Probing Biological Sex Differences in Cancer

An extensive analysis of data from The Cancer Genome Atlas (TCGA) has revealed considerable differences, at the molecular level, between men and women with certain cancers. The findings could help explain why the sexes vary in terms of disease incidence, prognosis, and response to therapy—well-documented cancer disparities that have not been systematically investigated before (Cancer Cell 2016;29:711–22).

“We needed a large-scale data set with sufficient samples to enable rigorous statistical control for confounding factors such as race, age, and disease stage, which wasn't possible until TCGA's advent,” explains senior author Han Liang, PhD, an associate professor of bioinformatics and computational biology at The University of Texas MD Anderson Cancer Center in Houston.

Liang's team focused on 13 cancers in TCGA with molecular data on 30 or more samples from each sex. They found that these cancers could be separated into two categories. One, the strong sex-effect group, comprised eight cancers—including lung adenocarcinoma and head and neck squamous cell carcinoma (HNSCC)—with unequal male-to-female incidence and mortality ratios. Five cancers, including glioblastoma multiforme and acute myeloid leukemia, were in the weak sex-effect group, which had more balanced incidence and mortality ratios.

The researchers uncovered a variety of sex-biased somatic mutations and copy-number changes in cancers in the strong sex-effect group. For example, men with lung adenocarcinoma had a higher frequency of STK11 mutations: This kinase activates the AMPK pathway, and mutations in its gene are thought to predict sensitivity to the mitochondrial inhibitor phenformin, at least in mouse models of lung adenocarcinoma. Meanwhile, PIK3CA was amplified more often in women with clear-cell renal cell carcinoma, potentially influencing sensitivity to mTOR inhibition; in the same cancer in men, PDCD1, which encodes the immune checkpoint protein PD-1, was more frequently deleted.

The team then scrutinized 114 clinically actionable genes, including 86 whose alterations are targeted by FDA-approved drugs. They reported that 53% showed sex-biased differences in mRNA and protein expression, among other molecular signatures; these 60 genes were almost all found in cancers of the strong sex-effect group. For instance, women with HNSCC had higher levels of the SRC protein.

This finding could be relevant to the failure of two clinical trials assessing the SRC inhibitor dasatinib (Sprycel; Bristol-Myers Squibb) for the disease, Liang notes, because less than half of the patients in each study were women. Meanwhile, EGFR showed female-biased mRNA expression in lung adenocarcinoma, which he says is consistent with the higher response rate to EGFR inhibitors in women.

Biological sex is often not explicitly considered when treating cancer, Liang says. “Although this may be appropriate for cancers in the weak sex-effect group, the strong group merits special consideration in terms of drug development and clinical practice.”

The next steps for Liang's team include validating the sex-biased molecular signatures unveiled in this study, and extending their analysis to encompass more cancers.

“There are certainly cancers in which biological sex is a prognostic factor, so this is a very intriguing study,” says Michael Davies, MD, PhD, deputy chair of melanoma at MD Anderson, who wasn't involved in the research. “These candidates [sex-biased signatures] will need to be explored in greater detail, including in the context of systemic therapies. Additional confirmation may lead to the development of sex-specific therapeutic approaches.” –Alissa Poh

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/content/early/by/section.