Editor’s Note: For more on air pollution and lung cancer, read epidemiological insights from Dr. Christine Berg’s ELCC 23 presentation.
Pioneering investigator and University College London Professor Charles Swanton, PhD, FRCP, recently returned to the European Lung Cancer Congress to discuss his latest research. On March 31, Prof. Swanton, who is principal group leader at the Francis Crick Institute, presented a keynote address on the links between air pollution and lung cancer at ELCC 23. He recently discussed his work during an interview with ILCN.
ILCN: For those who are unfamiliar with your work, tell us a little about fine particulate matter (PM) 2.5 pollution.
Prof. Swanton: The definition we’re using are particles under 2.5 microns in diameter. They derive from internal combustion engines. They derive from brake dust, tire dust, but also agriculture. Essentially heavy industry churns out these particles, and they’re very heterogeneous. And the composition can differ over short ranges within cities and between cities, depending on where you are in different parts of the world. It’s a very heterogeneous composition of organic and inorganic matter.
ILCN: You have previously described potential mechanisms that help explain how particulate matter may be carcinogenic, particularly looking at links between EGFR and KRAS mutations and air pollution (read more). Remind us about your research on these mechanisms. Where are these mutations coming from?
Prof. Swanton: Broadly speaking, there are two hypotheses about the way cancers are initiated. The first, conventional model is that mutations directly initiate cancer. So, we acquire mutations as we age or through carcinogens like tobacco or UV light and by chance a driver mutation will occur, and a tumor will grow up.
Now, there are some serious problems with that model that the field—me included—has not fully explored. The first is we know that our normal tissue is an array of complex clones, many of which harbor mutations that have been seen and found in cancer. So that tells us that cancer mutations may be necessary for cancer formation, but they’re not always sufficient.
Secondly, the commonest mutation found in melanoma—the BRAF V600E mutation—is not an ultraviolet light mutation. If we say ultraviolet light causes melanoma because it causes mutations, that’s simply not true because the commonest driver, BRAF V600E, is not a UV light-induced mutation.
And then we find about 8% of lung cancer patients who smoke don’t have evidence of a tobacco carcinogen signature. You can be exposed for 15 to 20 pack years, but not have any obvious mutational signature in your tumor genome linked to tobacco exposure. So, how do these tumors start?
The last piece of evidence is work from Allan Balmain and colleagues, who showed that 17 out of 20 of environmental carcinogens tested do not induce mutated cancers in mice. They induce cancer, but they do so in a way that is independent of mutations.
So that led us to think there must be a second alternative model of cancer formation. We turned to Allan Balmain, who has been a great proponent of the work of cancer biologist Isaac Berenblum, who in 1947 published an alternative way in which he thought cancer was initiated through a two-step process—an initiator and a promoter. Now of course Isaac Berenblum didn’t know what DNA was in 1947, but subsequently his model has been elaborated upon in the following way.
The initiator is the mutational step, that’s where you need the mutations. Those are necessary, but not sufficient to drive cancer. You need something else. And that’s the promoter step. The promoter step can be an inflammatory process, and that’s where we think pollution comes in. Pollution causes inflammation of tissues, which I can explain a bit in a minute, and that allows a cell with an EGFR mutation to then grow up and form a cancer. But you need both a cell with an EGFR mutation and the pollution for the cancer to form.
As for where the mutations come from; we find that these EGFR and KRAS mutations are present everywhere. They are there in normal lung tissue in people who have never smoked. In fact, as you age, the chance of finding a cancer driver mutation increases. We’re acquiring them as we age, and they’re sitting in cells that won’t become cancerous unless there’s another insult, we think.
Of course, this raises questions about what a carcinogen is because these carcinogens are causing cancer independent of driving DNA mutations, which is fascinating because it also suggests that there must be other carcinogens. It’s likely there are other carcinogens that may be working through similar ways in these tissues—driving inflammation and cancer. I guess that’s not new. We’ve seen this. The role of inflammation and cancer is well established.
ILCN: What have you learned since about how that inflammatory process works?
Prof. Swanton: With the new data we have, we think the inflammatory process works directly through macrophages, which are recruited to tissue—the lung, in this case—on exposure to pollution. The macrophages gobble up the pollutants but can’t metabolize them; thus they release inflammatory mediators like interleukin 1-beta. And that interleukin 1-beta is sufficient to transform the EGFR mutant cell into a cancer cell. If you block IL1-beta, you can prevent pollution-induced tumors in mice. So, we think IL1-beta is central to the inflammatory process.
ILCN: If blocking IL1-beta in mouse models can stop the process, how might this knowledge be leveraged in future research?
Prof. Swanton: One of the reasons we became interested in IL1-beta is the CANTOS trial, which showed very compelling data suggesting a dose dependent reduction in lung cancer incidence with rising doses of canakinumab, the anti-IL1-beta antibody.1 So that shows very clear evidence in humans that anti-IL1-beta can prevent lung cancer initiation. What we’ve done is to suggest that one way this may be working is by dampening down inflammation caused by smoking or pollution and thus reducing the risk of lung cancer initiation.
What is fascinating is the new data we’ve generated in the past several months shows that if you take macrophages from pollution-exposed mice and incubate them with cells that have EGFR mutations from a non-pollution-exposed mouse, those macrophages can induce those EGFR mutant cells to form a cancer. But you need the macrophages to do so. And those macrophages must be exposed to pollution to increase their stem cell capacity.
ILCN: Does this research offer new insights into the driving forces behind lung cancer in general?
Prof. Swanton: My feeling is that, yes, it does offer insights into cancer initiation in lung cancer in general. I think it’s very likely that you need both a promoter and an initiator. You need the mutation and the inflammatory mediator. The reason tobacco is such a potent carcinogen, I think, is because it can do both. It can induce mutations and inflammation. Whereas pollution, and probably secondhand smoke and other factors, just do the latter—the promotion.
I think anything that we inhale into the lung is likely to induce an inflammatory process that acts as the promoter, but you still need the initiator—i.e. the mutation—to develop cancer. We have evidence that the combination of air pollution and tobacco is additive; there’s additional risk if you smoke and live in a highly polluted area.
ILCN: For clinicians and patients, what do they need to know about the risks of cancer from air pollution and how that compares to the risks from tobacco smoke?
Prof. Swanton: This is important. The crucial thing is if you live in a highly polluted area, don’t panic. The increased risk of lung cancer with smoking is about 30-fold. And the risk from air pollution is somewhere between 1.07- and 1.5-fold, depending on the concentration you’re exposed to. The risk associated with air pollution is an order of magnitude less than smoking. But the crucial thing about pollution is that about seven times more people are exposed to unsafe levels of air pollution than tobacco smoke.
To give you an example, in London, the concentration of PM 2.5 is about 15 micrograms per meter cubed. Let’s say you lived in the center of London; the risk of EGFR-mediated lung cancer would be about 15 cases per 100,000 of the population per year. That is roughly—very roughly—one extra case per 100,000 of the population per year for every one microgram rise in PM 2.5.
Now that may not sound very much, but over a 60-year time frame, that translates to many patients who are at risk. We think about 7 to 8 million deaths a year can be attributed to PM 2.5 worldwide. That’s as many as die per year from tobacco.
ILCN: So, it is possible to look at air pollution data for a given area and quantify the risk for people living in that region. Do you know if there are there any efforts yet to leverage this knowledge to reduce the risks at a population level?
Prof. Swanton: We’re in touch with the mayor of London’s office about clean air. They’ve been very interested in our work as have Public Health England, who helped us with this research. There has also been a lot of interest in the press worldwide. People have reached out to us from other countries interested in the work and continue to do so. I don’t know what’s happening as a direct result of our data, but I think it’s fair to say that our research is another piece in the puzzle. We’ve known for decades that air pollution is bad, so it’s not groundbreaking in that respect.
Of course, it’s not just lung cancer. Diabetes, cardiovascular disease, strokes, dementia, premature births, low birth weight: All these things are induced by PM 2.5 pollution exposure.
ILCN: What are your next steps?
Prof. Swanton: First, there are a thousand other carcinogens out there that probably don’t mutate DNA. We want to know how they work. How are they causing inflammation?
We also want to develop new carcinogen assays, because clearly, if all carcinogen assays consider whether something mutates DNA or not, that doesn’t work. We’ve shown that pollutants don’t mutate DNA, but they do cause cancer. This raises serious questions about the chemical matter that we’re allowing into our environment—chemical matter that we’re probably being exposed to daily that may be carcinogenic, but because it doesn’t mutate DNA, it doesn’t score very highly on any carcinogenic assays.
The other thing we’re trying to do is establish what is it about PM 2.5? Is it size? Is it quality? Is it a particular heavy metal? We don’t know. And that’s something we’re looking at now. And of course, we’re also looking at microplastics, because we’ve got some early data that microplastics pose similar risks as PM 2.5.
- Ridker PM, Everett BM, Thuren T, et al. Antiinflammatory Therapy with Canakinumab for Atherosclerotic Disease. N Engl J Med. 2017;377(12):1119-1131. doi:10.1056/NEJMoa1707914