Incidence, Characteristics, and Outcomes of Patients with Newly Diagnosed, De Novo KIT-Mutated Non-Core Binding Factor Acute Myeloid Leukemia without Systemic Mastocytosis

Background

Mutations in the receptor tyrosine kinase KIT (located on chromosome 4q12) have been reported in systemic mastocytosis (SM, in over 90% of cases), gastrointestinal stromal tumors, melanoma, and acute myeloid leukemia (AML). In AML, KIT mutations are enriched in patients (pts) with the core-binding factor (CBF) abnormality (20-40% of pts). These mutations can occur in various exons throughout the KIT gene, with "hot spots" identified in exons 8, 10-11, and 17. This study aims to characterize the baseline features, molecular patterns, and survival outcomes in pts with newly diagnosed non-CBF AML that is not associated with SM, a population that has not previously been well-described.

Methods

A retrospective analysis was conducted on newly diagnosed adults with AML treated with first-line therapy between 2014 and 2024 at our institution to identify pts with KIT-mutated AML. Pts with wild-type KIT or KIT variants of unknown significance were excluded from the study. Among patients with KIT mutations, those with CBF AML, evidence of underlying SM, or history of antecedent hematological neoplasms (AHN) were also excluded.

Results

Among 3,240 newly diagnosed AML pts, 77 carried a pathogenic/likely pathogenic KIT mutation. After excluding patients with CBF AML (n=46), those with underling SM (n=5), or AHN (n=10), 16 patients remained in the analysis, resulting in an overall incidence of 0.5% in patients with newly diagnosed non-CBF, non-SM de novo AML. The median age of the cohort was 69 years (range: 36-90). The median KIT variant allelic frequency (VAF) was 15% (range, 5%-48%). All patients had a single missense point mutation in exon 17 of KIT, with no involvement of other exons. Specifically, the D816V mutation was present in 14 pts (88%), the D816H mutation in one pt, and the N822K mutation in one pt. In contrast, pts with KIT-mutated CBF AML had a wider distribution of KIT mutations, with 52% occurring outside of exon 17 and 5% of pts harboring more than one KIT mutation. NPM1 was the most common co-mutation, detected in 38% of patients. IDH1 or IDH2 mutations were detected in 31% (5/16) of the cohort, a higher frequency than expected in the general AML population. 31% of pts presented with either trisomy 4 or trisomy 8, and one pts exhibited a complex karyotype. 7 pts (44%) were treated with intensive chemotherapy, and nine patients (56%) received low-intensity chemotherapy, with or without venetoclax.

12 pts (75%) achieved either CR/CRi following first-line treatment. There was no correlation between KIT VAF and response rates. Despite a relatively high remission rate, the OS was poor, with a median OS of 9.0 months and a 12-month OS rate of 29%. The median RFS was 3.2 months and the 12-months RFS rate was 11%. Poor outcomes were observed in both intensively and non-intensively treated pts (median OS of 15 months and 5 months, respectively). OS was poor regardless of NPM1 mutation status, with a median OS of 9.2 months for those with mutated NPM1 compared to 10.7 months for pts with wild-type NPM1 (p=0.7). Overall, 7 patients (44%) relapsed. Among the 5 re-tested for KIT at relapse, only 1 showed an increase in KIT VAF, suggesting KIT mutations are unlikely to be the primary factor driving relapse in this cohort.

Conclusion

This is the first report specifically evaluating the incidence and clinical impact of KIT mutations in newly diagnosed non-CBF, non-SM AML. KIT mutations were very rare in this population and appear to exhibit a poor prognosis, regardless of co-mutation status. KIT missense point mutations in this population were exclusively located in exon 17, unlike what has been observed in CBF AML. Larger studies are needed to confirm the prognostic impact of KIT mutations in AML.