NEW YORK (GenomeWeb News) – An international research group led by investigators at the Ontario Cancer Institute has identified a gene expression signature in histologically normal tissue at the periphery of surgically removed oral cancers that coincides with cancer recurrence.
Using cancerous and non-cancerous samples from two-dozen individuals with oral squamous cell carcinoma — combined with a meta-analysis of gene expression data for almost 200 more samples from five datasets — the team looked for genes with elevated expression in cancer samples.
As they reported in BMC Cancer, the search led them to four genes that were differentially expressed in oral cancer samples, but also in some apparently normal tissue at the edges of surgically resected tumors. And, they found, over-expression of the genes in this histologically normal tumor margin tissue corresponded to oral cancer recurrence risk.
"Over-expression of the [four]-gene signature in histologically normal surgical margins was validated and highly predictive of recurrence in an independent patient cohort," co-corresponding author Suzanne Kamel-Reid, a researcher affiliated with the Ontario Cancer Institute and the University of Toronto, and co-authors wrote. "Our findings may be applied to develop a molecular test, which would be clinically useful to help predict which patients are at a higher risk of local recurrence."
At the moment, researchers explained, the risk of oral squamous cell carcinoma is typically predicted based on histological analyses of tissue at the edges of surgically resected tumors. But even when seemingly normal tissues are present at these surgical margins, they noted, cancer still returns to the same area in some 10 to 30 percent of cases.
Consequently, they speculated that it may be possible to find genetic signals in tissue from the margins of surgical resections and/or in the oral tumors themselves that might more accurately gauge recurrence risk.
"[M]olecular analysis of histologically normal resection margins and the corresponding [oral squamous cell carcinoma] may aid in identifying a gene signature predictive of recurrence," the study authors explained.
To explore this possibility, the researchers used Affymetrix HG-U133A 2.0 plus oligonucleotide microarrays to assess gene expression patterns in 96 cancerous and non-cancerous (histologically normal) oral tissues collected from 24 individuals who'd been surgically treated for oral squamous cell carcinoma at the Toronto General Hospital.
All of the individuals tested had tumors with resection margins that had been deemed histologically normal.
To this data, the team added information from a meta-analysis that they did using publicly available raw gene expression data for 141 oral squamous cell carcinoma samples, 38 nearby normal tissues, and 20 unaffected tissues from five published datasets.
In the process, they found 138 genes whose expression was at least twice as high in the oral cancer samples than in non-cancerous tissue samples.
By comparing gene expression in tumor, margin, and normal oral tissues, they then narrowed in on four genes whose over-expression tissue at the margins of resected tumors corresponded to cancer recurrence: MMP1, COL4A1, P4HA2, and THBS2.
"This signature is based on genes found to be consistently over-expressed in [oral squamous cell carcinoma] as compared to normal oral mucosa," the researchers noted, adding, "these genes are also over-expressed in a subset of histologically normal surgical resection margins, and their over-expression in such margins provides an indication of the presence of genetic changes before histological alteration can be detected by histology."
Indeed, the team went on to verify the predictive value of this four-gene expression signature using quantitative real-time PCR for another 136 samples from 30 individuals treated for oral squamous cell carcinoma at the Toronto General Hospital.
Data from the study is available through the NCBI Gene Expression Omnibus.