Study identifies EGFR T790M variant as a risk factor for lung cancer in nonsmokers, with 25-fold higher risk

A large-scale study has identified a genetic variant that may help explain why people who have never smoked develop lung cancer.

According to a cover paper published on the 17th in the international journal Science, a team led by Pasi A. Jänne of Harvard Medical School analyzed the association between the EGFR T790M variant and lung cancer risk using genotyping data from more than 3.3 million people from the genetic testing company 23andMe.

The study found that the variant was rare, occurring in about 1 in 15,850 people, but carriers had a 25.18-fold higher risk of developing lung cancer than noncarriers.

In particular, the variant showed no significant association with 17 other cancers or noncancerous lung diseases, suggesting it is essentially a genetic risk factor specific to lung cancer.

The researchers said the variant acts independently, without clear interactions with other genetic predispositions.

Its effect also differed by smoking status. Carriers who had never smoked had a higher relative risk, raising the possibility that the variant is a previously overlooked genetic cause of lung cancer in families with a history of the disease among nonsmokers.

The variant also had a distinct regional distribution. It occurred in about 1 in 8,920 people in U.S. families, and its frequency rose to about 1 in 2,078 in the southern Appalachian region through a founder effect about 200–225 years ago.

Later, population movements during the U.S. Civil War spread the variant to groups with African and Native American ancestry.

The researchers added that testing for the variant could help identify high-risk families and guide prevention strategies, such as low-dose CT screening, for people with a family history of lung cancer, multiple primary lung cancers, lung nodules or ancestry from the southeastern United States.

However, the researchers noted that the study was limited by the predominantly European ancestry of the 23andMe data, and that separate studies are needed to determine the variant’s frequency in other regions, including East Asia.