Improved overall survival in patients with advanced systemic mastocytosis treated with avapritinib versus real-world therapy based on mutation-adjusted risk score (MARS) stratification

Abstract Introduction Avapritinib (Ava) is a potent, selective KIT D816V inhibitor indicated for adult patients (pts) with advanced systemic mastocytosis (AdvSM) based on results from the PATHFINDER (NCT03580655) and EXPLORER (NCT02561988) studies. MARS is a prognostic tool classifying AdvSM pts into low-, intermediate-, and high-risk groups based on age (>60 years), anemia (hemoglobin <10 g/dL), thrombocytopenia (platelet count <100x10⁹/L), and the presence and number of SRSF2, ASXL1, or RUNX1 (S/A/R) mutations. Prior retrospective analyses of overall survival (OS) showed a hazard ratio (HR) of 0.19 for Ava vs midostaurin (Mido) in frontline (1L) therapy and 0.34 for Ava vs best available therapy (BAT) in second or later-lines (2L+) of treatment in AdvSM, regardless of MARS (Reiter et al., ASH 2024). MARS-defined intermediate- and high-risk pts face particularly poor prognoses, making it essential to evaluate treatment outcomes in this underserved population. This analysis builds on the earlier work, comparing OS between intermediate- and high-risk MARS pts treated with Ava (200 mg/day starting dose) in the PATHFINDER trial and those treated with BAT in real-world (RW) clinical practice. Methods Data from the PATHFINDER trial (median follow-up: 38.0 months) and a RW retrospective chart review study (NCT04695431) conducted at six global sites, were used to compare OS between pts treated with 1L Ava vs 1L Mido, and 2L+ Ava vs 2L+ BAT, which was predominantly Mido and cladribine (Clad). Analyses were restricted to lines of therapy (LOTs) in the combined cohorts of MARS- intermediate- or high-risk (score >1) pts. Subgroup analyses were conducted among pts with SM with an associated hematologic neoplasm (SM-AHN). Inverse probability of treatment weighting (IPTW) was used to adjust for baseline differences in key prognostic covariates; covariate balance was confirmed by standardized mean differences <0.1. IPTW-weighted Cox proportional hazards models were used to compare OS between cohorts. Results 1L analysis. Ava (n=24) vs Mido (n=43); mean age 70.3 vs 70.9 years; mean follow-up 33.1 vs 22.5 months. Before weighting, more Ava pts had anemia (75.0% vs 60.5%) and fewer had an elevated serum tryptase level at baseline (≥125 ng/mL: 62.5% vs 72.1%). Distributions of AdvSM subtypes were similar between the cohorts (SM-AHN: 79.2% vs 79.1%; Aggressive SM [ASM]: 12.5% vs 14%; Mast Cell Leukemia [MCL]: 8.3% vs 7.0%). IPTW-weighted median OS was not reached (NR) (95% confidence interval [CI]: not estimable [NE], NE) in the Ava cohort, and 26.8 months (95% CI: 14.4, 49.8) in the Mido cohort. In IPTW-weighted Cox analysis, OS was significantly longer in Ava vs Mido pts (HR [95% CI]: 0.08 [0.02, 0.29]; p<0.001). 1L SM-AHN subgroup. Ava (n=19) vs Mido (n=34); IPTW-weighted median (95% CI) OS was NR (NE, NE) in the Ava cohort, and 22.1 months (13.0, 61.5) in the Mido cohort. After adjustment, OS was significantly improved in Ava vs Mido pts (HR [95% CI]: 0.09 [0.02, 0.36]; p<0.001). 2L+ analysis. Ava (n=41) vs BAT (n=55, 71 LOTs); mean age 71.0 vs 67.3 years; mean follow-up 28.4 vs 17.6 months. Common 2L+ BAT agents included Mido (47.5%), Clad (34.4%), and hydroxyurea (8.2%). Before weighting, more Ava vs BAT LOTs had elevated serum tryptase at baseline (78.0% vs 63.4%) and received prior treatment with tyrosine kinase inhibitors (85.4% vs 46.5%). Fewer Ava vs BAT LOTs had ≥1 S/A/R mutation (56.1% vs 66.2%); distributions of AdvSM subtypes were similar (SM-AHN: 70.7% vs 70.4%; ASM: 17.1% vs 14.1%; MCL: 12.2% vs 15.5%). IPTW-weighted median (95% CI) OS was 50.2 months (50.2, NE) in the Ava cohort, and 14.8 months (13.2, 32.6) in the BAT cohort. In IPTW-weighted Cox analysis, OS was significantly longer in Ava vs BAT pts (HR [95% CI]: 0.28 [0.13, 0.61]; p=0.001). 2L+ SM-AHN subgroup. Ava (n=29) vs BAT (n=38, 50 LOTs); IPTW-weighted median (95% CI) OS was 50.2 months (23.0, NE) in the Ava cohort, and 13.2 months (7.7, 27.5) in the BAT cohort. After adjustment, OS was significantly improved in Ava vs BAT pts (HR [95% CI]: 0.31 [0.13, 0.74]; p=0.008). ConclusionsAmong the combined cohort of MARS intermediate- and high-risk AdvSM pts, avapritinib was associated with significantly improved OS compared to midostaurin in the 1L setting and BAT in the 2L+ setting, including in the SM-AHN subgroup. This analysis further characterizes the survival benefit of avapritinib, and supports treatment decisions, in this traditionally underserved patient population.