Correlation of ELN-Risk and Gene Mutations with Quizartinib Efficacy in Patients with FLT3-ITD–Negative Newly Diagnosed Acute Myeloid Leukemia in the Phase 2 Quiwi Pethema Trial

Background: Quizartinib (Quiz) is a potent type II inhibitor showing clinical activity in FLT3-ITD negative acute myeloid (AML) patients (pts), as suggested by the interim analyses of the PETHEMA randomized, double-blind, placebo (PBO)-controlled phase II QUIWI trial (NCT04107727). Here, we describe the frequency of baseline gene mutations among FLT3-ITD negative (allelic ratio by PCR <0.03) randomized subjects and evaluate their impact on OS in QUIWI final dataset.

Methods: Eligible pts were 18–70 years old, with newly diagnosed FLT3-ITD negative AML (FLT3-ITD allelic ratio <0.03, centralized analyses). Pts were randomized 2:1 to receive 3+7 induction chemotherapy with PBO or Quiz (60 mg x 14 days). A second identical induction cycle was allowed in case of failure to achieve CR/CRi after the first cycle. Consolidation (up to 4 cycles) consisted of high dose Cytarabine on Days 1, 3, and 5 plus PBO or Quiz for 14 days. Patients with high genetic risk or intermediate with MRD positivity were recommended for allo-SCT. A maintenance phase with 60 mg Quiz or PBO started after the consolidation or after allo-SCT. A blinded independent review committee revised response assessment and European Leukemia Net (ELN) risk classification (2017 and 2022). Analyses were performed on an intent-to-treat basis. Baseline mutational statuses for 26 genes relevant to AML were analyzed in bone marrow via next-generation sequencing panels in the PETHEMA central lab plataform (PLATAFO-LMA) (Quiz, n=180; placebo, n=93 pts). A gene was considered mutated if it exhibited at least one somatic mutation with a VAF of ≥ 5%. For FLT3-TKD and NPM1 mutations, conventional PCR was also used for positivity diagnosis (sensitivity 3%). The impact of Quiz according to ELN 2017 and 2022 risk categories on OS was assessed. Based on their prevalence and prognostic significance, the effect of NPM1, DNMT3A, IDH1, IDH2, FLT3-TKD, and TP53 mutations on OS was explored. Hazard ratios (HRs) comparing Quiz vs. PBO were calculated using unstratified Cox proportional hazards models. These analyses were exploratory in nature and not powered for formal hypothesis testing.

Results: At baseline, gene mutations were detected in 256/273 (93.8%) analyzed pts (all pts had available NGS results). The most common mutations were: DNMT3A (24 %), NPM1 (21 %), TET2 (19 %), IDH2 (18 %), RUNX1 (16 %), SRSF2 (13 %), TP53 (12 %), ASXL1 (10 %), FLT3-TKD (10 %), NRAS (9 %), IDH1 (9 %), CEBPa (8 %), KRAS (7 %), PTPN11 (7 %), SF3B1 (6 %), EZH2 (5 %), U2AF1 (5 %), WT1 (5 %), GATA2 (3.3 %), and SETBP1 (3.3 %). ELN 2017 risk distribution for Quiz and PBO arms was low 28.3% vs. 25.8%, intermediate 27.2% vs. 25.8%, and high 44.4% vs. 48.4%, p=0.82; and ELN 2022 risk distribution for Quiz and PBO was low 29.4% vs. 25.8%, intermediate 12.8% vs. 13.9%, and high 57.8% vs. 60.2%, p=0.8).

OS benefit with Quiz vs PBO was observed among ELN 2017 low-risk (3-years OS 90% and 71%, respectively; HR, 0.3 [0.095, 0.946]) and intermediate-risk (3-years OS 67% and 44%, respectively; HR, 0.466 [0.227, 0.956]); with no detrimental OS effect among high-risk group (HR,0.8 [0.5-1.3]). OS benefits with Quiz vs PBO was observed among ELN 2022 low-risk (3-years OS 92% and 71%, respectively; HR, 0.23 [0.067, 0.786]) and intermediate-risk (3-years OS 68% and 53%, respectively; HR, 0.53 [0.185, 1.513]); with no detrimental OS effect among high-risk group (HR, 0.7 [0.5-1.13]).

OS benefits with Quiz vs. PBO was observed across most of the assessed subgroups, including NPM1^mut (3-years OS 92% and 53%, respectively; HR, 0.123 [0.033, 0.467]) and DNMT3A^mut (3-years OS 71% and 48%, respectively; HR, 0.458 [0.205, 1.024]); FLT3-TKD ^mut (3-years OS 95% and 33%, respectively; HR, 0.057 [0.006, 0.514]); IDH2^mut (3-years OS 72% and 47%, respectively; HR, 0.503 [0.204, 1.24]). No detrimental OS effect was observed in IDH1^mut (3-years OS 65% and 62%, respectively; HR, 0.88 [0.22, 3.52]), and TP53^mut (3-years OS 16% and 0%; HR, 0.989 [0.473, 2.071]).

Conclusions: The addition of Quiz to intensive chemotherapy was associated with significant OS advantage among ELN low-intermediate risk pts. NPM1^mut and FLT3-TKD ^mut appeared to have special sensitivity to Quiz, resulting in high survival rates. The addition of Quiz vs. PBO was not associated with improved or worsened survival among ELN adverse-risk patients. These results suggest that baseline mutational status of FLT3-ITD negative pts could be relevant to predict benefit from Quiz treatment.