Gail Roboz, MD: The softening of the marrow concept, this is a particularly challenging one to figure out, but actually, I would argue that that challenge extends to you've got blasts of 7, you've got blasts of 11, you've got blasts of 14. We are still in the lousy data zone about figuring out what actually are we doing with hypomethylating agents before patients going on to myeloablative therapies. I am not talking about 75-year-olds now. I'm talking about younger patients who are intended to have myeloablative or nearly myeloablative therapies. The transplanters, again, depending on their mood and where you are, it needs to be two. It needs to be three. It needs to be five. I can generally give them whatever number they want without even looking under the microscope, but that's a different problem. I think that the goal is here, are we helping people do the data from the hutch? Do we have to give therapy of any sort, chemotherapy hypomethylating agents? If you have a patient who doesn't have 20% blast, but who has between 5% and 20% blasts and is going on to an Allo, does that patient get treated first or not. Amy of the soften the marrow group I imagine is going to say yes. How soft, what is the flame under that satay? How much are you softening it? What number has to happen or how many cycles have to happen before that patient gets transplanted?

Amy DeZern, M.D., M.H.S.: I will say, I think sometimes we fall into the paradigm of the enemy of better is best in the transplant world and we wait too long. It used to be six cycles. Then maybe it only needed to be four, but in somebody who I am giving it with the intention of lowering the blast, this may sound overly harsh to a patient's backside, but I try and marrow after every cycle. As long as the blasts are on the way down and are less than 5% we go. I do think, well, the caveat is we don't tend to myeloablate above age 40. That does matter. The choice of conditioning needs to play into these decisions with your transplant colleagues, but the better disease control you have going in the better outcomes there will be on the back end of transplant. I think that's pretty well accepted.

Gail Roboz, MD: Rami, you have a complex karyotype patient, no TP53, but complex karyotype patients going into a transplant. Is that the role of the hypomethylating agent to just keep whacking away at that marrow until those chromosomal abnormalities go away or the commas go away? Does that need to be normal before you go into a transplant?

Rami Komrokji, MD: I think this is one of the most controversial areas, as you even mentioned from the beginning is like the role of treatment prior to transplant. I think many of the times we've functioned by beliefs or what we think than really data. Even just about the blast reduction prior to transplant, people would say to data that it matters. Some other people will say to data it doesn't matter. Same for the cytogenetic response. There is studies from MD Anderson looking at decitabine and those patients that achieved cytogenetic response did better. We did a project with the French group, looking at the azacytidine, the cytogenetics response did not matter because in reality, most of the time we are looking at those at the level of the common myeloid progenitors, and you are not really seeing what's happening at the stem cell level. My bias is like what we know from transplant, if you go to transplant with less than 5% blasts and good cytogenetics, you do better, but we don't know if you started about 5% and had bad cytogenetics normalizing them, is it really better or not? We really don't know. My bias is to try to get that. I would feel maybe more comfortable if they are in cytogenetic response. I think a lot of this is really, unfortunately also in the past would have been lack of effective therapies in this disease that we could not prove those points. Like you do probably more AML than I do as well. You will not take a patient to transplant AML with 25% blasts or in ALL now. They want an MRD-negative disease to take patient to a transplant. I think we are lagging behind in the MDS with that. There is not solid data, but I do almost like what Amy would like. I would start hypomethylating agents. Try to get to the best I can get, but not necessarily deny the transplant if I didn't get to the best point. Again, the P53s are an exception a little bit, but that's my general approach.

Transcript edited for clarity.