Ambient Air Pollution and Lung Cancer: Nature and Nurture

During a typical day, the average adult inhales about 10,000 L of air. Consequently, even the carcinogens present in the air at low concentrations are of concern as a risk factor for lung cancer in large populations. Outdoor (ambient) air can contain a number of hazardous agents, and many of these are generated by the combustion of fossil fuels, including carcinogens such as polycylic aromatic hydrocarbons and metals such as arsenic, nickel, and chromium. Depending on the pollution sources, the constituents of “air pollution” vary by locale and over time. Particulate matter (PM), which has multiple sources in urban air, has been studied the most as a potential lung cancer risk factor, and studies from around the world are generally consistent in finding increased cancer risk with increased exposure to PM ⩽2.5 μm in aerodynamic diameter. In 2013, the International Agency for Research on Cancer classified ambient air pollution as a Group 1 carcinogenic to humans (1). PM, a major component of air pollution, was evaluated separately and also classified as a Group 1 carcinogen—a strong signal to the international community to take immediate action to reduce exposures.

There is also growing evidence for heritable susceptibility to lung cancer. A meta-analysis of all multiple 28 case-control and 17 cohort studies found an approximately twofold increased risk of lung cancer associated with family history (2). Risk was generally higher for relatives of people in whom lung cancer was diagnosed at a young age and when multiple family members were affected.

Studies of risk for lung cancer among relatives of never-smokers are limited. Those studies do find some increased risk, but the association is usually weaker than among smokers. Germline mutations in the TP53 gene cause the inherited Li-Fraumeni syndrome. Individuals with this syndrome are at increased risk for many cancers, including lung cancers.

Studies of families with multiple relatives affected by lung cancer identified a region on chromosome 6q23-25 harboring a susceptibility region in families that had four or more affected relatives in two or more generations (3). Haplotype studies indicate that light or heavy smoking conferred high risk, demonstrating that the individuals in these families are particularly sensitive to tobacco exposure. Rare deleterious cancer risk variants have also been described recently to significantly impact lung cancer risk (4).

As for common genetic variants and lung cancer risk, the region identified in early genome-wide association studies studies included a neuronal nicotinic acetylcholine receptor gene cluster comprising cholinergic receptor nicotine α 5 CHRNA5, CHRNA3, and CHRNB4 subunits. Since the 2008 studies, 51 susceptibility loci have been found for lung cancer among a variety of populations and ethnicities, each one accounting for a small to moderate proportion of risk, in smokers (5). Polygenic or genetic risk score (GRS) is a parameter that summarizes the estimated effect of many genetic variants on a person’s phenotype, typically calculated as a weighted sum of disease-associated alleles. Recent evidence suggests that an individual’s genetic background may inform the optimal lung cancer low-dose computed tomography screening strategy (6). To date, however, there is little information on the combined effect of genetic risk factors and environmental factors, such as ambient air pollution, while accounting for smoking.

In this issue of the Journal, Huang and colleagues (pp. 817–825) conducted a study using the UK Biobank cohort of over 455,000 participants, 95% of whom are of European descent (7). The study has the advantage of a large size, increasing the power to examine both main effects and gene–environment interactions while adjusting for multiple comparisons. Data on exposure to common air pollutants were available as well as data on covariates and potential confounders such as smoking and obesity. In addition to recapitulating the association between air pollution and lung cancer, the authors calculated a polygenic risk score utilizing 18 single-nucleotide polymorphisms. The higher exposure category of pollution was associated with a 63% increased risk of lung cancer and the higher GRS with a 50% increased risk. More importantly, the air pollution–lung cancer risk existed across GRS categories, and there was additive interaction between these two parameters. Hence, compared with those with low GRS and low air pollution, the high GRS and high pollution group had the highest risk of lung cancer. This occurred using multiple measures of pollution. Of note, exposures below the current regulated threshold levels were significantly linearly associated with lung cancer risk.

The study is not without limitations. The exposure measures were based on a single sampling at baseline. Therefore, there is a potential for exposure misclassification during follow up. Moreover, this cohort design may be prone to left truncation, whereby subjects at risk before baseline do not remain observable until the start of follow up, leading to potential bias in the presence, direction, or magnitude of an association. Some other sources of lung carcinogen exposure were not characterized, such as occupational exposures. Finally, the smoking variables were simple, with 15% missing and imputed, although sensitivity analysis found robust results. The generalizability to other world populations, with different population admixture, requires further study.

Nevertheless, the results are impressive and highlight the importance of improving air quality, even below current thresholds values. Although additional gene–environment interaction studies such as this one are needed to identify specific interactions, the highest priority should remain enforceable societal-level actions to reduce exposure to air pollution globally. Susceptibility to diseases such as lung cancer is conferred not only by heritable factors but also acquired ones, such as comorbidities, age, and socioeconomic disparities. The most effective public health policy is to protect those among us who are more vulnerable, thus affording a wider umbrella of protection for the entire population.