Introduction: Standard first-line therapy for chronic graft-versus-host disease (cGVHD) includes systemic oral glucocorticoid (GC) which, although effective initially, are associated with serious long-term toxicities and possibly steroid resistance/dependence. Belumosudil (BEL), a ROCK2 inhibitor, is a US FDA-approved drug for the treatment of steroid-refractory cGVHD which was associated with 74-77% overall response rate in advanced cGVHD in the phase 2 ROCKstar trial. Despite promising efficacy and safety data with use of BEL in clinical trials and real-world settings, there remains a limited understanding of real-world use of BEL in combination with systemic GC. This study focuses on understanding real-world prescribing patterns and assessing the GC-sparing effects (GC dose taper or discontinuation) of BEL.

Methods: An analysis was conducted using two complementary real-world databases: ROCKreal (academic transplant centers) and MarketScan (mixed community/academic centers). In ROCKreal, eligible patients were ≥12 years of age with cGVHD who received BEL after 2-5 prior systemic therapies. In MarketScan, patients were identified using ICD-9/10 codes for cGVHD (279.52, D89.811) or unspecified GVHD (279.5, D89.813), required ≥6 months continuous enrollment prior to cGVHD diagnosis, and BEL initiation in line of therapy (LOT) 3-6 between March 2015 and March 2024. Those with cancer relapses prior to LOT 3 were excluded from both cohorts. Patients were stratified at BEL initiation into BEL+GC and BEL without (w/o) GC; other concomitant therapies were allowed in both groups. Primary objectives included characterization of baseline demographics, disease characteristics, and treatment history across both databases. Secondary objective was to assess cumulative incidence of GC-dose reduction to ≤5 mg/day and discontinuation within 1 year (censored at treatment switch, death, or end of LOT) from BEL start, using ROCKreal data.

Results: The analysis included 113 patients from ROCKreal and 87 patients from MarketScan. In ROCKreal cohort, 77 (68.1%) received BEL+GC and 36 (31.9%) received BEL w/o GC, while in MarketScan, 67 (77.0%) received BEL+GC and 20 (23.0%) received BEL w/o GC, showing consistent utilization patterns across practice settings.

Baseline characteristics in the ROCKreal cohort were generally balanced between BEL+GC (N=77, 68%) and BEL w/o GC (N=36, 32%) groups including median age (63.1 [IQR 51.0, 69.8] vs 62.8 [48.8, 68.8] years), gender (56% vs 61% males), and race (83% vs 89% white). The most common organs involved were skin (79% vs 66%), eyes (64% vs 66%), and muscles/joints/fascia (52% vs 49%). In the MarketScan cohort, which had lower median age (53.0 [43.0, 62.5] vs 45.5 [40.2, 58.8] years), baseline characteristics were also generally balanced between BEL+GC and BEL w/o GC groups except male gender (54% vs 40%), and higher proportion of patients with cardiovascular (63% vs 30%), diabetes (31% vs 10%) and hepatic (27% vs 15%) comorbidities at baseline.

In the ROCKreal cohort, 73 patients were receiving GC at BEL initiation of which 56 patients received GC dose >5 mg. Among the cohort receiving GC at BEL initiation, 23.7% (n=13; 95% CI: [14.2%, 37.9%]) of patients have discontinued GC at one year. Among those who discontinued (n=13), the median time to CS discontinuation was 119 days (IQR 67-273 days). Among patients receiving GC dose > 5 mg, 45.2% (19; 95% CI: [30.9% ,62.3%]) have either had a steroid dose reduction or discontinued GC at 1-year. Steroid dose reduction below 5 mg was observed in 31.6% (12; 95% CI: [18.8%, 50%]) patients out of 56 (76.7%) patients. Additional analyses for outcomes in the MarketScan data are ongoing.

Conclusions: We found BEL to be most often used concurrently with GC among cGVHD patients receiving therapy in LOT 3-6. ROCKreal data points at GC reduction or discontinuation potential with the use of BEL in the majority of treated patients in concordance with previously published studies. By reducing long-term GC exposure, the use of BEL holds promise to lessen the burden of GC-related adverse events, thereby potentially mitigating morbidity and improving the quality of life for patients with cGVHD.

This content is only available as a PDF.
2025
Sign in via your Institution