Efficacy and Safety of Single-Agent Quizartinib (Q), a Potent and Selective FLT3 Inhibitor (FLT3i), in Patients (pts) with FLT3-Internal Tandem Duplication (FLT3-ITD)-Mutated Relapsed/Refractory (R/R) Acute Myeloid Leukemia (AML) Enrolled in the Global, Phase 3, Randomized Controlled Quantum-R Trial

Introduction:FLT3-ITD mutations are among the most common molecular abnormalities in AML, occurring in ≈ 25% of pts. These driver mutations are associated with high leukemic burden and poor prognosis, eg, high risk of relapse, decreased response to salvage therapy, and shorter overall survival (OS). Pts with R/R FLT3-ITD AML have a worse prognosis and represent a population with high unmet medical need. Q is a once-daily, oral, highly potent and selective FLT3i shown in phase 2 trials to have promising single-agent antileukemic activity and a manageable safety profile. QuANTUM-R was the first global, phase 3, randomized controlled trial (NCT02039726) to show that an FLT3i prolonged OS compared with salvage chemotherapy (SC) in pts with R/R FLT3-ITD AML. Final efficacy and safety data from this pivotal phase 3 trial are reported.

Methods: Pts aged ≥ 18 years with FLT3-ITD AML refractory to or relapsed (duration of first remission

≤ 6 mo) after standard AML therapy, w/wo hematopoietic stem cell transplant (HSCT) were randomized 2:1 to receive Q (60 mg [30-mg lead-in]) or 1 of 3 preselected investigator's choice (IC) SC: low-dose cytarabine (LoDAC); mitoxantrone, etoposide, and intermediate-dose cytarabine (MEC); or fludarabine, cytarabine, and granulocyte-colony stimulating factor (G-CSF) with idarubicin (FLAG-IDA). Up to 2 cycles of MEC or FLAG-IDA were permitted; Q and LoDAC were given until lack of benefit, unacceptable toxicity, or HSCT. Prior therapy with midostaurin was allowed, but all other FLT3i were not. Pts receiving HSCT in the Q arm could resume Q after HSCT. Primary and secondary endpoints were OS and event-free survival (EFS), respectively. Sensitivity analyses for OS and EFS were conducted: (1) using the per-protocol set (randomized and treated patients without major protocol deviations), (2) censoring at HSCT, (3) censoring at the use of other postbaseline FLT3i (for OS only). Predefined subgroup analyses of OS were also performed. Exploratory endpoints included response rates, duration of CRc, and transplant rate.

Results: 367 pts were randomized; 245 to Q and 122 to IC SC (LoDAC, n=29; MEC, n=40; FLAG-IDA, n=53). Four pts randomized to Q and 28 pts randomized to SC did not receive therapy. Median follow-up was 23.5 mo. Six pts were still on initial Q treatment at data cutoff vs 0 in the SC arm. Treatment groups were well balanced for baseline characteristics, including age, response to prior therapy, transplant history, and FLT3-ITD allelic burden.

OS hazard ratio (HR) of Q relative to SC was 0.76 (95% CI, 0.58-0.98; stratified log-rank test, 1-sided P=0.0177). Median OS was 6.2 (95% CI, 5.3-7.2) vs 4.7 (95% CI, 4.0-5.5) mo, with an estimated 12-mo OS probability of 27% vs 20% in Q and SC arms, respectively. EFS HR was 0.90 (95% CI, 0.70-1.16; stratified log-rank test, 1-sided P=0.1071); median EFS was 6.0 (95% CI, 0.1-8.3) vs 3.7 (95% CI, 0.4-5.9) wk, respectively. Sensitivity analyses of OS and EFS all supported benefit of quizartinib compared with SC, as did OS analyses across subgroups, including varying allelic ratio, prior HSCT, AML risk score, and response to prior therapy (Tables 1 and 2, Figure).

CRc was 48% (95% CI, 42%-55%) and 27% (95% CI, 19%-36%) in Q and SC arms (nominal P=0.0001), respectively. Duration of CRc was 12.1 (95% CI, 10.4-27.1) vs 5.0 (95% CI, 3.3-12.6) wk. Transplant rate was 32% and 12% in Q and SC arms (nominal P<0.0001), respectively; of 79 eligible pts, 49 (62%) resumed single-agent Q after HSCT (15 ongoing Q treatment at data cutoff). Median duration of post-HSCT Q was 129 d.

Rates of treatment-emergent adverse events (TEAEs) were comparable between the 2 arms, despite higher total drug exposure in Q vs SC arms (101.9 vs 3.7 patient-years [pt-y], respectively). Exposure-adjusted TEAEs were 2.3 vs 25.2 per pt-y, respectively. Most common grade ≥ 3 TEAEs in both arms were infections and those associated with cytopenia. Only 2 pts discontinued Q due to QTcF prolongation. QTcF >500 ms (grade 3) by central laboratory was 3% in the Q arm; no grade 4 QTcF occurred. Q-treated pts post-HSCT had a similar AE profile to those overall.

Conclusions: This report confirms the survival benefit observed with single-agent Q compared with SC in pts with R/R FLT3-ITD AML and the favorable Q safety profile, providing evidence of meaningful clinical benefit in pts who have few options. These results are paradigm changing in the R/R FLT3-ITD AML treatment setting.

View largeDownload slide

View largeDownload slide

Close modal