Search for a command to run...
Abstract Introduction: Acute myeloid leukemia (AML) with FLT3-internal tandem duplication (ITD) mutations represent ~25% of all AML cases and are associated with high relapse rates, even after allogeneic hematopoietic cell transplantation (HCT). A common strategy to mitigate relapse is the use of FLT3-ITD tyrosine kinase inhibitor maintenance (TKIm) therapy post-HCT. Recently, the MORPHO trial found an improvement in relapse-free survival (RFS) in patients (pts) with pre- or post-HCT detectable measurable residual disease (MRD) treated with post-HCT gilteritinib maintenance. Real-world data evaluating this approach remain limited. Herein, we report our experience with FLT3i maintenance at the University of California, San Francisco. Methods: We conducted a retrospective analysis including pts with FLT3-ITD mutated AML who underwent their first HCT between January 2017 and June 2024 while in a complete remission (CR) or a morphologic leukemia free state with a known MRD status at the time of HCT. MRD-negativity (MRD-) was assessed either by multiparameter flow cytometry (<0.01%) or FLT3 next generation sequencing assessment (allelic sensitivity 5 x 10-5). The primary endpoint was 3-year overall survival (OS), while secondary endpoints included 3-year relapse-free survival (RFS), TKIm completion rate as per treating physician, and incidence of TKIm discontinuation. Baseline characteristics and toxicity data were summarized with descriptive statistics and survival analyses were performed via the Kaplan Meier method. Results: In total, 58 pts were included, of whom 32 (55%) received TKIm post-HCT. The median age at HCT was 54.5 years (range, 19-79), 32 (55%) were female, and 4 (7%) had secondary AML. Forty-seven (81%) patients were in CR1, 4 (7%) were in CR2, and 41 (71%) were MRD- pre-HCT. Thirty-nine (67%) received myeloablative conditioning (MAC), 56 (97%) received peripheral blood stem cells, and 39 (67%) had an 8/8 matched unrelated or sibling donor. Fourteen (44%) received gilteritinib, 13 (41%) received sorafenib, and 5 (16%) received midostaurin as their first TKI post-HCT. TKIm was started at a median of 83.5 days (range, 38-412) post-HCT. The median follow-up time for the entire cohort was 3.76 years (IQR, 2.18-6.62). Among pts who received TKIm, 24 (75%) received MAC while 15 (58%) received MAC in the no TKIm group (p= 0.26). Among pts who were MRD-, 22 pts (54%) received TKIm, 18 (82%) of whom received MAC, while 19 (46%) did not receive TKIm, 14 (74%) of whom received MAC (p-value for MAC between TKI and no TKI groups: 0.71). Of the 17 pts who were MRD+, 10 (59%) pts received TKIm, 6 (60%) of whom received MAC, while 7 (41%) did not receive TKIm, of whom only 1 (14%) received MAC (p-value for MAC between TKI and no TKI groups: 0.13). The 3-year OS and RFS for the entire cohort was 69.7% (95% CI, 58.5-82.9) and 60.1% (48.2-74.9), respectively. Among pts who were MRD-, the 3-year OS in pts who received TKIm vs no TKIm was 95.5% (87.1-100) vs 66.2% (47.3-92.8) (p=0.046), while the 3-year RFS was 80.7% (65.4-99.8) vs 73.7% (56.3-96.4) (p=0.2), respectively. Among pts who were MRD+ pre-HCT, the 3-year OS in those who received TKIm vs no TKIm was 46.7% (19-100) vs 14.3% (2.33-87.7) (p=0.027), while the 1-year RFS was 50% (26.9-92.9) vs 0% (p <0.001), respectively. Of the 13 pts who received sorafenib, 3 (23%) switched to a second TKI at a median of 105 days while 1 pt each (20% and 7.7%) initially treated with midostaurin or gilteritinib switched to a 2nd TKI at 370 and 54 days, respectively. The median cumulative exposure of TKIm was 312 days (range, 28-1393). For gilteritinib, sorafenib, and midostuarin as the first TKI, the median cumulative exposures for each TKI were 250 days (range, 28-1393), 241 days (range, 28-886), and 363 days (range, 128-676), respectively. Twelve (38%) completed predetermined TKIm and 5 (16%) are receiving ongoing TKIm at the time of study cutoff. Reasons for discontinuation include: cytopenias (n=4, 13%), gastrointestinal intolerance (n=4, 13%), lack of insurance (n=2, 6%), relapse (n=2, 6%), graft-vs-host disease (n=1, 3%), peripheral neuropathy (n=1, 3%), and transaminitis (n=1, 3%). Conclusions: Despite limitations, including limited sample size, we observed an improvement in OS in patients who received post-HCT TKIm, irrespective of pre-HCT MRD status and conditioning regimen intensity. Consistent TKIm post-HCT is challenging due to the high discontinuation rate from toxicity.