MPN updates from iwMDS-iwMPN 2024 | Considering the use of symptom response as an endpoint in myelofibrosis clinical trials

We’ve learned a lot around symptoms for patients with MPNs, both in clinical care but also as endpoints in clinical trials. So in myelofibrosis, first we’ve learned: one, that many patients have difficult baseline symptoms, but not all. So, as it relates to a clinical trial endpoint, I strongly think we should have a threshold for patients to be evaluable. Not to necessarily limit clinical trials to a subset that are only symptomatic, but if you’re symptomatic, then you’re evaluable. Because we’ve learned that if you’re not symptomatic, it’s still worth tracking because we again want to see: is there any toxicity from a therapy that then makes you symptomatic or really has that negative component? So that’s first, are folks evaluable? 

Two, symptoms are always part of a much more comprehensive and holistic assessment of response. That can be spleen, that can be symptoms, that can be other factors such as progression-free survival, overall survival, or a variety of different candidate biological endpoints at this point – variant allele frequency decreases, fibrosis, etc. 

Third, as we look at symptoms, we realize it’s important that patients act as their own control in terms of you’re looking at differences from baseline, and that the goal isn’t necessarily for the patients to have a score of zero. So, for example, in each of the symptoms, there has been work done before – there was a study done by Leslie Anderson, in which I and many participated, in the UK, where they looked at age-matched controls. And indeed, with each of the symptoms, there is a little bit of a floor effect out there in the general population, with fatigue having perhaps the greatest kind of elevation as a baseline. Now, it’s clear from all of the studies done in myelofibrosis that there is a clear increase in fatigue over what that floor is. Let’s say that that was an average of three out of ten for the general population, and in myelofibrosis studies it’s more like five, six, seven. So there is possibly a ceiling effect in terms of response in fatigue, because again, expecting that to get down to zero may be unrealistic because any number of things: people are older, they don’t sleep well, they’re busy with work. So again, it is good to have realistic expectations of what improvement looks like. 

So symptoms have been very reproducible in terms of their conduct as an endpoint in clinical trials. The Phase III trials and Phase II trials we’ve done have been very consistent. Responses have been relatively consistent. They behave very similarly across different populations as well as different translated values. But I think it’s important that they really be part of a holistic approach. People have said, boy, it’s a time for different endpoints in MPNs and in part, I think, trying to see deeper responses, particularly as we look at combination therapies, two drug therapies, three drug therapies, I think that’s a good expectation. 

However, I think another factor that’s not discussed enough is, in addition to kind of expanding the endpoints that we have, because spleen and symptoms, I think are still very relevant, we may need to really conduct longer clinical trials. The concept of trying to fully assess the benefit of a disease modification within 24 weeks is probably unrealistic. You know, the 24-week number really came from a realization that the rapidity of response to JAK inhibition in JAK inhibitor-naive patients was very quick. Many patients have their benefit within two weeks, or three weeks, so 24 weeks was more than generous for really fully assessing that benefit. But improvements in things like survival, anemia, fibrosis, variant allele frequency really are going to need longer clinical trials and also will need to be sensitive to what are the the needs and realistic asks of patients who enroll in trials who end up on control arms.