Real-world effectiveness of gilteritinib beyond second-line therapy in relapsed/refractory FLT3-mutated acute myeloid leukemia: A european multicenter study of 171 patients

Abstract Background: Gilteritinib is a selective FLT3 inhibitor approved for the treatment of relapsed or refractory (R/R) FLT3-mutated acute myeloid leukemia (AML) following at least one prior line of therapy. However, evidence regarding its efficacy in later-line settings remains limited. Methods: We conducted a multicenter, retrospective cohort study involving 171 adult patients with R/R FLT3-mutated AML who received gilteritinib as third-line therapy or beyond between August 2017 and March 2024 across centers in Italy, Spain (PATHEMA group), the United Kingdom (MRC group), and Turkey. The primary endpoint was overall survival (OS). Secondary endpoints included overall response rate (ORR) defined as the proportion of patients achieving complete remission (CR), CR with incomplete hematologicrecovery (CRi), morphological leukemia-free state (MLFS), or partial response (PR), composite complete remission (cCR) rate defined as the sum of CR and CRi, duration of response, and safety. Results: The median age was 54.9 years (range, 22.7–77.6), with 30% of patients aged >65 years. FLT3-ITD mutations were identified in 144 patients (84%), and 115 (67%) had FLT3 mutations confirmed at initial diagnosis. At the time of gilteritinib initiation, 110 patients (64%) had relapsed disease, while 61 (36%) had primary refractory AML. Prior exposure to FLT3 inhibitors (FLT3i) and venetoclax (VEN) was reported in 98 (57%) and 59 (35%) patients, respectively. A total of 42 patients (26%) had received ≥3 prior lines of therapy; this subgroup was younger (median age 50.3 vs. 61.2 years) and more frequently exposed to multiple FLT3i (10%) or gemtuzumab ozogamicin (24%). Regarding first-line treatment, 36% received intensive chemotherapy (IC), 48% IC plus midostaurin, 7% hypomethylating agents (HMAs)-VEN, 5% HMAs alone, and 5% underwent upfront allogeneic hematopoietic stem cell transplant (HSCT). For second-line therapy, 61% received IC, 11% IC plus a FLT3i, 25% HMA-VEN, and 3% proceeded to HSCT. The ORR and cCR rates were 47% and 28%, respectively. Patients received a median of 5 gilteritinib cycles (range, 1–22), with a median treatment duration of 3.04 months (95% CI, 0.19–20.8). Treatment discontinuation was primarily due to refractory disease (36%), death (15%), toxicity (10%), or transition to HSCT (12%). Favorable outcomes were observed in younger patients and those with relapsed rather than refractory disease. Among cCR responders, the median time to best response was 1.52 months (95% CI, 0.59–6.16), with 77% receiving ≥4 cycles. The median duration of response was 9.62 months (95% CI, 1.04–26.03) for CR/CRi and 5.75 months (95% CI, 0.94–12.17) for PR. On multivariate logistic regression, prior exposure to VEN (p=0.023; 95% CI, 0.81–5.95) and prior use of FLT3i (p=0.041; 95% CI, 1.04–6.61) were independently associated with reduced response probability. Gilteritinib enabled HSCT in 12% of patients, who experienced significantly longer median OS (16.3 months; 95% CI, 8.75–NR) compared to those not undergoing HSCT (6.5 months; 95% CI, 5.6–9.02; p<0.001). The median OS was 7.1 months (95% CI, 5.9–10.1), with 12- and 18-month OS rates of 25% and 12%, respectively. Responders exhibited significantly prolonged survival, with 12-month OS rates of 59% and 54% in patients achieving CR and CRi, respectively. In univariate analyses, age, sex, FLT3 mutation subtype (ITD vs. TKD), number of prior therapies, prior FLT3i exposure, or acquisition of FLT3 mutation at relapse were not significantly associated with OS. Conversely, prior VEN exposure was associated with inferior survival (median OS 5.7 vs. 9.0 months; p=0.021). In multivariate Cox regression, no baseline clinical or molecular features were independently associated with OS, apart from achievement of CR/CRi and subsequent HSCT. The median event-free survival (EFS) was 3.1 months (95% CI, 2.8–3.7), and neither prior FLT3i nor VEN exposure significantly affected EFS. Conclusions: Despite substantial prior treatment exposure, gilteritinib demonstrated clinically meaningful activity in heavily pretreated patients with R/R FLT3-mutated AML. Its administration beyond second-line may serve as a viable bridge to HSCT in select patients. Resistance mechanisms—particularly following venetoclax exposure—pose a persistent therapeutic challenge. These findings support the continued use of gilteritinib in later-line settings and underscore the need for prospective studies to optimize therapeutic sequencing strategies.