User login
Moderator: Matt Kalaycio, MD1
Discussants: Michael Mauro, MD2; Michael Deininger, MD, PhD3
From Cleveland Clinic, Cleveland, OH1; Memorial Sloan Kettering Cancer Center, New York, NY2; Huntsman Cancer Institute, Salt Lake City, UT3
Address for correspondence: Matt Kalaycio, MD, Cleveland Clinic Main Campus, Mail Code R32, 9500 Euclid Avenue, Cleveland, OH 44195
E-mail: [email protected]
Biographical sketch:
Dr. Kalaycio has been published in numerous scientific publications including Bone Marrow Transplantation, Journal of Clinical Oncology, and Leukemia. He also is the editor of a book on leukemia and co-editor of a book on clinical malignant hematology. His research interests focus on testing new treatments for leukemia.
Dr. Kalaycio received his degree from West Virginia University School of Medicine in Morgantown. He completed his residency in internal medicine at Mercy Hospital of Pittsburgh and fellowships in hematology and medical oncology and bone marrow transplantation at Cleveland Clinic.
Michael Deininger, MD, PhD, is Professor and Chief of Hematology and Hematologic Malignancies for the Department of Internal Medicine and for the Huntsman Cancer Institute (HCI) at the University of Utah. He is an HCI investigator and member of the Experimental Therapeutics program. He has extensive experience treating patients with blood cancers, including chronic myeloid leukemia (CML) and myeloproliferative neoplasms, a group of blood cancers related to leukemia.
Dr
Dr. Deininger’s scientific focus is leukemia, specifically myeloproliferative neoplasms including chronic myeloid leukemia (CML). As a clinician-scientist with a translational research focus Dr. Deininger is heading an extramurally funded research laboratory that is dedicated to the study of signaling pathways, drug resistance and new molecular therapies in leukemia. Dr. Deininger’s work describing the selective effects of imatinib on CML cells provided the rationale for clinical trials that led to the approval of Gleevec as the first molecularly-based therapy for leukemia. Current work in his lab is focused on understanding the role of the bone marrow microenvironment in leukemia drug resistance, discovering novel therapeutic targets and developing more specific signal transduction inhibitors. Dr. Deininger’s work encompasses more than 170 articles in the peer-reviewed literature, including journals like Blood, Journal of Clinical Investigation and the New England Journal of Medicine. He has co-authored more than 10 book chapters, with contributions in leading textbooks such as deVita’s Principles of Oncology. He is a regular speaker at major international scientific meetings, such as the American Society of Hematology and the European Hematology Association and a peer reviewer for journals like Nature Genetics and Cancer Cell. His honors include the Alexandra Kefalides Prize for Leukemia Research and membership on the Editorial Board of Blood, the leading journal in Hematology. Dr. Deininger was named among the world's Highly Cited Researchers by Thomson Reuters in 2014.
Dr. Kalaycio: My name is Matt Kalaycio and I'm the Chairman of the Department of Hematology and Medical Oncology at the Cleveland Clinic. Today I’m joined by Drs. Mike Deininger, Division Chief of Hematology and Hematologic Malignancies at the University of Utah Huntsman Cancer Institute and Michael Mauro, Myeloproliferative Neoplasm Program Director at Memorial Sloan Kettering Cancer Center, New York, New York – together we will discuss practical issues surrounding CML diagnosis, management and treatment options including TKIs and investigational therapies.
Initial Presentation: Assessment & Treatment Options
Dr. Kalaycio: Often, the patient with chronic myelogenous leukemia (CML) gets admitted to the hospital, or an emergency consult is called, because the white count is up for a concern of leukemia. The treatment team sees the differential circulating blasts, and they worry about acute leukemia. Then the hematologist comes and needs to make a decision about whether or not to do a bone marrow biopsy.
When the patient presents in such a manner, I often see that bone marrow biopsies are not performed. I would like to start by asking where you both stand with regard to the necessity of a bone marrow biopsy at the time of diagnosis. I'll start with Dr. Deininger.
Dr. Deininger: I would strongly be in favor of a diagnostic biopsy as well as a smear. The reason is that I think this is the one and only chance to get a clear disease classification into chronic phase, accelerated phase, and blast crisis.
The third may be rare. There are occasional patients who have sheets of blasts who would not be seen on just a differential. For these patients, of course, the treatment decisions are going to be very different compared to a patient in chronic phase.
I think this is an opportunity that shouldn't be missed and I would always recommend that.
Dr. Kalaycio: Dr. Mauro.
Dr. Mauro: I couldn't concur more. With so much focus on the change in disease status from presentation to early response, I think understanding the scope of the disease at diagnosis, including the bone marrow, is essential and either revealing or reassuring. Although we focus mostly on early molecular response, when expectations go awry, not having as full a picture of the disease prior to treatment leaves you less informed about best treatment.
Making sense of accelerated phase features is a good example; the difference in outcomes between chronic phase patients and those who have morphologic features of accelerated phase—with or without cytogenetic features (clonal evolution)—compared with those with clonal evolution can only drive initial and long term treatment decisions. Without cytogenetics and morphology together, such key pieces of the puzzle are missing; it is worth it for the practitioner and patient to have all the data in hand at all times.
Dr. Kalaycio: Once the diagnosis has been made and the biopsy was not done and now you're seeing the patient having either been on hydroxyurea for a month or having had tyrosine kinase inhibitor (TKI) for a month, do you bother with the bone marrow biopsy at that point?
Dr. Deininger: Well, that is a very good question. I think we would probably still do it most of the time. I think it's very clear that the information you can gather from that is less valuable. I think one should try to make up for the omissions as much as possible. We would still go for it.
Dr. Kalaycio: Interesting. The other thing that happens a lot is a patient will be started on hydrea while you're waiting for BCR-ABL to return either by fluorescence in situ hybridization or quantitative polymerase chain reaction (PCR).
I wonder if you have your own set of internal guidelines for when hydroxyurea should be employed or not.
Dr. Mauro: If you look back at some of the original imatinib trials—the phase I trials—patients initiated TKI with higher blood counts, a median around 25,000 and up to 200,000, and responded with excellent tolerance.1 I think there's a bit of overapplication of hydroxyurea early in CML, prior to initiation of TKI, which often confounds and complicates the early myelosuppressive toxicity of TKIs. With use of more potent TKIs, there may be greater amounts of myelosuppression as the leukemic clone may clear faster.
I think we may get ourselves into a bit of a bind by overusing hydroxyurea and then needing to hold and lower TKI dosages quickly when that may not have been necessary had we simply deployed the TKI sooner.
My general rule would be to use hydroxyurea for symptoms if necessary or for more extreme counts where leukostasis is a concern. – Michael Mauro, MDMy general rule would be to use hydroxyurea for symptoms if necessary or for more extreme counts where leukostasis is a concern. The other question that often comes up is about tumor lysis and hydration, and how closely these need to be managed. The likelihood of tumor lysis is low in CML treated with TKIs, but more frequent early labs and good hydration are always the right thing to do; better safe than sorry!
Dr. Deininger: I second what Michael just said. To your question about whether we have an algorithm, I have to admit we don't. I think it's up to the discretion of the treating physician to initiate hydroxyurea or not.
Dr. Kalaycio: Sure. When you start hydroxyurea, do you routinely add allopurinol?
Dr. Deininger: We tend to do that. I know that some people think it's unnecessary. It's such a low-risk and low-cost intervention that I think that it's hard to get anything wrong here.
Dr. Mauro: Right, we tend to do the same. Whether we need to or not is a different question, I suppose.
Dr. Kalaycio: One more thing about the initial presentation and assessment of a patient with what you think might be a myeloproliferative disorder, CML, how important do you find it, Dr. Mauro, to measure splenomegaly and calculate a risk score?
Dr. Mauro: I think it's very useful information. The spleen size factors heavily into the risk score and the risk score does forecast response to a degree. We've looked at calculation of risk score in recent large observational studies and it is under-reported and underutilized. It winds up being useful for two reasons. One, it does set expectations for response, and US treatment guidelines (National Comprehensive Cancer Network [NCCN] guidelines) note that treatment choice may be different for low versus high risk Sokal score.2
I think the second and most intriguing reason to assess Sokal risk seems now to be the impact that risk score has on the ability to proceed to a treatment-free remission (TFR). There appear to be differences in outcomes in patients with high-risk versus low-risk disease despite both having what is required to proceed to TFR in trials, namely consistent and deep molecular remission over a number of years.
Given these implications, we really ought to be gathering the initial risk stratification and quantifying spleen size. It's an important part of our initial assessment.
Dr. Kalaycio: Dr. Deininger?
Dr. Deininger: I think there's a lot of agreement today. I absolutely think that measuring the spleen size ascertains that you've got all these diagnostic parameters available. I think that should be part of the initial evaluation and work-up and will allow some prognostication.
Testing, Risk Factors & Considerations of Treatment: Nilotinib & Dasatinib
Dr. Kalaycio: I'm starting with the less controversial questions to begin with. I think the next set have the potential for some more controversy. Before we leave the initial assessment of CML, do either of you have observations regarding referrals that you get about which you would like to either dispel myths or remind practitioners about best practices in patients newly diagnosed with CML?
I think, moving forward it would be really important to have documentation of concomitant risk factors such as smoking and so on that are essentially driving outcomes more than the TKI treatment or the CML itself. – Michael Deininger, MD, PhDDr. Mauro: I can mention one thing. I think when thinking about initial molecular diagnostic results, it is important to point out that testing should screen broadly for different fusions, namely P190 and P210, or variant (p230) transcripts. There are rare patients in chronic phase with non-p210 fusion who need to be followed with specific PCR. On this same topic, the measurement of transcripts at presentation (ie, before treatment) has become quite important and whereas formerly had not been emphasized, presently all patients should have ”baseline” transcript levels.
Dr. Deininger: I think one issue that comes up once in a while is that spleen measurements are done by ultrasound. Technically it's more accurate, but all the clinical risk scores and the prognostication is based on the old fashioned—but probably highly inaccurate—technology of palpation. This is what counts, this is the value to document. I think, moving forward it would be really important to have documentation of concomitant risk factors such as smoking and so on that are essentially driving outcomes more than the TKI treatment or the CML itself. Getting a good handle of those risk factors at diagnosis is also important.
Dr. Kalaycio: I think that's a very important point. That's where I was going to go next with this conversation. I was going to avoid the conversation about which TKI to choose as initial treatment. I think that's a debate unto itself.
I would like to ask you how you might assess cardiovascular risk before placing a patient on nilotinib. You got to that a little bit, Dr. Deininger. Could you expand on what else you might review as far as whether or not you feel a patient is a good candidate to start on nilotinib?
Dr. Deininger: Specifically, with regard to nilotinib, we would always get a baseline electrocardiogram. We would do a clinical exam. We would not do an echocardiogram just routinely in the absence of a cardiovascular history or any clinical evidence for heart failure or other cardiovascular issues.
We've adopted the practice of doing a lipid panel. Of course we would include fasting glucose as well. Some of these recommendations are probably somewhat on the soft side, because it's not yet clear what to do with the information.
On the other hand, I think for a patient who is being considered for nilotinib one wants to make sure that one really does the best to minimize the cardiovascular risk factors. Of course that would include smoking history and taking blood pressure and making sure that these risk factors are controlled.
If people have a presentation that is really out of whack in terms of their risk factor management, I would send them to an internist or even a cardiologist to help me optimize the cardiovascular prevention strategy.
Dr. Kalaycio: Great. Similarly, Dr. Mauro, how do you assess pulmonary risk before placing a patient on dasatinib?
Dr. Mauro: I think here we are focused on the less frequent and also less well understood potential toxicity of pulmonary hypertension, coupled with the more common risk of pleural and pericardial effusions.
I'm not sure how much we've learned in clinical studies looking at baseline chest X-rays or timing of X-rays during treatment. I think our best tool in the prevention and management of pleural and pericardial effusions is full discussion with patients about risk and what to look for, attention to any and all symptoms, and appropriate deployment of diagnostics as indicated.
It's interesting to consider whether baseline echocardiography for measurement of pulmonary pressures is warranted. I would say now we're on probably somewhat softer ground, first because on routine echocardiogram pulmonary pressure can't be measured readily unless there is some valvular regurgitation. As well, it is stated that pulmonary hypertension is only properly diagnosed by right heart catheterization. While I'm tempted to do routine echocardiogram studies, I think that such a recommendation still may be perhaps the realm of a clinical study. We need to explore that further. I think with dasatinib there may be certain patients at higher risk, although the data are somewhat limited. There seem to be certain conditions potentially associated with more pleural and pericardial toxicity, including cardiovascular disease and autoimmune disease. There may be circumstances during treatment—lymphocytosis, for example—that may be associated with greater risk. I think expectant management may still be the right approach and echocardiography and more aggressive diagnostics be reserved for patients in whom there might be much more clinical consequence.
Dr. Kalaycio: I'd like to pursue that a little bit further because sometimes the patients will come to us having already had an echocardiogram that may actually show some mild pulmonary hypertension and maybe they've got significant cardiovascular risk factors where you would otherwise be thinking about using dasatinib. Here's someone with pulmonary hypertension, at least by echocardiographic criteria, would that be enough to dissuade you from the use of dasatinib?
Dr. Mauro: I think it would certainly require significant consideration, understanding what the basis of the pulmonary hypertension is for that patient, and risk with adding dasatinib. I think the good news is the low incidence and the reversibility for the most part of dasatinib-associated pulmonary hypertension.
Again, I think the mechanism of action and the pathophysiology isn't completely understood, although there is the intriguing notion that imatinib has been reported to potentially mitigate pulmonary hypertension whereas dasatinib triggers it—a ”closed loop” if you will and an area ripe for research.
I would probably think that a patient with preexisting pulmonary hypertension in the new diagnosis setting might be the kind of patient for whom you really might weigh the pluses versus the minuses of a second generation TKI versus imatinib.
After Evaluation & Diagnosis: Following the Patient
Dr. Kalaycio: I agree. We have fully evaluated our patient and we've made the diagnosis. Now, we start therapy with a TKI based on patient risk profile and side effect potentials. It's time to follow the patient and determine next steps.
Current guidelines3,4 suggest monitoring quantitative PCR after 3 months of therapy and to gauge response. Dr. Deininger, how do you interpret and act on those results following the first 3 months of therapy?
Dr. Deininger: I think what you're getting at is the 10% mark that is a highly predictive value in terms of subsequent achievement of major molecular response and also overall progression free and overall survival.
We'll certainly get this data point. Then we'll put it in a clinical context. I think this context needs to take into account the initial BCR-ABL transcript level. I think Dr. Mauro mentioned that they always determine that. I think that is really good advice, so you can make a comparison with the diagnostic value.
One scenario, of course, is the patient is well below 10% and then things are just in the green range and you would wave people through and reassess in 3 months. If people are very high, I think then you have to ask yourself why that's the case.
Some people have a lot of toxicity issues. For example, they may not have been able to take the required amount of medication. It's not always easy to clearly distinguish between resistance and intolerance.
No matter what, I think it's going to be critical to consider a potential change in treatment if people are in the 70%—80% range. In my mind, there's a gray area.
These are the people who are maybe between 10%—20%. Here, I think this initial value can really help. If there's a significant reduction compared to baseline, I would not necessarily change at that point.
If there is literally no change compared to baseline, I would strongly consider a change unless I am concerned about other issues like noncompliance or drug interactions. What I'm trying to say in a rather long-winded way is that the 10% value shouldn't be seen as a dogma.
It still needs to be placed in a clinical context. One should not rush to any conclusions. Ten percent, 11%, and 9% are identical values in the world of PCR testing. One should not over-interpret that.
Dr. Kalaycio: I think that you're making an important point about absolutes in the interpretation of these tests. Dr. Mauro, how do you interpret the 3-month data that comes back?
Dr. Mauro: I agree with my colleague on the approach that it has to be put in a clinical context. I think what we've learned is the importance of the starting value and relative reduction and to not consider response milestones as black and white guides.
I think there are certain scenarios that require some caution. The patient who is on imatinib who is close but has not reached the landmark may be different that a similar patient on dasatinib or nilotinib who has not had significant reduction—that's probably a more pressing situation.
It's ironic that guidelines—maybe because of lack of options—don't encourage us to think about changing therapy early in someone who hasn't met milestones in 3 months when they've been put on a more potent agent.
I also think that careful consideration is needed regarding treatment intensity and the impact of interruptions, in conjunction with all the other facets—the rate of change, the absolute change from the patient's baseline, the starting response level and the timing of the PCR. We need to look at the actual day it's performed. If it's really not 3 months of therapy, we can misjudge.
Dr. Kalaycio: Right. Now, as you're monitoring patients, Dr. Mauro, at what point would you consider testing for kinase domain mutations?
Dr. Mauro: I think we have pretty good guidance from studies to date regarding when mutation testing is indicated and of higher yield. It's generally earlier on into treatment when we have clinical scenarios that are of greater risk, namely failure to achieve cytogenetic responses. In such scenarios if mutation testing informs treatment decision making it is very helpful.
Patients who have defined themselves as not achieving early molecular response—which we discussed earlier—especially someone who's had a second-generation TKI, warrants mutation testing in my view. Patients who don't achieve classical cytogenetic response landmarks at 6 and 12 (or 18) months, who thus have a higher residual volume of disease and perhaps as a function more clonal instability, I think also warrant attention.
I think we run into trouble when we start to, in an overly critical manner, assess patients’ longer term and deeper molecular response trajectories. Mutation testing becomes difficult or impossible for patients who are at or near major molecular remission (MMR). Mutation yield is generally very low for patients who have not lost MMR and also likely those with small volume of change around the MMR threshold. Looking ahead, I think further investigation into early time points and the setting of minimal residual disease may yield data to be able to predict potential resistance earlier than observing it clinically.
Dr. Kalaycio: Right. Dr. Deininger, if you're monitoring a patient and he’s missing milestones and you do obtain a mutation analysis and find a T315I mutation, do you offer that patient an opportunity to see how well they're going to do with ponatinib, or do you refer that patient to your transplant team for consideration of transplant as soon as a donor is available?As a transplanter, even I would not suggest transplant upon the recognition of the T315I. I would wait until all available TKIs have been tried and have either not worked or weren't tolerated. – Matt Kalaycio, MD
Dr. Deininger: I think we do the first and half of the second. We would certainly offer a trial of ponatinib or, if possible, a clinical trial unless there are insurmountable contraindications.
I don't really think there are any such circumstances with optimized cardiovascular management at the same time, and we may involve a cardiologist at that point to help us if there are cardiovascular risk factors.
We would also do a referral to a transplant center unless the patient is as per performance status, just not a transplant candidate or is too old. Otherwise, we would always make a referral, but we would not pursue a transplant as the first modality, provided that the patient is in chronic phase. Progression to accelerated phase or blastic phase would be looked at totally differently. Here we would certainly put people on a transplant course.
Dr. Kalaycio: I agree. As a transplanter, even I would not suggest transplant upon the recognition of the T315I. I would wait until all available TKIs have been tried and have either not worked or weren't tolerated.
As the two of you know, we're doing fewer and fewer transplants for CML these days. The results we're getting in folks who are failing all of these agents are not as good as they used to be when we were transplanting in the first 3 months of a new diagnosis. Bone marrow transplant is something to defer for as long as possible, at least in my mind.
TKI Dosage Considerations
Dr. Kalaycio: Now we touched a little bit on the side effects of nilotinib and dasatinib. We've talked about progressive disease, perhaps with or without mutations. Dr. Mauro, could you comment on where you think bosutinib might play a role in patients with relapsed or progressive CML?
Dr. Mauro: I think bosutinib remains an underutilized TKI. It is potent and thus offers good salvage activity. I think it offers a fairly distinct side effect profile making it a good choice for cases of intolerance. I think we've learned more about how to initiate bosutinib and manage its side effects; one element of this is that current trials start at slightly lower doses of bosutinib to avoid some of the early gastrointestinal toxicity.
Bosutinib continues to seek a place as a first-line therapy. I think this may be worthwhile as we are looking for the right balance of safety and side effect risk to maximize early response. Regarding the salvage setting—intolerance and resistance to other agents—I find myself speaking to patients about bosutinib as a potent and viable alternative. Bosutinib offers a similar spectrum of activity as dasatinib but can allow us to avoid cardiovascular toxicity and/or allow clearance of pleural and pericardial toxicity occurring with dasatinib.
The one place I would say bosutinib may not hold up as strongly to its competition would be as a third-line therapy after failure of a second-generation TKI. I think while there is reasonable activity with bosutinib in this setting of somewhat fairly drug-resistant CML, the performance of ponatinib is better. There is likely often a struggle with the long-term safety question of ponatinib, making some feel that a trial of bosutinib is often a worthwhile, logical, or even necessary step before ponatinib. My concerns about this approach are related to treatment decision-making based on safety and theoretical risk of adverse effects rather than efficacy and risk of progression; these have to be weighed carefully and appropriately against each other.
Dr. Kalaycio: That's an interesting perspective. You mentioned starting with a lower dose. Dr. Deininger, I'd like to ask you your thoughts about dose modifications of the TKIs in general. It used to be anathema to either start with a low dose or to maintain patients on a lower dose.
I'm aware of at least some data suggesting that those patients who tolerate these TKIs poorly particularly with regard to myelosuppression can be treated long term at lower than typically prescribed doses without adverse effect. What are your thoughts surrounding dose modification of the TKIs?
Dr. Deininger: I would stratify according to TKIs. I think imatinib at 400 mg is probably just rightly dosed, maybe not even at the optimal dose. It could be 600 mg. If you go to 300 mg, that is still acceptable.
If you go to 200 mg, I think I would make double sure that I monitor patients very frequently. People should be, in my mind, in a major molecular response to treat them with 200 mg long term. With the second-generation TKIs, it's a little different.
I cannot really speak to bosutinib because I'm not aware of a lot of data on those modifications. What seems to be clear from dasatinib is that the initial recommended dose of 100 mg is quite high, especially in older people.
I believe that's been corroborated by a study that hasn't been published yet. Apparently the drug excretion in the older individuals is quite a bit slower than in younger people. I think in our practice and in other people's practices as well, about 50% of those people end at doses substantially lower than 100 mg, maybe 50 mg, maybe 40 mg, some even 20 mg per day.
I think in the case of dasatinib you have a lot of maneuvering space. You can adjust the dose according to tolerability and molecular response. With nilotinib I think it's kind of similar. We have a few patients who are on maybe 150 mg twice a day because they have issues in terms of side effects at the higher doses.
I should again thoroughly qualify that by saying if you give up through dose reductions the goal of a major molecular response, then I think you really have to think about the strategy in general because at that point, I think you basically say "this patient is not ever going to be reaching a safe haven and is not tolerating inhibitors well enough to get him there."
That's quite a significant statement to make. Myelosuppression is frequently a situation where you cannot deliver enough dose but you also don't get a good response, whereas other side effects like pleural effusions or excessive fluid retention may well be cause for dose reduction and yet responses may be acceptable. Maybe Dr. Mauro could chime in here and share his experience.
Dr. Mauro: I agree. I think some patients, based on their disease status, may not tolerate the dose that is going to—at least in theory—get them to a deep molecular remission. The biggest problem we often face is myelosuppression. Some patients may have indolent disease and still have a good outcome, at least in the short to medium term, with lower doses of drug. In general I try to work through myelosuppression, especially early, as the combination of myelosuppression and suboptimal or response failure is dangerous.
I think treatment shouldn't be too chaotic and change too much, if possible. That was a very nice summary of the way we view the doses. There's still a fair bit of flexibility amongst the TKI doses, although I don't advocate for starting low and titration up.
In general I think, with regard to optimal TKI dose, we might have overshot; for example, with dasatinib. I think we might have overshot with bosutinib. Trials ongoing now initiate bosutinib at 400 mg rather than 500 mg. That's what I was alluding to.
Treatment Free Remission: Approaches
Dr. Kalaycio: Very interesting discussions. As we wind down the conversation, I want to get to the idea of stopping therapy. We're all aware of data that suggest it's at least possible for a proportion of patients to stop treatment.5
However, the guidelines3,4 such as they are, suggest that stopping should not be done outside the context of a clinical trial. I think most of us would agree with that. I wonder in a practical sense, Dr. Mauro, in your practice, do you have your own set of internal criteria for stopping a TKI and observing patients in the absence of any therapy?
Dr. Mauro: I think I try to incorporate all the experience we have to date when considering this question. That being said I'm a little hesitant to agree entirely with the current thinking regarding retreatment during a Treatment Free Remission trial, which is waiting until patients lose MMR in order to resume treatment but agree it might be hard to retreat based on lesser degrees of molecular relapse. I just worry a little bit about that amount of proliferation without treatment.
On the other hand I think we're coming to the realization that TFR may be feasible in patients who are simply nonproliferative and have low volume of disease based on newer data that patients may not need to have consistent "complete molecular response" or disease reduction below 4.5 logs in order to consider TFR. This will mean more patients may be eligible for such a strategy.
With that being said, I strongly advocate for discontinuation occurring in trials still, especially in the United States. I think we still have to ensure regular monitoring and not run the risk of loss of CML remission as a result of this endeavor. I think we have to have a clear message that this is still an investigational approach.
As we apply TFR strategies more broadly and explore it in patients with different circumstances, rather than those already studied in clinical trials, we may see slightly different results. We may need to exercise more caution.
Outside of a trial I don't discontinue TKI therapy unless there's a clear medical indication and there's no way around it, such as pregnancy or other illness that precludes TKI treatment. I would still pursue TFR only in a clinical trial.
Dr. Kalaycio: Dr. Deininger, I'll give you the last word for your thoughts regarding both treatment discontinuation and the future of CML therapies.
Dr. Deininger: As far as TFR is concerned, I agree with Dr. Mauro. Unfortunately, in the United States we're not quite as far advanced as the Europeans at monitoring diligence. There's a bit of a concern that if we elevate that into daily practice, we may see that patients are not monitored frequently enough and then relapse, but that doesn't get caught. You could imagine a scenario where we're actually seeing a decline in outcomes because people are discontinued and then not restarted if they have a recurrence at the molecular level.
All in all, where this is going, I think it will be really interesting to see more confirmatory data for second-generation TKIs. Let me put that differently: whether a similar proportion of patients in deep molecular response can maintain responses and have TFR if they needed a second-generation TKI to get there rather than just imatinib.
If so, this would be very promising. That would mean that second-generation TKIs actually somewhat impact the natural history of the disease and its biology. Then TFR may become a reality for a substantial proportion of our patients.
I don't think we have the data yet, but at least there are some suggestions that this may be the case. If you do the math, you will still see that many patients with CML will never reach a state where they can consider TFR.
They'll never get into a deep molecular response or they will have been diagnosed with accelerated phase. For all these patients who are not candidates for TFR, we still have to think about treatment optimization in order to make them catch up with the rest.
Now I have a couple of interesting developments. One is clearly to see whether ponatinib administered at a lower dose will have a more practical, therapeutic window in terms of toxicity and yet maintain the excellent efficacy that it has in the setting.
These data need to be produced. Another scenario would be that people would get started on intense induction treatment with higher-risk drugs such as ponatinib. Then if they are a good responder, they can switch to something that is lower-risk and that may be better tolerated.
Actually some of these trials are underway in Europe. Of course there's the question, are there still other drugs that will enter the CML space. There's one promising molecule called ABL001, a TKI with a different mode of action, which exploits an allosteric site rather than the catalytic center of the kinase.
It’s really conceptually very interesting. The expectation would be that this molecule has fewer side effects but still may be quite potent. This could be a very interesting development and add something to the armamentarium that we currently don't have.
I also think there will be patients whose diseases are just beyond the reach of a TKI alone for many reasons, maybe additional mutations or things that have to do with the host and metabolism.
In these cases, we'll still have to think about combination treatments and non-TKI treatments. Here, I think an honest answer is the labs have pulled out a lot of interesting leads, but so far, nothing has really made it into a serious clinical context, either because of side effects or because the target may not be as good in humans as it seemed from mouse models.
I think a lot more work is required here to find the best combination therapies and to define those pathways that need to be inhibited together with BCR-ABL. I think there's a field that we'll develop further and that will be the cutting edge.
The Wrap-Up
Dr. Kalaycio: Although an uncommon diagnosis, CML in chronic phase can be treated with an expectation for nearly 100% 5-year survival. For that reason, clinicians need to manage their patients expertly. Steps taken at diagnosis are critical to subsequent decision-making. Once treatment begins, close monitoring is required to screen for side effects as well as ensuring treatment success. With so many effective agents available, patients can be selected for the agent least likely to cause them long-term harm. Perhaps in the future we will learn which patients can stop treatment altogether.
Gentleman, thank you so much for your time. Every time I talk to you I learn something.
1. Peng B, Hayes M, Resta D, et al. Pharmacokinetics and pharmacodynamics of imatinib in a phase I trial with chronic myeloid leukemia patients. J Clin Oncol. 2004;22(5):935–942.
2. National Comprehensive Cancer Network guidelines for treatment of cancer by site. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp. Accessed August 20, 2015.
3. O'Brien S, Radich JP, Abboud CN, et al. Chronic myelogenous leukemia, version 1.2015. J Natl Compr Canc Netw. 2014;12(11):1590–1610.
4. Baccarani M, Deininger MW, Rosti G, et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood. 2013;122(6):872–884.
5. Mahon FX, Réa D, Guilhot J, et al. Discontinuation of imatinib in patients with chronic myeloid leukemia who have maintained complete molecular remission for at least 2 years: the prospective, multicentre Stop Imatinib (STIM) trial. Lancet Oncol. 2010;11(11):1029–1035.
Moderator: Matt Kalaycio, MD1
Discussants: Michael Mauro, MD2; Michael Deininger, MD, PhD3
From Cleveland Clinic, Cleveland, OH1; Memorial Sloan Kettering Cancer Center, New York, NY2; Huntsman Cancer Institute, Salt Lake City, UT3
Address for correspondence: Matt Kalaycio, MD, Cleveland Clinic Main Campus, Mail Code R32, 9500 Euclid Avenue, Cleveland, OH 44195
E-mail: [email protected]
Biographical sketch:
Dr. Kalaycio has been published in numerous scientific publications including Bone Marrow Transplantation, Journal of Clinical Oncology, and Leukemia. He also is the editor of a book on leukemia and co-editor of a book on clinical malignant hematology. His research interests focus on testing new treatments for leukemia.
Dr. Kalaycio received his degree from West Virginia University School of Medicine in Morgantown. He completed his residency in internal medicine at Mercy Hospital of Pittsburgh and fellowships in hematology and medical oncology and bone marrow transplantation at Cleveland Clinic.
Michael Deininger, MD, PhD, is Professor and Chief of Hematology and Hematologic Malignancies for the Department of Internal Medicine and for the Huntsman Cancer Institute (HCI) at the University of Utah. He is an HCI investigator and member of the Experimental Therapeutics program. He has extensive experience treating patients with blood cancers, including chronic myeloid leukemia (CML) and myeloproliferative neoplasms, a group of blood cancers related to leukemia.
Dr
Dr. Deininger’s scientific focus is leukemia, specifically myeloproliferative neoplasms including chronic myeloid leukemia (CML). As a clinician-scientist with a translational research focus Dr. Deininger is heading an extramurally funded research laboratory that is dedicated to the study of signaling pathways, drug resistance and new molecular therapies in leukemia. Dr. Deininger’s work describing the selective effects of imatinib on CML cells provided the rationale for clinical trials that led to the approval of Gleevec as the first molecularly-based therapy for leukemia. Current work in his lab is focused on understanding the role of the bone marrow microenvironment in leukemia drug resistance, discovering novel therapeutic targets and developing more specific signal transduction inhibitors. Dr. Deininger’s work encompasses more than 170 articles in the peer-reviewed literature, including journals like Blood, Journal of Clinical Investigation and the New England Journal of Medicine. He has co-authored more than 10 book chapters, with contributions in leading textbooks such as deVita’s Principles of Oncology. He is a regular speaker at major international scientific meetings, such as the American Society of Hematology and the European Hematology Association and a peer reviewer for journals like Nature Genetics and Cancer Cell. His honors include the Alexandra Kefalides Prize for Leukemia Research and membership on the Editorial Board of Blood, the leading journal in Hematology. Dr. Deininger was named among the world's Highly Cited Researchers by Thomson Reuters in 2014.
Dr. Kalaycio: My name is Matt Kalaycio and I'm the Chairman of the Department of Hematology and Medical Oncology at the Cleveland Clinic. Today I’m joined by Drs. Mike Deininger, Division Chief of Hematology and Hematologic Malignancies at the University of Utah Huntsman Cancer Institute and Michael Mauro, Myeloproliferative Neoplasm Program Director at Memorial Sloan Kettering Cancer Center, New York, New York – together we will discuss practical issues surrounding CML diagnosis, management and treatment options including TKIs and investigational therapies.
Initial Presentation: Assessment & Treatment Options
Dr. Kalaycio: Often, the patient with chronic myelogenous leukemia (CML) gets admitted to the hospital, or an emergency consult is called, because the white count is up for a concern of leukemia. The treatment team sees the differential circulating blasts, and they worry about acute leukemia. Then the hematologist comes and needs to make a decision about whether or not to do a bone marrow biopsy.
When the patient presents in such a manner, I often see that bone marrow biopsies are not performed. I would like to start by asking where you both stand with regard to the necessity of a bone marrow biopsy at the time of diagnosis. I'll start with Dr. Deininger.
Dr. Deininger: I would strongly be in favor of a diagnostic biopsy as well as a smear. The reason is that I think this is the one and only chance to get a clear disease classification into chronic phase, accelerated phase, and blast crisis.
The third may be rare. There are occasional patients who have sheets of blasts who would not be seen on just a differential. For these patients, of course, the treatment decisions are going to be very different compared to a patient in chronic phase.
I think this is an opportunity that shouldn't be missed and I would always recommend that.
Dr. Kalaycio: Dr. Mauro.
Dr. Mauro: I couldn't concur more. With so much focus on the change in disease status from presentation to early response, I think understanding the scope of the disease at diagnosis, including the bone marrow, is essential and either revealing or reassuring. Although we focus mostly on early molecular response, when expectations go awry, not having as full a picture of the disease prior to treatment leaves you less informed about best treatment.
Making sense of accelerated phase features is a good example; the difference in outcomes between chronic phase patients and those who have morphologic features of accelerated phase—with or without cytogenetic features (clonal evolution)—compared with those with clonal evolution can only drive initial and long term treatment decisions. Without cytogenetics and morphology together, such key pieces of the puzzle are missing; it is worth it for the practitioner and patient to have all the data in hand at all times.
Dr. Kalaycio: Once the diagnosis has been made and the biopsy was not done and now you're seeing the patient having either been on hydroxyurea for a month or having had tyrosine kinase inhibitor (TKI) for a month, do you bother with the bone marrow biopsy at that point?
Dr. Deininger: Well, that is a very good question. I think we would probably still do it most of the time. I think it's very clear that the information you can gather from that is less valuable. I think one should try to make up for the omissions as much as possible. We would still go for it.
Dr. Kalaycio: Interesting. The other thing that happens a lot is a patient will be started on hydrea while you're waiting for BCR-ABL to return either by fluorescence in situ hybridization or quantitative polymerase chain reaction (PCR).
I wonder if you have your own set of internal guidelines for when hydroxyurea should be employed or not.
Dr. Mauro: If you look back at some of the original imatinib trials—the phase I trials—patients initiated TKI with higher blood counts, a median around 25,000 and up to 200,000, and responded with excellent tolerance.1 I think there's a bit of overapplication of hydroxyurea early in CML, prior to initiation of TKI, which often confounds and complicates the early myelosuppressive toxicity of TKIs. With use of more potent TKIs, there may be greater amounts of myelosuppression as the leukemic clone may clear faster.
I think we may get ourselves into a bit of a bind by overusing hydroxyurea and then needing to hold and lower TKI dosages quickly when that may not have been necessary had we simply deployed the TKI sooner.
My general rule would be to use hydroxyurea for symptoms if necessary or for more extreme counts where leukostasis is a concern. – Michael Mauro, MDMy general rule would be to use hydroxyurea for symptoms if necessary or for more extreme counts where leukostasis is a concern. The other question that often comes up is about tumor lysis and hydration, and how closely these need to be managed. The likelihood of tumor lysis is low in CML treated with TKIs, but more frequent early labs and good hydration are always the right thing to do; better safe than sorry!
Dr. Deininger: I second what Michael just said. To your question about whether we have an algorithm, I have to admit we don't. I think it's up to the discretion of the treating physician to initiate hydroxyurea or not.
Dr. Kalaycio: Sure. When you start hydroxyurea, do you routinely add allopurinol?
Dr. Deininger: We tend to do that. I know that some people think it's unnecessary. It's such a low-risk and low-cost intervention that I think that it's hard to get anything wrong here.
Dr. Mauro: Right, we tend to do the same. Whether we need to or not is a different question, I suppose.
Dr. Kalaycio: One more thing about the initial presentation and assessment of a patient with what you think might be a myeloproliferative disorder, CML, how important do you find it, Dr. Mauro, to measure splenomegaly and calculate a risk score?
Dr. Mauro: I think it's very useful information. The spleen size factors heavily into the risk score and the risk score does forecast response to a degree. We've looked at calculation of risk score in recent large observational studies and it is under-reported and underutilized. It winds up being useful for two reasons. One, it does set expectations for response, and US treatment guidelines (National Comprehensive Cancer Network [NCCN] guidelines) note that treatment choice may be different for low versus high risk Sokal score.2
I think the second and most intriguing reason to assess Sokal risk seems now to be the impact that risk score has on the ability to proceed to a treatment-free remission (TFR). There appear to be differences in outcomes in patients with high-risk versus low-risk disease despite both having what is required to proceed to TFR in trials, namely consistent and deep molecular remission over a number of years.
Given these implications, we really ought to be gathering the initial risk stratification and quantifying spleen size. It's an important part of our initial assessment.
Dr. Kalaycio: Dr. Deininger?
Dr. Deininger: I think there's a lot of agreement today. I absolutely think that measuring the spleen size ascertains that you've got all these diagnostic parameters available. I think that should be part of the initial evaluation and work-up and will allow some prognostication.
Testing, Risk Factors & Considerations of Treatment: Nilotinib & Dasatinib
Dr. Kalaycio: I'm starting with the less controversial questions to begin with. I think the next set have the potential for some more controversy. Before we leave the initial assessment of CML, do either of you have observations regarding referrals that you get about which you would like to either dispel myths or remind practitioners about best practices in patients newly diagnosed with CML?
I think, moving forward it would be really important to have documentation of concomitant risk factors such as smoking and so on that are essentially driving outcomes more than the TKI treatment or the CML itself. – Michael Deininger, MD, PhDDr. Mauro: I can mention one thing. I think when thinking about initial molecular diagnostic results, it is important to point out that testing should screen broadly for different fusions, namely P190 and P210, or variant (p230) transcripts. There are rare patients in chronic phase with non-p210 fusion who need to be followed with specific PCR. On this same topic, the measurement of transcripts at presentation (ie, before treatment) has become quite important and whereas formerly had not been emphasized, presently all patients should have ”baseline” transcript levels.
Dr. Deininger: I think one issue that comes up once in a while is that spleen measurements are done by ultrasound. Technically it's more accurate, but all the clinical risk scores and the prognostication is based on the old fashioned—but probably highly inaccurate—technology of palpation. This is what counts, this is the value to document. I think, moving forward it would be really important to have documentation of concomitant risk factors such as smoking and so on that are essentially driving outcomes more than the TKI treatment or the CML itself. Getting a good handle of those risk factors at diagnosis is also important.
Dr. Kalaycio: I think that's a very important point. That's where I was going to go next with this conversation. I was going to avoid the conversation about which TKI to choose as initial treatment. I think that's a debate unto itself.
I would like to ask you how you might assess cardiovascular risk before placing a patient on nilotinib. You got to that a little bit, Dr. Deininger. Could you expand on what else you might review as far as whether or not you feel a patient is a good candidate to start on nilotinib?
Dr. Deininger: Specifically, with regard to nilotinib, we would always get a baseline electrocardiogram. We would do a clinical exam. We would not do an echocardiogram just routinely in the absence of a cardiovascular history or any clinical evidence for heart failure or other cardiovascular issues.
We've adopted the practice of doing a lipid panel. Of course we would include fasting glucose as well. Some of these recommendations are probably somewhat on the soft side, because it's not yet clear what to do with the information.
On the other hand, I think for a patient who is being considered for nilotinib one wants to make sure that one really does the best to minimize the cardiovascular risk factors. Of course that would include smoking history and taking blood pressure and making sure that these risk factors are controlled.
If people have a presentation that is really out of whack in terms of their risk factor management, I would send them to an internist or even a cardiologist to help me optimize the cardiovascular prevention strategy.
Dr. Kalaycio: Great. Similarly, Dr. Mauro, how do you assess pulmonary risk before placing a patient on dasatinib?
Dr. Mauro: I think here we are focused on the less frequent and also less well understood potential toxicity of pulmonary hypertension, coupled with the more common risk of pleural and pericardial effusions.
I'm not sure how much we've learned in clinical studies looking at baseline chest X-rays or timing of X-rays during treatment. I think our best tool in the prevention and management of pleural and pericardial effusions is full discussion with patients about risk and what to look for, attention to any and all symptoms, and appropriate deployment of diagnostics as indicated.
It's interesting to consider whether baseline echocardiography for measurement of pulmonary pressures is warranted. I would say now we're on probably somewhat softer ground, first because on routine echocardiogram pulmonary pressure can't be measured readily unless there is some valvular regurgitation. As well, it is stated that pulmonary hypertension is only properly diagnosed by right heart catheterization. While I'm tempted to do routine echocardiogram studies, I think that such a recommendation still may be perhaps the realm of a clinical study. We need to explore that further. I think with dasatinib there may be certain patients at higher risk, although the data are somewhat limited. There seem to be certain conditions potentially associated with more pleural and pericardial toxicity, including cardiovascular disease and autoimmune disease. There may be circumstances during treatment—lymphocytosis, for example—that may be associated with greater risk. I think expectant management may still be the right approach and echocardiography and more aggressive diagnostics be reserved for patients in whom there might be much more clinical consequence.
Dr. Kalaycio: I'd like to pursue that a little bit further because sometimes the patients will come to us having already had an echocardiogram that may actually show some mild pulmonary hypertension and maybe they've got significant cardiovascular risk factors where you would otherwise be thinking about using dasatinib. Here's someone with pulmonary hypertension, at least by echocardiographic criteria, would that be enough to dissuade you from the use of dasatinib?
Dr. Mauro: I think it would certainly require significant consideration, understanding what the basis of the pulmonary hypertension is for that patient, and risk with adding dasatinib. I think the good news is the low incidence and the reversibility for the most part of dasatinib-associated pulmonary hypertension.
Again, I think the mechanism of action and the pathophysiology isn't completely understood, although there is the intriguing notion that imatinib has been reported to potentially mitigate pulmonary hypertension whereas dasatinib triggers it—a ”closed loop” if you will and an area ripe for research.
I would probably think that a patient with preexisting pulmonary hypertension in the new diagnosis setting might be the kind of patient for whom you really might weigh the pluses versus the minuses of a second generation TKI versus imatinib.
After Evaluation & Diagnosis: Following the Patient
Dr. Kalaycio: I agree. We have fully evaluated our patient and we've made the diagnosis. Now, we start therapy with a TKI based on patient risk profile and side effect potentials. It's time to follow the patient and determine next steps.
Current guidelines3,4 suggest monitoring quantitative PCR after 3 months of therapy and to gauge response. Dr. Deininger, how do you interpret and act on those results following the first 3 months of therapy?
Dr. Deininger: I think what you're getting at is the 10% mark that is a highly predictive value in terms of subsequent achievement of major molecular response and also overall progression free and overall survival.
We'll certainly get this data point. Then we'll put it in a clinical context. I think this context needs to take into account the initial BCR-ABL transcript level. I think Dr. Mauro mentioned that they always determine that. I think that is really good advice, so you can make a comparison with the diagnostic value.
One scenario, of course, is the patient is well below 10% and then things are just in the green range and you would wave people through and reassess in 3 months. If people are very high, I think then you have to ask yourself why that's the case.
Some people have a lot of toxicity issues. For example, they may not have been able to take the required amount of medication. It's not always easy to clearly distinguish between resistance and intolerance.
No matter what, I think it's going to be critical to consider a potential change in treatment if people are in the 70%—80% range. In my mind, there's a gray area.
These are the people who are maybe between 10%—20%. Here, I think this initial value can really help. If there's a significant reduction compared to baseline, I would not necessarily change at that point.
If there is literally no change compared to baseline, I would strongly consider a change unless I am concerned about other issues like noncompliance or drug interactions. What I'm trying to say in a rather long-winded way is that the 10% value shouldn't be seen as a dogma.
It still needs to be placed in a clinical context. One should not rush to any conclusions. Ten percent, 11%, and 9% are identical values in the world of PCR testing. One should not over-interpret that.
Dr. Kalaycio: I think that you're making an important point about absolutes in the interpretation of these tests. Dr. Mauro, how do you interpret the 3-month data that comes back?
Dr. Mauro: I agree with my colleague on the approach that it has to be put in a clinical context. I think what we've learned is the importance of the starting value and relative reduction and to not consider response milestones as black and white guides.
I think there are certain scenarios that require some caution. The patient who is on imatinib who is close but has not reached the landmark may be different that a similar patient on dasatinib or nilotinib who has not had significant reduction—that's probably a more pressing situation.
It's ironic that guidelines—maybe because of lack of options—don't encourage us to think about changing therapy early in someone who hasn't met milestones in 3 months when they've been put on a more potent agent.
I also think that careful consideration is needed regarding treatment intensity and the impact of interruptions, in conjunction with all the other facets—the rate of change, the absolute change from the patient's baseline, the starting response level and the timing of the PCR. We need to look at the actual day it's performed. If it's really not 3 months of therapy, we can misjudge.
Dr. Kalaycio: Right. Now, as you're monitoring patients, Dr. Mauro, at what point would you consider testing for kinase domain mutations?
Dr. Mauro: I think we have pretty good guidance from studies to date regarding when mutation testing is indicated and of higher yield. It's generally earlier on into treatment when we have clinical scenarios that are of greater risk, namely failure to achieve cytogenetic responses. In such scenarios if mutation testing informs treatment decision making it is very helpful.
Patients who have defined themselves as not achieving early molecular response—which we discussed earlier—especially someone who's had a second-generation TKI, warrants mutation testing in my view. Patients who don't achieve classical cytogenetic response landmarks at 6 and 12 (or 18) months, who thus have a higher residual volume of disease and perhaps as a function more clonal instability, I think also warrant attention.
I think we run into trouble when we start to, in an overly critical manner, assess patients’ longer term and deeper molecular response trajectories. Mutation testing becomes difficult or impossible for patients who are at or near major molecular remission (MMR). Mutation yield is generally very low for patients who have not lost MMR and also likely those with small volume of change around the MMR threshold. Looking ahead, I think further investigation into early time points and the setting of minimal residual disease may yield data to be able to predict potential resistance earlier than observing it clinically.
Dr. Kalaycio: Right. Dr. Deininger, if you're monitoring a patient and he’s missing milestones and you do obtain a mutation analysis and find a T315I mutation, do you offer that patient an opportunity to see how well they're going to do with ponatinib, or do you refer that patient to your transplant team for consideration of transplant as soon as a donor is available?As a transplanter, even I would not suggest transplant upon the recognition of the T315I. I would wait until all available TKIs have been tried and have either not worked or weren't tolerated. – Matt Kalaycio, MD
Dr. Deininger: I think we do the first and half of the second. We would certainly offer a trial of ponatinib or, if possible, a clinical trial unless there are insurmountable contraindications.
I don't really think there are any such circumstances with optimized cardiovascular management at the same time, and we may involve a cardiologist at that point to help us if there are cardiovascular risk factors.
We would also do a referral to a transplant center unless the patient is as per performance status, just not a transplant candidate or is too old. Otherwise, we would always make a referral, but we would not pursue a transplant as the first modality, provided that the patient is in chronic phase. Progression to accelerated phase or blastic phase would be looked at totally differently. Here we would certainly put people on a transplant course.
Dr. Kalaycio: I agree. As a transplanter, even I would not suggest transplant upon the recognition of the T315I. I would wait until all available TKIs have been tried and have either not worked or weren't tolerated.
As the two of you know, we're doing fewer and fewer transplants for CML these days. The results we're getting in folks who are failing all of these agents are not as good as they used to be when we were transplanting in the first 3 months of a new diagnosis. Bone marrow transplant is something to defer for as long as possible, at least in my mind.
TKI Dosage Considerations
Dr. Kalaycio: Now we touched a little bit on the side effects of nilotinib and dasatinib. We've talked about progressive disease, perhaps with or without mutations. Dr. Mauro, could you comment on where you think bosutinib might play a role in patients with relapsed or progressive CML?
Dr. Mauro: I think bosutinib remains an underutilized TKI. It is potent and thus offers good salvage activity. I think it offers a fairly distinct side effect profile making it a good choice for cases of intolerance. I think we've learned more about how to initiate bosutinib and manage its side effects; one element of this is that current trials start at slightly lower doses of bosutinib to avoid some of the early gastrointestinal toxicity.
Bosutinib continues to seek a place as a first-line therapy. I think this may be worthwhile as we are looking for the right balance of safety and side effect risk to maximize early response. Regarding the salvage setting—intolerance and resistance to other agents—I find myself speaking to patients about bosutinib as a potent and viable alternative. Bosutinib offers a similar spectrum of activity as dasatinib but can allow us to avoid cardiovascular toxicity and/or allow clearance of pleural and pericardial toxicity occurring with dasatinib.
The one place I would say bosutinib may not hold up as strongly to its competition would be as a third-line therapy after failure of a second-generation TKI. I think while there is reasonable activity with bosutinib in this setting of somewhat fairly drug-resistant CML, the performance of ponatinib is better. There is likely often a struggle with the long-term safety question of ponatinib, making some feel that a trial of bosutinib is often a worthwhile, logical, or even necessary step before ponatinib. My concerns about this approach are related to treatment decision-making based on safety and theoretical risk of adverse effects rather than efficacy and risk of progression; these have to be weighed carefully and appropriately against each other.
Dr. Kalaycio: That's an interesting perspective. You mentioned starting with a lower dose. Dr. Deininger, I'd like to ask you your thoughts about dose modifications of the TKIs in general. It used to be anathema to either start with a low dose or to maintain patients on a lower dose.
I'm aware of at least some data suggesting that those patients who tolerate these TKIs poorly particularly with regard to myelosuppression can be treated long term at lower than typically prescribed doses without adverse effect. What are your thoughts surrounding dose modification of the TKIs?
Dr. Deininger: I would stratify according to TKIs. I think imatinib at 400 mg is probably just rightly dosed, maybe not even at the optimal dose. It could be 600 mg. If you go to 300 mg, that is still acceptable.
If you go to 200 mg, I think I would make double sure that I monitor patients very frequently. People should be, in my mind, in a major molecular response to treat them with 200 mg long term. With the second-generation TKIs, it's a little different.
I cannot really speak to bosutinib because I'm not aware of a lot of data on those modifications. What seems to be clear from dasatinib is that the initial recommended dose of 100 mg is quite high, especially in older people.
I believe that's been corroborated by a study that hasn't been published yet. Apparently the drug excretion in the older individuals is quite a bit slower than in younger people. I think in our practice and in other people's practices as well, about 50% of those people end at doses substantially lower than 100 mg, maybe 50 mg, maybe 40 mg, some even 20 mg per day.
I think in the case of dasatinib you have a lot of maneuvering space. You can adjust the dose according to tolerability and molecular response. With nilotinib I think it's kind of similar. We have a few patients who are on maybe 150 mg twice a day because they have issues in terms of side effects at the higher doses.
I should again thoroughly qualify that by saying if you give up through dose reductions the goal of a major molecular response, then I think you really have to think about the strategy in general because at that point, I think you basically say "this patient is not ever going to be reaching a safe haven and is not tolerating inhibitors well enough to get him there."
That's quite a significant statement to make. Myelosuppression is frequently a situation where you cannot deliver enough dose but you also don't get a good response, whereas other side effects like pleural effusions or excessive fluid retention may well be cause for dose reduction and yet responses may be acceptable. Maybe Dr. Mauro could chime in here and share his experience.
Dr. Mauro: I agree. I think some patients, based on their disease status, may not tolerate the dose that is going to—at least in theory—get them to a deep molecular remission. The biggest problem we often face is myelosuppression. Some patients may have indolent disease and still have a good outcome, at least in the short to medium term, with lower doses of drug. In general I try to work through myelosuppression, especially early, as the combination of myelosuppression and suboptimal or response failure is dangerous.
I think treatment shouldn't be too chaotic and change too much, if possible. That was a very nice summary of the way we view the doses. There's still a fair bit of flexibility amongst the TKI doses, although I don't advocate for starting low and titration up.
In general I think, with regard to optimal TKI dose, we might have overshot; for example, with dasatinib. I think we might have overshot with bosutinib. Trials ongoing now initiate bosutinib at 400 mg rather than 500 mg. That's what I was alluding to.
Treatment Free Remission: Approaches
Dr. Kalaycio: Very interesting discussions. As we wind down the conversation, I want to get to the idea of stopping therapy. We're all aware of data that suggest it's at least possible for a proportion of patients to stop treatment.5
However, the guidelines3,4 such as they are, suggest that stopping should not be done outside the context of a clinical trial. I think most of us would agree with that. I wonder in a practical sense, Dr. Mauro, in your practice, do you have your own set of internal criteria for stopping a TKI and observing patients in the absence of any therapy?
Dr. Mauro: I think I try to incorporate all the experience we have to date when considering this question. That being said I'm a little hesitant to agree entirely with the current thinking regarding retreatment during a Treatment Free Remission trial, which is waiting until patients lose MMR in order to resume treatment but agree it might be hard to retreat based on lesser degrees of molecular relapse. I just worry a little bit about that amount of proliferation without treatment.
On the other hand I think we're coming to the realization that TFR may be feasible in patients who are simply nonproliferative and have low volume of disease based on newer data that patients may not need to have consistent "complete molecular response" or disease reduction below 4.5 logs in order to consider TFR. This will mean more patients may be eligible for such a strategy.
With that being said, I strongly advocate for discontinuation occurring in trials still, especially in the United States. I think we still have to ensure regular monitoring and not run the risk of loss of CML remission as a result of this endeavor. I think we have to have a clear message that this is still an investigational approach.
As we apply TFR strategies more broadly and explore it in patients with different circumstances, rather than those already studied in clinical trials, we may see slightly different results. We may need to exercise more caution.
Outside of a trial I don't discontinue TKI therapy unless there's a clear medical indication and there's no way around it, such as pregnancy or other illness that precludes TKI treatment. I would still pursue TFR only in a clinical trial.
Dr. Kalaycio: Dr. Deininger, I'll give you the last word for your thoughts regarding both treatment discontinuation and the future of CML therapies.
Dr. Deininger: As far as TFR is concerned, I agree with Dr. Mauro. Unfortunately, in the United States we're not quite as far advanced as the Europeans at monitoring diligence. There's a bit of a concern that if we elevate that into daily practice, we may see that patients are not monitored frequently enough and then relapse, but that doesn't get caught. You could imagine a scenario where we're actually seeing a decline in outcomes because people are discontinued and then not restarted if they have a recurrence at the molecular level.
All in all, where this is going, I think it will be really interesting to see more confirmatory data for second-generation TKIs. Let me put that differently: whether a similar proportion of patients in deep molecular response can maintain responses and have TFR if they needed a second-generation TKI to get there rather than just imatinib.
If so, this would be very promising. That would mean that second-generation TKIs actually somewhat impact the natural history of the disease and its biology. Then TFR may become a reality for a substantial proportion of our patients.
I don't think we have the data yet, but at least there are some suggestions that this may be the case. If you do the math, you will still see that many patients with CML will never reach a state where they can consider TFR.
They'll never get into a deep molecular response or they will have been diagnosed with accelerated phase. For all these patients who are not candidates for TFR, we still have to think about treatment optimization in order to make them catch up with the rest.
Now I have a couple of interesting developments. One is clearly to see whether ponatinib administered at a lower dose will have a more practical, therapeutic window in terms of toxicity and yet maintain the excellent efficacy that it has in the setting.
These data need to be produced. Another scenario would be that people would get started on intense induction treatment with higher-risk drugs such as ponatinib. Then if they are a good responder, they can switch to something that is lower-risk and that may be better tolerated.
Actually some of these trials are underway in Europe. Of course there's the question, are there still other drugs that will enter the CML space. There's one promising molecule called ABL001, a TKI with a different mode of action, which exploits an allosteric site rather than the catalytic center of the kinase.
It’s really conceptually very interesting. The expectation would be that this molecule has fewer side effects but still may be quite potent. This could be a very interesting development and add something to the armamentarium that we currently don't have.
I also think there will be patients whose diseases are just beyond the reach of a TKI alone for many reasons, maybe additional mutations or things that have to do with the host and metabolism.
In these cases, we'll still have to think about combination treatments and non-TKI treatments. Here, I think an honest answer is the labs have pulled out a lot of interesting leads, but so far, nothing has really made it into a serious clinical context, either because of side effects or because the target may not be as good in humans as it seemed from mouse models.
I think a lot more work is required here to find the best combination therapies and to define those pathways that need to be inhibited together with BCR-ABL. I think there's a field that we'll develop further and that will be the cutting edge.
The Wrap-Up
Dr. Kalaycio: Although an uncommon diagnosis, CML in chronic phase can be treated with an expectation for nearly 100% 5-year survival. For that reason, clinicians need to manage their patients expertly. Steps taken at diagnosis are critical to subsequent decision-making. Once treatment begins, close monitoring is required to screen for side effects as well as ensuring treatment success. With so many effective agents available, patients can be selected for the agent least likely to cause them long-term harm. Perhaps in the future we will learn which patients can stop treatment altogether.
Gentleman, thank you so much for your time. Every time I talk to you I learn something.
Moderator: Matt Kalaycio, MD1
Discussants: Michael Mauro, MD2; Michael Deininger, MD, PhD3
From Cleveland Clinic, Cleveland, OH1; Memorial Sloan Kettering Cancer Center, New York, NY2; Huntsman Cancer Institute, Salt Lake City, UT3
Address for correspondence: Matt Kalaycio, MD, Cleveland Clinic Main Campus, Mail Code R32, 9500 Euclid Avenue, Cleveland, OH 44195
E-mail: [email protected]
Biographical sketch:
Dr. Kalaycio has been published in numerous scientific publications including Bone Marrow Transplantation, Journal of Clinical Oncology, and Leukemia. He also is the editor of a book on leukemia and co-editor of a book on clinical malignant hematology. His research interests focus on testing new treatments for leukemia.
Dr. Kalaycio received his degree from West Virginia University School of Medicine in Morgantown. He completed his residency in internal medicine at Mercy Hospital of Pittsburgh and fellowships in hematology and medical oncology and bone marrow transplantation at Cleveland Clinic.
Michael Deininger, MD, PhD, is Professor and Chief of Hematology and Hematologic Malignancies for the Department of Internal Medicine and for the Huntsman Cancer Institute (HCI) at the University of Utah. He is an HCI investigator and member of the Experimental Therapeutics program. He has extensive experience treating patients with blood cancers, including chronic myeloid leukemia (CML) and myeloproliferative neoplasms, a group of blood cancers related to leukemia.
Dr
Dr. Deininger’s scientific focus is leukemia, specifically myeloproliferative neoplasms including chronic myeloid leukemia (CML). As a clinician-scientist with a translational research focus Dr. Deininger is heading an extramurally funded research laboratory that is dedicated to the study of signaling pathways, drug resistance and new molecular therapies in leukemia. Dr. Deininger’s work describing the selective effects of imatinib on CML cells provided the rationale for clinical trials that led to the approval of Gleevec as the first molecularly-based therapy for leukemia. Current work in his lab is focused on understanding the role of the bone marrow microenvironment in leukemia drug resistance, discovering novel therapeutic targets and developing more specific signal transduction inhibitors. Dr. Deininger’s work encompasses more than 170 articles in the peer-reviewed literature, including journals like Blood, Journal of Clinical Investigation and the New England Journal of Medicine. He has co-authored more than 10 book chapters, with contributions in leading textbooks such as deVita’s Principles of Oncology. He is a regular speaker at major international scientific meetings, such as the American Society of Hematology and the European Hematology Association and a peer reviewer for journals like Nature Genetics and Cancer Cell. His honors include the Alexandra Kefalides Prize for Leukemia Research and membership on the Editorial Board of Blood, the leading journal in Hematology. Dr. Deininger was named among the world's Highly Cited Researchers by Thomson Reuters in 2014.
Dr. Kalaycio: My name is Matt Kalaycio and I'm the Chairman of the Department of Hematology and Medical Oncology at the Cleveland Clinic. Today I’m joined by Drs. Mike Deininger, Division Chief of Hematology and Hematologic Malignancies at the University of Utah Huntsman Cancer Institute and Michael Mauro, Myeloproliferative Neoplasm Program Director at Memorial Sloan Kettering Cancer Center, New York, New York – together we will discuss practical issues surrounding CML diagnosis, management and treatment options including TKIs and investigational therapies.
Initial Presentation: Assessment & Treatment Options
Dr. Kalaycio: Often, the patient with chronic myelogenous leukemia (CML) gets admitted to the hospital, or an emergency consult is called, because the white count is up for a concern of leukemia. The treatment team sees the differential circulating blasts, and they worry about acute leukemia. Then the hematologist comes and needs to make a decision about whether or not to do a bone marrow biopsy.
When the patient presents in such a manner, I often see that bone marrow biopsies are not performed. I would like to start by asking where you both stand with regard to the necessity of a bone marrow biopsy at the time of diagnosis. I'll start with Dr. Deininger.
Dr. Deininger: I would strongly be in favor of a diagnostic biopsy as well as a smear. The reason is that I think this is the one and only chance to get a clear disease classification into chronic phase, accelerated phase, and blast crisis.
The third may be rare. There are occasional patients who have sheets of blasts who would not be seen on just a differential. For these patients, of course, the treatment decisions are going to be very different compared to a patient in chronic phase.
I think this is an opportunity that shouldn't be missed and I would always recommend that.
Dr. Kalaycio: Dr. Mauro.
Dr. Mauro: I couldn't concur more. With so much focus on the change in disease status from presentation to early response, I think understanding the scope of the disease at diagnosis, including the bone marrow, is essential and either revealing or reassuring. Although we focus mostly on early molecular response, when expectations go awry, not having as full a picture of the disease prior to treatment leaves you less informed about best treatment.
Making sense of accelerated phase features is a good example; the difference in outcomes between chronic phase patients and those who have morphologic features of accelerated phase—with or without cytogenetic features (clonal evolution)—compared with those with clonal evolution can only drive initial and long term treatment decisions. Without cytogenetics and morphology together, such key pieces of the puzzle are missing; it is worth it for the practitioner and patient to have all the data in hand at all times.
Dr. Kalaycio: Once the diagnosis has been made and the biopsy was not done and now you're seeing the patient having either been on hydroxyurea for a month or having had tyrosine kinase inhibitor (TKI) for a month, do you bother with the bone marrow biopsy at that point?
Dr. Deininger: Well, that is a very good question. I think we would probably still do it most of the time. I think it's very clear that the information you can gather from that is less valuable. I think one should try to make up for the omissions as much as possible. We would still go for it.
Dr. Kalaycio: Interesting. The other thing that happens a lot is a patient will be started on hydrea while you're waiting for BCR-ABL to return either by fluorescence in situ hybridization or quantitative polymerase chain reaction (PCR).
I wonder if you have your own set of internal guidelines for when hydroxyurea should be employed or not.
Dr. Mauro: If you look back at some of the original imatinib trials—the phase I trials—patients initiated TKI with higher blood counts, a median around 25,000 and up to 200,000, and responded with excellent tolerance.1 I think there's a bit of overapplication of hydroxyurea early in CML, prior to initiation of TKI, which often confounds and complicates the early myelosuppressive toxicity of TKIs. With use of more potent TKIs, there may be greater amounts of myelosuppression as the leukemic clone may clear faster.
I think we may get ourselves into a bit of a bind by overusing hydroxyurea and then needing to hold and lower TKI dosages quickly when that may not have been necessary had we simply deployed the TKI sooner.
My general rule would be to use hydroxyurea for symptoms if necessary or for more extreme counts where leukostasis is a concern. – Michael Mauro, MDMy general rule would be to use hydroxyurea for symptoms if necessary or for more extreme counts where leukostasis is a concern. The other question that often comes up is about tumor lysis and hydration, and how closely these need to be managed. The likelihood of tumor lysis is low in CML treated with TKIs, but more frequent early labs and good hydration are always the right thing to do; better safe than sorry!
Dr. Deininger: I second what Michael just said. To your question about whether we have an algorithm, I have to admit we don't. I think it's up to the discretion of the treating physician to initiate hydroxyurea or not.
Dr. Kalaycio: Sure. When you start hydroxyurea, do you routinely add allopurinol?
Dr. Deininger: We tend to do that. I know that some people think it's unnecessary. It's such a low-risk and low-cost intervention that I think that it's hard to get anything wrong here.
Dr. Mauro: Right, we tend to do the same. Whether we need to or not is a different question, I suppose.
Dr. Kalaycio: One more thing about the initial presentation and assessment of a patient with what you think might be a myeloproliferative disorder, CML, how important do you find it, Dr. Mauro, to measure splenomegaly and calculate a risk score?
Dr. Mauro: I think it's very useful information. The spleen size factors heavily into the risk score and the risk score does forecast response to a degree. We've looked at calculation of risk score in recent large observational studies and it is under-reported and underutilized. It winds up being useful for two reasons. One, it does set expectations for response, and US treatment guidelines (National Comprehensive Cancer Network [NCCN] guidelines) note that treatment choice may be different for low versus high risk Sokal score.2
I think the second and most intriguing reason to assess Sokal risk seems now to be the impact that risk score has on the ability to proceed to a treatment-free remission (TFR). There appear to be differences in outcomes in patients with high-risk versus low-risk disease despite both having what is required to proceed to TFR in trials, namely consistent and deep molecular remission over a number of years.
Given these implications, we really ought to be gathering the initial risk stratification and quantifying spleen size. It's an important part of our initial assessment.
Dr. Kalaycio: Dr. Deininger?
Dr. Deininger: I think there's a lot of agreement today. I absolutely think that measuring the spleen size ascertains that you've got all these diagnostic parameters available. I think that should be part of the initial evaluation and work-up and will allow some prognostication.
Testing, Risk Factors & Considerations of Treatment: Nilotinib & Dasatinib
Dr. Kalaycio: I'm starting with the less controversial questions to begin with. I think the next set have the potential for some more controversy. Before we leave the initial assessment of CML, do either of you have observations regarding referrals that you get about which you would like to either dispel myths or remind practitioners about best practices in patients newly diagnosed with CML?
I think, moving forward it would be really important to have documentation of concomitant risk factors such as smoking and so on that are essentially driving outcomes more than the TKI treatment or the CML itself. – Michael Deininger, MD, PhDDr. Mauro: I can mention one thing. I think when thinking about initial molecular diagnostic results, it is important to point out that testing should screen broadly for different fusions, namely P190 and P210, or variant (p230) transcripts. There are rare patients in chronic phase with non-p210 fusion who need to be followed with specific PCR. On this same topic, the measurement of transcripts at presentation (ie, before treatment) has become quite important and whereas formerly had not been emphasized, presently all patients should have ”baseline” transcript levels.
Dr. Deininger: I think one issue that comes up once in a while is that spleen measurements are done by ultrasound. Technically it's more accurate, but all the clinical risk scores and the prognostication is based on the old fashioned—but probably highly inaccurate—technology of palpation. This is what counts, this is the value to document. I think, moving forward it would be really important to have documentation of concomitant risk factors such as smoking and so on that are essentially driving outcomes more than the TKI treatment or the CML itself. Getting a good handle of those risk factors at diagnosis is also important.
Dr. Kalaycio: I think that's a very important point. That's where I was going to go next with this conversation. I was going to avoid the conversation about which TKI to choose as initial treatment. I think that's a debate unto itself.
I would like to ask you how you might assess cardiovascular risk before placing a patient on nilotinib. You got to that a little bit, Dr. Deininger. Could you expand on what else you might review as far as whether or not you feel a patient is a good candidate to start on nilotinib?
Dr. Deininger: Specifically, with regard to nilotinib, we would always get a baseline electrocardiogram. We would do a clinical exam. We would not do an echocardiogram just routinely in the absence of a cardiovascular history or any clinical evidence for heart failure or other cardiovascular issues.
We've adopted the practice of doing a lipid panel. Of course we would include fasting glucose as well. Some of these recommendations are probably somewhat on the soft side, because it's not yet clear what to do with the information.
On the other hand, I think for a patient who is being considered for nilotinib one wants to make sure that one really does the best to minimize the cardiovascular risk factors. Of course that would include smoking history and taking blood pressure and making sure that these risk factors are controlled.
If people have a presentation that is really out of whack in terms of their risk factor management, I would send them to an internist or even a cardiologist to help me optimize the cardiovascular prevention strategy.
Dr. Kalaycio: Great. Similarly, Dr. Mauro, how do you assess pulmonary risk before placing a patient on dasatinib?
Dr. Mauro: I think here we are focused on the less frequent and also less well understood potential toxicity of pulmonary hypertension, coupled with the more common risk of pleural and pericardial effusions.
I'm not sure how much we've learned in clinical studies looking at baseline chest X-rays or timing of X-rays during treatment. I think our best tool in the prevention and management of pleural and pericardial effusions is full discussion with patients about risk and what to look for, attention to any and all symptoms, and appropriate deployment of diagnostics as indicated.
It's interesting to consider whether baseline echocardiography for measurement of pulmonary pressures is warranted. I would say now we're on probably somewhat softer ground, first because on routine echocardiogram pulmonary pressure can't be measured readily unless there is some valvular regurgitation. As well, it is stated that pulmonary hypertension is only properly diagnosed by right heart catheterization. While I'm tempted to do routine echocardiogram studies, I think that such a recommendation still may be perhaps the realm of a clinical study. We need to explore that further. I think with dasatinib there may be certain patients at higher risk, although the data are somewhat limited. There seem to be certain conditions potentially associated with more pleural and pericardial toxicity, including cardiovascular disease and autoimmune disease. There may be circumstances during treatment—lymphocytosis, for example—that may be associated with greater risk. I think expectant management may still be the right approach and echocardiography and more aggressive diagnostics be reserved for patients in whom there might be much more clinical consequence.
Dr. Kalaycio: I'd like to pursue that a little bit further because sometimes the patients will come to us having already had an echocardiogram that may actually show some mild pulmonary hypertension and maybe they've got significant cardiovascular risk factors where you would otherwise be thinking about using dasatinib. Here's someone with pulmonary hypertension, at least by echocardiographic criteria, would that be enough to dissuade you from the use of dasatinib?
Dr. Mauro: I think it would certainly require significant consideration, understanding what the basis of the pulmonary hypertension is for that patient, and risk with adding dasatinib. I think the good news is the low incidence and the reversibility for the most part of dasatinib-associated pulmonary hypertension.
Again, I think the mechanism of action and the pathophysiology isn't completely understood, although there is the intriguing notion that imatinib has been reported to potentially mitigate pulmonary hypertension whereas dasatinib triggers it—a ”closed loop” if you will and an area ripe for research.
I would probably think that a patient with preexisting pulmonary hypertension in the new diagnosis setting might be the kind of patient for whom you really might weigh the pluses versus the minuses of a second generation TKI versus imatinib.
After Evaluation & Diagnosis: Following the Patient
Dr. Kalaycio: I agree. We have fully evaluated our patient and we've made the diagnosis. Now, we start therapy with a TKI based on patient risk profile and side effect potentials. It's time to follow the patient and determine next steps.
Current guidelines3,4 suggest monitoring quantitative PCR after 3 months of therapy and to gauge response. Dr. Deininger, how do you interpret and act on those results following the first 3 months of therapy?
Dr. Deininger: I think what you're getting at is the 10% mark that is a highly predictive value in terms of subsequent achievement of major molecular response and also overall progression free and overall survival.
We'll certainly get this data point. Then we'll put it in a clinical context. I think this context needs to take into account the initial BCR-ABL transcript level. I think Dr. Mauro mentioned that they always determine that. I think that is really good advice, so you can make a comparison with the diagnostic value.
One scenario, of course, is the patient is well below 10% and then things are just in the green range and you would wave people through and reassess in 3 months. If people are very high, I think then you have to ask yourself why that's the case.
Some people have a lot of toxicity issues. For example, they may not have been able to take the required amount of medication. It's not always easy to clearly distinguish between resistance and intolerance.
No matter what, I think it's going to be critical to consider a potential change in treatment if people are in the 70%—80% range. In my mind, there's a gray area.
These are the people who are maybe between 10%—20%. Here, I think this initial value can really help. If there's a significant reduction compared to baseline, I would not necessarily change at that point.
If there is literally no change compared to baseline, I would strongly consider a change unless I am concerned about other issues like noncompliance or drug interactions. What I'm trying to say in a rather long-winded way is that the 10% value shouldn't be seen as a dogma.
It still needs to be placed in a clinical context. One should not rush to any conclusions. Ten percent, 11%, and 9% are identical values in the world of PCR testing. One should not over-interpret that.
Dr. Kalaycio: I think that you're making an important point about absolutes in the interpretation of these tests. Dr. Mauro, how do you interpret the 3-month data that comes back?
Dr. Mauro: I agree with my colleague on the approach that it has to be put in a clinical context. I think what we've learned is the importance of the starting value and relative reduction and to not consider response milestones as black and white guides.
I think there are certain scenarios that require some caution. The patient who is on imatinib who is close but has not reached the landmark may be different that a similar patient on dasatinib or nilotinib who has not had significant reduction—that's probably a more pressing situation.
It's ironic that guidelines—maybe because of lack of options—don't encourage us to think about changing therapy early in someone who hasn't met milestones in 3 months when they've been put on a more potent agent.
I also think that careful consideration is needed regarding treatment intensity and the impact of interruptions, in conjunction with all the other facets—the rate of change, the absolute change from the patient's baseline, the starting response level and the timing of the PCR. We need to look at the actual day it's performed. If it's really not 3 months of therapy, we can misjudge.
Dr. Kalaycio: Right. Now, as you're monitoring patients, Dr. Mauro, at what point would you consider testing for kinase domain mutations?
Dr. Mauro: I think we have pretty good guidance from studies to date regarding when mutation testing is indicated and of higher yield. It's generally earlier on into treatment when we have clinical scenarios that are of greater risk, namely failure to achieve cytogenetic responses. In such scenarios if mutation testing informs treatment decision making it is very helpful.
Patients who have defined themselves as not achieving early molecular response—which we discussed earlier—especially someone who's had a second-generation TKI, warrants mutation testing in my view. Patients who don't achieve classical cytogenetic response landmarks at 6 and 12 (or 18) months, who thus have a higher residual volume of disease and perhaps as a function more clonal instability, I think also warrant attention.
I think we run into trouble when we start to, in an overly critical manner, assess patients’ longer term and deeper molecular response trajectories. Mutation testing becomes difficult or impossible for patients who are at or near major molecular remission (MMR). Mutation yield is generally very low for patients who have not lost MMR and also likely those with small volume of change around the MMR threshold. Looking ahead, I think further investigation into early time points and the setting of minimal residual disease may yield data to be able to predict potential resistance earlier than observing it clinically.
Dr. Kalaycio: Right. Dr. Deininger, if you're monitoring a patient and he’s missing milestones and you do obtain a mutation analysis and find a T315I mutation, do you offer that patient an opportunity to see how well they're going to do with ponatinib, or do you refer that patient to your transplant team for consideration of transplant as soon as a donor is available?As a transplanter, even I would not suggest transplant upon the recognition of the T315I. I would wait until all available TKIs have been tried and have either not worked or weren't tolerated. – Matt Kalaycio, MD
Dr. Deininger: I think we do the first and half of the second. We would certainly offer a trial of ponatinib or, if possible, a clinical trial unless there are insurmountable contraindications.
I don't really think there are any such circumstances with optimized cardiovascular management at the same time, and we may involve a cardiologist at that point to help us if there are cardiovascular risk factors.
We would also do a referral to a transplant center unless the patient is as per performance status, just not a transplant candidate or is too old. Otherwise, we would always make a referral, but we would not pursue a transplant as the first modality, provided that the patient is in chronic phase. Progression to accelerated phase or blastic phase would be looked at totally differently. Here we would certainly put people on a transplant course.
Dr. Kalaycio: I agree. As a transplanter, even I would not suggest transplant upon the recognition of the T315I. I would wait until all available TKIs have been tried and have either not worked or weren't tolerated.
As the two of you know, we're doing fewer and fewer transplants for CML these days. The results we're getting in folks who are failing all of these agents are not as good as they used to be when we were transplanting in the first 3 months of a new diagnosis. Bone marrow transplant is something to defer for as long as possible, at least in my mind.
TKI Dosage Considerations
Dr. Kalaycio: Now we touched a little bit on the side effects of nilotinib and dasatinib. We've talked about progressive disease, perhaps with or without mutations. Dr. Mauro, could you comment on where you think bosutinib might play a role in patients with relapsed or progressive CML?
Dr. Mauro: I think bosutinib remains an underutilized TKI. It is potent and thus offers good salvage activity. I think it offers a fairly distinct side effect profile making it a good choice for cases of intolerance. I think we've learned more about how to initiate bosutinib and manage its side effects; one element of this is that current trials start at slightly lower doses of bosutinib to avoid some of the early gastrointestinal toxicity.
Bosutinib continues to seek a place as a first-line therapy. I think this may be worthwhile as we are looking for the right balance of safety and side effect risk to maximize early response. Regarding the salvage setting—intolerance and resistance to other agents—I find myself speaking to patients about bosutinib as a potent and viable alternative. Bosutinib offers a similar spectrum of activity as dasatinib but can allow us to avoid cardiovascular toxicity and/or allow clearance of pleural and pericardial toxicity occurring with dasatinib.
The one place I would say bosutinib may not hold up as strongly to its competition would be as a third-line therapy after failure of a second-generation TKI. I think while there is reasonable activity with bosutinib in this setting of somewhat fairly drug-resistant CML, the performance of ponatinib is better. There is likely often a struggle with the long-term safety question of ponatinib, making some feel that a trial of bosutinib is often a worthwhile, logical, or even necessary step before ponatinib. My concerns about this approach are related to treatment decision-making based on safety and theoretical risk of adverse effects rather than efficacy and risk of progression; these have to be weighed carefully and appropriately against each other.
Dr. Kalaycio: That's an interesting perspective. You mentioned starting with a lower dose. Dr. Deininger, I'd like to ask you your thoughts about dose modifications of the TKIs in general. It used to be anathema to either start with a low dose or to maintain patients on a lower dose.
I'm aware of at least some data suggesting that those patients who tolerate these TKIs poorly particularly with regard to myelosuppression can be treated long term at lower than typically prescribed doses without adverse effect. What are your thoughts surrounding dose modification of the TKIs?
Dr. Deininger: I would stratify according to TKIs. I think imatinib at 400 mg is probably just rightly dosed, maybe not even at the optimal dose. It could be 600 mg. If you go to 300 mg, that is still acceptable.
If you go to 200 mg, I think I would make double sure that I monitor patients very frequently. People should be, in my mind, in a major molecular response to treat them with 200 mg long term. With the second-generation TKIs, it's a little different.
I cannot really speak to bosutinib because I'm not aware of a lot of data on those modifications. What seems to be clear from dasatinib is that the initial recommended dose of 100 mg is quite high, especially in older people.
I believe that's been corroborated by a study that hasn't been published yet. Apparently the drug excretion in the older individuals is quite a bit slower than in younger people. I think in our practice and in other people's practices as well, about 50% of those people end at doses substantially lower than 100 mg, maybe 50 mg, maybe 40 mg, some even 20 mg per day.
I think in the case of dasatinib you have a lot of maneuvering space. You can adjust the dose according to tolerability and molecular response. With nilotinib I think it's kind of similar. We have a few patients who are on maybe 150 mg twice a day because they have issues in terms of side effects at the higher doses.
I should again thoroughly qualify that by saying if you give up through dose reductions the goal of a major molecular response, then I think you really have to think about the strategy in general because at that point, I think you basically say "this patient is not ever going to be reaching a safe haven and is not tolerating inhibitors well enough to get him there."
That's quite a significant statement to make. Myelosuppression is frequently a situation where you cannot deliver enough dose but you also don't get a good response, whereas other side effects like pleural effusions or excessive fluid retention may well be cause for dose reduction and yet responses may be acceptable. Maybe Dr. Mauro could chime in here and share his experience.
Dr. Mauro: I agree. I think some patients, based on their disease status, may not tolerate the dose that is going to—at least in theory—get them to a deep molecular remission. The biggest problem we often face is myelosuppression. Some patients may have indolent disease and still have a good outcome, at least in the short to medium term, with lower doses of drug. In general I try to work through myelosuppression, especially early, as the combination of myelosuppression and suboptimal or response failure is dangerous.
I think treatment shouldn't be too chaotic and change too much, if possible. That was a very nice summary of the way we view the doses. There's still a fair bit of flexibility amongst the TKI doses, although I don't advocate for starting low and titration up.
In general I think, with regard to optimal TKI dose, we might have overshot; for example, with dasatinib. I think we might have overshot with bosutinib. Trials ongoing now initiate bosutinib at 400 mg rather than 500 mg. That's what I was alluding to.
Treatment Free Remission: Approaches
Dr. Kalaycio: Very interesting discussions. As we wind down the conversation, I want to get to the idea of stopping therapy. We're all aware of data that suggest it's at least possible for a proportion of patients to stop treatment.5
However, the guidelines3,4 such as they are, suggest that stopping should not be done outside the context of a clinical trial. I think most of us would agree with that. I wonder in a practical sense, Dr. Mauro, in your practice, do you have your own set of internal criteria for stopping a TKI and observing patients in the absence of any therapy?
Dr. Mauro: I think I try to incorporate all the experience we have to date when considering this question. That being said I'm a little hesitant to agree entirely with the current thinking regarding retreatment during a Treatment Free Remission trial, which is waiting until patients lose MMR in order to resume treatment but agree it might be hard to retreat based on lesser degrees of molecular relapse. I just worry a little bit about that amount of proliferation without treatment.
On the other hand I think we're coming to the realization that TFR may be feasible in patients who are simply nonproliferative and have low volume of disease based on newer data that patients may not need to have consistent "complete molecular response" or disease reduction below 4.5 logs in order to consider TFR. This will mean more patients may be eligible for such a strategy.
With that being said, I strongly advocate for discontinuation occurring in trials still, especially in the United States. I think we still have to ensure regular monitoring and not run the risk of loss of CML remission as a result of this endeavor. I think we have to have a clear message that this is still an investigational approach.
As we apply TFR strategies more broadly and explore it in patients with different circumstances, rather than those already studied in clinical trials, we may see slightly different results. We may need to exercise more caution.
Outside of a trial I don't discontinue TKI therapy unless there's a clear medical indication and there's no way around it, such as pregnancy or other illness that precludes TKI treatment. I would still pursue TFR only in a clinical trial.
Dr. Kalaycio: Dr. Deininger, I'll give you the last word for your thoughts regarding both treatment discontinuation and the future of CML therapies.
Dr. Deininger: As far as TFR is concerned, I agree with Dr. Mauro. Unfortunately, in the United States we're not quite as far advanced as the Europeans at monitoring diligence. There's a bit of a concern that if we elevate that into daily practice, we may see that patients are not monitored frequently enough and then relapse, but that doesn't get caught. You could imagine a scenario where we're actually seeing a decline in outcomes because people are discontinued and then not restarted if they have a recurrence at the molecular level.
All in all, where this is going, I think it will be really interesting to see more confirmatory data for second-generation TKIs. Let me put that differently: whether a similar proportion of patients in deep molecular response can maintain responses and have TFR if they needed a second-generation TKI to get there rather than just imatinib.
If so, this would be very promising. That would mean that second-generation TKIs actually somewhat impact the natural history of the disease and its biology. Then TFR may become a reality for a substantial proportion of our patients.
I don't think we have the data yet, but at least there are some suggestions that this may be the case. If you do the math, you will still see that many patients with CML will never reach a state where they can consider TFR.
They'll never get into a deep molecular response or they will have been diagnosed with accelerated phase. For all these patients who are not candidates for TFR, we still have to think about treatment optimization in order to make them catch up with the rest.
Now I have a couple of interesting developments. One is clearly to see whether ponatinib administered at a lower dose will have a more practical, therapeutic window in terms of toxicity and yet maintain the excellent efficacy that it has in the setting.
These data need to be produced. Another scenario would be that people would get started on intense induction treatment with higher-risk drugs such as ponatinib. Then if they are a good responder, they can switch to something that is lower-risk and that may be better tolerated.
Actually some of these trials are underway in Europe. Of course there's the question, are there still other drugs that will enter the CML space. There's one promising molecule called ABL001, a TKI with a different mode of action, which exploits an allosteric site rather than the catalytic center of the kinase.
It’s really conceptually very interesting. The expectation would be that this molecule has fewer side effects but still may be quite potent. This could be a very interesting development and add something to the armamentarium that we currently don't have.
I also think there will be patients whose diseases are just beyond the reach of a TKI alone for many reasons, maybe additional mutations or things that have to do with the host and metabolism.
In these cases, we'll still have to think about combination treatments and non-TKI treatments. Here, I think an honest answer is the labs have pulled out a lot of interesting leads, but so far, nothing has really made it into a serious clinical context, either because of side effects or because the target may not be as good in humans as it seemed from mouse models.
I think a lot more work is required here to find the best combination therapies and to define those pathways that need to be inhibited together with BCR-ABL. I think there's a field that we'll develop further and that will be the cutting edge.
The Wrap-Up
Dr. Kalaycio: Although an uncommon diagnosis, CML in chronic phase can be treated with an expectation for nearly 100% 5-year survival. For that reason, clinicians need to manage their patients expertly. Steps taken at diagnosis are critical to subsequent decision-making. Once treatment begins, close monitoring is required to screen for side effects as well as ensuring treatment success. With so many effective agents available, patients can be selected for the agent least likely to cause them long-term harm. Perhaps in the future we will learn which patients can stop treatment altogether.
Gentleman, thank you so much for your time. Every time I talk to you I learn something.
1. Peng B, Hayes M, Resta D, et al. Pharmacokinetics and pharmacodynamics of imatinib in a phase I trial with chronic myeloid leukemia patients. J Clin Oncol. 2004;22(5):935–942.
2. National Comprehensive Cancer Network guidelines for treatment of cancer by site. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp. Accessed August 20, 2015.
3. O'Brien S, Radich JP, Abboud CN, et al. Chronic myelogenous leukemia, version 1.2015. J Natl Compr Canc Netw. 2014;12(11):1590–1610.
4. Baccarani M, Deininger MW, Rosti G, et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood. 2013;122(6):872–884.
5. Mahon FX, Réa D, Guilhot J, et al. Discontinuation of imatinib in patients with chronic myeloid leukemia who have maintained complete molecular remission for at least 2 years: the prospective, multicentre Stop Imatinib (STIM) trial. Lancet Oncol. 2010;11(11):1029–1035.
1. Peng B, Hayes M, Resta D, et al. Pharmacokinetics and pharmacodynamics of imatinib in a phase I trial with chronic myeloid leukemia patients. J Clin Oncol. 2004;22(5):935–942.
2. National Comprehensive Cancer Network guidelines for treatment of cancer by site. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp. Accessed August 20, 2015.
3. O'Brien S, Radich JP, Abboud CN, et al. Chronic myelogenous leukemia, version 1.2015. J Natl Compr Canc Netw. 2014;12(11):1590–1610.
4. Baccarani M, Deininger MW, Rosti G, et al. European LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood. 2013;122(6):872–884.
5. Mahon FX, Réa D, Guilhot J, et al. Discontinuation of imatinib in patients with chronic myeloid leukemia who have maintained complete molecular remission for at least 2 years: the prospective, multicentre Stop Imatinib (STIM) trial. Lancet Oncol. 2010;11(11):1029–1035.
