Srdan Verstovsek, MD: The other drugs listed in the NCCN [National Comprehensive Cancer Network] guidelines for lower risk are interferon and hydroxyurea, but I think they are really for those patients who have a high white blood cell count, high platelets, are at an earlier stage, have prefibrotic myelofibrosis, or those low-risk patients who are transforming from ET and PV for whom you need to control myeloproliferation. Inflammation is not such a big problem. They do not have symptoms, so therefore Hydrea, and interferon are just fine. That is giving us the opportunity to individualize and assess the patients properly and choose what you want to do.

Now, for the high-risk patients, we have, in addition to ruxolitinib, fedratinib. Fedratinib was approved 2 years ago in the United States based on some old studies. JAKARTA was a randomized phase 3 study comparing a placebo and fedratinib. On the face value of the results, not really having a study that compared head-to-head ruxolitinib and fedratinib, it seems to me that the control of the spleen and symptoms is about the same. What is your take on that?

Andrew Kuykendall, MD:Yes, it seems so. I think it is really difficult. We are always counseled on not comparing across trials, but we have to because these were never compared head-to-head. The trials are also designed relatively similarly. I think fedratinib, with the JAKARTA trial, was given at 2 different dose levels of fedratinib versus a placebo, looking again at spleen response at around 24 weeks. The JAKARTA study also required a confirmatory spleen response 4 weeks later, but it was very similar. The percentage of patients who were able to achieve spleen responses or symptom responses were quite similar. I think, from a therapeutic benefit standpoint, ruxolitinib and fedratinib seem to be both able to control symptoms and spleen size.

I think the utilization standpoint comes down to a comfort with ruxolitinib, availability of long-term data, and the toxicity profile, which is very different between the 2 agents. That is the challenge when we are looking at patients and deciding how to treat based on a frontline study. Fedratinib gained approval in the frontline or second-line setting. The approval is agnostic to line of therapy, but when we are evaluating patients between using ruxolitinib or fedratinib in the front line, you are dealing with one agent that you have long-term data for, long-term follow-up, and survival improvement. Then, you have another where you do not have as much long-term data and it has a toxicity profile that has more GI [gastrointestinal] [adverse] effects.

Srdan Verstovsek, MD: I think the toxicity and [adverse] effect part is one that bothers me a little bit here—not that we cannot manage; we do. Nobody really stops fedratinib because of the nonhematological toxicities. By the way, the degree of hematological toxicities, anemia and thrombocytopenia, also appears to be similar to ruxolitinib if you look at these numbers. The GI toxicity is low grade in two-thirds of the patients, and it does not really cause a problem long term in terms of patients stopping it. That is not very common because of GI toxicity, but it requires education of the patients. Tell the patients, “Hey, this is highly likely; you need to take some antinausea and antidiarrheal medications.” I actually give my patients a prescription.

Then, this black box warning shows the possibility—a distant possibility, to be fair—of a central nervous system toxicity, which appears to be related to fedratinib inhibiting the uptake of thiamine, vitamin B1, from the GI tract into the body of the person. If that is deficient, then people develop central nervous system toxicity. What this means is that you need to measure the thiamine before you give fedratinib, supplement it, and then periodically measure it.What I do is I give patients thiamine to take—why not—instead of continuously measuring and worrying if it is going to cause some troubles or not. This management style is a little different, and it is difficult to find convincing evidence to use it in the frontline setting much. Perhaps it can be used in an individualized approach, in patients with low platelets for example, but not much beyond that. It is good to have the drug, but perhaps we should be using it in a second-line setting later.

Andrew Kuykendall, MD:I think it has been a challenge to try to figure out what its role is, and if there is a cohort of patients in the frontline setting for whom it may be better, but we really do not know the answer to those questions, even in the thrombocytopenic setting. Obviously, it included patients with platelet counts greater than 50,000, as opposed to the COMFORT trials, which required 100,000 or more. There are some data; I think about 12% of patients in the JAKARTA trial had platelet counts between 50,000 and 100,000 and showed benefit there. Then, with the results of the EXPAND study, which was looking at this patient population with ruxolitinib, we have data with ruxolitinib for patients with lower platelet counts as well.

Transcript edited for clarity.