Moshe Talpaz, MD: There are several risk models that have been developed over the years. IPSS [International Prostate Symptom Score], which was followed by DIPSS [Dynamic International Prognostic Scoring System], which was followed by a molecular-based prognostication. Generally the IPSS, which is a onetime deal in which DIPSS can be repeated, is based on clinical features.

Then another feature is age. For a long time I wondered, why is age a bad sign? It turned out it’s not just the age. With age there are more mutations. The high-risk mutations of ASXL1 and so forth go up with age. There is a reason that age is bad.

That said, the other element that goes into this prognostication includes symptoms. Then the level of anemia seems to be the most profound prognosticator. In DIPSS it counts for more. It counts for 2 points with the white blood cell count greater than 25,000 and the appearance of blasts.

All those constitute a fairly good and reproducible prognostic model. Later the cytogenetics were added in some of the models and, as stated, the molecular. It turns out that the ASXL1 mutation is the high-risk feature. It seems that we are capable of identifying patients with a relatively low-risk disease, where they can live 10-plus years plus, and patients with high-risk disease who live less than 2 years. Those are apparently very useful in treatment decisions. Perhaps the most important decision is when and if to go to transplant.

Harry Erba, MD, PhD: John, do you use this molecular testing for choosing patients to go to transplant?

John Mascarenhas, MD: I think it aids in the discussion. I still rely primarily in DIPSS. I think DIPSS is a valuable risk-stratification tool. If you’re intermediate or high-risk with a good performance status and no significant competing comorbidities, you really should be seen by a transplanter or evaluated, at least to have that discussion to figure if that’s going to be in the treatment paradigm. There are data that your risk score upticks if you have a molecular mutation, such as EZH2 or ASXL1. I definitely use them. There’s not a loss of risk-stratification tools these days. Increasingly we are incorporating these prognostic markers. The only thing I would say in addition is, the unfortunate part is that we use them; we get them. I don’t know that we totally understand how to interpret them, and they may help in prognostication. We have not gotten to the point where they truly inform us in therapeutic decision making, outside perhaps transplant.

Harry Erba, MD, PhD: For the audience, the high-risk features that I remember, they’ve been mentioned in different groups, ASXL1, EZH1 and EZH2, IDH1 and IDH2, SRSF2. Those are the ones.

John Mascarenhas, MD: And TP53.

Harry Erba, MD, PhD: Of course, yes. I’m going to put you on the spot, Mary Frances. You have a patient who has low-risk or intermediate 1—risk disease because they’re 66 years old but have a high-risk molecular profile. Is that someone that would go to transplant?

Mary Frances McMullin, MD: That’s the difficult question. We kill people with transplant, and that’s really the issue, particularly in my book, and I do transplants as well. Myelofibrosis transplants are not good news. These patients do badly, so how do you decide? I heard somebody saying recently, which is quite a good one, that if life expectancy is less than 5 years, then probably you want to go down the transplant route. You’ve still got the problem of, how bad are these things? What does it actually tell you? I’ve been following a patient with an ASXL1 mutation for 7 or 8 years, and he’s only now going to transplant, so that’s a bad risk from the beginning. It’s very difficult, and I think it’s a discussion not just for us but for the patient as well.

Harry Erba, MD, PhD: Absolutely. One of the challenges is when I look at the transplant literature, the best outcomes are from centers that transplant patients with earlier disease before they get sick.

Mary Frances McMullin, MD: Of course, they’re going to do well.

Moshe Talpaz, MD: Still, the results are fairly disappointing. Here you have a problem with 2 suboptimal therapies. And you say, what the heck do you choose between the 2? The transplant in myelofibrosis has not matured enough when you compare it with other diseases, let’s say even chronic myeloid leukemia, where the results are much more in favor of transplant. The statistics vary, but don’t be surprised to see figures of 50% success rate. That’s disappointing and even scary sometimes in some patients. There is an issue of graft rejection, graft failure. The fibrosis doesn’t go away, and even the relapse still creates a problem with transplant. I think the Wisconsin group has collected a large number of data and are about to publish a paper to argue the superiority of transplant. I would be cautious because of the patient selection, which stage of the disease, and so forth. You can never do really a controlled study. That’s not doable in the context of transplant, so the matching may be suboptimal and I would be cautious then.

Harry Erba, MD, PhD: Mary Frances, because you volunteered that you are a transplanter…

Mary Frances McMullin, MD: In a small way.

Harry Erba, MD, PhD: There are 2 very important practical issues. For the patient whom you do want to take to transplant but has massive splenomegaly, how do you manage that before the transplant?

Mary Frances McMullin, MD: Ruxolitinib to try to shrink the spleen. That of course is an issue in itself, which is where we’re going to come next, because they get much better when they’re on the ruxolitinib. Then you have the issue of, well, do I have to have the transplant? That’s common in practice. You can take that even further. The patient who turns up who’s not a candidate for transplant, but then once they’ve been treated with ruxolitinib for a while, they’re in a much better position. That’s really a difficulty.

Harry Erba, MD, PhD: Do you still do splenectomies if you have?

Mary Frances McMullin, MD: No, no, no.

Moshe Talpaz, MD: We do.

Mary Frances McMullin, MD: Do you?

Moshe Talpaz, MD: Yes. Do you keep the patient on ruxolitinib throughout the transplant?

Mary Frances McMullin, MD: No, until the day before.

Moshe Talpaz, MD: Because it’s been approved for GBA.

Mary Frances McMullin, MD: Yes.

Robyn Scherber, MD: I’ve seen some people actually we have done that where they’ve continued ruxolitinib throughout the transplant.

Harry Erba, MD, PhD: The second important question with transplant, and then we’ll move on to therapies. Transplanters I’ve spoken to feel very strongly about the preparative regimen that needs to be myeloablative. Do you think there’s any truth to that?

Mary Frances McMullin, MD: You use a more myeloablative regimen in the REC1 but not the full regimen, because you want your patient alive at the end.

Transcript Edited for Clarity