The Landscape of RUNX1 Mutations in Acute Myeloid Leukemia: Investigations On Stability of Mutations At Relapse and Utility As a Marker for Minimal Residual Disease

Abstract

AbstractAbstract 657 Introduction:RUNX1 mutations constitute disease-defining aberrations in acute myeloid leukemia (AML) and were demonstrated to be particularly frequent in secondary and de novo AML with normal karyotype or non-complex alterations and to confer an unfavorable prognosis. Monitoring minimal residual disease (MRD) in AML has been shown to provide prognostic information and is increasingly used for treatment decisions. A variety of molecular markers has been identified suitable for MRD assessment, yet there still is a lack of such markers in a significant number of patients. The use of RUNX1 mutations may bridge a gap. Aims:1.Evaluate the stability of RUNX1 mutations between diagnosis and relapse.2.Test the utility and clinical relevance of RUNX1 mutations as MRD marker. Patients and Methods:RUNX1 mutation screening was prospectively performed in 814 patients with AML at diagnosis (645 de novo, 109 s-AML, and 60 t-AML). The median age of the patients was 69.6 years (range: 1 – 93 years), including 375 female and 439 male patients, respectively. 50.5% (411/814) of cases presented with a normal karyotype, 38.8% (316/814) with non-complex cytogenetic alterations, 9.6% (78/814) with a complex aberrant karyotype, and 1.1% (9/814) with prognostically favorable cytogenetics. Mutation analysis was performed using a sensitive next-generation amplicon deep-sequencing assay (454 Life Sciences, Branford, CT). Moreover, in a subset of 44 AML patients and additional 59 retrospectively analyzed cases the prognostic impact of MRD levels of RUNX1 mutations was studied at a second time point after completion of intensive induction therapy (median sampling interval: 128 days after diagnosis; range 60 – 180 days). In these follow-up samples the RUNX1 mutations already detected at diagnosis were investigated with a higher coverage (835-fold median coverage) as compared to the diagnostic assessment (759-fold median coverage) resulting in a sensitivity level of 1%. Furthermore, in 57 patients paired samples from diagnosis and relapse were analyzed to assess the stability of RUNX1 mutations. Results:211/814 patients (25.9%) were detected to carry RUNX1 mutations. The median clone size was 39% and revealed a significant heterogeneity ranging from 2% to 96%. 73.9% (156/211) of mutated patients carried one mutation only, whereas 26.1% (55/211) harbored two (n=46) or more (n=9) mutations. In detail, the 211 patients harbored a total number of 275 alterations in RUNX1: 42.5% (117/275) frame-shift mutations, 34.9% (96/275) missense, 14.2% (39/275) nonsense, 4.4% (12/275) exon-skipping/splicing, and 4.0% (11/275) in-frame insertion/deletion alterations, respectively. Regarding MRD assessment, patients were separated according to the median MRD level (3.92%; range 0.03% - 48.00%) into “good responders” (n=78) with MRD levels below 3.92% and “poor responders” (n=25) with MRD levels above 3.92%. This resulted in significant differences in both event-free survival (median 21.4 vs 5.7 months, p<0.001) and overall survival (73.3% vs 66.1% at 2 years, p=0.016). Moreover, in 57 cases the stability of individual RUNX1 mutations was studied at the time of relapse. In 46/57 (80.7%) cases the same alterations detected at diagnosis were present at relapse, whilst in 2/57 (3.5%) cases the RUNX1 mutation from the diagnostic sample was no longer detectable at relapse. Importantly, in 7/57 (12.3%) patients novel RUNX1 mutations were detected in regions different from those affected at diagnosis. Conclusion:Next-generation deep-sequencing accurately detects and quantifies RUNX1 mutations in AML with high sensitivity. RUNX1 mutations qualify as patient-specific markers for individualized disease monitoring. Thus, the measurement of mutation load by next-generation sequencing may contribute to refine the assignment into distinct risk categories in AML. Analysis of RUNX1 mutations should be considered for the complete coding region at relapse to detect new RUNX1 mutations. Disclosures:Kohlmann:MLL Munich Leukemia Laboratory: Employment; Roche Diagnostics: Honoraria. Nadarajah:MLL Munich Leukemia Laboratory: Employment. Grossmann:MLL Munich Leukemia Laboratory: Employment. Alpermann:MLL Munich Leukemia Laboratory: Employment. Kern:MLL Munich Leukemia Laboratory: Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Equity Ownership. Schnittger:MLL Munich Leukemia Laboratory: Equity Ownership.