Researchers have determined that breast cancer can be classified into 10 different subtypes instead of four subtypes as previously thought, a discovery that could play a significant role in tailoring personalized care for patients as these subtypes become more understood.
In the study, published online in April by the journal Nature
, researchers from Canada and the United Kingdom performed a genomic analysis on 1,992 breast tumor samples from women diagnosed with breast cancer five to 10 years ago who received long-term follow-up. The study found that approximately 40% of genes expressed inherited variants – including copy number variations and single nucleotide polymorphisms (SNP) – and acquired somatic copy number aberrations (CNA), especially cis
- and trans
-acting CNAs. These variations are the result of regions of the genome either being duplicated or deleted, resulting in either susceptibility or resistance to certain diseases.
When an analysis of DNA-RNA profiles was performed, the researchers found subtypes with distinct clinical outcomes. Traditionally, four subtypes have been associated with breast cancer, which in most cases are determined by testing for the estrogen receptor (ER) and HER2. Disparities existed in the clinical outcomes of patients within each of these four subtypes, suggesting heterogeneity that could be further analyzed through genomics.
The researchers were able to identify 10 subtypes with distinct clinical outcomes. These were determined in a discovery set of 997 tumors and confirmed in a validation set of 995 tumors.
One example of a new subtype identified by the researchers was a high-risk, ER+ 11q13/14 subtype that had a steep mortality trajectory and higher hazard ratio (HR=3.620, 95% CI, 1.381-8.141). Strong cis
-acting associations characterized this group.
Several groups were identified by the use of integrative clustering. One subtype, known as cluster 10, exhibited deletions of chromosome 5q. That deletion was closely associated with several signaling molecules, transcription factors and cell-signaling genes. This group was characterized by trans
Another subtype, known as cluster 4, had cases with generally favorable outcomes between both ER+ and ER- cases and was essentially devoid of CNAs because it did not have much variance in copy numbers. This was the largest subtype present and accounted for about 16% of all cases (n=167).
These classifications also led to the discovery of new genes associated with different subtypes.
“Our work provides a definitive framework for understanding how gene copy number aberrations affect gene expression in breast cancer and reveals novel subtypes that should be the target of future investigation,” the authors wrote.
Curtis C, Shah SP, Chin S, et al. The genomic and transcriptomic architecture of 2,000 breast tumors reveals novel subtypes. Nature. 2012; Epub 18 April.