Genomics tool could help predict tumour aggressiveness, treatment outcomes

A new method for measuring genetic variability within a tumour might one day help doctors identify patients with aggressive cancers that are more likely to resist therapy, according to a study led by researchers now at The Ohio State University Comprehensive Cancer Center – Arthur G. James Cancer Hospital and Richard J. Solove Research Institute (OSUCCC – James).

Researchers used a new scoring method they developed called MATH (mutant-allele tumour heterogeneity) to measure the genetic variability among cancer cells within tumours from 305 patients with head and neck cancer. High MATH scores corresponded to tumours with many differences among the gene mutations present in different cancer cells.

Cancers that showed high genetic variability – called ‘intra-tumour heterogeneity’ – correlated with lower patient survival. If prospective studies verify the findings, MATH scores could help identify the most effective treatment for patients and predict a patient’s prognosis.

Researchers have long hypothesized that multiple sub-populations of mutated cells within a single cancer lead to worse clinical outcomes; however, oncologists do not use tumour heterogeneity to guide clinical care decisions or assess disease prognosis because there is no single, easy-to-implement method of doing so in clinical practice.

To address this need, James Rocco, MD, PhD, and his colleagues developed MATH to make it easier for doctors to measure genetic variability in patients’ tumours and to help guide treatment decisions.

The new findings confirm that high genetic variability with a patient’s tumour is related to increased mortality in head and neck squamous cell carcinoma.

‘Genetic variability within tumours is likely why people fail treatment,’ says Rocco, Professor and John and Mary Alford Chair of Head and Neck Surgery and Director of the OSUCCC – James Division of Head and Neck Oncologic Surgery. ‘In patients who have high heterogeneity tumours it is likely that there are several clusters of underlying mutations – in the same tumour – driving the cancer. So their tumours are likely to have some cells that are already resistant to any particular therapy.’

For the current study, Rocco and his team used the MATH tool to analyse retrospective data from 305 head and neck squamous cell carcinoma patients from The Cancer Genome Atlas (TCGA). This National Institutes of Health repository of publicly available data was launched in 2006 as a pilot project and now includes samples from more than 11,000 patients across 33 tumour types. The MATH score was calculated from data obtained by TCGA with a genome sequencing technique called whole-exome sequencing.

Researchers confirmed that high intra-tumour heterogeneity was related to increased mortality in this sub segment of patients. Each 10 percent increase in MATH score corresponded to an 8.8 percent increased likelihood of death.

The relationship between MATH score and mortality was not dependent on HPV (human papilloma virus) status or other molecular characteristics of the tumour.

‘Our retrospective analysis showed that patients with high heterogeneity tumours were more than twice as likely to die compared to patients with low heterogeneity tumors,’ says Rocco. ‘This type of information could refine the dialogue about how we tackle cancer by helping us predict a patient’s treatment success and justify clinical decisions based on the unique makeup of a patient’s tumor.’ EurekAlert