Clinical and Prognostic Implications of WT1 Mutations in De Novo and Relapsed Acute Myeloid Leukemia

Introduction

Wilms Tumor 1 (WT1) is integral to cell proliferation and survival and is mutated in up to 10-15% of patients (pts) with acute myeloid leukemia (AML). WT1 mutated (WT1^m) AML frequently co-occurs with fms-like tyrosine kinase 3 (FLT3) and nucleophosmin 1 (NPM1) mutations, and is generally thought to impart poorer outcomes, however the impact of WT1^m remains incompletely understood.

Methods

We analyzed 307 pts with WT1^m AML, identified through next generation sequencing panels of recurrently mutated myeloid genes performed at two large academic cancer centers between 2015 - 2022. Of the 307 pts, 161 (52%) had WT1^m at AML diagnosis (ND) and 146 (48%) were identified at relapse (RR). Here we describe the baseline characteristics and outcomes in pts with WT1^m AML with emphasis on co-mutated FLT3 and NPM1.

Results

In pts with ND WT1^m AML, the median age was 59 years (range,22-87). Pts most commonly had intermediate (41%) or adverse (43%) risk disease by ELN 2022 criteria. 51% (n=82) had co-mutated FLT3 (ITD or TKD). Among pts with FLT3 mutated AML, 30% (37%) also had co-mutated NPM1. 10 (7%) pts had co-mutant NPM1^m without FLT3 ^m. Median OS in the ND cohort was 19.8 months. Variant allele frequency (VAF) ranges of WT1^m were 1-98 (median 27) and 1-97 (median 13) in the ND and RR cohorts respectively. OS was affected by mutational profile (table 1) with triple mutated NPM1/FLT3/WT1 exhibiting worse OS compared to others. 68% of ND pts received intensive chemotherapy (+/- venetoclax (VEN)) at time of diagnosis with mOS of 26.5 mos vs. 11.1 mos (p = <0.0001) for pts treated with lower intensity therapy (+/- VEN).

Improved OS was observed in pts with ELN2022 favorable risk disease (mOS not reached (NR)) and in pts receiving an allogeneic hematopoietic stem cell transplant (SCT). Pts with ND AML and ELN2022 intermediate-risk disease who underwent SCT had significantly longer OS of 47.9 mos vs. 11.8 mos (p<0.0001, figure 1). In the ND cohort, pts had significantly improved OS if they underwent SCT after achievement of CR1 compared to SCT after first relapse, mOS NR vs. 38 months (p=0.00039), respectively. Notably, the median age of pts who received SCT was 50 (22-74) with 16% having secondary (s-AML) or therapy related (t-AML). Of those who did not undergo SCT median age was 63 (23-87) and had with 28% of pts with (s- or t-AML).

Within the RR cohort, 44 (30%) pts acquired WT1 mutation post SCT. 56 (38%) were FLT3^m (25 pts with co-mutated NPM1) and 10 (7%) NPM1^m mutated (without FLT3 ^m). The mOS in the RR cohort (from time of WT1 ^m acquisition) was shorter in pts with co-mutated FLT3/WT1 without NPM1 (table 1). Pts who underwent SCT after WT1 ^macquisition had mOS of 25.7 mos (95% CI 23.8-38.7) compared to 18.3 mos (95% CI 13.2-21.2) (p=0.001) in those who did not undergo SCT. 28% of pts received intensive therapy (+/- VEN) as salvage after acquisition of WT1 ^m without difference in OS compared to those receiving low intensity therapy (66% with VEN). mOS (11.1 vs. 7.8 mos, p=0.12).

In pts ND AML and co-mutant NPM1/WT1, WT1 ^m was frequently (75%) subclonal to NPM1 ^m with a mOS NR. However, in pts who initially had NPM1^m disease and acquired a WT1 ^m at relapse, WT1 ^m was more clonally co-dominant with NPM1 ^m with a mOS of 15.9 mos (p=0.11). Amongst both cohorts, there was no difference in survival based on sex.

Conclusion

WT1 mutations are enriched in intermediate and adverse risk AML, frequently acquired at relapse, and are associated with overall poor OS in the absence of SCT. Our data suggests that pts with WT1^m AML benefit from high intensity induction regimens and SCT at CR1 should be considered in all WT1^m pts with ELN2022 intermediate or adverse risk disease. WT1 mutations should also be considered for inclusion into ELN risk stratification given poor overall survival without SCT.