Part 1: Emerging immunotherapies in AML

Naval Daver:
So to start the panel, I’ll maybe go through some of the immunotherapy updates that I found useful from the ASCO EHA meeting. And then turn it over to Dr Subklewe, and others who have some ideas and suggestions on what they think will be going forward. So I think this was a very interesting meeting. We are definitely seeing more and more immunotherapy emerging in both AML/MDS over the last year, year and a half.

Naval Daver:
I think to me, the big ones from this meeting were bispecific antibody updates, there was data from AMG 330 that was shown at the ASCO meeting by Dr. Farhad Ravandi. Marion was one of the co-authors on that one. And it does look like in my view that the safety profile is starting to get better and better with these agents, probably because we’re getting smarter and how to use them. Ramp up dosing approaches are being implemented. There also does seem to be early use of drugs such as tocilizumab which seemed to mitigate the CRS that can be seen with these agents.

Naval Daver:
The response rates still I think are in about 25-30% range, but hopefully as we get to the higher dose levels. And as was shown by Farhad Ravandi at the higher dose levels, especially the 240, 480 and above. There does seem to be responses especially in people who have lower blast disease, and favorable T-cell to blast ratio. So I think again, just like molecular, this looks like it’s going to be going more into a selected population where we need to identify patients that have favorable immune profile, high T-cell infiltration, and probably not high blast, which is kind of very similar to the profile we’re seeing with blinatumomab, for example in ALL. So that’s, an interesting area that’s developing I think couple of others were the CD47 antibody, which is a macrophage checkpoint. This inhibits CD47, which is a major inhibitory pathway to macrophage phagocytosis promoting a “eat me” signal.

Naval Daver:
So I presented some of that data at the EHA meeting, and David Solomon presented some of it at the ASCO. To me really the safety of this combination is very attractive. Now we’ve treated probably across 90 or so patients in AML/MDS, the 60-day mortality remains zero in the frontline setting, which is very encouraging in this multicenter study, and I think especially in the TP53 mutated AML, we are seeing a very early signal for high response rates, along with good duration of response. We had 12 patients presented more enrolled and hopefully will be updated later this year. But of those 12, nine had a CR/CRi and the duration response has not been reached yet with a median follow up of about 8-9 months.

Naval Daver:
So looks encouraging, safety is very good. On its own maybe very good combination in MDS is aza-magro maybe in TP53 AML. But I think the interest for us, at least Anderson, is how this could be potentially built into a triplet, or other combinations for the non TP53 AML. Where aza-ven of course is doing very well, and Andrew will talk about that. But we can always do better the two years survivals are still around 40%. So can you bring that up? So with that, I’ll kind of turn it over to Marion for some of her thoughts.

Marion Subklewe:
Yes, so also welcome from me. So just in general, I think at ASCO focusing now on BiTEs and CARs, we can see that there’s still some struggle in translating the success from B-cell lymphoid neoplasia treatment, and ALL treatment to other entities just in general. And I think that’s also demonstrated that besides AML the most interesting stuff that was presented in my eyes in BiTEs and CARs setting was in relation to BCMA multiple myeloma, where there was very interesting talks about bispecifics, and CAR T-cells.

Marion Subklewe:
Now coming back to AML. I think this underlines the still this struggle we are having with the target antigen. So, in B-cell neoplasia as well as in the other entities, we are still targeting allinit antigen and the presentation on AMG 330 targeting CD33. I just want to sort of maybe explain why it’s taking a little bit longer than we probably expect getting clinical results out. And that is due to the non-restricted target antigen that is expressed within the myeloid compartment, and I think what might have been underestimated is that we also have a large antigen sink.

Marion Subklewe:
So when monocyte macrophage is expressing CD33, and one has to point out that in all these trials that were presented at ASCO and also at EHA, we are still trying to find the dose, so we are still moving up. And are now up in the AMG 330 trial, and a three step approach going up to 960 mg. And if you consider what we are applying in blina, where the maximal dose is 28 mg.

Marion Subklewe:
It highlights that the different target antigen situation is a challenge, just to get enough molecules in these heavily pretreated patients. And that was also demonstrated by the presentation by Ravandi with the update of the 330 trial, where 60 patients were now presented and you can see from the patient demographics most of the patients had four or more prior treatment lines, and almost 50% of the patients presented were opposed allogeneic stem cell transplantation.

Marion Subklewe:
So these are really very heavily pretreated patients. And I think first of all, it’s really nice to see that the safety is and was advancing experience in this field is okay. So CRS is the most common side effect we are observing, we see no neurotoxicity and with toci and also steroids, I think this is feasible. And one can see in the higher doses as already unwell set responses, but clearly it’s still too early to say. When you look at the biomarker aspects you can also see that was also a poster at EHA, that clearly the CRS and other side effects correlate to tumor burden and is also negative factor for response rates.

Marion Subklewe:
So, I think what all these data and as was also already mentioned by you that we need to go in earlier treatment lines with a lower tumor burden, as already been shown actually, there’s a great similarity to blinatumomab. It was already shown in the TOWER trial, that there was a better response in patients in salvage 1 compared to salvage 2, which was 11-something months compared to five months overall survival. And then clearly the data and the MRD setting were really nice with 85 of the 113 patients in the blina trial showing MRD conversions.

Marion Subklewe:
So I think there is a development to use these kind of strategies, the bispecifics into a much earlier treatment line where you have less tumor burden, and clearly better T-cell fitness. And that was also actually shown in one of the posters now in the 673 trial. So that is the CD33 BiTE half-life extended, where you can clearly see that the dose also correlates to T-cell activation. There they assessed the CD8 T-cell activation, which is needed as a mode of action in this setting.

Marion Subklewe:
So, I think still early data and I think we’re probably getting a little bit impatient, but I think antigen sink is one of the factors that is taking more time that we expected, clear this year as the most common side effect, which I think we are struggling successfully but it is something to clearly watch out for. And now getting up to the higher doses we see some responses, but I think we are still not in the optimal patient setting and have to go in an earlier treatment line.

Naval Daver:
Yeah. I mean, thanks a lot Marion I think one of the big challenges right off the bat for AML is that, which is a good challenge is we have so many treatments for AML available today compared to what blina had to face eight years ago or so when it came in, where it was kind of easy to put it in first and second salvage. Ino was probably the only other thing competing with it.

Naval Daver:
And I think that’s the striking difference in the TOWER study is even in the blina early studies, there was at least one and two. So it kind of got a fair chance here, as you said, median was four salvages half with prior transplant. And I mean, there’s no therapy that I know of that works. So, I mean what do you think would be the way to develop this? Should we be looking at focused MRD populations where maybe there’s less competition? Should we restrict the trials to maybe only salvage one and two, which may impact enrollment or do you think there’s a role to start combining them with PD-1 thinking, okay there’s going to be potentially more toxicity but if we can really enhance the response rate, then people may accept it. What are your thoughts for the future for BiTEs?

Marion Subklewe:
So I think both things have to be implemented. So I think maybe starting with the second part of your question, I think PD-L1 is only playing a role if you have a T-cell compartment that is fit for a response. So I mean, it’s clearly been shown that PD-L1 is not that relevant in primary diagnosis. So it’s something that’s evolving, it is somehow an adaptive resistance mechanism in AML. And it’s clearly been shown that it is upregulated on AML cells upon IFN/TNF exposure, and it’s clearly upregulated in patients with a T-cell response receiving bispecifics.

Marion Subklewe:
So I think we need to move AMG 330 into the MRD setting, and I think ideally we would do it in the MRD setting actually in combination, because there you have still a T-cell compartment it is fit enough that is responsive. But I think there will be a relevant number of patients where you still have this adaptive immune escape mechanism with upregulation of PD-L1. So I actually believe that you need the combination in the MRD setting.

Marion Subklewe:
And clearly that would be the struggle if this is a bridge for transplant strategy, then you have the problem with the checkpoint inhibitor prior to transplant maybe there has to be I don’t know if that’s the ideal setting. And clearly you have to start first with just the bispecific and then do the combination. So I would really love to use the bispecifics in the MRD situation maybe without the checkpoint inhibitor initially, as maybe also possibility bridge for transplant. But I don’t think anything beyond salvage 1 is going to-

Andrew Wei:
Marion if you’re going to take immunotherapies, or BiTEs into the MRD setting. How would you define the entry point in AML? What level of MRD, what method?

Marion Subklewe:
That’s an excellent point. You know that the ELN, and everybody’s working hard to get MRD flow harmonized so that we can do it worldwide to a certain standard. Clearly, I think the NPM1-positive patients are a good population similar to the RELAZA trial, we have very established and a validated MRD marker. I’m still a little bit hesitant to use MRD flow in this setting. I think it really has to be harmonized approach at a maybe central lab where I would say, at the current time .10 to the -4 has to be the cutoff.

Andrew Wei:
I’m just reflecting upon the interesting paper presented at EHA with respect to the large British Mylotarg study where they looked at patients after one induction and then patients who were MRD positive got randomized to a variety of intensification strategies, and they found that at the end of the day there was no difference in terms of survival. And so I find it difficult to reconcile putting those sorts of patients into an MRD immunotherapy study with BiTEs when potentially you could get the same result with chemotherapy. So I still don’t know if we really know how to define MRD in a robust way, where the false positive and false negative rates from cut points at post-induction or post-consolidation are stringent enough to really identify a truly adverse risk population. I’m not sure what to what do others think, it’s difficult.

Naval Daver:
Yeah I mean, I agree. I think it’s very challenging. And again with ALL, I think even defining MRD was much more clear cut than AML. But I think starting with maybe higher threshold where people would really not disagree, whether it’s 0.1, or some companies are even showing 1% MRD, which I don’t think is MRD. But not going into these really small populations, because I agree with you it’s very hard to know in AML if you have 0.02% on a high sensitivity flow, whether this is going to be reflective of that outcome. But with ALL I think it was kind of almost a cycle. Where, blina brought MRD into the clinical limelight.

Naval Daver:
We didn’t really… We did it, but we didn’t really act on it until we had the drugs. So I think if we do start seeing the drug that frequently is able to get you MRD negative, and that’s associated with the long duration of the CR, then they will be interested. It’s almost kind of a reverse story where the drug may end up helping us all reconcile and define MRD, rather than having MRD definition completely set up. That’s kind of what I think from ALL world. But Amir, what do you think?

Amir Fathi:
Well, I think there are some challenges here AML is not ALL. ALL in many ways when it comes to bispecific, and CAR T-cell modalities is a much easier disease to treat for a variety of reasons, better targets, better therapeutic index. At the same time, we see a similar range of tolerability and toxicity issues with these agents in both AML and ALL. The other challenge with ALL is, MRD is much better defined in terms of both prognostic and predictive aspects. We know what to do with it, with AML we pretty much except for some very prominent exceptions, can only provide prognostic information with the MRD at this stage, albeit if it’s MPN1 perhaps or core binding factor alterations. Those would be the exceptions.

Amir Fathi:
I do agree with the group here that it does appear that patients who have a lower burden of disease tend to do better, which is probably not… Or tend to respond, which is probably not surprising. And the other aspect is of course, based on recent data. And data also in the past that primary refractory patients seem to be a group of interest for this subgroup. MRD has always been a challenge for me in terms of how to proceed even in ALL.

Amir Fathi:
Because in children especially, I feel that they are much more… The oncologist community are much more nimble in terms of what needs to be done earlier, and with lower thresholds of MRD. With us the data that has emerged really mainly from Germany is with blina with ALL. And we are now trying to figure out how to potentially use that also in AML. What do you do for example, if you get a value that’s slightly less than .01? How do you communicate that to the patient? What do you say you have MRD is probably not as good as having zero, but we’re not going to do something about it.

Amir Fathi:
Those conversations are challenging in the real world. Although, I do agree it’s much more important to get data from relapsed refractory patients in general. I think we’re still dose escalating, we still trying to figure out the schedule, trying to still figure out what dosing schedule is most effective. I think that’s most important. I think before taking the more prominent steps and trying to figure out what thresholds of MRD is ultimately going to be the one that we use.

Amir Fathi:
So I think defining efficacy thresholds is more important right now and then moving on to… And that’s what happened, I think with blinatumomab ALL. So those would be my thoughts.
Naval Daver:

Okay, in the interest. Oh, go ahead Marion.

Marion Subklewe:
No, I just want to… I completely agree and I think one of the strategies maybe to address at least the problem is to have always two time points of the MRD sort of a confirmatory MRD, right. And not rely maybe on one value, right. Which, can be very arbitrary maybe.

Part 2: Venetoclax-based regimens

Naval Daver:
Yeah. So I think in the interest of time we’ll move on, and I’m going to turn it over to Andrew to discuss some of the venetoclax updates. There was a lot at EHA and ASCO, and let us know his thoughts. And then we can have a panel discussion Andrew please.

Andrew Wei:
Thanks Naval. So obviously, at EHA the late breaking abstract session, probably the highlight for the AML field was the presentation of the VIALE-A randomized trial by Courtney DiNardo. And this was the culmination of a long development program which commenced in about 2014. Where low dose ara-C and hypermethylating agents were combined with venetoclax to try and improve the outcomes with that drug, because the monotherapy activity was only modest with about 19% response rates which were not particularly durable.

Andrew Wei:
So this is on a background of an area where it’s really frustrated clinicians for many decades. So, first of all in practice the Swedish registry tells us that patients are over the age of 75. Almost 50% of patients treated palliatively, with palliative intent. And for patients over the age of 80, 70% of patients are treated with palliative intent. So an enormous amount of therapeutic nihilism in the older AML setting.

Andrew Wei:
The second major issue which has faced trials in the older AML setting for decades, is that many trials have demonstrated improvements in response rates and also promising seeing outcomes in early phase studies, but this is not translated into any improvements in survival in the phase three setting. And so this has really been a common factor for many decades. So, the VIALE-A study was obviously positive, and was impressively positive because the curves remained separated, they remained widely apart with a 34% reduction in the risk of death with a large study of over 400 patients.

Andrew Wei:
And also the other major feature was a really high response rate. So 66% of older people, or patients unfit for intensive chemotherapy could achieve a response which is obviously associated with a higher rate of transfusion independence. And furthermore, the duration of these responses was quite impressive. So the patients that achieved CR with azacitidine, plus venetoclax, 37% of patients were in this category and the duration of response was 17 months. Furthermore, in subgroup analysis there were particular subgroups which were quite impressive with respect to their subgroup sensitivity to venetoclax, such as the IDH-1 and 2 mutant subgroup. The hazard ratio for this group was about .28, and I think this is particularly interesting because there was there for a large survival advantage for venetoclax plus azacitidine compared to azacitidine in the ADH mutant group.

Andrew Wei:
However, we didn’t see the separation in the randomized frontline trial with enasidenib plus azacitidine. So it will be interesting to know what the actual median survival was with venetoclax, because in the enasidenib combination was 22 months. And it will be interesting to compare whether that median survival was perhaps superior with venetoclax plus azacitidine.

Andrew Wei:
There was a companion study also presented at ASCO and EHA which was the VIALE-C study, which was low dose ara-C plus venetoclax, versus low dose ara-C plus placebo. Again, with some additional follow up of six months, there was a survival benefit from 4.1 to 8.4 months with a 30% reduction in the risk of death. Again, superior responses translating into higher rates of CR transfusion independence, event-free survival, and also quality of life. So I think these two sets of studies lay the groundwork for venetoclax perhaps being established as a new standard of care for the treatment of older patients. And this leads now onto many new questions for the future.

Andrew Wei:
How do we improve outcomes in those that don’t respond as well to these combinations? How do we augment responses and duration of responses with combination of targeted therapies? Again, Courtney DiNardo gave us an early insight into the potential for the addition of either ivosidenib to venetoclax, plus azacitidine as a potential triplet. And also the role of conventional therapies such as allogeneic transplant, to take advantage of the really high and rapid response rates achievable with these regimens, perhaps selecting patients who are potentially destined for a poorer risk outcome, and taking them to transplant to try and take advantage of the improvement in the quality of life in these older people. So, I think exciting times for venetoclax. And this is just the beginning, no doubt.

Naval Daver:
Yeah, thanks a lot, Andrew. I mean, I totally agree. We’ve been using venetoclax very extensively here at MD Anderson as well. And in the Phase II’s the data looked very encouraging with CR/CRi rates of 70 to 80%. And it’s quite interesting that in Phase III, we’re actually not seeing significant attrition of the early response rates. And the survival there is a little bit decreased but less than I think many of us were worried about.

Naval Daver:
So it’s really good to see that this was translatable from the initial 8, 10 center experience to almost 100 centers and continue to show almost similar response rates, early mortality and survival. And I mean, I think in the United States where it is already approved, I think this is now the standard of care. I don’t really see where one would use aza alone, and there may be some rare exceptions.

Naval Daver:
But aside from that, we did have some people who are still let’s say not convinced, or unsure just because of what you said that there had been a lot of doublets that had shown early high response rates, almost 70-80% and some early signals for survival, but then there was more toxicity potentially, that was not seen when the big centers are managing them.

Naval Daver:
So I think it is… I mean, I think there are two big groups where we’re still not clear about whether aza-ven is clearly that much more beneficial. And even if it is beneficial, there’s still a lot of room for improvement. So I think the two big ones in my mind are TP53 and FLT3. In the TP53 mutated population we’ve seen across the phase IB, phase II, our experience at MD Anderson along with you, as well as now the Phase III that the response rates are about 45-50%. They’re not bad, but unfortunately the survival usually ends up between 5 to 8 months, which is arguably maybe better than aza, maybe not. Depending on how you look at it, but definitely room for improvement.

Naval Daver:
So I was going to ask, Amir what’s your take on TP53 AML, HMA events. Is it something you’re using in Boston? Are you looking at other strategies? What do you think the future will be for that group?

Amir Fathi:
Well I think we are, ultimately this is a subgroup of patients that uniformly do poorly, regardless of the choice of treatment. And I think from my perspective, this is my view alone. It’s not that TP53 mutated patients routinely and consistently fail induction traditional chemotherapy, it’s just that they fail it at a high rate and tend to also get all the toxicity of that treatment.

Amir Fathi:
But now we have a more gentle combination, which may be as bad but not as severely toxic. So for that reason, if we are aware that a patient has TP53 mutation we generally move towards a more gentle hypermethylating therapy combination, which currently is with venetoclax. But it was also important to say on occasion, you’ll find another target in these patients such as an IDH mutation rarely a FLT3 mutation.

Amir Fathi:
On occasion, a complex karyotype will give you a clue that you may have a p53 mutation so it may be wise to wait and get that value. p53 staining can be a surrogate of p53 mutation. So all that is important to perhaps make a decision. I don’t think the data that’s emerged from HMA, venetoclax in p53 mutated patients is poor. It is decent it’s approximately half, maybe around there that respond to treatment. Unfortunately, the response to treatment is short lived, which is quite a challenge. And the only chance you have with these p53 mutated patients ultimately, is to try and transition to transplant and a lot of the time thats it’s not effective either.

Amir Fathi:
It’s also important to note that not all p53 mutated patients are the same, or are equal co-mutations matter. Probably a location of the p53 mutation matters, that as we all know, the allelic fraction of the p53 mutation may be important. Some data emerging now suggests that if you’ve brought down the burden of p53 prior to transport, you may have more success. So I think all of those aspects are important. And then the question of MRD comes up again here. Do you have better success if you suppress MRD prior to consolidation, after consolidation… You know, most people probably say yes, it’s just that it’s harder to do so, and sort of has to be the genetics of disease.

Amir Fathi:
So the challenging disease a heterogeneous population, no therapy is highly efficacious survival continues to be a problem, but yes because of its tolerability, relative efficacy, I do use HMA venetoclax in this group of patients.

Naval Daver:
Yeah, thanks Amir. I mean, I think for now in the US if a physician asked me should I use HMA-ven versus HMA. I probably would still be inclined towards HMA-ven in TP53. I guess the question is more in the future. There are drugs as you know APR, CD47 that are hopefully going to improve but again, this is Phase IB, and then maybe as doublets or triplets. So we will see more what emerges, which you said is very true about the VAF and the concomitant complex cytogenetics or not. Because we’ve done analysis, it was presented at ASH by one of our fellows showing that those who have low VF, less than 40%, was the cutoff we found statistically, and without having the chromosome 17P. They actually tend to do almost as good as people who don’t have a TP53. That’ll be published soon.

Naval Daver:
So you’re right. It’s not the TP53 per se, it’s a lot of variables that go into it. And Marion in Europe, how is the data with the HMA-ven low dose ara-C ven even going to change things using both of those will become available, will it be one? What’s the feeling in the European community?

Marion Subklewe:
Yeah, so it’s not approved yet. But the reality is, I think in a lot of centers, we already use the combination and aza and venetoclax is sort of our standard of care, although we still have to ask insurance and so on for coverage, but it’s granted in all cases, actually. So I expect approval, actually of both combinations within the next six to eight months in Europe. And it’s also reflected in the clinical trials, right. So even in the immunotherapy trials, where they for example the anti-CD70 antibody and so on.

Marion Subklewe:
They all randomized now against both substances and aza is not considered to be standard anymore. So I think that’s going to be the reality and we have to sort of then move on, adjust, combine. But in aza on its own is also although it’s still not have the approval used only in rare cases elderly, and we cannot do the monitoring that well or who don’t wish to have the more aggressive treatment.

Naval Daver:
Okay, great.

Andrew Wei:
Can I ask you both, or all three of you now. If you have availability of venetoclax, how do you now choose between a patient that’s fit or unfit? And who would you still give intensive chemotherapy to. Will you follow the yeah, all over 75 year olds would get the ven-aza, some would still get intensive chemotherapy, and vice versa for the younger people. Would you universally give intense because they fit for it or otherwise? So, how do you think your practice will evolve in the light of the data?

Amir Fathi:
You know this is an extremely prominent question, a lot of people are asking it. In the United States, a combination is approved for patients who are 75 and up and those who have comorbidities or functional limitations that render them unsuitable for induction chemotherapy. So we as of yet cannot recommend the combination HMA-venetoclax for patients who would otherwise be younger or induction eligible.

Amir Fathi:
Having said that, the data is very exciting, composite remission rates, especially among IDH mutated patients and FLT3 mutated patient. Some would argue, obviously P53 mutated patients not that great. Certainly, perhaps have not been compared head to head, but based on historical data seems to compare favorably. So I mean, I think I’m hopeful that we’ll be able to answer these questions over time, but as of yet, our approach remains. Induction chemotherapy for the fit patients. HMA-venetoclax for the unfit patients.

Amir Fathi:
I will say one thing it might vary a little bit from what my colleagues have said on this conversation. HMA-venetoclax for me hovers somewhere between induction and HMA, both in terms of tolerability and advocacy. And there are some patients that I do very much worry about with HMA-venetoclax. I work in Boston, the median age here at our General Hospital is quite larger. On occasion, I’ll see an 88 year old, a 94 year old in whom I worry about giving HMA-venetoclax honestly because I do think that they may get in trouble.

Amir Fathi:
So I think there is still a role of HMA monotherapy in a small subset of patients who I worry about tolerability, marrow suppression, complexity of treatment in whom I might use monotherapy.

Naval Daver:
I think that’s a great question Andrew, and I think at MD Anderson, we’re kind of probably moving our bar already. So above 75, for sure we’re not getting anybody induction. I think the big debate area that gets discussed pretty much on a weekly, or two times, three times weekly basis is those were between 65 and 75.

Naval Daver:
And I think there’s a little bit of a difference here where we’re not, let’s say convinced HMA-ven alone is enough. And maybe chemotherapy has some role but has a lot of toxicity. So this is where the triplets are starting to become for research purposes our area of focus. So if you have a 65, or 60 to 75 if you identify an IDH, or a FLT3, or TP53. We’re definitely prioritizing a HMA-ven plus IDH, HMA-ven plus FLT3, these are trials, or HMA plus TP53 targeted therapy, whether it’s APR, CD47, or even triplets of those.

Naval Daver:
So that’s, kind of moving the bar where we believe that the triplets will further push what else already looks like a good outcome with HMA-ven without potentially still having the same toxicity? And I think that’s the question with the triplets, how much is myelosuppression? Does it started getting close to induction? We don’t think so. Based on some of the early data, the early induction mortality especially. So I would say there are very few, now co-binding factor is an exception. Right, even if you have a 70 year old with co-binding factor, he’s reasonably healthy. Absolutely, we’re going to go for gemtuzumab with FLAG-Ida has been our backbone but anthracycline cytarabine based backbone, but on the other hand if I have a TP53.

Naval Daver:
And this is very recent based on our analysis, who’s 50 years old with a high VF 50% with complex cytogenetics. We’re actually starting to push those patients more towards HMA-ven plus one of the TP53 directed drugs APR, CD47. So I think it’s becoming more molecular selection, at least in our experience than the traditional age fit/unfit. We’ll hopefully have more data with some of these triplets over time, but I think in the targeted group, they will shift the curve is my feeling IDH, FLT3 maybe TP53, potentially over chemotherapy.

Marion Subklewe:
I think everything sets so it’s the patient disease related factors is discussing the interview case basis, and also importantly always the question, are we going to go for allo or not, so that also influences our decision, right. Patients who we think we can… They’re favorable risk profile and not heading for allo, are clearly still preferentially treated with intensive interaction chemotherapy.

Part 3: Targeted Therapies

Naval Daver:
Okay, well in the interest of time for the last section, I’ll turn it over to Dr Fathi to give us a few of his highlights and things he was excited about in the targeted therapy world FLT3 IDH. And then maybe we’ll have a few questions and wrap it up. All right, Amir please.

Amir Fathi:
Thank you. So IDH and FLT3 have been historically very exciting in the last five years, perhaps a little bit less so in the last couple of conferences has been relatively a small number of presentations compared to some of the other therapies. There was a lot of excitement around HMA-venetoclax, low dose cytarabine-venetoclax, the maintenance therapeutics have been studied. In comparison, the excitement for FLT3 and IDH was more muted I would say during the EHA and ASCO meetings, there were a few presentations that I personally thought were interesting and relevant.

Amir Fathi:
There was a presentation of the IDH-2 inhibitor enasidenib plus azacitidine in patients with IDH-2 mutated AML. It was a head to head study versus azacitidine. And it was small, it was less than 100 patients total that were studied. And ultimately, there was no significant difference in terms of overall survival or event-free survival. But overall survival between the two arms were the same. I think it was around 22 months, which was surprising in and of itself, and probably an outcome related to the size of the study.

Amir Fathi:
But the rates of composite remission, were markedly improved and individuals without enasidenib and azacitidine. Approximately 55% versus in terms of composite remission, as opposed to a composite remission of teens when it came to HMA monotherapy. So that, I think confirmed much of what we already know, which is if you combine these targeted therapies, IDH inhibitors with the three inhibitors with HMA, you tend to have a more pronounced rate of response. You may be able to bridge transplant eligible patients to transplant, you seem to drive down IDH mutational burden, which is a form of analysis on these patients.

Amir Fathi:
There’s also an abstract presented that looked at the biology of response in patients who received HMA and IDH inhibitors. It seemed, I should say the larger response on relapse, at least with relapsed/refractory IDH inhibitors, when there is a relapse, there tend to be secondary IDH mutations, or additional alterations, and receptor tyrosine kinase pathway such as FLT3 and RAS. It didn’t seem to be that this was a prominent modality of resistance, or disease progression of patients who received the combination.

Amir Fathi:
I think that data still probably early and emerging I suspect, but that was also an interesting presentation that emerged. I do think that, although early, the data that’s emerging from the MD Anderson, in terms of the combination of ivosidenib and venetoclax is highly intriguing and exciting at very high rates of composite remission in patients who received the combination.

Amir Fathi:
And I think, two or three cohorts that are presented well above 80%. That’s also exciting. And that study is ongoing, and we hope to have more long term results that emerge from that. Ultimately, to have doublets or triplets oral therapies that are highly effective for AML patients it is a huge step forward. If we do get there successfully in the years ahead. In terms of FLT3, I think probably one of the more prominent abstracts that was presented, at least in my view was the outdated data from the ADMIRAL study that looked at gilteritinib.

Amir Fathi:
As you know, the ADMIRAL study compared gilteritinib in randomized fashion to a variety of relapsed/refractory regimens, both lower intensity and higher intensity and showed a higher rate of remission, a higher rate of positive remission, higher rates of overall survival which was the primary endpoint. And this study showed that overall survival advantage persisted years out, and that patients who ultimately have the better survival were also those patients who tended to respond which wasn’t a surprise. Overall, the composite remission rate of gilteritinib that study was around 50%, the composite that that complete remission rate was somewhere mid 30% range.

Amir Fathi:
And that’s very exciting in terms of this highly specific, potent FLT3 inhibitor that is currently still under study in combination with conventional regimens such as HMA, and induction chemotherapy. There were also a few abstracts that were presented on other FLT3 inhibitors, some data that’s emerging from crenolanib another targeted FLT3 inhibitor that hits both TKD and ITD alterations in combination with induction chemotherapy.

Amir Fathi:
That data is also early and hopefully emerging over the course of the next few years. So although perhaps not as exciting as some of the other targeted therapies that are emerging and presented EHA and ASCO, I believe that the realm of the FLT3, and IDH remains very active and hopefully promising.

Naval Daver:
Yeah, I mean, I think it’s almost kind of a established area right where we’re all checking FLIT3 at diagnosis and relapse, which is good. I mean, they become kind of the established standard approaches in I think most of the world. And of course the next questions like you said is and I think we will see a lot of this data in the next two years of doublets versus triplet in salvage.

Naval Daver:
One of the things I found interesting was Catherine Smith’s presentation, she had shown it at ASH and updated it here looking at emergent mutations at the time of relapse post gilteritinib with ADMIRAL study. They had a small subset about 40 patients, where they were able to get pre- and post- NGS broad sequencing, and they found that in about 40% of the patients there was a acquisition. I wouldn’t say acquisition emergence because it may have been there and expanded, or it may have been a truly new acquisition but emergence of a MAP, RAS arrest pathway mutation, which includes of course, your N- and K-RAS, and RAF, PTPN11 and NF-1.

Naval Daver:
This is also something we have frequently seen as you know, Amir and Andrew and others in IDH patients as well, the same kind of pathway MAP kinase RAS is emerging and also we’re seeing in venetoclax. So, I think this is going to be a challenging area. And of course, there have not been any great RAS inhibitors in AML treatment. So how do we overcome this? The other thing she showed was that there are people who can still have on target escape with the F691 which is kind of similar to the D315 in CML world activation lock mutation.

Naval Daver:
So, even with these highly potent second generation FLT3 inhibitors, we’re still seeing most of these people are having parallel escape pathways and hopefully with combinations, whether it’s HMA, or HMA-ven, or chemo we can overcome these, but we will have to see, any other thoughts?

Amir Fathi:
I was just going to say I think it’s what you mentioned is very important, interesting. Catherine Smith is doing some very good work in this area, developing real science. Around the mechanisms of resistance when it comes to FLT3 inhibitors in general. I think it’s incredibly important when it comes to FLT3 inhibitors for individuals that are treating patients to understand sort of the breadth and unique aspects of individual FLT3 inhibitors, for example drugs like sorafenib and quizartinib are not particularly active in FLT3 disease. Whereas gilteritinib is far as we think midostaurin is, and even with gilteritinib you mentioned that F691.

Amir Fathi:
We all talk about DA35 as the prominent TKD alteration and other resistance mutations, but F691 is not too far behind. And most of our FLT3 inhibitors do not inhibit that alteration. We published the paper not too long ago, at our institution looking at a drug that’s been approved in other solid tumors, cabozantinib which tends to be active in F691. On occasion, you’ll get a FLT3 resistance mutation TKD alteration as a mechanism of therapeutic escape, and you have to think outside the box, perhaps find the targeted therapy that may have a role in suppressing that clone. The alterations of other RTK pathways are as you mentioned, also an increasingly important. And we need drugs to also target those alterations too.

Naval Daver:
Marion or Andrew, any thoughts from the FLT3 IDH, anything you were specifically excited about or things we haven’t covered.

Andrew Wei:
I guess I’d wonder where do we go with respect to frontline phase III studies, particularly in the older AML space mentioned in the enasidenib study was underpowered. But how do you do a pilot study now, would you do aza-enasidenib versus aza? Which country would you do that in, or do we really have to rethink our strategy with respect to taking new drugs into the front line spacing.

Andrew Wei:
Also Naval with the magrolimab data I mean, obviously really promising and potentially well tolerated combination with aza. I’d love to see a front line study of aza-magrolimab. But how do you do that now? What’s the control arm, or do you basically restrict that combination to p53 mutant AML. It makes it very difficult to bring new drugs. So be interested in all of your thoughts.

Amir Fathi:
That’s a great question. It’s very difficult now, because azacitidine may not be the comparator of choice anymore. So, aza-venetoclax is now the new backbone I would say, because it’s the new comparator at least for AML, MDS that’s another question. The other aspect is aza is going to be the backbone going forward, now that you have an oral azacitidine and those studies are now underway.

Amir Fathi:
So and we’re working with a moving target when it comes to our standards comparer, we all hope that… At least I hope that in the future, we will have well tolerated oral medicines for patients. If that is the direction we’re going, do we really engage in three year, four year, five year long randomized studies for subtypes that in AML take a long time because they’re uncommon. When we know that an oral well tolerated medicine may replace our comparator in the years ahead. So I think this is an amazing challenge. And we have to be very thoughtful about it. I don’t think I for one can come up with a solution. I don’t think azacitidine is going to be a comparator of choice in the future.

Naval Daver:
Yeah, I mean I agree. I think there’s a couple of paradigms that I like to look at, and maybe we will go that way, maybe we won’t. I mean, I think myeloma faced the same problem in 2005. And one of the things that had to change is your endpoints of studies, because if you have multiple drugs which are all going to go into triplets. Which is going to happen whether it’s through smaller studies or larger studies, and all of them of course want to head towards approval. I don’t see how you can do a large phase III, whether you say aza-en versus aza-IDH, or aza-ven-IDH versus aza-IDH, or aza-ven-magro versus aza-ven.

Naval Daver:
These are all studies where I think the industry is going to be very reticent to support a lot of those. And I think people are not going to want to wait and the standard may move, because you may have a Phase II that shows okay aza-ven-IDH is excellent in 60 people. So who wants to do aza-IDH then or aza-ven, so I think hopefully whether it’s going to be things like EFS, or whether it’s going to be MRD endpoints.

Naval Daver:
Hopefully these are validated. I think we have to move on to those. I don’t see how we can be doing 50 Phase IIIs to get every triplet its endpoint. The other thing is I mean with the CLL world I’m always amazed how they keep using chlorambucil as the backbone. It’s fantastic everything else has been approved, but they still beat chlorambucil. So that question is going to be, is the FDA going to agree and say, “Hey, you can still beat aza, and we’ll be fine with that since it’s approved.” I don’t think that’s going to happen. What were your saying?

Amir Fathi:
But maybe we should use chlorambucil.

Naval Daver:
So aza is like our chlorambucil, right. So I think the answer is a lot of these hopefully will be single-arm maybe based on MRD negativity safety, and see if the… Now the US is always different in the way the FDA approaches, I think from others. So I think in the US things like ven-gilteritinib, or ven-IDH. The FDA will consider label expansion with robust data, especially safety not being impacted significantly. For the rest of the world, I think new endpoints are going to be needed otherwise, it’s going to be very hard to get these drugs approved. Marion, any thoughts?

Marion Subklewe:
No, I completely agree. Just to make something positive as the last comment, I mean, we’ve been using 7+3 since 1972 or something. So now we are confronted with this dilemma of combinations which is fantastic, right?

Naval Daver:
Good, yeah. Good problem. Okay with that we’re at the end of our hour, and thank you all very much. I think we really covered a lot of the important ASCO EHA abstracts. I’m sure there are others we missed. But looking forward to talking to you all and doing another update in the near future. And good night. Good afternoon and good morning in reality to everybody here. All right, thanks bye.

Andrew Wel:
Thanks Naval.

Marion Subklewe:
Yes, bye.