Researchers have identified strong genetic relationships between major cancers, using data from some of the largest genome-wide association studies of cancer to date.
The study, published today in Nature Communications, sought to quantify the genetic correlations and heritability—the extent to which disease risk is due to genetics—of six solid cancers and 38 non-cancer traits. It was led by genetic researchers from all over the world, including University of Washington School of Public Health Assistant Professor of Epidemiology Sara Lindström.
“This study shows that it makes sense to study multiple cancers together because they, to some extent, have the same underlying biology,” said Lindström, senior author of the study. “This will help us understand the biological pathways that lead to cancer.”
Lindström and her colleagues looked at the genetic data for 296,215 participants diagnosed with breast, colorectal, head/neck, lung, ovary, and prostate cancer and 301,319 controls free of cancer, all of European ancestry. The researchers confirmed a link between lung and head/neck cancers, two cancers associated with tobacco use, and observed for the first time a significant shared genetic basis between breast and ovarian cancer, two cancers known to share the BRCA1/2 mutations.
The study also found statistically significant genetic links between colorectal and lung cancer, breast and lung cancer, and breast and colorectal cancer. Researchers saw no correlation between ovarian and prostate cancer, which were previously thought to be related cancers.
The researchers also looked at genetic correlations between each cancer and non-cancer traits such as smoking, psychiatric diseases, body mass index (BMI), education, and sleep, among other traits. Some of the study’s findings confirmed previous medical knowledge. For example, there were strong genetic links between lung and head/neck cancer and smoking, which is a known primary risk factor for those cancers. Colorectal cancer had genetic links to BMI, and breast cancer shared a genetic basis with schizophrenia. In contrast, there was no evidence that linked the six cancers to commonly thought risk factors such as sleep characteristics or cardiovascular traits like hypertension, blood pressure or coronary artery disease.
Finally, the researchers also investigated if specific functions of the human genome contributed to cancer risk more than expected. They found that regions that have been preserved throughout evolution contribute to 25 percent of cancer heritability, despite only making up 2.6 percent of the human genome.
“This study was a stepping stone for us,” said Lindström. “Because there is a shared genetic basis to cancer, why not combine data from multiple cancers to help us find new genetic variants that contribute to cancer risk?”
Contributing researchers from the University of Washington (UW) include Lisa Newcomb, deputy director of the Canary Prostate Active Surveillance Study with Fred Hutch and the UW; Polly Newcomb, research professor of epidemiology; Ulrike Peters, research professor of epidemiology; Amanda Phipps, assistant professor of epidemiology; Mary Anne Rossing, research associate professor of epidemiology; Janet Stanford, research professor of epidemiology; and Emily White, professor emeritus of epidemiology.