Novel Induction and Maintenance Strategies in Acute Myeloid Leukemia — Jeffrey E Lancet, MD

DR LANCET: Well, thank you very much for the invitation to speak today and to review some of the latest and greatest developments in acute myeloid leukemia. Today’s focus will be on some of the novel induction and maintenance strategies that have come to the forefront over the past few years. And it's a good time, I guess, to be an AML provider compared to maybe how it was 15 or 20 years ago when we didn’t have a whole lot, unfortunately, to offer our patients. Now we’ve seen really quite an explosion of novel therapies for AML in the last 5 years alone and we’re very excited about what the next 5 years holds in terms of novel and targeted therapies, immunotherapies, cellular therapies. So, there’s a lot going on right now. And I think the future is certainly looking brighter for our patients with this devastating disease.

So one area that I thought we could focus on today, and part of this discussion, would be the area of secondary AML and this really is a bit of a confusing term. And what it refers to is really a composite definition of AML, either from prior malignancy, so AML arising from a prior myeloid malignancy such as MDS or myeloproliferative neoplasms or AML that arises following cytotoxic chemotherapy or radiation therapy. And this is traditionally referred to as therapy-related AML.

So both AML that arises from a prior myeloid malignancy or following prior chemoradiation therapy fall into the umbrella of secondary AML. And the most common type of secondary AML would be AML with MDS related changes, and these include patients with prior MDS, patients with MDS-related cytogenetics, such as deletion 7 or deletion 5 or a complex karyotype, or patients with morphologic multilineage dysplasia. So all of these subtypes of MDS-related changes fall within the overall category of secondary AML. We also have a smaller fraction of patients with AML from prior myeloproliferative neoplasm and AML that's therapy-related, as I just mentioned.

Now, in large part, secondary AML has been defined by these prior definitions. And what we're learning in the molecular age is that we can better define AML by the genetic characteristics as opposed to purely the clinical or the microscopic characteristics. And an important paper came out from Dr Lindsley in his laboratory in Clinical Partners a few years that demonstrated a very precise signature of genetic alterations that would define secondary AML as traditionally defined. So, in other words, patients that have or had secondary AML by the traditional definitions, very often carried specific mutations that were distinct from the mutations that you would see with de novo AML. And these mutations included many of the mutations you might see in myelodysplastic syndrome for example, so SRSF2, ZRSR2, SF3B1, ASXL1 and others, are commonly seen in secondary AML and are helpful in distinguishing secondary AML from de novo AML. But probably more important is that the fact that even de novo AML that is associated with these secondary like mutations behaves like secondary AML.

So, if you have a patient with de novo AML that happens to carry an ASXL1 mutation or a UTAF1 mutation or any of these certain mutations, they’re clinical behavior will mimic that of a patient with true secondary AML as traditionally defined. So I think we’re learning, we’re entering an era where we can define these diseases more precisely through genetic testing and molecular profiling than purely by clinical characteristics.

Now going back to a little more clinical discussion about the epidemiology of secondary AML, it's a fairly common phenomenon and it’s certainly something that increases with age and older patients are more likely to have secondary or therapy-related AML than younger patients. And as you can see here, there’s an incidence of approximately 20 to 30% of all AML cases being associated with a secondary AML phenotype. So, it's a clinically common scenario to have secondary AML. So, certainly, you’re going to want to be able to target the appropriate biologic and clinical targets in this disease because of the overall epidemiological importance.

And we know that historically, secondary AML is a poor subtype of AML prognostically. And these are data from a registry study that really depicts the fact that de novo AML does better than secondary or therapy-related AML. Again, I’m using the term somewhat interchangeably, but therapy AML is considered a type of secondary AML, but in some registry studies it's considered as a somewhat distinct entity. But the message here is that secondary AML, as defined traditionally, does more poorly than de novo AML. And the difference is not quite as obvious in the older age population where there’s really no subtype of AML that does well. So even the de novo AML patients have poor outcomes, very similar to what would be seen in secondary AML or therapy-related AML.

So, the overall message is that secondary AML has a worse clinical outcome, but that outcome is a little bit more blunted in the older age population where outcomes are quite poor anyway. And, also, taking into consideration that more patients who are older do have secondary AML or AML that arose from MDS or, as we’re learning, AML with secondary-like mutations that I just described earlier which may explain some of the overlapping in the survival curves.

Now when we’re talking about therapeutics, we often don’t distinguish between AML subtypes unless, of course, it's acute promyelocytic leukemia or something of the like. So traditionally, we had never looked at differences in treatments for patients with secondary versus de novo AML.

But there is a new drug that's hit the market relatively recently that has focused primarily on patients with secondary AML, that drug is known as CPX-351, also known as liposomal formulation of daunorubicin and cytarabine. And what this drug is really, is nothing more than a liposome, a bilamilar liposome that encapsulates both cytarabine and daunorubicin together in a fixed molar concentration ratio. And that molar concentration ratio is what is felt to be synergistic as a ratio for killing leukemic cells.

So, once it was determined what the appropriate molar ratio was for synergy, then we were able to package both of these drugs together in that particular molar ratio to allow for drug delivery directly through the liposome at the preferred synergistic ratio between the two drugs. And that's the whole concept behind CPX-351.

And preclinically, as you can see here, there’s actually a lot of interest in understanding whether this drug is actually behaving the way it's intended. And we think that is. So shown here from preclinical modeling using cyan fluorescent lipid labels that show liposomes, in the blue, are taken up whole by the cell into vacuoles, into these leukemia cells, assuring that the delivery of these two drugs occurs in the preferred synergistic molar ratio. And that subsequently, the liposomal structure is broken down and the free drug is released intracellularly and accumulates in the nucleus. So the concept of drug delivery of both of these drugs simultaneously, in the appropriate molar ratio, I think is demonstrated in this preclinical model here which is kind of interesting.

And in the clinical setting, we did a Phase I trial several years ago testing CPX-351 in patients with relapsed and refractory AML. And I think one of the most important points to come out of this study was that CPX-351 maintained the synergistic molar ratio during the course of treatment. As you can see from the upper curve here, this relative molar ratio of 5:1 was really maintained for up to 24 hours after the final dose of the CPX-351 was administered.

And maybe just as importantly, you can detect the plasma levels of both cytarabine and daunorubicin well after the final dose of CPX is administered, and even up 7 days at the completion of the infusion of the compound you can still see circulating daunorubicin and cytarabine in the plasma. So, this suggests a unique pharmacokinetic profile of the drug that allows it to stick around for a while and to be maintained at the synergistic molar ratio that we think is important in dictating its overall activity.

So this led to a randomized Phase II trial comparing CPX-351 against 7+3 in older patients with AML, not selected for any specific type of AML in this trial, but just over age 60 — between ages 60 and 75, with the primary endpoint being response rate. And this was a 2:1 randomization between CPX and 7+3. And just to summarize the results of this trial, we did observe an overall benefit within the CPX-351 arm, but most notably in the subset of patients that had traditionally defined secondary AML, as I described to you before, primarily patients with AML from MDS or therapy-related AML, the outcomes were much more distinct between CPX and 7+3, with a longer overall survival and a longer event-free survival, as well as a higher response rate favoring the CPX in this secondary AML subset.

So this piqued our interest in further assessing the efficacy of CPX-351 in a poor prognostic subset of AML. So, we took it into a Phase III trial of patients with secondary AML where we randomized, 1:1, between CPX and 7+3, again in older patients with newly diagnosed but secondary AML, either AML from MDS or therapy-related AML. And these were patients who were felt to be more on the fit side and be able to tolerate more intensive therapy. And many of these patients were considered transplant candidates as well.

So the primary efficacy results of this trial are shown here, and you recall that the primary endpoint was overall survival. And at the end of this trial we did observe that the CPX-351 demonstrated superior overall survival compared to 7+3 traditional induction, with approximately a 3-1/2-month overall survival advantage favoring CPX. And I think more importantly, a reduced hazard risk of death by about 30% favoring the CPX arm over the entire course of the study and the follow up. So, clearly, a survival improvement attributed to CPX in this patient population studied in the Phase III trial.

We also recently completed 5 years of follow up and we were pleased to see that the efficacy results were holding up quite well at the end of 5 years. And as you can see here, there remains a significant difference in survival between patients receiving CPX and patients receiving 7+3, as shown in the purple and the yellow curves respectively here. And I think noteworthy is the fact that a fraction of patients were still alive and well at 5 years, so almost 20% of patients with CPX were still living at 5 years compared to only 8% with the 7+3.

You may be interested in hearing about the response rates, and not surprisingly, as is often the case with chemotherapy trials that demonstrate overall survival with one drug versus the other, the response rates do tend to correlate with survival, and this was no different here. So in the Phase III trial that I just described, the overall response rate was close to 50% for the CPX treated patients, and about 33% for the 7+3 treated patients. So, a significant difference in the overall response rate as well as the overall survival.

And what I think is important to recognize also, is that this is not a drug that should be considered low intensity or lite therapy because it really isn’t. And we know that because patients who go through it have expectedly marked and prolonged cytopenias prior to the evolution of a full remission. And as you can see here in our Phase III trial, we demonstrated that there was prolonged count recovery, and actually, CPX-351 was associated with a longer time to count recovery, both neutrophils and platelets, compared to 7 + 3.

So, the efficacy was higher, but the hematologic toxicity was also higher in the sense that there was a delay in blood count recovery. Now this did not translate into any overt signs of increased mortality at an early timepoint or other clear toxicity signals, but it does serve as a warning that this is not therapy to be taken lightly and patients do require typically to be in the hospital for the typical length of time. And that attention needs to be paid to these blood counts and the appropriate prophylactic measures taken.

And then another important offshoot of this trial was the fact that many of the patients eventually went on to a transplant. And as I mentioned at the beginning, we were accruing patients to the trial who were considered fit enough for intensive therapy, many of whom were also considered fit enough for an allogeneic transplant. So a fraction of these patients eventually went on to a transplant as a potentially curative modality once they had achieved a remission of some type of with the frontline therapy.

So, we were interested to find out whether there was a difference in transplant outcomes, post-transplant outcomes, among patients who were assigned originally to the CPX arm versus the 7 + 3 arm. And what we found interestingly, was that again in a non-randomized fashion, that if you landmarked patients from the time that they received the transplant, the ones that received CPX-351 as their initial therapy lived longer than those who had received 7 + 3 as initial therapy. So even though both sets of patients went into transplant with what we think is a similar clinical situation in terms of remission and performance status and all that, the CPX-351 patients seemed to fare better. And we’re not sure why that is. The obvious hypothesis would be that CPX-351 induces a deeper molecular remission so that there’s better overall disease control at the time of transplant. And that hypothesis is currently being explored. Or another possibility would be that the patients that went to transplant perhaps were in better physical and functional condition following the CPX treatment than the 7 + 3 treatment.

So these are issues that need to be ironed out to gain a better understanding as to why transplant outcomes were better among the CPX-351 patients. But I think the data that are evident at this point is that the CPX-351 can serve as an effective bridge to transplant in some patients being treated with curative intent with secondary AML.

And there have been a lot of new clinical trials that have been spawned since the Phase III trial of the CPX that demonstrated superiority over 7+3. And some of these are depicted here.

There’s a lot of interest in combining targeted agents with CPX-351. So some of these targeted agents include FLT3 inhibitors such as quizartinib and more recently, gilteritinib. Glasdegib, of course, is a hedgehog pathway inhibitor that's being tested in combination. Gemtuzumab, the monoclonal anti-CD33 conjugated drug antibody. Ivosidenib and enasidenib, the IDH1 and IDH2 inhibitors respectively. And then venetoclax, which, of course, is the newest drug to the market in AML that appears to be very, very promising, especially in combination with hypomethylating agents. Also now being combined with CPX-351 in clinical trials.

So I thought as a next step I would focus on another very important and intriguing development over the past year, and that relates to maintenance therapy in AML. And I think we’ve recognized for a long time that while remissions are achievable in many cases of patients with AML, the large majority of patients will, unfortunately, have a relapse. And this is a particular problem in older patients, or people with an adverse molecular profile where you may be able to achieve remission, but the relapse occurs regularly and early.

So while potentially curative allotransplant is an option for some patients with AML, particularly younger patients and patients with a donor and patients who are in remission of course, the fact remains that allotransplant is still not suitable for the majority of patients, especially older patients who may not be fit enough or who just may outright refuse to go through the rigors of a transplant. So to have an option, a post-remission option, is a very attractive consideration for these patients.

I think we say that we need effective AML maintenance therapy strategies in order to reduce relapse and prolong survival. And we have to be able to do so without causing too much toxicity or impact quality of life. But, unfortunately, up until recently, there have really been no clear and demonstrable and reproducible data showing improvement in overall survival amongst patients with AML receiving maintenance therapy.

So it's really been kind of the Achille’s heel of AML therapy is to find something that you can give in the post-remission setting that moves the needle and makes a difference and improves the overall outcomes of these patients.

So that brings us to a new drug, which is known as CC-486, or also orally bioavailable azacitidine. We’re all familiar with azacitidine in the MDS setting, where it was approved many years ago, and its development as one of the backbone drugs in AML as well, especially in older adults who may not be candidates for more intensive therapy. And as you know, the aza/venetoclax combination looks extremely good and promising and has demonstrated efficacy to the point where it was recently FDA-approved as frontline therapy for patients with poor risk, elderly subtypes of AML. Anyway, the ability to utilize azacitidine as a lower intensity therapy is very attractive because it can be given for a long period of time. It has a favorable toxicity profile. And it's been shown to be effective in the upfront setting.

So, as an oral formulation, it has the potential for an extended dosing schedule, even more so than what you would give parenterally, and the fact that parenteral aza has a short half-life means that its overall exposure may not be that much when you can only give it for 5 or 7 days. So being able to give it for an extended dosing schedule as an oral drug is quite attractive because you could potentially increase leukemic cell exposure to azacitidine over a longer period of time.

Early studies, looking at this regimen, demonstrated that you could actually give this drug safely in both 14-day and a 21-day schedule and that with some evidence of efficacy and feasibility in these earlier studies, it made sense to consider it as maintenance therapy later on.

So, this was undertaken in the form of the Phase III QUAZAR study, which was a fairly simple clinical trial, Phase III clinical trial, that was presented last year at the annual ASH meeting as a late-breaking abstract. And this was a Phase III randomized multicenter, placebo-controlled, double-blind Phase III trial that compared CC-486 or oral azacitidine, against placebo in older age patients with AML and first remission following induction with intensive chemotherapy.

Now these were patients who were considered to be ineligible for subsequent allogenic transplant. So, patients who had undergone successful induction therapy but were not planning a subsequent transplant. And the overall survival was the primary endpoint, with other key endpoints including relapse-free survival, quality of life and safety.

So, these are the baseline characteristics of the patients who entered the study. And as you can see both arms of treatment are well balanced with respect to the patient characteristics. Similar numbers of patients were over age 65 and similar percentages of the patients had various WHO AML subtypes, including AML with MDS related changes and others. A relatively small number of patients had poor risk cytogenetics by NCCN criteria. And as you can see the other factors were quite well matched between the 2 arms of the study.

So, the primary endpoint of the study, as I mentioned, was overall survival. And, in fact, this was demonstrated. This was overall survival from the time of registration. And you can see that the patients assigned to the CC-486 drug lived longer than patients assigned to placebo as maintenance by approximately 9 months. And also important was the fact that the 2-year survival rate was significantly higher, favoring the CC-486 or the oral azacitidine, compared to placebo. And relapse-free survival was about double that of placebo in the CC-486 arm. So, about a 10-month relapse-free survival median for patients receiving CC-486. And this was a fairly long follow-up period as well. So these are not data that I think are going to change because of very short follow up of the initial dataset. The 1-year relapse rate was 53% in the CC-486 arm and as high as 71% in the placebo arm. So, by all measures, the CC-486 outperformed placebo as maintenance therapy in this AML population.

And these are some of the adverse events depicted here that resulted in either dose modification or discontinuation in this trial. And really, not surprisingly, there was a little bit of a higher toxicity profile with the actual drug in the placebo arm, with 43% of patients experiencing dose interruptions in the CC-486 arm, and 16% requiring dose reduction.

Neutropenia was the most common reason for dose modification. The median treatment duration was about a year for the CC-486 drug, so it's encouraging to see that this drug could be given for a fairly long period of time with good overall tolerance. And the study also demonstrated that the overall quality of life metrics were preserved with CC-486 compared to placebo. In terms of other side effects, I think that we can, non-hematologic types of side effects, GI occurred as one of the more common side effects, but rarely resulted in discontinuation of the drug. So, overall, well tolerated as a maintenance therapy with clearly a disease modifying activity profile.

So, I think we’ve seen a couple of examples of new drugs that have hit the market over the past couple of years now, and I should point out the CC-486, oral azacitidine, was just recently FDA-approved as maintenance therapy in AML, post-induction. So it's an option to consider for your patients who may fit the bill for needing some type of post-remission therapy, but not maybe being eligible for a transplant.

And then we have the CPX-351 drug of course, that received approval a couple of years ago and is considered a standard of care frontline option for patients with secondary AML in previously untreated disease.

So, there’s certainly a lot more exciting, new drugs to talk about that we don’t have time to go through today, but, like I mentioned at the beginning, it's an exciting time to really be working in the field and to be able to learn more about these new compounds and where they fit into the overall armamentarium and how they will impact the natural history of the disease in ways that we have never seen before.

Case: A man in his late 60s with secondary AML, deletion 7q and mutations in RUNX1, TET2 and CUX1 receives induction therapy with CPX-351

DR LANCET: I think now what we can do is go to a couple of cases.

So, the first case that we’ll discuss is a 69-year-old gentleman who presented with a history of monoclonal gammopathy of unknown significance. And this patient also had a history of prostate cancer, treated with localized radiation therapy. Went to see his primary care doc with about a 1-month history of increasing fatigue, shortness of breath with exertion. Was an avid golfer, but lately not able to play much due to his symptoms.

His workup included a CBC that showed a pancytopenia, and most notably with a hemoglobin of 7.6. And he went to see a hematologist for further workup.

This led to a bone marrow biopsy, about one week later, that showed hypercellularity at 85%, with increased myeloblasts at 41% and trilineage dysplasia. Cytogenetics depicted a deletion 7q in the majority of the metaphases and next-generation sequencing revealed mutations of CUX1, TET2 and RUNX1.

So the patient went on to receive induction chemotherapy with CPX-351, as fit the clinical indication. He had a day 14 bone marrow biopsy performed that showed good response with hypocellularity in less than 5% blasts. And at about day 32, he recovered his neutrophils to greater than 1000 and platelets greater than 100, at which point he was discharged home.

His hospital course was relatively uncomplicated, save for febrile neutropenia, a rash and a nodular pneumonia, which was treated successfully with antibiotics.

So this patient is currently under consideration for an allogeneic transplant, given the high-risk features of his AML, namely the deletion 7q and the RUNX1 mutation, both of which are considered adverse in the NCCN and ELN risk stratification guidelines. And he’s currently awaiting final donor confirmation and dental clearance, which is always an important step in transplant.

So, I think this case illustrates some of the utility of a drug like CPX-351 in an older adult with previously untreated AML that is felt to be secondary in nature by virtue of the deletion 7q and the trilineage dysplasia. So, this is the type of patient that was actually deemed to benefit from CPX-351 in the Phase III trial that I showed you. And would seem to be a good candidate for a potentially curative transplant having received CPX-351 as initial therapy.

So I think that we want to think carefully about our patients in the older age range who may be quite fit and well enough to go through intensive therapy and certainly to strongly consider induction therapy in this patient population with an emphasis on CPX-351, given the survival advantages that have been shown to date.

Case: A man in his mid-70s with a history of myelodysplastic syndrome for 2 years develops secondary AML and receives CPX-351

DR LANCET: And then a second case is a little bit different, but in some ways some of the same decision points. So, this is a 74-year-old man with a history of myelodysplastic syndrome for 2 years. He was originally treated with erythropoietin stimulating agents, and then went on to lenalidomide with a response that lasted for a number of months. And then with again worsening of blood counts that necessitated discontinuation of the lenalidomide regimen. History of hypertension, gout, arthritis, pulmonary hypertension.

So at the time of the workup of his worsening anemia, after the lenalidomide therapy, we saw that his white count was 1.9, hemoglobin 8.4, platelets were 22. So clearly, a worsening of his hematologic parameters. Bone marrow biopsy was hypercellular with 21% blasts. Trilineage dysplasia. Normal cytogenetics. And next-generation sequencing revealing mutations of TET2 and ASXL1.

So, given his excellent performance status at the time of initial therapy, despite some coexisting medical illnesses, his treating physician’s chose to pursue therapy with CPX-351 in the setting of secondary AML. The day 14 bone marrow biopsy was looking pretty good with hypocellularity and less than 5% blasts. Count recovery was delayed but did occur. So, as I described previously, you often do see longer and more protracted count recovery of neutrophils and platelets after CPX.

His hospital course was not the greatest. He had problems with C difficile colitis, some significant fluid overload and pulmonary edema and a lot of deconditioning. So he was, in fact, discharged to a skilled nursing facility with a rehab unit for about 2 weeks prior to going home. So this treatment took a lot out of him. And the repeat bone marrow biopsy done following his return home showed that he was actually in a complete remission state with incomplete count recovery, with less than 5% blasts, both a normal white count, mild anemia and platelets in the 70s. So clearly, a CRi state but not without some price to pay in getting there.

So, after a lot of discussion, the patient was deemed not considered — or was deemed to
not be a suitable candidate for consolidation chemotherapy or bone marrow transplant, given the comorbidities and the deconditioning that he experienced. So, in light of the lack of suitable options, this patient opted for more conservative monitoring, which is where things stand today.

And I think this case illustrates a scenario that we often experience in treating older patients with AML in that they often do have some trouble getting through the intensive portion of the therapy with hospital-related complications, not necessarily fatal, but enough to affect performance status and quality of life and decision making down the road.

So, this is a patient who I think up until recently would have had very few options other than just waiting and watching and hoping for the best. But now with the advent of the oral azacitidine regimen, this may be the type of patient who would actually benefit from a lower intensity post-remission regimen that would allow for maintenance and quality of life, but at the same time a disease control that would occur at a higher degree than simple observation alone.

So, I think this is a case of a scenario where there was an unmet need for more effective post-remission therapy in patients who may not be eligible for other forms of post-remission therapy. And certainly, an avenue to explore in this type of patient population.

So I think these are the 2 cases that illustrate some of the scenarios that we do encounter in the initially more fit patient population and some of the decisions that go into the initial therapy and post-remission decisions as well, that are impacted by the things that occur along the way.

DR LOVE: In terms of this second case — so I’m kind of curious if you could access it, I don’t what the situation is in terms of accessing oral azacitidine, would you give it to him or would you wait a little bit until he — kind of better condition?

DR LANCET: Well, it's a good question. I think there is not a lot of clarity on that issue. I think the idea really would be get somebody started on treatment with this type of a regimen within a reasonable timeframe. And you could maybe argue a little bit about what that timeframe might be precisely. I think that in general, we would want to start a patient on the maintenance therapy regimen within approximately 3 to 4 months of achieving the initial response and this was actually the design of the Phase III trial as well, that I just described, was the patients were started on post-remission therapy within 4 months I believe, of achieving that remission.

When you get beyond that, you would have to potentially think that the benefit would be less clear. But at the same time, I think that given the very high likelihood of a relapse in this patient without question, that you would want to get treatment started at some point, but ideally within the first 4 months to really take advantage of what the data from the study showed. But I think that if you’ve demonstrated remission persistence in a patient with deconditioning, you’re okay to wait until you think that they're well enough to go through this treatment. Although, the actual impact, if you wait beyond 3 or 4 months, I think is less clear.

So, I think we just have to go into it with that little bit of a knowledge gap as to what the efficacy is when you started a treatment later than earlier.