Managing GVHD Following HCT: New Uses for Old Treatments?

Graft-versus-host disease (GVHD) remains a major complication following allogeneic hematopoietic cell transplantation (alloHCT) and, when severe, is associated with poor long-term overall survival. Two studies recently published in Blood evaluated two new drugs for their effects on GVHD: off-label anti-PD-1 monoclonal antibodies, and the anti-CD30 antibody-drug conjugate brentuximab vedotin (BV).

Brentuximab Vedotin in Acute GVHD

In a multicenter, phase I study, Yi-Bin Chen, MD, from Massachusetts General Hospital, and co-authors assessed the activity of BV in patients with steroid-refractory GVHD, after murine models suggested that CD30 may mediate acute GVHD (aGVHD).1

The study included 34 patients (median age = 56 years; range = 22-73 years) who had aGVHD that progressed after at least three days of treatment with systemic corticosteroids, that did not improve after at least seven days of daily oral prednisone ≥1 mg/kg or equivalent and warranted treatment with another agent, or that flared during the tapering of corticosteroids. Patients were excluded if they had aGVHD or chronic GVHD (cGVHD) overlap syndrome, had significant organ dysfunction, or used more than one systemic agent besides corticosteroids for aGVHD.

The investigators determined the maximum tolerated dose (MTD) of BV using a 3+3 cohort design, with an initial dose of 0.6 mg/kg weekly, escalating to 0.9 and 1.2 mg/kg weekly. Six patients received BV weekly for three weeks, followed by maintenance dosing every three weeks for four additional doses. After two patients died because of neutropenia, the study protocol was revised, and the next 28 patients were treated with 0.6 mg/kg (n=10) or 0.8 mg/kg (n=18) every two weeks for four doses.

The authors established BV 0.8 mg/kg as the MTD, and sepsis was the only dose-limiting toxicity.

At 28 days of treatment, the five patients achieved a complete response (CR; defined as resolution of aGVHD, per consensus criteria) and eight achieved a very good partial response (VGPR), for an overall response rate (ORR; defined as CR and VGPR) of 38.2 percent. An additional seven patients achieved CR by day 56. One patient in CR at day 28 later died, resulting in an overall CR rate of 32.4 percent at day 56.

Four of the 12 patients who achieved CR by day 56 later required additional therapy for recurrent aGVHD symptoms.

The ORRs at six and 12 months were 41 percent (95% CI 25-57) and 38 percent (95% CI 22-54), respectively. Of the 18 patients whose GVHD responded to treatment by day 56, 12 were alive at 12 months of follow-up; six died from infection (n=3), disease relapse (n=2), or GVHD (n=1). Of the 16 non-responders, only one patient was alive at 12 months of follow-up and 11 of those deaths were related to GVHD.

The remaining causes of death were infection (n=3) or disease relapse (n=1).

In the 29 patients with peripheral blood mononuclear cells collected at baseline, CD30 expression was not associated with clinical response. “The lack of correlation of CD30 expression in central memory CD8+ T cells in this trial may be [because] this population of patients [was] already steroid-refractory when samples were collected,” the authors noted.

The study is limited by its small and heterogenous patient population. Though the authors observed “significant activity” for BV in patients with steroid-refractory aGVHD, “future investigation should focus on identifying a biologically defined population most likely to respond to BV.”

Anti-PD-1 Monoclonal Antibodies

In a multicenter, retrospective study, Bradley M. Haverkos, MD, MPH, MS, from the Division of Hematology at the University of Colorado, and co-authors assessed whether anti-PD-1 monoclonal antibodies had activity in the setting of lymphoma that relapsed following alloHCT; the study evaluated the risk of GVHD after PD-1 blockade in that patient population.2

“There have been a few [patients with] severe and even fatal transplant-related complications, including GVHD, when anti-PD-1 monoclonal antibodies were given for disease control prior to alloHCT, which has led to a package insert warning,” the authors explained.

For this study, the researchers contacted transplant programs with high volumes of patients with post-transplant lymphoma to identify 31 evaluable patients (median age at time of transplant = 37 years; range = 20-68 years) who were receiving nivolumab or pembrolizumab. Most patients (87%) received salvage therapy (median = 2 therapies; range = 0-6 therapies) before initiating anti-PD-1 therapy, and many patients (n=19; 61%) experienced GVHD prior to anti-PD-1 treatment.

Patients received a median of two doses of anti-PD-1 therapy (range = 1-31 doses). After a median follow-up of 428 days (range = 133-833 days) after the first dose of anti-PD-1 therapy, the ORR (defined as CR and partial response [PR]) was 77 percent in 30 evaluable patients, including 15 who achieved CR and eight who achieved PR. Three patients had stable disease and four had progressive disease.

“While disease response rates were high, the frequency of GVHD after anti-PD-1 was also high,” the authors wrote. More than half of the patients (55%; n=17) developed treatment-emergent GVHD after initiating anti-PD-1 therapy:

  • 6 with aGVHD, following a median of 1 dose (range = 1-2 doses)
  • 7 with cGVHD, following a median of 2 doses (range = 1-25 doses)
  • 4 with aGVHD/cGVHD overlap, following a median of 2 doses (range = 1-2 doses)

Responses to treatment for GVHD were also poor, the authors added, with 14 of 17 patients requiring two or more systemic therapies, including steroids, calcineurin inhibitors, extracorporeal photopheresis, topical or local therapy, rituximab, and sirolimus. Of the 10 deaths during follow-up, eight (26%) were deemed related to new-onset GVHD following anti-PD-1 therapy.

The majority of treatment-emergent GVHD occurred in patients with a history of GVHD (n=12; 71%), but the only factor that was associated with development of GVHD after anti-PD-1 therapy was alloHCT with matched sibling donor (p=0.02).

“PD-1 blockade [in these patients] appears to be highly efficacious but frequently complicated by rapid onset of severe and treatment-refractory GVHD,” the authors concluded. “Providers should be cognizant of this potentially fatal and morbid complication. … Administration of anti-PD-1 after alloHCT should be done with extreme caution and preferably carried out within the context of a clinical trial.”

The study is limited by its small patient population and retrospective design. Responses were also assessed by individual investigators at each participating center, rather than by central monitoring.

Drs. Chen and Perales report financial relationships with Seattle Genetics, Inc., which provided BV and funding for the clinical trial. Drs. Haverkos and Abbott report financial relationships with Merck.


  1. Chen YB, Perales MA, Li S, et al. Phase I multicenter trial of brentuximab vedotin for steroid refractory acute graft-vs.-host disease (GVHD). Blood. 2017 May 4. [Epub ahead of print]
  2. Haverkos BM, Abbott D, Hamadani M, et al. PD-1 blockade for relapsed lymphoma post allogeneic hematopoietic cell transplant: high response rate but frequent GVHD. Blood. 2017 May 3. [Epub ahead of print]