Background: KIT mutation, which was detected in 4%-6% de novo acute myeloid leukemia (AML), most frequently occurred in core binding factor AML (CBF AML), definitely associated with unfavorable prognosis. However, less attention has been given to CBF-negative (CBF-neg) AML and there is still a lack of understanding regarding the distribution and prognosis of KIT mutation in the specific population.

Methods: Weconducted a retrospective study in our single center and identified non-M3 de novo AML with KIT mutations by next-generation sequencing (NGS) from January 2018 to June 2024. Core binding factor (CBF) AML or patients with underlying systemic mastocytosis were excluded. The clinical data of KIT mut/CBF-neg AML was collected and analyzed.

Results: From January 2018 to June 2024, a total of 45 patients were enrolled in the cohort, including 3 patients with secondary AML. Twenty-six patients (26/45, 57.8%) exhibited normal karyotype, six patients (6/45, 13.3%) exhibited MRC and three patients (3/45, 6.7%) presented t(v;11q23). Thirty-seven patients were available for analysis of ELN 2022 risk stratification at diagnosis: 29.7%, 43.2% and 27% in favorable, intermediate, and adverse, respectively.

The median variant allele frequency (VAF) of KIT mutation was 41.2% and the most frequent comuations were CEBPA (27/45, 60%), WT1 (12/45, 26.7%) and NRAS (11/45, 24.4%). The proportion of male was significantly higher in patients with exon 17 (76.9% vs. 40.0%, p=0.04). Meanwhile, patients with KIT exon 17 showed a significantly lower median variant allele frequency (VAF) (29.1% vs. 48.3%, p=0.0005). Additionally, the frequency of several somatic mutations was relatively higher in the subgroup of patients with exon 17, including CEBPA, WT1, NRAS, CSF3R and NPM1, among which the incidence of CEBPA and NRAS was significantly higher in patients with exon 17 (CEBPA: 40.0% vs 11.1%, p=0.0257; NRAS: 22.2% vs 0.0%, p=0.006).

Forty-three patients received at least one cycle of induction therapy and available for response assessment: 27.9% (12/43) patients received non-intensive therapy (NIT), including ten patients receiving venetoclax and HMA, and the others (31/43, 72.1%) received intensive therapy (IST), among which seventeen, thirteen and one patients received standard regimen, priming with G-CSF and venetoclax in combination with standard regimen. The rate of complete remission (CR) after induction therapy was comparable between patients with NIT and IST (83.3%, 10/12 vs 80.6%, 25/31). NIT, primarily venetoclax + hypomethylating agents (HMA) presented a comparable remission rate, a rapid MRD elimination and a superior survival compared with intensive therapy (IST).

After a median follow-up of 14.9 months, the median event-free survival (EFS) and overall survival (OS) of the cohort were 15.3 months and 26 months, respectively. Patients with KIT exon 17 mutations tended to harbor an inferior EFS and OS, significantly shortened relapse-free survival (RFS). Cox regression identified TP53 as an independent risk factor and allo-HSCT as an independent protective factor for EFS and OS.

Conclusion:In conclusion, our results indicated that KIT mut/CBF-neg AML was a specific subgroup, which might be overlooked in our clinical practice, and differed from CBF AML in the respect of clinical characteristics and molecular biological features and harbored a dismal prognosis. Induction with venetoclax+HMA might bring benefits for those patients, which might induce an early MRD elimination and remission, subsequently bridging to allo-HSCT and attain a sustainable remission.

This content is only available as a PDF.
2025
Sign in via your Institution