-Author name in bold denotes the presenting author
-Asterisk * with author name denotes a Non-ASH member
Clinically Relevant Abstract denotes an abstract that is clinically relevant.

PhD Trainee denotes that this is a recommended PHD Trainee Session.

Ticketed Session denotes that this is a ticketed session.

1066 Comparative Genomic Analysis of Adolescents and Young Adults Versus Elderly with Acute Myeloid Leukemia

Program: Oral and Poster Abstracts
Session: 617. Acute Myeloid Leukemia: Biology, Cytogenetics, and Molecular Markers in Diagnosis and Prognosis: Poster I
Hematology Disease Topics & Pathways:
AML, Diseases, Myeloid Malignancies
Saturday, December 5, 2020, 7:00 AM-3:30 PM

Hassan Awada, MD1*, Carmelo Gurnari, MD2*, Arda Durmaz3*, Simona Pagliuca, MD4, Misam Zawit, MBBS4*, Cassandra M Kerr, MS4* and Valeria Visconte, PhD4

1Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland, OH
2Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic, Cleveland
3Department of Systems biology and Bioinformatics, CWRU, cleveland
4Department of Translational Hematology and Oncology Research, Taussig Cancer Institute, Cleveland Clinic Foundation, Cleveland, OH

Prognosis and survival outcomes of acute myeloid leukemia (AML) strongly depend on age at diagnosis. In fact, long-term survival rates range from 60% in children and adolescent/young adults (AYA-AML: aged 15-39 years) to 10-15% in elderly patients (E-AML: aged >70 years). The currently available classification and prognostication models incorporate genomic features that include cytogenetics and molecular mutations based on adult AML clinical trials that mainly enroll patients outlying these two age categories. Consequently, prognostication models might still not account for the invariant genomic background of each subgroup. In this study, we aimed to dissect the genomic landscape of AYA-AML and E-AML cases in order to uncover patterns of distinct genomic features of important pathogenetic information, knowing how fundamental it is for the development of novel treatment approaches and more importantly for providing opportunities of individualization therapies.

We studied the clinical and genomic characteristics of fully annotated sets of 198 AYA-AML and 414 E-AML cases from the Cleveland Clinic and publicly available data. Median age at diagnosis was 31 years for AYA-AML (range: 16-39 years) and 76 years (range: 71-100 years) for E-AML patients with a significant male preponderance noted in the latter cohort (65% vs 46%, P<0.0001). Based on the 2016 WHO classification, AML subtypes were differently distributed within the two subgroups. Low-risk AML (PML-RARA AML, core-binding factor AML, and CEBPA biallelic AML) was more prevalent among AYA cases while a higher odd of secondary AML (sAML) was found in E-AML. Compared to AYA-AML, E-AML were more typically presenting with a less proliferative disease phenotype, as outlined by a lower bone marrow blast infiltration (median: 34% vs 51%, P=0.04) and a higher frequency of leukopenia at baseline (66% vs 27%, P=0.009). In terms of cytogenetic abnormalities, when compared with AYA-AML, E-AML cases harbored more complex karyotype (32% vs 13%, P=0.002) and other cytogenetic abnormalities e.g., -5/del(5q) (14% vs 2%, P<0.0001), -7/del(7q) (14% vs 5%, P=0.001), -17/del(17p) (5% vs 1%, P=0.02), del(20q) (4% vs <1%, P=0.01) and trisomy 8 (14% vs 5%, P=0.0007) while a significantly lower odd of -9/del(9q) (<1% vs 4%, P=0.006).

A total of 1092 somatic mutations were identified by next-generation sequencing of common myeloid genes. The AYA-AML cohort had a total of 249 mutations (ratio 1.25) while 843 mutations were reported in E-AML patients (ratio 2.04). Older patients, as compared to AYA-AML, had more mutations in age-related myeloid genes linked to clonal hematopoiesis, including mutations in DNMT3A (15% vs 6%, P=0.001), ASXL1 (14% vs 4%, P=0.0005), and TET2 (19% vs 7%, P=0.0002). As inferred from the cytogenetic comparisons, TP53 mutations were more frequently detected in E-AML (10% vs 2%, P=0.002) compared to AYA-AML, as they carried more chromosomal 5 and 17 abnormalities which are commonly associated with this genomic lesion. On the other hand, AYA-AML had a higher frequency of mutated WT1 gene (9% vs 2%, P=.0009). Multivariate Cox-Proportional hazard was performed to identify group-specific genomic lesions associated with survival in each cohort. In AYA-AML, normal karyotype (HR:0.59, P=0.02), t(15;17) translocation (HR:0.24, P=0.0001), inv(16)/t(8;21) (HR:0.52, P=0.02) and CEBPA biallelic (HR:0.50, P=0.02) were favorable lesions while complex karyotype had an adverse impact on survival (HR:3.0, P=0.0003). Moreover, while normal karyotype and t(15;17) translocation had a favorable impact in E-AML (HR:0.73, P=0.0001 and HR:0.28, P<0.0001, respectively), mutations in RUNX1 (HR:1.2, P=0.4), TET2 (HR:1.2, P=0.01), and TP53 (HR:1.97, P<0.0001) lead to worse outcomes.

In conclusion, we demonstrated that E-AML cases are typically characterized by mutations belonging to the genomic subgroups of DNA methylation and chromatin modifiers, which have been reported to occur in age-related hematopoiesis or CHIP. Moreover, the complexity of aberrations (including cytogenetics) emphasizes that in this group of patients AML arises from a stepwise mutation acquisition model. In contrast, AYA-AML are typically characterized by favorable cytogenetic translocations and less clonal instability, dominated by the enrichment in myeloid leukemia driver gene mutations such as WT1.

Disclosures: No relevant conflicts of interest to declare.

*signifies non-member of ASH