The safety and efficacy of belumosudil and ruxolitinib combination in ruxolitinib refractory chronic graft-versus-host disease

Abstract Background Chronic graft-versus-host disease (cGVHD) remains a leading cause of non-relapse mortality post-allogeneic hematopoietic stem cell transplantation (allo-HSCT), with 30-50% of patients developing steroid-refractory disease. Despite advances with JAK inhibitors (ruxolitinib) and ROCK2 inhibitors (belumosudil) in cGVHD, long-lasting responses remain uncommon, a significant number of patients need next immunosuppressive therapies. Refractory cases demand personalized, multi-targeted approaches. Preclinical data suggest synergistic effects through dual JAK/ROCK pathway inhibition, potentially addressing fibrotic and Th17-mediated resistance mechanisms. While clinical data on this specific combination are limited, whether the combined belumosudil and ruxolitinib improves efficacy without safety concerns. Methods We conducted a retrospective multicenter study to investigate the safety and efficacy of belumosudil and ruxolitinib combination in the treatment of ruxolitinib refractory cGVHD. Eligibility required progression or suboptimal response after ≥3 months of ruxolitinib, more than 2 prior lines of treatment. Primary endpoints was 3 months overall response rate (ORR). Best of response (BOR), Failure-free survival (FFS), adverse events (AEs), change in Lee Symptom Scale (LSS) summary score, duration of response (DOR) and change in CS dose were secondary end points. Results A total of 42 patients at median 42 (17-68) years were enrolled, 61.9% (n=26/42) had severe NIH-grade cGVHD, and 61.9% (n=26/42) had ≥4 organs involved. The cohort was heavily pretreated, with 52.4% (n=22/42) having received ≥5 prior lines of therapy (LOTs). Lung involvement occurred in 57.1% (n=24/42) (23.8% with lung score 3). The median time from transplantation to cGVHD diagnosis was 217 days (68-743). Median time from ruxolitinib initiation to combination therapy was 303 days (69–2471). At median follow-up of 178 days (57–577), ORR was 54.8% (n=23/42) (95%CI 39.8–67.0), with BOR reaching 76.2% (n=32/42) (95%CI 60.0–87.1). Patients achieved responses at a median of 48 days (21-112) after starting combination treatment. Subgroup analyses revealed that ORR was 53.9% (n=14/26) in severe cGVHD and 63.6% (n=14/22) in ≥5 prior LOTs. Lung cGVHD showed 87.5% (PR:58.3%; CR:29.2%) symptom-based ORR and 12.5% (PR:12.5%) FEV1%-based ORR. Ocular cGVHD demonstrated higher sensitivity to therapy (ORR 45.5% [18.2% CR + 27.3% PR]) than joint/fascia disease (36.0% ORR [12.0% CR + 24.0% PR]). FFS was 92.5% (82.3–100.0) at 12 months. CS doses were reduced in 90% of patients (median reduction 70.8%). The safety analysis revealed leukocytopenia as the most common hematologic adverse event, occurring in 7.2% of patients (n=3/42). Infectious complications included pneumonia in 14.3% of cases (n=6/42), with grade ≥3 events observed in 4.8% (n=2/42). Notably, no treatment-related discontinuations or grade ≥4 hematologic toxicities were reported. The combination was well tolerated with preserved immune reconstitution and improved LSS. Clinically meaningful improvement (≥7-point reduction) in LSS summary scores was achieved by 19.1% of patients (8/42) at 3 months. Responders demonstrated significantly greater LSS improvement compared to non-responders. Conclusion The belumosudil and ruxolitinib combination demonstrated clinically meaningful efficacy in ruxolitinib-refractory cGVHD. High response rates (ORR 54.8%, best ORR 76.2%), durable disease control (12-month FFS 92.5%), and significant CS reduction (median 70.8% dose decrease) were achieved without severe toxicity (grade ≥3 infections: 4.8%; no treatment discontinuations). These findings support the mechanistic rationale for concurrent JAK/ROCK inhibition and warrant prospective validation.