Today we discuss chronic myeloid leukemia (CML) treatment and research. CML is a type of leukemia that is characterized by the unregulated growth of myeloid cells that grow and divide in the bone marrow. CML is associated with a driver chromosomal mutation called the Philadelphia chromosome that results in the fusion of the BCR and ABL genes, resulting in continuous activation of the ABL tyrosine kinase. We are speaking with Dr. Michael Deininger, chief of the division of hematology at the University of Utah School of Medicine and expert on the treatment and study of leukemia. Dr. Deininger studies the biology of CML including resistance mechanisms to treatment.
—Interviewed by Anna Azvolinsky, PhD
Cancer Network: Imatinib, the first treatment that specifically inhibits the BCR-ABL kinase was first introduced into practice in 2001. Now, there are several next-generation tyrosine kinase inhibitors including dasatinib, nilotinib, and the newest one is bosutinib that are available for treatment of CML. Could you briefly describe the improvement in patient response and long-term survival rates that have results from these treatments?
Dr. Deininger: Well, as you pointed out, the first tyrosine kinase inhibitor or TKI that was introduced was imatinib. Imatinib really revolutionized how we treat CML. It increased survival rate to an unprecedented level. In previous years, patients would be proposed to undergo an allogeneic stem cell transplant if they were candidates and now we treat everyone who is newly diagnosed with chronic-phase CML with a TKI. So imatinib made the really big, big dent in helping to manage the disease. It is a very good drug, however, there is some degree of resistance, so second-line or second-generation TKIs were developed to contend with resistance and those are dasatinib, nilotinib, and more recently bosutinib. The first two of these agents, dasatinib and nilotinib, have also been approved for the treatment of newly diagnosed patients. So the big distinction we need to make here is the frontline setting of chronic-phase patients vs those patients who have failed imatinib therapy. As far as the frontline setting of newly diagnosed patients goes, the second generation inhibitors, dasatinib and nilotinib have improved response rates (ie, cytogenetic response or molecular responses), which are basically an indicator of how deep the leukemic clone is suppressed by the treatment. And if you look at these endpoints, specifically the so-called major molecular responses, the newer generation inhibitors, dasatinib and nilotinib, have improved these responses compared to imatinib. There is also a slight reduction of progression to the accelerated or blastic phase. However, as of now, there is no difference in overall survival. This might reflect the efficacy of these drugs to salvage patients who failed imatinib. The bottom line here is more profound and for the most part, faster responses that have not perhaps not yet, translated into the improvement of overall survival.
Cancer Network: What are the differentiating factors of these tyrosine kinase inhibitors? How should clinicians think about choosing which inhibitor to give to which patient? And I realize that some of these agents are only approved for first line and some for subsequent lines, but are there differences in terms of safety profiles for example?
Dr. Deininger: So if we look at frontline treatment, typically patients are diagnosed in the chronic phase, at least in the Western world, then the differentiating features between imatinib, dasatinib, and nilotinib are that the later generations induce deeper molecular responses. In a given patient, the question arises, which agent to chose? The first consideration is, in my mind, the comorbidities of the patient that may make a clinician chose one agent over another. For example, one of the side effects of nilotinib is to increase blood glucose levels; thus in a diabetic patient nilotinib may not be the best choice. Dasatinib on the other hand, is known to cause pleural effusions. These are fluid collections between the chest walls and the lungs, and if somebody has a history of that, obviously dasatinib is then not a good idea. Imatinib has fluid retention, that is lower extremity edema and periorbital edema, as a leading and sometimes significant side effect. So patients with any of these pre-existing conditions are not ideal candidates for imatinib.
In reality however, many times even these considerations will still leave us with a choice between the three inhibitors. Here, I think the considerations are complicated by financial constraints or financial considerations. The second-generation inhibitors are more expensive than imatinib and they will become even more expensive in comparison with imatinib once imatinib runs out of its patent. So we are faced with the dilemma of having to choose between an extremely active second-line inhibitor with better short-term responses but as yet no impact on overall survival, and imatinib as an agent that is certainly a little less active but on the other hand, has an extremely impressive long-term safety record. So to give definitive recommendations is somewhat difficult and it will boil down in some instances to what individual physicians are comfortable with—ie, which drugs they know.
Another consideration that I want to bring up, although it is not completely evidence-based is the aggressiveness of the disease. Chronic-phase CML is not a homogeneous disease. There are patients who are bordering advanced phase whereas other patients have a very early, rather benign disease. In the latter, imatinib may well suffice, whereas for the former candidates I would chose a second-generation inhibitor. All three agents are approved, and I think what we need to learn as a community is how to use these agents as a treatment approach to CML rather than individual agents. For example, we don’t have good data to tell us whether we might start everyone on imatinib, and then take an early decision point to see which patients actually require a second-generation inhibitor. Conversely, we don’t know what would happen if we treated everybody with a second-generation inhibitor such as dasatinib or nilotinib and then de-escalated it in those patients with a very good response. I do hope that clinical trials will be constructed to clarify these questions. The hope, of course, is that the higher rate of complete molecular responses (that is, even with the most sensitive technology, there is no detectable residual leukemia), will eventually lead to the ability of more patients to discontinue treatment altogether. But, while this is possible and of course everyone is hoping for that, I think we need to be cognizant that this is a research question, and it is very hard to base definitive recommendations on what is essentially evidence that still needs to be gathered.
Cancer Network: Along those lines, the long-term follow up data of the BELA trial which showed that long-term outcomes were similar for both imatinib or bosutinib, which is the newly approved next-generation CML agent. Bosutinib was approved in September for those CML patients who are resistant or intolerant to prior therapy including imatinib. What do the BELA results suggest about targeting the resistance mutations that develop in patients as a result of treatment?
Dr. Deininger: The BELA trial was a trial in newly diagnosed patients, and it failed in its primary endpoint, that is the complete cytogenetic response rate after 12 months of treatment. There are some reasons to believe that it failed because the treating physicians on the trial may have responded too aggressively to side effects, especially diarrhea, and reduced the dose or even pulled patients off the trial because of that. So I would predict that it would have met the primary endpoint had that not happened. However, as far as long-term outcome is concerned, the long-term outcome here means overall survival, the picture is quite similar to that in the other second-line tyrosine kinase inhibitors. That is, there is yet no difference visible and while with longer term follow-up, the difference might well show up, at this point, we do not have the data. A quick qualifier here is that the hope, of course, is that the higher rate of profound molecular responses, that is, even with the most sensitive technology, there is no detectable residual leukemia, will eventually lead to the ability of more patients to discontinue treatment altogether. But, while this is possible and of course everyone is hoping for that, I think we need to be cognizant that this is a research question and it is very hard to base very definitive recommendations on what is essentially evidence that still needs to be gathered.
Cancer Network: What do we know now about the range of resistance mechanisms to these agents? Are most of these within the BCR-ABL fusion gene or are there other types of resistance mechanisms that develop?
Dr. Deininger: There are clearly other resistance mechanisms. Those mechanisms that are due to changes in the BCR-ABL fusion gene itself are typically point mutations that prevent binding of the TKIs to the target. But, there is an increasing realization in the field that mutations are only part of the story. There are clearly patients who fail TKIs without any detectable changes in the BCR-ABL genes. The thinking is that in these cases, other mechanisms substitute for BCR-ABL activity. So the cells learn to use another transforming principle and of course, in these patients, addressing BCR-ABL is not that promising anymore.
Cancer Network: As a last question, what are other therapies in development for CML that are promising with a different mechanism of action, so ones that don’t target the ABL gene? Are there any?
Dr. Deininger: I think it has been difficult to address this type of resistance, and it is most likely quite heterogeneous, making it more challenging to develop specific treatments. One agent that has just recently been approved is omacetaxine. That is a drug that has been around for a very long time. It is a plant alkaloid-based nonspecific protein synthesis inhibitor. It has shown activity in patients in situations where patients have not responded to approved TKIs anymore, including patients without mutations. Beyond that, I think there is currently a lack of understanding of the pathways that drive this type of non-BCR-ABL resistance. So you could enumerate a very long list of agents that have been tried with various degrees of success such as hypomethylating agents such as decitabine, azacitidine, HDAC inhibitors, arsenic and conventional cytotoxics. I think what we need to do is study these relapses more carefully and hopefully identify common pathways that are consistently activated in these circumstances, or at least in a majority of patients. I would predict that this would lead to combination trials of BCR-ABL tyrosine kinase inhibitors with PI3-Kinase inhibitors or maybe JAK2 inhibitors, but there is clearly a big knowledge gap here at the moment. So for practical purposes, I think patients who fail tyrosine kinase inhibitors including the most advanced one, which is called ponatinib, don’t have very good options at the current time, and they should definitely be considered for a stem cell transplant.
Cancer Network: Thank you so much for joining us today, Dr. Deininger.
Dr. Deininger: You are very welcome.