Post-hoc analysis of the GRAPHITE phase III trial: Vedolizumab for lower-GI aGvHD prophylaxis in Japanese patients

Acute GvHD (aGvHD) remains a leading cause of morbidity and mortality post allogeneic hematopoietic stem cell transplant (allo-HSCT), with lower-gastrointestinal (GI) aGvHD as the main driver for aGvHD-associated mortality. Vedolizumab is a gut-selective anti-lymphocyte trafficking antibody that blocks α₄β₇ integrin signaling with mucosal vascular addressin cell adhesion molecule 1. Vedolizumab has demonstrated efficacy in lower-GI aGvHD prophylaxis after allo-HSCT compared with placebo in the phase III GRAPHITE trial (NCT03657160). A post-hoc analysis assessing the efficacy and safety of vedolizumab vs placebo in lower-GI aGvHD prophylaxis in Japanese and non-Japanese patients who underwent allo-HSCT was published by Goto et al. in the International Journal of Hematology.¹

In the GRAPHITE trial, patients were randomized 1:1 to receive intravenous vedolizumab 300 mg or placebo. All patients received GvHD prophylaxis with tacrolimus or cyclosporine and either methotrexate or mycophenolate mofetil. The primary endpoint was lower-GI aGvHD-free survival by Day 180 post allo-HSCT. Key secondary endpoints included lower GI-aGvHD-free and relapse-free survival, International Bone Marrow Transplant Registry database Grades C–D aGVHD (any organ)-free survival, non-relapse mortality, and overall survival. This post-hoc analysis focused on Japanese (n = 35) and non-Japanese (n = 298) subgroups.  

Key learnings

In Japanese patients, the lower-GI aGvHD-FS rates were numerically higher (94% vs 81%; p = 0.2201), while in the non-Japanese patients, these were significantly higher (84% vs 70%; p = 0.0018) in the vedolizumab vs placebo arms, respectively. 

Lower GI aGvHD-FS RFS (11.1% vs 17.6%), IBMTR Grades C–D aGvHD (11.1% vs 29.4%), IBMTR Grades B–D aGvHD (22.2% vs 35.3%), NRM (0% vs 11.8%), and OS (0% vs 11.8%) were lower in the vedolizumab and placebo arms, respectively, for Japanese patients.

No new safety signals were identified, with Grade ≥3 TEAEs reported in 26.3% vs 17.6% of Japanese patients and 8.7% vs 10.8% of non-Japanese patients in the vedolizumab and placebo arms, respectively.

In Japanese patients, vedolizumab demonstrated a trend toward reduced rates of lower-GI aGvHD-FS with no new safety signals. The findings are limited by the small sample size, warranting robust real-world data.