by ecancer reporter Clare Sansom

Approximately sixteen percent of breast tumours are classified on diagnosis as primary triple-negative, meaning that none of three signature genes – the oestrogen receptor, the progesterone receptor and the herceptin (HER2) receptor – is amplified in the tumour tissue.

 In general, triple negative breast cancer has a relatively poor prognosis, with lower five-year survival rates than other types of breast cancer regardless of the stage at diagnosis. 

It is known to be genetically heterogeneous, but its development is still poorly understood.

A large group of researchers led by Samuel Aparicio, Sohrab Shah and Marco Marra from the University of British Columbia, Vancouver, Canada, and Carlos Caldas from the Cancer Research UK Cambridge Research Institute, Cambridge, UK has now sequenced the genomes and transcriptomes from 104 cases of primary triple negative breast cancer, sub-classified into basal-like and non-basal subtypes.

 All tumour samples were obtained at the time of diagnosis and the sequence data was annotated with clinical information.

A total of 2,414 somatic mutations were identified in these tumours and validated using deep sequencing. 

Alterations in the copy number of genes were also identified in the tumour tissue. The number of somatic mutations observed in each tumour was seem to be extremely variable, and the number of mutations was not correlated with the percentage of the tumour genome affected by copy number alteration.

 Copy number variation was observed in a number of known oncogenes and tumour suppressor genes, and intragenic deletions were observed in the tumour suppressor PARK2, linking this gene to triple negative breast cancer for the first time. Comparisons of the sequences of the tumour genomes with those of the transcriptomes showed that only 36% of validated somatic mutations were found in the expressed RNA sequences.

The researchers determined the enrichment patterns of the observed mutations in the tumour genomes by examining the mutation frequencies of single genes, gene families and genes involved in the same expression networks. 

The tumour suppressor gene TP53 was found to be the most frequently mutated gene, mutated in 62% of basal-like and 43% of non-basal tumours. Frequent mutations were also observed in a number of other genes including kinases such as PIK3CA, the Usher syndrome gene USH2A and the retinoblastoma gene RB1.

Approximately 20% of the tumours had mutations associated with known clinical interventions, including the V600E mutation in the kinase BRAF.

The Reactome functional protein interaction database was used to search for gene families and functionally related gene sets with significantly over-represented mutations in the breast tumour genomes. 

Protein functions and pathways associated with significant numbers of mutations in this tumour set included p53-related pathways; chromatin re-modelling; PIK3 and ERBB signalling; growth hormone and nuclear receptor co-activation; and focal adhesion. None of these pathways and cellular processes were previously linked specifically to triple negative breast cancer although many of them are known to be associated with tumour progression. 

All exons of 29 genes with frequent mutations in these tumours were then sequenced in an additional 159 breast cancers, 82 oestrogen receptor positive and 77 oestrogen receptor negative, and many were found to also be mutated frequently in this more general breast cancer set.

The researchers then estimated the clonal evolution of the triple negative breast cancers by investigating the allelic abundance in each tumour. 

This was extremely varied; some tumours were found to consist of only one or two genetically distinct sub-clones and others to show extensive clonal evolution. In general, more clonal frequency modes were observed in basal-like than in non-basal tumours.

 Somatic mutations in TP53, PIK3CA and PTEN were observed in most clones, indicating that these are often, but not always, founder mutations. In contrast, mutations in genes involved in controlling cell shape and the cytoskeleton were found in fewer clones indicating that these genes mutate at a later stage of tumour progression.

Taken together, these results should help researchers to classify this heterogeneous disease further and determine the best treatment options for each sub-type.

Reference

Shah, S.P., Roth, A., Goya, R. and 56 others (2012). The clonal and mutational evolution spectrum of primary triple-negative breast cancers. Nature, published online ahead of print 4 April 2012. DOI: 10.1038/nature10933