Munshi Explains Staging, Prognosis, and Treatment for a Patient With Acute Graft-vs-Host-Disease

During a virtual Targeted Oncology Case-Based event, Pashna N. Munshi, MD, associate clinical director, Stem Cell Transplant and Cellular Immunotherapy Program, assistant professor of Medicine, Georgetown University School of Medicine at MedStar Georgetown University Hospital, discussed the case of a 48-year-old male patient with acute graft-versus-host-disease (GVHD).

Targeted OncologyTM: What factors contribute to the risk of acute GVHD in a patient like this one?

MUNSHI: A lot of donor-recipient factors and other conditions increase the risk of acute GVHD. [These include] gender matching, human leukocyte antigen disparity, degree of mismatch, and having an older donor. There’s also [blood group] incompatibility and definitely CMV mismatched status. Though now that the FDA has approved letermovir [Prevymis] for patients who are undergoing allogeneic transplant if they have a CMV-positive donor,¹ we’re seeing very little CMV reactivation. That has been a bit of a game changer for the good.

Patients have an increased risk of GVHD if they receive a transplant from a peripheral blood stem cell source versus a bone marrow graft, because the peripheral blood has more T cells in its composition. The myeloablative regimens [are associated with greater risk of GVHD than] reduced-intensity regimens.²

Do you agree with these poll results? Would you start with systemic therapy for this patient?

It can get a little tricky whether you want to give patients systemic steroids or wait and see if something gentler might work. I tend to agree that, at this point, the patient needs to immediately start with systemic steroids, because there are 2 organ systems involved. Once the lower gastrointestinal [GI] tract gets involved, it surely portends a poor prognosis if the grade becomes worse. And they become refractory to steroids very quickly: 50% of these patients will eventually not respond to steroids.

How would you stage this patient’s GVHD?

There are many criteria for staging GVHD. The criteria that most clinical trials use are the Mount Sinai Acute GVHD International Consortium [MAGIC] criteria.3 They are adapted from the Glucksberg criteria, which are very similar.⁴

Three organ systems [are involved in] acute GVHD: the skin, the liver, and the GI tract. Skin involvement is graded on the basis of the body surface area involved. Liver involvement is graded on the basis of the total bilirubin level. Upper GI involvement is graded on the basis of anorexia, nausea, and vomiting, and it just comes in stage 0 or stage I, depending on if it’s persistent or not. To determine lower GI tract involvement, we measure stool volume, especially when patients are admitted to the hospital. But once they go home, we can’t do that, so we ask them how many times a day they have diarrhea. Is it watery? Is it muddy? What’s the volume? Is it large or small?

The patient can characterize the stool and tell their doctor how many times per day: 4 times, 5 times, 6 times. This patient is having 4 episodes per day; that puts them in stage I lower GI GVHD. But with a 60% body rash, that puts them in stage III skin GVHD. So really getting up there with skin, but not so much yet for GI. Once each organ [involvement is] staged, there’s an aggregate score based on the combination of these organs. Then we come up with the grade.

In this patient, with a stage III rash, stage I upper GI, and stage I lower GI GVHD, they have a total score of a grade 2 acute GVHD. This is still in the mild to moderate zone. Anything above grade 2 is considered very severe GVHD.

Would you recommend that this patient receive systemic steroids?

In the scheme of things, somebody who didn’t have symptoms and now is having active symptoms, especially with lower GI tract involvement, definitely needs high-dose steroids to get in there and [stop] the inflammation.

On what would you base a prognosis for this patient?

We can risk stratify these patients on the basis of the stage of organ involvement.5 Broadly, they can be at a standard risk or at a high risk [of poor response to treatment, mortality, and transplant-related mortality]. The patient is at high risk once they have very active GI involvement [or] if they have 2 organs involved. This is one more reason to think about starting these patients early on steroids. Why is this important? Because once a patient has high-risk GVHD, the chance of response to steroids is even lower, and once they don’t respond to steroids, there is a higher [risk of] transplant-related mortality. The probability of transplant-related mortality is 44% for patients with high-risk acute GVHD flares, [versus] 22% for patients with low-risk GVHD [P < .001]. These are a few things to think about. Act very swiftly if a patient has 2-organ involvement, especially the lower GI tract.

Can biomarkers guide treatment decisions in this case?

In the field of GVHD, biomarkers are a very exciting advancement. We want a prognostic model of which patients will get GVHD. Can biomarkers in the blood [help] prevent GVHD and improve transplant outcomes?

A large prospective trial was done through the Bone and Marrow Transplant Clinical Trials Network where a set of 6 biomarkers were tested at several time points after the transplant.6 They saw that they could predict when GVHD happened by using these biomarkers. They could see that as the levels of these biomarkers increased, the patients had higher scores of GVHD. Once treatment was started, if specific biomarkers went down it was predictive of response at day 28 [56% vs 17%; odds ratio, 6.32; P = .001] and also predictive of [decreased] transplant-related mortality by day [180 (49% vs 87%; P < .0001)]. If all these biomarkers went up aggressively, overall survival was lower [P < .0001].

The MAGIC Consortium also tried to test biomarkers.⁷ They looked at 2 biomarkers, REG3A—the regenerating islet-derived 3-alpha, which is specific for the GI tract— and ST2. Looking at these 2 biomarkers, they came up with an algorithm of prediction. On the basis of how these biomarkers responded at the time of GVHD and to treatment, they could predict mortality by 6 months. In clinical practice, it is difficult to use this day in and day out. We still use our clinical skills to assess the degree of GVHD. But all patients eventually get treated the same way—with high-dose steroids—despite biomarkers being elevated or not.

At this point, [biomarker data] may tell us an association rather than a causality. We’re not openly using biomarkers to guide our practice, but I think we’re learning to use them a bit more and knowing that there’s something out there that could be used as a predictive tool. It is an exciting development.

Are there alternatives to systemic steroids?

Steroids remain the mainstay. We need to see if we can move to other therapies that are coming down the pipeline.

Data from the REACH1 [NCT02953678] and REACH2 [NCT02913261] trials led to ruxolitinib [Jakafi] approval.^8,9 If we can use ruxolitinib in an up-front setting, [maybe we] can use the newly approved rho-kinase or ROCK2 inhibitors as well.10 We want to think about steroid-sparing agents. Maybe biomarkers can guide us in the future for that. But right now, in terms of, “Do I start my patient on treatment?” or “Will they respond to this treatment,” I find that [biomarkers are] still not a very useful tool because at the end of the day, the patients all still need to be started on steroids.

The minute you see that your patient is not responding to steroids, very quickly start them on a JAK2 inhibitor.

How do you dose steroids?

This patient received 2 mg/kg of prednisone per day for 14 days. Two mg/kg is a very high dose. The standard is 1 to 2 mg/kg.¹¹ There are data to show that 2 mg isn’t any different from 1 mg.¹² But a lot of times, if it’s a very active, severe flare, we will use 2 mg/kg. I’m not sure if I would have done 2 mg/kg in this case, but it’s certainly not out of the realm of treating these patients.

The goals of primary therapy for acute GVHD are to stabilize the organ manifestations, or improve them, and limit long-term treatment toxicity. We want to improve functional capacity and prevent any reduction in quality of life. First-line therapy is always with corticosteroids. Now ruxolitinib is approved for second-line therapy.8 There have been data to show that it can improve overall survival.

How do you taper glucocorticosteroids after achieving initial response?

If the patient is taking 2 mg/kg of steroids, an average 70-kg person, that’s over 100 mg of steroids. After 2 weeks, they probably are not getting up from a seated position anymore with all the muscle wasting that can happen.

[As soon as they start to show improvement, it would be safe to start to taper the dose.] Traditionally, [the patient receives the full dose for] at least a week or 10 days. Then it is traditional to decrease the dose 10% every 5 to 7 days, gently coming down, making sure that the patient is not having any flares.

Describe the multidisciplinary team–based approach that you use for acute GVHD.

The incidence of acute GVHD in the patient population is anywhere from 30% to 50%, despite the best [efforts at] prophylaxis. Most patients will get some form of acute GVHD—it can go up to even 80%. This [necessitates] a multidisciplinary team approach. If the patient is having diarrhea, they’re having malnourishment. There’s nausea or anorexia, so they’re not eating on top of that. Then there’s skin rash, so the risk of infections and cellulitis. They’re in pain. A dermatologist probably should be involved at some point. A nutrition team is also needed. If they’re on high-dose steroids, physical therapy should be involved up front. So early involvement of a whole team is very important. That’s usually how I treat my patients and usually how centers of excellence continue to treat active patients with GVHD after transplantation.

How do you determine if a patient’s GVHD is steroid refractory?

The strict definition of steroid refractoriness or resistance is if there’s progression of acute GVHD within 3 to 5 days of starting high-dose steroids, or there’s failure to improve within 1 week of starting these steroids, or there’s incomplete response after more than 28 days of any immunosuppressive treatment.13 So, by and large, in 3 days or a maximum of 7 days, [it will be clear] if the patient’s GVHD is going to be steroid refractory or not.

Steroid dependence is [defined as when] the patient’s GVHD initially responded to steroids, but the disease flares when the dose is tapered, so they cannot be taken off the steroids.

Steroid intolerance is when the patient develops [unacceptable toxicity from steroids such as] uncontrolled diabetes or myopathies. Then it becomes hard to keep them on steroids.

What are the treatment options for patients with steroid-refractory GVHD?

Ruxolitinib now has been FDA approved for steroid-refractory acute GVHD, and it’s a category 1 definition.^8,11 Ibrutinib [Imbruvica] has also been approved—it’s only FDA-approved indication is for chronic GVHD.¹⁴ There are many other treatment options [in the National Comprehensive Cancer Network guidelines].¹¹ Oncologists always end up using some combination or other depending on which of these different immune suppression medications they are comfortable using.

What new treatments are in the pipeline?

In terms of BTK inhibitors, I don’t think there’s anything other than ibrutinib at this time point. There are many JAK inhibitors being studied.¹⁵ Baricitinib is another JAK inhibitor that’s actively being studied for chronic GVHD, as well as for pulmonary GVHD.¹⁶ Then there are other rho-kinase inhibitors, called ROCK2 inhibitors. This is really making waves. We’re very excited about this drug because the response rates are very high, about 70%.10 It’s a smaller study, but clearly it has antifibrotic pathways. So I think that’s going to be used much more in the up-front setting.

Then there’s also alpha-1 antitrypsin, which targets the liver and macrophages and has very promising results from trials done at Dana-Farber Cancer Institute and Michigan.¹⁷ So I think we’re going to see very different characteristics of how to approach GVHD.

What data support the use of ruxolitinib in this setting?

The REACH1 study led to the approval of ruxolitinib for steroid-refractory acute GVHD.^9,18 In this phase 2 trial, patients with steroid-refractory acute GVHD got ruxolitinib (5 mg twice a day) with or without a calcineurin inhibitor. They were allowed to remain on steroids. The primary end point of this trial was overall response rate [ORR] at day 28. They also looked at response rates at day 56 and day 100, biomarkers, failure-free survival, and durability of these responses. The ORR at day 28 was very high: 54.9%.¹⁸ The best ORR, which was at any given time during the treatment, which was as high as 73.2%. The median time to response was 7 days. So this was very quick. The median duration of response was 345 days, with more than 6 months follow-up. Nonrelapse mortality at 6 months was 44.4%. There were deaths from infections, etc, but not related directly to ruxolitinib.

Subsequently there was a phase 3 trial, REACH2.¹⁹ They looked at higher doses of ruxolitinib in steroid-refractory acute GVHD. They started off with 10 mg [of ruxolitinib] twice a day. This study had a similar primary end point of ORR at day 28. This was compared with best available therapy. This was done in Europe, so [the comparison was to the] best available therapy used in Europe, like anti-thymocyte globulin, sirolimus [Rapamune], etanercept [Enbrel], photopheresis, or other therapies; all things that we would use in the United States as well. They looked at similar key secondary end points, [including] duration of response at day 56.

The ORR for ruxolitinib was 62% at day 28, compared with the best available therapy arm, which was 39% [odds ratio, 2.64; 95% CI, 1.65 to 4.22; P < .001].19 Durable overall response at day 56 [was higher in the ruxolitinib group than it was in the control group (40% vs 22%, odds ratio, 2.38; 95% CI, 1.43-3.94; P < .001)].¹⁹

The lower grade acute GVHD, which was grade 2, had the highest complete response rate with ruxolitinib: 50.9% compared with just 26.4% with best available therapy.19 This is quite remarkable to have a complete response in GVHD so quickly. When you get to higher grades of GVHD, the complete response rate for ruxolitinib is not as impressive; it’s less than 30%. But it’s still much higher than the [response rates of] other therapies we would have otherwise treated these patients with in steroid-refractory disease. The key point is to diagnose steroid refractoriness early. Then get ruxolitinib in there to break the cycle and break the progression of organ grade to something higher.

The loss of response wasn’t statistically significant. The estimated cumulative incidents for the loss of response at 6 months was 10% in ruxolitinib compared with 39% in the control arm.¹⁹ So patients continued to maintain responses, which, again, is what we want to see. We don’t want to see flares if they come off steroids.

[Of the 4 organ systems involved in GVHD], the skin responses were the best with ruxolitinib. Lower GI and liver GVHD did have good responses, but the responses were not as remarkable. Ruxolitinib is an ideal drug in this setting, on the basis of the organ responses.

A secondary end point was failure-free survival, basically indicating a time point from randomization to either non–relapse-related death or any new GVHD. This was not statistically significant because it was not designed to compare ruxolitinib survival outcomes with control therapy. But there were 5.0 months median failure-free survival with ruxolitinib compared with 1.0 month with control [hazard ratio for relapse or progression of hematologic disease, non–relapse-related death, or addition of new systemic therapy for acute GVHD, 0.46; 95% CI, 0.35-0.60]. That tells you that the responses were maintained, and the treatment was still working.

[Most of the adverse events associated with ruxolitinib] were expected; the bone marrow is recovering so it’s a bit fragile. [The most common was] thrombocytopenia. You can reduce the dose of ruxolitinib down to 5 mg adjusted accordingly or support patients with transfusions. CMV reactivation was also common. But again, with letermovir, that happens less and less.

<sub>References:
1. Merck receives FDA approval of Prevymis (letermovir) for prevention of cytomegalovirus (CMV) infection and disease in adult allogeneic stem cell transplant patients. News release. Merck. November 9, 2017. Accessed April 7, 2021. https://bit.ly/3fS6S0Q</sub>

_{2. Scott BL. Long-term follow up of BMT CTN 0901, a randomized phase 3 trial comparing myeloablative (MAC) to reduced intensity conditioning (RIC) prior to hematopoietic cell transplantation (HCT) for acute myeloid leukemia (AML) or myelodysplasia (MDS) (MAvRIC Trial). Biol Blood Marrow Transplant. 2020;26(3):S11. doi:10.1016/j.bbmt.2019.12.07}

_{3. Harris AC, Young R, Devine S, et al. International, multicenter standardization of acute graft-versus-host disease clinical data collection: a report from the Mount Sinai Acute GVHD International Consortium. Biol Blood Marrow Transplant. 2016;22(1):4-10. doi:10.1016/j.bbmt.2015.09.001}

_{4. Martino R, Romero P, Subira M, et al. Comparison of the classic Glucksberg criteria and the IBMTR Severity Index for grading acute graft-versus-host disease following HLA-identical sibling stem cell transplantation. International Bone Marrow}