Contributions
Abstract: EP423
Type: E-Poster Presentation
Session title: Acute myeloid leukemia - Biology & Translational Research
Background
Acute myeloid leukemia (AML) is mainly a disease of older patients (pts) with median age of 70 years (yrs) at diagnosis. Age and age-related unfavorable prognostic factors, like adverse genetic changes, impact management and outcome. While large cohorts of younger AML have been extensively studied at the molecular level, genomic aberrations underlying elderly AML have by far been less well characterized.
Aims
To evaluate the mutational landscape of 294 elderly AML (median age 76 yrs; range, 70-90 yrs) entered into the AMLSG BiO Registry (NCT01252485). Primary treatment was intensive chemotherapy (IC, n=33), hypomethylating agents (HMA, n=164) or best supportive care (BSC, n=97).
Methods
Custom Haloplex HS target enrichment system (Agilent) for resequencing of the entire coding regions of 94 myeloid disease related genes (Illumina MiSeq).
Results
Based on an average gene coverage of 300x(fold) for consensus reads and a defined variant allele frequency (VAF) ≥5%, we identified 1095 mutations (mut) in 83 genes in 286 of the 294 pts with a median of 4 mut/pt (range 0-9). The vast majority were missense mut (n=743, 67.9%) followed by indels (n=214, 19.5%), nonsense (n=112, 10.2%) and splice site mut (n=26, 2.4%). Within our cohort of older pts most frequently mutated genes were TET2 (23.2%), TP53 (22.8%), FLT3 (ITD and TKD, 24.5%), SRSF2 (22.8%), DNMT3A (21.8%), RUNX1 (18.0%), NRAS (13.3%), NPM1 (12.6%), IDH2 (10.5%), ASXL1 (9.5%), SF3B1 (8.5%) and U2AF1 (7.8%). Observing a broad variance for VAF most pts displayed a dominant clone with a VAF in ~50% range. For mut affecting the functional gene classes methylation, chromatin-spliceosome, signaling, transcription, tumor suppressors, and others, VAF differed significantly (P<.001).
According to ELN 2017 risk classification 34 (12.2%) had favorable, 76 (27.3%) intermediate and 168 (60.4%) adverse risk. Risk groups did not correlate with age, categorized as 70-74 yrs, 75-80 yrs, and ≥80 yrs, but with functional classes. Within the favorable, intermediate and adverse risk groups 68%, 67% and 49% of pts (P=.002) had methylation mut and 74%, 57%, and 68% (P=.002) signaling mut, respectively. Methylation mut were correlated with higher white blood cell counts (P<.001), lactate dehydrogenase, and bone marrow blasts (P=.002 each).
Genomic classification based on cytogenetics and patterns of comutations (Papaemmanuil et al, NEJM 2016) revealed chromatin-spliceosome (n=101), TP53-aneuploidy (n=86), not classifiable (n=36), NPM1 (n=33), other classes (n=17), and no class (n=21).
Median follow-up was 3.1 years (95% CI: 2.1-4.1). Due to the different treatment regimens, outcome analyses were restricted to overall survival (OS). Median OS was 0.5 years (0.3-0.7 yrs) with no difference for ELN risk groups (P=.302). According to treatment median OS was 0.97 yrs for IC, 0.76 yrs for HMA and 0.15 yrs for BSC (P<.001) with no significant difference between IC and HMA (P=.475). The general beneficial effect for HMA compared to BSC, decreased with age (70-74 yrs, P<.001; 75-80 yrs, P=.029; ≥80 yrs, P=.889). The TP53-aneuploidy group had the worst and the spliceosome group the best OS (P=.046). A significant benefit for HMA compared to BSC was seen for the TP53-aneuploidy group (P<.001), but not the spliceosome group (P=.394).
Conclusion
The mutational landscapes of elderly AML differs significantly from that of younger AML. Adapting genomic classification of younger AML, distinct subgroups of prognostic relevance were also observed in elderly AML.
FGR and MH, KD and LB contributed equally.
Keyword(s): Acute myeloid leukemia, Elderly, Mutation analysis
Abstract: EP423
Type: E-Poster Presentation
Session title: Acute myeloid leukemia - Biology & Translational Research
Background
Acute myeloid leukemia (AML) is mainly a disease of older patients (pts) with median age of 70 years (yrs) at diagnosis. Age and age-related unfavorable prognostic factors, like adverse genetic changes, impact management and outcome. While large cohorts of younger AML have been extensively studied at the molecular level, genomic aberrations underlying elderly AML have by far been less well characterized.
Aims
To evaluate the mutational landscape of 294 elderly AML (median age 76 yrs; range, 70-90 yrs) entered into the AMLSG BiO Registry (NCT01252485). Primary treatment was intensive chemotherapy (IC, n=33), hypomethylating agents (HMA, n=164) or best supportive care (BSC, n=97).
Methods
Custom Haloplex HS target enrichment system (Agilent) for resequencing of the entire coding regions of 94 myeloid disease related genes (Illumina MiSeq).
Results
Based on an average gene coverage of 300x(fold) for consensus reads and a defined variant allele frequency (VAF) ≥5%, we identified 1095 mutations (mut) in 83 genes in 286 of the 294 pts with a median of 4 mut/pt (range 0-9). The vast majority were missense mut (n=743, 67.9%) followed by indels (n=214, 19.5%), nonsense (n=112, 10.2%) and splice site mut (n=26, 2.4%). Within our cohort of older pts most frequently mutated genes were TET2 (23.2%), TP53 (22.8%), FLT3 (ITD and TKD, 24.5%), SRSF2 (22.8%), DNMT3A (21.8%), RUNX1 (18.0%), NRAS (13.3%), NPM1 (12.6%), IDH2 (10.5%), ASXL1 (9.5%), SF3B1 (8.5%) and U2AF1 (7.8%). Observing a broad variance for VAF most pts displayed a dominant clone with a VAF in ~50% range. For mut affecting the functional gene classes methylation, chromatin-spliceosome, signaling, transcription, tumor suppressors, and others, VAF differed significantly (P<.001).
According to ELN 2017 risk classification 34 (12.2%) had favorable, 76 (27.3%) intermediate and 168 (60.4%) adverse risk. Risk groups did not correlate with age, categorized as 70-74 yrs, 75-80 yrs, and ≥80 yrs, but with functional classes. Within the favorable, intermediate and adverse risk groups 68%, 67% and 49% of pts (P=.002) had methylation mut and 74%, 57%, and 68% (P=.002) signaling mut, respectively. Methylation mut were correlated with higher white blood cell counts (P<.001), lactate dehydrogenase, and bone marrow blasts (P=.002 each).
Genomic classification based on cytogenetics and patterns of comutations (Papaemmanuil et al, NEJM 2016) revealed chromatin-spliceosome (n=101), TP53-aneuploidy (n=86), not classifiable (n=36), NPM1 (n=33), other classes (n=17), and no class (n=21).
Median follow-up was 3.1 years (95% CI: 2.1-4.1). Due to the different treatment regimens, outcome analyses were restricted to overall survival (OS). Median OS was 0.5 years (0.3-0.7 yrs) with no difference for ELN risk groups (P=.302). According to treatment median OS was 0.97 yrs for IC, 0.76 yrs for HMA and 0.15 yrs for BSC (P<.001) with no significant difference between IC and HMA (P=.475). The general beneficial effect for HMA compared to BSC, decreased with age (70-74 yrs, P<.001; 75-80 yrs, P=.029; ≥80 yrs, P=.889). The TP53-aneuploidy group had the worst and the spliceosome group the best OS (P=.046). A significant benefit for HMA compared to BSC was seen for the TP53-aneuploidy group (P<.001), but not the spliceosome group (P=.394).
Conclusion
The mutational landscapes of elderly AML differs significantly from that of younger AML. Adapting genomic classification of younger AML, distinct subgroups of prognostic relevance were also observed in elderly AML.
FGR and MH, KD and LB contributed equally.
Keyword(s): Acute myeloid leukemia, Elderly, Mutation analysis