An in-depth characterization of the genomics, proteomics, and methylation patterns of nearly 400 endometrial cancers has revealed 4 new molecular tumors subtypes, information which could change diagnosis and treatment. The findings are the latest to be released by The Cancer Genome Atlas Research (TCGA), a joint effort of the National Cancer Institute and the National Human Genome Research Institute to characterize 20 major cancers.
Mardis and researchers at dozens of institutions from around the country collected and analyzed samples from 373 endometrial cancer patients. The tumors were assessed for somatic copy number alterations (SCNAs), which immediately produced four clusters of tumor types. When associated with patient follow-up data, each SCNA cluster had a distinct pattern of survival.
Then the team sequenced the exomes of endometrial tumors from their collection and analyzed messenger RNA expression, protein expression, microRNA expression, and methylation patterns. Overall, 48 genes had mutation frequencies that varied between the four SCNA subtypes, and some genes in particular stood out as being clearly associated with one subtype.
To group the endometrial cancers, the team developed a new data algorithm, dubbed SuperCluster, to integrate the different forms of data. The algorithm identified four clusters, similar to those identified through SCNA clustering. The four new groups are POLE ultramutated, microsatellite instability hypermutated, copy number low, and copy number high. Mardis and colleagues also used integrated data to identify molecular pathways that were repeatedly affected in the tumors.
“I think the strength of the TCGA project is the large numbers of samples and the fact that we’re being very comprehensive across the spectrum of DNA, RNA, and protein in terms of the questions that we’re asking,” said Mardis.
Currently, doctors classify endometrial cancers into two groups—endometriod and serous—based on the appearance of tumor cells under a microscope. But by testing for the genetic signatures of the four new subtypes, clinicians can make more informed decisions on the best treatment plan for women. Moreover, clinical trials can look at a drug’s effect on each molecular subtype rather than in all endometrial cancers without considering genetic and phenotypic variation.
The team also found that serous endometrial cancer, the most aggressive subtype, has similar molecular characteristics to some types of breast and ovarian cancer. This could lead to clinical trials of drugs that could treat all three types of tumors, said Mardis.
The TCGA researchers plan to continue collecting endometrial tumor samples to strengthen their dataset, and pursue questions on the effect of ethnicity on the cancer’s heritability and outcomes. They’ll also be performing full genome sequencing on a subset of the samples.
1. The Cancer Genome Atlas Research Network (2013). Integrated genomic characterization of endometrial carcinoma. Nature 497:67-73