Drs Michael Bishop and John DiPersio comment on common challenges in measuring response in chronic GvHD and review clinical implication from the REACH3 trial.
John DiPersio, MD, PhD: One of the things about a lot of clinical trials previously was that patients had to fail a half a mg [milligram] or a mg/kg [kilogram] of steroids for a month. The problem is that a lot of these patients that have chronic GvHD [graft-versus-host disease] is they’re weak, and they have a lot of muscle loss and you put them on that much steroids for a month, they just fall apart. There were a lot of trials we did not participate in because we couldn’t bring ourselves to torture people with that much steroids. A lot of them, get a lot of symptoms from high-dose steroids. They can’t sleep, they’re miserable, and that limited our participation in some of these chronic GvHD studies. It also begs for alternative therapies for chronic GvHD that are not steroids.
Michael Bishop, MD: You know, you hit on something very important is you’re right, you had to be defined as steroid refractory. I would give the NIH [National Institutes of Health] Consensus Conference some things where they go, you don’t have to be on a mg/kg. It could be that it’s just an extensive amount and you’re just not going anywhere. But I agree with you, we know that we need interventions that will allow us to be steroid sparing. We just know that these are going to be, the complications that will occur are unlimited. From steroid psychoses to myopathies, as you described, it’s very important to, I agree with you that having agents that we can add on and minimize toxicity in these patients is key.
John DiPersio, MD, PhD: We have had a lot of exposure to many drugs over the years, too, in the context of small single-arm, open-label phase 2 studies in chronic GvHD which have led us in every direction but the right direction. I can tell you that there was a lot of interest and there were lots of believers and nonbelievers in things like Rituxan [rituximab], or regulatory T cells, and low dose IL-2 [interleukin-2], lots of things. I’m not saying that they don’t have merit, I’m just saying that none of these things were followed up with properly designed randomized trials to see if they benefit patients.
One of the problems is that we have a lot of difficulty, even though we have NIH consensus criteria for measuring chronic GvHD, whether it’s limited, extensive, or severe, and severe or not severe, it still is very complicated, multiorgan system staging which almost no one has time to do. And even if you have independent stagers at your center that do this on a regular basis, there’s enormous variability from visit to visit and from node to node. And it’s very hard to do this in a way that’s meaningful.
If you can’t measure the disease quantitatively in any way, then you can’t determine quantitatively if you’re making things better or not. And what you’re left with is your observation and the patient’s reported outcomes. One could argue that patient-reported outcomes are the most important. If someone has chronic GvHD and they have fascial involvement, or tendon involvement, or thickened skin and they can’t move and they can’t bend their joints, then you could give the patient a drug and if the patient clearly can move better and skin is clearly softer after 2 or 3 months of treatment, then you’d have some confidence that the drug is working. But, oftentimes, what happens is you walk into the room, and you know you’ve given the patient a new drug and you look at the patient and you ask the patient, “Well, how do you feel?” And the patient will say, “Well, I don’t know.” And then you look at the patient and you say, “Well, you look better,” and then the patient says, “Well, yeah, I might feel a little better.” And then you write down improved because it’s your bias. It’s also the patients responding to your cues. You want the drug to work, you walk in the room, and you say, “Ah, I think you’re a little bit better, what do you think?” “Ah, maybe I am a little bit better.”
There’s an enormous amount of physician and patient bias in measuring responses in chronic GvHD trials which, in my humble opinion, have limited the interpretation of the small trials that have been done, both with ibrutinib [Imbruvica] and with the ROCK [rho-kinase] inhibitor. I say that they both may be very effective and obviously they’re both approved now for the treatment of chronic GvHD. But, in a single-arm small open-label study where you have enormous observational biases, both from the patient and the physician, I have concerns about the real relative benefit of treatment.
Michael Bishop, MD: Before the NIH Consensus criteria, it was like, I know graft-versus-host disease when I see it and we had limited and chronic. This is trying to put some objective criteria because even in everything that we do, they’re subjective and I totally agree with you. We need better, and it’s because the disease is so complex, but at least we have some guidelines to do it. Just like the REACH3 study uses NIH criteria, and I think there is a difference. Before we had these, the comparisons, I mean you would go back to talk about the ECP [Extracorporeal Photopheresis] trials. I mean trying to interpret those is nearly impossible. We couldn’t even agree on how we grade skin. I know it’s imperfect, but at least it gives us some guidance and it’s evidenced by the number of large trials that we’re able to do now. I think that’s moved the field forward.
John DiPersio, MD, PhD: And it’s funny you should say ECP because I would say, I don’t want to sound too heretic here, but I would say having treated lots of patients with all of these drugs that if there’s one approach that seems to work and actually reduce the extent of pure chronic GvHD in some patients, that’s clearly, in my humble opinion, that’s clinically obvious it’s ECP.
Michael Bishop, MD: Yes. I have that same bias.
John DiPersio, MD, PhD: It doesn’t help that many patients. But when patients have severe chronic GvHD and they get on ECP, some of these patients just dramatically improve.
Michael Bishop, MD: I mean, coming back to the REACH3 study where investigators had a choice of about 10 different treatments that they could use, including ibrutinib, which is I always find it interesting now that it only got a minority and that was a large European components, they may not have had access to it. But the number one choice was ECP, and there were patients who did respond, right? The number one choice for alternative was ECP, yet the ruxolitinib [Jakafi] arm was markedly superior in almost every aspect.
John DiPersio, MD, PhD: Yes. And I also think that these medical therapies, the ROCK inhibitor, Jakafi as per the REACH3 study, and ibrutinib, the major take-home message and the newer Syndax trial with the CSF1 [colony stimulating factor 1] receptor antibody, this take-home message that I get from all of these studies is that the most important primary end point is time to progression or percentage of patients that progress. And that’s the thing that seems to be impressive in the REACH3 study, and that’s the best study that’s been done so far in chronic GvHD because it’s a randomized trial. Again, I go back to this issue of doing properly designed studies before you put all your money on your front runner here. Even though the ROCK inhibitor and ibrutinib were approved, I actually think that there’s much more solid evidence to support a JAK [janus kinase] inhibitor in this disease because of this well-designed randomized trial than anything else. Even though these other drugs may work just fine, and they could even work in the long-term better than Jakafi. I don’t know, but they haven’t stood the test of time.
Michael Bishop, MD: Yes. I don’t know if they have either. They’ll have to go through the randomized trials as well. With REACH3 is the durability of response and seeing the increase over time because again looking at this 24-week evaluation. But then, subsequently when you take it out to the 48 and beyond is seeing sustained and increasing degree of response. And that to me and the ability to get patients off steroids or a greater percentage of patients, at least having a dose reduction, those are my goals and for number one to be relatively well tolerated and convenient.
John DiPersio, MD, PhD: No. I think that’s a good point.
Transcript Edited for Clarity