In a phase 3 clinical trial, ruxolitinib (Jakafi) led to superior overall response rates and durable responses compared with best available therapy (ie, control group) in patients with steroid-dependent or steroid-refractory chronic graft-versus-host disease (GVHD), with acceptable safety, according to results presented at ASH 2020.
“Ruxolitinib is the first agent to demonstrate superior efficacy to best available therapy in a phase 3 trial of patients with chronic GVHD and an inadequate response to steroids. It shows a significant advantage for ruxolitinib,” stated lead investigator Robert Zeiser, MD, Head, Section of Tumor Immunology and Immune Modulation, Medical Center–University of Freiburg, Germany. “It is likely that this trial will lead to approval for this indication and change the guidelines for the treatment of this disease.”
GVHD is a leading cause of mortality and morbidity in patients with nonrelapsed disease who are undergoing allogeneic stem-cell transplant. Systemic steroids are the standard therapy for GVHD, but approximately 50% of patients will become steroid-dependent or -refractory, and no standard second-line treatment is available. Ruxolitinib, a Janus kinase 1/2 inhibitor, may fulfill this unmet need.
The REACH3 Study
Patients with steroid-dependent or steroid-refractory chronic GVHD enrolled in the phase 3 REACH3 study were randomized in a 1:1 ratio to ruxolitinib 10 mg twice daily (N = 165) or to best available therapy with steroids plus or minus calcineum inhibitor (N = 164).
The primary efficacy end point was overall response rate at 24 weeks. After week 24, crossover to ruxolitinib was allowed during an extension phase that lasted until week 156 (ie, end of treatment). The safety follow-up period was an additional 30 days.
The patients’ baseline characteristics were well-matched for disease and demographic characteristics. The patients were also well-matched for stem-cell source, donor type, and donor or recipient of cytomegalovirus type.
At the end of the study, 50.3% of the patients in the ruxolitinib group and 25.6% of the patients in the best available therapy group were still continuing with their treatment.
Treatment discontinuations were reported in 49.7% and 74.4%, respectively. More treatment discontinuations because of adverse events occurred in the ruxolitinib group than the best available therapy group (17.9% vs 4.9%, respectively). The reverse was the case because of lack of efficacy (14.5% vs 42.7%, respectively), demonstrating greater efficacy with ruxolitinib. A total of 37.2% of patients crossed over to ruxolitinib during the extension phase.
The overall response rate was significantly higher with ruxolitinib than in the control group at week 24 (49.7% vs 25.6%, respectively; P <.0001). The complete response rate was 6.7% versus 3%, and the partial response rate was 43% versus 22.3%, respectively.
At week 24, patients in the ruxolitinib group had a significantly better improvement in symptoms (24.2%) versus the control group (11%; P = .001). There was a significant 63% relative improvement in failure-free survival favoring ruxolitinib at week 24 (P <.0001).
The best overall response rate was 76.4% versus 60%, respectively, and the duration of response was also significantly longer with ruxolitinib—not reached versus 6.24 months in the control group.
“This is the first successful phase 3 trial in chronic GVHD ever,” stated Robert A. Brodsky, MD, Professor of Medicine, and Director, Division of Hematology, Johns Hopkins University School of Medicine, Baltimore, MD. “Results suggest that ruxolitinib is a viable second-line treatment for patients whose symptoms are not fully resolved with corticosteroids,” Dr Brodsky said at a premeeting press conference.
The rates of adverse events of any grade and grade 3 or higher were similar in both arms. Serious adverse events were reported in 33% of the ruxolitinib group and 36.7% of the control group. The number of deaths was not significantly different between the 2 study arms.
The most frequent adverse events with ruxolitinib were anemia and thrombocytopenia. The rates of grades 3 or 4 anemia were 12.7% with ruxolitinib versus 7.6% in the control group; grades 3 or 4 thrombocytopenia were 15.2% and 10.1%, respectively.
No significant difference was observed between the treatment arms in the rate of viral or bacterial infections and in cytomegalovirus infection or reactivation.