Carfilzomib for Treatment of Refractory Chronic Gvhd: A Chronic GVHD Consortium Pilot Trial

Abstract

Advances are needed in the therapy of treatment-refractory chronic graft-vs.-host disease (GVHD). Prior experimental and clinical data suggest proteasome inhibition may have immunomodulatory activity relevant to GVHD. Carfilzomib is a specific, irreversible, next-generation proteasome inhibitor with extensive experience in the therapy of multiple myeloma. To explore the safety and activity of carfilzomib in advanced chronic GVHD, we conducted a pilot multi-center phase II trial through the Chronic GVHD Consortium. Carfilzomib was administered at 20 mg/m2 on day 1, and then 36mg/m2 on days 8 and 15 of a 28-day treatment cycle (cycle 1), and then 36 mg/m2 on days 1, 8, and 15 of a 28-day treatment cycle (cycles 2-6). The primary endpoint was 6 month treatment failure, a composite endpoint including death, relapse, and requirement for an additional line of systemic immune suppressive (IS) therapy. Considering a historical 6 month treatment failure rate of 44%, this 20 subject pilot trial allowed estimation of the 6 month treatment failure rate with a standard error of 10 percentage points. Secondary objectives included safety, physician- and NIH-based response rates, failure-free survival, and patient-reported outcomes. Key eligibility criteria included age ≥ 18, NIH Consensus defined chronic GVHD, and failure of at least one prior line of systemic IS therapy. A total of 20 subjects were enrolled at 4 institutions. Median age was 53 years. Median time from transplant to enrollment was 2.8 years (IQR 1.7-5), and from chronic GVHD onset to enrollment 1.5 years (IQR 0.5-3.7). Chronic GVHD was NIH moderate (30%) or severe (70%), predominantly classic (90% vs. overlap 10%), and involved multiple diverse organ sites. Prior lines of systemic therapy for chronic GVHD were ≤ 2 (n=6, 30%) or ≥ 3 (n=14, 70%). Doses were held primarily for infection (50% of total held doses); only 4 patients (20%) completed all planned 6 months of carfilzomib. SAEs occurred in 40% of subjects, and 7 deaths occurred among study participants occurring between 0.3-9 months after last carfilzomib dose (none attributed to carfilzomib). The 6 month treatment failure rate was not significantly improved vs. the historical benchmark (40% vs. 44%, p=0.36), figure 1. NIH response rates at 3, 6, and 12 months are presented in figure 2. Overall survival at 6 and 12 months was 79% and 58%. Failure-free survival at 12 months was 32%. These pilot phase II data suggest that carfilzomib therapy in this very advanced chronic GVHD population did not improve expected 6 month treatment failure rates achieved under conventional practices and is not recommended for further study for this indication.