Acute myeloid leukemia (AML) consists of at least 11 disease classes that represent distinct paths in the evolution of AML and have prognostic implications, based on an analysis of somatic driver mutations in 1,540 patients.
In total, 5,234 driver mutations were identified in 76 genes or regions, with 96% of patients having at least one mutation and 86% having two or more mutations. However, nearly one-half of the cohort did not fall into one of the molecular groups defined by the World Health Organization in 2008.
Elli Papaemmanuil, Ph.D.
“The characterization of many new leukemia genes, multiple driver mutations per patient, and complex co-mutation patterns prompted us to reevaluate genomic classification of AML from the beginning,” wrote Elli Papaemmanuil, Ph.D., a molecular geneticist at Memorial Sloan Kettering, New York, and of the Cancer Genome Project, Wellcome Trust Sanger Institute, and her colleagues (N Engl J Med. 2016 Jun 9; 374:2209-21).
The team developed a Bayesian statistical model to define 11 mutually exclusive subtypes based on patterns of co-mutations. The schema unambiguously classified 1,236 of 1,540 patients (80%) into a single subgroup and 56 (4%) into two or more groups. A subset of patients (166, 11%) remained unclassified, possibly due to mutations in genes not sequenced in the study.
NPM1-mutated AML was the largest class (27% of the cohort), followed by the chromatin-spliceosome group (18% of the cohort) that included mutations in genes regulating RNA splicing (SRSF2, SF3B1, U2AF1, and ZRSR2), chromatin (ASXL1, STAG2, BCOR, MLLPTD, EZH2, and PHF6), or transcription (RUNX1). Another subgroup consisted of mutations in TP53, as well as complex karyotype alterations, cytogenetically visible copy-number alterations (aneuploidies), or a combination. While broader than previous classifications, such as “monosomal karyotype AML” and “complex karyotype AML,” this group emerged from correlated chromosomal abnormalities and was mutually exclusive of other class-defining lesions. In general, patients in this group were older and had fewer RAS pathway mutations.
The groups had considerable differences in clinical presentation and overall survival, according to the report. The TP53-aneuploidy subgroup had poor outcomes, as previously described. Patients in the chromatin-spliceosome group had lower rates of response to induction chemotherapy, higher relapse rates, and poorer long-term outcomes, compared with other groups. Most of these patients (84%) would be classified as intermediate risk under current guidelines, but the characteristics were more similar to those of subgroups with adverse outcomes.
Overall survival was correlated with the number of driver mutations, and deleterious affects of mutations often were found to be additive. In some cases, complex gene interactions accounted for variation in outcomes, suggesting the clinical effect of some driver mutations may depend on the occurrence of co-mutations in a wider genomic context.