker of poor outcomes, and as a surrogate for the proliferative signature in MCL that can be incorporated into clinical practice (as opposed to gene expression profiling). Statistically significant differences in OS have emerged between groups of MCL patients with Ki-67–positive cells comprising less than 30% of their tumor sample (favorable) and those with Ki-67–positive cells comprising 30% or greater (unfavorable).¹⁰

Recent data has also identified the importance of the transcription factor SOX 11 (SRY-related HMG-box), which regulates multiple cellular transcriptional events, including cell proliferation and differentiation, apoptosis, and angiogenesis.¹¹ MCL expressing SOX 11 behaves more aggressively than MCL variants lacking SOX 11 expression, and tends to accumulate more genetic alterations.¹² Moreover, lack of SOX 11 expression characterizes a subset of MCL that does not carry the t(11;14) translocation.

DIAGNOSIS AND STAGING

CASE PRESENTATION

A 62-year-old man with a history of diabetes mellitus and hypertension presents with cervical lymphadenopathy, fatigue, and early satiety over the past several months. He is otherwise in good health. His Eastern Cooperative Oncology Group (ECOG) performance status is 1. On physical examination, 3-cm lymphadenopathy in the bilateral cervical chain is noted. Bilateral axillary lymph nodes measure 2 to 4 cm. His spleen is enlarged and is palpable at approximately 5 cm below the costal margin. A complete blood count reveals a total white blood cell (WBC) count of 14,000 cells/μL, with 68% lymphocytes and a normal distribution of neutrophils. Hemoglobin is 11 g/dL, and platelet count is 112,000/μL. The lactate dehydrogenase (LDH) level is 322 U/L (upper limit of normal: 225 U/L).

• How is MCL diagnosed?

Diagnosis of MCL requires review by expert hematopathologists.¹³ Whenever possible, an excisional biopsy should be performed for the adequate characterization of lymph node architecture and evaluation by immunohistochemistry. Aside from the characteristic expression of CD5 and CD20 and absence of CD23, MCL should express cyclin D1, which reflects t(11;14). If cyclin D1 is inconclusive or unavailable, fluorescent in situ hybridization (FISH) for t(11;14) should be performed.⁸ Patients often have circulating malignant lymphocytes, or leukemic phase MCL. Flow cytometry of the peripheral blood can detect traditional surface markers, and FISH can also be performed on circulating abnormal lymphocytes.

RISK STRATIFICATION

At diagnosis, patients should undergo risk stratification in order to understand prognosis and possibly guide treatment. In MCL, the MCL international prognostic index (MIPI) is used. The MIPI is a prognostic tool developed exclusively for patients with MCL using data from 455 patients with advanced-stage MCL treated within 3 European clinical trials.¹⁶ The MIPI classified patients into risk groups based on age, ECOG performance status, LDH level, and WBC count. Patients were categorized into low-risk (44% of patients, median OS not reached), intermediate-risk (35%, median OS 51 months), and high-risk groups (21%, median OS 29 months). This is done through a logarithmic calculation, which can be accessed through online calculators (a prototype example can be found at www.qxmd.com/calculate-online/hematology/prognosis-mantle-cell-lymphoma-mipi). Cell proliferation using the Ki-67 index was evaluated in an exploratory analysis (the biologic [“B”] MIPI), and also demonstrated strong prognostic relevance.¹⁶ Currently, treatment of MCL patients is not stratified by MIPI outside of a clinical trial, but this useful tool assists in assessing patient prognosis and has been validated for use with both conventional chemoimmunotherapy and in the setting of autologous stem cell transplant (autoSCT).^16,17 At this point in time, the MIPI score is not used to stratify treatment, although some clinical trials are incorporating the use of the MIPI score at diagnosis. Nonetheless, given its prognostic importance, the MIPI should be performed for all MCL patients undergoing staging and evaluation for treatment to establish disease risk.

As noted, the proliferative signature, represented by the Ki-67 protein, is also highly prognostic in MCL. Ki-67 is expressed in the late G1, S, G2, and M phases of the cell cycle. The Ki-67 index is defined by the hematopathologist as the percentage of lymphoma cells staining positive for Ki-67 protein, based on the number of cells per high-power field. There is significant interobserver variability in this process, which can be minimized by assessing Ki-67 quantitatively using computer software. The prognostic significance of Ki-67 at diagnosis was established in large studies of MCL patient cohorts, with survival differing by up to 3 years.^18,19 Determann et al demonstrated the utility of the proliferative index in patients with MCL treated with standard chemoimmunotherapy.¹⁰ In this study, 249 patients with advanced-stage MCL treated within randomized trials conducted by the European MCL Network were analyzed. The Ki-67 index was found to be extremely prognostic of OS, independent of other clinical risk factors, including the MIPI score. As a continuous variable, Ki-67 indices of greater than 10% correlated with poor outcomes. The Ki-67 index has also been confirmed as prognostic in relapsed MCL.²⁰ It is important to note that, as a unique feature, the Ki-67 index has remained an independent prognostic factor, even when incorporated into the “B” MIPI.

TREATMENT

CASE CONTINUED

The patient undergoes an excisional biopsy of a cervical lymph node, which demonstrates an abnormal proliferation of small-medium–sized lymphocytes with slightly irregular nuclear contours. Immunohistochemistry shows that the abnormal lymphocytes are positive for CD20 and CD5, negative for CD10 and CD23, and diffusely positive for cyclin D1, consistent with a diagnosis of MCL. The proliferative index, as measured by the Ki-67 immunostain, is 40%. A bone marrow aspirate and biopsy are then obtained, which show a clonal population of B lymphocytes expressing the same immunophenotype as the lymph node (positive for CD20 and CD5, negative for CD10 and CD23, cyclin D1 positive). A CT scan of the neck, chest, abdomen, and pelvis with contrast is obtained, along with a PET scan. These studies identify extensive hypermetabolic lymphadenopathy in the bilateral cervical chains, supraclavicular areas, mediastinum, and hilum. Mesenteric lymph nodes are also enlarged and hypermetabolic, as are retroperitoneal lymph nodes. The spleen is noted to be enlarged with multiple hypermetabolic lesions. Based on the presence of extensive lymphadenopathy as well as bone marrow involvement, the patient is diagnosed with stage IV MCL. He undergoes risk-stratification with the MIPI. His MIPI score is 6.3, high risk.

FRONTLINE THERAPY

Role of Watchful Waiting

A small proportion of MCL patients have indolent disease that can be observed. This population is more likely to have leukemic-phase MCL with circulating lymphocytes, splenomegaly, and bone marrow involvement and absent or minimal lymphadenopathy.^4,5 A retrospective study of 97 patients established that deferment of initial therapy in MCL is acceptable in some patients.⁵ In this study, approximately one third of patients with MCL were observed for more than 3 months before initiating systemic therapy, and the median time to treatment for the observation group was 12 months. Most patients undergoing observation had a low-risk MIPI. Patients were not harmed by observation, as no OS differences were observed among groups. This study underscores that deferred treatment can be an acceptable alternative in selected MCL patients for a short period of time. In practice, the type of patient who would be appropriate for this approach is someone who is frail, elderly, and with multiple comorbidities. Additionally, expectant observation could be considered for patients with limited-stage or low-volume MCL, low Ki-67 index, and low-risk MIPI scores.

Approach to Therapy

Treatment of MCL is generally approached by evaluating patient age and fitness for treatment. While there is no accepted standard, for younger patients healthy enough to tolerate aggressive approaches, treatment often involves an intensive cytarabine-containing regimen, which is consolidated with an autoSCT. This approach results in the longest remission duration, with some series suggesting a plateau in survival after 5 years, with no relapses.²¹ Nonintensive conventional chemotherapy alone is often reserved for the frailer or older patient. Given that remission durations with chemotherapy alone in MCL are short, goals of treatment focus on maximizing benefit and remission duration and minimizing risk of toxicity.

Standard Chemotherapy: Elderly and/or Frail Patients

Conventional chemotherapy alone for the treatment of MCL results in a 70% to 85% overall response rate (ORR) and 7% to 30% complete response (CR) rate.²² Rituximab, a mouse humanized monoclonal IgG1 anti-CD20 antibody, is used as standard of care in combination with chemotherapy, since its addition has been found to increase response rates and extend both progression-free survival (PFS) and OS compared to chemotherapy alone.^23,24 However, chemoimmunotherapy approaches do not provide long-term control of MCL and are considered noncurative. Various regimens have been studied and include anthracycline-containing regimens such as R-CHOP (rituximab with cyclophosphamide, doxorubicin, vincristine, prednisone),²² combination chemotherapy with antimetabolites such as R-hyper-CVAD (hyper-fractionated rituximab with cyclophosphamide, vincristine, doxorubicin, dexamethasone, alternating with methotrexate and cytarabine),²⁵ purine analogue–based regimens such as R-FC (rituximab with fludarabine and cyclophosphamide),²⁶ bortezomib-containing regimens,²⁷ and alkylator-based treatment with BR (bendamustine and rituximab) (Table 1).^28,29 Among these, the most commonly used are R-CHOP and BR.

Two large randomized studies compared R-CHOP for 6 cycles to BR for 6 cycles in patients with indolent NHL and MCL. Among MCL patients, BR resulted in superior PFS compared to R-CHOP (69 months versus 26 months) but no benefit in OS.^28,29 The ORR to R-CHOP was approximately 90%, with a PFS of 21 months in the Rummel et al study.²⁹ This study included more than 80 centers in Germany and enrolled 549 patients with MCL, follicular lymphoma, small lymphocytic lymphoma, marginal zone lymphoma, and Waldenström macroglobulinemia. Patients were randomized in a 1:1 fashion. Among these, 46 patients received BR and 48 received R-CHOP (18% for both, respectively). It should be noted that patients in the BR group had significantly less toxicity and experienced fewer side effects than did those in the R-CHOP group. Similarly, BR-treated patients had a lower frequency of hematologic side effects and infections of any grade. However, drug-associated skin reactions and allergies were more common with BR compared to R-CHOP. The study by Flinn and colleagues was an international randomized, noninferiority phase 3 study designed to evaluate the efficacy and safety of BR compared with R-CHOP or R-CVP (rituximab plus cyclophosphamide, vincristine, and prednisone) for treatment-naive patients with MCL or other indolent NHL. The primary endpoint was CR. In this study, BR was found to be noninferior to R-CHOP and R-CVP based on CR rate (31% versus 25%, respectively; P = 0.0225). Response rates in general were high: 97% for BR and 91% for R-CHOP/R-CVP (P = 0.0102). Here, BR-treated patients experienced more nausea, emesis, and drug-induced hypersensitivity compared to the R-CHOP and R-CVP groups.

Bendamustine in combination with other agents has also been studied in the frontline setting. Visco and colleagues evaluated the combination of bendamustine with rituximab and cytarabine (R-BAC) in older patients with MCL (age 65 or older).⁶³ This phase 2, two-stage study enrolled 40 patients and had a dose-finding arm for cytarabine in combination with BR. It permitted relapsed/refractory patients, but 50% had newly diagnosed, previously untreated MCL. The regimen had an impressive ORR of 100%, with CR rates of 95% for previously untreated patients. PFS at 2 years was 95%. R-BAC was well tolerated, with the primary toxicity being reversible myelosuppression.

BR was combined with the proteasome inhibitor bortezomib and dexamethasone in a phase 2 study.⁶⁴ This Lymphoma Study Association (LYSA) study evaluated 76 patients with newly diagnosed MCL older than age 65 years. BR was administered in standard doses (bendamustine 90 mg/m² on days 1 and 2 and rituximab 375 mg/m² IV on day 1) and bortezomib was administered subcutaneously on days 1, 4, 8, and 11, with acyclovir for viral prophylaxis. Patients received 6 cycles. The ORR was 87% and the CR was 60%. Patients experienced toxicity, and not all bortezomib doses were administered due to neurotoxic or hematologic side effects.

A randomized phase 3 study compared R-CHOP to the VR-CAP regimen (R-CHOP regimen but bortezomib replaces vincristine on days 1, 4, 8, 11, at 1.3 mg/m²) in 487 newly diagnosed MCL patients.²⁷ Median PFS was superior in the VR-CAP group compared with R-CHOP (14.4 months versus 24.7 months, respectively). Additionally, rates of CR were superior in the VR-CAP group (53% compared to 42% with R-CHOP). However, there was more hematologic toxicity with VR-CAP. On the basis of these findings, the U.S. Food and Drug Administration approved bortezomib for the frontline treatment of MCL.

Other chemoimmunotherapy combinations containing bortezomib have been studied in frontline MCL treatment, with promising results. These include bortezomib in combination with R-CHOP or modified R-hyper-CVAD, as well as bortezomib in combination with CHOP-like treatments and purine analogues.^27,30–32 The ongoing ECOG 1411 study is currently evaluating bortezomib added to BR for induction therapy of newly diagnosed MCL in a 4-arm randomized trial. Patients receive BR with or without bortezomib during induction and are then randomly assigned to maintenance with either rituximab alone or rituximab with lenalidomide. Other novel combination agents are actively being studied in frontline MCL treatment, including lenalidomide and rituximab and BR with lenalidomide.

Intensification of Therapy and AutoSCT: Fitter and/or Younger Patients

Short response duration has created the need for post-remission therapy in MCL. One approach to improve remission duration in MCL is to intensify induction through the use of cytarabine-containing regimens and/or consolidation with high-dose chemotherapy, typically using BEAM (carmustine, etoposide, cytarabine, melphalan) and autoSCT (Table 2). The cytarabine-containing R-hyper-CVAD regimen, developed at the MD Anderson Cancer Center, resulted in a 97% ORR and an 87% CR rate, with TTF of nearly 5 years. However, nearly one third of patients were unable to complete treatment due to toxicity, and 5 patients developed secondary myelodysplastic syndrome or acute myeloid leukemia.³³ The feasibility of this R-hyper-CVAD regimen was tested in a multicenter cooperative group setting, but similar results were not seen; in this study, nearly 40% of patients were unable to complete the full scheduled course of treatment due to toxicity.³⁴

​Other ways to intensify therapy in MCL involve adding a second non-cross-resistant cytarabine-containing regimen to R-CHOP after remission, such as DHAP (dexamethasone, high-dose cytarabine, cisplatin), followed by consolidation with an autoSCT. A retrospective registry from the National Comprehensive Cancer Network sought to compare the efficacy of different treatment approaches in the frontline setting. They studied 167 patients with MCL and compared 4 groups: treatment with R-hyper-CVAD, either with or without autoSCT, and treatment with R-CHOP, either with or without autoSCT. This study found that in patients younger than 65, R-CHOP followed by autoSCT or R-hyper-CVAD without autoSCT resulted in similar PF and OS, but was superior to R-CHOP alone for newly diagnosed MCL patients.³⁵ These data support more intensive regimens in younger and fitter patients. Several other prospective and randomized studies have demonstrated clinical benefit for patients with MCL undergoing autoSCT in first remission. Of particular importance is the seminal phase 3 study of the European MCL Network, which established the role of autoSCT in this setting.⁶¹ In this prospective randomized trial involving 122 newly diagnosed MCL patients who responded to CHOP-like induction, patients in CR derived a greater benefit from autoSCT.

More recent studies have demonstrated similar benefits using cytarabine-based autoSCT. The Nordic MCL2 study evaluated 160 patients using R-CHOP, alternating with rituximab and high-dose cytarabine, followed by autoSCT. This study used “maxi-CHOP,” an augmented CHOP regimen (cyclophosphamide 1200 mg/m², doxorubicin 75 mg/m², but standard doses of vincristine [2 mg] and prednisone [100 mg days 1–5]), alternating with 4 infusions of cytarabine at 2 g/m² and standard doses of rituximab (375 mg/m²). Patients then received conditioning with BEAM and autoSCT. Patients were evaluated for the presence of minimal residual disease (MRD) and for the t(11;14) or clonal immunoglobulin heavy chain gene rearrangement with polymerase chain reaction (PCR). Patients with MRD were offered therapy with rituximab at 375 mg/m² weekly for 4 doses. This combination resulted in 10-year OS rates of 58%.³⁶ In a multicenter study involving 78 patients from the Cancer and Leukemia Group B (CALGB), R-CHOP followed by high-dose cytarabine and BEAM-based autoSCT resulted in a 5-year OS of 64%.³⁷ A single-arm phase 2 study from the Netherlands also tested R-CHOP followed by high-dose cytarabine and BEAM-based autoSCT. Nonhematologic toxicities were 22% after high-dose cytarabine, and 55% after BEAM. The ORR was 70%, with a 64% CR rate and 66% OS at 4 years.³⁸ The French GELA group used 3 cycles of R-CHOP and 3 cycles of R-DHAP in a phase 2 study of young (under age 66) MCL patients. Following R-CHOP, the ORR was 93%, and following R-DHAP the ORR was 95%. Five-year OSA was 75%.³⁹ A large randomized phase 3 study by Hermine and colleagues of the EMCLN confirmed the benefit of this approach in 497 patients with newly diagnosed MCL. R-CHOP for 6 cycles followed by autoSCT was compared to R-CHOP for 3 cycles alternating with R-DHAP for 3 cycles and autoSCT with a cytarabine-based conditioning regimen. The addition of cytarabine significantly increased rates of CR, TTF, and OS, without increasing toxicity.⁴⁰

CASE CONTINUED

The patient is treated with R-CHOP chemotherapy for 3 cycles followed by R-DHAP. His course is complicated by mild tinnitus and acute kidney injury from cisplatin that promptly resolves. Three weeks following treatment, a restaging PET/CT scan shows resolution of all lymphadenopathy, with no hypermetabolic uptake, consistent with a complete remission. A repeat bone marrow biopsy shows no involvement with MCL. He subsequently undergoes an autoSCT, and restaging CT/PET 3 months following autoSCT shows continued remission. He is monitored every 3 to 6 months over the next several years.

He has a 4.5-year disease remission, after which he develops growing palpable lymphadenopathy on exam and progressive anemia and thrombocytopenia. A bone marrow biopsy is repeated, which shows recurrent MCL. Restaging diagnostic imaging with a CT scan reveals lymphadenopathy above and below the diaphragm. An axillary lymph node biopsy also demonstrates recurrent MCL. At this time the patient is otherwise in fairly good health, except for feeling fatigued. His ECOG performance status is 1. He begins therapy with bortezomib at a dose of 1.3 mg/m² intravenously on days 1, 4, 8, and 11 for 6 cycles. His treatment course is complicated by painful sensory peripheral neuropathy of the bilateral lower extremities. Restaging studies at the completion of therapy demonstrate that he has achieved a partial response, with a 50% reduction in the size of involved lymphadenopathy and some residual areas of hypermetabolic uptake. His peripheral cytopenias improve moderately.

TREATMENT OF RELAPSED MCL

Single-Agent and Combination Chemotherapy

Whenever possible, and since there is no standard, patients with relapsed MCL should be offered a clinical trial. Outside of a clinical study, many of the treatment regimens used at diagnosis can also be applied in the relapsed setting. In relapsed MCL, Rummel et al showed that BR for 4 cycles resulted in an ORR of 90%, with a CR of 60%. The median PFS was 24 months.⁴¹ Bortezomib, an inhibitor of the proteasome-ubiquitin pathway, leads to apoptosis and cell cycle arrest in MCL.⁴² Multiple studies have evaluated bortezomib both as a single agent and in combination for patients with relapsed MCL. In 2006, bortezomib became the first agent approved by the FDA in relapsed or refractory MCL, based on the phase 2 PINNACLE study. This prospective multicenter study involving 155 patients demonstrated an ORR of 33%, CR rate of 8%, and median treatment duration of 9 months. The median time to progression was 6 months.⁴³ Subsequently, bortezomib-containing combinations evolved. In a multicenter study of relapsed and refractory indolent NHL and MCL, Friedberg and colleagues evaluated bortezomib in combination with BR.⁴⁴ In the MCL cohort, the ORR was 71%. These promising results led to the study of this combination in the frontline setting. The ongoing ECOG 1411 study is using BR for the frontline treatment of MCL with or without bortezomib as induction. This study also includes rituximab maintenance, and randomizes patients to undergo maintenance with or without the immunomodulator lenalidomide. Bortezomib has been associated with herpes simplex and herpes zoster reactivation. Neuropathy has also been observed with bortezomib, which can be attenuated by administering it subcutaneously.

Lenalidomide is an immunomodulatory agent derived from thalidomide. It has significant activity and is a mainstay of treatment in multiple myeloma. Lenalidomide acts by enhancing cellular immunity, has antiproliferative effects, and inhibits T-cell function leading to growth inhibitory effects in the tumor microenvironment.⁴⁵ In MCL, lenalidomide has demonstrated clinical activity both as a single agent and in combination, as well as in preclinical studies establishing its pro-apoptotic effects.⁴⁶ The pivotal EMERGE study evaluated monotherapy with lenalidomide in heavily pretreated relapsed and refractory MCL. This multicenter international study of 134 patents reported an ORR of 28% with a 7.5% CR rate and median PFS of 4 months. All patients had relapsed or progressed following bortezomib. This led to the approval of lenalidomide by the FDA in 2013 for the treatment of patients with MCL whose disease relapsed or progressed following 2 prior therapies, one of which included bortezomib.⁴⁷ Lenalidomide has been associated with neutropenia, secondary cancers, and deep venous thrombosis.

In combination with other agents in the relapsed setting, lenalidomide shows broader activity. A phase 1/2 study by Wang and colleagues demonstrated an ORR of 57%; the median response duration was 19 months when lenalidomide was combined with rituximab for relapsed/refractory MCL.⁴⁸

Novel Therapies

More recently, novel treatment approaches have been tested in MCL based on an increased understanding of aberrant signaling pathways in this disease (Table 3). Constitutive activation of B-cell receptor signaling is critical for the survival and proliferation of lymphomas, and has led to the development of targeted agents inhibiting B-cell receptor–associated protein kinases. Bruton’s tyrosine kinase (BTK) is one essential component of the B-cell receptor.⁴⁹ In particular, proteins upstream of the BTK pathway have been implicated in growth and proliferation of MCL, suggesting that inhibition of BTK may impede lymphomagenesis.⁵⁰ Ibrutinib is an oral inhibitor of BTK, and demonstrates activity in multiple lymphoma subtypes. In a phase 1 study of ibrutinib in relapsed and refractory hematologic malignancies, an ORR of 60% was observed in 50 evaluable patients, with 16% CR. Median PFS was 13 months. Among these, 7 of 9 patients with MCL responded, including 3 CRs.⁵¹ Given these promising results, a phase 2 multicenter study evaluating ibrutinib in relapsed and refractory MCL was completed.⁵² At a dose of 560 mg daily, the response rate was 68%, with CR of 21%. The most common observed treatment-related side effects included diarrhea, fatigue, and nausea. Neutropenia and thrombocytopenia were also observed. Of importance, 5% of patients had grade 3 or higher bleeding events, including subdural hematoma, gastrointestinal bleeding, and hematuria. The estimated OS rate was 58% at 18 months. On the basis of this study, the FDA approved ibrutinib for relapsed and refractory MCL in November 2013.

The PI3K pathway is another survival pathway that is dysfunctional in several hematologic disorders, including MCL. Overexpression of PI3K and its downstream targets contributes to MCL pathogenesis.⁵³ Idelalisib is an oral small molecule inhibitor of the delta isoform of PI3K that is dosed daily; it was approved by the FDA for the treatment of relapsed and refractory follicular lymphoma, small lymphocytic lymphoma, and chronic lymphocytic leukemia. It is being further evaluated in MCL. A dose-escalation phase 1 study in heavily pre-treated MCL patients established safety and tolerability.⁵⁴ Efficacy analysis showed an ORR of 40%, CR of 5%, and 1-year OS of 22%. Further phase 2 studies testing idelalisib as a single agent and in combination for MCL are ongoing. Side effects of idelalisib include elevated liver enzymes, pneumonitis, and diarrhea.

The BCL family of proteins is involved in both pro-and anti-apoptotic functions. BCL2 is an intracellular protein that blocks apoptosis. ABT-199 is an oral BCL2 inhibitor that in early clinical trials has shown very promising activity in MCL. In a phase 1 study of 31 relapsed and refractory NHL patients, all 8 MCL patients (100% ORR) responded to ABT-199 therapy.⁵⁵ Given these promising initial results, multiple studies evaluating ABT-199 are ongoing in MCL as part of first-line treatment as well as for relapsed disease. ABT-199 has been implicated in tumor lysis syndrome, and in early studies of chronic lymphocytic leukemia, fatal tumor lysis was observed.

The mammalian target of rapamycin (mTOR) inhibitor temsirolimus has been evaluated in relapsed MCL. It is given weekly at 250 mg intravenously. Response rates to single-agent temsirolimus are approximately 20% to 35%, and are higher when combined with rituximab.^56,57 The phase 2 study evaluating temsirolimus as a single agent enrolled 35 heavily pre-treated patients. ORR was 38% with only 1 CR. The duration of response was 7 months. Temsirolimus is approved for relapsed MCL in Europe but not in the United States. Similar to the other targeted agents, temsirolimus is actively being studied in combination with other active agents in MCL. Adverse effects noted with temsirolimus include diarrhea, stomatitis, and rash. Thrombocytopenia requiring dose reductions is another frequently observed complication.

Radioimmunotherapy

Radioimmunotherapy (RIT) has been studied extensively in MCL. RIT consists of anti-CD 20 antibodies coupled to radioactive particles that deliver radiation to targeted cells, minimizing toxicity to surrounding tissues. RIT is not used as frequently in the modern era as it had been in the past. At this time, only yttrium-90-ibritumomab tiuxetan is available.

RIT has been evaluated in MCL both at the time of relapse⁵⁸ and more recently, as part of a conditioning regimen prior to autoSCT, with good tolerability.^65–67 Averse events noted with RIT include hematologic toxicity (can be prolonged), hypothyroidism, and in rare cases, myelodysplastic syndrome and acute leukemia. The bone marrow must have less than 25% involvement with disease prior to administration. Wang and colleagues evaluated yttrium-90-ibritumomab tiuxetan in 34 heavily pretreated patients with MCL.⁵⁸ They observed an ORR of 31%. The median event-free survival (EFS) was 6 months, but in patients achieving either CR or PR, EFS was 28 months. A 21-month OS was noted.

In the upfront setting, RIT has been added as a mechanism of intensification. A recent Nordic group study of RIT with autoSCT did not find benefit with the addition of RIT.⁵⁹ An ECOG study recently added yttrium-90-ibritumomab tiuxetan after CHOP chemotherapy in the upfront treatment of MCL, with good tolerability.⁵⁵ However, when added to R-hyper-CVAD, the combination had unexpected high rates of hematologic toxicity, including grade 3/4 cytopenias and an unacceptably high rate of secondary malignancies.⁶⁸

AutoSCT or Allogeneic Transplant

While many studies noted above have established the beneficial role of autoSCT in MCL in first remission, the role of allogeneic transplant (alloSCT) in MCL remains controversial. A recent large retrospective study conducted by the Center for International Blood and Marrow Transplant Research (CIBMTR) evaluated 519 patients with MCL who underwent both autoSCT and alloSCT.⁶⁰ Patients were grouped into an early cohort (transplant in first PR or CR, and 2 or fewer treatments) and late cohort (all other patients). The analysis had mature follow up. A multivariate analysis demonstrated that early autoSCT was associated with superior outcomes compared to autoSCT performed later. While it was not possible to demonstrate a survival benefit favoring autoSCT over reduced intensity (RIC) alloSCT, patients transplanted later in their disease course had shorter OS. For patients receiving autoSCT in CR 1 following only 1 prior line of therapy, OS at 5 years was 75% and PFS was 70%. Patients undergoing RIC followed by alloSCT had fewer relapses, but this was negated by higher nonrelapse mortality (25%), resulting in a PFS similar to autoSCT.

CASE CONCLUSION

After treatment with bortezomib the patient is well for 9 months. Subsequently, however, he develops increasing lymphadenopathy and progressive fatigue. He is then started on lenalidomide 25 mg orally daily for 21 out of 28 days. He experiences significant fatigue with lenalidomide and prolonged neutropenia requiring dose delays, despite dose modification to 10 mg orally daily. He requires discontinuation of lenalidomide. Given persistent disease, the patient then begins treatment with ibrutinib. Within a few days of starting ibrutinib therapy, he experiences a marked but transient leukocytosis. Two months later, the patient’s palpable lymphadenopathy has decreased, and his anemia and thrombocytopenia related to MCL are improving. He has tolerated treatment well. His course has been complicated only by a mild, pruritic maculopapular eruption on his chest, back, and arms, that was responsive to topical low-dose steroids. He remains on ibrutinib 1 year later.

CONCLUSION

Advances in our understanding of MCL treatment are revolutionizing the approach to this once deadly disease. Over the next several years, these gains will weave themselves into the current treatment paradigm and likely alter the treatment landscape for MCL as we know it.

The gut microbiomes of children with autism spectrum disorder (ASD) and functional gastrointestinal disorders (FGID) had significantly higher levels of several Clostridium species and lower concentrations of other bacteria compared with neurotypical children with and without FGIDs, which correlated with increases in inflammatory cytokines, decreased tryptophan, and increased serotonin, according to a small, single-center, cross-sectional study.

This “unique multi-omic profile [was] specific to ASD-FGID and ASD-FGID with abdominal pain,” wrote Ruth Ann Luna, PhD, of Texas Children’s Microbiome Center at Texas Children’s Hospital, Houston, and her associates. The report was published online in Cellular and Molecular Gastroenterology and Hepatology (doi: 10.1016/j.jcmgh.2016.11.008).

Children with ASD are at increased risk for FGIDs such as functional constipation, nonretentive fecal incontinence, functional abdominal pain, abdominal migraines, and irritable bowel syndrome, compared with their neurotypical peers. Changes in the gut microbiome can affect immunologic pathways and the balance between tryptophan and serotonin. This altered “microbial-gut-brain axis” has been reported in both ASD and FGID, suggesting “that altered gut-brain communications not only may play a role in the increased occurrence of FGIDs in ASD individuals, but could advance our understanding of potential risk factors for FGID in the ASD community,” the researchers wrote.

Previous studies of stool specimens have found higher levels of several species of Clostridium in pediatric ASD compared with neurotypical children. To confirm and expand on that work, the investigators examined microbial and neuroimmune markers in rectal biopsies and blood specimens from 14 children with ASD-FGID, 15 neurotypical children with FGID, and 6 asymptomatic neurotypical children. Participants were recruited from Nationwide Children’s Hospital in Columbus, Ohio. The researchers quantified microbial 16S ribosomal DNA community signatures, cytokines, chemokines, and serotonergic metabolites, and correlated results with parental responses to the Questionnaire on Pediatric Gastrointestinal Symptoms–Rome III version.

The ASD-FGID group had significantly higher numbers for ribosomal DNA sequences for Clostridium lituseburense (P = .002), Lachnoclostridium bolteae (P = .02), Lachnoclostridium hathewayi (P = .03), Clostridium aldenense (P = .04), and Oscillospira plautii (P = .04), compared with neurotypical children with and without FGID. Children with ASD-FGID also had significantly lower levels of Dorea formicigenerans (P = .006), Blautia luti (P = .02), and Sutterella species (P = .03). “Overall, our identification of clostridial species aligns with previous autism studies that have identified microbiome alterations,” the researchers noted.

They also looked specifically at abdominal pain. Children with ASD-FGID and abdominal pain had significantly higher gut mucosal levels of Turicibacter sanguinis (P = .03), Clostridium aldenense (P = .004), Clostridium lituseburense (P = .003), Oscillospira plautii (P = .01), Clostridium disporicum (P = .049), and Clostridium tertium (P = .045) than did any other subgroup, the investigators found. Patients with both ASD-FGID and abdominal pain also had significantly higher levels of C. aldenense (P = .03), O. plautii (P = .04), Tyzzerella species (P = .045), and Parasutterella excrementihominis (P = .04) than did ASD-FGID patients without abdominal pain.

Both C. disporicum and C. tertium correlated with increases in the proinflammatory cytokines IL6 and interferon-gamma. Levels of these cytokines were highest in patients with ASD-FGID, and IL6 was highest of all among children with ASD-FGID with abdominal pain. Another proinflammatory cytokine, IL17A, also correlated with Clostridia species that were enriched in children with ASD-FGID. Both IL6 and IL17A have been implicated in autism-like phenotypes in rodents, the researchers noted. Several other cytokines also were linked to ASD-FGID, and abdominal pain correlated significantly with increases in MCP-1 (P = .03) and eotaxin (P = .03).

Gut mucosal levels of tryptophan were significantly lower among children with ASD-FGID compared with neurotypical children, either with (P = .006) or without (P = .009) FGID. In contrast, gut mucosal levels of 5-HIAA, the primary metabolite of serotonin, were significantly higher among children with ASD-FGID compared with neurotypical children (P = .01). Increased 5-HIAA also correlated significantly with abdominal pain (P = .04). Six species of bacteria correlated significantly with tryptophan or serotonin, implicating the gut microbiome in the serotonin pathway.

“Although these initial findings are correlative, these data form the framework for future studies targeting tryptophan-serotonin metabolism and inflammatory pathways in FGID in ASD,” the researchers concluded.

The U.S. Department of Health and Human Services funded the work. The investigators had no relevant disclosures.

The gut microbiomes of children with autism spectrum disorder (ASD) and functional gastrointestinal disorders (FGID) had significantly higher levels of several Clostridium species and lower concentrations of other bacteria compared with neurotypical children with and without FGIDs, which correlated with increases in inflammatory cytokines, decreased tryptophan, and increased serotonin, according to a small, single-center, cross-sectional study.

This “unique multi-omic profile [was] specific to ASD-FGID and ASD-FGID with abdominal pain,” wrote Ruth Ann Luna, PhD, of Texas Children’s Microbiome Center at Texas Children’s Hospital, Houston, and her associates. The report was published online in Cellular and Molecular Gastroenterology and Hepatology (doi: 10.1016/j.jcmgh.2016.11.008).

Children with ASD are at increased risk for FGIDs such as functional constipation, nonretentive fecal incontinence, functional abdominal pain, abdominal migraines, and irritable bowel syndrome, compared with their neurotypical peers. Changes in the gut microbiome can affect immunologic pathways and the balance between tryptophan and serotonin. This altered “microbial-gut-brain axis” has been reported in both ASD and FGID, suggesting “that altered gut-brain communications not only may play a role in the increased occurrence of FGIDs in ASD individuals, but could advance our understanding of potential risk factors for FGID in the ASD community,” the researchers wrote.

Previous studies of stool specimens have found higher levels of several species of Clostridium in pediatric ASD compared with neurotypical children. To confirm and expand on that work, the investigators examined microbial and neuroimmune markers in rectal biopsies and blood specimens from 14 children with ASD-FGID, 15 neurotypical children with FGID, and 6 asymptomatic neurotypical children. Participants were recruited from Nationwide Children’s Hospital in Columbus, Ohio. The researchers quantified microbial 16S ribosomal DNA community signatures, cytokines, chemokines, and serotonergic metabolites, and correlated results with parental responses to the Questionnaire on Pediatric Gastrointestinal Symptoms–Rome III version.

The ASD-FGID group had significantly higher numbers for ribosomal DNA sequences for Clostridium lituseburense (P = .002), Lachnoclostridium bolteae (P = .02), Lachnoclostridium hathewayi (P = .03), Clostridium aldenense (P = .04), and Oscillospira plautii (P = .04), compared with neurotypical children with and without FGID. Children with ASD-FGID also had significantly lower levels of Dorea formicigenerans (P = .006), Blautia luti (P = .02), and Sutterella species (P = .03). “Overall, our identification of clostridial species aligns with previous autism studies that have identified microbiome alterations,” the researchers noted.

They also looked specifically at abdominal pain. Children with ASD-FGID and abdominal pain had significantly higher gut mucosal levels of Turicibacter sanguinis (P = .03), Clostridium aldenense (P = .004), Clostridium lituseburense (P = .003), Oscillospira plautii (P = .01), Clostridium disporicum (P = .049), and Clostridium tertium (P = .045) than did any other subgroup, the investigators found. Patients with both ASD-FGID and abdominal pain also had significantly higher levels of C. aldenense (P = .03), O. plautii (P = .04), Tyzzerella species (P = .045), and Parasutterella excrementihominis (P = .04) than did ASD-FGID patients without abdominal pain.

Both C. disporicum and C. tertium correlated with increases in the proinflammatory cytokines IL6 and interferon-gamma. Levels of these cytokines were highest in patients with ASD-FGID, and IL6 was highest of all among children with ASD-FGID with abdominal pain. Another proinflammatory cytokine, IL17A, also correlated with Clostridia species that were enriched in children with ASD-FGID. Both IL6 and IL17A have been implicated in autism-like phenotypes in rodents, the researchers noted. Several other cytokines also were linked to ASD-FGID, and abdominal pain correlated significantly with increases in MCP-1 (P = .03) and eotaxin (P = .03).

Gut mucosal levels of tryptophan were significantly lower among children with ASD-FGID compared with neurotypical children, either with (P = .006) or without (P = .009) FGID. In contrast, gut mucosal levels of 5-HIAA, the primary metabolite of serotonin, were significantly higher among children with ASD-FGID compared with neurotypical children (P = .01). Increased 5-HIAA also correlated significantly with abdominal pain (P = .04). Six species of bacteria correlated significantly with tryptophan or serotonin, implicating the gut microbiome in the serotonin pathway.

“Although these initial findings are correlative, these data form the framework for future studies targeting tryptophan-serotonin metabolism and inflammatory pathways in FGID in ASD,” the researchers concluded.

The U.S. Department of Health and Human Services funded the work. The investigators had no relevant disclosures.

The gut microbiomes of children with autism spectrum disorder (ASD) and functional gastrointestinal disorders (FGID) had significantly higher levels of several Clostridium species and lower concentrations of other bacteria compared with neurotypical children with and without FGIDs, which correlated with increases in inflammatory cytokines, decreased tryptophan, and increased serotonin, according to a small, single-center, cross-sectional study.

This “unique multi-omic profile [was] specific to ASD-FGID and ASD-FGID with abdominal pain,” wrote Ruth Ann Luna, PhD, of Texas Children’s Microbiome Center at Texas Children’s Hospital, Houston, and her associates. The report was published online in Cellular and Molecular Gastroenterology and Hepatology (doi: 10.1016/j.jcmgh.2016.11.008).

Children with ASD are at increased risk for FGIDs such as functional constipation, nonretentive fecal incontinence, functional abdominal pain, abdominal migraines, and irritable bowel syndrome, compared with their neurotypical peers. Changes in the gut microbiome can affect immunologic pathways and the balance between tryptophan and serotonin. This altered “microbial-gut-brain axis” has been reported in both ASD and FGID, suggesting “that altered gut-brain communications not only may play a role in the increased occurrence of FGIDs in ASD individuals, but could advance our understanding of potential risk factors for FGID in the ASD community,” the researchers wrote.

Previous studies of stool specimens have found higher levels of several species of Clostridium in pediatric ASD compared with neurotypical children. To confirm and expand on that work, the investigators examined microbial and neuroimmune markers in rectal biopsies and blood specimens from 14 children with ASD-FGID, 15 neurotypical children with FGID, and 6 asymptomatic neurotypical children. Participants were recruited from Nationwide Children’s Hospital in Columbus, Ohio. The researchers quantified microbial 16S ribosomal DNA community signatures, cytokines, chemokines, and serotonergic metabolites, and correlated results with parental responses to the Questionnaire on Pediatric Gastrointestinal Symptoms–Rome III version.

The ASD-FGID group had significantly higher numbers for ribosomal DNA sequences for Clostridium lituseburense (P = .002), Lachnoclostridium bolteae (P = .02), Lachnoclostridium hathewayi (P = .03), Clostridium aldenense (P = .04), and Oscillospira plautii (P = .04), compared with neurotypical children with and without FGID. Children with ASD-FGID also had significantly lower levels of Dorea formicigenerans (P = .006), Blautia luti (P = .02), and Sutterella species (P = .03). “Overall, our identification of clostridial species aligns with previous autism studies that have identified microbiome alterations,” the researchers noted.

They also looked specifically at abdominal pain. Children with ASD-FGID and abdominal pain had significantly higher gut mucosal levels of Turicibacter sanguinis (P = .03), Clostridium aldenense (P = .004), Clostridium lituseburense (P = .003), Oscillospira plautii (P = .01), Clostridium disporicum (P = .049), and Clostridium tertium (P = .045) than did any other subgroup, the investigators found. Patients with both ASD-FGID and abdominal pain also had significantly higher levels of C. aldenense (P = .03), O. plautii (P = .04), Tyzzerella species (P = .045), and Parasutterella excrementihominis (P = .04) than did ASD-FGID patients without abdominal pain.

Both C. disporicum and C. tertium correlated with increases in the proinflammatory cytokines IL6 and interferon-gamma. Levels of these cytokines were highest in patients with ASD-FGID, and IL6 was highest of all among children with ASD-FGID with abdominal pain. Another proinflammatory cytokine, IL17A, also correlated with Clostridia species that were enriched in children with ASD-FGID. Both IL6 and IL17A have been implicated in autism-like phenotypes in rodents, the researchers noted. Several other cytokines also were linked to ASD-FGID, and abdominal pain correlated significantly with increases in MCP-1 (P = .03) and eotaxin (P = .03).

Gut mucosal levels of tryptophan were significantly lower among children with ASD-FGID compared with neurotypical children, either with (P = .006) or without (P = .009) FGID. In contrast, gut mucosal levels of 5-HIAA, the primary metabolite of serotonin, were significantly higher among children with ASD-FGID compared with neurotypical children (P = .01). Increased 5-HIAA also correlated significantly with abdominal pain (P = .04). Six species of bacteria correlated significantly with tryptophan or serotonin, implicating the gut microbiome in the serotonin pathway.

“Although these initial findings are correlative, these data form the framework for future studies targeting tryptophan-serotonin metabolism and inflammatory pathways in FGID in ASD,” the researchers concluded.

The U.S. Department of Health and Human Services funded the work. The investigators had no relevant disclosures.

There is strong evidence linking adiposity to esophageal adenocarcinoma, multiple myeloma, and cancer of the colon, rectum, biliary tract, pancreas, endometrium, kidney, and postmenopausal breast, according to the authors of an umbrella review published in the Feb. 28 edition of the BMJ.

“Several meta-analyses support the link between obesity and cancer, but substantial heterogeneity exists between studies,” wrote Maria Kyrgiou, MD, of Imperial College London and her coauthors. “The reported associations may be causal, but they may also be flawed, as inherent study biases such as residual confounding and selective reporting of positive results may exaggerate the effect of obesity on cancer.”

In this umbrella review, researchers analyzed 49 papers that included a total of 204 meta-analyses, which in turn summarized 2,179 individual study estimates from 507 unique cohort or case-control studies.

okeyphotos/iStockphoto

There is strong evidence linking adiposity to esophageal adenocarcinoma, multiple myeloma, and cancer of the colon, rectum, biliary tract, pancreas, endometrium, kidney, and postmenopausal breast, according to the authors of an umbrella review published in the Feb. 28 edition of the BMJ.

“Several meta-analyses support the link between obesity and cancer, but substantial heterogeneity exists between studies,” wrote Maria Kyrgiou, MD, of Imperial College London and her coauthors. “The reported associations may be causal, but they may also be flawed, as inherent study biases such as residual confounding and selective reporting of positive results may exaggerate the effect of obesity on cancer.”

In this umbrella review, researchers analyzed 49 papers that included a total of 204 meta-analyses, which in turn summarized 2,179 individual study estimates from 507 unique cohort or case-control studies.

okeyphotos/iStockphoto

There is strong evidence linking adiposity to esophageal adenocarcinoma, multiple myeloma, and cancer of the colon, rectum, biliary tract, pancreas, endometrium, kidney, and postmenopausal breast, according to the authors of an umbrella review published in the Feb. 28 edition of the BMJ.

“Several meta-analyses support the link between obesity and cancer, but substantial heterogeneity exists between studies,” wrote Maria Kyrgiou, MD, of Imperial College London and her coauthors. “The reported associations may be causal, but they may also be flawed, as inherent study biases such as residual confounding and selective reporting of positive results may exaggerate the effect of obesity on cancer.”

In this umbrella review, researchers analyzed 49 papers that included a total of 204 meta-analyses, which in turn summarized 2,179 individual study estimates from 507 unique cohort or case-control studies.

okeyphotos/iStockphoto

ORLANDO – In the largest assessment to date of circulating tumor DNA (ctDNA) in patients with metastatic renal cell carcinoma (mRCC), the majority of patients were found to have clinically relevant genomic alterations.

The most frequently occurring alterations for the entire cohort were TP53, VHL, NF1, EGFR, and ARID1A, but, importantly, the genetic profiles differed between patients receiving first-line therapy and those receiving second-line treatments.

“Compared to patients receiving first-line therapy, patients receiving post–first-line agents had increased genomic alterations in TP53, NF1, EGFR, and PIK3CA,” said lead study author Sumanta K. Pal, MD, a urologic oncologist at City of Hope, Duarte, Calif.

“These alterations underscore potential mechanisms of resistance,” said Dr. Pal, who presented the findings of his study in a press briefing held at the 2017 Genitourinary Cancers Symposium sponsored by the American Society of Clinical Oncology, ASTRO, and the Society of Urologic Oncology..

Several targeted therapies have been approved for mRCC, including vascular endothelial growth factor–targeted therapies, mammalian target of rapamycin inhibitors, and checkpoint inhibitors. However, treatment of mRCC is generally distinctly different for the first- and second-line settings.

Dr. Pal noted that, while “efforts such as the TCGA [The Cancer Genome Atlas] have shed some light on the tumor biology, it is important to keep in mind that these datasets reflect earlier stages of disease. Certainly, there may be an evolution of tumor biology as patients progress toward metastasis.”

Circulating tumor markers represent a practical means of serially assessing tumor biology, and ctDNA can account for tumor heterogeneity. In this study, the authors sought to determine the mutational landscape of mRCC as well as to assess changes across patients receiving first-line and subsequent therapies by using ctDNA.

Data were obtained from 224 patients who received ctDNA profiling at progression as part of routine clinical care using Guardant360, a CLIA-certified comprehensive plasma assay that evaluated 70 genes. Of this group, 64 and 56 patients were coded as receiving frontline and post–first-line agents, respectively.

Genomic alterations were pooled for the entire group, and first and second (subsequent) therapies were compared, based on conventional practice patterns (first-line regimens included sunitinib, pazopanib and bevacizumab, and second line included everolimus, axitinib, cabozantinib, and nivolumab).

Genomic alterations were found in 78.6% of patients, with an average of 3.3 genomic alterations per patient. For patients receiving first-line therapy, the average number of ctDNA alterations was 2.9, compared with 3.7 for those in the cohort who were receiving second-line therapy. The median (range) ctDNA variant allele fractions were 0.23 (0.05-9.92) in the first-line group and 0.24 (0.04-47.14) in second line.

The authors observed that there were disparities in genomic alterations between both patient cohorts, with the highest disparity seen in (second vs. first line) TP53 (49% vs. 25%), VHL (29% vs. 25%), NF1 (20% vs. 15%), EGFR (17% vs. 21%), and PIK3CA (17% vs. 8%).

“These alterations underscore potential mechanisms of resistance,” said Dr. Pal.

He also pointed out that there were significant differences between the current dataset and other published reports, which may reflect the advanced state of the disease of the patients in this study.

Efforts are also ongoing to add detailed data on demographics and clinical outcomes to the current dataset, Dr. Pal added.

Acting as the paper’s discussant, Primo N. Lara Jr., MD, of the University of California, Davis, Comprehensive Cancer Center pointed out that as there are no validated biomarkers of drug resistance or tumor evolution and that liquid biopsy offers a potential platform.

The use of ctDNA is a “convenient technology that offers new means to assess RCC biology,” said Dr. Lara, but the caveat is that it is “still in its infancy and has no immediate clinical application.”

The current study is “hypothesis generating only.” ctDNA changes need to be related to outcome following treatment, and the functional role of genomic alterations in RCC biology must be validated, he said.

ORLANDO – In the largest assessment to date of circulating tumor DNA (ctDNA) in patients with metastatic renal cell carcinoma (mRCC), the majority of patients were found to have clinically relevant genomic alterations.

The most frequently occurring alterations for the entire cohort were TP53, VHL, NF1, EGFR, and ARID1A, but, importantly, the genetic profiles differed between patients receiving first-line therapy and those receiving second-line treatments.

“Compared to patients receiving first-line therapy, patients receiving post–first-line agents had increased genomic alterations in TP53, NF1, EGFR, and PIK3CA,” said lead study author Sumanta K. Pal, MD, a urologic oncologist at City of Hope, Duarte, Calif.

“These alterations underscore potential mechanisms of resistance,” said Dr. Pal, who presented the findings of his study in a press briefing held at the 2017 Genitourinary Cancers Symposium sponsored by the American Society of Clinical Oncology, ASTRO, and the Society of Urologic Oncology..

Several targeted therapies have been approved for mRCC, including vascular endothelial growth factor–targeted therapies, mammalian target of rapamycin inhibitors, and checkpoint inhibitors. However, treatment of mRCC is generally distinctly different for the first- and second-line settings.

Dr. Pal noted that, while “efforts such as the TCGA [The Cancer Genome Atlas] have shed some light on the tumor biology, it is important to keep in mind that these datasets reflect earlier stages of disease. Certainly, there may be an evolution of tumor biology as patients progress toward metastasis.”

Circulating tumor markers represent a practical means of serially assessing tumor biology, and ctDNA can account for tumor heterogeneity. In this study, the authors sought to determine the mutational landscape of mRCC as well as to assess changes across patients receiving first-line and subsequent therapies by using ctDNA.

Data were obtained from 224 patients who received ctDNA profiling at progression as part of routine clinical care using Guardant360, a CLIA-certified comprehensive plasma assay that evaluated 70 genes. Of this group, 64 and 56 patients were coded as receiving frontline and post–first-line agents, respectively.

Genomic alterations were pooled for the entire group, and first and second (subsequent) therapies were compared, based on conventional practice patterns (first-line regimens included sunitinib, pazopanib and bevacizumab, and second line included everolimus, axitinib, cabozantinib, and nivolumab).

Genomic alterations were found in 78.6% of patients, with an average of 3.3 genomic alterations per patient. For patients receiving first-line therapy, the average number of ctDNA alterations was 2.9, compared with 3.7 for those in the cohort who were receiving second-line therapy. The median (range) ctDNA variant allele fractions were 0.23 (0.05-9.92) in the first-line group and 0.24 (0.04-47.14) in second line.

The authors observed that there were disparities in genomic alterations between both patient cohorts, with the highest disparity seen in (second vs. first line) TP53 (49% vs. 25%), VHL (29% vs. 25%), NF1 (20% vs. 15%), EGFR (17% vs. 21%), and PIK3CA (17% vs. 8%).

“These alterations underscore potential mechanisms of resistance,” said Dr. Pal.

He also pointed out that there were significant differences between the current dataset and other published reports, which may reflect the advanced state of the disease of the patients in this study.

Efforts are also ongoing to add detailed data on demographics and clinical outcomes to the current dataset, Dr. Pal added.

Acting as the paper’s discussant, Primo N. Lara Jr., MD, of the University of California, Davis, Comprehensive Cancer Center pointed out that as there are no validated biomarkers of drug resistance or tumor evolution and that liquid biopsy offers a potential platform.

The use of ctDNA is a “convenient technology that offers new means to assess RCC biology,” said Dr. Lara, but the caveat is that it is “still in its infancy and has no immediate clinical application.”

The current study is “hypothesis generating only.” ctDNA changes need to be related to outcome following treatment, and the functional role of genomic alterations in RCC biology must be validated, he said.

ORLANDO – In the largest assessment to date of circulating tumor DNA (ctDNA) in patients with metastatic renal cell carcinoma (mRCC), the majority of patients were found to have clinically relevant genomic alterations.

The most frequently occurring alterations for the entire cohort were TP53, VHL, NF1, EGFR, and ARID1A, but, importantly, the genetic profiles differed between patients receiving first-line therapy and those receiving second-line treatments.

“Compared to patients receiving first-line therapy, patients receiving post–first-line agents had increased genomic alterations in TP53, NF1, EGFR, and PIK3CA,” said lead study author Sumanta K. Pal, MD, a urologic oncologist at City of Hope, Duarte, Calif.

“These alterations underscore potential mechanisms of resistance,” said Dr. Pal, who presented the findings of his study in a press briefing held at the 2017 Genitourinary Cancers Symposium sponsored by the American Society of Clinical Oncology, ASTRO, and the Society of Urologic Oncology..

Several targeted therapies have been approved for mRCC, including vascular endothelial growth factor–targeted therapies, mammalian target of rapamycin inhibitors, and checkpoint inhibitors. However, treatment of mRCC is generally distinctly different for the first- and second-line settings.

Dr. Pal noted that, while “efforts such as the TCGA [The Cancer Genome Atlas] have shed some light on the tumor biology, it is important to keep in mind that these datasets reflect earlier stages of disease. Certainly, there may be an evolution of tumor biology as patients progress toward metastasis.”

Circulating tumor markers represent a practical means of serially assessing tumor biology, and ctDNA can account for tumor heterogeneity. In this study, the authors sought to determine the mutational landscape of mRCC as well as to assess changes across patients receiving first-line and subsequent therapies by using ctDNA.

Data were obtained from 224 patients who received ctDNA profiling at progression as part of routine clinical care using Guardant360, a CLIA-certified comprehensive plasma assay that evaluated 70 genes. Of this group, 64 and 56 patients were coded as receiving frontline and post–first-line agents, respectively.

Genomic alterations were pooled for the entire group, and first and second (subsequent) therapies were compared, based on conventional practice patterns (first-line regimens included sunitinib, pazopanib and bevacizumab, and second line included everolimus, axitinib, cabozantinib, and nivolumab).

Genomic alterations were found in 78.6% of patients, with an average of 3.3 genomic alterations per patient. For patients receiving first-line therapy, the average number of ctDNA alterations was 2.9, compared with 3.7 for those in the cohort who were receiving second-line therapy. The median (range) ctDNA variant allele fractions were 0.23 (0.05-9.92) in the first-line group and 0.24 (0.04-47.14) in second line.

The authors observed that there were disparities in genomic alterations between both patient cohorts, with the highest disparity seen in (second vs. first line) TP53 (49% vs. 25%), VHL (29% vs. 25%), NF1 (20% vs. 15%), EGFR (17% vs. 21%), and PIK3CA (17% vs. 8%).

“These alterations underscore potential mechanisms of resistance,” said Dr. Pal.

He also pointed out that there were significant differences between the current dataset and other published reports, which may reflect the advanced state of the disease of the patients in this study.

Efforts are also ongoing to add detailed data on demographics and clinical outcomes to the current dataset, Dr. Pal added.

Acting as the paper’s discussant, Primo N. Lara Jr., MD, of the University of California, Davis, Comprehensive Cancer Center pointed out that as there are no validated biomarkers of drug resistance or tumor evolution and that liquid biopsy offers a potential platform.

The use of ctDNA is a “convenient technology that offers new means to assess RCC biology,” said Dr. Lara, but the caveat is that it is “still in its infancy and has no immediate clinical application.”

The current study is “hypothesis generating only.” ctDNA changes need to be related to outcome following treatment, and the functional role of genomic alterations in RCC biology must be validated, he said.

Up to 90% of reproductive women around the world describe experiencing painful menstrual periods (dysmenorrhea) at some point. Younger women struggle more than older women. Dysmenorrhea can lead to absenteeism and presenteeism to the tune of about $2 billion annually.

Dr. Ebbert is a professor of medicine and general internist at the Mayo Clinic in Rochester, Minn. and a diplomate of the American Board of Addiction Medicine. The opinions expressed are those of the author and do not necessarily represent the views and opinions of the Mayo Clinic. The opinions expressed in this article should not be used to diagnose or treat any medical condition, nor should they be used as a substitute for medical advice from a qualified, board-certified practicing clinician. Dr. Ebbert has no relevant financial disclosures about this article.

Up to 90% of reproductive women around the world describe experiencing painful menstrual periods (dysmenorrhea) at some point. Younger women struggle more than older women. Dysmenorrhea can lead to absenteeism and presenteeism to the tune of about $2 billion annually.

Dr. Ebbert is a professor of medicine and general internist at the Mayo Clinic in Rochester, Minn. and a diplomate of the American Board of Addiction Medicine. The opinions expressed are those of the author and do not necessarily represent the views and opinions of the Mayo Clinic. The opinions expressed in this article should not be used to diagnose or treat any medical condition, nor should they be used as a substitute for medical advice from a qualified, board-certified practicing clinician. Dr. Ebbert has no relevant financial disclosures about this article.

Up to 90% of reproductive women around the world describe experiencing painful menstrual periods (dysmenorrhea) at some point. Younger women struggle more than older women. Dysmenorrhea can lead to absenteeism and presenteeism to the tune of about $2 billion annually.

Dr. Ebbert is a professor of medicine and general internist at the Mayo Clinic in Rochester, Minn. and a diplomate of the American Board of Addiction Medicine. The opinions expressed are those of the author and do not necessarily represent the views and opinions of the Mayo Clinic. The opinions expressed in this article should not be used to diagnose or treat any medical condition, nor should they be used as a substitute for medical advice from a qualified, board-certified practicing clinician. Dr. Ebbert has no relevant financial disclosures about this article.

To improve patient care and outcomes, leading dermatologists offered their recommendations on adult acne products. Consideration must be given to:

• Bioclear Face Lotion and Face Cream
  Jan Marini Skin Research, Inc
  “These products contain a powerful combination of glycolic, salicylic, and azelaic acids to smooth and brighten acne-prone skin.”—Larisa Ravitskiy, MD, Gahanna, Ohio

• Neutrogena Clear Pore Cleanser/Mask
  Johnson & Johnson Consumer Inc
  “This is a good daily product for acne-prone skin. It is formulated with benzoyl peroxide and can be used as a daily wash or mask.”—Anthony M. Rossi, MD, New York, New York

• Offects Sulfur Masque Acne Treatment
  ZO Skin Health Inc
  “This nonirritating mask reduces inflammation and oiliness and is safe to use in pregnancy.”—Larisa Ravitskiy, MD, Gahanna, Ohio

• PanOxyl Acne Foaming Wash and Acne Creamy Wash
  Stiefel, a GSK company
  Recommended by Gary Goldenberg, MD, New York, New York

Cutis invites readers to send us their recommendations. Athlete’s foot treatments, as well as products for dry cuticles, hyperhidrosis, and sensitive skin will be featured in upcoming editions of Cosmetic Corner. Please e-mail your recommendation(s) to the Editorial Office.

Disclaimer: Opinions expressed herein do not necessarily reflect those of Cutis or Frontline Medical Communications Inc. and shall not be used for product endorsement purposes. Any reference made to a specific commercial product does not indicate or imply that Cutis or Frontline Medical Communications Inc. endorses, recommends, or favors the product mentioned. No guarantee is given to the effects of recommended products.

[polldaddy:9711250]

To improve patient care and outcomes, leading dermatologists offered their recommendations on adult acne products. Consideration must be given to:

• Bioclear Face Lotion and Face Cream
  Jan Marini Skin Research, Inc
  “These products contain a powerful combination of glycolic, salicylic, and azelaic acids to smooth and brighten acne-prone skin.”—Larisa Ravitskiy, MD, Gahanna, Ohio

• Neutrogena Clear Pore Cleanser/Mask
  Johnson & Johnson Consumer Inc
  “This is a good daily product for acne-prone skin. It is formulated with benzoyl peroxide and can be used as a daily wash or mask.”—Anthony M. Rossi, MD, New York, New York

• Offects Sulfur Masque Acne Treatment
  ZO Skin Health Inc
  “This nonirritating mask reduces inflammation and oiliness and is safe to use in pregnancy.”—Larisa Ravitskiy, MD, Gahanna, Ohio

• PanOxyl Acne Foaming Wash and Acne Creamy Wash
  Stiefel, a GSK company
  Recommended by Gary Goldenberg, MD, New York, New York

Cutis invites readers to send us their recommendations. Athlete’s foot treatments, as well as products for dry cuticles, hyperhidrosis, and sensitive skin will be featured in upcoming editions of Cosmetic Corner. Please e-mail your recommendation(s) to the Editorial Office.

Disclaimer: Opinions expressed herein do not necessarily reflect those of Cutis or Frontline Medical Communications Inc. and shall not be used for product endorsement purposes. Any reference made to a specific commercial product does not indicate or imply that Cutis or Frontline Medical Communications Inc. endorses, recommends, or favors the product mentioned. No guarantee is given to the effects of recommended products.

[polldaddy:9711250]

To improve patient care and outcomes, leading dermatologists offered their recommendations on adult acne products. Consideration must be given to:

• Bioclear Face Lotion and Face Cream
  Jan Marini Skin Research, Inc
  “These products contain a powerful combination of glycolic, salicylic, and azelaic acids to smooth and brighten acne-prone skin.”—Larisa Ravitskiy, MD, Gahanna, Ohio

• Neutrogena Clear Pore Cleanser/Mask
  Johnson & Johnson Consumer Inc
  “This is a good daily product for acne-prone skin. It is formulated with benzoyl peroxide and can be used as a daily wash or mask.”—Anthony M. Rossi, MD, New York, New York

• Offects Sulfur Masque Acne Treatment
  ZO Skin Health Inc
  “This nonirritating mask reduces inflammation and oiliness and is safe to use in pregnancy.”—Larisa Ravitskiy, MD, Gahanna, Ohio

• PanOxyl Acne Foaming Wash and Acne Creamy Wash
  Stiefel, a GSK company
  Recommended by Gary Goldenberg, MD, New York, New York

Cutis invites readers to send us their recommendations. Athlete’s foot treatments, as well as products for dry cuticles, hyperhidrosis, and sensitive skin will be featured in upcoming editions of Cosmetic Corner. Please e-mail your recommendation(s) to the Editorial Office.

Disclaimer: Opinions expressed herein do not necessarily reflect those of Cutis or Frontline Medical Communications Inc. and shall not be used for product endorsement purposes. Any reference made to a specific commercial product does not indicate or imply that Cutis or Frontline Medical Communications Inc. endorses, recommends, or favors the product mentioned. No guarantee is given to the effects of recommended products.

[polldaddy:9711250]

ORLANDO – Active surveillance prior to initiating targeted therapy could be an option for some patients with metastatic renal cell carcinoma (mRCC), according to new findings.

“Active surveillance does not affect the efficacy of subsequent therapies,” said lead study author Davide Bimbatti, MD, of the Azienda Ospedaliera Universitaria Integrata, University of Verona (Italy). “In selected patients, active surveillance allows us to delay the start of systemic treatment, and patients in active surveillance rarely have a worsening of prognostic class,” he said in a press briefing held at the 2017 Genitourinary Cancers Symposium sponsored by the American Society of Clinical Oncology, ASTRO, and the Society of Urologic Oncology.

Targeted therapies can improve survival in mRCC but treatment is not curative and associated toxicities can interfere with quality of life and force patients to discontinue treatment. Active surveillance delays the use of systemic therapy and associated toxicities, and is a feasible strategy for patients with indolent disease.

However, the effect of active surveillance on tumor burden and prognosis have not been investigated.

To address this issue, Dr. Bimbatti and his colleagues conducted a retrospective study and evaluated the effect of active surveillance on overall survival and postsurveillance overall survival, progression-free survival, and tumor burden (a measure of the number of sites and tumor size).

The study cohort included 48 patients with mRCC who underwent active surveillance during 2007-2016 and changes in the International mRCC Database Consortium (IMDC) prognostic class and tumor burden were analyzed for associations with these endpoints.

At baseline, 69% of patients had a favorable prognostic IMDC class, 25% were at an intermediate class, and 6% had a poor class designation. The main sites of metastases were lung (56%), lymph nodes (25%), pancreas (14%), adrenal gland (8%), CNS (8%), and bone (6%).

At a median follow-up of 37.3 months, 79.2% of patients were still alive and, at a median surveillance duration of 16.7 months, 71% of patients had begun a targeted therapy. There were a total of 17 deaths (33%).

Median progression-free survival was 16.6 months and median postsurveillance overall survival was 39.1 months.

During active surveillance, only four patients transitioned from good to intermediate IMDC prognostic class. IMDC classes overall maintained their prognostic value during active surveillance, Dr. Bimbatti said. IMDC class was also the only factor that was associated with time on surveillance.

At baseline, tumor burden was in one site for 65% of patients, two sites in 31%, and three or more sites for 4%, but during active surveillance, changes occurred in one site in 35% of patients, two in 48%, and more than two in 17%.

A change in tumor burden (greater than or equal to 2.2 times the original burden), however, was related to poorer postsurveillance survival (hazard ratio,1.23; P less than .01), but not with overall survival (HR, 1.0; P less than .05).

Conversely, any increase in metastatic sites was associated with both significantly worse postsurveillance and overall survival (HR = 2.6, P = 0.04; and HR = 3.3, P less than 0.01).

Commenting on the study, Primo N. Lara Jr., MD, of the University of California, Davis, Comprehensive Cancer Center explained that active surveillance “remains a reasonable option for highly selected patients.”

“[The investigators] provide reassuring data that active surveillance does not alter the IMDC risk grouping,” he said, and that the results of this study overall, were similar to those of prior research.

There were some limitations to the study, in that it was retrospective and conducted at a single institution. “It also failed to account for psychosocial issues such as anxiety during the active surveillance phase,” he said.

Critical questions also remain, Dr. Lara added, such as what are the validated selection criteria for patients entering active surveillance or what is the threshold of tumor burden to warrant active surveillance discontinuation.

The funding source was not disclosed. Dr. Bimbatti and his coauthors have no disclosures. Dr. Lara reports financial ties to multiple pharmaceutical companies.

ORLANDO – Active surveillance prior to initiating targeted therapy could be an option for some patients with metastatic renal cell carcinoma (mRCC), according to new findings.

“Active surveillance does not affect the efficacy of subsequent therapies,” said lead study author Davide Bimbatti, MD, of the Azienda Ospedaliera Universitaria Integrata, University of Verona (Italy). “In selected patients, active surveillance allows us to delay the start of systemic treatment, and patients in active surveillance rarely have a worsening of prognostic class,” he said in a press briefing held at the 2017 Genitourinary Cancers Symposium sponsored by the American Society of Clinical Oncology, ASTRO, and the Society of Urologic Oncology.

Targeted therapies can improve survival in mRCC but treatment is not curative and associated toxicities can interfere with quality of life and force patients to discontinue treatment. Active surveillance delays the use of systemic therapy and associated toxicities, and is a feasible strategy for patients with indolent disease.

However, the effect of active surveillance on tumor burden and prognosis have not been investigated.

To address this issue, Dr. Bimbatti and his colleagues conducted a retrospective study and evaluated the effect of active surveillance on overall survival and postsurveillance overall survival, progression-free survival, and tumor burden (a measure of the number of sites and tumor size).

The study cohort included 48 patients with mRCC who underwent active surveillance during 2007-2016 and changes in the International mRCC Database Consortium (IMDC) prognostic class and tumor burden were analyzed for associations with these endpoints.

At baseline, 69% of patients had a favorable prognostic IMDC class, 25% were at an intermediate class, and 6% had a poor class designation. The main sites of metastases were lung (56%), lymph nodes (25%), pancreas (14%), adrenal gland (8%), CNS (8%), and bone (6%).

At a median follow-up of 37.3 months, 79.2% of patients were still alive and, at a median surveillance duration of 16.7 months, 71% of patients had begun a targeted therapy. There were a total of 17 deaths (33%).

Median progression-free survival was 16.6 months and median postsurveillance overall survival was 39.1 months.

During active surveillance, only four patients transitioned from good to intermediate IMDC prognostic class. IMDC classes overall maintained their prognostic value during active surveillance, Dr. Bimbatti said. IMDC class was also the only factor that was associated with time on surveillance.

At baseline, tumor burden was in one site for 65% of patients, two sites in 31%, and three or more sites for 4%, but during active surveillance, changes occurred in one site in 35% of patients, two in 48%, and more than two in 17%.

A change in tumor burden (greater than or equal to 2.2 times the original burden), however, was related to poorer postsurveillance survival (hazard ratio,1.23; P less than .01), but not with overall survival (HR, 1.0; P less than .05).

Conversely, any increase in metastatic sites was associated with both significantly worse postsurveillance and overall survival (HR = 2.6, P = 0.04; and HR = 3.3, P less than 0.01).

Commenting on the study, Primo N. Lara Jr., MD, of the University of California, Davis, Comprehensive Cancer Center explained that active surveillance “remains a reasonable option for highly selected patients.”

“[The investigators] provide reassuring data that active surveillance does not alter the IMDC risk grouping,” he said, and that the results of this study overall, were similar to those of prior research.

There were some limitations to the study, in that it was retrospective and conducted at a single institution. “It also failed to account for psychosocial issues such as anxiety during the active surveillance phase,” he said.

Critical questions also remain, Dr. Lara added, such as what are the validated selection criteria for patients entering active surveillance or what is the threshold of tumor burden to warrant active surveillance discontinuation.

The funding source was not disclosed. Dr. Bimbatti and his coauthors have no disclosures. Dr. Lara reports financial ties to multiple pharmaceutical companies.

ORLANDO – Active surveillance prior to initiating targeted therapy could be an option for some patients with metastatic renal cell carcinoma (mRCC), according to new findings.

“Active surveillance does not affect the efficacy of subsequent therapies,” said lead study author Davide Bimbatti, MD, of the Azienda Ospedaliera Universitaria Integrata, University of Verona (Italy). “In selected patients, active surveillance allows us to delay the start of systemic treatment, and patients in active surveillance rarely have a worsening of prognostic class,” he said in a press briefing held at the 2017 Genitourinary Cancers Symposium sponsored by the American Society of Clinical Oncology, ASTRO, and the Society of Urologic Oncology.

Targeted therapies can improve survival in mRCC but treatment is not curative and associated toxicities can interfere with quality of life and force patients to discontinue treatment. Active surveillance delays the use of systemic therapy and associated toxicities, and is a feasible strategy for patients with indolent disease.

However, the effect of active surveillance on tumor burden and prognosis have not been investigated.

To address this issue, Dr. Bimbatti and his colleagues conducted a retrospective study and evaluated the effect of active surveillance on overall survival and postsurveillance overall survival, progression-free survival, and tumor burden (a measure of the number of sites and tumor size).

The study cohort included 48 patients with mRCC who underwent active surveillance during 2007-2016 and changes in the International mRCC Database Consortium (IMDC) prognostic class and tumor burden were analyzed for associations with these endpoints.

At baseline, 69% of patients had a favorable prognostic IMDC class, 25% were at an intermediate class, and 6% had a poor class designation. The main sites of metastases were lung (56%), lymph nodes (25%), pancreas (14%), adrenal gland (8%), CNS (8%), and bone (6%).

At a median follow-up of 37.3 months, 79.2% of patients were still alive and, at a median surveillance duration of 16.7 months, 71% of patients had begun a targeted therapy. There were a total of 17 deaths (33%).

Median progression-free survival was 16.6 months and median postsurveillance overall survival was 39.1 months.

During active surveillance, only four patients transitioned from good to intermediate IMDC prognostic class. IMDC classes overall maintained their prognostic value during active surveillance, Dr. Bimbatti said. IMDC class was also the only factor that was associated with time on surveillance.

At baseline, tumor burden was in one site for 65% of patients, two sites in 31%, and three or more sites for 4%, but during active surveillance, changes occurred in one site in 35% of patients, two in 48%, and more than two in 17%.

A change in tumor burden (greater than or equal to 2.2 times the original burden), however, was related to poorer postsurveillance survival (hazard ratio,1.23; P less than .01), but not with overall survival (HR, 1.0; P less than .05).

Conversely, any increase in metastatic sites was associated with both significantly worse postsurveillance and overall survival (HR = 2.6, P = 0.04; and HR = 3.3, P less than 0.01).

Commenting on the study, Primo N. Lara Jr., MD, of the University of California, Davis, Comprehensive Cancer Center explained that active surveillance “remains a reasonable option for highly selected patients.”

“[The investigators] provide reassuring data that active surveillance does not alter the IMDC risk grouping,” he said, and that the results of this study overall, were similar to those of prior research.

There were some limitations to the study, in that it was retrospective and conducted at a single institution. “It also failed to account for psychosocial issues such as anxiety during the active surveillance phase,” he said.

Critical questions also remain, Dr. Lara added, such as what are the validated selection criteria for patients entering active surveillance or what is the threshold of tumor burden to warrant active surveillance discontinuation.

The funding source was not disclosed. Dr. Bimbatti and his coauthors have no disclosures. Dr. Lara reports financial ties to multiple pharmaceutical companies.

Investment in advanced palliative care planning has the potential to improve the quality of care for vascular surgery patients, according to investigators from Oregon Health and Science University, Portland.

Dale G. Wilson, MD, and his colleagues performed a retrospective review of electronic medical records for 111 patients, who died while on the vascular surgery service at the OHSU Hospital during 2005-2014.

Almost three-quarters (73%) of patients were transitioned to palliative care; of those, 14% presented with an advanced directive, and 28% received a palliative care consultation (JAMA Surg. 2017;152[2]:183-90. doi: 10.1001/jamasurg.2016.3970).

While palliative care services are increasing in hospitals, accounting for 4% of annual hospital admissions in 2012 according to the study, they are not implemented consistently. “Many teams from various specialties care for patients at end of life; however, we still do not know what prompts end-of-life discussions,” Dr. Wilson said. “There is still no consensus on when to involve palliative services in the care of critically ill patients.”

While the decision to advise a consultation is “variable and physician dependent,” the type of treatment required may help identify when consultations are appropriate.

Of the 14 patients who did not choose comfort care, 11 (79%) required CPR. Additionally, all had to be taken to the operating room and required mechanical ventilation.

Of 81 patients who chose palliative care, 31 did so despite potential medical options. These patients were older – average age, 77 years, as compared with 68 years for patients who did not choose comfort care – with 8 of the 31 (26%) presenting an advanced directive, compared with only 7 of 83 patients (8%) for those who did not receive palliative care.

Dr. Wilson and his colleagues found that patients who chose palliative care were more likely to have received a palliative care consultation, as well: 10 of 31 patients who chose comfort care received a consultation, as opposed to 1 of 83 who chose comfort care but did not receive a consultation.

The nature of the vascular surgery service calls for early efforts to gather information regarding patients’ views on end-of-life care, Dr. Wilson said, noting that 73% of patients studied were admitted emergently and 87% underwent surgery, leaving little time for patients to express their wishes.

“Because the events associated with withdrawal of care are often not anticipated, we argue that all vascular surgical patients should have an advance directive, and perhaps, those at particular high risk should have a preoperative palliative care consultation,” Dr. Wilson wrote.

Limitations to the study included the data abstraction, which was performed by a single unblinded physician. Researchers also gathered patients’ reasons for transitioning to comfort care retrospectively.

[email protected]

Investment in advanced palliative care planning has the potential to improve the quality of care for vascular surgery patients, according to investigators from Oregon Health and Science University, Portland.

Dale G. Wilson, MD, and his colleagues performed a retrospective review of electronic medical records for 111 patients, who died while on the vascular surgery service at the OHSU Hospital during 2005-2014.

Almost three-quarters (73%) of patients were transitioned to palliative care; of those, 14% presented with an advanced directive, and 28% received a palliative care consultation (JAMA Surg. 2017;152[2]:183-90. doi: 10.1001/jamasurg.2016.3970).

While palliative care services are increasing in hospitals, accounting for 4% of annual hospital admissions in 2012 according to the study, they are not implemented consistently. “Many teams from various specialties care for patients at end of life; however, we still do not know what prompts end-of-life discussions,” Dr. Wilson said. “There is still no consensus on when to involve palliative services in the care of critically ill patients.”

While the decision to advise a consultation is “variable and physician dependent,” the type of treatment required may help identify when consultations are appropriate.

Of the 14 patients who did not choose comfort care, 11 (79%) required CPR. Additionally, all had to be taken to the operating room and required mechanical ventilation.

Of 81 patients who chose palliative care, 31 did so despite potential medical options. These patients were older – average age, 77 years, as compared with 68 years for patients who did not choose comfort care – with 8 of the 31 (26%) presenting an advanced directive, compared with only 7 of 83 patients (8%) for those who did not receive palliative care.

Dr. Wilson and his colleagues found that patients who chose palliative care were more likely to have received a palliative care consultation, as well: 10 of 31 patients who chose comfort care received a consultation, as opposed to 1 of 83 who chose comfort care but did not receive a consultation.

The nature of the vascular surgery service calls for early efforts to gather information regarding patients’ views on end-of-life care, Dr. Wilson said, noting that 73% of patients studied were admitted emergently and 87% underwent surgery, leaving little time for patients to express their wishes.

“Because the events associated with withdrawal of care are often not anticipated, we argue that all vascular surgical patients should have an advance directive, and perhaps, those at particular high risk should have a preoperative palliative care consultation,” Dr. Wilson wrote.

Limitations to the study included the data abstraction, which was performed by a single unblinded physician. Researchers also gathered patients’ reasons for transitioning to comfort care retrospectively.

[email protected]

Investment in advanced palliative care planning has the potential to improve the quality of care for vascular surgery patients, according to investigators from Oregon Health and Science University, Portland.

Dale G. Wilson, MD, and his colleagues performed a retrospective review of electronic medical records for 111 patients, who died while on the vascular surgery service at the OHSU Hospital during 2005-2014.

Almost three-quarters (73%) of patients were transitioned to palliative care; of those, 14% presented with an advanced directive, and 28% received a palliative care consultation (JAMA Surg. 2017;152[2]:183-90. doi: 10.1001/jamasurg.2016.3970).

While palliative care services are increasing in hospitals, accounting for 4% of annual hospital admissions in 2012 according to the study, they are not implemented consistently. “Many teams from various specialties care for patients at end of life; however, we still do not know what prompts end-of-life discussions,” Dr. Wilson said. “There is still no consensus on when to involve palliative services in the care of critically ill patients.”

While the decision to advise a consultation is “variable and physician dependent,” the type of treatment required may help identify when consultations are appropriate.

Of the 14 patients who did not choose comfort care, 11 (79%) required CPR. Additionally, all had to be taken to the operating room and required mechanical ventilation.

Of 81 patients who chose palliative care, 31 did so despite potential medical options. These patients were older – average age, 77 years, as compared with 68 years for patients who did not choose comfort care – with 8 of the 31 (26%) presenting an advanced directive, compared with only 7 of 83 patients (8%) for those who did not receive palliative care.

Dr. Wilson and his colleagues found that patients who chose palliative care were more likely to have received a palliative care consultation, as well: 10 of 31 patients who chose comfort care received a consultation, as opposed to 1 of 83 who chose comfort care but did not receive a consultation.

The nature of the vascular surgery service calls for early efforts to gather information regarding patients’ views on end-of-life care, Dr. Wilson said, noting that 73% of patients studied were admitted emergently and 87% underwent surgery, leaving little time for patients to express their wishes.

“Because the events associated with withdrawal of care are often not anticipated, we argue that all vascular surgical patients should have an advance directive, and perhaps, those at particular high risk should have a preoperative palliative care consultation,” Dr. Wilson wrote.

Limitations to the study included the data abstraction, which was performed by a single unblinded physician. Researchers also gathered patients’ reasons for transitioning to comfort care retrospectively.

[email protected]