The Teleological Fallacy in Microbiome-Cancer Research

The Teleological Fallacy in Microbiome-Cancer Research Dr. Love Ogbeyalu Charles-Ukeagu  Department of Microbiology, Immunology and Pharmacology, St George’s University, St George’s, Grenada Abstract The discovery of tumor-associated microbes like Fusobacterium nucleatum has led to striking descriptions of bacteria "exploiting" tumors or "evolving to promote cancer" a language that smuggles in teleological assumptions about purpose in evolution. This paper interrogated such narratives through Gould and Lewontin's critique of adaptationism and Dennett's concept of the intentional stance, asking whether microbiologists risk conflating selected effects with goal-directed behavior. Through discourse analysis of 20 key oncology papers (2015-2024) and two case studies (F. nucleatum's immunesuppressing Fap2 protein and colibactin-producing E. coli), the study found that 65% of studies employ teleological language without evidence of bacterial intent. These metaphors as a finding, while heuristically useful, may obscure the complex evolutionary pressures shaping microbial traits in tumors. The paper recommended replacing purpose-laden terms like "oncobiont" with mechanistic descriptions ("trait X enhances persistence in niche Y") and designing experiments that distinguish true adaptation from exaptation in tumor microbiomes. Introduction The association between Fusobacterium nucleatum and colorectal cancer is one of the most striking findings in microbiome-oncology over the last decade. [1] demonstrated that the bacterium not only colonizes tumor tissue but also potentiates intestinal tumorigenesis and modulates the tumor-immune microenvironment. Such findings were quickly taken up in the literature, where F. nucleatum was described as if it had the “capacity to exploit tumors” or even as an “oncobiont” that evolved to promote cancer [2]. While these terms convey the severity of microbial involvement in cancer, they also smuggle in assumptions of purpose and agency. To speak of a bacterium “exploiting” a tumor is to imply intent, when in reality, bacterial survival is driven by environmental pressures and chance mutations, not foresight. This challenge between scientific description and metaphorical shorthand has been noticed by some microbiome researchers themselves. [3] for instance, reported that F. nucleatum persists in human colorectal tumors even after metastasis, leading them to propose that it may “hitchhike” with cancer cells as they spread through the body. The image of hitchhiking captures the phenomenon vividly, but it also ascribes a goal to the bacterium, as if it deliberately chooses to travel. The problem here is not only rhetorical because by portraying bacteria as agents with purposes, researchers may inadvertently confuse proximate mechanisms with ultimate evolutionary explanations. This is what [4] criticized in their classic essay on the “adaptationist programme,” urging against explanations that assume every trait is shaped by direct adaptive purpose. In the context of tumor microbiomes, the danger is that scientists interpret microbial traits in teleological terms when they may simply be exaptations by-products of other evolutionary processes. [5] showed that Fap2 binds to the human inhibitory receptor TIGIT, suppressing anti-tumor immunity. The way this result was received in the literature often implied that F. nucleatum had evolved the protein to manipulate human immune defenses in tumors. Yet Fap2’s immune interactions may have arisen under very different selection pressures in the oral cavity, where the bacterium normally resides [2]. Its role in More Information How to cite this article: Charles-Ukeagu LO. The Teleological Fallacy in Microbiome-Cancer Research. Eur J Med Health Res, 2025;3(6):71-9. DOI: 10.59324/ejmhr.2025.3(6).12 Keywords: Teleology, microbiome-cancer interactions, adaptationism, oncobionts, philosophy of science, evolutionary biology. This work is licensed under a Creative Commons Attribution 4.0 International License. The license permits unrestricted use, distribution, and reproduction in any medium, on the condition that users give exact credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if they made any changes. EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 72 cancer may therefore be an exaptation which is a functionally relevant trait in a new environment that was not originally selected for cancer-related purposes. To describe this as bacteria “hijacking the tumor microenvironment” [6] introduces a teleological fallacy, equating survival outcomes with intentional design. [7] notion of the “intentional stance” helps explain why such metaphors are misrepresenting. This is because attributing beliefs and desires to a system, scientists can generate useful predictions. Saying for instance that bacteria “want to persist” in tumors is shorthand for describing ecological pressures favoring traits that enhance survival in that niche. However, the stance is a heuristic, not a literal truth because when microbiologists adopt the intentional stance without caution, they may push forward metaphors into causal explanations. This is particularly concerning in oncology, where the language of microbial “oncobionts” frames bacteria as purposeful cancerpromoting entities, rather than organisms opportunistically thriving in disturbed tissue environments demanding a close scrutiny because of its many implications. Objectives of the Study i. The central objective of this study is to interrogate the persistence of teleological reasoning in microbiome– cancer discourse and to provide a more rigorous conceptual framework for analyzing microbial traits within tumors. Others are to: ii. Examine how teleological expressions such as “oncobiont,” “tumor-promoting microbe,” or “bacteria exploiting tumors” are deployed in influential oncology papers. iii. Analyze philosophical resources and perspectives on the subject matter. iv. Analyze two case studies in depth which are the F. nucleatum’s Fap2 protein and the colibactin-producing Escherichia coli, which induces DNA damage and mutations. v. Propose alternatives to teleological phrasing in scientific writing. v. Generate a broader reflection on how metaphors influence scientific understanding. Research Questions i. How frequently and in what forms does teleological language appear in microbiome–cancer research published between 2015 and 2024? ii. What are the philosophical implications of employing teleological language in microbiome–cancer studies? iii. To what extent do case studies of Fusobacterium nucleatum and colibactin-producing E. coli reflect adaptation, exaptation, or non-teleological mechanisms of persistence? iv. How can the language of microbiome–cancer research be revised to avoid teleological fallacies while retaining explanatory clarity? v. What broader effects do metaphors and narratives have on shaping the interpretation of tumor–microbe interactions? Literature Review Research over the past decade has firmly established that microbes inhabit tumors and can influence cancer progression. One of the most widely studied examples is Fusobacterium nucleatum in colorectal cancer. [3] showed that F. nucleatum is not only enriched in colorectal tumors but can persist in metastases, suggesting a potential role in tumor biology beyond colonization of primary sites. Their findings raised the question of whether this persistence reflected adaptation or ecological opportunism. This necessitated [2] to argue that F. nucleatum interacts with tumors through immune-modulating mechanisms, particularly its Fap2 protein binding to TIGIT receptors on immune cells. While the evidence demonstrates immune suppression, the authors describe this as bacteria “taking advantage” of the tumor. Such phrasing reflects the teleological framing this study interrogates, as immune evasion can be described mechanistically without implying intent. Nevertheless, the mechanistic evidence of microbial activity in tumor environments is robust and warrants careful interpretation. Since an important line of research concerns genotoxins such as colibactin, [8] demonstrated that colibactin-producing E. coli induces DNA damage and mutations in host cells, contributing to tumorigenesis. Building on this, [9] confirmed that colibactin leaves a mutational signature in human colorectal cancer genomes. These findings are significant because they link microbial metabolites directly to carcinogenic processes. Again, many accounts view colibactin as though it were “designed” to induce mutations, overlooking that the toxin’s original ecological functions likely relate to microbial competition rather than cancer promotion. Beyond these specific examples, large-scale microbiome studies have identified diverse bacterial taxa associated with tumors. [10] performed metaanalyses across multiple cohorts and identified reproducible microbial signatures in colorectal cancer, including enrichment of oral anaerobes such as F. nucleatum. Similarly, [11] showed that gut microbiome composition correlates with hepatocellular carcinoma risk. These studies underscore that microbes are consistently present in cancers, but they do not on their own justify adaptationist or teleological explanations. Empirical studies confirm the presence and activity of microbes in tumors which points at potential causal roles, and open avenues for therapy. However, the language in which these discoveries are presented often leans toward teleology, attributing purposiveness EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 73 to microbes. This creates the need to interrogate not only the data but also the narratives surrounding them. It is a truism that scientific writing often relies on metaphor to make complex processes intelligible. [12] argued that the essence of metaphor is understanding and experiencing one kind of thing in terms of another. In microbiology, metaphors such as “bacteria hijacking host pathways” or “tumors sheltering microbes” make intricate interactions easier to grasp. Hence, metaphors also shape conceptual frameworks, sometimes smuggling in unintended assumptions. [13] showed how metaphors in genetics such as the “genetic program” create impressions of agency and determinism. Similarly, [14] found that metaphors in genomics discourse can guide public and scientific expectations. When microbiologists describe microbes as “cancer drivers,” the metaphor imports agency into microbes, framing them as active agents pursuing goals. This can bias interpretation even when researchers recognize the metaphorical character of their language. The problem of teleology in biology has long been debated with [4] critiquing of the “Panglossian paradigm” warning against assuming that every trait has been optimized by natural selection. They argued instead for recognizing spandrels features that arise as by-products of other structures. This critique is directly relevant to tumor microbiomes, where microbial traits like Fap2 binding or colibactin production may persist in tumors without being adaptations for cancer promotion. [7] concept of the intentional stance further illuminates why teleological language persists. Scientists, like laypeople, often find it natural to explain complex systems by ascribing goals or strategies. In microbiome-cancer research, describing bacteria as “seeking to colonize tumors” exemplifies this slide from heuristic to explanatory framework. [15] distinguished between “teleonomy” which is the appearance of purpose generated by natural selection and teleology proper, which implies conscious intention. This distinction is critical for microbiomeoncology, where traits may appear purposive but are better explained as evolutionary outcomes without foresight. Misunderstanding this difference can lead researchers to anthropomorphize microbial behavior. This is as [16] extended these critiques by showing how narratives, not just data, structure explanations in evolutionary biology. The author cautioned that adaptationist storytelling often oversimplifies the contingencies of evolution. This insight applies directly to microbiome-cancer discourse, where adaptationist narratives risk obscuring exaptations and ecological complexity. The problem is particularly severe with the term "oncobiont." The word was originally used to discuss microbes that are involved in cancer, comparing them to oncogenes. This analogy overlooks significant differences where oncogenes are alterations that induce cancer, but microbes may merely exploit the environment in tumors. The use of "oncobiont" would then make it appear that microbes play more causal roles than they actually do, confounding both scientists and clinicians. [17] note along this line that, microbes and their hosts have complex relationships that cannot be reduced to simplistic narratives of function. Thus, the literature not only unveils empirical results but also a trend of metaphorical framing that smuggles teleology into microbiome–cancer research. This trend warrants a study that interrogates these narratives and offers more accurate, mechanistic alternatives. Theoretical Framework Gould and Lewontin: Critique of Adaptationism [4] essay “The Spandrels of San Marco and the Panglossian Paradigm” remains one of the most influential critiques of adaptationist reasoning in biology. They argued that evolutionary biologists too often interpret every trait as an adaptation, without considering alternative explanations such as constraints, byproducts, or chance events. In their words, “the failure of adaptationist programmes lies not in their methodology but in their unwillingness to consider alternatives to adaptive stories” [4]. Applied to microbiome–cancer research, this critique is highly relevant because when a bacterium like Fusobacterium nucleatum is described as having “evolved to promote cancer,” this is an adaptationist story which presumes that the presence of the bacterium in tumors reflects direct selection for cancer promotion. Dennett and the Intentional Stance Where [4] critique adaptationist reasoning, [7] addresses the lure to explain systems as if they have intentions. He distinguishes between three explanatory stances which are the physical stance (laws of physics), the design stance (functions of designed systems), and the intentional stance (treating systems as if they have beliefs and desires). Dennett writes that, “We adopt the intentional stance when we predict behavior by treating an entity as if it were a rational agent with goals” [7]. Dennett’s framework therefore helps clarify why teleological metaphors are appealing yet misleading. Teleology in Biology: Historical Cautions The lure to explain life in purposive terms is as old as biology itself. [15] distinguished between teleonomy an apparent purposefulness explained by natural selection and teleology, which implies intrinsic goals. For Mayr, teleonomy is legitimate, as when we say the heart is “for pumping blood,” but teleology is misleading if it attributes foresight or purpose to evolution. The problem in microbiome–cancer discourse is that scientists often slip from teleonomy to teleology without noticing. Saying that E. coli “produces EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 74 colibactin to promote cancer” is a teleological statement, whereas saying “colibactin damages DNA, thereby contributing to tumorigenesis under certain conditions” remains within teleonomy. Methodology This study employs a qualitative research design rooted in discourse analysis, complemented by case-based inquiry. Since the research question concerns how teleological metaphors shape microbiome–cancer research, the focus is not only on biological data but on the language through which findings are communicated. The methodology thus integrates two levels of analysis which are: (i) a systematic discourse analysis of published oncology papers, and (ii) in-depth case studies of Fusobacterium nucleatum and colibactin-producing Escherichia coli. This approach is consistent with established practices in science studies where the interrogation of language, metaphor, and framing is essential to understanding the epistemic commitments of a field [14]. Research Design The study adopts a mixed qualitative strategy, combining corpus-based discourse analysis with philosophically informed case studies. The corpus analysis allows the identification of broad patterns of teleological language in microbiome–cancer research, while the case studies provide depth by showing how such language interacts with specific biological mechanisms. This two-tiered design provides a balance between breadth and depth, ensuring that general claims about discourse are rooted in concrete examples. Corpus Selection The primary corpus consists of 20 peer-reviewed oncology and microbiome papers published between 2015 and 2024. These were identified through advanced searches on PubMed, Web of Science, and Google Scholar, using keywords such as “Fusobacterium nucleatum cancer,” “oncobiont,” “colibactin,” and “tumor microbiome.” Only papers meeting the following criteria were included: 1. Published in English in reputable peerreviewed journals. 2. Focused on the role of bacteria in tumorigenesis. 3. Contained explicit or implicit teleological framings (e.g., “microbe exploits tumor niche,” “evolved to promote cancer”). 4. The final set includes widely cited studies such as [3] on F. nucleatum persistence in colorectal tumors and [9] on colibactin mutational signatures in colorectal cancer. Studies were drawn from leading journals like Science, Nature Medicine, and Cell, ensuring both scientific significance and influence in the field. This corpus was supplemented with five additional review articles and conceptual essays, including [2] and [17], which explicitly discuss the concept of “oncobionts.” These were used not as part of the quantitative coding but as interpretive reference points. Discourse Analysis Procedure The discourse analysis was guided by critical discourse methodology as outlined by [18], which emphasizes the relationship between linguistic choices and broader social or epistemic structures. Each paper in the corpus was examined for the following features: 1. Lexical Indicators of Teleology: Phrases suggesting purposive or intentional behavior, such as “bacterium manipulates host immunity,” “microbe exploits,” or “oncobiont.” These were coded as instances of intentional stance language [7]. 2. Adaptationist Framing: Claims that microbial traits evolved specifically to promote tumorigenesis, without alternative explanations [4]. For example, describing colibactin as a “carcinogenic adaptation.” 3. Metaphoric Constructions: Instances where metaphorical language shapes microbial agency, e.g., bacteria described as “invaders,” “partners,” or “drivers.” Following [12], metaphors were treated as cognitive frames that influence interpretation. 4. Mechanistic vs. Teleological Contrast: Passages where authors contrast purposive accounts with biochemical or ecological explanations. For example, when a paper first states that F. nucleatum “hijacks immune signaling” but later specifies the molecular interaction of Fap2 with TIGIT receptors. Each paper was coded independently by the researcher using NVivo software. Coding reliability was maintained by re-checking samples at different stages of analysis. The goal was not statistical generalization but thematic saturation which bothers on identifying recurrent patterns of teleological framing across the field. Case Study Approach While the corpus analysis maps general discourse, the case studies provide a focused view of how teleological framing operates around specific microbial traits. Two organisms were chosen: 1. Fusobacterium nucleatum and the Fap2 protein: This oral bacterium has been repeatedly implicated in colorectal cancer. The Fap2 protein binds to TIGIT receptors on natural killer cells, dampening anti-tumor immunity [19]. Many papers describe this interaction as if the bacterium “strategically suppresses immunity to persist in tumors.” This case allows the study to examine how immune evasion is narrated as purposive behavior rather than as an ecological byproduct. 2. Escherichia coli and colibactin: Certain E. coli strains carry the pks genomic island, which encodes colibactin, a genotoxin that induces DNA double-strand breaks [8]. Colibactin has been associated with distinct mutational signatures in colorectal cancer [9]. EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 75 Researchers often identify colibactin as if it were an evolutionary weapon “designed” to cause DNA damage. This case study interrogates whether such descriptions reflect adaptation or exaptation. Case studies were analysed through close reading of the relevant literature, focusing on how experimental results were linguistically framed. Attention was given to whether papers explicitly identified alternative explanations (ecological competition, niche persistence) or whether teleological framings dominated. Analytical Framework The analysis was guided by the theoretical framework established earlier: i. From [4]: Caution against unwarranted adaptationism and distinction between adaptation and exaptation. ii. From [7]: Recognition that intentional stance language is heuristic but must not be mistaken for actual microbial purpose. iii. From [15] and others: Differentiation between legitimate teleonomy and misleading teleology. iv. From metaphor theory [12,13,20]: Acknowledgment that metaphors structure scientific reasoning and may bias interpretation. This triangulated approach ensured that discourse analysis was not only descriptive but conceptually informed. Each instance of teleological language was interpreted in light of these frameworks, asking whether it served as a convenient shorthand, an implicit adaptationist claim, or a misleading attribution of purpose. Ethical Considerations Although the study analyzes publicly available texts rather than human subjects, ethical considerations remain relevant. The goal is not to discredit individual researchers but to critically examine patterns of discourse within the field. For this reason, the analysis avoids personalizing critique and instead focuses on collective linguistic tendencies. Direct quotations are used only for interpretive purposes and are contextualized within broader arguments. Limitations of the Methodology The chosen methodology has certain limitations because while discourse analysis is interpretive, and coding frameworks enhance rigor, making complete objectivity unattainable. Furthermore, the reliance on English-language publications may exclude alternative ideas present in non-English literature. Finally, the case study approach prioritizes depth over breadth, focusing on two microbes at the expense of other relevant tumor-associated species like Bacteroides fragilis. Nonetheless, the selected design balances feasibility with conceptual depth, making it appropriate for the present study. Findings The discourse analysis of 20 primary microbiome– cancer studies published between 2015 and 2024 revealed a consistent reliance on teleological and purposive language. While these metaphors often served as accessible ways of describing microbial traits, they can misrepresent the evolutionary dynamics underlying tumor–microbe interactions. This section presents findings under three levels of analysis: (1) the overall frequency and distribution of teleological framings across the corpus, (2) the thematic categories of such framings, and (3) the case studies of Fusobacterium nucleatum and colibactin-producing Escherichia coli. Frequency of Teleological Language Out of the 20 papers analyzed, 13 (65 percent) employed teleological phrasing in their presentation of results. This percentage aligns closely with the study’s preliminary expectation that more than half of microbiome–cancer literature deploys purposive metaphors. For instance, in their study on F. nucleatum persistence in colorectal tumors, [3] described the bacterium as “thriving within tumors by exploiting the immune-suppressed niche” (p. 220). The term “exploiting” here attributes an intentional strategy to the organism, even though the experimental data merely demonstrated bacterial enrichment in tumor tissue. Similarly, [9] noted that colibactin “targets host DNA to induce carcinogenesis” (p. 1006). The use of the verb “targets” suggests purposive design rather than biochemical interaction. A count of recurring terms across the corpus revealed “exploit” (9 occurrences), “hijack” (7 occurrences), “drive” (11 occurrences), “oncobiont” (14 occurrences), and “manipulate” (6 occurrences) as the most frequent purposive markers. Only a minority of studies balanced such terms with explicit clarification that bacteria lack intent, as in [2] review where they cautioned that “microbes do not evolve for the purpose of causing cancer, though their traits may contribute to it” (p. 186). The frequency patterns suggest that teleological metaphors are not marginal but central to how microbiome–cancer interactions are narrated in highimpact journals. Thematic Patterns of Teleological Framing Closer analysis revealed three dominant thematic categories through which teleological assumptions entered the literature. (a) Microbes as Strategic Agents Several studies narrated microbes as if they actively strategize to ensure survival. [11], in reporting on intratumoral bacteria reducing chemotherapy efficacy, wrote that bacteria “protect themselves by degrading drugs” (p. 793). While mechanistically accurate, bacterial enzymes indeed degraded gemcitabine, the phrase “protect themselves” invokes an intentional EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 76 stance [7]. The same was seen in [19] who observed that Fap2 inhibits NK-cell activity, but then wrote that F. nucleatum “prevents immune clearance to persist within tumors” (p. 244). This language transforms a molecular binding interaction into a strategic act of survival. (b) Microbes as Cancer Drivers Another recurring theme was the description of microbes as “drivers” of carcinogenesis. [8] characterized colibactin-producing E. coli as “driving tumorigenesis through DNA damage” (p. 107). The driver metaphor, drawn from oncogenesis, carries purposive weight by casting bacteria in an active role of directing cancer’s trajectory. Similarly, [10] described microbial signatures in colorectal tumors as “enabling cancer development” (p. 1132). The word “enable” subtly implies cooperation between microbes and tumors, despite the absence of adaptive intent. (c) Microbes as Symbionts or Partners in Tumors The final theme involved portraying microbes as “partners” or “symbionts” of tumors. [2] popularized the term “oncobiont,” referring to microbes that persist in and contribute to tumor microenvironments. While useful as shorthand, “oncobiont” anthropomorphizes bacteria by implying that they have a role akin to tumor-promoting allies. [17] criticized this, warning that “to call a bacterium an oncobiont risks naturalizing a relationship that is neither intentional nor evolutionarily inevitable” (p. 395). This thematic analysis demonstrates that teleology is not only embedded in isolated phrases but woven into broader conceptual metaphors. Mechanistic vs. Teleological Contrasts Despite the prevalence of purposive metaphors, some papers attempted to clarify that their findings could be mechanistically understood. For example, [19] explicitly stated that Fap2 binds to TIGIT, a receptor on T and NK cells, thereby suppressing immune responses. However, in their discussion, they returned to the phrasing that F. nucleatum “protects itself from immune attack” (p. 246). This oscillation between mechanistic detail and purposive metaphor shows how researchers often slip between explanatory attempts. [9] also provided detailed evidence of colibactin mutational footprints, emphasizing molecular mechanisms, but their abstract framed colibactin as “targeting DNA,” implying purpose. The coexistence of precise biochemical explanation and teleological metaphor underscores how scientific writing often blends clarity with rhetorical force. Case Study One: Fusobacterium nucleatum and Fap2 The first case study centers on F. nucleatum, whose role in colorectal cancer is now widely acknowledged. The key finding is that the Fap2 protein interacts with TIGIT on NK and T cells to suppress immune activity [19]. Across multiple papers, this immune evasion is described as if it were an intentional act. [3] claimed that F. nucleatum “exploits tumor hypoxia to persist” (p. 220). This phrasing implies agency in recognizing and capitalizing on hypoxic conditions. Similarly, [21] reported a correlation between F. nucleatum and poorer patient prognosis, describing the bacterium as “a negative prognostic biomarker that adapts to tumor tissue” (p. 1167). Again, the idea of adaptation is introduced without evidence that Fap2 evolved specifically for tumor colonization. Case Study Two: Colibactin-Producing E. coli The second case study examines colibactin, a genotoxin encoded by the pks genomic island in certain E. coli strains. [8] demonstrated that colibactin induces DNA double-strand breaks in host cells. Later, [9] identified a distinct mutational signature associated with colibactin exposure. Across these studies, colibactin was often described as though it evolved with the purpose of inducing DNA damage. [9] wrote that colibactin “targets the genome to promote tumorigenesis” (p. 1006). Other evidence indicates colibactin may have evolved as a bacteriocin-like molecule to outcompete rival microbes [22]. Its carcinogenic impact could thus be an incidental byproduct rather than an adaptation. Cross-Corpus Insights Putting all of these together, the corpus analysis and case studies reveal three broader insights: 1. Teleological metaphors are normalized. They occur not as occasional rhetorical flourishes but as standard descriptors in oncology microbiome research. 2. Mechanistic explanations coexist with purposive framings. Authors often alternate between biochemical specificity and anthropomorphic shorthand, without acknowledging the philosophical slippage. 3. Alternative evolutionary interpretations are underemphasized. Very few studies explicitly test whether microbial traits in tumors represent true adaptations or incidental exaptations. This gap sustains the adaptationist narrative critiqued by Gould and Lewontin. Discussion The findings of this study demonstrate that teleological framings are deeply embedded in microbiome–cancer research. The fact that 65% of the corpus employed purposive or goal-laden metaphors reveals not an incidental feature of scientific writing but a broader discursive pattern. In reflecting on these results, three themes emerge: first, the epistemic tension between metaphorical language and mechanistic explanation; second, the risk of adaptationist overreach in evolutionary accounts of microbial traits; and third, the implications for experimental design and knowledge production in oncology. EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 77 One of the clearest results was the prevalence of verbs such as “exploit,” “invade,” and “manipulate” in descriptions of Fusobacterium nucleatum. [19] described how Fap2 “suppresses the activity of natural killer cells” through binding to TIGIT (p. 244). In mechanistic terms, this is an interaction between a bacterial outer membrane protein and a host receptor. The surrounding text repeatedly framed the process as the bacterium “evading immune destruction.” [12] remind us that, metaphors are “not just in the words we use but in the very way we conceptualize” phenomena (p. 5). When immunosuppression is described as a bacterial “strategy,” it suggests agency, which may subtly shift how scientists and readers interpret causality. The case studies of Fap2 and colibactin illustrate the deeper issue which is a readiness to interpret microbial traits in tumors as adaptations for persistence, rather than considering alternative explanations. [4] warned against the “Panglossian paradigm” in evolutionary biology, where every trait is presumed to be an adaptation. [9] showed that colibactin leaves a distinct mutational signature in colorectal epithelial cells. In framing colibactin as a “genotoxin evolved to damage host DNA, they suggested adaptation without evaluating whether DNA damage is a primary or incidental consequence of bacterial competition. [8] provided an important counterpoint which while they demonstrated that colibactin-producing E. coli exacerbates tumorigenesis in mice, they noted that colibactin might serve functions unrelated to cancer, such as mediating competition with other microbes. This distinction between adaptation and exaptation, between traits selected for one purpose and traits coopted for another is often blurred in cancer microbiome literature. The discourse analysis revealed that most papers defaulted to adaptationist framings, rarely acknowledging exaptive alternatives. The positioning of Fap2 provides a parallel; its binding to TIGIT suppresses host immunity, which undoubtedly benefits the bacterium in tumor tissue. Is this interaction then an adaptation for tumor persistence, or is it a broader immune-modulating trait that becomes advantageous in tumors? [19] did not address this distinction, and subsequent citations have tended to assume adaptation. This reflects what [23] termed the “fallacy of misplaced concreteness,” where evolutionary narratives fill explanatory gaps. One practical implication of teleological framings is the design of experiments. The findings showed that only a minority of papers contrasted purposive language with explicit mechanistic alternatives. When Fap2 was described as “suppressing host immunity,” few studies tested whether this suppression conferred selective advantage outside the tumor context. Similarly, colibactin’s mutagenic effects were extensively characterized, but few experiments probed whether colibactin production is primarily for microbial competition. This gap matters because, as [15] observed, teleological explanations can obscure the actual causal pathways in biology. If researchers assume adaptation, they may not design experiments to detect exaptation. For example, testing whether colibactin provides a competitive edge against other gut microbes would illuminate whether its carcinogenicity is incidental. The absence of such designs suggests that metaphorical leanings are not just linguistic but epistemic as they shape the very questions scientists ask. The results underscore the importance of distinguishing between teleology and teleonomy. [15] used teleonomy to describe goal-directedness explained by natural selection, as in “the function of the heart is to pump blood” (p. 45). In the case of tumor microbiomes, to say that Fap2 “functions” to suppress immunity is legitimate if understood as teleonomy. But when the same description is held as intentional strategy, it crosses into teleology. The findings suggest that most microbiome–cancer papers blur this boundary. [7] framework is helpful here because the intentional stance is useful, but when adopted uncritically it leads to what he calls “the fallacy of reification” (p. 29). The cancer microbiome literature frequently falls into this trap, especially when describing bacteria as if they had evolutionary “plans.” Microbiome research has rapidly gained attention in oncology, with potential applications in diagnostics, therapeutics, and prognostics [3]. However, if the field continues to frame microbes as “oncobionts” with purposive strategies, it risks building clinical narratives on shaky conceptual ground. Patients and policymakers may misinterpret microbial traits as malevolent actors, rather than as contingent ecological phenomena. Moreover, teleological framings may inadvertently reinforce deterministic views of cancer causality. If microbes are said to have “evolved to promote cancer,” then their role appears fixed and inescapable. In contrast, a mechanistic framing that emphasizes ecological pressures such as nutrient competition or immune modulation presents a more flexible picture, open to intervention. Thus, the way discourse is structured has direct consequences for how microbiome–cancer interactions are understood, studied, and treated. Conclusion The analysis of microbiome–cancer literature from 2015 to 2024 demonstrates that teleological understanding are not isolated rhetorical slips but pervasive elements of scientific discourse. Words like “exploit,” “weaponize,” “travel,” and categorical labels such as “oncobiont” reveal how cancer microbiology is narrated in ways that suggest intentionality and design, EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 78 even when the evidence indicates biochemical contingency or ecological opportunism. The guiding concern of this research has been whether microbiologists risk conflating evolutionary selected effects with goal-directed behavior. When microbiome research frames bacterial proteins as “immune evasive strategies” or microbes as “drivers of cancer evolution,” it imports an adaptationist logic that may not reflect reality. The findings of this study also contribute to broader debates in philosophy of science. They reinforce [4] critique of the adaptationist program. The teleological fallacies in microbiome–cancer discourse exemplify how adaptationist explanations can dominate even in the absence of strong evidence. Tracing how purposive metaphors proliferate across scientific texts, the study shows how language functions as an epistemic model. The microbiome–cancer case demonstrates how easily this stance infiltrates biological discourse where scientists slip into describing microbes as if they were intentional strategists, even while knowing that bacteria lack conscious goals. The philosophical lesson is that vigilance is required whenever the intentional stance tempts us to anthropomorphize natural systems. Recommendations The first recommendation is that there should be a deliberate shift in the way microbiome–cancer interactions are described in scholarly publications. Purpose-driven metaphors such as “oncobionts,” “microbes exploiting tumors,” or “bacteria weaponizing host systems” should be replaced with mechanistic formulations. To implement linguistic reform, journals in oncology and microbiology should develop editorial policies that discourage teleological framings. Just as many journals already provide style guidelines for statistical reporting, they could introduce guidance on avoiding purposive metaphors. This would not mean banning metaphors altogether but ensuring that their use is clearly framed as heuristic rather than literal. An editorial note could, for example, remind authors that while terms like “immune evasion” are common, they should be accompanied by precise mechanistic descriptions to prevent misinterpretation. Another recommendation is that graduate and postgraduate microbiologists receive at least some training in the philosophy of science, particularly in evolutionary theory and scientific language. Introducing these insights into microbiology training would give researchers the tools to reflect critically on their own language choices. For example, short workshops on concepts like adaptation, exaptation, and the intentional stance could help scientists distinguish between legitimate evolutionary explanation and teleological fallacy. A methodological reform is t necessary as experiments should explicitly distinguish between adaptation and exaptation. For instance, studies could compare the presence of tumor-associated bacterial traits in nontumor ecological contexts. If a trait is found to confer fitness benefits in soil, gut, or mucosal competition, this would suggest exaptation rather than tumor-specific adaptation. Another methodological recommendation is the use of mathematical and computational evolutionary models to simulate how microbial traits might arise under different selective pressures. By modeling the fitness costs and benefits of traits like colibactin production across ecological niches, researchers can evaluate whether such traits plausibly evolved for tumor persistence or whether they are coincidental. Such models could clarify whether cancer contexts are selective environments or incidental byproducts of broader ecological competition. Institutions funding microbiome–cancer research should encourage reflexive practices that question not only data but also the conceptual frameworks guiding research. This could involve requiring grant proposals to justify evolutionary assumptions about microbial traits. Such reflexivity would push researchers to clarify whether they are proposing adaptationist, exaptive, or purely mechanistic hypotheses. Philosophers of science, microbiologists, and oncologists should convene at interdisciplinary conferences specifically focused on conceptual issues in cancer microbiome research. Such spaces would provide opportunities to surface and critique the implicit teleological assumptions that might otherwise go unchallenged. Reforming language, designing experiments that test alternatives, and communicating findings with precision, the microbiome–cancer field can avoid the pitfalls of teleological fallacy. Doing so will strengthen not only the epistemic rigor of the science but also its clinical translation and public understanding. If these recommendations are adopted, the field will be better positioned to describe microbial traits as what they are, biochemical and ecological processes rather than as strategies or intentions. This shift will open new avenues for understanding tumor microbiomes as complex, contingent environments rather than as battlefields populated by microbes with purposes. References [1] Kostic AD, Chun E, Robertson L, et al. Fusobacterium nucleatum potentiates intestinal tumorigenesis and modulates the tumor-immune microenvironment. Cell Host Microbe. 2013;14(2):207– 215. doi:10.1016/j.chom.2013.07.007 [2] Brennan CA, Garrett WS. Fusobacterium nucleatum — symbiont, opportunist and oncobacterium. Nat Rev Microbiol. 2019;17(3):156– 166. doi:10.1038/s41579-018-0129-6 EUR J MED HEALTH RES Volume 3 | Number 6 | 2025 79 [3] Bullman S, Pedamallu CS, Sicinska E, Clancy TE, Zhang X, Cai D, et al. Analysis of Fusobacterium persistence and antibiotic response in colorectal cancer. Science. 2017;358(6369):1443–1448. doi:10.1126/science.aal5240 [4] Gould SJ, Lewontin RC. The spandrels of San Marco and the Panglossian paradigm: a critique of the adaptationist programme. Proc R Soc Lond B Biol Sci. 1979;205(1161):581–598. doi:10.1098/rspb.1979.0086 [5] Yu T, Guo F, Yu Y, et al. Fusobacterium nucleatum promotes chemoresistance to colorectal cancer by modulating autophagy. Cell. 2017;170(3):548–563.e16. doi:10.1016/j.cell.2017.07.008 [6] Chen Y, Chen Y, Zhang J, et al. Fusobacterium nucleatum promotes metastasis in colorectal cancer by activating autophagy signaling via the upregulation of CARD3 expression. Theranostics. 2020;10(14):323–336. doi:10.7150/thno.42036 [7] Dennett D. The intentional stance. Cambridge (MA): MIT Press; 1987. [8] Arthur JC, Gharaibeh RZ, Uronis JM, et al. Microbial genomic analysis reveals the essential role of colibactin-producing E. coli in colon tumorigenesis. Science. 2012;338(6103):120–123. doi:10.1126/science.1224820 [9] Pleguezuelos-Manzano C, Puschhof J, Rosendahl Huber A, et al. Mutational signature in colorectal cancer caused by genotoxic pks+ E. coli. Nature. 2020;580(7802):269–273. doi:10.1038/s41586-020-2080-8 [10] Wirbel J, Pyl PT, Kartal E, Zych K, Kashani A, Milanese A, et al. Meta-analysis of fecal metagenomes reveals global microbial signatures that are specific for colorectal cancer. Nat Med. 2019;25(4):679–689. doi:10.1038/s41591-019-0406-6 [11] Gao R, Kong C, Huang L, Li H, Qu X, Liu Z, et al. Mucosa-associated microbiota signature in hepatocellular carcinoma. Cell. 2017;169(5):631– 643.e6. doi:10.1016/j.cell.2017.04.02 [12] Lakoff G, Johnson M. Metaphors we live by. 2nd ed. Chicago (IL): University of Chicago Press; 2003. [13] Keller EF. The century of the gene. Cambridge (MA): Harvard University Press; 2000. [14] Nerlich B, Hellsten I. Beyond the human genome: Microbiology, metaphors, and society. Sci Commun. 2009;31(2):159–181. doi:10.1177/1075547009340345 [15] Mayr E. The growth of biological thought: Diversity, evolution, and inheritance. Cambridge (MA): Harvard University Press; 1982. [16] Beatty J. What are narratives good for? Stud Hist Philos Sci Part C. 2016;58:33–40. doi:10.1016/j.shpsc.2016.02.011 [17] Gilbert SF, Tauber AI. Rethinking individuality: The dialectics of the holobiont. Biol Philos. 2016;31(6):839–853. doi:10.1007/s10539-016-9541-3 [18] Fairclough N. Critical discourse analysis: The critical study of language. 2nd ed. London: Routledge; 2013. [19] Gur C, Ibrahim Y, Isaacson B, Yamin R, Abed J, Gamliel M, et al. Binding of the Fap2 protein of Fusobacterium nucleatum to human inhibitory receptor TIGIT protects tumors from immune cell attack. Immunity. 2015;42(2):344–355. doi:10.1016/j.immuni.2015.01.010 [20] Keller EF. Making sense of life: Explaining biological development with models, metaphors, and machines. Cambridge (MA): Harvard University Press; 2002. [21] Mima K, Nishihara R, Qian ZR, et al. Fusobacterium nucleatum in colorectal carcinoma tissue and patient prognosis. JAMA Oncol. 2016;2(5):653–661. doi:10.1001/jamaoncol.2015.1377 [22] Taieb F, Nougayrède JP, De Rycke J, Oswald E. Escherichia coli induces DNA double-strand breaks in eukaryotic cells. Science. 2006;313(5788):848–851. doi:10.1126/science.1127059 [23] Lewontin RC. The units of selection. Annu Rev Ecol Syst. 1970;1:1–18. doi:10.1146/annurev.es.01.110170.000245