Introduction: Allogeneic Hematopoietic-cell transplantation (allo-HCT) may elicit immunological graft versus-tumor effects against tumor cells. However, these immune responses can be misdirected towards normal host organs, resulting in graft-versus host disease (GVHD). Acute GVHD (aGVHD) develops after activation and maturation of antigen-presenting cells, which then leads to donor T-cell activation, expansion, and destruction of host tissue by effector cells. The programmed death (PD) pathway serves as a checkpoint to limit T-cell-mediated immune responses. Blocking the PD1 receptor results in T-cell activation, proliferation, and can induce a potent immunotherapeutic antitumor effect. In the post-allogeneic HCT setting there is concern that activating T-cells via PD1 blockade may induce GVHD (Blazar, J Immunol 2003). While there have been several cases of severe and even fatal transplant-related complications, including GVHD, when nivolumab was given for disease control prior to allo-HCT, less is known about its use in managing relapse after HCT. A few case reports suggest it may be safe (Herbeaux, ASH 2015; Angenendt, BMT 2015; Yared, BMT 2016).

Methods:With IRB approval, we conducted a multicenter, retrospective analysis of lymphoma patients (pts) who received monoclonal antibodies against PD1 after allo-HCT. We contacted 10 U.S. transplant programs with the highest volumes of lymphoma pts undergoing HCT, as provided by the Center for International Bone Marrow Transplant Registry (CIBMTR), to derive an estimate of the safety and toxicity of PD1 blockade post allograft. Additional sites were surveyed based on recommendation from these initial 10 sites. Descriptive statistics were used to summarize patient characteristics and clinical outcomes. Response assessments were defined according to revised Lugano criteria (Cheson, 2014). GVHD stage/grade was recorded according to Consensus scoring severity index. PD1 and PDL1 immunohistochemical (IHC) stains were performed on 2 pts' liver biopsies treated at University of Colorado (CU).

Results:We surveyed 23 sites with 21 replies to date. 8 sites reported no experience using anti-PD1 after allo-HCT. 13 sites identified 27 lymphoma pts (26 classical Hodgkin lymphoma) who received a monoclonal antibody against PD1 for relapsed disease after allo-HCT (24 nivolumab, 3 pembrolizumab). Table 1 highlights patient characteristics prior to treatment with anti-PD1. At a median follow up of 217 days (range 26-560) after the first dose of PD1 blockade, 22 of 27 (81%) remain alive. 3 pts died from GVHD and 2 died from lymphoma progression complicated by GVHD. Overall response rate (ORR) to anti-PD1 was 79% (CR+PR) in 24 assessable pts (13 CR, 6 PR, 2 SD, 3 PD). 10 of 27 pts (37%) developed aGVHD and in 19% it was severe (5 grade IV, 3 grade II, & 2 grade I) after PD1 blockade. 9 of 10 experienced liver aGVHD (4 stage IV, 1 stage III, 4 stage I). 3 of 4 cases with stage IV liver aGVHD were biopsy proven and distinct from typical hepatopathy associated with PD1 blockade. In the 2 cases with PD1 IHC stains, PD1+ lymphocytes were more frequently involved in the endotheliitis rather than the cholangitis component of GVHD. 8 of 10 experienced skin aGVHD (1 stage IV, 5 stage III, 1 stage II, & 1 stage I). Only 2 experienced gut aGVHD (1 stage IV, 1 stage III). The 10 who developed aGVHD received a total of 1-3 doses (median 1) of anti-PD1. Median day of aGVHD onset from first dose of anti-PD1 was 15 (range 8-28). Of the 10 pts who developed aGVHD, 4 of 10 (40%) responded to treatment (2 CR, 2 PR). Only 1 patient responded to steroids alone. An additional 4 pts who did not experience aGVHD had new or worsening classic chronic GVHD (cGVHD) after anti-PD1. Overall 52% (14/27) of the pts in this analysis developed new or worsening acute or chronic GVHD after PD1 blockade (10 aGVHD, 4 cGVHD).

Discussion/Conclusion: PD1 blockade was associated with a high ORR (79%) in heavily pretreated Hodgkin lymphoma patients relapsing following allo-HCT, suggesting a strong graft versus tumor effect. However, some patients developed severe and treatment refractory GVHD in this setting. Prospective study of PD1 blockade after allo-HCT is warranted. Potential risks and benefits of PD1 blockade after allo-HCT should be discussed with each patient, and they should be carefully monitored for aGVHD after PD1 blockade.


Kamdar:Seattle Genetics: Speakers Bureau. Saad:Astellas: Research Funding; American Porphyria foundation: Research Funding; Alexion: Honoraria; Spectrum: Honoraria. Ganguly:Janssen: Research Funding; Onyx: Speakers Bureau; Seattle Genetics: Speakers Bureau. Hari:Merck: Research Funding; BMS: Honoraria. Hamadani:Celgene: Honoraria, Research Funding; Takeda: Research Funding; Pharmacyclics: Consultancy; Merck: Research Funding. Jagasia:Therakos: Consultancy. Armand:Sequenta Inc: Research Funding; Merck: Consultancy, Research Funding; Pfizer: Research Funding; Roche: Research Funding; Infinity Pharmaceuticals: Consultancy; Bristol-Myers Squibb: Consultancy, Research Funding.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract