Ruxolitinib Induces Prolonged Efficacy in Steroid-Refractory Acute GVHD


Ruxolitinib has demonstrated sustained efficacy with a lower probability of progression or additional systemic therapeutics in patients with steroid-refractory acute graft-versus-host disease.

Mohamad Mohty, MD, PhD

Mohamad Mohty, MD, PhD

Ruxolitinib (Jakafi) has demonstrated sustained efficacy with a lower probability of progression or additional systemic therapeutics in patients with steroid-refractory acute graft-versus-host disease (GVHD), according to 6-month follow-up data from the phase 3 REACH2 study (NCT02913261) presented during the Virtual 47th Annual Meeting of the EBMT.1

Patients who received ruxolitinib experienced a longer median duration of response (DOR) compared with the best available therapy (BAT), at 163 days (range, 22-623 days) vs 101 days (range, 10-456 days), respectively. Additionally, the ruxolitinib cohort had a lower probability of loss of response at both 6 months (8.73%) and 12 months (10.16%) vs the BAT arm (37.34% and 37.34%, respectively). 

Moreover, the median failure-free survival (FFS) was over 3 times longer in the ruxolitinib arm vs the BAT arm, at 4.86 months and 1.02 months, respectively (HR, 0.49; 95% CI, 0.37-0.63). The agent also resulted in fewer incidences of hematologic disease progression or relapse, non-relapse mortality, or the addition of new systemic treatments. Ruxolitinib also demonstrated a longer median event-free survival (EFS) vs BAT, at 8.18 months and 4.17 months, respectively (HR, 0.80; 95% CI, 0.60-1.08); this translated to a 20% reduction in the risk of hematologic relapse or progression, graft failure, or death because of any cause.

The risk of malignancy relapse or progression was relatively low in both cohorts. The probability of events at 6 months in the ruxolitinib arm was 8.5% vs 13.5% in the BAT arm; at 12 months, these rates were 11.1% and 15.1%, respectively. The cumulative incidence of NRM proved to have similar event rates over time in both treatment arms. The probability of events at 6 months in the investigative and control arms were 37.7% vs 42.4%, respectively; at 12 months, these rates were 43.8% and 46.5%, respectively.

"The ruxolitinib-treated patients had a longer DOR compared with [those who received] BAT,” Mohamad Mohty, MD, PhD, a professor of hematology and head of the Department of Hematology and Cellular Therapy at Saint-Antoine Hospital, University Pierre and Marie Paris, said during the presentation. “They have a longer median FFS, a longer EFS, and more pronounced improvement in health outcomes [with ruxolitinib]…This is really good news for our patients who are struggling with this severe complication post-allogenic stem cell transplantation."

In May 2019, ruxolitinib was approved by the FDA for use in patients with steroid-refractory acute GVHD, making it the only agent to receive regulatory approval for this patient population.2 The initial results of the phase 3 REACH2 study indicated that ruxolitinib demonstrated superiority over BAT;3 the agent elicited an overall response rate (ORR) of 62.3% at day 28 vs 39.4% with BAT (P <.001) in the 309 patients enrolled on the study. At day 56, the ORRs in the investigative and control arms were 39.6% and 21.9%, respectively (P <.001). At the time, no unexpected safety signals were reported with ruxolitinib and cytopenias were the most common adverse effect (AE) observed with its use.

To be eligible for participation on the phase 3 trial, patients needed to be 12 years of age or older, have steroid-refractory acute GVHD, and have progressed following 3 or more days of high-dose systemic steroid therapy with or without calcineurin inhibitors (CNIs), had a lack of response after 7 days, or had progressed while being treated with a steroid taper. Evidence of myeloid and platelet count engraftment was also required.

Participants were randomized 1:1 to receive either 10 mg of ruxolitinib twice daily and steroids with or without CNI (n = 154), or BAT and steroids with or without CNI (n = 155). Patients were stratified by acute GVHD grade. Investigators planned for follow-up at 6 months, 9 months, 12 months, 18 months, and 24 months. Notably, the study allowed for crossover to ruxolitinib. 

Key secondary end points evaluated after 6 months of follow-up included DOR, FFS, EFS, malignancy relapse/progression, non-relapse mortality (NRM), patient-reported outcomes (PROs), corticosteroid dose, chronic GVHD crossover, and safety.

Overall survival data are not yet mature but are expected to be presented during the study’s final analysis, according to Mohty.

The median age of participants was 53.25 years and the majority of patients were male. Thirty-three percent of patients had grade II disease, 44.4% had grade III disease, and 20.4% had grade IV disease. The majority (62.3%) of patients on the ruxolitinib arm had acute GVHD organ involvement in the lower gastrointestinal tract; the same was true for the BAT arm (74.2%). Moreover, 30.5% of those in the ruxolitinib arm experienced failure during the steroid taper vs 31.6% of those in the BAT arm.

"[The baseline characteristics] were very well balanced between treatment arms, especially the—and this is of importance when it comes to acute GVHD—organ involvement, but also the criteria for being steroid refractory,” Mohty noted.

In the ruxolitinib arm, treatment is ongoing in 1.9% of patients, 22.7% completed treatment, and 75.3% of patients discontinued treatment. The most common reasons for discontinuation included lack of efficacy (20.8%), AEs (17.5%), and death (16.2%). By comparison, no patients in the BAT arm are still receiving treatment, 13.5% completed treatment, and 86.5% discontinued treatment. Here, 44.5% discontinued due to lack of efficacy, 13.5 discontinued because of death, and 8.4% did so because they experienced disease relapse. Notably, 31.6% of patients from the BAT cohort crossed over to receive ruxolitinib.

Additionally, Mohty spoke to the results of a PRO analysis from the study. “Quality of life is very important when it comes to GVHD.” Mohty said. “The mean EQ-5D-5L scores improved in both treatment arms at week 24; however, this improvement was more pronounced in the ruxolitinib arm. This is in line with the data that [have been] previously described." 

Key secondary findings indicated that 22.1% of patients in the ruxolitinib arm and 14.8% of those in the BAT arm had completed steroid taper by day 56 (odds ratio, 1.63; 95% CI, 0.91-2.92). Moreover, the ruxolitinib cohort had more patients completely tapered off steroids at day 56 vs the BAT cohort by this time point.

Notably, 29.2% and 18.7% of patients in the investigative and control arms, respectively, had developed chronic GVHD, although fewer severe cases were observed in the ruxolitinib arm (4 vs 7, respectively). Patients who crossed over to the ruxolitinib arm experienced responses that were consistent with what had been reported in the primary analysis of ruxolitinib, with a ORR at crossover day 28 of 67.3% and 42.9% at crossover day 56.

Regarding safety, patients in the ruxolitinib arm (n = 152) had a median duration of exposure of 63.0 months (range, 6.0-463.0 months) vs 29.0 (range, 1.0-188.0 months) in the BAT arm (n = 150). Any-grade toxicities were reported in 99.3% of those given ruxolitinib vs 98.7% of those given BAT; 91.4% vs 87.3% of patients, respectively, experienced AEs that were grade 3 in severity. Slightly more patients in the investigative arm experienced serious AEs vs those in the control arm, at 66.4% vs 53.3%, respectively. 

Additionally, more patients who received ruxolitinib experienced toxicities that resulted in discontinuation vs those given BAT, at 27.0% vs 9.3%, respectively. Just under 55% of patients on the investigative arm experienced toxicities that led to dose modifications vs 13.3% of those on the control arm. Death rates in the investigative and control arms were 53.9% and 57.3%, respectively.

"We didn't see any new safety signals with this longer follow-up,” Mohty said. “The toxicity profile of ruxolitinib reflects the longer treatment duration compared with the control BAT arm.”

The most common all-grade AEs associated with ruxolitinib included thrombocytopenia (234.9/100 PTY), anemia (181.7/100 PTY), and neutropenia (131.3/100 PTY), with common grade 3 or higher AEs including thrombocytopenia (200.3/100 PTY), anemia (181.7/100 PTY), and neutropenia (131.3/100 PTY). The most frequently experienced any-grade AEs in the BAT arm comprised anemia (216/100 PTY), thrombocytopenia (211.9/100 PTY), and cytomegalovirus infection (169.8/100 PTY), with common grade 3 or higher AEs including thrombocytopenia (178.9/100 PTY), anemia (148.6/100 PTY), and neutropenia (115.6/100 PTY).

Other all-grade AEs associated with ruxolitinib included diarrhea (56.6/100 PTY), hypomagnesemia (53.0/100 PTY), and hypertension (48.2/100 PTY), with other grade 3 or higher AEs including white blood cell count decrease (46.2/100 PTY), sepsis (26.0/100 PTY), and hypertension (21.2/100 PTY).


  1. Mohty M. Ruxolitinib vs best available therapy in patients with steroid-refractory acute graft-vs-host-disease: 6-month follow-up from the randomized, phase 3 REACH2 study. Presented at: Virtual 47th Annual Meeting of the EBMT; March 14-17, 2021; Virtual. Abstract OS8-4. Accessed March 16, 2021.
  2. FDA approves Jakafi (ruxolitinib) for the treatment of patients with acute graft-versus-host disease. News release. Incyte. May 24, 2019. Accessed March 16, 2021.
  3. Zeiser R, von Bubnoff N, Butler J, et al. Ruxolitinib for glucocorticoid-refractory acute graft-versus-host disease. N Engl J Med. 2020;382(19):1800-1810. doi:10.1056/NEJMoa1917635
Related Videos
Michael R. Cook, MD
Andre Goy, MD
Peter Schmid, MD, PhD
Seema A. Khan, MD
Stephanie L. Graff, MD
Senthil Damodaran, MD, PhD
Jason Jincong Freeman, PhD
Heather A. Parsons, MD, MPH
Nicole Lamanna, MD, a hematologist-oncologist