Introduction: Chronic graft-versus-host disease (cGVHD) is a heterogenous, multiorgan complication of allogeneic hematopoietic cell transplantation that is characterized by inflammation and fibrosis. Belumosudil, a selective Rho-associated coiled-coil containing protein kinase 2 (ROCK2) inhibitor, is approved by the FDA for patients with cGVHD who have failed at least two prior lines of therapy (LOT). While belumosudil has demonstrated efficacy across all organ systems involved in cGVHD, the timing and dynamics of organ-specific responses have not previously been characterized. This secondary analysis aimed to evaluate time to response (TTR) and patterns of organ-specific improvement over the first year of belumosudil therapy in clinical trial participants with long-term follow up.

Methods: This analysis utilized data from the Phase 2 ROCKstar trial (NCT03640481), which evaluated belumosudil in patients with cGVHD. Clinical response was assessed using the 2014 NIH consensus response criteria for cGVHD and was limited to participants identified as responders (those who achieved complete [CR] or partial response [PR]). Cumulative overall and organ-specific response rates were assessed; TTR was defined as the interval from initiation of belumosudil to the first documented CR or PR. Late responses were defined as those occurring after at least 6 months of belumosudil therapy.

Results: Of the 152 participants enrolled in the ROCKstar trial, 114 (75%) were identified as responders, with 57 in each of the 200 mg once-daily and 200 mg twice-daily dosing cohorts. The median age was 55 years (range,18–77). cGVHD severity at screening was mild in 1 participant (0.9%), moderate in 34 (29.8%), and severe in 79 (69.3%), with a median of 4 organs involved (range, 1–7) at baseline. Participants had received a median of 3 prior systemic LOTs; with 74.7% being refractory to their last systemic therapy prior to enrollment. Organ involvement at baseline was as follows: skin (n=95); joint/fascia (n=86); eyes (n=85); mouth (n=68); lungs (n=37); esophagus (n=26); upper gastrointestinal (GI) tract (n=14); liver (n=12); and lower GI tract (n=10). Median follow-up was 30.4 months (range, 2.6–40.5). Median TTR was 4.4 weeks (range, 3.7–80.1). Cumulative overall response rates at 1, 2, 3, 6, and >12 months were 14.0%, 65.8%, 86.8%, 94.7%, and 100%, respectively. Cumulative organ-specific response rates at >12 months were as follows: skin, 48.4%; joint/fascia, 91.9%; eyes, 55.3%; mouth, 77.9%; lungs, 40.5%; esophagus, 76.9%; upper GI tract, 85.7%; liver, 25.0%; and lower GI tract, 90.0%. The shortest median TTR was observed for lower GI tract involvement (4.1 weeks; range, 3.7–16.3), while the three longest median TTRs were seen in the lungs (16.1 weeks; range, 8.0–104.3), mouth (11.9 weeks; range, 3.7–152.1), and skin (9.6 weeks; range, 3.7–127.9). The organs with the most rapid responses were lower GI tract (80.0% within 3 months), upper GI tract (78.6% within 3 months), and joints/fascia (74.4% within 3 months); late responses were rare or absent in these organs. Conversely, the lungs (21.6% and 40.5% at 6 and >12 months) and eyes (38.8% and 55.3% at 6 and >12 months) had the highest proportions of late responses, with responses after 6 months accounting for 46.7% of all lung responses and 29.8% of all eye responses. Of note, early responses in lungs were rare (10.8% at 3 months). For skin, the most commonly involved organ, response rates were 29.5%, 40.0%, and 48.4% at 3, 6, and >12 months, respectively.

Conclusion: Belumosudil achieved clinical responses across all involved organs in cGVHD. While most participants experienced their first clinical response within approximately 3 months of therapy, a small proportion had late responses, with 5.3% of all overall responses occurring after 6 months of treatment. Response kinetics varied by organ, with the most rapid responses seen in GI and joint/fascia involvement, and the highest rates of late responses observed in the lungs and eyes. These data underscore the importance of tailoring clinical expectations and treatment duration based on organ-specific response dynamics.

Funding: This study was funded by Sanofi. Medical writing assistance was provided by Helen Jones, PhD, of Envision90Ten, an Envision Medical Communications agency, part of Envision Pharma Group, and was funded by Sanofi.

This content is only available as a PDF.
Sign in via your Institution