Ruxolitinib: A Long-Awaited Standard for Steroid Refractory Acute Graft-Versus-Host Disease

Allogeneic stem-cell transplantation (allo-SCT) offers long-term disease-free survival for a variety of hematologic diseases. However, patients may experience the treatment-related toxicity acute graft-versus-host disease (GVHD). Despite major advances such as high-resolution human leukocyte antigen (HLA) typing, improved donor selection algorithms, and novel GVHD prophylactic regimens, approximately 50 percent of allo-SCT recipients will still develop acute GVHD.^1-3  High-dose glucocorticoids have long been used as standard first-line therapy for acute GVHD.⁴  However, complete response rates are only approximately 40 percent.⁵  Multiple studies have attempted, unsuccessfully, to establish a preferred second-line therapy.⁴  Without a clear treatment strategy, overall survival (OS) in glucocorticoid refractory acute GVHD has been less than 25 percent.⁶  There is a critical need for novel therapies.

The Janus kinase (JAK) and signal transducers and activators of transcription (STAT) signaling pathways were recently shown to contribute to cytokine transduction and downstream regulation of immune cells critical to the pathophysiology of acute GVHD in murine models.^7,8  Moreover, blockade of these pathways seemed to reduce acute GVHD while preserving the allogeneic graft-versus-tumor (GVT) effect, an important finding as most therapies for GVHD concomitantly decrease GVT.⁹  These findings established a potential role for ruxolitinib, a JAK1/2 inhibitor approved for the treatment of myelofibrosis, in the treatment of GVHD. The single-arm REACH1 trial established the efficacy and led to U.S. Food and Drug Administration approval of ruxolitinib for the treatment of steroid refractory GVHD.¹⁰ 

The REACH2 randomized, multicenter, international phase III trial compared ruxolitinib (154 patients) to investigator’s choice therapy (155 patients) in patients with glucocorticoid refractory GVHD. Ruxolitinib was started at 10 mg by mouth twice daily, to be tapered after 56 days. The control arm included treatment with antithymocyte globulin, extracorporeal photophoresis, mesenchymal stromal cells, low-dose methotrexate, mycophenolate mofetil, sirolimus/everolimus, etanercept, or infliximab. Crossover was permitted to the ruxolitinib arm for 49 patients in the control arm who did not respond to therapy within 28 days.

The study met the primary endpoint, showing higher overall response (OR) in the ruxolitinib arm (62% vs. 39%; p<0.001). Complete response rates were 34 percent with ruxolitinib compared to 19 percent with control. Separated by grade, responses to ruxolitinib versus control were 75 percent versus 51 percent for grade II, 56 percent versus 38 percent for grade III, and 53 percent versus 23 percent for grade IV. The key secondary endpoint was durable response defined as response by day 28 and maintained to day 56. Durable response was significantly higher in the ruxolitinib group (40% vs. 22%; p<0.001). Of responders at day 28, only 10 percent of patients in the ruxolitinib group lost their response by day 56, compared to 39 percent in the control arm. Median failure-free survival was 5.0 months for ruxolitinib and 1.0 month for controls (HR, 0.46; 95% CI, 0.35-0.60). OS was similar between the two groups (HR, 0.83; 95% CI, 0.60-1.15).

For ruxolitinib, the most common adverse events (AEs) higher than grade 3 were thrombocytopenia (33%), anemia (22%), and infection (22%). First infection occurred at a median of 0.8 months, and the most common infection was cytomegalovirus (7%). AEs leading to discontinuation of ruxolitinib occurred in 11 percent compared to 5 percent in the control arm. By day 28, serious AEs occurred in 38 percent of patients treated with ruxolitinib and 34 percent in the control arm.