Patient Reported Outcomes in Patients with Newly Diagnosed FLT3^mut+ acute Myeloid Leukemia Ineligible for Intensive Induction Chemotherapy from Lacewing: A Randomized Phase 3 Trial of Gilteritinib and Azacitidine Versus Azacitidine Alone

Background: The symptom burden for patients (pts) with acute myeloid leukemia (AML) is substantial, and patient reported outcomes (PROs) in pts with newly diagnosed (ND) AML ineligible for intensive induction chemotherapy (IIC) are poor. Baseline health status and conditions (ie, fatigue and physical functioning) are independent prognostic factors for overall survival and pts may benefit from improved quality of life (QoL) while undergoing treatment. To fully evaluate the impact of AML and its treatments, assessing health-related QoL and PROs has become increasingly important. Gilteritinib (GIL), an FMS-like tyrosine kinase 3 gene (FLT3) inhibitor, demonstrated efficacy/safety in pts with FLT3^mut+ relapsed/refractory AML. In the phase 3 LACEWING (NCT02752035) trial, GIL was evaluated in combination with azacitidine (AZA) in pts with ND AML with FLT3 mutations who were unable to receive IIC.

Aim/Objective: To evaluate prespecified PROs in pts with ND FLT3^mut+ AML ineligible for IIC receiving GIL+AZA or AZA in LACEWING.

Methods: PROs were obtained from pts in a safety cohort receiving GIL plus AZA and pts randomized (2:1) to GIL (120 mg/day orally on Days 1-28) plus AZA (75 mg/m ²/day SC/IV on Days 1-7) or AZA alone (same regimen) during 28-day cycles. All pts receiving GIL plus AZA (safety cohort and randomized) were included in the 'GIL+AZA' group. Patient reported fatigue, assessed by the Brief Fatigue Inventory (BFI), was a secondary endpoint; other PROs, assessed by the Functional Assessment of Chronic Illness Therapy-Dyspnea Short Form (FACIT-Dys-SF), Functional Assessment of Cancer Therapy-Leukemia (FACT-Leu), dizziness and mouth sore symptoms, and EuroQol Group 5-Dimension 5-Level (EQ-5D-5L), were exploratory endpoints. The BFI was administered at Cycle 1 (Days 1, 8, and 15), Cycle 2 (Days 1 and 15), and subsequent cycles on Day 1; all other assessments were administered on Day 1 of each cycle. All assessments were evaluated using analysis of variance models.

Results: A total of 15 pts in the safety cohort and 123 randomized pts (GIL+AZA, n=74; AZA, n=49) were included. Median age was 78 years with GIL+AZA and 76 years with AZA; ECOG PS ≥2 was 50.6% and 32.7%, respectively, and FLT3-ITD was the only FLT3 mutation present in 76.4% and 81.6% of pts, respectively. Questionnaire completion rates for the BFI were 73% and 85% for patients in the safety and randomized cohorts, respectively. Baseline scores on assessment scales were similar between groups. A difference in favor of GIL+AZA was seen at Cycle 1, Day 8 on the BFI (least squares mean [LSM] difference [95% CI] -1.0 [-1.9, -0.2]; P=.019); differences were not significant at other time points. On the FACIT-Dys-SF dyspnea subscale score, pts receiving AZA reported improvement in dyspnea compared with pts receiving GIL+AZA (LSM difference [95% CI] 7.7 [0.5, 14.8]; P=.037) at Cycle 3, Day 1. No significant differences were seen at other time points on the FACIT-Dys-SF dyspnea subscale score or at any time for the functional limitation subscale score. On the FACT-Leu at Cycle 2, Day 1, LSM differences in favor of GIL+AZA were observed for physical and emotional well-being but were in favor of AZA on social well-being; at most time points, well-being did not differ between groups. At Cycle 2, Day 1, LSM differences favored GIL+AZA (8.1 [95% CI 0.5, 15.7]; P=.038) on the FACT-Leu Trial Outcome Index. No differences were observed between groups on other measures of the FACT-Leu. At baseline, very little dizziness (mean scores: GIL+AZA, 0.49 vs AZA, 0.63) and mouth sores (mean scores: GIL+AZA, 0.47 vs AZA, 0.48) were reported; no significant differences were observed at any time point while on treatment. A small but significant difference was seen in favor of GIL+AZA at Cycle 2, Day 1 on the EQ-5D-5L utility index (LSM difference [95% CI] 0.1 [0.0, 0.2]; P=.049), but not at other time points. No differences were observed between groups on the EQ-5D-5L visual analog scale scores measuring self-rated health status.

Conclusions: Results from analyses of PRO data indicate that the addition of GIL to AZA for the treatment of patients with ND FLT3^mut+ AML ineligible for IIC does not result in clinically significant change or worsening of fatigue, leukemia-related symptoms (including dizziness and mouth sores), and QoL compared with AZA treatment alone. From the patient's perspective, combination therapy with GIL+AZA is a feasible and well tolerated treatment strategy.