Acute Myeloid Leukemia — Andrew H Wei, MBBS, PhD

PROF WEI: Hello. My name is Dr Andrew Wei from the Peter MacCallum Cancer Centre in Royal Melbourne Hospital in Melbourne, Australia. And I am going to give you an AML update as part of the Research To Practice Year in Review.

So let me just start by showing you an overview of the current landscape for AML which includes some drugs which are close to FDA approval such as quizartinib, but not quite there yet. And I will go through various aspects of this landscape and what are the new pieces of information which are relevant for our clinical practice. So let me start with the FLT3 arena where we currently have midostaurin as the main front-line therapy for patients with FLT3 mutant AML. We know that last year, Harry Erba presented the very important randomized study showing that quizartinib was superior to placebo as front-line therapy for patients with FLT3 mutant AML. So the question now becomes, which drug is now standard of care once both these drugs are available, midostaurin or quizartinib?

Well firstly, we can't compare these trials directly because they don't have completely overlapping populations. For patients with FLT3-TKD, the obvious choice is midostaurin because quizartinib is inactive against FLT3-TKD abnormalities. Second, midostaurin, the indication doesn't allow for maintenance therapy, particularly posttransplant, whereas quizartinib does offer 3 years of posttransplant maintenance. This becomes critical because we recently heard of the results of the MORPHO study which was gilteritinib versus placebo as maintenance therapy for patients after allograft for FLT3 mutant AML. And we all expected that to be positive. With that study not meeting its defined primary endpoint, we now would only be able to look to quizartinib as offering the potential for posttransplant maintenance therapy.

The other key question is whether all patients should get posttransplant maintenance therapy in view of the MORPHO study not meeting its primary endpoint. We do hope for more information about the MORPHO study in the coming months and I think the subgroup analyses will be very informative.

However, there have been a number of clinical studies now published which are really putting forth the notion that perhaps we should be using the FLT3-ITD MRD assay which is performed by highly sensitive NGS techniques as a way of trying to stratify risk in patients with FLT3 mutant AML. This is a test which is sensitive down to 10^-6, so a very sensitive PCR NGS based assay which can pick up very low levels of FLT3 mutant disease.

This is one of the papers published which was from our group which shows, on the left, about 100 patients with FLT3 mutant disease and MRD being done prior to transplant. On the top, you can see patients who are positive by the conventional method which is capillary electrophoresis which you can see only detects a small proportion of patients with positivity whereas in the blue in the middle, you can see this new PCR NGS based assay picking up a whole bunch of patients who are positive pretransplant compared to those who are negative or below the limit of detection at the bottom. On the right, you can see the relapse-free survival after transplant in these 100 patients according to where the FLT3 level was negative, which is the green curve at the top, or positive in various degrees shown by the orange, blue, and red lines. And so the main message here is if you're FLT3 positive pretransplant, the risk of relapse is high whereas if you're FLT3 negative pretransplant, the risk of relapse is low. And one hypothesis would be that if you're negative for FLT3 that perhaps posttransplant FLT3 maintenance is not required thus sparing the patient the risk of graft-versus-host disease.

The next area I'd like to talk about is the indication for patients with secondary and therapy-related AML. We know that CPX-351 has been around for a number of years now because of the pivotal studies showing benefit in this group compared to conventional 7+3. A new piece of information which came out at ASH last year was trying to address the question whether the new classification and designation of patients with myelodysplasia-related mutations, or MR-AML, might also be an indication for CPX-351.

This was based upon data presented by Jad Othman at ASH showing that this subanalysis of a UK trial comparing FLAG-idarubicin versus CPX-351 in patients who had adverse cytogenic risk. And what they did was they looked at this group of patients with MDS-related mutations, which you can see in the third box there, which is how patients with these mutations led by ASXL1 are classified by the new ICC classification which was published in 2022.

So in patients that had one of these mutations, ASXL1, BCOR, et cetera, Jad showed that there was a benefit for CPX-351 shown in the red curve compared to FLAG-idarubicin amongst patients that had adverse cytogenic risk. You can see still a small study, needs to be confirmed further, but certainly an intriguing possibility for the further use of CPX-351 defined by mutations rather than by clinical features.

The next area is a discussion about what to do with patients that are fit for intensive chemotherapy, but don't have a FLT3 mutation and don't have secondary AML and don't have nonadverse cytogenic risk.

DR LOVE: Before you go on, I just want to clarify. That last paper that you talked about, so these were patients who did not have secondary AML, they just had high-risk AML?

PROF WEI: Some of these patients would have had secondary AML, but not all of them. And many of them were defined purely by the presence of mutations rather than a prior history of MDS or a prior history of therapy-related AML.

DR LOVE: So were these mutations the kind that are normally seen with MDS or were those just high-risk in general?

PROF WEI: No, these now are a group of mutations which are highly related to patients with a history of MDS. And so the thought is that using mutations to define MDS-related AML is perhaps more accurate than just a clinical history.

DR LOVE: Okay, that makes sense. Thank you.

PROF WEI: So this study which was performed as a subanalysis of the pivotal trial by Jorge Cortes and published last year looked at patients with adverse-risk AML and asked the question whether CPX-351 displayed any potential benefit over 7+3. And as you can see here, there was a small benefit with respect to CPX-351. And this is important because the key question for clinicians at the moment is in patients with adverse-risk AML, age over 60, should you use CPX-351 or is ven/aza a possibility? Looking at this curve on the right, you can see that perhaps the benefit is not large. However, when you look at the posttransplant outcomes, you can see that CPX-351 does produce a 3-year survival which looks quite substantially improved over 7+3 along with the slight increase in response rate on the left, 41% versus 29%. And so it really does ask the question whether we should be using CPX-351 followed by allograft or whether we should go with ven/aza followed by allograft. And really, future randomized studies will be needed to fully define the answer to this question.

The next area I'd like to address is the area of unfit AML. And currently, there are 2 indications with overlap between patients who have an IDH1 mutation with respect to ivosidenib + aza or whether ven/aza should be the standard of care. Obviously, for ivosidenib, one needs to do rapid testing to identify the IDH1 mutation whereas for ven/aza, this doesn't require pretreatment risk stratification with genetic testing.

And so really, we don't have a randomized study between these 2 treatment options, but 2 separate studies which show, on the left, the benefit of ivosidenib over placebo with respect to its combination with aza. And on the right, the results with ven/aza both in the IDH1 mutant subgroup. So in the table, you can see the number of patients treated. The median age was identical between the 2 studies. The response rates were slightly in favor of ven/aza, but not a randomized comparison. The true CR rate is slightly higher with ivosidenib. And the median survival down at the bottom, fairly similar between the 2 treatment options. And so really, it comes down to, one, can you do rapid IDH1 testing to allow ivosidenib to be selected? Two, I think most clinicians feel that perhaps the myelosuppression is less with ivosidenib. And three, it still opens the possibility of using ven/aza after ivo/aza if patients fail. So I think, in general, most clinicians are in favor of using the targeted approach.

The next area to address is to update you on some of the long-term outcome data which was recently presented with respect to ven/aza which is now currently used for many patients who are considered unfit for intensive chemotherapy.

So on the left, there is outcome data now with over 40 months of median follow-up presented by Keith Pratz at ASH last year. And on the right is a publication which came out in Blood also updating the long-term outcomes of low dose Ara-C + ven in the elderly, unfit population. And I think the main message from both of these studies is that although there has been a substantial improvement with the addition of venetoclax in older people, you can still see that the survival curve does decline and long-term survival is only going to be apparent in about 15 to 20% of patients long term. So cure is still a very difficult concept to achieve in older people. Yes, there has been substantial improvement but clearly, understanding the mechanisms of resistance and relapse and further improving this backbone with additional drugs or additional strategies will be required.

Another message which has come out from the long-term follow-up studies is the question as to which molecular subgroups benefit most from these 2 approaches. And on the left, you can see that the group which benefits the most from ven/aza are patients with IDH mutation. You can see a very big survival benefit compared to no addition of venetoclax. And on the right, with low dose Ara-C + venetoclax, it has now been shown in multiple studies that NPM1 mutation, the green survival curve on the top, appears to be the subgroup that benefits the most from this particular regimen.

CPX-351 has also been combined with venetoclax for patients who are considered unfit for intensive chemotherapy. Early results, but very promising response data shown on the right. And more information clearly needs to be achieved in the future in terms of comparison with other conventional agents.

The other area I'd like to address is, are there any attempts to further improve the role of 7+3 in patients fit for intensive chemotherapy? And one strategy which is being looked at is whether the addition of venetoclax could provide a therapeutic opportunity for the future.

And so listed here is a table of the various trials which have been done showing the feasibility and the efficacy of adding venetoclax to an intensive chemotherapy backbone. At the top is the only study showing intensive chemotherapy in combination with venetoclax, the CAVEAT trial. And in the red squared box, you can see the various trials combining intensive therapy are the 7+3, CLIA, or FLAG-idarubicin in combination with venetoclax. You can see very high response rates with these various options. And I'll talk about 2 of the key treatment options on the next 2 slides.

So the first is the older population. This was a trial that we published first in JCO in 2020 and updated at ASH last year. You can see the CAVEAT trial median age 72 years and median follow-up over 3 years now. This is a very well tolerated regimen in older people. And there were 2 particular subgroups, NPM1 mutation and IDH2 mutant patients where you can see the response rates are almost 80% in this subpopulation of patients and the 3-year survival over 50%. And so I think this becomes a very important and useful option to consider because patients only require 1 cycle of therapy and after that, treatment is minimal and producing very good outcomes.

In the younger population, FLAG-idarubicin and venetoclax is a regimen which many clinicians are considering, especially in patients that have adverse-risk. And you can see the 2 major studies published, one by DiNardo and the other by Shahswar, showing slightly different approaches. And on the right, under modified, I've listed our approach which is sort of a combination of the 2 which I think is most clinically relevant. So with the fludarabine, we do use the same dose as DiNardo, but we do incorporate a dose reduction for people over the age of 60 as recommended by Shahswar. With respect to Ara-C, we also do the same thing by recommending a dose reduction for people over the age of 60. With respect to G-CSF, we use pegylated G-CSF at the completion of therapy which we think makes most sense. With respect to the idarubicin, we've elected to use the lower doses of idarubicin as illustrated by DiNardo rather than by Shahswar because we feel there is a level of toxicity of combining idarubicin with venetoclax that would favor a lower dose. And lastly, the venetoclax. We've used the regimen by Shahswar which allows incorporation of posaconazole because most of these people will develop severe neutropenia and we've elected for a day 1 run in with venetoclax without combination with chemo just to mitigate the risks of TLS. We hope that this modified regimen will guide clinicians out there as to how best to use FLAG-idarubicin with venetoclax.

The next area to discuss is a new indication or a new drug for patients with relapsed and refractory AML which is targeting a new subgroup not currently targeted by existing drugs.

And this is in relation to the exciting development of menin inhibitors of which there are a number in clinical trials of which the AUGMENT-101 using SNDX-5613 is one of them. You can see here, the 2 populations targeted by this menin inhibitor are patients with KMT2A rearrangements and also patients with an NPM1 abnormality. The main potential adverse risk is differentiation syndrome which is manageable if appropriate measures are instituted. And you can see here very, very promising response rates in relapsed and refractory patients with multiple prior therapy failures, 60% almost with KMT2A rearrangements and almost 40% with mutant NPM1. Both of these studies now have been published recently in Nature. And on the bottom was some very interesting and intriguing data presented from Scott Armstrong's lab where you can see exposure of patients to menin inhibitors has now resulted in the emergence of some menin mutations, on-target mutations which block the activity of menin which interestingly seemed to occur quite quickly. And so this will be an incredibly important area for the field to try and understand further.

I mentioned SNDX-5613. However, there are other menin inhibitors coming through with ziftomenib, another major option which you can see on the top left there. And interestingly, this agent appears to have slightly more activity in patients with NPM1 mutation, but small numbers and I think we need more follow-up data to really distinguish between the 2 menin agents in front-line clinical development.

And lastly, this is a slide on the development of SYK inhibitors, entospletinib. And this was an agent which was being examined in a pivotal front-line trial in combination with intensive chemotherapy with NPM1 MRD clearance as the primary endpoint. And unfortunately, it was announced that in November last year, this study was terminated by the company because of difficulties of enrollment. And the company is now focusing on the development of a next generation SYK inhibitor called lanraplenib in combination with FLT3 inhibitors.

And so it remains to be seen whether SYK inhibitors will eventually have a role in AML.

The next area is patients with a FLT3 mutation who have relapsed after prior therapy. Currently, we have gilteritinib available for this setting. And there was long-term data presented by Sasha Perl in this publication showing that really, there has actually been a long-term separation of the 2 curves between gilteritinib and conventional care in red at the bottom. And so the main message from this is that gilteritinib even in the salvage setting can actually produce long-term remissions in some patients. You can see all the censored events there on the blue curve stretching out beyond 3 to 4 years. And really, this drug has had a very important impact in this patient subgroup.

So the next question is, how can we improve upon gilteritinib? And there have been many attempts to try and combine gilteritinib with conventional agents such as venetoclax and also venetoclax + azacitidine. I'm going to discuss with you the results presented by Nick Short from MD Anderson at ASH last year showing the triplet combination of aza/ven/gilteritinib in the front-line setting with the survival curves on the bottom left, and in the relapsed refractory setting on the right. The key message is that this is a myelosuppressive regimen because of the very long half-life of gilteritinib. And this results in the need to reduce the doses post-induction which you can see on the right with azacitidine truncated to 5 days compared to 7 days, venetoclax halved to 7 days instead of 14 or more days, and gilteritinib the dose reduced from 120 to 80 mg. And it's important to consider these dose adjustments. Otherwise, severe and prolonged myelosuppression can result. However, with these dose adjustments, you can see in the green curve on the left the survival in front-line use for patients with FLT3-ITD. You can see the neutrophil recovery is 37 days, platelet recovery 25 days. As I mentioned before, myelosuppression being the major issue. And so it's hope that this combination will be taken forth into validation studies in the future.

The next study is in relation to IDH1 mutant AML. And there's a new drug which has come out called olutasidenib which has been approved and now provides another option in conjunction with ivosidenib for patients with IDH1 mutant AML.

So olutasidenib, or FT-2102, is a new option for IDH1 mutant AML with relapsed refractory disease. You can see response rate almost 50%, and the blue curve showing the survival of patients with a response which looks quite promising. A well tolerated drug as you would expect from a targeted agent and now definitely a competitive alternative to ivosidenib in this setting.

DR LOVE: I'm just kind of curious. I'm kind of — I don't if there's a nickname for the new IDH inhibitor. But any way to compare indirectly either efficacy or tolerability to ivosidenib?

PROF WEI: I think because the population was different, it's going to be difficult to say that one is superior to the other. Both agents are well tolerated. Both have potential risks of differentiation syndrome. Both can cause some reversible liver complications. Both have shown good response rates and molecular clearance. So I think it's going to be a difficult choice at this stage until we have more information on the tox profile from the 2 agents in terms of full publication and also whether these agents will be able to be combined effectively with other agents with respect to drug-drug interactions and PK issues. So I think at this stage, they're both available but too early for us to pick one over the other.

DR LOVE: All right. Please continue.

PROF WEI: The next area I wanted to focus on was the situation of IDH2 mutant AML in the relapsed and refractory setting. Enasidenib is available in the United States, but unfortunately not available outside of the US.

And there have been some further studies. This was a study presented by Stephane DeBotton presented at ASCO last year. And this was an interesting study because he looked at a post hoc subgroup analysis of the pivotal trial of enasidenib versus conventional care in patients with relapsed and refractory AML and looked at whether there were any differences between patients that had the R140 isoform of IDH2 versus the R170. On the right, you can see that overall, enasidenib compared to conventional care had a better response rate, but there was no improvement in survival. However, if you look at the 2 subgroups, R140 which was present in almost 3/4 of patients, you can see that there was a slightly different co-mutation profile with more FLT3 and NRAS mutations whereas the R172 patients had more p53 abnormalities as co-variants. And you can see that the response rate interestingly was higher for enasidenib in the R172 subpopulation as well as the median overall survival suggesting that perhaps these isoforms don't behave the same and that IDH2 inhibition may have a more selective and more beneficial role for patients with the R172 isoform.

And there have been attempts to combine enasidenib with venetoclax and also the triple combination of enasidenib, venetoclax, and azacitidine in IDH2 mutant AML. This is the doublet first, enasidenib + venetoclax, data presented previously by Steve Chan showing that in 11 patients, 10 of whom had relapsed refractory disease, that there were 5 responders. So early days on the right showing that molecular reductions can be achieved with this doublet.

And on the next slide, you can see here this is new data presented with respect to the triplet, azacitidine, enasidenib + venetoclax. On the left, you can see the results amongst patients who received this triple combination in first salvage or second or subsequent salvage. And on the right, the results of the triplet versus the doublet in first-line therapy. So really, you can see the triplet combination does appear to have its best efficacy in first salvage or front-line therapy. Whether this is going to be better than ven/aza or other possibilities remains to be seen, but definitely an interesting alternative for using venetoclax in combination with other standard agents.

Lastly, I just wanted to finish up by highlighting the use of decitabine in combination with cedazuridine, an oral combination in one tablet versus IV decitabine. This was the results of the ASCERTAIN trial which was a pharmacokinetic study which started off with oral decitabine versus IV decitabine, and then there was a crossover in the second cycle, and then all patients continued on with oral decitabine thereafter. And really, the main result and message from this trial was that these 2 agents are bioequivalent, and so people have resorted to using oral decitabine where IV decitabine was currently used. Now although they are pharmacokinetically equivalent, many clinicians have found that the 5-day schedule of oral decitabine/cedazuridine is quite myelosuppressive resulting in some clinicians reducing, particularly in combination with venetoclax, the dose of this oral agent to 3 times a week, Monday, Wednesday, Friday compared to 5 days a week. Now, this is a clinical practice option not shown in clinical trials yet whether this is sufficient and beneficial but certainly, there's the observation that perhaps this agent is a bit more myelosuppressive even though pharmacokinetically similar concentrations were achieved.

So that was the Year In Review in AML. I hope some of this information was of use to you and I look forward to speaking to you again in the future. Thank you.

DR LOVE: So that was great. Just a couple of follow-up questions. When you look back over this past year and think about some of the newer developments in AML, from your point of view, what are some of the ones, maybe 2 or 3 developments, that you think are most relevant to the general medical oncologists in community-based practice as opposed to AML investigators?

PROF WEI: I think the use of MRD is definitely becoming a more important aspect, so measuring small amounts of disease which allows us to make more rational decisions before patients become clinically unwell. I think the notion of targeted therapy with menin inhibitors in conjunction with previously approved drugs targeting FLT3 and IDH are important. But also, recognition that drug resistance and mutations resulting in clonal evolution will almost be inevitable, particularly for a rapidly evolving disease like AML and that utilizing MRD strategies to identify these clonal changes and perhaps in the future adapting our therapy rather than waiting for clinical relapse I think for me has emerged as a future theme.

DR LOVE: In terms of clinical trials, clinical research, what are some of the trials that are sort of ongoing that perhaps we're going to see some data on in the next 6 months or a year that you think might have an important impact both on practice and future research?

PROF WEI: Yes, the MORPHO results we hope will be presented at a congress in the near future is one which is critically important and one which was a bit baffling to many clinicians where we really thought this was going to be a homerun. The results of 2 major front-line venetoclax — front-line trials in high-risk MDS I think are incredibly important, the VERONA study which is ven/aza in the front-line MDS setting, and also the sabatolimab + aza trials. I think the results of these are really important because we don't have any new drugs in MDS. And furthermore, the ven/aza option is important because with the new revisions of the AML classification, there are more people using ven/aza for people with blast counts below 20%. So I think having data to inform this practice is going to be really important. There's also a trial called AMADEUS which is oral azacitidine randomized to placebo after allograft as a maintenance option. And I think the results of this will also be important because we already have oral azacitidine as a maintenance option for patients with AML in remission. So whether we can use it posttransplant also I think will be critically important. And also, we hope that the results of the QuANTUM-First study with quizartinib, we're really waiting for the FDA approval of this which is actually expected any time now. And so once this is available, this will become a new FLT3 inhibitor that will ultimately change practice. So those are the key studies which I think are on the near horizon.

DR LOVE: Anything new in terms of magrolimab?

PROF WEI: Yes. I forgot to mention that one, but magrolimab is being examined in a number of pivotal front-line studies. So the major one is in MDS, high-risk MDS, which has been completed which is aza/magro versus aza/placebo, and we hope that will be positive. There's also trials looking at its role in FLT — in sort of p53 mutant AML and also in triple combination with venetoclax/aza/magrolimab in the front-line setting, and that's called ENHANCE-3. That study is still early in development. With the magro, magro/aza option, 2 papers have been published recently both in JCO. First is already out, the one is about to come out in MDS and AML respectively. And this agent does appear to substantially improve the response rates. And the key question is whether survival will be improved with this option. And we'll have to wait for the result of the ENHANCE-1 trial to find that out for MDS.

Myelodysplastic Syndromes — Uma Borate, MD, MS

DR BORATE: Hello everyone. My name is Dr Uma Borate. I am an Associate Clinical Professor at the Ohio State University James Cancer Hospital. And today, I’ll be talking to you about the year in review for myelodysplastic syndromes.

So let’s start off with a very novel and new classification system that was presented at ASH in the end of 2022. And I think we all are very familiar with using both the IPSS and the IPSS-R classification systems when we see an MDS patient in our clinics to let the patient know what category they fall in and what their prognosis is.

So this new system is called the IPSS-M. And M stands for molecular. What the group that presented the study did is they looked at 2,500 MDS patients and they incorporated not just the cytogenetic abnormalities, but also the molecular mutations that we know are commonly seen in our myelodysplastic patients and then generated different scores to classify these patients into much more, I think, neater categories that then correlated better with prognosis than we’re able to do with the IPSS-R which as you all know does not incorporate molecular mutations. So if you look at the slide in front of you, you can see that there’s a bell curve. There’s your average patient in the middle and then you have patient scores that go from very low all the way to very high. And you can see that if there’s a change in the score by 1, the patient has double the risk of progression of, you know, poor survival as opposed to if the score is lower by -1 then it’s half the risk of that average patient that you see right in the middle. And again, you can see on the slide, these categories broke out really well in terms of leukemia, transformation to leukemia or leukemia free survival as well as overall survival which was much more discrete than the previous IPSS-R because of the incorporation of these molecular mutations.

So to summarize this new classification system, it had a large cohort of patients that profiled very closely. They had a validation cohort. And they saw that the TP53 multi-hit, FLT3 mutations and the MLL PTD mutations were really identified as the top genetic predictors. Clinically, for us that see these patients, this prognostic discrimination was improved across all clinical endpoints compared to previous versions.

So next, I want to talk about lower-risk MDS. And I want to discuss specifically some trials that maybe address the point of, do our interventions in this population actually impact patient outcomes? Traditionally, the mantra has been these are patients that you support, you provide supportive care, these patients do well for a long time and you’re not really impacting survival necessarily with some of your outcome — with some of your supportive care measures in these patients.

So let’s talk about this study presented at EHA 2022 which addressed a really important question. When is the optimal time to start erythropoietin-stimulating agents or ESAs? Is it based on the hemoglobin threshold and need for transfusions? Or do we actually intervene before a patient is transfusion dependent? Again, think of the patient in your clinic. You know they’re anemic, you know they have MDS, maybe their hemoglobin is 10. Maybe it’s 9. You wait for the hemoglobin to drop to 8 and below and for them to start needing transfusions before you intervene. So in this study, the patients that were enrolled had lower-risk MDS, had a hemoglobin less than 10 and they had comparable propensity scores for analysis to estimate the effects of these ESAs.

And what they really looked at was what is beneficial to the patient before — to see if ESA treatment is beneficial to the patient before transfusion dependence in terms of quality of life, but also in terms of outcomes. So as you can see on this slide, there’s a clear survival advantage to the ESA treated group when the patients were initiated on treatment before they became transfusion dependent compared to those patients that were transfusion dependent. They also, and not shown on the slide, but those patient also endorsed a much better quality of life when they were initiated with ESAs before than later when they were already needing transfusions.

In follow-up to this, I also wanted to discuss the long-term follow-up data from the MEDALIST trial. Again, this might be a trial that you are familiar with. We’re now using luspatercept in the clinic. And the MEDALIST trial was really evaluating the efficacy of luspatercept as opposed to placebo. And they focused on low-risk MDS patients, patients who had become intolerant or unresponsive to ESAs. And then obviously, they also focused on patients with ring sideroblasts and patients with SF3B1 mutations. And as you — just a quick reminder, the trial randomized patients 2:1 to luspatercept versus placebo. Luspatercept was started at the dose as you can see on the slide and then escalated as is now noted in the package insert. And then the patients entered a long-term follow-up study where they continued to get the same treatment as they got on the study including placebo, so they weren’t really allowed to cross over unless they had a clinical benefit. So these patients continued to be followed for almost up to 3 years or some patients even up to 5 years depending on the response.

And I think the findings that they saw were quite interesting in that if your patient received luspatercept and received a response, and by response I mean they became transfusion independent where they did not need transfusions for over 8 weeks, these patients tended to do really, really well and they tended to maintain the response with a median time of response of over 80 weeks. And this is compared to patients that obviously, they got placebo or did not have a response. And I think the message here potentially could be the same message you heard from the prior study that maybe intervening with ESAs/luspatercept earlier affects the biology of low-risk MDS patients to where maybe we’re altering something in the bone marrow where they are able to continue with normal hematopoiesis and not be as transfusion dependent and also have a better quality of life. There’s something that changes in our patients when they start needing transfusions. And even though you are able to maintain their hemoglobin clinically with ESAs, with transfusions, they just don’t feel as well. So, to me, that’s very intriguing and definitely something we need to think about as clinicians.

And then to summarize again, and this is not data that is actually already published, it’s a topline release about the COMMANDS study. As you can see, the COMMANDS study here now went one step further in where they were actually randomizing patients to get either erythropoietin-stimulating agents or luspatercept in your low-risk MDS patients and then patients were continued to be followed for, you know, for 24 weeks or until progression. And the press release that we have is listed on this slide. And basically, what they released was that the study met its primary endpoint which was a highly statistically significant and clinically meaningful, we don’t know what that is yet, improvement in transfusion independence or response to this treatment with luspatercept as opposed to the control arm. So I think, again, this is something that we’re going to be watching eagerly for the publication. And I think, to me, it brings the question of, what is the role of these agents and when do we initiate them clinically to help our patients live the best lives that they can?

So moving on to another sort of workhorse of low-risk MDS patients, and that’s lenalidomide. We’ve been using this for a long time for our low-risk deletion 5q MDS patients. And this trial, I thought, was very interesting in which they looked at lenalidomide, but they looked at a specific dose of lenalidomide which is 5 mg per day on days 1 to 28 versus placebo for this study. And I think the same question is being asked. Can you initiate lenalidomide at a lower dose earlier to improve outcomes in these patients? We know they do well. We know lenalidomide helps in transfusion independence. The response rates are over 60%. The cytogenetic response rates are up to 70%. But can we intervene earlier and potentially give patients more benefit?

And this study seems to suggest that that is possible, that using lenalidomide at a lower dose, but intervening early can decrease a risk of transfusion dependence. That happened in about 70% of patients in this study compared to placebo. They did look like they were achieving some erythroid and cytogenic response. There was no sign of clonal evolution. This was a big concern. We know lenalidomide maintenance in multiple myeloma and other diseases can lead to secondary AML, so people are obviously waiting eagerly for long-term follow-up to make sure that doesn't happen. And then the side effects of initiating lenalidomide early, as we know, is neutropenia. And this study suggests that that neutropenia was not clinically significant.

So I think this is something that we really want to look forward to the analyses.

So in summary, for low-risk MDS, it appears that early initiation of ESAs is more beneficial than waiting until the emergence of transfusion dependence for outcomes and for quality of life for our low-risk MDS patients. If our low-risk MDS patients that have initiated luspatercept respond, they seem to really maintain that response. Early initiation of low dose lenalidomide in these deletion 5q patients appears to be beneficial in this study with no long-term clonal evolution, but we still need more time to look at the data. And so I guess I'll leave you with a provocative thought. Can these be considered disease modifying therapies versus just supportive care only?

DR LOVE: I'm curious if any new hypotheses about the mechanism of how lenalidomide ameliorates anemia. If it's not like a direct effect on the cancer, like what would it be?

DR BORATE: That is a great question. I'm not aware if there's new information on why it's affecting patients or why it's benefiting patients. I know there's a lot of interest, for example, with other sort of — I guess lenalidomide would be a disease modifying agent in this case whether it affects overall inflammation, the inflammasome. I think there's a lot of interest in looking at that and I know there's potential data on whether agents like luspatercept do the same thing with their mechanism of action. But I'm not sure if we really still understand how lenalidomide works.

DR LOVE: Well actually, I guess if you think about it, because I don't think you see amelioration of anemia in myeloma when you give lenalidomide, just the opposite. But maybe the microenvironment is different, you know, and that it actually isn’t affecting the microenvironment except the microenvironment is different than in myeloma or whatever, wherever else they're using it, lymphoma.

DR BORATE: Agreed. Agreed. I think the microenvironment is a hugely underexplored area of research in MDS. I know there's a lot of interest in exploring that and potentially targeting that with therapeutic agents.

DR LOVE: Okay, please continue.

DR BORATE: Next, I wanted to address the use of hypomethylating agents in low-risk MDS. I know clinically when I see a low risk MDS patient, they're not responding to ESAs or luspatercept. I always have to be very thoughtful on when do I actually initiate a hypomethylating agent because we know that these are agents that the patient might be on potentially for the rest of their MDS life, so as to speak. And so I think understanding the data on what these agents do and how helpful they may be in the low-risk MDS population is very important.

So I wanted to talk a little bit about the data from the oral decitabine/cedazuridine study looking at low-risk MDS patients. And this study specifically, again, was looking at, what is the optimal dose of this oral dec/c? Which as you know, is decitabine/cedazuridine. And it is equivalent to the IV formulation as shown in the Phase III study. And they wanted to look at different dosages. You can see here on the slide, the decitabine can be from 5 mg, 10 mg, to 15 mg. But they also wanted to look at different schedules to see what dose and schedule would actually be the most beneficial in the low-risk MDS patients. This is a Phase I study, so this was not designed to look at efficacy, but we always get a few clues when we look at these type of studies.

So what they did see was even looking at all the different doses, they appeared to be fairly safe. But clearly, there were dose limiting toxicities and most of them were related to Grade 4 neutropenia which was, again, very dose dependent. So the higher the dose, the more the number of days, the more patients were neutropenic for. They did see responses. There were hematological improvements and there were signs of transfusion independence. And based on this Phase I study, they determined that the 5-day schedule with the dose of 10 mg of dec/c with 100 mg of cedazuridine was actually the best to move forward. And it is now moving forward in a randomized Phase II study comparing it to the 3-day regimen of 35 mg of dec/c. So is it better to have a lower dose for 5 days or do we go higher up on the dose of decitabine for 3 days? And this is, I think, a study that will be helpful because our patients don't like to take too much treatment. So if they have to be on this for maybe several months or years, maybe the 3-day schedule would actually be better, but we don't know until we get the results of that study.

So moving forward with high-risk MDS and the therapeutic advances, I think all of us are excited because we all have had 1 or maybe 2 therapies for these patients in the last decade or so and that's either azacitidine or decitabine.

So let's talk about what's going on in this field. And I think the first thing that comes to everyone’s mind is, well can we use venetoclax in combination with HMAs? We did it in AML. It's a natural sort of transition to do it in MDS and see how that works. So let's talk about what's going on in MDS and CMML with the combination of HMAs and venetoclax.

So this study that was presented at ASH really looked at the oral dec/c drug that I just mentioned right now which is oral dec/c, cedazuridine and venetoclax, so it's an all-oral combination. And this was looked at with various doses of venetoclax to see, again, Phase I study. What are the dose limiting toxicities? Can patients tolerate it? And if they cannot, then what are we seeing? And are we seeing any early responses? So all the data we have so far is there's definitely some signal that this appears to be tolerable. There wasn't a lot of GI toxicity which comes to mind when you use 2 oral regimens. And that they are moving forward in different doses and cohorts and studying the efficacy to see which is the best dose to move forward with in this patient population. So I think we need to watch this space and look at how we can potentially give our patients an all-oral combination.

Another study, again, looked at the same thing, but this time looked at azacitidine, again, in combination with venetoclax. And these were patients that were newly diagnosed as well as relapsed refractory. And in this study, there were about 23 patients enrolled. This study is actually completed and published. They had a very standard Phase III design. And they actually did see myelosuppression in these patients as can be expected. Count recoveries were delayed, there were high rates of myelosuppression, and most of the responses they saw were marrow CRs, meaning the marrow is empty, excess blasts were gone, but there's not a lot of healthy hematopoietic cell recovery. So I think this gives us a clue that this is a very potent combination and something that we really need to tease out the dose and identify the right dose, the right duration to use in our MDS and CMML patients.

So, again, as I just mentioned, we don't have an MTD in this population. But looking at the cytopenias, they decided that the best dose to move forward with would be 75 mg/m2 for 5 days for azacitidine, and 400 for venetoclax for 14 days. Very similar side — I guess very similar side effects to what you would see with AML with prolonged cytopenias, you get infections. And they did have 3 deaths due to sepsis when patients were on treatment. So in this study, they deemed overall response rate about 87%. But as I mentioned, most of these were marrow CRs.

So then looking at what happens when a patient is failing hypomethylating agents. So this was a study that explored using the combination of azacitidine, maybe the patient has gotten azacitidine for a various number of cycles as was outlined in the study, and the patient is losing a response. This is a scenario unfortunately we see clinically all the time. Can we rescue the patient by combining azacitidine with venetoclax in this patient population? This was also looked at with various doses of venetoclax to determine what dose is safe. And as you can see here on the slide, there does appear to be clinical efficacy of this combination across the different mutations. It does appear that the more the blasts are BCL2 dependent, the better response they seem to get.

And in this study, we know that patients actually had marrow responses, there were a few CRs, and the value of the marrow CR is actually kind of debated in MDS. However, if you have a patient that has a marrow CR, all the excess blasts are gone, but they also have a hematological improvement, then that is clinically meaningful for the patient. And in this study, they actually salvaged patients where about 9 patients had less than 5% blasts and used it as a bridge to transplant. So I think this is intriguing. We don't have a large number of patients treated with this combination, but something to consider clinically as we are figuring out what to do with our MDS patients once they failed a hypomethylating agent.

So in summary, azacitidine, oral decitabine, various HMAs and venetoclax are being studied in high risk MDS. However, we don't have clear data as yet on what is the safest combination dose of the HMA and the venetoclax. We're all familiar with this combination in AML. We all know this can cause prolonged cytopenias. And I think the key is identifying the right dose and the right intensity of this combination to give the best efficacy for our MDS patients. And there's a large Phase III study called the VERONA study which is exploring just that. I think we'll be anxiously waiting for results in the next year or so.

DR LOVE: Any hypotheses about why you see more cytopenias with MDS and AML?

DR BORATE: You mean with venetoclax? With the combination?

DR LOVE: Yeah, with venetoclax. Right.

DR BORATE: I don’t other than we know that both MDS and AML are myeloid malignancies that are heavily BCL2 dependent. And somehow, the combination of venetoclax with azacitidine or HMA is, the combination just is hard on the marrow. It kills off the blasts very effectively, but it also kills off normal hematopoietic stem cells. And for that recovery to take place, I think the more complex the cytogenetic abnormalities are, the more the mutations, the more damaged the stem cells are, it just takes them longer. And I think that’s where we really need to finesse what agents do we use in combination with HMAs to move forward depending on what the abnormalities in MDS are, whether they’re p53 or complex karyotype or so on.

DR LOVE: Do you have the feeling or have you seen patients that are getting — I hear docs talking about using HMA/venetoclax in MDS. Are you hearing about cases where people are having problems with cytopenias because they’re not aware of the difference? Like in second opinion, do you see cases like that?

DR BORATE: I do see cases like that. I’ve seen several where people have started at — so they’ve started with an HMA, patient hasn’t had an adequate response, usually azacitidine. They’ve added the venetoclax. Then, they’ve been really alarmed because the cytopenias are so prolonged and profound that they’ve just backed off completely and then the patient comes to us to say, what’s going on an what do we do? So I think that’s definitely a concern. And I think that’s why I think these trials are so important to clarify the dosing of venetoclax for MDS and AML provided the MDS trial is positive, of course.

DR LOVE: Yeah. The message I’m getting is there’s a CME need there.

DR BORATE: I completely agree.

DR LOVE: What you’re saying is they start out with the same — they start out using the same dose, the same approach as AML. They see bad cytopenias, they send it to you. And really, they need to know they don’t start out that way.

DR BORATE: The other thing that I think there’s a huge CME need for in addition to what you just said is while they’re doing all this, I would say most docs in practice are not using the adequate prophylaxis. So these patients are being cytopenic and neutropenic for a long time, but what we do is we are starting off using broad spectrum antifungal azoles like posaconazole or voriconazole. We are just seeing the roles of venetoclax because we know there’s an interaction. And we’re being very careful to make sure that patients continue on them when they’re cytopenic. I don’t see that being done regularly in the community setting. And I think then patients end up not only with cytopenias, but an additional fungal infection. So now, you’re stuck because there’s nothing you can do.

DR LOVE: That’s really interesting. I wonder whether or not they’re sort of getting scared about using antifungals because they know they’re going to have to change the dose of venetoclax and maybe they get — they’re not secure about what to do.

DR BORATE: You’re right. You’re absolutely right. It’s hard to get the azole approved, so there’s an insurance issue there. And then if you look at the package insert of venetoclax, you’re right, for certain azoles, it’s 70 mg, for others, it’s 100. And it is confusing. And if I’m a busy doc in practice, I’m not sure I want to deal with all of these nuances.

So what about novel combinations? We already talked about HMA with venetoclax. Maybe that's not considered that novel anymore since we use it in AML, but there's several other combinations that are moving forward in high-risk MDS.

So let's talk about another unique mechanism of action called an anti-TIM-3 antibody. So what is TIM-3? TIM-3 is actually a molecule, an immunomodulatory molecule if you want to call it that, which is expressed both on leukemic stem cells, but not on normal hematopoietic stem cells. So it's expressed on leukemic stem cells and it's also expressed on a number of immune cells like your T cells, your dendritic cells, and so on. And so the drug that we're about to discuss is called sabatolimab. It's an anti-TIM-3 antibody. And the hypothesis is now, you have an antibody that maybe kills two birds with one stone. Not only does it attack the leukemic stem cells, but it also prevents this suppression of the immune system and the microenvironment as we were talking about earlier, and helps to activate a tumor immune response at the same time. So it potentially has 2 mechanisms of action.

So we do have data that was presented at ASH for a randomized Phase III — Phase II double-blind study looking at sabatolimab, and it was dosed every 2 weeks at the dose as you can see on the slide, so on day 8 and day 22, in combination with decitabine or azacitidine compared to placebo. And the primary endpoints were complete remission and progression free survival.

So what we saw in the presentation by Dr Zeidan and colleagues at ASH is this study actually did not meet its primary endpoint for CR or for progression free survival, but there seems to be a signal that there might be some benefit to this combination compared to azacitidine or decitabine plus placebo. And you can see the hazard ratios here. You can see the sort of separation of curves. And the hypothesis is we know this drug is well tolerated from the Phase I study, maybe patients have to be on it for a certain period of time to derive the benefit and maybe that is the key to treating MDS patients in general. We need to keep them on active treatments for as long as possible for them to get a clinical benefit. So the randomized Phase III study is actually ongoing. And, again, as mentioned before with the VERONA study, these are results that should be available in the next couple of years and we're very eager to see if this is a potential option for our patients.

So moving on to another antibody but a different target, and the target this time is CD47. This is expressed on a wide variety of cancer cells. And this is really, as very eloquently put, a don't eat me signal. The tumor cell wears a cloak with CD47 and sort of evades the immune system and tells it, don't eat me, I'm one of you. But what we have shown with antibodies like magrolimab is we can really effectively uncloak the tumor cell, present it to the immune microenvironment and the immune cells, and really activate that immune response. So we now have data, published data from the Phase Ib study that was looking at magrolimab in combination with azacitidine in patients to find the optimal dose of this combination because, as I'll mention in a minute, there's definitely some off-target effects that we need to be careful for in this mechanism of action.

So as you can see here, we have the final results of the Phase Ib study and we saw some very intriguing responses. We saw several CRs. Not only did we see several CRs, they happened to be in your high-risk patients with TP53 mutations which I think was very intriguing. But the other thing that we saw was a clear dose dependent anemia when patients first started magrolimab. So in the first week when you give patients this antibody, CD47 is actually expressed on the red blood cells and you have this very dose dependent hemolysis. And you can see the hemoglobin levels drop pretty dramatically in the first couple of weeks. And over time, as these old red blood cells get cleared and the body has — is able to produce normal red blood cells, you can see the hemoglobin slowly improve. But this is important to note because you can have drops of up to 2 grams in the hemoglobin in the first couple of weeks. And so the dose that is moving forward has initiated a priming dose in the magrolimab study and this is moving forward, again, in a Phase III study. It's already completely accrued and we're anxiously waiting results, again, in the next couple of years. So there's a lot of exciting drugs moving forward in combination in the high-risk MDS space.

I did want to mention a negative Phase III study that was also presented last year and that was a combination with a NEDD8 inhibitor called pevonedistat. And the reason I mention this negative study because, as you can see, the interesting thing in the study was the azacitidine arm, which is the control arm, actually performed really, really well. These patients did far better than expected and as a result, obviously, when you combine an effective therapy with another therapy, if your control arm is effective, your study is not going to be positive. And the question is, well why did patients on this trial do so well with azacitidine alone? And I think it gives us another clue that maybe we're not utilizing and dosing azacitidine or whatever HMA you use for our MDS patients the way we probably should. Are we abandoning therapy too early? Because you can see here on the slide, patients that actually got therapy for greater than 3 and greater than 6 cycles actually did the best whether it be in the control arm or in the study arm. So I think there's maybe a lesson here. And hopefully, as we learn more with these ongoing trials, we'll not only learn how to use combination therapies if they get approved, but also how to use HMA therapy better for our patients.

So in summary, several novel combinations are being explored in large Phase III studies. There appears to be some preliminary data that shows that these combinations are tolerable. There's a benefit to our high-risk MDS patients though obviously, we're anxiously waiting for our confirmatory studies. And maybe, as I said, there's some learnings here to tell us how to best dose our hypomethylating agents that we have available at this time.

So I wanted to briefly mention this precision medicine in MDS. We talked about we've identified multiple mutations. We've talked about we have an IPSS-M score that incorporates mutations. But are we actually targeting these mutations specifically? And how do we do that, again, borrowing from the AML world?

So we know that MDS patients also have IDH mutations just like AML patients, maybe not as frequent. And so this is a really unique set of studies looking at both ivosidenib and enasidenib in MDS patients with IDH1 or IDH2 mutated disease.

And we don't have a lot of data on this and it's a small number of patients, but I did want to point out both in high-risk MDS, patients had a failed azacitidine as you can see here in cohort A, and in low-risk MDS, patients had a failed EPO as well as high-risk naïve patients with no treatment. They received single agent ivosidenib or single agent enasidenib, very similar dosing to AML, and they actually saw a fair number of patients that had a response. And so now, we're talking single agent mutational directed therapy for our MDS patients. We're not talking about combination with hypomethylating agents. So I think this is an intriguing study. We're anxious to see more. But I'm quite excited to explore the potential of doing these targeted trials for our patients and not always relying on that HMA backbone if they do have a mutation that can be targeted.

So again, response rate in these cohorts were, I think, pretty decent, 42 to 69%. The overall survival was comparable to what you would see with HMAs. And these drugs, we have experience from AML, appear to be well tolerated. We know how to dose them. We know about differentiation syndrome. So again, something to be watched.

DR LOVE: Do you see FLT3 in MDS?

DR BORATE: It's very rare. And people now think if you see FLT3 in MDS, it's like NPM1, you just caught it as it was evolving to AML.

DR LOVE: The other thing I was going to ask you is, in the IDH patients, I don't know if there are enough of them, but the IDH MDS patients, do they respond better to venetoclax/HMA?

DR BORATE: So I don't know in the Phase Ib study that was published whether there were enough because IDH mutations in MDS are actually even more rare than in AML. It's maybe 5, maybe at the most 8 to 9%. So I don't think we've really treated enough to know. I would suspect yes, but we don't have that data yet. Hopefully, the Phase III will tell us.

DR LOVE: Please continue.

DR BORATE: So last, but not the least, I think this is a vexing problem across multiple cancers but especially in the myeloid malignancies, what do we do about TP53 mutated MDS? I think we all know these mutations are associated with poor survival despite similar response rates. So you start the patient on treatment, they actually respond. The problem is they lose their response and overall, these patients just do poorly. You can see here, the data showing maybe 9 months versus 20 months in TP53 mutated versus wild-type. So this is a study, again, the ASCERTAIN study, where they were looking at the bioavailability of oral decitabine/cedazuridine compared to IV decitabine to establish that these are bioequivalent. And there was maybe a signal that this therapy could potentially help the TP53 mutated patients. So this — the design of the study is patients got the first cycle of either the oral agent and then they alternated to the IV agent, and vice versa. And this was a way to establish bioequivalence in the same patient. So this was not necessarily looking at response. It was really looking at bioequivalence, so just to clarify that point.

So looking at the ASCERTAIN study, you can see that the patients that were enrolled in the study had a very similar spread of MDS type mutations as you could see in a normal MDS cohort except they did seem to have a higher proportion of TP53 mutated patients. So they analyzed this population to see how these patients responded and they defined these mutations as either monoallelic, that means one TP53 mutation, or biallelic, meaning they have one TP53 mutation or a 17p deletion and at least one TP53 mutation. They did not look at loss of heterozygosity in this patient population.

So when they looked at leukemia free survival and overall survival, obviously the TP53 mutated patients did slightly worse. But overall, they saw that their survival was not that different from the TP53 wild-type patients, presenting an intriguing hypothesis. Maybe this oral combination is something that we could employ in some of these TP53 mutated patients maybe to build on with other novel agents to see if we can improve outcomes.

So as I mentioned before with magrolimab and others, there's multiple new therapies being explored in TP53 mutated MDS in combination with hypomethylating agents. And as with most MDS but especially with this subtype, allogenic stem cell transplant remains the only curative therapy that we have for this patient population.

So I wanted to end with, you know, what do we do when we see an MDS patient in clinic? Starting with, will the IPSS-M change the way we clinically approach these patients? You have a patient in front of you, you calculate their score, maybe you tell them based on the IPSS-M, you have a very high-risk disease. How do we clinically apply that? And does it even matter? Does it actually change anything?

So this was a study, again, presented at ASH where they looked at 2 large MDS studies that I mentioned before, MDS STIMULUS-1 and STIMULUS-2, and they looked to see, how does a patient population change if they're scored by IPSS-R and IPSS-M? And just to refresh people's memories, these are 2 randomized placebo-controlled trials looking at sabatolimab. One was STIMULUS-MDS1 which I just discussed, and STIMULUS-MDS2 which is a Phase III study which is ongoing.

So they compared the patients, they looked at IPSS, IPSS-R and IPSS-M, and they actually found there was substantial upstaging from IPSS to IPSS-R to IPSS-M. And I discussed this specifically to think clinically of the MDS patient that you see in your office, but I do think there is value when using the new classification because I think it gives us the most accurate prognosis for the patient. Right now, the standard of care for MDS remains hypomethylating agents, but in the next year, maybe next 2 years, you might have more options available. And then, you may decide based on the scoring system you use how to treat this patient right from the get-go. Are you going to start with ESAs? Are you starting with hypomethylating agents? Or are you starting with a combination of hypomethylating agents plus Y once we have data from these large Phase III studies?

So I wanted to summarize this by saying, as the IPSS-M becomes more clinically utilized, we are incorporating this mutational data to stratify our patients. And I think this will change how we treat and counsel our patients, both in the low-risk and high-risk MDS settings, especially if you're now going to upstage your low-risk patients.