Rethinking Environmental Carcinogenesis

It is widely assumed by cancer biologists, clinicians, and cancer organizations that chemical substances are tested for carcinogenicity prior to being released for public use, that agents known to be or designated as probable carcinogens are carefully monitored, and that regulations surrounding carcinogens are strictly enforced. The 2010 President's Cancer Panel (1) noted that none of these assumptions is correct. This realization led the panel to explore whether chemicals used in the production of goods and services—from food to manufacturing to transportation—might be more important contributors to the cancer burden than previously suspected, and whether the often-quoted estimate that 2%–4% of human cancers within the United States derive from exposures to “pollution” or “industrial” risk factors (2) might be an underestimate. New research on the “key characteristics” of carcinogens showing that chemicals can activate a wide range of biological pathways in cancer causation (3), for example, immunosuppression, chronic inflammation, and endocrine disruption (Fig. 1); the possibility that mixtures of chemical exposures could act synergistically to cause cancer (3, 4); and the existence of “windows of susceptibility” to action by carcinogens during specific stages of development (5) suggest that more attention to the role of chemical exposures in cancer development is warranted at this time.

To explore these issues, the American Association for Cancer Research (AACR) convened a special conference entitled, “Environmental Carcinogenesis: Potential Pathway to Prevention,” held June 22–24, 2019, in Charlotte, NC, and cochaired by Margaret Kripke, Ernest Hawk, and Timothy Rebbeck. A goal of the meeting was to review and analyze evidence for the role of environmental factors in cancer development and discuss how this information could be applied to prevent cancer. The meeting also provided a forum for discussion of these issues among diverse groups, including cancer biologists, epidemiologists, toxicologists, advocates, community leaders, and communication specialists, who rarely have opportunities to interact.

Historically, chemical carcinogenesis constituted a major focus of cancer research. Rodent models of chemically induced cancers dominated cancer research efforts in the 1950s and 1960s. This research area was overshadowed by a new focus on the role of viruses in cancer causation in the 1970s and was nearly completely supplanted in the 1980s–1990s by the focus on cancer genetics. In the 2000s, carcinogenesis was proposed as the cumulative evolutionary result of 6 to 10 discrete, functional derangements in key aspects of cellular identity, function, and communication which seemed to characterize almost all cancers (i.e., “hallmarks” of cancer development) and which might be exploited as “targets” of cancer treatment; that paradigm persists today (6).

Although still a major concern of toxicologists, epidemiologists, and regulatory agencies, the field of environmental carcinogenesis has virtually disappeared from the mainstream of cancer research. It has been overwhelmed by exciting advances in understanding genetic alterations in cancers, the use of genetically engineered animal models, the promise of new cancer therapies targeting altered biochemical pathways in cancer cells, and most recently, advances in immunotherapy. However, there is growing concern that the current emphasis in cancer research on treatment has stalled progress on both primary and secondary cancer prevention (7). Current prevention and early detection efforts are largely directed toward screening for precancerous lesions and early-stage cancers as well as modifying behavioral risk factors including tobacco use, obesity, sunlight exposure, and physical inactivity. Recently, prevention research has been extended toward “precision prevention,” which involves the identification of individuals and populations at high risk of cancer and the implementation of individually tailored interventions during early stages of cancer development (8).

Recent studies of environmental carcinogenesis suggest that this research may provide yet another avenue for preventing certain cancers. It is important to note that cancer incidence in children and adolescents continues to rise (9). The NCI estimates that up to 10% of cancer incidence in these groups is caused by a heritable germ line mutation, suggesting that environmental influences, rather than genetic susceptibilities alone, may be significant and modifiable contributors to cancer development (10). Prevention via reduction in exposure to and use of environmental carcinogens, as well as early detection, has the potential to substantially reduce cancer incidence and mortality in the general population, along with their devastating emotional, physical, and financial consequences for patients with cancer and their families.

A number of concepts are being integrated into the field of environmental carcinogenesis, including health disparities, cancer genetics, and measurement of the early effects of carcinogen exposure. These include a new conception of how to define a possible environmental carcinogen, extending beyond mutagenesis alone, to include the wide array of identified “mechanistic hallmarks” including inflammation and immune suppression, as well as effects on development and other processes (3).

Environmental causes of cancer may be discerned when higher incidence of certain cancers is found in groups of people that have higher than average exposure to toxic agents in their living or work environment. For example, studies have documented disparities in rates of childhood leukemias associated with residence: children living near freeways and industrial facilities are more likely to develop leukemia than children living in neighborhoods with lower levels of air pollutants (11, 12). Workers in the construction trades are disproportionately exposed to carcinogenic agents such as asbestos, crystalline silica, and diesel exhaust, which are associated with increased lung cancer risk in these workers (13). However, the lack of research examining cancer risk in the most vulnerable (e.g., migrant workers exposed to agricultural chemicals or janitorial workers exposed to cleaning and building maintenance chemicals) is a consequence of scientists' difficulties in recruiting and retaining study participants from these populations, among other challenges.

At the molecular level, studies of cancer genetics are beginning to reveal new avenues for measuring early markers of exposure to carcinogens (14–16), leading to a better understanding of how cancers arise (17, 18). Whole-genome sequencing of tumors has been used to identify over 50 mutational signatures that reflect the history of genome-altering events, such as chemical exposure or altered DNA repair (17–19). These histories may indicate underlying causes and mechanisms of disease. For example, specific mutational signatures have been associated with cancers caused by UV radiation and tobacco smoke (20, 21), and these signatures can be replicated in cell culture (17). Other signatures are associated with inherited polymorphisms in DNA repair processes such as BRCA1 and BRCA2, while about half the signatures so far identified do not have a known etiology (19). Sequencing of tumors in experimental animals and mutations in cell culture can validate associations seen in epidemiology and also provide insight into the biological activity of agents being tested (17, 18).

Technologies that reveal the mechanisms of established human or animal carcinogens can be used to predict the activity of chemicals that have unknown carcinogenic potential (22, 23). For example, patterns of gene expression in cancer cell lines have been shown to differ in response to exposure to liver carcinogens, and the U.S. Environmental Protection Agency (EPA) is generating transcriptomics data for 2,000 common chemicals, including many carcinogens, to identify predictors of carcinogenicity (23, 24). In vitro methods are rapidly developing with the potential to provide low-cost tests that can predict carcinogenicity using human cells or prioritize chemicals for testing in animal models so that chemical carcinogens can be identified and exposures reduced (25, 26).

Continuing investigations of endocrine-disrupting chemicals are providing new information regarding critical “windows of susceptibility” to cancer induction, reflecting an interaction between chemical exposure and developmental stage. For example, experimental animals exposed to bisphenol A (BPA) early in life show both alterations in prostate cells that are suggestive of increased prostate cancer risk later in life, and mammary gland hyperplasia and tumors (27), with effects observed at lower and not higher doses (26). A transgenerational study of women exposed to the endocrine disrupter DDT showed a higher incidence of breast cancer in women with the highest blood levels of DDT during pregnancy and in their daughters, compared with women with lower blood levels of DDT (28, 29).

Another evolving concept centers on the idea that mixtures of chemicals could act additively or synergistically or affect multiple pathways that together produce cancer (4). As exemplified by the work of the recent Halifax Project, chemicals that are ubiquitous in the environment, but not considered carcinogens individually, can exhibit actions on key cancer hallmarks at low doses (4). This raises the possibility that chemicals that are not carcinogenic in and of themselves could produce cancers when combined, or that some chemicals thought to be noncarcinogens could cause mutated cells to proliferate, thus promoting cancer emergence (4). The observation that cancer can develop by means of multiple pathways has implications for classifying and regulating carcinogens, and points to the need to recognize them as harmful substances that facilitate cancer development even though they may not cause cancer by themselves. Moreover, given that most toxicologic and epidemiologic studies investigate the role of single agents as carcinogens, these findings raise the need for continued research on the combined action of chemical mixtures at low doses—studies that better mirror the exposure environment in which we all live.

These advances in the field of environmental carcinogenesis indicate that reducing exposure to certain toxic substances can help reduce the burden of cancer in the population. However, applying this knowledge to prevent cancer will require more effort in several areas. It will also require recognizing and overcoming challenges particular to the field.

Establishing disease causation in humans is sometimes assumed to be a prerequisite for developing prevention strategies. However, methodologic challenges to doing so are substantial. Evidence from human studies requires relevant data on historical exposure timing and dosage, extended duration of follow-up, and highly powered studies. However, these are difficult and costly to obtain, and there are thousands of chemicals and mixtures to be studied. While clear evidence of carcinogenicity for some exposures is available—including hundreds of agents classified as carcinogenic by International Agency for Research on Cancer (IARC), the U.S. National Toxicology Program (NTP), or the U.S. EPA Integrated Risk Information System (IRIS) program—most evidence is less certain. Thus, it is critical for the scientific community to continue to strengthen research and tools for assessing the potential of chemicals and mixtures to contribute to cancer, to develop evidence-informed guidelines for implementation of risk and use reduction strategies, and to communicate across institutional and disciplinary silos about the streams of evidence that comprise robust science in this context.

Even when a strong evidence base exists for a role of chemical exposure in carcinogenesis, population approaches aimed at remediating exposures and reducing the development of exposure-related cancers face challenges. Some population-based approaches involve tradeoffs. Moreover, some preventive measures require near-term expenditures by entities that may never realize the associated downstream cost savings. Advice about risk reductions may change over time as new knowledge becomes available, which may confuse those following the advice. Thus, it can be difficult to discern a balance between population risks and the costs and sometimes unknown harms associated with lowering those risks.

Among the most significant challenges to reducing exposure to carcinogens are the limitations of the federal approach to regulating chemicals and the lack of communication and overlap between the many federal health agencies involved in cancer prevention and cancer research and those involved in various aspects of environmental regulation (Fig. 2). On the environment side, the Toxic Substances Control Act is the U.S. primary chemicals management law that governs the regulation of chemicals in commerce. Although recently amended in 2016, the U.S. EPA has not yet dealt with the thousands of chemicals that were grandfathered for use in commerce without rigorous assessments of carcinogenicity or other toxicologic properties.

Moreover, efforts to address cancer risks in workplaces, consumer products, or in the general environment are impeded by the disciplinary divisions between scientists working in cancer biology versus environmental health, which results in a tendency of some decision-makers to discount specific lines of evidence, because of the inherent ethical and practical limitations of environmental research. In addition, the influence of powerful vested interests may have an outsized influence on policymakers (30). This decision-making context also means that action to restrict a particular hazardous chemical can lead to “regrettable substitutions,” wherein a replacement turns out to pose health risks as problematic as the original chemical of concern.

Businesses investing in safer alternatives stand to benefit from accelerating attention to environmental carcinogens, but this requires aligned market and government signals. There are many potential impediments to such alignment, among them cultural resistance to a strong role for government, and lack of mechanisms for internalizing the costs of environmental degradation, whereby health risks would be accounted for in costs of production. Without such mechanisms, policy changes are needed to change the economic drivers toward safer alternatives.

Finally, an important challenge in translating reduced use of and exposure to environmental chemicals to prevention is the lack of a compelling endpoint of a successful cancer prevention strategy. We will never know the faces of people whose cancer has been prevented. Efforts to limit carcinogenic exposures may result in lower incidence rates, but these statistical differences over a long period have little emotional impact and make prevention strategies less interesting to policymakers or the general population than finding cures (31). For these reasons, along with the methodologic challenges of designing population-level cancer research, academic careers in cancer prevention are challenging. These barriers limit the number of people pursuing cancer prevention careers, and in turn, the volume of relevant research.

The path forward to advance understanding and action on environmental carcinogenesis requires investment and engagement by a range of stakeholders, including cancer research sponsors, research scientists, public health professionals, cancer advocates, policymakers, and community leaders. Some of the needs and opportunities pertain primarily to a particular constituency. Others will benefit from collaboration by stakeholders to influence systemic change.

New strategies are needed to prioritize chemicals for use or exposure reduction despite uncertainties about causation and risk level. These include ongoing investment in tools that can efficiently characterize chemicals of concern, strengthening regulation to reduce exposures, and incentives for and investment in the design of safer substitutes for hazardous materials. Government has an important role to play in supporting relevant programs, implementing regulations, and helping to grow new markets. Educating health professionals about the role of environmental chemicals in causing cancer and other health disorders, including the diverse types of scientific evidence that support reasonable risk reduction, is critical to bringing about changes that will lead to safer environments. Researchers and health professionals should not shy away from engagement in policy debates; their expertise and the public's perception of them as trusted messengers can help shift dominant narratives about what is needed and what is possible.

There is clearly a need for greater investment in research on environmental contributors to cancer and their impact on the “hallmarks” of cancer development and progression. Although cancer prevention is a major focus, the NCI does not explicitly identify environmental carcinogenesis as a high priority (32). Without sufficient funding to support basic, clinical, and population research on the roles of environmental chemicals in cancer—as well as intervention research—actionable information will lag, and new investigators will not be attracted to this field.

Critical for this area of research is engagement by scientists across the siloed disciplines of genetics, molecular biology, toxicology, epidemiology, and exposure assessment, among others, to advance understanding of the contribution of chemical exposures to cancer and the opportunities for prevention strategies. For example, new learnings about the key characteristics of carcinogens and tumor etiology, taken together with novel technologies for detecting biological changes after exposure, can inform the development of new toxicology screens to predict whether a chemical is likely a carcinogen. Although not without challenges, given that each discipline has its own language, research methods, and even separate scientific journals, there remains tremendous opportunity to create synergies and collaborations across cancer research simply by bringing people together to exchange and build upon knowledge and insights from multiple disciplines. Cross-training scientists across these disciplines and fostering interdisciplinary journals will accelerate this engagement.

Of special note is the importance of engaging advocates for the environment, public health, and cancer and community leaders in the design of research studies as well as in priority settings for research and prevention initiatives. Community advocates and other stakeholders bring with them the perspective of those impacted by cancer and by disproportionate chemical exposures, adding representation of cultural and other factors relevant to the research. Including these perspectives can help strengthen the research, its translation and dissemination, and financial support.

There is a need for greater understanding of and dialogue about the types of evidence needed to inform understanding of environmental carcinogenesis and to justify and guide interventions to reduce exposure and prevent disease. In clinical cancer research, evidence from double-blinded randomized control studies is considered the gold standard for determining efficacy and safety of treatment options. Such human experimental designs are infeasible for studying environmental carcinogenesis for both ethical and practical reasons. A range of evidence, from toxicology to molecular mechanistic studies to epidemiology, provides robust information about the role of environmental factors in cancer etiology that can and should inform the integration of environmental risk factors into cancer prevention activities at the individual and population levels.

Yet there remains significant resistance to drawing conclusions about the risk chemicals pose on the basis of these types of evidence. Faced with these less-familiar types of evidence and not wishing to cause alarm, clinical cancer experts and other health professionals frequently offer reassurance that chemical exposures have not been “proven” to cause cancer. Helpful approaches would instead communicate what we do know about the biological effects of chemicals on cancer pathways and the results of animal and human studies, so that people can contribute to decisions in accordance with their personal values about health risk. This discussion of the range of evidence will also enable decision-makers in the public and private sectors to support evidence-informed risk reduction strategies. In a similar vein, the 2010 President's Cancer Panel report on reducing environmental cancer risk encouraged the development of an “environmental health paradigm for long latency diseases,” given the significant cost to individual families and to society that would be incurred by waiting decades for robust cohort studies to demonstrate cause and effect in humans, especially given the expense of such studies and the thousands of chemicals that would need to be examined (1). Conversation and agreement about such a paradigm within the cancer research and prevention communities would help accelerate attention to and action on this important category of cancer risk factors. Engagement by cancer research leaders in integrating environmental risk factors into prevention initiatives holds tremendous promise for reducing both rates of disease and the burden of cancer on our society.

No potential conflicts of interest were disclosed.

J.G. Brody and R.A. Rudel were supported by the Cedar Tree Foundation and the Garfield Foundation. P.J. Hoppin and M.M. Jacobs were supported by the Garfield Foundation and the Heinz Endowments.