Firstline Treatment with Ruxolitinib Versus Best Available Therapy in Patients with Polycythemia Vera: Pre-Specified Interim Analysis of the Randomized Phase 2b Ruxobeat Clinical Trial of the German Study Group for Myeloproliferative Neoplasms (GSG-MPN)

Introduction: Polycythemia vera (PV) is characterized by trilineage blood cell expansion, PV-associated symptoms, and the risk of thromboembolic complications, progression to post-PV myelofibrosis, and acute leukemia. Cytoreductive treatment with hydroxyurea (HU) or ropeginterferon-alpha is approved in EU for the treatment of high-risk patients (pts) with PV. In addition, ruxolitinib (RUX) is approved for HU-intolerant/-resistant PV. However, RUX vs. best available therapy (BAT) has not been investigated as first-line treatment of pts with untreated PV. We hypothesized that RUX may have higher efficacy in such early-line PV pts.

Methods: The clinical trial entitled “Ruxolitinib versus Best Available Therapy in pts with high-risk PV or high-risk ET” (RuxoBEAT; NCT02577926) is a multicenter, open-label, two-arm phase-IIb trial with a target population of 190 pts in each stratum. Crossover from BAT to RUX is possible in eligible pts after 6 months. Pts with PV in the RUX arm received a starting dose of 10 mg BID and may increase their dose up to 20 mg BID. The primary endpoint is the clinicohematologic complete response (CR) rate at month 6, as defined by Barosi et al (Blood 2009), using a strict score of zero for the four symptoms by patient-reported outcome measures (PROM, Table 2) and an adjusted level alpha = 0.005 at a power of 80%. Secondary endpoints include overall response rate (CR+PR), freedom from phlebotomy, changes in blood counts, spleen size, and PROM, using p values descriptively. The interim trial data were recently reviewed by the DMC, and, due to the low likelihood of reaching the primary endpoint (CR rate improvement), the DMC has recommended that the trial should continue without recruitment of new pts. Therefore, this pre-specified interim analysis was performed after 78 (instead of 95) PV pts were randomized 1:1 to RUX vs. BAT. A maximum of 6 weeks of previous PV-directed therapy was allowed. Differences between RUX and BAT at month 6 and between baseline and end of month 6 data for each treatment were calculated using Fisher´s exact test/ McNemar Test (for unpaired/ paired binary variables) or the Mann-Whitney-U test/ Wilcoxon signed-rank test (for unpaired/ paired ordered or continuous variables).

Results: 78 first-line pts (33% were pre-treated for a median of 22 days [range 1-42]) were included in the intention-to-treat set (n=44 vs 34 randomized to RUX vs. BAT, resp.) and analyzed at the 6-month time point. Baseline characteristics are listed in Table 1. In pts on RUX or BAT, CR rates at 6 months were comparable (2.3% and 2.9%, p=1.0), while overall response (CR+PR) rates were 77.3% with RUX and 55.9% with BAT (p=0.054). Also, at month 6 (Table 2), pts on RUX showed significantly lower hematocrit (40.8% vs 42.1%, p=0.046), and PROM symptom points for pruritus (1 vs 4, p=0.001) and fatigue (2.5 vs 5, p=0.031), and a trend towards less headache, abdominal discomfort, and weight loss, while the number of platelets was lower with BAT vs. RUX (254/nl vs 386/nl, p=0.013). There was no difference between RUX and BAT in the number of white blood cells (WBC), spleen size, or other symptoms at this time point.

When each treatment arm was analyzed separately for changes between baseline and month 6, RUX treatment significantly reduced platelet counts, WBC, hemoglobin and hematocrit, phlebotomy rates, spleen size and fraction of pts with splenomegaly (Table 2), and significantly improved PROMs for pruritus, early satiety, and quality of life. During the same time period, BAT significantly reduced platelet counts, WBC, hematocrit, and phlebotomy rates, but failed to impact on hemoglobin, spleen size, fraction of pts with splenomegaly, or any of the symptoms (Table 2).

Safety analysis revealed 301 adverse events (AE) in both assigned treatment groups, including 153 AE in the RUX and 148 AE in the BAT group (11.8% and 8.1% grade ≥3, resp.). There was no significant difference in the percentage of grade ≥3 AE between both groups (p=0.358), and there were no consistent patterns of grade ≥3 AE in either of the two arms.

Conclusion: In this interim analysis, first-line treatment with ruxolitinib for 6 months in high-risk pts with PV led to clinically meaningful improvements in overall response, hemoglobin and hematocrit, phlebotomy rates, splenomegaly, and patient-reported pruritus and fatigue severity, while BAT only improved platelet counts, WBC, hematocrit, and phlebotomy rates, without having an impact on symptoms.