Ruxolitinib for Steroid-Refractory Acute Graft-Versus-Host Disease: A Single Centre Retrospective Study

Background Acute graft-versus-host disease (aGVHD) is the major cause of non-relapse morbidity and mortality after allogeneic hematopoietic stem cell transplantation(allo-HSCT). But only 40% of the patients will respond to first-line therapies Corticosteroids. Ruxolitinib, a selective Janus kinase (JAK) 1/2 inhibitor, reduces the incidence and severity of GVHD while preserving graft-versus-leukemia effects in preclinical models. The retrospective study evaluated the efficacy of ruxolitinib compared with other second-line therapies for steroid-refractory aGVHD (SR-aGVHD) in a single-center. Methods A total of 90 patients who developed SR-aGVHD after HSCT were evaluated in this retrospective study.They were treated with ruxolitinib (n=45) or other salvage-therapies (n=45) including MTX, basiliximab, etanercept and MSC at our institution. The primary endpoint was the overall response rate (ORR) at Day 28. Additional endpoints included overall survival (OS), cumulative incidence rates of failure-free survival (FFS), relapse, non-relapse mortality (NRM) and cGVHD. Treatment failure was defined as 1) addition of new systemic therapy for aGVHD, 2) NRM, 3) relapse, 4) progression of hematologic disease. FFS and OS were calculated from the day of starting the use of second-line treatments for aGVHD. Results The median age was 34 years (range 14-69). At the time of enrollment 25 patients had grade Ⅱ disease, 39 with grade Ⅲ disease, 26 with grade IV disease. The skin was involved in 54.4%, lower gastrointestinal (GI) tract in 78.9% and liver in 20.2%. With a median follow-up of 1.33 years, the ORR at day 28 was higher in ruxolitinib group than non-ruxolitinib group (62.2% [95% CI, 47.5%-77.0%] vs. 26.7% [95% CI, 13.2%-40.1%], P=0.001) (Table 1). The 1-year OS was 64.4 % and 45.5% in the two groups, respectively (P=0.0382). Ruxolitinib treatment also improved the 1-year cumulative incidence of FFS (57.8% vs. 26.6%, P=0.002), while the 1-year cumulative incidence of relapse did not differ significantly (9.6% vs. 20.0%, P=0.195). The 1-year cumulative incidence of NRM was lower in the ruxolitinib group than the non-ruxolitinib group (24.4% vs. 45.3%, P=0.023). The 1-year and 3-year cumulative incidence of cGVHD were 17.8% vs. 33.3% and 26.8% vs. 44.4% between the ruxolitinib group and non-ruxolitinib group (P=0.10 and P=0.04). Conclusions Our study demonstrated that ruxolitinib is effective than other second-line treatments in patients with SR-aGVHD due to the higher r