Pacritinib in patients with high-risk myelofibrosis: Outcomes from post-hoc analyses of two Phase 3 studies Free

Introduction: Patients with high-risk myelofibrosis (MF; Dynamic International Prognostic Scoring System [DIPPS] score of 5-6) have poor prognosis and frequently require red blood cells (RBCs) transfusions, thereby increasing their risk of complications and shortening life expectancy (median overall survival [OS] of 1.5 years) compared to those with lower risk MF. The efficacy of approved and experimental treatments for MF is usually lower in higher risk patients. Pacritinib (PAC), a JAK1 sparing inhibitor of JAK2/IRAK1/ACVR1 has been shown to consistently reduce spleen size and improve MF-associated symptoms and quality of life independent of baseline degrees of cytopenias. As cytopenias correlate with advanced disease and higher risk, here we present a post-hoc analyses from two Phase 3 studies to understand efficacy of PAC versus best available therapy (BAT) in high-risk patients with MF.

Methods: This pooled analysis of PERSIST-1 (NCT01773187) and PERSIST-2 (NCT02055781) evaluated week 24 efficacy in the intention-to-treat (ITT) population with high-risk MF DIPPS score. PERSIST-2 ITT population included patients who were randomized ≥22 weeks prior to study end. The PAC treatment arms from PERSIST-1 (400 mg once daily [QD) and PERSIST-2 (400 mg QD and 200 mg twice daily [BID]) were pooled, as was the BAT group (including ruxolitinib) from each study. Outcomes included ≥35% spleen volume response (SVR35), ≥50% 6-symptom (excluding tiredness) Total Symptom Score (TSS50) response, Patient Global Impression of Change (PGIC) response (reporting symptoms as “Very Much” or “Much” improved) and transfusion independence response (TI-R) rates. TI-R was assessed among patients who required RBC transfusion at baseline (within the prior 90 days) and defined as the absence of RBC transfusions over any 12-week period through 24 weeks (Gale criteria). A subgroup of patients with baseline platelets ≥50x10⁹/L was also analyzed.

Results: This analysis included 77 patients randomized to PAC and 37 to BAT, including 10 receiving ruxolitinib. Age, gender, spleen length, platelets and hemoglobin at baseline were similar between groups. Over 50% of patients in both groups were JAK-inhibitor naïve.

The proportion of patients who achieved SVR35 was significantly higher with PAC (17%) compared to BAT (3%; P<0.0340). In the PAC group, 25% achieved TSS50 response vs 13% in the BAT group (P=NS). The proportion of patients who achieved PGIC response was 29% on PAC vs 8% on BAT (P=0.0152). Among the 56 patients on PAC and 31 on BAT who received RBC transfusions at baseline, 16% of patients on PAC and 10% on BAT achieved TI-R, and 27% of patients on PAC and 16% on BAT had ≥50% transfusion burden reduction (P=NS).

In patients with baseline platelet count ≥50x10⁹/L (n=41, PAC; n=16, BAT), the PAC group had consistently numerically higher response across the endpoints: SVR35 (15% vs 0%), TSS50 (26% vs 17%), and PGIC (27% vs 6%) compared to the BAT group. Among the 27 patients on PAC and 12 on BAT who received RBC transfusions at baseline, 22% of patients on PAC and 0% on BAT achieved TI-R, and 30% of patients on PAC and 8% on BAT had ≥50% transfusion burden reduction.

In the safety population for PAC (n=91) vs BAT (n=40), Common Terminology Criteria for Adverse Events grade ≥3 treatment emergent adverse events reported in ≥10% included anemia (29.7% vs 25.0%) and thrombocytopenia (30.8% vs 25.0%). Median platelet count and hemoglobin remained stable from baseline to week 24 for PAC and BAT group.

Median OS was 126 weeks (95% confidence interval [CI]: 86.1, NE) for the PAC group and 62 weeks (95% CI: 43.6, NE) for the BAT group. Numerically higher survival probabilities were observed in the PAC group compared to the BAT group at the assessed timepoints. At 48 weeks, the survival probability was 77.5% for PAC and 60% for BAT.

Conclusions: Patients with high-risk MF treated with PAC had significant improvements in spleen size and symptom burden when compared to BAT. The safety profile reported for this subgroup was similar to what has already been reported for PAC in the primary analysis. The prolonged survival of PAC-treated patients in this high-risk cohort, as well as the overall efficacy and safety of PAC in this group supports PAC as an important therapeutic option for high-risk MF patients, addressing a critical unmet need in this population with historically poor outcomes.