Skip to main content
Premium Trial:

Request an Annual Quote

TCGA Team Reports on Findings from Comprehensive Breast Cancer Analysis

This article has been updated with a comment from one of the study authors and additional information on TCGA data.

NEW YORK (GenomeWeb News) – In a study appearing online yesterday in Nature, members of the Cancer Genome Atlas presented a multi-faceted genetic analysis of breast cancer, characterizing four main subtypes of the disease and uncovering shared molecular features between one of these subtypes and tumors from another part of the body.

"This study has now provided a near complete framework for the genetic causes of breast cancer," corresponding author Charles Perou, a genetics researcher with the University of North Carolina at Chapel Hill and the Lineberger Comprehensive Cancer Center, said in a statement, "which will significantly impact clinical medicine in the coming years as these genetic markers are evaluated as possible markers of therapeutic responsiveness."

Along with other investigators from UNC and Washington University, Perou helped to lead a large international team that brought together multiple types of data for tumor and matched normal samples from more than 800 women with breast cancer. This included information generated through exome and microRNA sequencing experiments, as well as DNA methylation, copy number, gene expression, and protein profiling.

As found in prior breast cancer studies, breast tumors displayed a wide range of genetic glitches. Even so, across 466 tumors assessed by five of the six methods, researchers saw breast cancers clustering in groups that more or less corresponded to gene expression-based breast cancer classifications.

"Our ability to integrate information across platforms provided key insights into previously defined gene expression subtypes," the TCGA team explained, "and demonstrated the existence of four main breast cancer classes when combining data from five platforms, each of which shows significant molecular heterogeneity."

"It very much reconfirms some previous findings that breast cancer is not one disease, but it's at least four different diseases," co-lead author Perou told GenomeWeb Daily News. "And now this extends those findings from basically gene expression findings into the realm of all the other different technologies [used in the study]."

Among the most intriguing findings were the identification of new mutations and potential treatment targets for the most common breast cancer subtype: ER-positive, luminal A tumors. Also surprising was the researchers' realization that tumors belonging to the basal-like subtype, which includes difficult-to-treat "triple-negative" breast cancers, tended to share genetic features with high-grade serous tumors found in the ovary.

"These findings suggest that basal-like breast cancer, while arising in the same anatomical location, is potentially a completely different disease," co-author Katherine Hoadley, a genetics researcher affiliated with UNC at Chapel Hill and the Lineberger Comprehensive Cancer Center, said in a statement.

For their study, the researchers set out to do an integrated analysis using tumor and matched germline samples from 825 women with a range of breast cancer types. Along with whole-exome sequencing experiments on 510 tumors from 507 women and miRNA sequencing on 697 samples, these assessments included array-based gene expression, DNA methylation, and genotyping analyses on tumors from between around 550 and 800 women each.

All told, the team had access to exome sequence, miRNA sequence, gene expression, copy number, and methylation information for 466 of the tumors. Of these, nearly 350 were also profiled at the protein level using a reverse-phase protein array method.

Despite the pronounced genetic heterogeneity in these samples, the team noted, tumors tended to cluster in line with expression-based subtypes for the disease. And the search for somatically mutated genes unearthed almost all of the genes linked to breast cancer in the past.

Still, new candidate genes turned up as well. And while tumors within each subtype each shared certain defining features, mutational profiles within a given subtype were far from uniform.

"The biological finding of the four main breast cancer subtypes caused by different subsets of genetic and epigenetic abnormalities raises the hypothesis that much of the clinically observable plasticity and heterogeneity occurs within, and not across, these major biological subtypes of breast cancer," Perou and his co-authors noted.

In its analyses, the team delved into the diverse genetic patterns within each of the breast cancer subtypes, known as HER2-enriched breast cancer, luminal A and luminal B (estrogen receptor expression-positive) breast cancers, and basal-like breast cancer. The latter subtype typically includes most triple-negative cases, known for having negligible expression of estrogen receptor, progesterone receptor, and HER2 genes.

When breast tumors from all four subtypes were considered as a single group, the researchers found just three genes that were altered with more than 10 percent frequency: TP53, PIK3CA, and GATA3. All three genes have each been linked to breast cancer in past studies, including an analysis published by American and Mexican researchers in Nature this spring.

In the luminal A breast cancers, though, mutations in two of these genes — PIK3CA and GATA3 — tended to turn up at specific places in each gene, as did mutations affecting the MAP3K1 gene. The luminal A tumors also tended to have a much lower TP53 mutation frequency than more aggressive luminal B tumors.

Other subtype-specific patterns emerged from the study as well, including an intriguing overlap between mutation, gene expression, and copy number patterns in basal-like breast cancers and those described for serous ovarian cancer. In particular, basal-like breast cancer and serous ovarian cancers shared similar alterations affecting genes such as TP53, BRCA1/2, RB1, cyclin E1, and MYC, researchers reported. And both appear prone to pronounced genomic instability.

That prompted the study's authors to propose that "common therapeutic approaches should be considered" for basal-like and serous ovarian cancers.

"For basal-like breast tumors, it's clear they are genetically more similar to ovarian tumors than to other breast cancers," Washington University researcher Matthew Ellis, a co-lead author on the study, said in a statement. "Whether they can be treated the same way is an intriguing possibility that needs to be explored."

The TCGA team is continuing to analyze samples described in the current study and ultimately plans to test as many as 1,100 breast cancers by all six methods described, Perou noted.

In addition, the researchers have already completed RNA-sequencing for roughly 800 breast cancer cases, as well as whole-genome sequencing on a few dozen of the breast cancers — work that will be detailed in upcoming publications.

Sequence data from the current study is available online through the Cancer Genomic Hub, maintained by the University of California at Santa Cruz. Additional data and information on the samples can be accessed through a TCGA data coordinating center site maintained by the National Cancer Institute and the National Human Genome Research Institute, through the Institute for Systems Biology and MD Anderson Cancer Center's Regulome Explorer site, and via the Memorial Sloan-Kettering Cancer Center's cBio Cancer Genomics Portal.

The Scan

Push Toward Approval

The Wall Street Journal reports the US Food and Drug Administration is under pressure to grant full approval to SARS-CoV-2 vaccines.

Deer Exposure

About 40 percent of deer in a handful of US states carry antibodies to SARS-CoV-2, according to Nature News.

Millions But Not Enough

NPR reports the US is set to send 110 million SARS-CoV-2 vaccine doses abroad, but that billions are needed.

PNAS Papers on CRISPR-Edited Cancer Models, Multiple Sclerosis Neuroinflammation, Parasitic Wasps

In PNAS this week: gene-editing approach for developing cancer models, role of extracellular proteins in multiple sclerosis, and more.