Shining a light on lung cancer in non-smokers

We’ve known for many years that smoking is the main cause of lung cancer, and we understand a lot about why that is: tobacco contains a vast number of chemicals, many of which are ‘genotoxic’, i.e. they cause direct damage to DNA. Repeated exposure to inhaled tobacco smoke causes the cells lining the lungs to acquire mutations in their DNA. All cancer cells contain DNA mutations, but the sheer number of mutations in smoking-related lung cancers is astounding.

The image below shows results from sequencing the DNA in non-small cell lung cancer (NSCLC) cell lines from a non-smoker and a smoker. Each piece of grey text outside the coloured circle corresponds to a mutated gene. Although these are both lung cancers, they are biologically quite different diseases, and might therefore need different approaches to treatment:

Image credit: Govindan et al, 2012: Genomic Landscape of Non-Small Cell Lung Cancer in Smokers and Never-Smokers: Cell

The challenge, then, is to understand how and why lung cancer develops in people who have never smoked (or been chronically exposed to second-hand smoke, like the late Roy Castle). Lung cancer in never-smokers kills around 6000 people in the UK each year. Now that smoking rates have dropped to very low levels, overall lung cancer rates will fall over the coming years, meaning that tumours in non-smokers will begin to make up a larger proportion of all lung cancers. After smoking, air pollution is thought to be the second biggest risk factor for lung cancer, but scientists have not been certain of the biological mechanism behind this – as there is no evidence that air pollution directly damages DNA.

In the NDRS genomics team, we collect data from NHS labs that test lung tumours for DNA mutations. One gene that is often tested in NSCLC is the EGFR gene, which is mutated in around 10-12% of NSCLC tumours. EGFR is interesting for two reasons: firstly, if an EGFR mutation is present, the cancer can be treated with a drug that is specifically targeted to block the mutated EGFR; secondly, a lung cancer from a non-smoker is more likely than one from a smoker to have an EGFR mutation.

A couple of years ago, we were approached by Professor Charles Swanton from the Francis Crick Institute in London. With funding from Cancer Research UK, his team were studying the relationship between air pollution and lung cancer in non-smokers. International data showed that almost all cases of lung cancer in never-smokers occurred in people living in areas with high air pollution. (i.e. areas where pollutants under 2.5 micrometres in diameter, known as ‘PM2.5’, exceed the World Health Organisation (WHO) acceptable annual levels). The scientists wondered if there might be a link between PM2.5 and EGFR mutations, given that both had an association with lung cancer in non-smokers.

Since 2016, NDRS has recorded which lung cancers have EGFR mutations: using patient area of residence, we linked this cancer data to regional air pollution figures held by the Centre for Radiation, Chemical and Environmental Hazards (CRCE). This enabled us to see, in fine geographical detail, any possible correlation between EGFR-mutated lung cancer rates and air pollution. The scientists at the Francis Crick Institute analysed the linked data, alongside similar information from Asian countries, and established that rates of EGFR-mutated lung cancer were indeed higher in areas with high air pollution.

The next step was to investigate a biological mechanism for a link between PM2.5 pollution and EGFR mutations in lung cancer, having already established that air pollution was not directly causing EGFR mutations. The scientists looked at normal human lung tissue from a tissue bank, and found that it was quite common to see EGFR mutations in patches of normal lung tissue in older people. In other words, the EGFR mutations were just an age-related change, and were not in themselves sufficient to cause a cell to become cancerous. They then studied the effects of air pollution upon mice with EGFR mutations in some of their lung cells. It became apparent that air pollution caused inflammation in the lungs of the mice. If the inflammation occurred in an area of the lung where existing EGFR mutations were present, a particular inflammatory signalling molecule called IL-1β acted upon the EGFR mutated cells and caused them to grow and divide. Over time, and with chronic inflammation, the cell growth became uncontrolled, and a cancer developed. The scientists were able to prevent the mice from developing cancer by treating them with a drug that blocks the IL-1β inflammatory molecule.

Taken together, these results show that, in the absence of inflammation, EGFR-mutated lung cells lie dormant but primed for growth. However, when PM2.5 pollutants cause the local environment around the EGFR-mutated cells to become inflammatory, the mutant cells become activated, and a cancer develops.

Interestingly, these findings tie in with some very old lab observations on how cancer develops. In the 1940s, two scientists called Mottram and Berenblum were studying how chemicals might cause cancer. They found that substances causing cancer in laboratory animals could be divided into two types – tumour initiators (e.g. benzo(a)pyrene, a chemical found in coal tar and tobacco smoke), and tumour promoters (e.g. a skin injury). Neither on its own was sufficient to cause cancer, and, for cancer to develop, the animal had to be exposed to the initiator and promoter in that specific order. Exposure to the promotor (skin injury) followed by the initiator (coal tar) would not cause cancer. At the time, these observations were not fully understood. With more modern knowledge of the role of DNA mutations in cancer cells, we now understand that the old ‘tumour initiators’ are substances that cause DNA mutations, whereas ‘tumour promoters’ are substances or conditions that stimulate cell growth, such as hormones, wound healing, or, in this case, the inflammatory response to air pollution.

It’s fascinating how the combination of lab research, global epidemiological data, and real-world NHS data collated in the NDRS cancer registry can all come together to shine a light on why non-smokers can develop lung cancer, whilst also backing up an 80-year-old model of cancer development.

More research will be needed to investigate whether this has implications for the way lung cancer is treated in non-smokers. However, it does add weight to the already compelling argument that we need to collaborate at a global level to reduce air pollution. In the words of Professor Swanton, who presented this work at the European Society of Medical Oncology (ESMO) Congress in Paris on 10 September 2022: