During a Targeted Oncology Case-Base Roundtable event, Yi-Bin A. Chen, MD, discussed options for treating acute and chronic graft versus host disease.
CHEN: We believe that the primary curative mechanism of an allogeneic transplant is an immunologically based graft vs malignancy effect, also called graft vs leukemia or graft vs tumor. The issue, though, is that oftentimes this phenomenon is after allogeneic hematopoietic cell transplantation, and the [further] complication of GVHD has been common over the past few decades. Both acute and chronic GVHD have been a huge area of research for those of us in transplantation and [the disease is] a leading cause of nonrelapse mortality. In acute GVHD, the main organs affected are the patient’s skin, gastrointestinal [GI] tract, and liver; however, chronic GVHD is far more variable, and it can affect almost any organ and the syndrome itself can mimic classic autoimmune disease.1
The clinical risk factors traditionally for both acute and chronic GVHD have been quite similar. In fact, if you look at older studies, the occurrence of acute GVHD was thought to be the leading risk factor for chronic disease. That may not be true at this point, though I think large studies may now disagree with each other. With standard methods of prophylaxis, still, about 25% to 50% of patients receiving a conventionally matched transplant will develop some spectrum of acute GVHD that requires some treatment and systemically that remains high-dose steroids. Of those patients who are put on steroids, up to 50% or more will require a second agent because of either failure of a response or an inadequate response.
How GVHD Develops in Patients
Acute GVHD starts with damage or inflammation caused by the conditioning regimen, which is often myeloablative at that time, and that created a local environment that was ripe to generate an immunological response. Once donor cells were introduced, and allorecognition happened, donor T-cell activation occurred in secondary lymphoid tissue, and they acquired the trafficking phenotype to go back to those host tissues upon recirculation. Expansion then happened, and those T cells recirculated to the tissue and basically caused damage leading to the clinical manifestations of acute GVHD. This model has been popularized, but I believe we all believe it’s far more complicated at this point and often its specific arms have failed to improve the field.
Prophylaxis in GVHD
Every patient after an allogeneic transplant received some form of prevention or prophylaxis against GVHD, not just for acute but also for chronic. The international standard now for a matched transplant has been for the past couple of decades built upon the backbone of a calcineurin inhibitor, which includes either cyclosporin or tacrolimus.
The calcineurin inhibitors were classically paired with short courses of methotrexate after transplant, and recently methotrexate has been substituted for mycophenolate mofetil....Most recently, the newest form of pharmacological prophylaxis includes high doses of cyclophosphamide given on days 3 and 4 after transplant to eliminate alloreactive cells, and ongoing prospect trials are comparing posttransplant cyclophosphamide with our previous standard of calcineurin inhibitor– based prevention.
The other general category for prophylaxis has been called T-cell depletion. This is far less popular because it involves a laboratory expertise and resources, and this involves taking a donor graft and manipulating it either through positive selection of donor stem cells, or negative selection by eliminating a subset of T cells to try and prevent GVHD. Trials have tried to compare T-cell depletion with pharmacological methods and fail to show superior outcomes. In the future, this may change as more trials are done with more sophisticated methods of graft manipulation.
Risk Stratification Methods in GVHD
One of our main weaknesses for why we haven’t made progress in GVHD clinical trials is that we treat everybody the same, and when we do a clinical trial, we write it as grade 2 through 4—grade 2 being the most common. We are overwhelmed with low-risk patients, so any intervention we design, we fail to see because it is diluted by a lower-risk population. Now, there have been calls on the national level and international level to how we best risk-stratify patients, and if we’re able to risk-stratify patients, we can then do risk-stratify trials, and have a more homogeneous risk population, higher risk, lower risk, and do different things. There are 2 ways to do this. One is by a clinical score, which is the Minnesota clinical score that they published in about 1800 of their patients with acute GVHD treated over the years.2
[Researchers] separated them out into standard risk and high risk, and you can do this when you see your patient; if you can clinically grade their acute GVHD, you can separate them into standard vs high risk. The strength of this criterion is you can do it immediately, but the biggest weakness here is only 16% of patients are high risk and 84% of patients are standard risk. It isn’t really a great delineator into groups, and if you try and do a high-risk trial based on this, you’re going to be doing a trial for a long time, even when only 1 in 6 patients is high risk.
The other model for risk stratification is what we’ve mentioned, which is the biomarkers, that are based on ST2 and Reg3A.3 These are GI-mucosal biomarkers that hint at mucosal injury and track, we think, with GI GVHD in the right setting. So that is another way of how we’re going to biologically restratify patients through the mucosal GI GVHD, biomarkers, and ultimately do trials. I think the discussion on the national level has been if we’re going to make progress, we need to be able to identify high-risk patients and probably do high-risk trials with steroids plus drug X. Then for low-risk patients, if we’re accurately able to restratify, we either do lower-dose steroids or avoid steroids, as has been done in the recent Clinical Trials Network trial, or start steroids and taper really a lot faster.
Approaches to Risk Stratification
We have simple approaches; we generally give [1 mg/kg] for skin only. There’s visceral involvement with lower GI and we generally escalate to 2 mg/kg, though you do much less hepatic, and we generally question if this is upper GI infection, but that’s our own issue....We now use clinical criteria, and we reached out to find results based on those.
I think we all understand the downsides to steroids, too. So we’re all afraid to fire away and overtreat patients as well, but I think we can all admit that there are some patients with skin disease that you get to 1 mg/kg and they don’t get better or a week later, they have GI disease and you wonder how that happened.
There are also patients with gut disease that you’re sure they have the disease, you started [treatment], and a week later, they have normal bowel movements, and you wonder how that happened. It does really call into need our lack of tools that better understand the biology of the patients in front of us.
Therapy Options for Patients With GVHD
Refractory Response to Steroids
I think many clinicians can agree that if you put the patient on steroids, it may get worse after 3 to 5 days and then we would move on to something else. If we gave them a week, and they didn’t get better, I think we’d all agree to move on to something else. They’re steroid dependent, which involves an initial response, but then once you taper it, the disease flares. I think that’s got shifted to gray because I think tapering below 1 mg/kg is different from tapering below 10 mg a day of prednisone.4
Those are different biological populations, but they have often been mixed for the purposes of trials. Then there’s steroid intolerance, which isn’t truly steroid refractory, but that can lead you to go to another agent soon, be it because of things like diabetes, horrible osteoporosis, or manic breaks that may make steroid use unacceptable. [A transplant-oriented] National Comprehensive Cancer Netwrok committee happened in the past 2 years or so [that suggested several treatment options, but] the only one FDA approved is ruxolitinib.5
Clinical Benefit of Ruxolitinib
The primary end point [of the phase 2 REACH-1 trial (NCT02953678) was] the day 28 overall response rate [ORR] as has been agreed upon by consensus criteria. There was a 55% ORR and about 73% best ORR at any time. The complete response was 27% at day 28, about half complete responses and half partial responses.6
The median duration of response was about a year, which is pretty good, and the time to response was about a week, which tells you this course [of treatment was] quickly released. I think, disappointingly, the nonrelapse mortality rate at 6 months was still high at 44%, but those who obviously responded did a lot better than those who didn’t, as you might expect. REACH1 was the trial that got ruxolitinib FDA approved, and REACH2 [NCT02913261] was the validating trial, and it was a phase 3 trial. We rarely see these in this disease and REACH2 was an international trial that randomized patients with refractory acute GVHD to roughly 10 mg twice a day vs best available therapy [BAT]. The investigator had to specify what they would use as BAT, but I think many people can criticize the BAT as a control. All of us know, there’s no national consensus at that time on the second-line therapy, but this was the best trial that could be done. [The trial compared ruxolitinib with BAT and there was a] big benefit for ruxolitinib of 62% vs 39% ORR and these ORRs did mirror the REACH-1 study.7
The durable ORR at 8 weeks was 40% vs 22%. [Without BAT as the] control, I don’t know how else you do a phase 3 trial in this population, and these results can only be viewed as positive progress in the field. Failure-free survival was a key secondary end point from [this study] and was a composite end point that showed a big benefit for ruxolitinib vs control. In terms of adverse events [AEs], there were more cases of thrombocytopenia, and when specifically inhibiting JAK, 2 [cytomegalovirus] infections were about the same, but there weren’t any huge AE signals otherwise.8
REACH3 Trial (NCT03112603) Outcomes
Patients were randomized between ruxolitinib at 10 mg twice a day vs BAT, with ibrutinib [Imbruvica] added as an option because it became FDA approved. I think you can criticize the BAT control because it’s so heterogeneous, but I think you can see that there is no standard, there’s no consensus [for second-line treatment]. So the only way to do this trial would be in this way, and just like the REACH2 trial, I think this was a massive undertaking and really what needed to be done to move the field forward.9
The primary end point here was 6 months’ ORR based on the National Institutes of Health criteria with the key secondary end points of failure-free survival, as well as symptom score. The ORR with ruxolitinib was about 50% compared with BAT at about 25%, so a 2-fold increase in ORR. The trial certainly wasn’t blinded, this is an open-label trial, so investigators and patients knew what they were on, and at 6 months, patients on the BAT arm were allowed to cross over if they had not received ruxolitinib.
Thinking About A Multidisciplinary Approach to GVHD
As a multidisciplinary approach, I think it’s useful to point out all the complications of steroid use. I also think a big push in moving treatment forward is just like in acute GVHD, where we’re trying to stratify patients. I think in chronic disease treatment, we’re trying to do the same thing. I think treating all chronic [GVHD] as 1 patient population is probably going to lead to nonsuccess here, as we’ve had over the past couple of decades.
There has to be a subset of chronic [GVHD] that we have to be very aggressive with and hit harder with steroids and maybe some other drugs up front, and then there have to be some sort of chronic treatment group that you can possibly avoid steroids in and use alternative therapy once we have more access, with ibrutinib and hopefully ruxolitinib.
1. Hill GR, Ferrara JL. The primacy of the gastrointestinal tract as a target organ of acute graft-vs-host disease: rationale for the use of cytokine shields in allogeneic bone marrow transplantation. Blood. 2000;95(9):2754-2759.
2. MacMillan ML, Robin M, Harris AC, et al. A refined risk score for acute graft-vs-host disease that predicts response to initial therapy, survival, and transplant-related mortality. Biol Blood Marrow Transplant. 2015;21(4):761-767. doi:10.1016/j.bbmt.2015.01.001
3. Choi SW, Reddy P. Current and emerging strategies for the prevention of graft-versus-host disease. Nat Rev Clin Oncol. 2014;11(9):536-547. doi:10.1038/ nrclinonc.2014.102
4. Schoemans HM, Lee SJ, Ferrara JL, et al. EBMT-NIH-CIBMTR Task Force position statement on standardized terminology & guidance for graft-versus-host disease assessment. Bone Marrow Transplant. 2018;53(11):1401-1415. doi:10.1038/ s41409-018-0204-7
5. NCCN. Clinical Practice Guidelines in Oncology. Hepatobiliary cancers, version 2.2021. Accessed September 2, 2021. https://jnccn.org/view/journals/ jnccn/19/5/article-p541.xml
6. Jagasia M, Zeiser R, Arbushites M, Delaite P, Gadbaw B, Bubnoff NV. Ruxolitinib for the treatment of patients with steroid-refractory GVHD: an introduction to the REACH trials. Immunotherapy. 2018;10(5):391-402. doi:10.2217/imt-2017-0156
7. Jagasia M, Perales MA, Schroeder MA, et al. Ruxolitinib for the treatment of steroid-refractory acute GVHD (REACH1): a multicenter, open-label phase 2 trial. Blood. 2020;135(20):1739-1749. doi:10.1182/blood.2020004823
8. Zeiser R, von Bubnoff N, Butler J, et al. Ruxolitinib for glucocorticoid-refractory acute graft-versus-host disease. N Engl J Med. 2020;382(19):1800-1810. doi:10.1056/NEJMoa1917635
9. Zeiser R, Polverelli N, Ram R, et al. Ruxolitinib vs best available therapy in patients with steroid-refractory/steroid-dependent chronic graft-vs-host disease: primary findings from the phase 3, randomized REACH3 study. Presented at: 62nd ASH Annual Meeting and Exposition; December 5-8, 2020; virtual. Abstract 77. Accessed September 2, 2021. https://bit.ly/3kPNthX