NEW YORK – Researchers in France have confirmed an epidemiological association between air pollution and EGFR mutations in certain lung cancer patients.
At the European Society for Medical Oncology Congress in Barcelona, Spain, this week, Alexis Cortot, a professor of thoracic oncology at the Centre Hospitalier Régional Universitaire de Lille, described how his research team explored the relationship between air pollution and molecular characteristics of lung cancer using data from thousands of patients in France.
The findings, he said, raise questions about whether lung cancer screening strategies, which are largely based on smoking history, should also factor in patients' exposure to air pollution. Additionally, he proposed that future studies should explore the association between air pollution and response-predictive biomarkers in lung tumors, such as STK11 mutations associated with poor response to immune checkpoint inhibitors.
An estimated 260,000 lung cancer deaths per year can be attributed to air pollution, mainly due to particulate matter less than 2.5 micrometers in diameter, dubbed PM2.5. There is data showing that any increment of 5 micrograms per cubic meter increase in mean annual exposure to PM2.5 is associated with greater risk of dying from lung cancer and other diseases. "We all know that the air we breathe can cause lung cancer," Cortot said.
However, plenty of people exposed to air pollution don't develop lung cancer, and scientists don't fully understand the biological processes that lead to malignancy or how PM2.5 impacts the characteristics of lung tumors. Two years ago at the same meeting, researchers led by the Francis Crick Institute's Charles Swanton made a major advance with their finding of a significant association between levels of PM2.5 and the incidence of EGFR-mutated lung cancer in never-smokers, based on an analysis of nearly 447,000 lung cancer cases in the UK Biobank.
Swanton's group didn't show that air pollution directly causes DNA to mutate, but rather that it contributes to cancer development by triggering cells already carrying cancer-linked mutations. When they exposed mice with EGFR-mutated and KRAS-mutated cells in their lungs to PM2.5 levels typically found in cities, they found that the mice were more likely to develop cancer in cells carrying these mutations. In these models, "they showed very nicely that EGFR mutations are already present … in healthy lung tissues, and … PM2.5 promotes the transformation of EGFR-mutant cells through non-genomic mechanisms involving IL1β and macrophages," Suzette Delaloge, associate professor of medical oncology at Gustave Roussy, who was not involved in Cortot's research, said at the meeting. "This is quite specific for EGFR but they also showed [an association] for KRAS but with less impact."
In the study presented at the ESMO Congress, Cortot and colleagues aimed to confirm these earlier findings using data from lung cancer patients in France, but also to expand the field's understanding of the role of air pollution in lung cancer. Specifically, they explored whether there is a link between EGFR mutations and other types of pollution, if radon exposure plays a role, if air pollution is associated with other oncogenes, and if air pollution has an impact on lung cancer biomarkers predictive of immunotherapy response, such as PD-L1 expression and STK11 mutations.
To do this analysis, Cortot and colleagues sifted through data from nearly 9,000 lung cancer patients diagnosed in 2020 at 82 public non-academic hospitals in France, representing 20 percent of all lung cancer cases diagnosed in the country that year. They focused on patients with non-squamous NSCLC with a known home zip code, so they could use data collected by the French National Institute for Industrial Environment and Risks to determine their exposure to air pollution, including PM2.5 , particulate matter that is 10 micrometers or less in diameter (PM10), nitrogen dioxide (NO2), and ozone (O3). They also calculated radon exposure according to data from the French National Research and Safety Institute. In addition, they gathered data on patients' PD-L1 expression status based on testing performed at their local cancer center.
Ultimately, around 5,800 non-squamous NSCLC patients met the inclusion criteria for the analysis, 500 of whom had an EGFR mutation, 1,068 a KRAS mutation, and 156 an STK11 mutation. The researchers also identified NSCLC patients with mutations in other oncogenic driver genes, such as ALK, BRAF, HER2, ROS1, and MET. Approximately 42 percent of patients were negative for PD-L1 expression; 31 percent had PD-L1 expression in less than half of tumor cells; and 27 percent had expression levels in at least half their tumor cells.
Overall, levels of PM2.5 were low across the regions of France represented in the study, averaging less than 10 micrograms per cubic meter. As expected, PM2.5 levels were the highest in large cities like Paris, where there is more exhaust from cars and other sources.
Cortot and colleagues conducted a statistical analysis and found, just as Swanton's group had, a significant correlation between EGFR-mutated NSCLC and exposure to air pollution. "Patients living in the most [PM2.5] exposed areas have a greater chance of harboring an EGFR mutation than those living in the less exposed areas," Cortot said, adding that the association appeared to hold strong regardless of the type of common driver EGFR mutations patients had.
"These data are pretty clear-cut," Delaloge said, reviewing the findings at the meeting. "This is real-world [data] but very high quality."
However, Cortot's group didn't find any association between KRAS-mutated lung cancer and exposure to PM2.5. This may reflect the fact that tobacco exposure, rather than air pollution, may have a more pronounced impact on the development of KRAS-mutated lung cancer, Cortot suggested. There were also no significant associations between PM2.5 exposure and other driver oncogenes, which may be because the study wasn't sufficiently powered to detect these associations, he noted.
Looking at other types of air pollutants, the researchers found similar associations between EGFR-mutant NSCLC and PM10 and NO2, which are linked to PM2.5, but they found an inverse association with O3 exposure and EGFR-mutated lung cancer, which Cortot said was expected because "ozone is inversely correlated to nitrogen dioxide because of chemical interactions." Radon exposure was not associated with the presence of any driver oncogenes.
Looking at lung cancer biomarkers predictive of response to immunotherapies, Cortot and colleagues reported no significant impact of air pollution and the presence of PD-L1 expression in NSCLC. They did find an association between STK11 mutations and PM10 exposure but not with other pollutants. Delaloge acknowledged this was an interesting finding but noted that the association between STK11 mutations and PM10 exposure was statistically weak.
The results from Swanton's and Cortot's research groups come as the World Health Organization has issued recommendations for reducing PM2.5 levels down to specific targets that would improve air quality and reduce health risks. "These results are an incentive to pursue pollution reduction efforts worldwide, at both collective and individual levels," Cortot said.
Cortot further wondered if it was time to consider air pollution exposure to determine who should undergo lung cancer screening. Delaloge pointed out that there is emerging data in mice and humans showing that carcinogens are associated with specific mutational signatures that can be useful in prevention strategies, but these data are not being applied today.
She agreed that the confirmation of the link between PM2.5 and EGFR-mutated NSCLC presents the healthcare community with an "actionable carcinogenic pathway" and suggested the field explore potential new ways of detecting lung cancer beyond CT scans and opportunities for new therapeutic strategies.
Given the association with STK11 mutations, the new study also raises questions about the impact air pollution may have on patients' ability to respond to treatments. "This is the next step for this study," Cortot said, noting that the datasets his team has been working with contain information on the treatments lung cancer patients received and their survival.
"This is a key question when you see there may be an impact [of air pollution] on STK11 mutations, which is associated, at least when KRAS mutations are also present, with resistance to immunotherapy," Cortot said. "And we also need to look at response to EGFR [tyrosine kinase inhibitors and], whether there is some different sensitivity based on if the patient has been exposed or not to PM2.5."