Should we screen for lung cancer in some nonsmokers?

Lung cancer in people who have never smoked is no longer a statistical footnote; it’s an increasingly visible clinical reality with concentrated risk in specific groups. Still, the evidence does not yet justify universal screening of all never-smokers. However, we may want to prioritize those with the highest pretest probability.

Overall lung cancer mortality has dropped as smoking has waned, but a sizable slice of cases now occur in people who have never smoked—roughly 10–25% worldwide, and up to ~50% in some Asian and Asian-American women (i.e. 50% of lung cancer cases in Asian women may occur in never-smokers, Figure 1). The burden is not evenly distributed: women, people of Asian ancestry, and even adults under 50 show a disproportionate impact, and a first-degree family history further elevates risk. These facts challenge the view that equate lung cancer almost exclusively with smoking and underscore the need to think differently about detection.

The current USPSTF (U.S. Preventive Services Task Force) recommendations for lung cancer screening are adults ages 50–80 with ≥20 pack-years (smoking one pack of cigarettes per day for 20 years). USPSTF is an independent panel of experts that evaluates scientific evidence on the effectiveness of clinical preventive services. However, these criteria were designed around smoking risk and therefore miss most never-smoker cases, which tend to surface at later, less curable stages. Taiwan, on the other hand, now offers low-dose CT (LDCT) to never-smokers with a family history of lung cancer, creating an instructive template for risk-based eligibility. Screening involves LDCT which uses much less radiation than a standard diagnostic CT scan. The goal is to find lung cancers when they are small and localized, making them more treatable and improving survival chances.

Biologically, never-smoker tumors tend to look different from those in long-time smokers, often featuring actionable driver mutations (QH) such as EGFR. In addition, analyses link higher PM2.5 (inhalable fine particulate matter/pollution less than 2.5 microns in diameter; QH) exposure to increased mutations, evidence that environment and biology intersect in ways that could inform risk stratification.

A recent Taiwan study (ages 55–75, never-smokers with ≥1 risk factor) found a 2.6% baseline cancer detection rate (2.1% invasive; 0.5% AIS = adenocarcinoma in situ, a pre-invasive stage of lung adenocarcinoma that can often be cured with surgery) and a striking shift toward earlier (less dangerous) Stage I tumors. Risk was adjusted by family history, female sex, and age >60.

Among never-smokers, air pollution (especially long-term exposure to fine particulate matter e.g. PM2.5) likely carries the largest population-level burden because exposure is widespread. Secondhand smoke has a modest individual effect size compared with active smoking. Genetics and family history often exert the strongest signal for a given individual even though they account for a smaller share of cases than environmental exposures. In short, air pollution likely dominates at the population level, secondhand smoke adds a real but smaller increment shaped by context, and genetics/family history most sharply identify which never-smokers warrant priority for targeted screening.

The potential upside of screening is likely to be clinically meaningful: earlier stage at detection enables curative surgery and timely targeted therapy, and screening yield improves when you focus on groups with higher baseline risk (e.g., family history, older age, some Asian/Asian-American women). Counterbalancing harms include false positives (and the downstream anxiety, procedures, and complications), overdiagnosis of indolent lesions like AIS, and radiation exposure from imaging (albeit low with LDCT). Most importantly, because no large never-smoker RCTs exist, the mortality benefit remains to be demonstrated.

One option is to offer LDCT to never-smokers ≥55 with a first-degree family history, and consider carefully monitored substudies that lower the age threshold for Asian/Asian-American women. Other groups eligible for screening may include long-term residents of high-PM2.5 regions and people with genetic/biologic risk markers—but thresholds, durations, and biomarkers need validation before clinical rollout.

In summary, we are not ready to screen all never-smokers; the costs would outweigh the benefits. However, we are ready to consider narrow, risk-based LDCT screening for never-smokers with a first-degree family history, and, in carefully monitored settings, for Asian/Asian-American women at appropriate ages—backed by rigorous data capture and clear stopping rules.

Figure 1. "Annie Chen was diagnosed with lung cancer after two years of reporting persistent shortness of breath." She was 48 when she was diagnosed with Stage 4 lung cancer (Shuran Huang for The New York Times).