Gilteritinib Remains Clinically Active in Relapsed/Refractory FLT3 Mutated AML Previously Treated with FLT3 inhibitors

Background: Gilteritinib is approved for the treatment of relapsed/refractory (R/R) AML and FLT3-mutation (FLT3^mut+). However, the gilteritinib phase 3 ADMIRAL study (Perl et al NEJM 2019) enrolled prior to widespread adoption of either midostaurin as a component of standard intensive induction and consolidation or post-transplant FLT3inhibitor (FLT3i) maintenance. Some mechanisms of drug resistance can be shared across FLT3i’s, suggesting response to gilteritinib might differ in patients treated with frontline FLT3i. A better understanding of how prior therapy modulates response to gilteritinib is necessary to clarify this novel agent’s role in the current FLT3-mutated AML treatment algorithm.

Methods: This is an ongoing multi-institutional analysis from 13 US centers identifying patients who received gilteritinib alone or as combination therapy for the treatment of R/R FLT3^mut+ AML. Patients who received gilteritinib as a part of an ongoing trial were excluded. Response criteria were identical to the ADMIRAL trial. For patients with available data and a composite complete remission (CRc), we defined clinically measurable residual disease (cMRD) negative status by bone marrow flow cytometry using a cutoff of <1 x 10x10⁵ cells as well as polymerase chain reaction (PCR) for FLT3 mutation with a minimum sensitivity of 5%. Survival from the time of gilteritinib initiation was recorded. Multivariate analysis included all variables collected to determine interaction with patient outcome (CR and survival data). Kaplan-Meier curves and log rank test were used for survival analysis after gilteritinib initiation.

Results: 72 patients treated with prior FLT3i exposure received gilteritinib for treatment of R/R FLT3^mut+ AML. Patient characteristics are presented in table-1 with 46 (64%) previously receiving midostaurin, 19 (26%) sorafenib, and 7 (10%) other FLT3i. 8 (11%) received more than one prior TKI. NGS at diagnosis were available in 66 patients (92%) and co-mutations in DNMT3A, NPM1 and NRAS were observed in more than 10% of patients. Average duration of gilteritinib therapy was 5.7 months (range: 0.2-25 months). 27 (37.5%) received stem cell transplant (SCT) before gilteritinib and 15 (21%) underwent SCT after gilteritinib. The composite CR rate (CRc, defined as CR + CRi + CRp) was 51.4% (n= 37 patients). With regard to specific FLT3i’s, we found no significant difference in median survival (7.1 months and 6.4 months for midostaurin and sorafenib, respectively) or remission rates (CRc 54% and 47% for prior midostaurin and sorafenib, respectively). The CRc rate for patients who received only prior 7+3 and midostaurin with or without consolidation was 58% with a median survival of 7.8 months. A trend toward higher CRc rate was noted in patients treated with gilteritinib in combination regimens rather than as a single agent (64% vs 43%, respectively, p=0.09 using Chi-square test). No survival advantage for combination therapy was seen over single agent. Survival was longest in patients who obtained a CR, particularly a cMRD negative response; this remained significant after censoring at the time of SCT (figures 1&2). Patients who received SCT after gilteritinib had significantly longer survival than non-transplanted patients (9.3 months vs 5.6 months, respectively, HR 0.44, 95% CI 0.2-0.8)(figure-3). Patients with concurrent mutations of NPM1/DNMT3A had a trend toward a higher CRc compared to FLT3 mutation alone (71% vs 50%, p= 0.2) but similar survival. With regard to mutations associated with drug resistance to other FLT3i’s, patients with both FLT3-ITD and FLT3-D835 mutation had similar response and survival to FLT3-ITD alone. However, patients with MAPK pathway activating mutations (e.g. NRAS and PTPN11) had lower CRc (37.5% vs 59.5%) and poorer median survival than other patients (3.3 months vs 7.5 months) (HR 2.2- 95% CI 1.1-4.4) p value <0.01(figure-4).

Conclusions: In this multi-center, retrospective analysis, gilteritinib remains a clinically active agent after treatment failure of prior FLT3i’s (midostaurin or sorafenib). Mutations activating MAPK pathway have been implicated in secondary gilteritinib resistance but are seldom co-mutated at initial FLT3^mut+ AML diagnosis. While further study is required, we hypothesize that their detection after frontline FLT3i therapy may represent treatment-emergent clones with propensity for pan-FLT3i resistance.