Photo courtesy of NIGMS
Using patient samples from 3 prospective multicenter clinical trials of the German-Austrian AML Study Group, researchers have identified 5234 driver mutations involving 76 genes or regions in 1540 patients with acute myeloid leukemia (AML).
They say the sample size afforded a more comprehensive analysis than previously conducted, and as a result, they found patients were divided into at least 11 subgroups of AML.
They also identified 3 genomic categories beyond the existing World Health Organization (WHO) subgroups. These genomic categories are chromatin–spliceosome mutations, TP53–aneuploidy, and provisionally, IDH2R172 mutations.
Their study, led by scientists at the Wellcome Trust Sanger Institute and international collaborators, could have an impact on clinical trial design and improve the way patients are diagnosed and treated in the future.
The scientists stated in their published paper that 736 patients (48%) in their study would not have fit into the molecular groups included in the 2008 WHO classification of adult AML. This prompted them to reevaluate genomic classification of AML from the beginning.
“We have shown that AML is an umbrella term for a group of at least 11 different types of leukemia,” said Peter Campbell, MBChB, PhD, co-leader of the study. “We can now start to decode these genetics to shape clinical trials and develop diagnostics.”
The scientists found NPM1-mutated AML to be the largest class in their cohort, accounting for 27% of the patients.
The second largest subgroup--the chromatin-spliceosome group—accounted for 18% of patients, the TP53-aneuploidy group accounted for 13%, and the IDH2R172 mutations for 1%.
The study also showed that most patients had a unique combination of genetic changes driving their leukemia. This genetic complexity helps explain why AML shows such variability in survival rates among patients.
Under their new schema, the scientists were able to unambiguously classify 80% of the 1540 patients with driver mutations in a single subgroup. Fifty-six of the patients (4%) met criteria for 2 or more categories, primarily in the TP53-aneuploidy and chromatin-spliceosome groups.
They were not able to classify 11% of patients with driver mutations. They explained that these patients might have had mutations in drivers that were either not sequenced or had mutations that were missed.
The scientists applied their classification schema to an independent cohort from the Cancer Genome Atlas (TCGA). The new schema was able to replicate the absence of overlap among subgroups, and relative frequencies of the mutations were equivalent to those in the AML cohort.
“For the first time we untangled the genetic complexity seen in most AML cancer genomes into distinct evolutionary paths that lead to AML,” joint first author Elli Papaemmanuil, PhD, of Memorial Sloan Kettering Cancer Center in New York, said.
“By understanding these paths, we can help develop more appropriate treatments for individual patients with AML. We are now extending such studies across other leukemias."
The investigators recommend prospective clinical trials to further validate the schema.
The work was supported by the Wellcome Trust, Bundesministerium fur Bildung und Forschung, Deutsche Krebshilfe and Deutsche Forschungsgemeinschaft, the European Hematology Association, Amgen and the Kay Kendall Leukaemia Fund.
