However, the major randomized clinical trials with momelotinib didn’t quite meet their primary efficacy endpoint. There has been a lot of discussion on why that was. Whether it is something with the compound itself, or perhaps [there was something] in the clinical trial design that could have been optimized to show more efficacy for the medication. We are going back to the drawing board for that one and see if it’s worth continuing to further develop this medication for our patients.
Plenty of combination and triplet regimens have been studied in other hematologic cancers. Will we see more of this in MPNs?
We have seen a number of combination studies done over the last several years, of ruxolitinib plus drug “X,” “Y,” or “Z.” One of the things that we are seeing often with these combination studies is that the drugs together can be pretty toxic and lead to AEs. In myelofibrosis in particular, this is a pretty frail population. They can succumb to AEs pretty readily, so we always need to be cautious when combining therapies; escalate slowly and try to pick agents that don’t have overlapping toxicity profiles to optimize that.
However, a number of studies were done of combinations of ruxolitinib and hypomethylating agents, ruxolitinib and lenalidomide (Revlimid), and ruxolitinib with interferons. For example, there are ongoing studies looking at ruxolitinib plus a PI3K-delta kinase inhibitor and many other compounds that we will hear about more at the various hematologic focused meetings coming up in the future.
Regarding the 2018 ASH Annual Meeting, what studies are you hoping to hear more about?
Some of the big headlines will be updates from PROUD-PV, which is a randomized study between a newer, longer-acting version of interferon and hydroxyurea in PV. We expect to get updates from the pivotal MPD-RC 112 study looking at frontline treatment of PV and ET with either interferon or hydroxyurea. We also expect to get updates from imetelstat, a telomerase inhibitor, as well as PRM-151, an antifibrotic agent. People have been keyed in on some of those drugs because they have new mechanisms of action and something different from what we already have in our armamentarium.
Are any of these newer agents targeting molecular abnormalities other than JAK1/2?
Thus far, many of the mutations that we have identified at this point are important for prognosis. There is a small population with mutations that are actionable. For example, there is a small population of patients with myelofibrosis with mutations in the gene IDH2
. There is an approved IDH2 inhibitor that is approved for acute myeloid leukemia.
The hope is that we can start to subclassify myelofibrosis, and the MPNs in general, into more mutation-specific subclasses where we can apply these targeted therapies. The studies have yet to be done. The major challenge is that this represents a small part of the population. If you are looking at these specific mutations, it may only be 5% of all patients with myelofibrosis. Therefore, you are already taking a very small pie because it is a rare disease, and you’re making it substantially smaller. Therefore, these studies are hard to do—to understand that there is efficacy and safety in using these medications, particularly in combination for these populations.
What does the future of MPN treatment look like 5 years from now?
The landscape in 5 years is going to look like a couple of different things. Ruxolitinib will still be a mainstay in treatment; it is a broadly applicable drug with a great track record and it will stick around. We will start to incorporate other treatment strategies based on patients’ disease risk and symptoms. There will be medications to alleviate anemia in the near future; there are certainly a lot of medications looking at that. In patients where anemia is the major issue, we may address that directly with a second medication.
Also, there is a push for antifibrotic agents—anything that can reduce fibrosis and reduce allele burden. We are going to start looking at medications that can do that readily, whether by themselves or in combination with a JAK inhibitor.
Regarding the available MPN risk models, does there need to be refinement?
Risk models are very imperfect. Often, when we look at the stats to predict how good these models perform in populations, their performance is almost closer to a coin toss than it is to perfection. There is a long way to go. We know that certain mutations have bad prognostic implications, but we often don’t understand how the different mutations interact with each other and how mutations interact with other prognostic factors within the model. While we are making great strides to better predict who has aggressive disease and who does not, there is still a long way to go.