An inherited EGFR variant carried 62-fold lung cancer odds among never-smokers

Across the whole cohort, smokers and nonsmokers together, carriers had about 25 times the odds of a lung cancer diagnosis. That average hides an odd internal structure. Put the strata on one scale, with a nonsmoker who lacks the variant as 1. Smoking alone sits at 4, a never-smoking carrier at 62, and a carrier who smoked at 11 times other smokers, which is about 44 against the same baseline. Two risks multiplying would have predicted 248. Carriers who smoked came out lower than carriers who never did. The subgroup counts are small, and the published accounts do not report confidence intervals or absolute risks, so 44 against 62 does not establish that smoking is neutral in a carrier.

The frequency figures need care too. The reported carrier rate covers more than 3 million participants of European ancestry. Divide the full 3.3 million consented cohort by the 641 carriers found and you get roughly 1 in 5,150. That is about three times as common as the reported rate, so the two numbers do not share a denominator. Everyone counted bought a consumer genetic test and consented to research use, and the main analysis was restricted to European ancestry. The 62-fold estimate has not been shown to hold in other ancestries.

The geographic pattern came from where participants were born. The researchers traced the variant to settlers from the British Isles who arrived in the early 1700s, and to families who later moved into southern Appalachia and stayed relatively isolated, passing it down within the area. In Alabama, Mississippi and Tennessee it runs about 7.6 times the overall rate.

Stephen Chanock, who directs the National Cancer Institute's Division of Cancer Epidemiology and Genetics and was not involved in the study, said the concentration of carriers may help explain the Southeast's elevated lung cancer rates. About 20% of adults in Appalachia report smoking, against 16% of adults elsewhere in the US, and the two factors may interact. A study of genotype and birthplace cannot apportion the region's excess between a founder variant and a higher smoking rate.

Current eligibility for annual low-dose CT is drawn from smoking history. "The prevalence of the T790M variant and its impact on lung cancer risk has previously been poorly understood," said Chris Amos, a genetic epidemiologist at Baylor College of Medicine who was not involved in the work. He called the result a "very important finding". Amos said the earlier age at diagnosis means screening needs to start earlier and to proceed regardless of smoking status.

How much of the never-smoker problem this explains is a separate question. Chanock said the variant is rare enough that it likely does not account for a large percentage of overall lung cancer cases. STAT reported experts saying it likely plays a role in only a small portion of never-smoker cases. The variant showed no link to any other type of cancer or to noncancerous lung conditions. The odds ratios describe diagnoses; mortality under earlier scanning is a different measurement. STAT also reported that among people diagnosed with lung cancer, the share who never smoked has been increasing.