Air Pollution and Breast Cancer Incidence

Journal of Clinical Oncology (JCO) Podcast

Host Dr. Davide Soldato and his guests Dr. Ann Wu and Dr. Alexa White discuss the article "Air Pollution and Breast Cancer Incidence in the Multiethnic Cohort Study" and the editorial "Growing Evidence for the Role of Air Pollution in Breast Cancer Development"

TRANSCRIPT

The guests on this podcast episode have no disclosures to declare.

Dr. Davide Soldato: Hello and welcome to JCO After Hours, the podcast where we sit down with authors from some of the latest articles published in the Journal of Clinical Oncology. I am your host, Dr. Davide Soldato, Medical Oncologist at Ospedale San Martino in Genoa, Italy.

Today, we are joined by JCO authors Dr. Anna Wu and Dr. Alexander White. Dr. Wu is a professor of Population and Public Health Sciences at the Keck School of Medicine of UCS, while Dr. White is an investigator in the Epidemiology branch of the Environment and Cancer Epidemiology Group at the National Institute of Health.

Today, we will be discussing the article titled, “Air Pollution and Breast Cancer Incidence in the Multiethnic Cohort Study,” and the accompanying editorial.

So, thank you for speaking with us, Dr. Wu, Dr. White.

Dr. Anna Wu: Thank you for having us.

Dr. Alexandra White: Yes, thank you so much for the invitation to be here.

Dr. Davide Soldato: So before going in depth about the results of the study that was published in the JCO, I was wondering if you could give us like a brief introduction and a little bit of background about what was known about air pollution as a risk factor for breast cancer and what was the evidence before this study was conducted.

Dr. Alexandra White: Okay. I can start with that question. So, there's been research for decades looking at the relationship between air pollution and breast cancer. And it's been a really challenging question to address for a number of reasons. One being that it can be really difficult to assess exposure to air pollution and many previous studies have had really limited information on people's residences over time. But in general, what we thought leading up to this study was that evidence was most consistent that exposure to traffic related pollutants such as nitrogen dioxide was more consistently related to a higher risk of breast cancer. The evidence for fine particulate matter or PM2.5 was less consistent. More recently, there have been a few large, well conducted studies that have supported a positive association. This new study in the multiethnic cohort led by Dr. Wu is really important because it really demonstrated that, in this large study of over 50,000 women in California, that they also do see an association with PM2.5.

Dr. Davide Soldato: Thank you very much for the introduction. So, Dr. Wu, we just want to hear a little bit more about the results. So, what was the association that was observed for PM2.5? And specifically, the study that you ran was focused on a very diverse population, a multiethnic cohort, and so I was wondering if you observed any type of differences when you consider the different populations that were included in your study. And if you could also give us a little bit of what was the composition of the women that were enrolled in this cohort.

Dr. Anna Wu: Thank you for the question. So, the multiethnic cohort study is a cohort of over 200,000 individuals who were enrolled when they lived in Hawaii or California. For the air pollution studies that we've been conducting, we have focused on primarily the California participants. And in this instance for the breast cancer study, it was based on roughly 56,000 individuals out of- there were about 100,000 because half of them were men and they were not included. Of the California participants, 75% of them were African Americans or Latinos and they were self-identified as these racial ethnic groups when they enrolled in the study. And this was a particularly important consideration for us because in most of the studies that have been published so far on-air pollution and breast cancer, as well as other cancer sites, most of those studies were conducted among whites in the US or whites in Europe. And even if they included non-white populations, the numbers tend to be small so that they were not able to conduct racial ethnic specific analysis. So, we were particularly interested in examining these other racial ethnic groups because we know from other studies that racial ethnic minority groups tend to live in communities of low socioeconomic status and those communities also tend to have higher levels of various types of environmental pollutants. And so, it was important for us to actually try to tease apart these various interrelated factors.

So, what we found was that per 10 micrograms per cubic meter, we had a 28% increased risk overall in all participants combined that meet across the racial ethnic groups. We actually did not see any differences or significant differences in the hazard ratios by race ethnicity and they were in general quite compatible with each other. But we did see a stronger finding among the white participants in our study.

Dr. Davide Soldato: Thank you, a lot, Dr. Wu. So, I think it's very interesting the fact that in the end you observed that air pollution is a significant risk factor across all the ethnicities that were included in the study. But I think that one very strong point of the manuscript and one very strong point of the analysis was that in the end you also corrected for a series of different factors because we know that the incidence of breast cancer can be modified, for example, by familial history or BMI or smoking habits or also alcohol consumption. And a lot of these risk factors were included in your analysis. And so, I was wondering if you could tell us a little bit whether you observed any significant differences when you observed or included also these risk factors in your analysis, or whether the association for air pollution as a risk factor stands even when we consider all of these other elements.

Dr. Anna Wu: Yes. So, we considered all the well-established breast cancer risk factors. And in this situation, we were particularly interested in considering smoking, alcohol intake, use of menopausal hormones, history of diabetes, body mass index, family history, as well as physical activity, because many of these risk factors, such as, for example, diabetes and body mass index, they are risk factors for breast cancer, and air pollution, have also been found to increase risk of these factors.

So, in our analysis, we first adjusted for all of these potential confounders in a mutually adjusted manner, so all of them were considered. In addition, we also conducted stratify analysis. So as an example, we stratified the analysis to examine whether the hazard ratio associated with PM2.5 provided comparable risk estimate or hazard ratio estimates for never smokers, former smokers, and current smokers. Although we did not see significant heterogeneity by these various subgroups, we did see a significantly stronger effect of PM2.5 among individuals who did not have a family history of breast cancer.

Interestingly, our finding was also stronger among individuals who were never smokers and light alcohol drinkers, even though the results were not significantly different. So, we surmised that maybe individuals who already had a high risk because of other established risk factors for breast cancer, we were less likely to be able to observe the effect of air pollution. But it's important to note that other studies, such as the ones that Dr. White has conducted, have also looked at various subgroups, and I think part of the limitation that all of us have is that once you subdivide the study population, even if you start out with a large sample size, often the sample size gets cut in half or a third. And so, we still lack the statistical power to be able to observe significant differences. But I think it is important to note that, in fact, the hazard ratio estimates are actually quite comparable, but we did see a hint of stronger effects among never smokers, and people who were light alcohol drinkers. So, I think this is an area that we certainly need to continue to investigate since there are other subgroups, such as menopausal status, such as hormone receptor status of breast cancer, that we need to consider in future studies. There's still a lot of work we need to do to sort this out, to actually figure out who are the women who are the most susceptible to the exposures.

Dr. Davide Soldato: Dr. White, I would really love a comment from you on this specific area and specifically on what still needs to be done. And related to this, a question actually, for both of you, because I think that from a methodological point of view, there is a lot of work that goes into deciding how we are going to assess the exposure to air pollution. So which type of data are we going to use? Which type of data are we currently using in the epidemiological studies that have been conducted and in the one that we are discussing right now in JCO? And what are the caveats for this data that we are using? Meaning, I think that we use mostly residential addresses, which means that we are looking at the exposure where people actually live, which might not be the place where they spend most of their time. For example, if someone is working, maybe they could be more exposed and have higher exposure when they are at work compared to when they are at home. So, I was wondering if you could give us a little bit of an overview as to what is the methodological standard of care right now in terms of this analysis and what can we do better to refine and understand this specific factor as Dr. Wu was mentioning?

Dr. Alexandra White: Yeah, so I'm happy to take a first stab at that question. So, I think it's important to note just how far we've come. I think even a few years ago, air pollution was really not considered a risk factor for breast cancer. And a lot of the work that we've been doing and others have really moved this forward in terms of understanding this as a risk factor. And as I mentioned earlier, there have been a lot of challenges in exposure assessment. And to get to your question, I think that our studies in general are doing better at looking at exposure over more years, residences, more time. We know that cancer takes time to develop, and we can't rely on just a single snapshot of exposure. But as you mentioned, almost all of the studies published have really exclusively focused on residential estimates of exposure. And so, there's a real need to understand the exposures that people are experiencing in other aspects of their life, from their commute to their jobs, to really capture that totality of exposure.

And then I think one of the points that Dr. Wu was alluding to as well as we know that breast cancer is a very heterogeneous disease, so risk factors for breast cancer vary by tumor subtypes, by menopausal status at diagnosis. And a lot of studies have really focused on considering breast cancer as a combined outcome, and that might be missing some really important signals where we might have a stronger effect for certain subtypes due to the fact that there's different biologic pathways that are underlying these subtypes or by menopausal status. And so having large study populations where, as we discussed earlier, would really give us the power to look among these smaller groups of women who might be more susceptible and those with younger women, we know that incidence of cancer is rising in young people, and we need to understand the risk factors for that. And most of our studies are really focused on older individuals, so I think that's one important gap, as well as having the power to really look at different differences by tumor subtypes.

Dr. Davide Soldato: I think it's very interesting, and I think one point both of you made in the original article and in the accompanying editorial is also the fact that we tend to look at these risk factors in people who are actually aged, while we maybe should be looking at this in an earlier phase of development and potentially during puberty. Do you think that we should design studies that are more focused on this population even though I think that they will take a lot of time to produce significant results?

Dr. Alexandra White: Yeah. I think that it is really important to consider how exposure during early life is related to breast cancer risk. We know that exposures during pregnancy or even as early as during puberty might be particularly relevant for breast cancer. And I think a lot of our studies have really been up against the challenge of the fact that exposure monitoring for air pollution really didn't start until the 1990s. And so, it's challenging, especially for these older cohorts, to get back at that time period that might be relevant. But I think that's something that definitely newer cohorts are going to be able to address, and I think it's going to be really important, and also will give us some clues to better understand the important windows of exposure, but also that might provide clues for the biologic pathways as well that are relevant.

Dr. Davide Soldato: And just a related question, because I'm not aware of this, but are there right now cohorts that are specifically looking at this in the US or in other parts of the world? If you are aware of that, of course.

Dr. Alexandra White: There have been some cohorts that have focused on exposure during these hypothesized windows of susceptibility, but I don't think they've been able to follow those women long enough to develop breast cancer. One of the things that we're working on in the sister study is trying to expand our assessment of air pollution exposure back in time to try to get at these earlier windows of exposure. So, I'm hoping that it's something we'll be able to comment on and at least for some of the women in our cohort who are younger. But I don't know, Dr. Wu, if you're familiar with any other populations that are doing this now?

Dr. Anna Wu: Well, NCI funded several new cohorts in the last couple years that are really focused on trying to get a much more refined exposure assessment. So, I know colleagues at University of Michigan that are peers and also Dr. Wei Zheng at Vanderbilt, they are putting together newer cohorts that are younger and also trying to include a range of exposure, not just air pollution, but really environmental exposures. Those cohorts I think have the potential in the future to try to address some of these questions, but again, it will take at least another number of years before there are a sufficient number of endpoints so that they can actually do these types of studies.

Another possibility is that there are a number of big cohort studies in Asia. The age of diagnosis tends to be earlier in Asia. I know that investigators in China are very interested and concerned with the air pollution effects in China. I think there are potentials that in other countries where the age of breast cancer diagnosis is actually younger than in the US and if they establish in a manner that allows them to assess air pollution that they may have opportunities.

And I think the other way to try to address this question, whether there are studies where you can actually tap into either biomarkers or pathology samples so you won't be actually studying air pollution in a large population, but you're actually narrowing it down to try to see if you see any signals in a way that would give you some additional clues and insights as to the mechanism. So I think we're going to have to piece together various types of study to try to answer the questions because one type of study like these observational air pollution studies, will allow us to address one slice of the questions that we have and then we need to put together other studies so that we can address other aspects that we're interested in to put it together.

Dr. Davide Soldato: Thank you very much both of you. That was very interesting.

Coming back to the results of the manuscript, we really focused up until now on PM2.5. But it's true that inside of the study you evaluated different pollutants. So, I was wondering whether you saw a similar association for other pollutants that were included in the study or whether the association for higher risk was observed only for PM2.5.

Dr. Anna Wu: The results for NO2, NOx, PM10, and carbon monoxide were actually very compatible with the risk estimates that other studies have published as well as from the meta-analysis. So, I would say that our results from the other pollutants are actually very consistent with other results. I think one difference is that our PM2.5 estimates were based on the satellite-based PM2.5 estimates, whereas all the other pollutants were based on monitoring station estimates from EPA sponsored air monitoring stations. So, they are not measured in the same way. And I think different studies over time have used either monitoring station type measures for other pollutants. And I think we were particularly interested in PM2.5 because the measurement of PM2.5 in the monitoring world didn't start until around 2000. So, studies up until that time were less able to actually provide the assessment of PM2.5 as good as we can for air pollution. There's always misclassification. So, I think it's a matter of how much misclassification in the assessment. But, again, we are really limited in really just having exposure over one part of adult life.

Dr. Davide Soldato: Thank you very much. And one potentially related question. We are speaking in general about air pollution, but I think that since we are considering residential addresses, probably we are capturing more either traffic pollution or pollution that comes from probably industries or stuff like that, which is mostly related to residential areas or the place where people live. But I think that in the end we also think about air pollution as something that can come from different forms. And one very interesting point, Dr. White, that you made in your editorial is also that there is a global change also in the way we are faced with air pollution. For example, you made the example of wildfires in your editorial and how this might potentially change exposure to air pollution, maybe for limited times, but with concentrations that are fairly higher compared to what we generally observed. So, I was wondering if you could comment a little bit on that and also, if there is potentially a way to also consider this in future epidemiological studies.

Dr. Alexandra White: Yeah, so when we talk about exposure to fine particulate matter, PM2.5, we're assessing exposure to particles that are based on the size of the particle, and we're really not evaluating the types of particles that people are experiencing exposure to. And we know that, in general, that PM2.5 composition really varies geographically due to differing sources of exposure. So, like you were saying, there might be a stronger contribution to industry or from agriculture or from traffic. And so that could really change the PM2.5 exposure profile that individuals experience. And so it could be that this is another really important area that this research needs to consider, which could really help us identify what sources of exposure are most relevant.

Wildfires are a really important growing concern. We know that wildfires are increasing in both intensity and duration and frequency, and we really don't understand the long-term health impacts of wildfires. But we know that wildfire associated PM2.5 might be one of the most dominant contributors to PM2.5 moving forward. And although we've seen historic declines in PM2.5 in the US after the Clean Air Act, those declines have really stalled. PM2.5 itself is projected to increase over the next few decades, so understanding different PM2.5 composition profiles and the sources that drive them can really help us identify the most important targets for any potential interventions. And wildfire PM2.5 in particular may be of concern because it's a combustion byproduct, and so it's thought to have more of the components that might, we hypothesize, are most relevant for breast cancer, such as PAHs or polycyclic aromatic hydrocarbons or metals. And so, these components are thought to act as endocrine disruptors, which may be particularly relevant for breast cancer. So, I think understanding this changing landscape of PM2.5 moving forward is going to be really important in understanding how PM2.5 contributes to cancers beyond just breast, but as well as other female hormone driven cancers and all of the cancers really.

Dr. Davide Soldato: Thank you very much. So, one closing remark, because I think that in general, we have been really in a field of primary prevention for breast cancer where we were focusing on individual behaviors, for example, smoking cessation, reduction in alcohol intake, reduction of BMI, increase of physical activity. But I think that the evidence that is accumulating in the last three years or so is telling us more and more that we also need to shift the perspective on prevention going not only on individuals, but also as including environmental risk. So, I was wondering, how can we include this new evidence in the policies that we implement and how policymakers should act on the data that we have available right now?

Dr. Anna Wu: I think it's really important that this new information is communicated to all the stakeholders, including our policymakers, so that they are, first of all, really aware that any changes and not actually adhering to current guidelines can have long lasting consequences, deleterious consequences. And I think it's important to also note that over 90% of the world actually live in areas where PM2.5 exceeds the limit. We have observed increases in breast cancer in many middle- and low-income countries, so I think it's particularly important to emphasize that this is really not just a western country issue, it is really a global issue.

Dr. Alexandra White: I agree. And I would just add to that that air pollution is not something that an individual can really change on their own. There are things you can do, you can monitor air quality, you can try to live in a home that's far away from traffic. But really these are large scale problems that really require large scale solutions. And we know that policy changes can be effective here and that this is something that, in my opinion, is not something that we leave to the individual to change. This is something that we as a society should encourage change for the health of everyone.

Dr. Davide Soldato: So, thank you very much again, Dr. Wu, Dr. White, for joining us today on the podcast.

Dr. Anna Wu: Thank you.

Dr. Alexandra White: Thank you so much for having us.

Dr. Davide Soldato: So we appreciate you sharing more on your JCO article and accompanying editorial titled, “Air Pollution and Breast Cancer Incidents in the Multiethnic Cohort Study.”

If you enjoy our show, please leave us a rating and review and be sure to come back for another episode. You can find all ASCO shows at asco.org/podcasts.

The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions.

Guests on this podcast express their own opinions, experience and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity or therapy should not be construed as an ASCO endorsement.