Contributions
Abstract: PB1380
Type: Publication Only
Session title: Acute myeloid leukemia - Biology & Translational Research
Background
Sequencing data from patients with Acute Myeloid Leukemia (AML) constitute a significant guidance at prognostic and therapeutic stages to support clinical decision making. In Argentina, next generation sequencing (NGS) on hematological diseases is performed in few institutions and the experience is currently based on small cohorts of patients. We have organized a collaborative working group in an effort to interactively increase the knowledge in our media.
Aims
To evaluate NGS real-world data from AML patients in a multicenter Argentine study.
Methods
Sequencing and clinical data of 90 adult AML patients were gathered from five centers performing NGS located across Argentina. Data was analyzed by using R-bioconductor (v4.0.3) and SPSS software. Statistical analyzes included Fisher’s exact /Chi2 and Kaplan Meir/Log-rank test. Median values are presented with interquartile ranges, unless otherwise specified. P values <0.05 were considered significant and, when appropriate, a false discovery rate (FDR) <0.05 was calculated.
Results
Sequencing data was obtained from eight different myeloid panels (customized or commercially available kits), targeting a median of 54 (range 30-81) complete or hot-spot gene regions. Our patients showed a median age of 58 (41-68) years and a sex ratio (M/F) of 1.36 (49/36). In this cohort, 44 genes showed at least one alteration, with a total of 291 pathogenic variants. The median of sequence variants per patient was 3 (2-5) with a median variant allele frequency of 0.4 (0.26-0.47). The most compromised genes were DNMT3A, NPM1, TET2, FLT3 and RUNX1 affecting 60% of the cohort (Figure 1). Sequence alterations in FLT3 (P<0.001) and NPM1 (P=0.008) were found differentially enriched in patients with normal karyotype. Secondary-type AML based on Lindsley’s 2015 classification (sAML-L) of variants were significantly associated with secondary-type AML (P<0.001). Within this classification, the most affected genes were SRSF2, BCOR and ASXL1 in 70% of the patients, also RUNX1 sequence variants were found significantly enriched (P=0.024, FDR=0.02) and TET2 mutations showed a tendency (P=0.04, FDR=0.23). Meanwhile, non-sAML-L were distinguished by NPM1 mutations (P<0.001; FDR=0.001). In addition, the somatic interaction of mutations was analyzed and, in accordance with the literature, NPM1-FLT3 (P<0.001), RUNX1-BCOR (P=0.003), TET2-SRSF2 (P=0.003), DNMT3A-IDH2 (P=0.006) were co-occurring events, while RUNX1-NPM1 (P=0.008) were mutually exclusive.
The median follow-up of this cohort was 18 (4-19) months with 51 reported deaths. The first-line therapy was 76% standard chemotherapy, 20% hypomethylating agents and 31 patients received hematopoietic stem cell transplantation. This procedure was favorable in terms of overall survival for patients <65 years with adverse molecular findings classified according to the ELN17 (38 vs 3 months, P=0.003) and to the Eisfeld’s 2020 proposal (29 vs 3 months, P<0.001), also with sAML-L or TP53 (19 vs 4 months, P=0.009) and for those with DNMT3A-TET2-ASXL1 (DTA) variants (29 vs 12 months, P=0.026).
Conclusion
Compilation and analysis of real-world NGS data from multiple institutions which are generated in diverse sequencing platforms represents a challenge in a collaborative setting. Nevertheless, our results are comparable to those reported in the literature and offer clinically relevant evidence that may help Argentine hematologists to make treatment decisions.
Keyword(s): Acute myeloid leukemia, Adult, Mutation analysis
Abstract: PB1380
Type: Publication Only
Session title: Acute myeloid leukemia - Biology & Translational Research
Background
Sequencing data from patients with Acute Myeloid Leukemia (AML) constitute a significant guidance at prognostic and therapeutic stages to support clinical decision making. In Argentina, next generation sequencing (NGS) on hematological diseases is performed in few institutions and the experience is currently based on small cohorts of patients. We have organized a collaborative working group in an effort to interactively increase the knowledge in our media.
Aims
To evaluate NGS real-world data from AML patients in a multicenter Argentine study.
Methods
Sequencing and clinical data of 90 adult AML patients were gathered from five centers performing NGS located across Argentina. Data was analyzed by using R-bioconductor (v4.0.3) and SPSS software. Statistical analyzes included Fisher’s exact /Chi2 and Kaplan Meir/Log-rank test. Median values are presented with interquartile ranges, unless otherwise specified. P values <0.05 were considered significant and, when appropriate, a false discovery rate (FDR) <0.05 was calculated.
Results
Sequencing data was obtained from eight different myeloid panels (customized or commercially available kits), targeting a median of 54 (range 30-81) complete or hot-spot gene regions. Our patients showed a median age of 58 (41-68) years and a sex ratio (M/F) of 1.36 (49/36). In this cohort, 44 genes showed at least one alteration, with a total of 291 pathogenic variants. The median of sequence variants per patient was 3 (2-5) with a median variant allele frequency of 0.4 (0.26-0.47). The most compromised genes were DNMT3A, NPM1, TET2, FLT3 and RUNX1 affecting 60% of the cohort (Figure 1). Sequence alterations in FLT3 (P<0.001) and NPM1 (P=0.008) were found differentially enriched in patients with normal karyotype. Secondary-type AML based on Lindsley’s 2015 classification (sAML-L) of variants were significantly associated with secondary-type AML (P<0.001). Within this classification, the most affected genes were SRSF2, BCOR and ASXL1 in 70% of the patients, also RUNX1 sequence variants were found significantly enriched (P=0.024, FDR=0.02) and TET2 mutations showed a tendency (P=0.04, FDR=0.23). Meanwhile, non-sAML-L were distinguished by NPM1 mutations (P<0.001; FDR=0.001). In addition, the somatic interaction of mutations was analyzed and, in accordance with the literature, NPM1-FLT3 (P<0.001), RUNX1-BCOR (P=0.003), TET2-SRSF2 (P=0.003), DNMT3A-IDH2 (P=0.006) were co-occurring events, while RUNX1-NPM1 (P=0.008) were mutually exclusive.
The median follow-up of this cohort was 18 (4-19) months with 51 reported deaths. The first-line therapy was 76% standard chemotherapy, 20% hypomethylating agents and 31 patients received hematopoietic stem cell transplantation. This procedure was favorable in terms of overall survival for patients <65 years with adverse molecular findings classified according to the ELN17 (38 vs 3 months, P=0.003) and to the Eisfeld’s 2020 proposal (29 vs 3 months, P<0.001), also with sAML-L or TP53 (19 vs 4 months, P=0.009) and for those with DNMT3A-TET2-ASXL1 (DTA) variants (29 vs 12 months, P=0.026).
Conclusion
Compilation and analysis of real-world NGS data from multiple institutions which are generated in diverse sequencing platforms represents a challenge in a collaborative setting. Nevertheless, our results are comparable to those reported in the literature and offer clinically relevant evidence that may help Argentine hematologists to make treatment decisions.
Keyword(s): Acute myeloid leukemia, Adult, Mutation analysis