Introduction: Old age and unfavorable cytogenetics are well-established adverse prognostic factors for outcome in patients (pts) with AML. Recently, high-throughput sequencing techniques have identified multiple somatically acquired driver mutations (muts) with a strong prognostic impact in AML. We explored the relative contribution of baseline molecular genetics versus a response-based sequential treatment approach on response and overall survival (OS) at two years in pts >60 years (y) with newly diagnosed AML treated within the multicenter RAS-AZIC study of the OSHO.

Patients and methods: In the RAS-AZIC trial, 112 pts received priming with Azacitidine (AZA) followed by a day (d) 15 bone marrow (BM) blast count-adapted sequential treatment with AZA or intensive chemotherapy (IC). Pts with d15 blasts ≥45% received IC. If the count was <45%, AZA was continued at d28. The second treatment adaptation was based on d56 response. If no complete remission (CR) or CR with incomplete recovery (CRi) were documented, IC was applied followed by AZA maintenance or allogeneic hematopoietic stem cell transplantation if a remission was achieved on d90. Conventional cytogenetics, NPM1, and FLT3 mutational screening were used for genetic risk stratification following the 2010 ELN recommendations. Response [CR/CRi/Partial remission (PR)] and treatment-related mortality on d90, the two-year OS, event-free survival (EFS), and relapse-free survival (RFS) were previously presented (Jaekel N, et al; ASH 2018).

Baseline BMs of 77 pts who signed the translational informed consent were analyzed by NGS using the NEOmyeloid panel of NEO New Oncology (Siemens Healthineers Company, Germany). This hybrid capture-based assay comprises 43 genes and detects point muts, small insertions and deletions in 30 genes as well as gene fusions in 18 genes.

Molecular results were correlated to outcome data.

Results: Median age was 70y. Secondary AML was present in 39%. Genetic risk was int-II and adverse in 25% and 27% respectively. Response on d90 was 67% (CR/CRi 58%, PR 9%). A median of 3 muts per pt were found. Mutations in DNA methylation-related-, spliceosome complex-, and chromatin-modifying-genes were 62%, 39%, and 30% respectively. Overall, muts in DNMT3A (34%), ASXL1 (25%), RUNX1 (24%), SRSF2 (23%), and TET2 (23%) were the most frequently detected (fig. 1). As expected, 72% of SRSF2 muts were associated with ASXL1 and/or RUNX1 muts. CR/CRi in the favorable and int-I 2010 ELN genetic risk groups was 81% and 57.5% in int-II and adverse risk pts. With the exception of muts in spliceosome complex-related genes, the mutational status (including the FLT3 status) seemed to have no impact on response in this cohort (table 1). In the presence of splicing factor gene muts a trend towards an inferior response of 53% compared to 77% in pts without such a mut was observed. High risk muts such as ASXL1, RUNX1, and TP53 muts which were included in the 2017 ELN adverse risk stratification were present in 20/32 (63%) of pts with favorable and int-I genetic risk. Yet, only 5/32 (16%) did not achieve a response. Similarly, CR/CRi was achieved in 7/9 (78%) pts with TP53 mut which was associated with an adverse karyotype in 8 pts. Likewise, OS tended to be similar across all molecular classes with the exception of KMT2A gene fusions with their well-known negative impact on outcome. The presence of such a fusion (n=4) was associated with a median survival time of only 8 compared to 16 months in pts without a detectable KMT2A fusion. These KMT2A gene fusions also correlated with a shorter EFS and RFS. The presence of myeloid transcription factor gene muts such as in RUNX1 (24%) and BCOR (13%) tended to be associated with an inferior EFS, while cohesin complex- (STAG2 8%), tumour suppressor-, and DNA-methylation-related gene muts with a negative RFS (table 1).

Conclusions: This explorative analysis with a limited number of patients describes the pattern of genetic mutations in elderly AML treated with a response-based sequential approach that integrates epigenetic and chemotherapeutic treatments and their association with outcome. Whether such a response-based treatment mitigates the negative impact of (some) high risk mutations on outcome needs to be addressed in a much larger trial.

Disclosures

Jaekel:Celgene: Other: travel grant. Siebolts:Merck: Honoraria; Boehringer Ingelheim: Honoraria; MSD: Honoraria; Pfizer: Honoraria; AstraZeneca: Honoraria; Roche: Honoraria; Novartis: Honoraria; BMS: Honoraria. Pazaitis:Novartis: Consultancy. Niederwieser:Daichii: Speakers Bureau; Cellectis: Consultancy. Hänel:Amgen: Honoraria; Roche: Honoraria; Novartis: Honoraria; Takeda: Other: advisory board; Celgene: Other: advisory board. Maschmeyer:Gilead, Janssen Cilag, Astra Zeneca; BMS, Merk-Serono: Honoraria. Heukamp:NEO New Oncology GmbH: Employment. Menon:NEO New Oncology GmbH: Employment. Bertrand:NEO New Oncology GmbH: Employment. Wickenhauser:BMS: Honoraria; Merck-Serono: Honoraria; MSD: Honoraria; Böhringer: Honoraria; Novartis: Honoraria; Takeda: Honoraria; Roche: Honoraria. Brosteanu:University Leipzig: Employment. Al-Ali:Celgene: Research Funding; Novartis: Consultancy, Honoraria, Research Funding; CTI: Honoraria.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution