Abstract 4165: Stable RUNX1p.(Leu56Ser) kinetics during the course of myeloid neoplasms disagree with putative RUNX1 poor prognosis status

Abstract

Abstract ‘De novo Acute Myeloid Leukemia (AML) with mutated RUNX1' is a provisional WHO entity with poor prognosis. Missense variant, RUNX1c.167T>C, NP_001745.2:p.(Leu56Ser), with COSMIC reference and FATHMM score of 0.90 suggests it is ‘Pathogenic'; while dbSNP reference and population frequencies of >1.5% suggests ‘Benign'. To address this discrepancy, we studied RUNX1p.(L56S) kinetics in patients with primary or secondary Myeloid neoplasms receiving varied chemotherapy regimen. Bone marrow (BM) samples received were subjected to routine diagnostic algorithm including targeted next-generation sequencing. ¶Out of 227 patients with RUNX1 mutated Myeloid neoplasms, 75 (33%) had RUNX1p.(L56S) making it the most frequent RUNX1 variant (Table-1). RUNX1p.(L56S) was present at ≈55% (46.6%-62.4%) Variant Allele Frequency (VAF) in all cases except one where VAF was 99%. Paired T-test (95% CI) analysis of RUNX1p.(L56S) VAF at diagnosis and remission (p-value=0.89) or relapse/AML transformation (p-value=0.7017) showed no significant difference; similar to that of single nucleotide polymorphisms (SNPs) identified in these cases (remission samples, p-value=0.15; disease evolution, p-value=0.79). In contrast, VAF of other coexisting gene variants was significantly different between diagnostic and remission (p-value=0.02) or relapse/transformation (p-value=0.01) samples, largely owing to acquisition or loss of pathogenic variants. RUNX1p.(L56S) VAF remained stable in sequential samples despite variable BM blasts; nor was it significantly different amongst patients with adverse risk cytogenetics versus those with favourable or intermediate risk (p-value=0.60). In the MDS cohort, RUNX1p.(L56S) was not associated with morphological subtypes. ¶Our data demonstrates RUNX1p.(Leu56Ser) kinetics remain stable during Myeloid disease course, similar to co-occurring SNPs. Our data strongly recommend considering clinical relevance of individual variant when assessing gene mutation-based disease entities. Table-1:Patient characteristics [includes only those with RUNX1p.(Leu56Ser)]AML (1o or 2o) [No./Median (Range)]MDS (1o or 2o)/MPN [No./Median (Range)]Fisher Exact (2-tailed; 95% CI)No. of patients1758No. of samples76105Genderp-value=0.27Male631Female1127Age (years)57.5 (13.6- 71.4)65 (20.8- 88.2)Cytogenetic risk groupp-value=1.00Favourable228Intermediate1215Adverse33Unavailable012Clinical outcomep-value= 0.24No. of patients in remission (immunophenotypic)79No. of patients with relapse/transformation63Time to event (months)21.0 (8-68)24.0 (10-49) Citation Format: Lena Rai, Sai Phyo, Elisabeth Nacheva. Stable RUNX1p.(Leu56Ser) kinetics during the course of myeloid neoplasms disagree with putative RUNX1 poor prognosis status [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 4165.