P1069: RETROSPECTIVE COMPARISON OF PATIENT OUTCOMES ON PACRITINIB VERSUS RUXOLITINIB IN PATIENTS WITH MYELOFIBROSIS AND THROMBOCYTOPENIA

Abstract

Background: Pacritinib is a JAK2/IRAK1 inhibitor for patients with myelofibrosis and thrombocytopenia. Unlike the JAK1/2 inhibitor ruxolitinib, which is used at lower doses in patients with thrombocytopenia, pacritinib has been studied at full dose regardless of platelet count. Aims: We retrospectively analyzed the phase 3 PERSIST-2 study, comparing pacritinib with ruxolitinib, with a focus on disease symptomatology. Methods: PERSIST-2 enrolled patients with platelet counts ≤100 x 109/L and randomized them 1:1:1 to pacritinib 200 mg twice daily (BID), pacritinib 400 mg once daily (QD), or best available therapy (BAT). This analysis focused on the 200-mg BID dose and patients who received ruxolitinib prior to week 24. Endpoints included the percentage of patients with ≥35% spleen volume reduction (SVR), ≥50% modified total symptom score (mTSS) reduction, and improvement in disease symptoms measured by Patient Global Impression of Change (PGIC). Safety analyses were based on all treated patients; efficacy analyses were based on the intention-to-treat population randomized at least 22 weeks prior to study end. The Fisher Exact test was used to describe differences in response. Logistic regression was used to adjust for differences in baseline characteristics. Results: Safety analysis included 106 patients on pacritinib and 44 on ruxolitinib; efficacy analysis included 74 on pacritinib and 32 on ruxolitinib (median daily dose of 10 mg). Baseline characteristics were similar between the groups, including median platelet count (55 vs 61 x 109/L) and percentage receiving RBC transfusion (46% vs 43%). The following differences in baseline characteristics were accounted for in multivariable modeling: percentage of patients with grade 3 fibrosis (58% vs 36%), primary myelofibrosis (77% vs 50%), ≥1% peripheral blasts (45% vs 61%), and prior JAK2 inhibitor (48% vs 73%). Patients treated with pacritinib vs ruxolitinib achieved higher rates of SVR (22% vs 3%, p=0.02) and numerically higher rates of mTSS response (35% vs 19%; p=0.11) at week 24 (Figure 1A). A greater percentage of patients on pacritinib reported “much” or “very much” improved symptoms (35% vs 16%, p=0.06; Figure 1A). Among ruxolitinib-treated patients with an available PGIC measure at week 24, 50% reported either no improvement or worsening symptoms, while 76% of pacritinib-treated patients reported improvement (Figure 1B). After adjusting for imbalances in baseline characteristics, there was no diminution of treatment effect on SVR or mTSS, and the hazard ratio for survival on pacritinib vs ruxolitinib was 0.46 [95% CI: 0.15-1.43]. Fatal adverse events occurred at slightly lower rates on pacritinib vs ruxolitinib (8% vs 11%). Bleeding occurred at similar rates on pacritinib and ruxolitinib (43% vs 41%). There were low rates of herpes zoster reactivation (n=0 vs 1), fungal skin infection (n=0 vs 1), pulmonary aspergillosis (n=1 vs 0), deep venous thrombosis (n=0 vs 1), and pulmonary embolism (n=1 vs 0) on pacritinib and ruxolitinib, respectively. Image:Summary/Conclusion: Pacritinib yielded higher response rates and a similar safety profile to lower-dose ruxolitinib in patients with myelofibrosis who have moderate or severe thrombocytopenia.