Cancers are fundamentally complex, multifactorial, genomic, and epigenomic diseases,1, 2, 3, 4, 5 which represent a major burden on societies globally. However, many cancers are potentially avoidable, with estimates suggesting that 60% of cancer deaths in the United States are attributable to a limited number of lifestyle factors.6 A better understanding of modifiable contributors to cancer initiation, evolution, and progression is a prerequisite for accurate risk prediction and the development of better strategies for prevention, early detection, treatment, and surveillance.1, 2, 3, 4, 5, 7, 8, 9, 10

The revolutionary theory of field effect (also known as field defect, field cancerization, or field carcinogenesis) has been continually adapted and updated since it was first consolidated by Slaughter et al11 in 1953. The description of ‘field cancerization’11 is regarded as one of the landmarks of the past 100 years of cancer research.12 Several authors have recently reviewed the field effect concept and its evolution,13, 14, 15, 16, 17, 18 and the presence of a recent textbook19 entirely devoted to the topic attests to its continued clinical and scientific importance.

In this article, we offer a reappraisal of field effect, approaching the concept from an etiologic perspective. We term this alternative model ‘etiologic field effect’, where endogenous and exogenous etiologic factors (such as dietary, lifestyle, environmental, microbial, hormonal, and genetic variations), and their interactions, predispose to an abnormal tissue microenvironmental milieu that can influence all stages of tumor evolution. We have taken into account the possible contribution of stromal cells and the microenvironment,20, 21, 22, 23, 24 and developed a paradigm where macroenvironmental and microenvironmental influences, in their totality, contribute to a field of etiologic predisposition to disease. The etiologic field effect concept embraces tumor–host interactions,3, 25, 26, 27, 28, 29, 30, 31, 32 and gene–environment interactions, which have become increasingly important in molecular epidemiology.33 There are several advantages to an etiologically oriented model of field effect, as elaborated upon in the following sections. An etiologic field effect concept can enhance the scope of the traditional field effect model and can effectively explain a variety of phenomena relevant to cancer causation and progression. Effective lifestyle interventions, such as dietary modification and physical activity, can be thought of as mechanisms through which etiologic fields can be attenuated throughout the body, preventing cancer occurrence and progression, and decreasing cancer burden in our society.34, 35

Evolution of the field effect concept

The concept of field effect was proposed by Slaughter et al,11 in their 1953 landmark paper, in an attempt to explain the phenomenon of synchronous or metachronous primary tumors arising within the oral mucosa. In this context, ‘field effect’ implied an inherent predisposition of the non-cancerous mucosa to malignant transformation.

Before the advent of molecular pathology, support for the field effect model relied on pathological observations describing histological abnormalities in grossly normal-appearing tissue adjacent to cancers.11 Subsequently, molecular genetic analyses of cancerous, precancerous, and normal tissues have yielded persuasive evidence that a field of cancer-predisposing molecular alterations can be present even in microscopically normal tissue.13, 14, 15, 16, 17, 18, 19, 36, 37 The existence of field effect has been described in a variety of tissues, spanning almost all organ systems in the human body.13, 14, 15, 16, 17, 18, 19, 38, 39, 40, 41, 42, 43, 44, 45 More recently, studies of epigenetic changes in tumors and normal cells,14, 15, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57 as well as analyses of stromal cells and the tissue microenvironment have contributed to the molecular field effect concept.20, 21, 22, 23, 24

Consequently, the prevailing interpretation of field effect is that a field of somatic molecular alteration in a given organ or tissue predisposes to tumor development within that field. The mechanisms through which geographic fields of molecularly abnormal cells arise are not fully understood, but clonal expansion and intraepithelial migration of genetically altered cells within contiguous epithelial structures has been proposed.36, 58 As a result of supporting evidence that has accrued through histopathological, genetic, and, most recently, epigenetic studies, the field effect concept has become firmly established.13, 14, 15, 16, 17, 18, 19 Importantly, altered molecular field may represent a potential therapeutic target.59, 60

Synchronous primary tumors and field effect: insights and intrigue

Field effects have been implicated in the co-occurrence of tumors in more than one organ. Examples of such multi-organ involvement by field effect include tumors arising within the ductal epithelia of the pancreas, ampulla of Vater, extrahepatic bile ducts, and gallbladder,44 respiratory epithelia of the lung, bronchi, trachea, larynx and nose,59, 61 and urothelium of the bladder, ureters, and renal pelves.45, 62, 63, 64 While shared ontogeny has been hypothesized to contribute to certain multi-organ field effects (eg in epithelia derived from embryonic foregut),44 multi-organ involvement can span tissues arising from more than one embryonic germ layer, as exemplified by urothelial tumors; upper urinary tract structures derive from the mesonephros (mesoderm), while the bladder and urethra arise from the urogenital sinus (endoderm).

The term ‘synchronous tumors’ generally refers to two or more primary tumors arising contemporaneously within a single individual. Synchronous primary neoplasms, particularly those originating within a single contiguous organ system, may develop as a result of field effect phenomena. It should be noted, however, that the absence of synchronous, or even a solitary cancer, does not exclude the existence of field effect; cancer development, per se, is not a requirement for defining field effect. This is an important consideration if we are to exploit field effect to develop screening and preventive strategies.

Synchronous cancers can provide a unique insight into the somatic molecular aberrations that might constitute a field effect.65, 66 Indeed, a number of studies have documented the presence of shared molecular features between synchronous primary tumors,67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77 which would support the influence of a field effect present in the ostensibly ‘normal’ tissue from which they have arisen.13, 14, 15, 16, 17 Alternatively, it can be considered that synchronous primary tumors arise through the interplay between common etiologic contributors, such as genetic predisposition, microbial and environmental exposures, and lifestyle factors, which facilitate progression through certain common carcinogenic pathways.65, 67, 71 Importantly, the latter model does not imply that the molecular features shared by synchronous tumors need to be present in the background ‘normal-appearing’ tissue.

In structurally continuous epithelial tissues, such as the orodigestive and respiratory mucosae, it is perhaps conceptually easy to envisage how multiple primary tumors could arise from a field of molecularly altered cells. It is well documented, however, that field effects appear to influence carcinogenesis in non-contiguous structures, for example, in bilateral primary breast cancer. Right and left breasts are separate organs, with no connections existing between glandular or ductal epithelia of the right and left breasts. Despite this anatomic independence, a number of studies have shown that synchronous and metachronous bilateral breast cancers tend to demonstrate concordance in expression status for the hormone receptors ESR1 (estrogen receptor-alpha) and PGR (progesterone receptor).71 These data imply that, in bilateral breast cancer, independent primary tumors tend to evolve through similar carcinogenic pathways, compared with cancers arising in two different individuals.

Similarly, spatially distinct primary tumors can arise within a single organ system despite the complete absence of detectable classical field change in the background normal mucosa. We have previously demonstrated that synchronous colorectal cancers tend to show concordant molecular features (including DNA hypomethylation and CpG island hypermethylation) without similar molecular changes in intervening normal colonic mucosa.67 While these observations are not necessarily at odds with the conventional field effect model, they underscore the need for emphasis to be placed on the putative factors that predispose to tumor initiation, or facilitate tumor evolution through specific, common carcinogenic pathways, where the end result is tumors with shared molecular features. It would also seem advantageous to develop a model where detectable somatic alterations, similar to those found in established cancers, are not prerequisite for defining a field.

Predisposition to neoplasia: exploring the geographic limits of field effect

It is incontrovertible that multiple primary tumors can arise within a background population of genetically predisposed cells. One concrete example of this mechanism is the presence of highly penetrant cancer syndromes,78 such as Lynch syndrome, where genetic predisposition to multiple primary tumors in one or more organ systems has been described.79, 80 In genetic predisposition syndromes, virtually every cell in the body carries a copy of the mutated gene and, as such, these syndromes may be considered whole-body mutational field effects. Germline inheritance of cancer-predisposing variants is perhaps beyond the intended scope of the conventional field effect concept. Furthermore, familial cancer predisposition syndromes contribute only a relatively small proportion of cases to the overall incidence figures for common cancers. Genetic influences, however, remain important in ‘sporadic’ cancers, which are considered to result from the interplay of genetic and environmental influences.81, 82

In non-syndromic cancers, evidence suggests that high-prevalence low-penetrance genetic variants (including those identified by genome-wide association study, GWAS) predispose to the acquisition of specific somatic molecular alterations.83, 84 For example, studies have shown a consistent association between a common single nucleotide polymorphism in the MGMT promoter (rs16906252), and MGMT promoter CpG island hypermethylation in several cell types, including colorectal cancer,85 normal colonic cells,86 normal peripheral blood cells,87 lung adenocarcinoma and premalignant lesions,88 and malignant pleural mesothelioma.89 In this example, a whole-body field effect, conferred by the common MGMT promoter single nucleotide polymorphism, appears to predispose many distinct cell types to an acquired epigenetic event, ie, somatic MGMT promoter hypermethylation.

In a manner analogous to inherited genetic variants, environmental, and other exogenous exposures, such as dietary and lifestyle factors, may also actuate or promote the accrual of specific somatic genetic or epigenetic alterations.90, 91 Indeed, epigenetic mechanisms are recognized to serve as a link between environmental influences and gene regulation.92

Beyond cancer syndromes, it is postulated that genetic determinants, endogenous, and exogenous environmental exposures can influence neoplastic transformation at multiple body sites. One would therefore expect there to be evidence of a wider, multi-organ, or even whole-body, field effect in cancer-predisposed individuals. It has been proposed that biomarkers at ‘surrogate anatomic/functional sites’ can be evaluated for presence of ‘an extended field effect’, indicative of elevated cancer risk.54, 61, 62, 93, 94 Likewise, epigenetic aberrations detectable in peripheral blood cells have been speculated to reflect constitutional cancer susceptibility.95, 96 Importantly, epidemiological studies have demonstrated clear pleiotropic effects for certain etiologic exposures, most notably smoking. Moreover, data from a large population-based study suggest increased familial clustering of cancers at different sites, which may be due to common genetic susceptibility as well as shared environmental exposures.97 Thus, evidence does exist to support a more systemic etiologic contribution to field effect, involving the interplay between germline genetic variants and environmental exposures.

The genesis of the ‘etiologic field effect’

The evolving field of molecular pathological epidemiology (MPE) has resulted in the discovery of a number of robust relationships between etiologic factors and somatic molecular alterations in human cancers and normal tissues.90, 91 These relationships include the associations between the following: reproductive and hormonal influences (such as age at menarche, parity, age at first full term-pregnancy, lactation, and hormone therapy) and risk of molecularly defined subtypes of breast cancer (by ESR1, PGR and ERBB2 expression);98, 99, 100, 101 the MLH1 rs1800734 single nucleotide polymorphism and MLH1 promoter hypermethylation (or microsatellite instability) in endometrial and colorectal cancers;102, 103, 104, 105, 106 genetic modifiers of one-carbon metabolism, micronutrient intake, and DNA methylation in colorectal cancer;107, 108, 109, 110, 111, 112, 113, 114 body mass index (or obesity), microsatellite instability, and fatty acid synthase (FASN) expression level in colorectal cancer;115, 116, 117, 118, 119, 120, 121 cigarette smoking and microsatellite instability, CpG island methylator phenotype, and BRAF mutation in colorectal cancer;122, 123, 124, 125, 126, 127 cigarette smoking and KRAS mutation in lung tumors;128, 129, 130 young age of onset and family history of colorectal cancer, and LINE-1-hypomethylated colorectal cancer;131, 132, 133, 134, 135 interactions between aspirin use and molecular features of colorectal cancer;136, 137, 138, 139 Epstein–Barr virus and CpG island hypermethylation in gastric cancer;140, 141, 142 H. pylori infection and CpG island methylation in gastric epithelial cells;143, 144 viral hepatitis and CpG island hypermethylation in hepatocellular carcinoma;145, 146, 147 and the MGMT rs16906252 promoter single nucleotide polymorphism and MGMT promoter hypermethylation in normal cells and various cancers.85, 86, 87, 88, 89 The MPE paradigm, which can encompass all human diseases,148 is firmly established,149, 150, 151, 152 and has gained widespread recognition.121, 153, 154, 155, 156, 157, 158, 159, 160, 161, 162, 163, 164, 165, 166, 167, 168, 169, 170, 171, 172, 173, 174, 175, 176

As a result of MPE research, considerable evidence has accumulated to support the concept that cells in one or more organs, or organ systems, can become predisposed, in a tissue-specific or less specific fashion, to the acquisition of well-defined somatic molecular alterations. Thus, internal susceptibility, in the form of heritable genetic and epigenetic variants, and exposure to exogenous influences, such as microorganisms, environmental toxins, dietary components, and lifestyle factors, converge at the level of the tissue microenvironment, and mediate the propensity to neoplastic transformation and progression through cell–cell and extracellular matrix–cell interactions.4, 177, 178 In other words, every cancer (or, indeed, every disease process) results from changes in interactomes, with interactomes ultimately dictating how tumors behave. Interactomes179 encompass the entirety of complex molecular interactions within a cell, between cells and stromal components in the tissue microenvironment, within tissues and organs, and even at the level of a whole organism. The interactome is the ultimate interface through which external stimuli interact with host biological systems. The interactome therefore includes gene–environment interactions, which have become increasingly important as the basis for molecular epidemiology studies. Early changes in interactomes can be regarded as an expanded notion of ‘field effect’, and could be utilized as a marker of increased tumor or disease risk.

There is no real provision for the interaction of heritable and environmental risk modifiers in the conventional interpretation of field effect. Taking into account the aforementioned MPE research findings, and cognizant of the importance of the potential for interplay between diverse etiologic exposures in carcinogenesis, we developed an alternative field effect model, which we term ‘etiologic field effect’. In contrast to the conventional field effect model, where the ‘cancer-susceptible field’ comprises a distinct molecular or cellular change in an anatomically defined area, the ‘etiologic field effect’ focuses principally on the dynamic interplay between fields of exposure to etiologic factors, which may alter a tissue’s microenvironmental milieu. An etiologic field can be defined as ‘a functional field of altered tissue microenvironment that predisposes to the acquisition of specific somatic molecular changes through alterations in cellular and extracellular interactomes’. Etiologic fields are characterized by the presence of common etiologic exposures, rather than by cellular molecular aberrations. Since exposures frequently transcend anatomic boundaries, etiologic fields are not restricted to contiguous epithelial structures. As etiologic exposures predate the establishment of pathological cellular and molecular aberrations that lead to neoplastic initiation and progression, ‘etiologic field effects’ are potentially reversible and represent modifiable targets for intervention. Table 1 contrasts key features of the conventional field effect with those of the ‘etiologic field effect’ concept.

Table 1 Comparison of the key features of the conventional and ‘etiologic field effect’ models

There are tangible examples that illustrate how an etiologically based model is better positioned to explain certain field effect phenomena. Smoking is well documented as a risk factor for colorectal cancers that display specific molecular features, namely, CpG island methylator phenotype, microsatellite instability, and BRAF mutation.122, 123, 124, 125, 126, 127 Smoking has been associated with genome-wide DNA methylation changes in blood leukocytes, which may imply its systemic effect on cellular epigenetic status.180, 181 Smoking has also been shown to be a strong risk factor for synchronous primary colorectal cancers (R Nishihara et al, unpublished data) and synchronous multiple polyps,182 especially serrated polyps,183 which are recognized as precursors for colorectal cancers with CpG island methylator phenotype, microsatellite instability and/or BRAF mutation.159, 184, 185, 186 CpG island methylator phenotype-high, microsatellite instability-high and BRAF mutation can co-occur in colorectal cancer,9, 30, 113, 184, 187, 188, 189, 190, 191, 192, 193, 194, 195 and are common characteristics of synchronous colorectal cancers.67, 68, 69, 70 Furthermore, synchronous primary colorectal cancers are considered to arise due to some form of predisposition, likely involving both genetic and environmental factors.196 However, there has been little evidence for conventional field effect in individuals with synchronous colorectal cancers, ie, CpG island methylator phenotype, microsatellite instability, and BRAF mutation are not found in normal colonic cells adjacent to synchronous colorectal cancers that demonstrate these somatic molecular aberrations.67 Furthermore, smoking has been consistently associated with BRAF-mutated colorectal cancer,122, 123, 124, 125, 126, 127 but not with KRAS-mutated colorectal cancer, where data are conflicting and complicated by publication bias.197, 198, 199, 200, 201 It is difficult to explain the gene specificity of mutations if one assumes the role of tobacco smoke as only a direct mutagen. Considering these pieces of evidence together, it seems plausible that smoking generates a field of tissue microenvironmental changes (as opposed to directly causing CpG island methylator phenotype, microsatellite instability or BRAF mutation), which may be advantageous for the growth of specific neoplastic/preneoplastic cells harboring BRAF mutation, but not so conducive to the grown of KRAS-mutated cells. The etiologic field of microenvironmental changes induced by smoking may predispose to the development of multiple cancers through similar carcinogenic pathways (Figure 1).

Figure 1
figure 1

An example of evidence for an ‘etiologic field effect’ phenomenon. Smoking has been shown to increase the incidence of colorectal cancer displaying CpG island methylator phenotype, microsatellite instability, and BRAF mutation, as well as the incidence of synchronous colorectal cancers. Smoking, as an etiologic exposure, creates a field effect manifest as altered colonic tissue microenvironment. Microenvironmental change induced by smoking promotes carcinogenesis via specific pathways resulting in synchronous tumors with shared molecular features. As one might expect of an etiologic field, somatic alterations (such as CpG island methylator phenotype, microsatellite instability, and BRAF mutation) are absent from non-neoplastic tissue within the field.

Another example that profits from the adoption of an ‘etiologic field effect’ model is the association between early breast neoplasia (including lobular carcinoma in situ and atypical ductal hyperplasia) in one breast, and increased risk of subsequent invasive breast cancer in the same breast, and also the contralateral breast.75, 76 This phenomenon is likely related to tissue-specific gene-environmental interactions, which can be considered to constitute a type of field effect. The presence of an etiological field therefore adequately explains this observation; certain shared etiologic exposures lead to changes in the breast tissue microenvironment resulting in cancer susceptibility in both breasts.

In Figure 2, using the colon as an example, we illustrate the temporal extent of ‘etiologic field effect’ concept and compare it with the conventional model of field effect. In the conventional model, cancers arise in a field of cells harboring acquired somatic molecular alterations. The ‘etiologic field effect’ model takes into account both host and exogenous factors, which, together, constitute a field of microenvironmental alterations and susceptibility to cancer development and progression.3, 90, 91

Figure 2
figure 2

Comparison of conventional and etiologic field effect models, using colon cancer as an example. (a) In the conventional field effect model, a field of alteration ‘X’ in normal tissue (eg, colon) makes an individual prone to cancer development within that field. Normal colon in the field, and resulting synchronous tumors, show the same molecular alteration ‘X’. (b) In an ‘etiologic field effect’ model, etiologic factors (which can be multifactorial) generate a field of tissue microenvironmental changes, favoring the development of cancers through a common carcinogenic pathway characterized by molecular alteration ‘X’. In this setting, resulting synchronous tumors demonstrate the same alteration ‘X’, but normal tissue (eg, colon) need not necessarily display the same molecular feature.

Tumor initiation, evolution and progression: exploring the temporal limits of field effect

In contrast to the conventional field effect concept, which essentially pertains to molecular events associated with the initiating phases of neoplasia, ‘etiologic field effect’ extends temporally to incorporate biologic and physical etiologic factors that promote microenvironmental changes leading to cellular transformation, invasion, and metastasis (Figure 3). For example, etiologic factors that provoke or modulate inflammation (eg microbes, genetic polymorphisms, drugs, and dietary exposures) could contribute to an ‘etiologic field effect’ that remains influential at all stages of tumor evolution. Indeed, accumulating evidence on the anti-neoplastic effects of aspirin, and other inhibitors of PTGS2 (cyclooxygenase-2), supports exactly such a model; inflammatory processes, susceptible to the effects of these drugs, appear to be important in the early phases of neoplasia (eg colonic adenomagenesis),202, 203 during cancer evolution,204, 205, 206 and after cancer diagnosis,137, 138 possibly including the development of distant metastases.205 Tobacco smoke is a further example of an exposure that influences multiple phases of tumor evolution. In addition to a role in the initiating phases of bronchial carcinogenesis, components of tobacco smoke are implicated in promoting lung cancer growth and metastasis.207, 208 Similarly, cigarette smoke, a risk factor for breast cancer, may promote epithelial-mesenchymal transition and increase the metastatic potential of breast cancer cells.209

Figure 3
figure 3

One of the principal differences between the ‘etiologic field effect’ and conventional field effect concepts is their temporal associations. The conventional field effect typically spans a relatively narrow part of neoplastic evolution spectrum, from the acquisition of somatic aberrations to histologically dysplastic pre-malignancy. The ‘etiologic field effect’, by comparison, is relevant at all stages from neoplasia initiation to patient death. The presence of ‘etiologic field effect’ precedes the acquisition of pathologic somatic changes, and extends to be influential in tumor evolution, invasion, and metastatic growth.

Tumor establishment at metastatic sites is dependent on physical cellular interactions and cross-talk between genetic, epigenetic, metabolomic and environmental factors occurring in the local tissue microenvironment.210 Thus, both tumor cell migration and the presence of a pro-metastatic microenvironmental niche, conducive to tumor seeding, could be ascribed to the presence of etiologic field effects.

The ‘etiologic field effect’ is not limited to epithelial cells, and embraces tumor–stromal interactions in the microenvironment, as well as macroenvironmental exposures and gene–environment interactions that effect microenvironmental change.33 Interestingly, tumor stroma and microenvironment may determine cancer molecular phenotype,211 and even response to molecularly targeted therapies.27, 212, 213 A small number of studies have highlighted the contribution of tumor stroma to field effect phenomena;20, 21, 22, 23, 24 however, the discussion in these studies20, 21, 22, 23, 24 has tended to be limited by the conventional notion of field effect.

Of note, ‘etiologic field effect’ phenomena do not need to be discrete from phenomena that constitute conventional field effect; rather, the two overlap within the spectrum of cancer predisposition. As with many features of biological systems, a ‘continuum’ model may afford a better representation of reality.78, 214, 215 Etiologic fields may seamlessly span multiple phases of neoplastic evolution, while their anatomic boundaries are likely to be gradients of tissue microenvironmental change, determined by variation in the magnitude of and sensitivity to a particular exposure.

Advantages and implications of the ‘etiologic field effect’ concept

The ‘etiologic field effect’ concept, as we perceive it, is attractive. First, it does not conflict or diminish the importance of the conventional notion of field effect; rather, the ‘etiologic field effect’ concept extends the temporal boundaries of the existing paradigm, to encompass the entire process of tumor evolution; exposure to etiologic factors generates a field of altered tissue interactome that remains influential throughout carcinogenesis, from initiation to progression and metastasis, and, ultimately, the demise of the patient. Second, the ‘etiologic field effect’ concept shifts the focus of attention from somatic genetic and epigenetic alterations to the influence of etiologic factors that might predispose to the acquisition of pathological molecular alterations in the first place. The ‘etiologic field effect’ therefore broadens the horizons of our inquiry into cancer susceptibility at molecular, cellular, and environmental levels.

Most cancer and pre-cancer surveillance protocols are based on the assumption of a persisting etiologic susceptibility. In clinical practice, modifiable components of etiological risk can be easily overlooked; in scheduling a polyp surveillance colonoscopy, we might miss opportunities for lifestyle interventions targeted at adenoma risk factors.216, 217 The term ‘etiologic field effect’ can successfully conceptualize the rather vague assumption of etiologic predisposition into a more concrete biomedical paradigm with a focus on risk modification and disease prevention. This concept can successfully act as model for dietary and lifestyle interventions, which have the potential to attenuate etiologic fields.

Epidemiologic and translational studies tend to use cancer occurrences or premalignant intermediaries (eg colorectal adenomas) as end points. For effective preventive strategies, we must strive to identify earlier markers of disease risk. Epidemiologic studies have linked low folate intake to increased incidence of colorectal adenomas and cancer.218, 219 It is also known that tissue folate status can modify molecular events in normal colonic mucosa.220 As not all individuals with low tissue folate will develop colorectal adenomas or cancer, the ability to interrogate the normal mucosal interactome for evidence of a pathological response to folate depletion could aid risk stratification, allowing for dietary or pharmacologic therapy to be instituted where appropriate. Biosensors of the presence and magnitude of ‘etiologic field effect’ could therefore facilitate the development of personalized prevention and treatment strategies.

An ‘etiologic field effect’ model may be informative for clinical and translational research. Clearly, the potential for etiologic field effect to act as a marker of the risk of cancer development, progression and/or metastasis constitutes an important area for future investigation. If etiologic field can act as a surrogate for disease risk, then evidence of etiological field could be used as an outcome measure in interventional studies of lifestyle modification or pharmacologic therapies. The possibility that reversal or modification of the etiologic field to a more ‘normal’ state can serve as a preventive strategy also demands scrutiny in future studies.

As described above, etiologic field effect can manifest as altered microenvironmental properties or abnormal patterns of cellular and tissue response to various endogenous and exogenous factors. We speculate that, in the future, the integrated efforts of ‘omics’ research, physical sciences, systems biology,2, 221 and nanotechnologies222 will help characterize complex molecular patterns (epigenomic, proteomic, metabolomic, etc) of host interactions with exogenous factors, giving rise to markers capable of indicating exposure to etiologic fields (Table 2). Markers of etiologic fields are likely to reflect altered exposome, with the exposome encompassing all exposures to which an individual is subjected.223 We would suggest that future biomarker discovery and validation efforts should focus on the identification of biosensors that signal the presence of disease susceptibility, rather than indicate established, albeit early pathologic changes. By means of illustration, recent data provide evidence for an epigenetic field of promoter methylation in normal colonic mucosa, involving genes distinct from those methylated in colorectal cancers.224 Furthermore, pathway analysis demonstrates that many of the differentially methylated genes are involved in carbohydrate metabolism, suggesting complex interactions between luminal contents and the gut microbiome and metabolome.

Table 2 Potential approaches to discovery of ‘etiologic field effect’ markers

Several biosensors and nanotechnologies have already demonstrated the capacity for in vivo assessment of tissue ultrastructural and microvascular correlates of the genetic and epigenetic aberrations that define conventional field effects.222, 225, 226 Indeed, nano-cytology and nano-cytoarchitecture have been proposed as screening targets for field carcinogenesis.226, 227 It is therefore conceivable that the sophistication of future biosensors will enable accurate assessment of in vivo real-time changes in the microenvironmental interactome with high resolution (ie, in vivo molecular pathology). By detecting the very earliest alterations in the biologic or physical properties of tissues, exposure to detrimental etiological fields may be revealed long before the acquisition of irreversible molecular aberrations.


The existence of the conventional field effect is irrefutable, and is supported by an abundance of published research. Historically, advances in science and technology have allowed us to define field effects at increasingly earlier stages in carcinogenesis. The new paradigm of ‘etiologic field effect’ represents a further advancement of the field effect concept. This overarching interpretation of field effect has not previously been conceptualized or consolidated, as far as we are aware. Importantly, the ‘etiologic field effect’ is better positioned to explain several intriguing observations that are inadequately accounted for by the conventional interpretation of field effect. An important agenda for future research in complex diseases, such as cancer, is to decipher the complex interaction of etiologic exposures in an attempt to understand how they predispose to the acquisition of specific molecular aberrations and facilitate malignant transformation, tumor growth, migration, and metastasis. We believe that an etiologically focused, holistic, approach to the field effect paradigm can lead to a better understanding of cancer predisposition and progression, which, in turn, can facilitate the design of improved personalized cancer prevention and treatment strategies.