Integrating exposomics into biomedicine

Abstract

Assessing a full range of environmental exposures will improve human health

For the past two decades, the notion that an individual’s exposure to multiple factors in the environment can influence health over their lifetime has been steadily maturing into an independent field of study (1). How do combined exposures to chemicals in the water, poor nutrition, and social stressors influence metabolism, cardiovascular function, or cognition? Such questions are complicated, and investigative research now spans a range of disciplines, from epidemiology and social science to biochemistry and molecular biology. A major challenge for this expanding field of exposomics has been conceptualizing studies and developing methodologies (2, 3) in ways that connect the relevance of environmental exposures to biomedical research. Strengthening this association could benefit from establishing definitions that guide and unify the systematic analysis of environmental factors affecting health and disease. This would enable discovery-based analysis of the environmental influences on health and potentially improve disease prevention, treatment, and public health policies.

Last year, an operational definition for exposomics and its underlying principles was proposed by researchers from diverse disciplines and areas of expertise to provide a foundation on which concepts, approaches, methods, and results can be compared across disciplines, studies, and laboratories (4). Briefly, the exposome is posited as an integrated compilation of all physical, chemical, biological, and (psycho)social influences that "impact biology." The field of exposomics thus examines the comprehensive and cumulative effects of these factors by integrating data from interdisciplinary methodologies and data streams to drive discovery.

The rationale for including exposomics in biomedical research is straightforward. Only a fraction of chronic diseases can be primarily attributed to genetic factors. Thus, considering factors that constitute the exposome will help achieve a more thorough understanding of a given health condition. Exposomics is distinct from environmental health research in that it involves a longitudinal assessment across multiple environmental domains, ultimately linking them with effects on health. This approach also extends epidemiological studies and toxicological risk assessments that address the effects of specific environmental factors on a phenotype at a given time—such as exposure to solvents in the workplace and cognitive function. Exposomics is therefore, by definition, explicitly multifactorial and is amenable to combining with other large-scale “omics” studies, such as genomics, proteomics, and metabolomics. The merging of exposomics and genomics, for example, is empowering new health research studies focused on the effects of industrial chemicals on cancer (5); of air pollution on Alzheimer’s and Parkinson’s diseases (6); of nontobacco exposures on chronic obstructive pulmonary disease (7); and of combined lifestyle, dietary, and social factors on diabetes (8).

Many research pursuits can benefit from this multifactorial exposomic approach. For example, the changing climate can affect human health through natural disasters (and the consequential disruption of socioeconomic structures) and vector-borne disease. Some chronic diseases are affected by suboptimal nutrition, exposure to chemical pollutants, low physical activity, and lack of access to green space; yet, the details of these relationships are unclear. Although some of these conditions are not amenable to classical biochemical measurement, many responses are, such as altered cellular responses to stress and epigenetic modifications. Exposomics has the potential to improve the ability to identify the environmental drivers, physiological targets, relevant cellular pathways, and molecular transducers of health impacts. Moreover, exposome-related factors associated with a medical disorder may be readily modifiable.

Humans are exposed to multiple dynamic factors throughout their lives, yet research methods and regulatory agencies have not kept pace with this complexity and continue to be overly reliant on a “one exposure at a time” mindset. Exposomics is starting to change this strategy through methods and tools that are currently available. These include low- and high-resolution mass spectrometry for detecting multiple chemicals, nutrients, molecular signals, metabolites, and chemical modifications to the genome (epigenetic changes) caused by exposure to the environment. Geospatial techniques allow mapping of exposure sources spatially and temporally. This includes satellite-based remote sensing of air pollutants and wearable devices that monitor an individual’s exposure to pollutants.

These technologies are uncovering previously unknown linkages between small molecules or lifestyle factors and physiological disruptions or human disease. By using approaches that provide systematic and relatively unbiased associations between external factors, their sources, and their connection to biological perturbations (3), the paths to discovery become widened.

Several examples illuminate the promise that exposomics can bring to tackling a broader spectrum of challenges. High-resolution mass spectrometry–based analysis of small molecules identified the nutrient precursor trimethylamine N-oxide—a plasma metabolite that enhances thrombosis and is linked to cardiovascular disease (9). Related approaches have identified chlorinated metabolites associated with renal carcinoma in the blood of workers exposed to trichloroethylene (10). Exposure to pesticides has been connected to altered metabolic pathways in primary sclerosing cholangitis, a complex bile duct disorder (11). Geospatial science helped link exposures to particulate matter in the air with a diameter of 2.5 micrometers with biological processes in Alzheimer’s disease by using residential address history (12). In addition, exposomics approaches extend the research beyond cohort and population studies. The value of deep exposome phenotyping, demonstrated in individuals, elucidated a range of health conditions and the notable gap in knowledge that exists between human exposures and health outcomes. For example, intensive personal monitoring revealed the dynamic nature of chemical and microbial exposures (13). A population-scale effort that examined a large database of health insurance claims involving twins discerned both genetic and environmental contributions to more than 500 disease phenotypes (14).

How can the field move forward? Several areas deserve attention. Improved measurement and data systems will be essential for exposomics to progress beyond small-scale and candidate studies to become more widely adopted. Increasingly sensitive technologies to interrogate exposure mechanisms and cascading biological responses are needed—lessons learned during the rise of genomics. Fortunately, many of the tools are familiar, and the types of data that come from exposomics are like those from related mass spectrometry–based methods (e.g., metabolomics or proteomics) or resemble geospatial features used in spatial epidemiology.

Efforts are also needed to develop biomedical measurement technologies (for biospecimens and environmental samples) and strategies for combining them with other omic technologies and studies to evaluate their use at the point of care. Another goal is to establish new methods to monitor or estimate individual-level exposures over the life course that can be applied in research, public health, and clinical applications. Creating a human exposome reference using systematic and spatially resolved characterization of exposures in biological systems through life spans would enable analysis and contextualization at the population scale to support epidemiological research and policy decisions. Developing wearable or minimally invasive tools that measure an individual’s exposome and strategies for commoditization would be a catalyst for research. And creating a framework for the standardization and harmonization of language, protocols, data collection, and large-scale and high-throughput analysis methods is needed. Quality control and quality assurance will maximize the value of data in a FAIR (findable, accessible, interoperable, and reusable) format. This framework would facilitate the application of machine learning and artificial intelligence to analyze the complex datasets from exposome studies.

As these initiatives are pursued, several high-reward but high-risk research avenues currently exist. Exposomics will complement pharmacogenomics by answering how combined environmental exposures and dietary factors alter drug metabolism and efficacy. Translational cancer research and clinical trials will benefit from the systematic evaluation of environmental factors on initiation, progression, metastasis, and response to therapy. In addition, its application will provide powerful approaches for identifying nongenetic drivers of health disparities and enhancing occupational health and safety by linking known work-related exposures to disruption of biological pathways and by identifying currently unknown exposures.

All large-scale studies of human disease, whether they are cohort, clinical, or biobank, could incorporate exposomics (1). Even if research teams lack expertise in exposomics, merely collecting extra aliquots of blood or geospatial information for future analysis can preserve potential knowledge and enhance reuse of cohort data samples. Moreover, performing exposomics on a population scale will allow the creation of exposure that combines individual-level measures using humans as sentinels with geospatial data from monitoring systems. Such approaches would empower policy-makers to make data-driven decisions for mitigation and prevention. It should be noted that improvements in exposomics have been uneven, with more attention paid to chemical exposures and less to social and physical influences. For humans, environmental exposures are undoubtedly associated with social determinants of health, and policy implications are therefore far-reaching.

Exposomics allows researchers to reframe studies from a reactive approach to a predictive and preemptive one. However, this requires attention to the ethical, legal, and social aspects of its use across many sectors (15). Issues such as data ownership and privacy, the effects of disparate exposures among socioeconomic groups, and uneven access to healthy environments need to be addressed. In addition, clear guidelines for returning results to study participants must be developed. Inclusion of actionable steps to improve health will help to establish public trust in exposomics studies and outcomes.

The interdisciplinarity that underlies exposomics requires investigators to cultivate a collaborative culture to solve major environmental and societal challenges. This includes engaging with community members who have often been research participants but not directly involved in the planning, formulation, and implementation process. The recently established coordinating centers for exposomics in the United States and Europe provide the foundation to foster the required global collaboration, establish best practices, support data harmonization, and train the next generation in the collaborative skills essential for success.

Human phenotype is determined by the dynamic combination of genetic and environmental factors, with stochastic events adding uncertainty on both sides independently and at the interaction level. Gaps in understanding human health and disease can be filled by integrating exposomics into the biomedical enterprise. Life exists at the interface of genetically encoded processes and environmentally driven realities. So too should the biomedical enterprise that studies it.