IntroductionTP53 mutant multi-hit MDS/AML patients have a median overall survival (OS) of 6-9 months. Despite continual improvements in allo-HCT outcomes, TP53 mutant MDS/AML patients continue to have very short long-term survival post HCT (~20%). Critically, novel pre- and post-HCT strategies are needed to improve outcomes and at Moffitt we have opted for HMA maintenance for TP53 mutant patients. Although HMA maintenance studies have been negative in all-comer populations, limited data exists on outcomes in TP53 mutant MDS/AML patients. Herein, we are reporting outcomes for all TP53 mutant MDS/AML patients who received HCT at our center and evaluated the impact of HMA maintenance.

Methods We retrospectively identified all TP53 mutant MDS/AML patients who underwent HCT at our center from May 2014 to December 2023 to allow for a minimum of 18-month survival follow-up. All patients who received HMA post-HCT were included in the maintenance group as long as it was initiated prior to overt relapse. In addition to baseline demographics, disease type (MDS vs AML), pre-HCT therapies, donor type and transplant conditioning; we captured serial pre-HCT/day +30/ day +100 bone marrow (BM) blasts, TP53 variant allele frequency (VAF), and cytogenetics. TP53 single-hit vs multi-hit status was based on the 2022 International Consensus Classification (ICC) definition. OS was calculated from the date of HCT. Multivariable analysis of OS used the Cox proportional hazards model, with clinical and molecular variables included if P value < 0.10 in univariable analysis.

Results Eighty-seven TP53 mutant MDS (n=66)/AML (n=21) patients having received HCT were identified with a median age of 64 years (30-75), 62% male (n=54) and 36% (n=31) therapy related. Median BM blasts at diagnosis were 7% (0%-75%), 86% had complex cytogenetics (68 of 79) and 91% were multi-hit TP53 by ICC (73 of 80). With a median follow up of 65 months from HCT for the entire cohort, the median OS was 18 months (95% CI 9-27) and 22% of patients were alive at 5 years. There was no difference in OS in patients treated with frontline intensive vs non-intensive chemotherapy (P=0.93) although patients refractory to first line therapy did uniformly poor (n=11; median OS 9 months from date of transplant; 9% alive at 2-year post-HCT). Additionally, there was no difference in OS based on baseline BM blast count, BM blasts prior to HCT (median 2%, range 0-10%) or conditioning intensity. Although baseline TP53 VAF >/= 23% had a trend for inferior OS (15 vs 43 months; P=0.08), this was independently predictive of OS in multivariable analysis (see below; HR 2.6; 95% CI 1.3-5.5); P=0.009). TP53 multi-hit status was strongly predictive of OS with a median OS of 14 months versus a median OS that was not reached (82% alive at data cutoff; P=0.009). TP53 VAF and/or cytogenetic clearance prior to HCT was not predictive of OS. However, abnormal cytogenetics at day + 30 (4 vs 21 months; P=0.007) and day +100 (9 vs 26 months; P=0.03) were strongly predictive of inferior outcomes.

HMA maintenance was utilized in 29% of the cohort (n=25), with median time to start of 3.1 months (1.4-7.8) and median duration of treatment of 10 months (0.13-44). HMA maintenance significantly improved OS of 32 months (95% CI 16-49) vs 10 months (95% CI 4-15; P=0.02). Type of HMA [i.e. azacitidine vs decitabine (PO or SC/IV) or HMA combination] was not further predictive of outcomes. Relapse free survival was also improved with HMA maintenance (23 vs 5 months; P=0.04). Cox regression analysis confirmed HMA maintenance improved OS independent of ICC multi-hit status, baseline cytogenetic complexity, MDS/AML diagnosis, donor type or TP53 VAF >/= 23% at diagnosis (HR 0.36; 95% CI .17-.79; P=0.01) and was confirmed on landmark analysis at 3 months (HR 0.34; 95% CI 0.15-.78; P=0.01).

Conclusions Although TP53 mutant MDS/AML patients have poor long-term OS (~20%), factors that predict outcomes include baseline TP53 VAF >/= 23% and TP53 ICC multi-hit status. Abnormal cytogenetics post-HCT strongly predict inferior OS. Utilization of post HCT HMA was independently associated with improved OS, supporting future pivotal studies in this molecular subgroup using HMA maintenance.

This content is only available as a PDF.
Sign in via your Institution