P503: MRD MONITORING DURING INTENSIVE CHEMOTHERAPY IN PEDIATRIC AML- DATA OF THE AML-BFM GROUP

Abstract

Background: During the past decades, minimal residual disease (MRD) has been established as a diagnostic tool, providing critical prognostic information in pediatric acute myeloid leukemia (AML). To measure initial treatment response, immunophenotyping was used by most cooperative study groups. However, known limitation of sensitivity and specificity supported the continuous evaluation of qPCR-based MRD detection in pediatric AML. Aims: To determine the prognostic role of MRD monitoring by RT-qPCR in the pediatric AML. Methods: A total of 238 pediatric AML patients were monitored for MRD by reverse-transcriptase quantitative PCR (RT-qPCR) assay from September 2012 to December 2020. Since the inclusion to this study is based on the presence of quantifiable genetic markers, the patient group is strongly biased. The treatment was conducted according to the AML-BFM study 2004 and 2012 as well as the register 12 and 17. All measurements were performed in bone marrow samples or in peripheral blood at initial diagnosis. MRD monitoring included the following aberrations: t(8;21), inv(16), inv(11), NPM1, FLT3-ITD, t(15;17), t(5;11), t(9;11), t(6;9), WT1, t(1;11), t(11;11), t(1;22), t(11;12), t(17;19), t(3;5), t(4;10), t(11;19), t(2;11), t(6;11), t(X;11), t(6;8), t(8;16) and t(8;22). Results: The patients had a mean age at diagnosis of 9.2 years (range 10 days to 18.6 years) and 51 % were females. They were distributed to the following stratification groups: standard risk (56 %), intermediate risk (27 %) and high risk (17 %). The event free survival was 71 % (±3 %) and the OS reached 90 %(±2 %). In core binding factor acute myeloid leukemia (CBF-AML) (n=91) a positive MRD level persists in the majority of cases. Following the 3rd treatment block 64% of cases remained MRD-positive (MRD threshold ≥1x10-5). However, MRD monitoring showed no prognostic relevance as relapses were detected in 18 % (6/33) and 17 % (10/58) of MRD-positive and negative cases, respectively. For patients with NPM1 mutations (n=26) MRD-positivity and -negativity (threshold of ≥1x10-6) showed a similar relapse rate of 10 %. This data are in contrast to reports in adults (Ivey et al. NEJM 2016). In patients with an AML and t(9;11) MRD-positivity after first and second induction indicated a significant higher risk of relapse. After first induction MRD-positive cases (n=19) had a relapse rate of 32 % compared to 13 % for negative cases (n=24). After second induction the relapse rates were 63 % and 22 % for MRD-positive (n=8) and negative (n=27) cases, respectively. Summary/Conclusion: MRD monitoring is a powerful tool to improve risk group stratification in pediatric AML. Despite its high sensitivity and specificity, the prognostic informativeness of MRD monitoring through RT-qPCR varies widely among different genetic subgroups. In CBL-AML or NPM1-positive AML, which are prognostically more favorable, persisting MRD-positivity alone was not associated with an increase in the risk of relapse. In contrast, MRD-positivity in AML with t(9;11) appears to be highly informative. Further studies with larger number of patients are needed to further define AML subgroups and thresholds for MRD-positivity in the pediatric AML.