Sequencing Strategies in Low-Risk MDS

Transcript:Jamile Shammo, MD: Can I be the devil’s advocate on that one? I was just struck by the very low response rate in the ECOG study as opposed to the HOVON-89 trial for single-agent lenalidomide. We’re talking about 10% versus 33%, and the 33% is closer to the MDS-002 study and the MDS-005 study. Explain to me: is this strictly looking at different response criteria or is it patient selection?

Ellen K. Ritchie, MD: The other information that was interesting about lenalidomide at ASH was SF3B1-mutated patients or ASXL1-mutated patients probably don’t respond so well to lenalidomide. But, it looks like DNMT3A-mutated patients may respond. So, it may depend on what the mix is.

Jamile Shammo, MD: And I totally understand.

Rami S. Komrokji, MD: If I may answer your question, if you don’t mind, there are a couple things: even in the Intergroup study, if you look at the responses, the 10% was all-comers in the intent-to-treat; like, if you look at the evaluated patients, it’s a little bit higher. Thomas had looked at the whole experience, where he combined data from several centers. In real life, outside the studies, he saw responses, I think, 24% to 25%. So, that’s what we see in real life: single-agent lenalidomide.

Jamile Shammo, MD: With non-deletion 5q.

Rami S. Komrokji, MD: Non-deletion 5q. In the study, there are 2 things: First, the criteria are not the IWG 2006; they are the older criteria, and they are modified to be even more strict. If you achieve transfusion independence, you have to have more than 1 gram, a gram-and-a-half. So, they were more than the usual criteria. It was a much more strict criteria than the usual. I’m sure we will see the follow-up later on, published in the data as ad hoc by the IWG 2006. It’s going to be higher. The other point is in the Intergroup study—there were patients that were treated by hypomethylating agents. That had not been the experience in either of the original MDS-002 and MDS-005 studies that looked at the non-deletion 5q, nor the HOVON study, because in Europe, you don’t have access to azacitadine in lower-risk. So, we have, and we’re presenting at this meeting, data on the sequence of those therapies in lower-risk MDS, and does it matter? And the rationale for that study was that most of the studies with lenalidomide, had no azacitidine exposure. We asked the question, “Does it matter?”, because the NCCN guidelines don’t put a preference. They just say lenalidomide and hypomethylators are an option for lower-risk.

It turned out in our experience, if patients were just anemic—again, if patients have thrombocytopenia—then, obviously we are going to go for hypomethylating agents. If we are treating hypochromic anemia particularly, basically sequence probably matters. In our experience, the response rate to lenalidomide as a first-line therapy is in the rate of 20% or so. If you do it as second-line after azacitadine, it was 10%. So, that explains part of what we’ve seen in the study. If you look at hypomethylating agents, when they were done first-line or second-line, there was no difference in the responses; they were in the range of 30%. When you look at survival and leukemia, it doesn’t seem that the sequence matters, but the responses do. So, in my mind, some strict criteria, and patients treated with hypomethylating agents on the Intergroup study, could explain the difference.

Thomas Prebet, MD, PhD: That would probably be one of the big challenges we’ll have to face in the next few years, to clearly have a strategy and some decision-making algorithm to know what kind of therapy for low-risk patients will go as a second-line treatment.

Vinod Pullarkat, MD: We should. Also, the point you brought up is important. We may identify certain genomic factors that predict response to lenalidomide. But, I just want to come back to this. For a clinician who is treating a patient with lenalidomide, what is the expectation? Do they expect improvement in hemoglobin? When do you call somebody a failure of lenalidomide?

Jamile Shammo, MD: If you’re talking about deletion 5q…

Vinod Pullarkat, MD: I’m talking about non-deletion 5q. Because, commonly patients are started on the drug, but then what is the expectation and when do you change?

Thomas Prebet, MD, PhD: I think we also have to face a little bit like what we have around chelation therapy: the issue of side effects and discontinuation. And we saw, from time to time, patients coming from the community who just have 1 month of lenalidomide for non-del 5q, and who probably don’t really favor the therapy. Based on the clinical trial, most of the patients who were on clinical trial were exposed for 6 months. That’s maybe kind of long, so I think most of the time we should consider 3 to 4 months before considering failure or not as a therapy, and thinking about maybe other options in that case. I think 3 to 4 months is good.

Jamile Shammo, MD: I think that is, yes. That’s exactly what the study showed. The response assessment was done at 6 months, but most of the people who responded did so by the end of the fourth month.

Rami S. Komrokji, MD: I think it tells us what you are looking at. Obviously, you are looking at hematological improvement. In deletion 5q, the responses are very high, and this is less. But, you are looking at patients becoming transfusion-independent, hemoglobin going up. Sometimes, some transfusion reduction could be meaningful for the patient. But, after 4 months, as Thomas mentioned, if it’s not there, then I don’t think there is a value to continue treatment. But, we are not looking at modifying the disease in the non-deletion 5q.

Vinod Pullarkat, MD: That has always been the challenge: how will you know if the survival is impacted? The only thing you can really look at is the transfusion.

Thomas Prebet, MD, PhD: On the non-del 5q population, I don’t see that yet as showing any improvement in survival in an untreated population. Response has been to have a benefit.

Vinod Pullarkat, MD: So, what you’re saying is you wait for a good 4 months, and if there is no improvement in hemoglobin, then we call that a failure.

Transcript Edited for Clarity