Abstract Topic: 3. Acute myeloid leukemia - Biology & Translational Research
Background: A novel genomically defined subset of non-APL AML and MDS patients with RARA gene overexpression was identified with an actionable target for tamibarotene (formerly SY-1425), an oral and selective RARα agonist. RARA overexpression is detectable by a blood-based biomarker test. In newly diagnosed (ND) unfit AML patients with RARA overexpression, tamibarotene/azacitidine treatment is associated with a 61% composite CR rate; median time to response was 1.2 months. In previous analyses, RARA overexpression was not correlated with mutations likely to confer prognostic risk (McKeown 2017, de Botton 2022). However, molecular characterization of this novel disease subset is ongoing.
Aims: Further characterize the molecular profile of AML and MDS patients positive for RARA overexpression in a Phase 2 multicenter trial SY-1425-201 (NCT02807558).
Methods: ND unfit AML patients and patients with relapsed/refractory (R/R) AML and MDS were screened for RARA overexpression. For RARA-positive patients, IPSS-R risk category (MDS patients), and AML/MDS subtype were recorded. Pretreatment peripheral blood mononuclear cells from a subset of AML and MDS patients with RARA overexpression were analyzed by NGS for 40 somatic DNA mutations (Oncomine Myeloid Assay Gx – ThermoFisher).
Results: 115 RARA-positive patients (80 AML, 35 MDS; median age 74; 57% male) were enrolled between September 2016 and May 2020.
80 RARA-positive AML patients had clinical annotation: 56 with R/R and 24 with ND unfit AML. Overall, 5% of patients (4/80) had therapy-related AML and 6% of patients (5/80) had secondary AML. 70 enrolled RARA-positive AML patients had NGS data: 51 with R/R and 19 with ND unfit AML. Mutation frequencies for the AML cohorts were generally similar to those described in published series. Frequencies for the R/R AML cohort included: 25% CEBPA, 22% RUNX1, 22% BCOR, 18% TET2, 16% TP53, 16% DNMT3A, 14% SRSF2, 12% IDH2, 10% SF3B1, and 8% FLT, and 18 other variants in four or fewer patients each. Frequencies for the ND unfit AML cohort included: 26% SRSF2, 21% RUNX1, 16% NPM1, 11% TET2, 11% TP53, 11% IDH2, 11% PTPN11, 11% NRAS, and 8 other variants in 1 patient each.
Mutation data were available for 6 patients with RARA-positive R/R MDS (3 IPSS-R high and 3 very high-risk disease). Pathogenic variants included mutations frequently identified in large published MDS datasets: DNMT3A (3 patients), SF3B1 (2 patients), U2AF1 (2 patients), ASXL1 (1 patient), ETV6 (1 patient), NRAS (1 patient), and STAG2 (1 patient).
Summary/Conclusion: AML and MDS with RARA overexpression present with molecular features similar to those described in published datasets. Notably, SRSF2 was the most common mutation in the ND older unfit AML cohort, consistent with published series. While the small number of RARA-positive MDS patients precludes definitive characterization of molecular profiles, data in AML patients suggest that RARA overexpression may be agnostic to known molecular features. Further investigation correlating genetic and mutational features in AML and MDS patients positive for RARA overexpression is ongoing in the Phase 2 SELECT-AML-1 trial of tamibarotene/venetoclax/azacitidine and Phase 3 SELECT-MDS-1 trial of tamibarotene/azacitidine, respectively.
Keywords: Acute myeloid leukemia, Myelodysplastic syndrome, Mutation analysis