Pathway alterations and expression of viral oncoproteins are observed in multiple viruses that are directly carcinogenic (Burd, 2003; Young and Murray, 2003; Levrero and Zucman-Rossi, 2016; Kuss-Duerkop et al., 2018; Virzì et al., 2018; Afzal et al., 2022; Burton and Gewurz, 2022; Marongiu et al., 2022). While other unidentified processes are likely, understanding known mechanisms may be helpful for discovering previously unrecognized microbial facilitators of oncogenesis. For example, HPV oncogenesis is mediated by viral oncogenes, such as E6 (Oyervides-Muñoz et al., 2018); E6 recruits intracellular E3 ubiquitin ligase (also named E6AP), which targets p53 for proteasomal degradation (Li et al., 2019). P53 is a tumor suppressor gene which plays critical roles in pathways to prevent DNA damage, marking cells for apoptosis or delaying cell cycle progression in the presence of DNA damage. HPV infection, therefore, inactivates p53 and leads to unregulated cell division, cell growth, cell survival, and DNA damage promotion. E7, another HPV oncogene, also interacts with the p53 pathway through retinoblastoma tumor suppressor protein (pRb), resulting in unregulated cell cycle progression. These oncogenes/proteins are also known to drive other oncogenic pathways, including telomerase regulation. The cancer cells depend on constitutive expression of these proteins, making them prime targets for therapeutic vaccines and biomarker detection (Burd, 2003; Ghebre et al., 2017; Marima et al., 2021).

Oncogene expression tends to interfere with important cell regulation or immortalization pathways. These can also be mediated by host-microbe interactions through proteins which mimic host proteins and interact with key signal pathways causing the overexpression of oncogenes and suppression of tumor suppressor proteins (Guven-Maiorov et al., 2019; Tempera and Lieberman, 2021). For example, EBV produces peptides (BHRF1 and BALF-1) and microRNAs which inhibit Bcl-2 and prevent apoptosis (Young and Murray, 2003; Brady et al., 2007; Carbone and Gloghini, 2018; Burton and Gewurz, 2022). Similarly, KSHV (HHV8) (for which seroprevalence ranges between 30-90% in SSA) inhibits ORF16 which is required for apoptosis (Wan et al., 2024). Viral proteins may also act synergistically with existing cancers by inducing the Warburg effect, i.e. where cells switch to glycolysis fermentation to generate energy instead of oxidative phosphorylation, allowing for the usage of alternative metabolites and the switching to anaerobic respiration (Liberti and Locasale, 2016). For example, HPV, KSHV and Merkel cell polyomavirus (MCPyV) affect glycolysis and induce increased glucose utilization in cancer cells, resulting in stress in healthy surrounding cells. EBV infection can modify lipid and cholesterol metabolism to induce anaerobic metabolism (Burton and Gewurz, 2022). Further mechanisms are reviewed elsewhere (Tempera and Lieberman, 2021)

Direct integration of viruses can cause differential expression of key proteins affecting the cell cycle, DNA repair mechanisms, and apoptosis, each important for oncogenesis. For example, HTLV-1 integrates into CD4+ T-cells causing chromosomal instability, mediated by its oncoprotein RNF8 (Zhi et al., 2020). Integration of HBV can both cause genome instability, and also leads to chronic inflammation due to sustained presence of HBV and its antigens (Borgia et al., 2021; Jin et al., 2023). For HPV and MCPyV, integration into epithelial cells (Yang and You, 2022; Karimzadeh et al., 2023), allows for cell immortalization of these cells due to the constant presence of the oncogenic driver (Elkhalifa et al., 2023).

Pathophysiology of some infectious diseases overlap with oncogenic mechanisms in ways which could promote cancer initiation and growth. Infection with HBV and HCV, for instance, are associated with chronic infection which may lead to chronic inflammation, thereby causing type 2 carcinogenic effects (Borgia et al., 2021; Yang et al., 2021). These viruses are synergistically carcinogenic with other factors such as aflatoxins and liver cirrhosis due to alcohol usage (Borgia et al., 2021; Jin et al., 2023). The links between HBV and HCV and hepatocellular carcinoma are well established, but chronic HBV and HCV infections also may be associated with other cancers, especially gastric adenocarcinomas. For instance, a metanalysis of ten studies found that patients with HBV infection had a higher risk (hazard risk =1.26;95% CI=1.08-1.47)) for gastric cancer when compared with controls (without HBV infection) (Chen et al., 2019; Yang et al., 2021; Yu et al., 2023b). Hypothesized mechanisms awaiting confirmation include chronic inflammation resulting in carcinogenesis, direct viral integration into gastric epithelial cells, expression of viral proteins (like HBx which interferes with cell signaling pathways, gene expression and apoptosis) and immune response modulation (Yang et al., 2021). Likewise, EBV may be associated with a subset of gastric adenocarcinoma; chronic inflammation is one hypothesized mechanism (Salnikov et al., 2024).

HIV causes cancer both by its integration into oncogenes, and, also through an indirect mechanism of immune suppression, which allows existing cancers to progress or oncoviruses to establish infection. HIV/AIDS progression is accompanied by AIDS-defining cancers, such as Non-Hodgkin lymphoma (NHL), Kaposi sarcoma, and cervical cancer) (Beral et al., 1990; Martínez-Maza and Breen, 2002; De Martel et al., 2015; Bohlius et al., 2016; Yarchoan and Uldrick, 2018), with cancer risk in people living with HIV (PLWHIV) significantly elevated, ranging between 25-40%, despite widely accessed antiretroviral therapy (De Martel et al., 2015). Lung cancer, Hodgkin lymphoma, hepatocellular cancer, and anal cancers are associated with HIV infection despite effective ART treatment, suggesting a possible direct oncogenic effect of HIV beyond immune suppression (Khandwala et al., 2021; Navarro et al., 2021; Haas et al., 2022; McGee-Avila et al., 2024). For Burkitt lymphoma, HIV may directly drive oncogenesis through its immunomodulation and engagement of C-C motif chemokine receptor 5 (CCR5) (Samson et al., 1996; Silverberg et al., 2007; Bohlius et al., 2009; Martorelli et al., 2015; Shindiapina et al., 2020). HIV appears to potentiate the oncogenic effect of some viruses to increase risk for cancer. For example, the attributable fraction of EBV-associated Hodgkin lymphoma in the general population is 20-50%, but in HIV-infected patients, 75%-100% of Hodgkin lymphoma is attributable to EBV, possibly due to aberrant CD4 T-cell responses to EBV infection (Carbone and Gloghini, 2018; Shindiapina et al., 2020; Navarro et al., 2021).

Other types of immunosuppression are similarly linked to cancer progression (Haas, 2019; Herrera et al., 2019); for example, advanced age leading to immune senescence and immune suppression drugs are both highly associated with cancer and cancer progression (Haas, 2019; Fu et al., 2023). In addition, tumors facilitate their growth and metastasis by actively suppressing the immune system in their direct microenvironment (Arner and Rathmell, 2023; Tie et al., 2022).

A retrovirus known as mouse mammary tumor virus (MMTV) has been shown to cause mammary tumors in rodents; multiple studies have shown that human mammary tumor virus (HMTV), which is 90-95% homologous to MMTV is present more often in human breast cancer tissue when compared with healthy breast tissue (Lawson and Glenn, 2022; Parisi et al., 2022). Using our proposed criteria for causation, HMTV/MMTV meets criterion 4 (with oncogenesis demonstrated in an animal model), and partially meets criterion 2 (Consistent detection of the microbe - virus, bacterium, fungus, or parasite- in cancer tissues compared to healthy controls) with consistent findings in some laboratories, but not across all geographies. When detected, it is not clear whether the virus is part of the causation pathway or whether breast cancer tissue is simply conducive to colonization with the virus (Lawson and Glenn, 2022). Infections with viruses like MMTV may induce signaling alterations which causes negative effects only under certain conditions. Microbiome characteristics might determine such a conducive environment. Recent studies have correlated gut microbiome features with breast cancer stages and progression, although a causative link has also not been established (Lee et al., 2023; Sohail and Burns, 2023; Viswanathan et al., 2023).

CMV is highly prevalent in SSA with a pooled prevalence of 81.9% (55–97%) (Bates and Brantsaeter, 2016). The potential oncogenicity of CMV has long been a focus for study. CMV proteins and nucleic acids have been found (meeting Criteria 5d and 5e [ Tables 2 , 3 ] in some, but not all, studies) within glioblastoma multiforme tumors suggesting that the virus may play a role in facilitating oncogenic transformation or progression of glioblastomas (Scheurer et al., 2008; Yang et al., 2022a; Mercado et al., 2024). Human astrocytes infected with CMV have formed glioblastoma-like cancers in mice models (Guyon et al., 2024) (Criterion 2) and infection with CMV leads to poorer prognosis for glioblastoma (Criterion 6); targeting CMV infected cells has shown potential for glioblastoma therapeutics (Yang et al., 2022a; Mercado et al., 2024). While the association has not been conclusively confirmed, potential mechanisms include chronic inflammation, immune evasion or modulation, and viral gene expression. A variety of characteristics of CMV infection might contribute to cellular transformation to cancer: induction of expression of pro-angiogenic factors, interaction with oncogenic signaling pathways, such as the phosphatidylinositol 3-kinase (PI3K)/Akt pathway, which is often dysregulated in cancers, and epigenetic changes like DNA methylation and histone changes, which can contribute to tumorigenesis (Chan et al., 2010). Among the cancers contributing to substantial burden in Africa, some studies have suggested that CMV may have a role in breast, prostate and colorectal cancers, among others, presumably with similar mechanisms that have been hypothesized for a putative role with glioblastomas (Yu et al., 2023a).

Polyomaviruses have been linked to colon cancers, with JCPyV and BK polyoma viruses showing the strongest causal links (Gorish et al., 2019; Shoraka et al., 2020). Although the majority of humans are latent carriers for these viruses and no comprehensive SSA data was found, immunosuppression either through disease or natural immune senescence can lead to reactivation. Higher levels of JCPyV and BK are found in colon cancer tissues (Criterion 2), as well as in solid B cell leukemia, when compared with non-cancerous tissue, and have been identified in many other cancers (Gorish et al., 2019; Loutfy et al., 2017). Injection of T antigen derived from JCPyV in mice was shown to cause a variety of cancers including neural and breast and hepatocellular cancers (Criterion 4) (Zheng et al., 2022). Similarly, Merkel Cell polyoma virus shows high prevalence in skin cancer tissues (Klufah et al., 2021).

A variety of studies have suggested that the oral anaerobic bacterium F. nucleatum may have a role in initiating oncogenic transformation in colonic and rectal cells (Kostic et al., 2013). While more study is needed to confirm this potential association, especially within African settings, there is evidence that colorectal cancer cells when colonized with specific subclades of F. nucleatum, may increase the potential for local tumor spread and metastasis (Kostic et al., 2013; Rubinstein et al., 2013; Bullman et al., 2017; Ou et al., 2022; Wang and Fang, 2022; Zepeda-Rivera et al., 2024). We further discuss this possibility within the section on microbiomes.

C. acnes (a skin commensal, implicated in superficial skin infections, especially acne vulgaris) has been shown to colonize the prostate, resulting in chronic inflammation, which appears to be a pivotal factor for prostate cancer (Davidsson et al., 2021; Goldstein and Mascitelli, 2024). Studies have demonstrated the presence of C. acne in prostate cancer specimens in much higher proportions than in non-cancerous prostate tissue (Criterion 2) (Kakegawa et al., 2017). Furthermore, cohort epidemiologic studies have shown that severe acne during adolescence is associated with a higher likelihood of prostate cancer later in life (Criterion 1) (Zhang et al., 2018). To demonstrate causation, future studies must evaluate and reproducibly confirm other criteria for causation, as well as validate the epidemiologic evidence. Investigations carried out thus far, have yielded inconsistent results (Drott et al., 2010; Sayanjali et al., 2016). Furthermore, existing data are from European studies with a lack of data from Africa, where prostate cancer appears to occur at an earlier age and can be more aggressive (Davidsson et al., 2016; Janivara et al., 2024). That gap needs to be filled to drive further research that could lead to preventive approaches, assuming a triggering role was established.

S Typhi can colonize the gall bladder following symptomatic typhoid fever or asymptomatic systemic infection, resulting in chronic carriage in 1-4% of patients acutely infected. In SSA there are 1.2 million acute typhoid fever cases annually; however, it is unclear how many of these infections lead to chronic carriage (Kim et al., 2024). Moreover, there are distinct genotypes in SSA that have a higher propensity for invasive disease and antibiotic resistance — both factors could influence potential for chronic infection (Kingsley et al., 2009). Gall bladder colonization results in chronic inflammation directly and by stimulated the formation of gallstones. A meta-analysis of 17 studies suggested that chronic S Typhi infection of the gall bladder is associated with gallbladder cancer (Criterion 1) (Nagaraja and Eslick, 2014; Upadhayay et al., 2022). S Typhi has a high prevalence in sub-Saharan Africa with a substantial proportion of infections being undiagnosed or untreated. This is coupled with the increasing incidence of multidrug resistance S Typhi, and low vaccination levels (Kariuki and Onsare, 2024; Khan et al., 2022).

While SARS-CoV-2 has not yet been linked to specific cancers (Jahankhani et al., 2023), there has not been sufficient follow-up period to observe an effect. However, this virus has distinct infection-associated patterns which may contribute to being an oncogenic virus. For instance, SARS-CoV-2 causes RAAS (renin-angiotensin-aldosterone system) pathway dysregulation, induces degradation of the tumor repressor retinoblastoma protein, via nsp15 and P53 via nsp3, affects cell cycle through among others nsp7, interferes with DNA methylation through NSP8, and generates reactive oxygen species (ROS), all common pathways involved in oncogenesis, making a link to cancer plausible (Criterion 9) (Jahankhani et al., 2023). Evidence of prolonged, persistence of replicating SARS CoV2 in tissues (Yang et al., 2024) raises a potential for chronic inflammation which may also increase a risk of cancer formation. Longitudinal cohorts, such as the Rotterdam study, maintained over time may provide insight into the role of infectious triggers including SARS-CoV2 in oncogenesis (Ikram et al., 2020; Sijtsma et al., 2022).

Candida has been observed in colorectal cancer tissue samples and is associated with decreased survival and metastatic disease in colon cancer. Similarly, Blastomyces is has been detected in lung cancer tumor tissues (Dohlman et al., 2022). However, for these and other examples of fungal colonization, association but no direct causal relationships have been established. Fungal colonization has been suggested to drive carcinogenesis through immune recruitment of TH2 cells in pancreatic and esophageal cancer. Pathogenic infections of Candida species correlate with a higher oral cancer incidence (Chung et al., 2017; Di Cosola et al., 2021). These associations may partially be a consequence of the lower levels of immunity in these individuals (increasing risk for colonization) or could indicate synergistic relationships that promote both cancer and fungal colonization.

Fungi play a demonstrable role in hepatocellular cancer, however in a more indirect way, where they (Aspergillus flavus and Aspergillus parasiticus mainly) infect food sources such as maize and grains, imparting high levels of hepatotoxic aflatoxins which in turn directly contribute to oncogenesis (Cui et al., 2015; Jin et al., 2023; Yu et al., 2023b). This is of particular importance in some regions of SSA (especially west Africa and parts of east Africa) where food storage conditions combined with high heat and humidity contribute to a high aflatoxin burden in staple foods (Falade et al., 2022). Other mycotoxins have been studied such as Ochratoxin A, produced by aspergillus or T-2 and Zearalenone, both fusarium toxins; these link to nephropathies and potentially neurological disorders such as Parkinsons and dementia (Khan et al., 2024). However, these toxins were linked to various cancers (esophageal, kidney, colon, urinary tract, gastrointestinal, uterine, breast) in animal and in in vitro models (Claeys et al., 2020; Ekwomadu et al., 2021, Ekwomadu et al., 2022; Khan et al., 2024). For example, fumonisins have been linked to esophageal cancer with recent studies suggesting that it affects PI3K/Akt pathway in human esophageal cells (Yu et al., 2021). Nonetheless, clear epidemiological evidence for these links has not been established (Claeys et al., 2020; Ekwomadu et al., 2022).

In healthy individuals the gut microbiota, consisting of bacteria, bacteriophages, viruses, archaea, and fungi, play a role in immune regulation through presentation of short chain fatty amino acids (SCFAs) (Mann et al., 2024). Firmicutes and Bifidobacteriaceae species present these SCFA’s which are taken up by the intestinal cells and regulate the pro-inflammatory cytokines, TNFa IL12 and IL6. Moreover, the microbiome also trains the immune system and inhibits the growth of pathogenic biota (such as Enterobacteriaceae) and the development of pathobionts. Pathobionts are microbes that under normal circumstances do not cause disease; however, in the context of cancer or microbiome dysregulation, they become pathogenic (Jochum and Stecher, 2020).

One such pathobiont is F. nucleatum, an oral commensal anaerobic bacterium, which as mentioned above, may play an important role in facilitating colorectal cancer incidence and metastasis. F. nucleatum colonizes colorectal cancer cells through Fap2, a galactose adhesion hemagglutinin. It produces virulence factors such as FadA, which provides a scaffold for colonization with other bacteria, contributing to dysbiosis, potentially inducing oncogenesis in host cells. FadA and other virulence factors (e.g. AvrA in Salmonella) bind to the E cadherin receptor, inducing the Wnt signaling pathway, one of the major pathways implicated in colorectal cancer oncogenesis and progression (Kostic et al., 2013; Rubinstein et al., 2013; Bullman et al., 2017; Silva-García et al., 2019). It may enhance colorectal cancer proliferation by upregulation of the wnt signaling pathways and metastasis by inducing the expression of CXCL1 and IL-8 which promotes migration and upregulating CCL20 (Ou et al., 2022). F nucleatum also induces immune evasion through binding of FapA to immune cells. It is similarly potentially implicated in oral cancers where it enhances proliferation and inhibits cell cycle control mechanisms through p27 (Chen et al., 2022; Wang and Fang, 2022). Lastly it was shown induce metastases by modulating mitogen-activated protein kinase p38, which is involved in mesenchymal transition (Lin et al., 2016).

While there has been a paucity of data characterizing microbiomes in SSA, recent studies have revealed unique taxa and diversity in both South African and Tanzanian samples (Nobels et al., 2025). Click or tap here to enter text. Some investigations have examined the role of the microbiome in cancer development in SSA, linking cervical microbiome characteristics and cervical cancer, as well as suggesting that changes to the gut microbiome after urbanization may correlate with development of colon cancer (Klein et al., 2020; Come et al., 2021; Yang et al., 2022b; Ramaboli et al., 2024), However, most microbiome research in SSA relies on time-intensive culture-based experiments, compared to the more rapid sequence-based technology applied in higher income settings (Paulo et al., 2023; Ayeni et al., 2024). Substantial knowledge gaps remain regarding links between microbiome characteristics and cancers in SSA, opening the door for prioritizing support for pivotal research on the topic

While some bacteria may play a causative role in cancer, contributing to immune evasion and cancer progression, dysregulation is often a consequence of opportunistic infections, indicating inconsequential presence of bacteria within the microbiome, rather than causation. There is a host of studies where non-commensal or dysbiotic bacteria such as Salmonella or Helicobacter bacteria are found in tumor tissue (Zhao et al., 2022; Zhu et al., 2022; Schorr et al., 2023). An analysis determined that for most solid tumors, 10⁵ to 10⁶ bacteria are present per palpable 1-cm³ tumor, which represents 34 bacterial cells per 5000 cancer cells. The levels of these bacteria are therefore generally low, which complicates analyses. Even when presence is established, presence is not sufficient to indicate causative or synergistic relationships. Indeed, studying the interaction between microbiota and cancer needs careful consideration as demonstrated by a recent re-analysis of links between pancreatic cancer and the microbiome, initially suggesting, then refuting an oncogenic role for microbiota (Guo et al., 2021; Fletcher et al., 2023; Eckhoff et al., 2024; Pourali et al., 2024). The authors argue that the low-biomass of human tissue specimens increases the risk for errors, including distinguishing between low-biomass microbial communities and contamination introduced during sample collection, and errors made during processing, and sequencing. Therefore, PCR confirmed presence and characterizations of the microbiome, cannot not indicate that these organisms were viable in this tissue, nor can determine whether they were causative or bystanders. To form a better understanding of this complex interplay, standardized methods are needed for generating and analyzing microbiome sequencing data to enhance the reproducibility of results across different studies (Aykut et al., 2019; Fletcher et al., 2023).

HIV has also been associated with significant changes in the microbiome. Upon infection, there is rapid spread throughout the lymph system of the gut through mechanisms of cell-to-cell transmission. HIV uses virological synapses to spread throughout the entire CD4 T cell network causing massive cell death and local immune dysregulation. These also result in permanent disruptions in the epithelium of the gut. Even when anti-retroviral therapy (ART) is given early during the course of HIV infection, the gut based immune system is not fully restored and the damage to the epithelial damage causes long lasting dysbiosis and microbial translocation (Zicari et al., 2019; Govindaraj et al., 2023). Microbiota associated with HIV infection are similar to those associated with other inflammatory diseases, such as inflammatory bowel disease. The dysbiosis in HIV may contribute to the continued systemic inflammation (with corollary impacts on cancer risks) observed in HIV, which persists in patients on ART (Serrano-Villar et al., 2017; Herrera et al., 2019).

Many of the microbes that have been shown to be oncogenic for specific cancers may cause additional cancers beyond what has already been demonstrated. HPV, implicated in a host of cancers, including cervical, head and neck, anal and penile cancers, may also be associated with prostate (Tsydenova et al., 2023), breast (Kudela et al., 2022), and colorectal (Hsu et al., 2022) cancers. For example, studies have found a predilection for immunohistochemistry-associated HPV presence in prostate cancer tissue when compared with healthy tissue (Zambrano et al., 2002; Lawson and Glenn, 2020). Further research to determine whether there is a facilitative role for HPV in prostate cancer could be considered a priority since, existing tools to prevent HPV infection would be used differently (in boys, perhaps with boosters later in adulthood) and could have dramatic public health benefits, should it be confirmed that a proportion of prostate cancer is triggered by HPV infection and persistence.

Likewise, there are data suggesting that EBV may be associated with breast cancer and gastric adenocarcinoma (Tavakoli et al., 2020; Jeong et al., 2022; Agolli et al., 2023). Determining such relationships could be pivotal for prioritizing EBV vaccine development. In addition to its role as an established trigger for gastric cancer, H. pylori, has been hypothesized to be linked to lower esophageal adenocarcinoma (Islami and Kamangar, 2008). Finding further cancer associations for H pylori, could increase the application of resources to utilize the knowledge to develop diagnostic and prevention tools.

Oncogenesis theories must consider “hit and run” cancer mechanisms, where the oncogenic driver may have initiated processes many years ago and now be undetectable; radiation and known mutagen exposure years prior to cancer detection are classic examples, but the concept also applies to microbial oncogenesis ( Smith and Saveria Campo, 1988 ; Tommasino, 2017 ) This is the case for HPV and head and neck cancers, as well as β-HPV and cutaneous cancers, and may also be implicated in the other oncogenic infectious diseases mentioned in this paper ( Ferreira et al., 2021, Ferreira et al., 2023). Other causal criteria may be fulfilled, but it may not be possible to find histopathological presence in tumor tissue nor persistence of the viral genome ( Ferreira et al., 2021). The determinant for causation, in a “hit and run” circumstance, may be a pattern of dysregulation, as discussed in this review, that if present, suggest an infectious trigger. Alternatively, microbes producing similar disruption as observed with known microbial cancer pairs (as with HPV and cervical cancer) may provide an indication despite the offending microbe not being present ( Irrazábal et al., 2014; Muhammad et al., 2019).