REACH2 Trial Limitations
John F. DiPersio, MD, PhD: There are lots of caveats to the study. As a purist designing clinical trials, one would say that some limitations in the study included the fact that it was an open-label study. Obviously, it’s very hard not to have an open-label study because you’re giving multiple therapies, including all sorts of different things to patients in the standard-of-care arm. So you know exactly who’s getting standard of care and who is getting ruxolitinib. That kind of open-label design can be undermined by investigator bias. That’s 1 major unavoidable limitation of the study.
The second limitation is that the study allowed for a crossover in the first 28 days, meaning patients who didn’t respond to standard-of-care therapy were able to cross over. That confabulates and results in significant difficulties in interpreting the results. If people who have bad GVHD—bad graft-vs-host disease—participate in a randomized study and are not responding to standard of care, there are real ethical issues about not providing them with the alternative as part of the study.
Sometimes people just come off the study and fail. That is a more clean design than actually allowing them to stay on study and cross over. Crossover designs are complicated. They require more patients. This was a fairly large study, around 300 patients, and around 150 to 155 were ultimately randomized to each group. So it was sufficiently powered. The primary and secondary end points, were met. There was a survival improvement. The exact reason why there was a survival improvement is not clear to me from the paper. The bottom line is that people who have acute GVHD have many problems—infections, their disease. They’re weak. All sorts of things can contribute to the death of a patient.
In the end, it doesn’t matter much. You just want to know who’s alive at a year and who is not alive at a year. Also, during that period of time, if you’re on therapy, how many are failing on therapy? That primary end point, an overall response of 28 days; secondary end point, an overall response of 56 days; and survival—for the patients who respond to therapy in both arms, failure over a year was much lower in the ruxolitinib group. All these things suggest that this is an effective therapy for established refractory and resistant acute GVHD.
It doesn’t provide any information on chronic GVHD. But these patients are continuing to be followed in this randomized study. One of the secondary end points over several years will be chronic GVHD. I suspect that that end point is coming up soon. That’s another analysis that will be a very important observation because, as you know, we have no good therapies. We have therapies, but they’re not particularly good for the treatment of chronic GVHD—1 approved drug, based on a very small open-label study, and that’s it. This is also an incredible unmet medical need. This trial, although it’s not given in a prophylactic setting, is given early and continued in patients with acute GVHD. In my humble opinion, it might have some impact on chronic GVHD rates as well. The best way to block both acute and chronic GVHD, however, using this drug—if patients’ blood counts can tolerate it—is by starting in the prophylactic setting early on after transplant, before these T cells reach the target organs.
Transcript edited for clarity.
