Itacitinib vs Placebo Combined With Corticosteroids for GVHD: Findings From the GRAVITAS-301 Trial

Yi-Bin Chen, MD, Dana-Farber Cancer Institute, Massachusetts General Hospital, Boston, MA, discusses results from the phase 3 GRAVITAS-301 trial, which evaluated itacitinib vs placebo, both in combination with corticosteroids, for the initial treatment of patients with acute graft-versus-host disease (GVHD).

The study found that the observed improvement in overall response rate at day 28 with the addition of itacitinib vs placebo did not reach the prespecified significance level.

Transcript:

Hi, my name is Yi-Bin Chen. I am the Director of the Hematopoietic Cell Transplant and Cell Therapy program at Massachusetts General Hospital, in Boston, Massachusetts.

Oncology Learning Network: In reference to the study we’re discussing today, what existing data led you and your co-investigators to conduct this research?

The study we're discussing today is Gravitas-301. Spelled out, the title is the efficacy and safety of Itacitnib versus placebo in combination with corticosteroids, for the initial treatment of acute graft-versus-host disease. This is a randomized multi-center, double blinded phase 3 trial. I think many of you know that research in graft-versus-host disease has been a huge priority for the transplant community. Acute graft-versus-host disease remains the leading cause of early death and morbidity and mortality, as I said. And these cases stick with you as a clinician, as they happen so early after transplant, patients end up back in the hospital, and watching patients go through that complication is really gut-wrenching. And so we've been trying to improve outcomes over the last couple of decades.

If you look, overall outcomes have improved, especially for severe disease. I think a lot of that is due to better transplant practices and improvements in medicine in general, and not specifically to treatment itself. Treatment, unfortunately, remains high dose systemic steroids as the initial therapy. And while that works for maybe 50% of patients, it doesn't work for the other 50% who need a second line therapy. And high dose steroids also tend to have their own toxicities and comorbidities that I think we all know well.

So, this research was the latest attempt to try and improve upon initial therapy for acute graft-versus-host disease. Now, JAK inhibitors have hit the world of medicine in a lot of different areas. And specifically in graft-versus-host disease, we've realized that JAK1 and JAK2 is sort of a bottleneck of signaling of pro-inflammatory cytokines that function in the pathophysiology of acute graft-versus-host disease. In 2019, Ruxolitinib, which is an oral JAK1 JAK2 inhibitor, was the first agent approved by the FDA for the treatment of steroid refractory acute graft-versus-host disease, supported by the landmark REACH-1 and REACH-2 trials.

Itacitnib is an oral JAK1 inhibitor; it's selective in a 6 to 1 ratio of JAK1 to JAK2. The theoretical benefit was that JAK1 appears to be where most of the signaling of pro-inflammatory cytokines go through. And JAK2 signaling tends to, theoretically at least, drive the side effects of cytopenias, most notably anemia and thrombocytopenia. So it was thought that if we used a selective JAK1 inhibitor, we could improve therapy, meaning improve efficacy of response in graft-versus-host disease, while better avoiding a cytopenia. So this was a big attempt.

 I think the company, Incyte, as well as the investigators, should be applauded for conducting a phase 3 trial, which is rare in the field of transplant, much less in graft-versus-host disease, to begin with. And this was an effort to improve upon the initial therapy of acute graft-versus-host disease, thinking that the addition of a JAK1 inhibitor might do so.

OLN: Please briefly describe your study and how it was conducted.

So, as I said before, Gravitas-301 was a randomized multicenter international double-blinded phase 3 trial. This trial was conducted at over 120 hospitals and community practices in 19 countries. Eligible patients were adults who developed newly diagnosed acute graft-versus-host disease. Acute graft-versus-host disease had to be grades 2 through 4, and thus requiring systemic therapy. Patients were stratified by clinical criteria by a system called the Minnesota Risk Score. And then they were randomized in a 1-to-1 ratio to receive steroids plus placebo, or steroids plus Itacitnib.

The primary endpoint here was a consensus-driven, agreed upon endpoint of day 28 overall response, comprising those patients who achieved the complete response, a very good partial response, or a partial response, 28 days after starting treatment. And this has been a traditional primary endpoint for a lot of graft-versus-host disease trials, and that's why it was chosen. It's been shown in large analyses to correlate with long-term outcomes, such as overall survival and non-relapse mortality. And those are the key secondary endpoints, would be non-relapse mortality and overall survival. About 440 patients were randomized in this study.

OLN. Please briefly describe the study findings. Were any outcomes surprising?

The results of the analysis showed that the patients were quite well-balanced. I mean, you'd expect this from a large phase 3 randomized study. In terms of the primary endpoint, the overall response at day 28 was 74% for patients who had received steroids plus Itacitnib, and it was 66% for patients who had received placebo. Now, this was not statistically significant that the p-value was 0.078, I believe, and thus fell short of the criteria of 0.05.

In looking at the other outcomes, more adverse events were seen in the patients who took Itacitnib by a small margin, but overall, Itacitnib was deemed fairly safe in this population. I think the expected cytopenias of thrombocytopenia and anemia were what we witnessed in both groups, including those who had received placebo. There did not seem to be an excess of serious opportunistic infections in patients who received Itacitnib, and the overall endpoints of non-relapsed mortality, overall survival, relapse of underlying malignancy, as well as cumulative steroid exposure, were all comparable between the group who received Itacitnib and those who received placebo.

In an unplanned, post-hoc analysis, it was noted that the proportion of patients who achieved a complete response at day 28 was 53% in patients in the experimental arm, versus only 40% in those who received steroids alone. And this was statistically significant, but obviously, an unplanned, post-hoc analysis.

Were any of these results surprising? I don't know. They were disappointing, I would say, in the sense that we didn't achieve our primary endpoint. There was a trend, but certainly, in such a large study, it did not meet statistical significance. And so, based on these results, we did not improve upon what is the standard of care for initial therapy in graft-versus-host disease.

The difference in the complete response rate is intriguing, because for those of us in the field, some of us feel that complete response at day 28 is a bit more meaningful than overall response, yet, if that was truly significant, I think we would expect to see differences in the other endpoints such as non-relapse mortality, or even overall survival, which we did not. So, [it took] a large effort of collaborators across the world, and, of course, patients, to put together this effort, and unfortunately, and disappointingly, we didn't hit our primary endpoint.

OLN: What do these results mean for the practicing clinician?

In terms of what should the practicing clinician draw from this, I think the practicing clinician's overall treatment paradigm is not going to change based on this study, because we didn't hit our primary endpoint, and we are not going to get regulatory approval for use of Itacitnib for the initial treatment of acute graft-versus-host disease. But I think what's informative from this is perhaps how we should conduct trials in acute graft-versus-host disease going forward.

So, if you look at perhaps why we didn't hit our primary endpoint, if you look at the patients that were enrolled on this trial, and the assumptions that were made, it's been known that in every transplant trial, and maybe every transplant in medicine, that if you use a historical control to then design a prospective study, that it's not surprising that your placebo arm will outperform what you had predicted, partly because medicine gets better [over] time, and so results are always better based on error compared to historical control. And also, when you do a clinical trial, there's a selection bias in patients who are stable enough and eligible enough to qualify through the screening criteria, and to get on trial.

So, if you look at the assumptions made for this trial, certainly there was a higher proportion of standard risk patients enrolled, and less high-risk patients, in terms of clinical risk and clinical severity of graft-versus-host disease. And thus the assumption made for the overall response rate of the placebo arm underestimated how the placebo arm actually did in this trial. And I think that's something for all of us in the community to learn moving forward.

I think another issue would be [that] perhaps it's naive to think [that] we can take a population that we can describe as grades 2 through 4 acute graft-versus-host disease, and treat them all the same. They're not one disease. There's a biological heterogeneity there, not just by organ distribution, but also just by overall biology and prognosis. And we've gotten a glimpse to this in the work done by colleagues looking at biomarkers of acute graft-versus-host disease, and the prediction of prognosis based on that.

I think moving forward, we should focus our efforts on risk-stratifying our population up front—understanding that very high-risk patients, biologically, likely still need steroids, but something else, something augmented on top of that, be it more immunosuppression, anti-inflammation, organ resiliency, or something that addresses those things—regardless, it would be steroids plus agent X. But yet, do other trials where the biologically low risk patients would perhaps not even need steroids, and be treated with agent X alone to avoid the toxicity of the steroids. And so I think moving forward along those lines is probably what we can learn from—not just this study, but the last couple of decades of our attempts to improve upon standard therapy.

OLN: What are next steps for research?

Itacitnib is still an agent that's being investigated for several other indications. We just recently helped to complete an upfront study using it without steroids for the treatment of low-risk graft-versus-host disease by itself. And hopefully we'll see those findings out in the next year. There's an ongoing phase 3 randomized trial studying steroids plus placebo, versus steroids plus Itacitnib for the initial treatment of chronic graft-versus-host disease, so we'll be eager to finish that trial and see the results. Itacitnib is also being used for other diseases such as aplastic anemia, or other diseases where JAK inhibition makes sense. And so, I think there's still a potential future for Itacitnib here, being that it certainly does have some level of biological activity. And so we look forward to seeing the results of these investigations.

OLN: Is there anything else pertaining to research/findings to add?

I would just encourage treating physicians to enroll patients in clinical trials. It's difficult, especially for transplant patients who have inherently been through a lot, and may be reluctant. But our treatments for graft-versus-host disease are not great. I think we've made progress in many areas, including prophylaxis and recently treatments for steroid refractory acute, and now three approvals for steroid refractory chronic disease too.

But the only way we're going to make progress, and the only way we have made progress, is these collaborative clinical trials that we can all work together on to improve the outcomes for our patients.

So, I'm hopeful we can continue this. Graft-versus-host disease has become an area that's of interest, and has generated a lot of research and resources. And I hope we can continue to build upon that.

Source:

Zeiser R, Socié G, Schroeder MA, et al. Efficacy and safety of itacitinib versus placebo in combination with corticosteroids for initial treatment of acute graft-versus-host disease (GRAVITAS-301): a randomised, multicentre, double-blind, phase 3 trial. Lancet Haematol. 2022;9(1):e14-e25. doi: 10.1016/S2352-3026(21)00367-7.