Environmental carcinogens – it’s not all about mutagens

One of the major breakthroughs in understanding cancer risk came with the recognition in the 1980s that DNA mutations are causative of cancer (Cancer. 1988;61:1963–1968; Br J Cancer. 1985;51:1–7). These exciting findings set a course for research centred on identifying environmental carcinogens based on their ability to cause DNA mutations. Unfortunately, this approach somewhat sidelined long-standing evidence that while mutations are essential to cancer development, the rate-limiting step is not the factors that cause mutations but the factors that promote them.

Research in the 1940s showed that although carcinogens can initiate skin cancer in animal models, tumour development does not necessarily follow unless a tumour promoter, known to induce inflammation and proliferation, is applied repeatedly to the target tissue. More recently, experiments my lab conducted in collaboration with the group of David Adams at the Sanger Institute revealed that most of the 20 known or suspected human carcinogens we tested did not directly cause mutations, suggesting that most carcinogens act through a promotional pathway to stimulate carcinogenesis. These findings were supported by research undertaken as part of a recent, extensive, world-wide collaboration on the role of air pollution in cancer risk, which demonstrated that carcinogenesis in cells carrying pre-existing mutations occurred more quickly in those exposed to air pollution compared with those not being exposed, as reported at the ESMO Congress 2022. The study revealed that lung cancers not related to smoking were concentrated in areas of high air pollution, confirming in a large human population what had previously been observed in laboratory models. In the light of new knowledge, a number of environmental factors are being re-examined. Tobacco smoke is an ‘ideal’ carcinogen, acting not only as a mutagen but also as a promoter. Similarly, UV light has been shown to expand clones of cells with pre-existing mutations and the pathways involved resemble those seen with known tumour promoters. There are also a whole variety of other environmental carcinogens, including the so-called ‘forever chemicals’, such as per- and polyfluoroalkyl substances (PFAS) and microplastics, which have links to promotion. A particularly worrying new problem involves vaping, which is known to cause inflammation and could be a ticking time bomb given the number of individuals, particularly young people, adopting the habit.

In addition, there are strong links between tumour promoters, lifestyle/behavioural factors – such as obesity, diet and exercise – and cancer, which were based originally on epidemiology but now, increasingly, are supported by mechanistic data (Cancer Cell. 2022;40:720–737). In this respect, pathways impacting on inflammatory responses that can naturally prevent cancer, particularly via promotion, are being described.

There is a need for increased research firstly to clarify the differences between cancers that develop primarily due to mutations and those that require the input of a promoter, and secondly to determine the mechanisms of action of existing and future environmental carcinogens – mutagen, promoter or both.

Many countries have national policies in place to try to tackle environmental pollutants, although the efficacy of these measures is debated. The gold-standard test for carcinogen detection is the National Toxicology Program’s 2-year rodent bioassay. These tests, which are expensive and time-consuming, measure the long-term effects of chronic exposure of mice and rats to candidate carcinogens. However, given that mutations can occur spontaneously, as well as be induced, carcinogens identified by this method could in fact be promoters – stimulating spontaneous mutations – rather than direct mutagens, an interpretation supported by a recent analysis of mutational signatures induced by these carcinogens (Nat Genet. 2020;52:1189–1197). This underlines the need for specific tests able to distinguish between mutagens and promoters. Shorter-term (90-day) assays are now becoming available and the quantitative information they provide could well prompt a review of national policies on environmental carcinogens. In the race to better define how environmental carcinogens work, AI could well be a real game changer. While we are only scratching the surface of what can be achieved with AI at the moment, its potential to make rapid advances in this area is hugely promising.