Johannes Schetelig: So my name is Johannes Schetelig. I’m the head of the clinical trials unit of DKMS, and at the same time, I head the cell therapy unit at the Technical University of Dresden. I’m going to summarize my impression from the SOHO meeting, especially the debate on whether or not patients with relapsed/refractory AML should receive treatment prior to transplantation. This was a debate between Roland Walter and me, and I presented on behalf of the position that patients may not need treatment to induce remission prior to transplantation, before a scheduled allogeneic transplantation. I would like to summarize the main arguments which I gave at the presentation. First and most important, there is no prospective data which shows that patients who receive remission induction chemotherapy benefit from this approach. The same holds true for trying to induce, or is even more true for patients where you try to induce a molecular remission prior to transplantation. Again for this concept, there is no evidence coming from prospective randomized trials. We can even go one step further than not even prospective observational intention to transplant studies which clearly describe the entire selection process for patients from the moment onwards, where the indication for transplantation was defined up to the point of transplantation and follow up after transplantation. So, when we talk of a standard, and we clearly admit that it is a standard to induce a remission prior to transplantation, then we talk of the standard which is mainly based on conceptual thinking and interpretation of retrospective study data mainly. So why are we discussing this topic? We are discussing it because we presented at last year’s ASH the results of the ASAP trial, and this was the very first trial in which we addressed this question. We asked the question whether patients benefit from remission induction chemotherapy prior to a scheduled transplantation, and we enrolled patients with active AML. We had two strata, one stratum for patients who failed the first induction chemotherapy, plus exposed an additional high-risk feature of the AML, which made them candidates for marrow transplant, and the second stratum for patients with untreated first relapse of AML. We’re talking of hematological relapse, not of molecular relapse here. The surprising finding of the ASAP trial was that we found virtually identical rates with respect to transplant success defined as achieving a complete remission after transplantation. Those patients who went on to transplantation had superimposable disease-free survival curves. Also, when we analyzed the data for all patients who were enrolled on this study, according to the intention to treat from randomization, we saw no difference in terms of overall survival. However, we saw important differences. Those patients who were randomized to the remission induction arm spent a median of 27 days more in hospital before transplantation and experienced more adverse events. So in this setting, we think that with these trial data, it suggests that it should be an alternative to go for a transplant directly. We’re not the only ones who generated data which shed a little bit of skepticism on this concept of inducing a complete remission or even a molecular remission prior to transplantation. There’s evidence coming from a related disease, myelodysplastic syndromes, where it is standard to offer transplantation for patients with active disease and not for patients who achieve their remission with any kind of treatment. There are also retrospective studies which suggest that the benefit of selecting patients with MRD negative disease is mainly the result of a selection mechanism which helps identify those patients who are the best candidates for good transplant outcome, but does not help in finding good treatment alternatives for those patients who are MRD positive. This is a special feature of the ASAP trial that, within this study patients were followed, and even those patients who did not achieve a complete remission with remission induction in the remission induction arm and who went on to transplantation- this information is part of the clinical trial and is accessible- so the surprising finding was also for this patient cohort who is very difficult to treat, that they had a surprisingly good transplant outcome with a four year survival- in the status of having primary refractory disease- of 40%. Which is so good that you shouldn’t deny transplant. But of course, there’s a lot of room for improvement. This is something which we clearly admit that should be built further on this, and need any possibility to further improve outcome after transplantation, especially for those patients who go on for transplantation with active disease. And with this, I would like to stop, let’s say, with this introductory part and maybe now try to carry on with the discussion. And we can start with, let’s say, reflecting on the on the criteria which we use for selecting patients for transplantation. So mainly we stick to the ELN criteria, and see that patients who have adverse risk genetics have a clear indication for transplantation. For those patients with intermediate risk, the body of information which has been published is less convincing for offering transplantation for all patients, but there is a clear mandate to think of transplantation and we start thinking about transplantation whenever additional factors predict a somewhat more predictable, let’s say, worse prognosis. And this is maybe the starting point for a discussion of who might be selected once the genetic risk assessment for a patient with a newly diagnosed AML is available and we made the experience or made the observation that a first attempt to induce a complete remission with intensive induction chemotherapy failed. There is the possibility to assess early treatment response with a day 15 or day 14 or day 16 marrow assessment. And this marrow assessment may result in a finding that the leukemic blast clearance was less than optimal. And this is where we as hematologists start thinking about a trigger point for going for a transplantation. Maybe this is the first point which we should discuss, Gesine, at your center, is a day 15 marrow assessment standard practice when you induce patients with newly diagnosed AML?

Gesine Bug: Yeah. Hi, My name is Gesine Bug. I’m head of the stem cell transplant and cellular therapy program at Goethe University in Frankfurt, Germany. And yeah, Johannes, that is what we do regularly because it is an early but very important time point for further treatment decision and planning, because patients who do not show blast clearance at day 15, they are at high risk to fail further induction treatment. And that’s what we’ve also shown in the ASAP trial because patients who then received salvage treatment with high-dose ara-C and mitoxantrone, only half of them actually had CR after this second induction. And so, it might be wise to prepare transplant upon receiving the results of the early bone marrow assessment.

Johannes Schetelig: Okay, so, let’s go through these three possibilities. We may have patients with favorable genetics, intermediate-risk or adverse risk genetics according to the current ELN criteria. If you have a patient who fails to clear marrow blasts on day 15 and has favorable genetics, is this is a candidate for you for a transplant?

Gesine Bug: It might become a candidate because we know that, for example, patients with NPM1 mutation might have a favorable outcome if we then go on with ara-C-based therapy and so it’s too early to decide upon transplant or not for these patients. But still, we should start a donor search at that time. So for patients with intermediate or unfavorable cytogenetics or genetics, unfavorable patients for sure need a transplant if they are eligible for one. And for patients in the intermediate risk group, they most probably need a transplant, so you need to be prepared.

Johannes Schetelig: So, for us, patients with intermediate risk AML who fail to clear blasts on day 15 after first induction, we factor in a variety of different factors. We have an MRD marker which allows us to monitor this patient closely so that we can be sure that molecular remission is achieved with a conventional approach. And when we don’t have a molecular marker, then we are inclined to go for a transplant if we have these three factors: Intermediate risk AML, poor response to first induction chemotherapy and no MRD marker, then this is a candidate for us to go for an ASAP transplant strategy. And of course for adverse risk candidates, I think there is no debate in Germany that those patients should go for a transplant right away and that there’s not much benefit with offering additional intensive chemotherapy. This is more or less what we think. Okay. So, we use this as a first trigger point. And when we think of the ASAP data, then we have we had those patients- you mentioned them already- who failed to achieve a remission after high-dose cytarabine plus mitoxantrone. Those patients had really primary refractory AML and this is a group of patients where we are still happy with a result that we achieved long- term disease control after transplantation for this group of patients, even though we transplanted them with active disease and the four-year overall survival was 40% for this group of patients who were transplanted with primary refractory AML in the ASAP trial. So maybe we should now move on to the to the second stratum: patients with untreated relapse, we’re talking about morphological relapse, so defined either by an increased blast count in the bone marrow or by extramedullary disease. These patients were eligible within the ASAP trial, and when we compared the two approaches, which were randomized in the ASAP trial- remission, induction and disease control have trended towards a better outcome for disease control compared to remission induction. So, when we talk about this group of patients, in an ideal world, this should be a group of patients with either intermediate-risk AML or favorable risk AML, because anyway, patients with adverse-risk AML should have been transplanted as part of their first line treatment. So, given the possibilities which we now have to monitor residual disease in remission by either molecular markers or multicolor flow cytometry. At least 50% of those patients should be patients where we have a chance to monitor disease during first remission and early on pick up a first signal that this might be a patient who develops relapse. So in an ideal world, we are talking about a group of patients who should be in good shape and have an imminent relapse, perhaps even a documented morphological relapse. Or, just at a molecular level, a molecular relapse defined according to standard criteria which we now all have available. So, once again, the question, Gesine, when you think of patients with an untreated hematological relapse, who are those patients for whom you think an immediate transplant is the best strategy? And for which group of patients do you think one should attempt to induce a remission with intensive chemotherapy before transplantation?

Gesine Bug: I think that depends on the time of relapse, so the duration of remission. Because if a patient presents with a hematologic relapse which is a late relapse, has higher blast counts, we would probably do a reinduction therapy. I think the chance that these patients would respond is so good that it is worth trying. However, in early relapses or if we are lucky enough to recognize the relapse while it is a molecular one, then we would really go to transplant as soon as possible because, what we haven’t mentioned yet is these patients might receive a sequential conditioning regimen, so we know that was also done in the ASAP study. Many patients received a sequential conditioning regimen, so we do apply quite intensive chemotherapy prior to transplant. So we can afford to do the transplant upfront and not lose time for additional intensive chemotherapy, given we have a donor available.

Johannes Schetelig: So at what time do you start? So let’s think of a patient who has favorable risk genetics and clear residual disease during first-line treatment. They have received consolidation therapy. And then, let’s say 15 months after last consolidation has a first MRD signal. At what time do you start the search? Has it already been done?

Gesine Bug: We are quite generous in starting donor search because it might take some time, but if we haven’t done this during first-line treatment, we would actually start doing a donor search at the first MRD signal. We know that from the first signal until hematologic relapse, there might be a time period of perhaps 6 to 8 weeks, so that would be enough to find a donor and to schedule transplant.

Johannes Schetelig: Yeah. So we do, and what we didn’t share so far is that there’s more information on patients with molecular relapse who go straight away to transplantation which will be presented next year, and the data look quite appealing. And so I think especially for this group of patients, it is not very attractive to give them salvage chemotherapy anymore. So, for patients with favorable risk AML, we are fortunate that we have molecular markers to monitor MRD. When you think of the possibility to offer transplantation at molecular relapse but still in complete remission, how important is it to monitor MRD? Is it something where you can really identify the right point in time for transplantation when you capture this at a very early moment, or do you think it doesn’t matter that much when to detect relapse for the first time?

Gesine Bug: Yeah. So if we do not have a very sensitive MRD marker like NPM-1, then you need a bone marrow assessment to do MRD analysis. And so in follow-up assessments after conventional therapy, we would not do a bone marrow more often than every three months. And so, of course, you can miss an early relapse. We will not have the chance to find early relapses in this group of patients. But still, even if we do only monitor peripheral blood counts, so we included patients in the ASAP trial that had hematologic relapses and still for these patients upfront transplant was successful. So it’s really a valid strategy.

Johannes Schetelig: Yeah. So what we do in order to improve, compliance with MRD monitoring is that for all patients with NPM-1 mutant AML, we now opt for monitoring in the peripheral blood because it’s easier and the same for core binding factor leukemias. Also for this subgroup of AML, we tend to monitor in the peripheral blood and not the bone marrow. The main issue is that the bone marrow is, even though it’s a minor intervention diagnostic procedure from our point of view, it time it takes, and patients don’t feel comfortable with getting a bone marrow aspirate often. Therefore, we are much more concerned about keeping a tight schedule of MRD monitoring to detect molecular relapse in time for this group of patients. It’s much more complicated for patients with intermediate risk, where we don’t have a good MRD marker. But hopefully we will improve things for this subgroup with NGS-based approaches or possibly also multi-color flow cytometry. So is there any genetic risk group where you would opt for salvage chemotherapy prior to transplantation in the setting of a molecular relapse?

Gesine Bug: No, actually not. Except, we’ve shown that in the setting of molecular relapse, azacitidine might have a role. So there is a study showing that you can achieve molecular CR by applying azacitidine at least for some time. This also gives you time for a donor search. And we haven’t talked about FLT3-ITD mutant AML because these patients have been shown to benefit from targeted therapy. These patients are underrepresented in the ASAP trial because they actually received a TKI even for salvage therapy.

Johannes Schetelig: Actually, this is also our practice. And this is what I always add as a disclaimer, that this group of patients was underrepresented in the ASAP trial. This drug is approved for patients with relapsed/refractory AML. So any patient with FLT3-ITD or TKD mutations who has either molecular or hematological relapse, once the criteria for approval are met, we give the drug and also continue with maintenance therapy after transplantation. Because our experience with this strategy is so good, we don’t want to withhold this for any patient. Now there’s another group of patients where we are inclined to offer a little bit of treatment, and those are those patients with IDH2 mutations who are especially susceptible to venetoclax. But the problem, from our point of view, is that this drug is not approved for relapsed/refractory AML. So at least in our country, it is difficult to offer this unless you have asked for approval from the health insurance company of this patient which is a cumbersome and often frustrating approach to get these approvals. And once again, although it is conceptually appealing to give this drug for those subsets of patients who have an especially high chance of achieving a remission, there is no data out, or few data out, which suggests that it is feasible. And of course no stringent evidence that you improve the prognosis of a patient. Okay. So, one last question, Gesine: when you have a patient with frank hematological relapse with leukocytosis or even hyperleukocytosis. 20, 30, 40, GBT [inaudible] white blood cells. Those are patients for whom you offer remission induction chemotherapy, or do you give just disease control measures to lower blast count and go for a transplant as soon as possible? What’s your approach?

Gesine Bug: Well, it depends on the time of relapse and on prior therapy, of course. So if a patient had salvage therapy during initial treatment then I would probably go to transplant as soon as possible. It also depends on the comorbidity of the patient, because if someone presents with hematologic relapse and we still aim for a curative approach, then we must be very careful not to bring the patient into a dangerous situation in which we cannot do the transplant anymore. So a severe infection or cardiac problems, for example, may actually hinder transplant. So these are aspects we take into consideration. And, of course, the donor availability is important. So that’s what I wanted to ask you. If we start a donor search and we realized that it might be difficult, would you go for a haploidentical donor in such a situation?

Johannes Schetelig: So, yes, we think that this transplantation is important for patients. We don’t have data from the ASAP trial on haploidentical transplants, but the available information- from my point of view- suggests that donor type should not cause a delay of transplantation, so when we think we need a transplant to achieve disease control and induce remission, then we would also go for a haploidentical transplant right away.

Gesine Bug: And in your experience, are sequential conditioning regimens feasible in the setting of haploidentical transplant?

Johannes Schetelig: Yes, it doesn’t depend on the donor type, also not on the type of GvHD prophylaxis to our current level of experience. We are both working on this topic, trying to find out whether ATG can be substituted by PTCy also in the context of sequential conditioning regimens. But it will take some time until we have real evidence from randomized control trials on this question. But the personal experience is that this is possible and therefore you have no restrictions with respect to donor choices when you want to go for an immediate transplant for a patient with active AML.

Gesine Bug: Yeah. And I think that’s really good news because that will increase the proportion of patients with relapse that actually might reach a transplant.

Johannes Schetelig: Yeah. When we think of those patients with hyperleukocytosis, quite a fraction of those patients might have FLT3- ITD mutations because this is a common feature of this genetic subtype. We said already before that those patients were underrepresented in the ASAP trial. So this is one group of patients where we are discussing a lot whether it is safer in specific contexts to induce a remission in order to have robust disease control prior to transplantation, especially when we have a plan for transplantation, where we do know that it will need four to six weeks. And this is our current experience- that once the donor has been identified and we want to go for an unrelated donor transplant, we need four weeks to get to transplant. Sometimes we have to accept a delay of one or two weeks. This is our current experience and especially in this setting, the haploidentical transplantation, when is of course a very convenient alternative because it is something which can often be organized faster. But there are also caveats, patients with donor specific antibodies especially, mothers who are immunized against their children. For them, it’s really hard to find haploidentical donors. Also for haploidentical donors, it may be a struggle to get to the transplantation.

Johannes Schetelig: Lecture fees from AstraZeneca, BMS, Novartis, Janssen, EuroCept. Advisory board: AbbVie, AstraZeneca, BeiGene, BMS, Janssen

Gesine Bug: Honoraria: Pfizer, Gilead, BMS, Jazz, Otsuka Consultancy: Novartis, Gilead, BMS, Affimed, Jazz Travel grants: Gilead, Jazz