P470: NGS-BASED MINIMAL RESIDUAL DISEASE DETECTION IN PERIPHERAL BLOOD SHOWS GOOD PROGNOSTIC VALUE FOR OS AND EFS IN PATIENTS WITH ACUTE MYELOID LEUKEMIA RECRUITED IN THE UNIFY TRIAL

Abstract

Background: Assessment of measurable residual disease (MRD) to evaluate the depth of remission at the time of achieving morphological complete remission (CR) or CR with incomplete blood count recovery (CRi) by multiparameter flow cytometry (MFC) or quantitative polymerase chain reaction (qPCR) has been shown to be predictive of outcome in patients (pts) with acute myeloid leukemia (AML), and is recommended according to European LeukemiaNet guidelines. However, accurate MRD assessment is still limited by lack of suitable markers in all pts (qPCR) and/or limited specificity/sensitivity (MFC). Next generation sequencing (NGS) holds promise to overcome some of these limitations and allows versatile and sensitive MRD assessment in almost all AML pts. However, data on NGS-MRD in this setting are still limited. Aims: To further extend the experience with NGS-based MRD assessment, we prospectively collected samples to explore prognostic implications of MRD as detected by NGS at CR/CRi in the context of a randomized, placebo-controlled phase 3 study (UNIFY; NCT03512197), investigating midostaurin added to conventional 7 + 3 based chemotherapy in FLT3wt pts. Following the recommendation by an independent DMC, the UNIFY study was stopped in Sep 2019 due to futility. The current subset analysis was performed to investigate the impact of NGS-MRD results on long-term outcomes. Methods: Peripheral blood mononuclear cell (PBMC) or bone marrow mononuclear cell (BMMC) samples were collected from pts treated in the UNIFY study. NGS-MRD was performed on genomic DNA using targeted NGS-error-controlled sequencing method (Ion-Torrent S5XL) with a sensitivity between 0.1 and 0.001%. Targets for MRD detection were identified by NGS of BM samples taken at diagnosis (Archer Myeloid Panel) covering 74 genes frequently mutated in myeloid disease. MRD positivity was defined as any mutation call >0.1%, excluding mutations in genes associated with clonal hematopoiesis, ie., DNMT3A, TET2 and ASXL1. Results: Total 731 BMMC (from 305 out of the total 501 pts) and 564 PBMC samples (from 204 pts) were collected and analyzed, of which 467 were paired (from 175 pts). The 10 most frequently used genes for MRD monitoring were NRAS, IDH2, RUNX1, SRSF2, TP53, PTPN11, IDH1, BCOR, CEBPA and GATA2. Analyses in paired samples showed numerically higher MRD+ rates in BMMC vs. PBMC (57% vs 46% MRD+ at the end of induction, respectively). No differences in MRD rates were seen between treatment arms, and data was pooled for analysis. A high correlation of variant allele frequency (Pearson’s r≥0.84) and a high level of concordance in MRD calls were seen between BMMC and PBMC (91% overall percent agreement). For pts who fulfilled the criteria of a successful induction treatment, detection of MRD in either BMMC or PBMC in CR/CRi pts at the end of induction was associated with significantly lower event free survival (EFS) and overall survival (OS) compared to MRD− status (1-year EFS of 26% vs 72% in BMMC and 34% vs 71% in PBMC; 1-year OS of 73% vs 93% in BMMC and 67% vs 94% in PBMC) (Figure). Preliminary multivariate analysis, controlling for age and sex, confirmed the independent prognostic value of NGS-MRD for OS. Additional analysis is ongoing. Image:Summary/Conclusion: Taken together, these data indicate that MRD by NGS at the time of CR/CRi in both BMMC and PBMC, is prognostic for EFS and OS. Although the results should be interpreted with caution due to the premature termination of the study and limited follow-up, these further support the use of molecular MRD assessment to predict outcome in pts with AML.