A rare inherited EGFR variant was strongly associated with lung cancer

The discovery

The researchers asked how strongly an inherited EGFR T790M variant raises lung-cancer risk, whether its association persists among never-smokers, whether it is specific to lung cancer and why the variant is more common in parts of the southeastern United States.

The research question and why it matters

A 2005 family report first linked inherited T790M to multiple cases of non-small-cell lung cancer, and later family registries described a hereditary syndrome concentrated among never-smokers. T790M is also well known as a tumor-acquired mechanism of resistance to older EGFR-targeted drugs. Previous samples were too small to estimate population risk precisely; the new work uses a very large research database to quantify the association and trace a founder lineage.

What researchers found

The variant occurred in about 1 of every 15,850 participants of European ancestry and was associated with lung cancer at an odds ratio of 25.18. Among never-smokers the odds ratio was 61.7; among people who had smoked it was about 10. The study found no increased risk across 17 other common cancers and no interaction with the tested polygenic risk score. Most U.S. carriers shared a lineage consistent with a Southern Appalachian founder event about 200 to 225 years ago, with local prevalence reported as high as roughly 1 in 2,000.

Results at a glance

Key results from the tested systems

3,372,531

research participants

The primary analysis combined genetic and lung-cancer information.

641

variant carriers

A huge cohort was needed because inherited T790M is rare.

OR 25.18

overall association

Carrier odds of lung cancer relative to noncarriers.

OR 61.7

never-smoker association

A relative estimate—not absolute lifetime risk.

How the research worked

Researchers compared lung-cancer histories between carriers and noncarriers in a large de-identified 23andMe research cohort, then stratified the analysis by smoking history and tested associations with 17 other cancers and a polygenic risk score. They used haplotypes, ancestry and geography to reconstruct a founder lineage, examined transmission within affected families and incorporated clinical information from the INHERIT Study.

Subjects or systemHuman

Research designPopulation genetic association study with ancestry and geographic analyses, family transmission testing and corroboration in a prospective inherited-risk cohort

Evidence base3,372,531 23andMe research participants with genetic and lung-cancer information, including 641 EGFR T790M carriers; supporting evidence included families and participants in the INHERIT Study (NCT05587439)

How to interpret this design

This design can measure patterns and associations in the observed population. It cannot, by itself, prove that the exposure caused the outcome because unmeasured differences, reverse causation and selection effects may contribute.

Because the research involved people, it speaks directly to the participants and outcomes measured. It may still apply differently to people outside the eligibility criteria, age range, clinical setting, geography or follow-up period.

What strengthens or limits the finding?

The exceptionally large cohort made a very rare variant measurable, the association was large and statistically strong, and family and prospective-cohort evidence pointed in the same direction. The analysis relied heavily on a voluntary consumer-genetics cohort, used odds rather than prospective lifetime incidence and has not shown that testing or CT surveillance improves survival.

The result is meaningfully informative, but identifiable limitations could alter the size, reach or causal interpretation of the finding.

Funding and disclosure context

The recorded funding source is: LUNGevity Foundation; Elaine & Gerald Schuster Fund; NIH grants R01 HG012133, U01 CA209414 and R35 CA220497; American Cancer Society grant CRP-17-111-01-CDD; Chen-Huang Center for EGFR-Mutant Lung Cancers; GO2 for Lung Cancer; Addario Family; and multiple family and private foundations listed in the institutional report. The complete conflict-of-interest declaration should be checked in the original publication rather than inferred. Funding or a disclosed relationship does not by itself invalidate a result, but it is relevant when judging design choices, analysis and the need for independent replication.

What it means

The study turns a hereditary-lung-cancer signal known from individual families into a population-scale risk estimate. It may help genetic counselors recognize families who warrant evaluation and provides a rationale for trials of tailored surveillance. Current routine lung-cancer screening remains based mainly on age and smoking history; this study alone does not establish a new screening standard.

Beyond the abstract

Deeper analysis

Relative risk sounds larger than absolute risk

A strong odds ratio can coexist with many carriers never developing disease, especially for a rare outcome. Clinical decisions require penetrance by age, sex, ancestry and exposure—not only a single relative estimate.

Smoking does not reverse the genetic finding

The variant's relative association was smaller among smokers because smoking raises the comparison group's baseline risk. The study found both exposures harmful; it did not find a protective interaction.

Scale made a rare variant visible

Traditional biobanks contained only a handful of carriers. Hundreds became available only in a cohort of millions, allowing risk, geography and shared ancestry to be studied together—but tying the analysis to the strengths and biases of a commercial research population.

Screening is the next question, not the result

Finding a high-risk group is necessary before testing targeted surveillance, but it is not enough to recommend it. Trials must show that earlier CT imaging changes meaningful outcomes and that benefits exceed false positives, radiation and downstream procedures.

Keep the claim in proportion

What it does NOT prove

It does not mean a carrier has a 25-fold or 62-fold absolute lifetime probability of developing lung cancer; those figures are odds ratios relative to noncarriers.
It does not show that smoking protects carriers—the smaller relative association among smokers reflects their already elevated background risk.
It does not prove that broad population genetic testing or more frequent CT scans improve survival or produce more benefit than harm.
It does not explain why many carriers never develop lung cancer or why the variant appeared specific to lung cancer in this analysis.
It does not justify interpreting a consumer-genetics result without clinical confirmation and professional genetic counseling.

Important limitations

23andMe participants are self-selected and are not demographically representative of the United States or world population.
The variant is extremely rare, so even a cohort of 3.37 million contained only 641 carriers and smaller smoking subgroups.
Cancer and smoking information in the consumer cohort may include self-report and incomplete clinical detail.
Odds ratios do not directly provide age-specific penetrance or an individual's absolute lifetime risk.
Founder history and European-ancestry concentration limit how precisely prevalence estimates transfer to other populations; the authors call for more diverse cohorts.
The accessible records did not provide complete conflict-of-interest declarations, while the main dataset and many authors came from 23andMe.

How this fits with previous research

Questions still unanswered

What are the age-specific and lifetime absolute risks for carriers?
Which environmental exposures or additional variants determine why some carriers develop cancer and others do not?
Can targeted CT surveillance detect disease earlier and reduce mortality without excessive radiation, false positives or overdiagnosis?
How common and consequential is T790M in populations underrepresented in consumer-genetics databases?
Do other rare inherited EGFR variants create comparable risk?

Government verification and context

Relevant U.S. government resources

These resources serve different purposes. A registry can verify what researchers planned, a repository can locate government-funded work, and an agency page can supply authoritative background. None automatically proves that this paper's conclusion is correct.

Study verificationNational Library of Medicine / NIH

PubMed record search ↗

Federal biomedical-index search keyed to this paper's DOI or title. It can confirm indexing and expose linked identifiers when a record is available; the journal paper remains the primary source.

Research registryNational Library of Medicine / NIH

ClinicalTrials.gov registry search ↗

A trial registry describes the planned design, outcomes and enrollment. Registration improves transparency, but it does not establish that a treatment works or that published reporting is complete.

Authoritative contextNational Institutes of Health

Understanding clinical research ↗

NIH background on how clinical research is designed, reviewed and interpreted. This is contextual guidance, not independent confirmation of the study's result.

Reuse note: Facts and discoveries are summarized here in original language. We link to government material instead of copying it wholesale, and we do not reuse agency logos, photographs, charts or third-party material unless the specific reuse rights are verified.

Sources and provenance