P1351: CHARACTERIZING ALLOGENEIC STEM CELL TRANSPLANTATION SAFETY IN PATIENTS WITH HEMATOLOGIC MALIGNANCIES AFTER PEMBROLIZUMAB THERAPY

Abstract

Background: Although allogeneic stem cell transplantation (allo-SCT) and programmed cell death 1 (PD-1) blockade have separately shown efficacy in some lymphoid malignancies, the safety of allo-SCT after anti–PD-1 therapy in these patients remains a high clinical interest. Aims: We present data from an analysis from 4 KEYNOTE phase 1-3 studies to describe predetermined complications in patients who received allo-SCT following pembrolizumab therapy. Methods: The analysis included patients from the KEYNOTE-013 (NCT01953692, n=20), KEYNOTE-087 (NCT02453594, n=31), KEYNOTE-170 (NCT02576990, n=5), and KEYNOTE-204 (NCT02684292, n=14) trials who have a known allo-SCT transplantation date received within 2 years of the last dose of study pembrolizumab. Descriptive statistics were used for predefined post–allo-SCT complications of interest. The cumulative incidence of acute grade 2-4 graft-versus-host disease (GVHD), acute grade 3-4 GVHD, and chronic GVHD was estimated. The corresponding competing risk events were death without acute grade 2-4 GVHD, death without acute grade 3-4 GVHD, and death without chronic GVHD, respectively. The cumulative incidence of post–allo-SCT transplant-related mortality (TRM) was estimated with relapse as a competing risk. Progression-free survival (PFS) and overall survival (OS) were estimated by the Kaplan-Meier method. Results: Seventy patients were included in the analysis. Median age was 30 y (range, 18-65), 57 (81.4%) had classical Hodgkin lymphoma (cHL), and the rest had non-Hodgkin lymphoma. 69 of 70 (98.6%) patients received pembrolizumab monotherapy; 1 patient received pembrolizumab + lenalidomide. Median duration on study treatment was 5.3 months (range, 0.7-29.6) and median time from last pembrolizumab dose to first allo-SCT was 4.6 months (range, 1-20). Before transplantation, 49 patients (70.0%) had intervening anticancer regimen; 34 (48.6%) had active disease, 31 patients (44.3%) were in remission at time of transplant, and 5 patients (7.1%) had unknown disease status. Overall, 55 patients (78.6%) developed GVHD (acute, 38; chronic, 17). The estimated 6-month post–allo-SCT cumulative incidence was 0.41 (95% CI, 0.30-0.53) for grade II-IV acute GVHD and 0.20 (95% CI, 0.12-0.30) for grade III-IV acute GVHD; 1 year post–allo-SCT was 0.21 (95% CI, 0.12-0.31) for chronic GVHD. Other predetermined complications, including critical illness, immune-mediated adverse events, pulmonary complications, and veno-occlusive liver disease, occurred in 32 patients (45.7%). After a median follow-up (defined as the time from allo-SCT to data cutoff) of 40.1 months (range, 5.1-71.2), the post–allo-SCT median PFS was not reached (NR) (95% CI, 14.5-NR) and the 30-month post–allo-SCT PFS rate was 56.8% (95% CI, 42.9-68.5) (Figure A); median OS was NR (95% CI, NR-NR) and the OS rate at 12 months was 82.2% (Figure B). The estimated post–allo-SCT cumulative incidence for TRM at 6 months was 0.09 (95% CI, 0.03-0.17) and for relapse at 26 months was 0.27 (95% CI, 0.16-0.38). PFS and OS outcomes were comparable in the subgroup of patients with cHL. Image:Summary/Conclusion: The incidence of GVHD in this population is comparable to historical data (40%-80% and 30%-70%, respectively). The incidence of severe acute GVHD may be higher than a typical modern allo-SCT series, although TRM was low. The observed PFS and OS outcomes compare favorably to historical benchmarks (2-year PFS ranging from 45%-48%; 1-year OS >80%). Altogether, this analysis lends confidence that allo-SCT is feasible for patients with lymphoid malignancies after PD-1 blockade.