Skip to main content


Bacteria and Cancer: Another Connection

Three months ago, I wrote about a report that some kinds of pancreatic cancer seem to be associated with particular infections, and wondered “How many similar stories are out there that we don’t know about yet?” Well, that didn’t take long to start being answered. This recent paper in Science (from a multicenter team: Dana-Farber, Broad, Harvard, Barcelona, Yale) shows a very similar relationship in some colon cancers – Fusobacterium nucleatum is the bacterial species this time, along with a few others that seem to hang along with it.

These species have been noticed as something associated with some colon cancer tissue samples, but this new work shows that things are a lot deeper than that. When you look at metastatic tumors derived from such tissues, the metastases have the same bacteria as the primary tumor. What’s more, it appears that the F. nucleatum cells are actually localized inside the tumor cells themselves (which would certainly account for the stability as they break loose and form metastases).

That’s one big step, but the team goes on to another. When rodents are transplanted with xenograft tumor tissues of this type, antibiotic treatment (metronidazole) actually slows tumor progression. That takes us from the “associated with” situation to the “caused/exacerbated by” one, which is similar to what was seen with the earlier work on pancreatic cancer and Mycoplasma species. And just as in that case, this strongly suggests that any patients with this sort of tumor should be tried out on antibiotic therapy; there would seem to be little to lose and potentially much to gain.

It’s been known for some time that stomach cancer is associated with H. pylori infection and there’s a less-well-known connection between Chlamydia infection and cervical tumors. I think that by now we can definitely add “bacterial infection” to the list of causes of cancer, just as viruses were added years ago. It’s surely not a coincidence that the known examples, when you think about it, are all associated with proximity to a large variety of different bacterial species (the digestive and reproductive tracts). Are there oral/throat cancers that will fall into this category as well? The next question, and one that I’m sure is being worked on right now, is a mechanistic one: how do these infections lead to tumorigenesis? How do the Fusobacteria invade colon tissue cells and survive once they’ve done so, and what are the changes that take place afterwards? I can think of a number of hypotheses, and this isn’t even really my field, so I’m sure that people who are in the area have plenty of experiments to run.

21 comments on “Bacteria and Cancer: Another Connection”

  1. anonymous says:

    So, can these bacteria be called bio-markers for these cancers? Meaning it detection when coupled with “genetic factors” could foretell the possibility of coming down with these cancers?

    1. MP says:

      These bacterial infections may have a potential use (assuming they are first validated) as biomarkers of susceptibility, or used in some other way in developing a prognosis. I don’t think they could be used on their own to predict whether an individual has or will develop cancer, because they’re likely to have very low specificity (i.e., many more people will have detectable levels of the bacteria than will develop cancer).

  2. UudonRock says:

    The connection with peptic ulcers and H. pylori has been well documented. It is interesting, but not surprising to read about the connection with Fusobacterium. I’m always hesitant to say wow when I see that a bacterium or virus cultures in immune compromised murine xenograft though. I’ve seen Epstein-Barr explode in mice after xenograft from clinical samples that were EBV negative. Grated, I’m talking apples and oranges here. Regardless the scope of this work is compelling and historically antibacterial treatment for peptic ulcers is efficacious. It would be great to see human trials on this course for colorectal cancers.

  3. Andy II says:

    There is another interesting one here. It may suggest that chronic inflammation in GI would lead to cancer?

    Recurrent infection progressively disables host protection against intestinal inflammation. Yang WH, et al., Science. 2017 Dec 22;358(6370). pii: eaao5610. doi: 10.1126/science.aao5610.

  4. MoMo says:

    This is news? Or is it news because Dana-Farber, Broad, Harvard, Barcelona, Yale now notice this 7 years since it was first reported?

  5. It seems entirely reasonable to me, as a non-biologist, that inflammation could increase the risk of cancer. Anything that increases mitosis rates is going to increase defective mitosis rates and the mutations that lead to cancer.

  6. Metacelsus says:

    It is worth noting that metronidazole becomes activated by intracellular reduction, which occurs more readily in bacterial cells than human cells. Once reduced, the compound indiscriminately damages DNA.

    It might be interesting to see whether metronidazole which becomes reduced in the bacterial cells can diffuse out into the cancer cell and help kill it.

    1. MrRogers says:

      Even more interesting is that xenografts are notoriously hypoxia. I’ll need to read the paper, but it seems plausible (likely even) that the mechanism of metronidazole’s anti tumor activity is direct activation in hypoxic tumor cells. In that case antibacterial effects would be irrelevant.

  7. dearieme says:

    I’m disappointed that the reasonable hypothesis that much CVD is caused by micro-organisms has yet to yield an identification of the m-o involved.

  8. R. Thomas Winters says:

    Here is a link to an earlier publication on the subject from 4 years ago.

    Wish I had seen this earlier would have had my wife’s primary check for this. Might have tried to have her Oncologist try the metronidazole in addition to here Chemo had this been discovered.

  9. Snap. I’ve recently written on this topic, along with other blog pieces on the “bugs and cancer” theme.

    A group at NYU Langone reported an association between certain oral bacteria and oesophageal adenocarcinoma risk, being elevated if Tannerella forsythia, was present and decreased if Streptococcus and Neisseria species were present. Risk also correlated with certain metabolic pathways.

    Some interesting observations are merging around the microbiome and responsiveness to PD-1 directected immunotherapy.

    Oral microbiome composition reflects prospective risk for esophageal cancers. Peters, BA et al. Cancer Res 2017;77:6777-6787.
    Microbes matter. Anna Dart. Nature Reviews Cancer,online December 8th, 2017 doi:10.1038/nrc.2017.120

  10. thank you for the article, i too have written about many associations of bacteria to cancer in my book “never brush your teeth again” my fda approved device provides 500% the desired outcome of periodontal therapy alone in reducing Fn and Aa, Pg and other bugs inplicated in cancer associations from breast cancer to esophegeal, lung , colon, prostrate, etc. these are oral pathogens. the take away is it serves people well to put your money where your mouth is. follow me on twitter ppierodds or see for more info.

  11. Imaging guy says:

    Though I want to believe that cancers are caused by bacteria, I doubt that is the reality. Professor Robert Weinberg is a very famous cancer researcher and his take of 40 years of cancer research was described in this article (1). If you think cancer research is smooth sailing, you should read that.
    1) “Coming Full Circle—From Endless Complexity to Simplicity and Back Again”, 2014, Cell, PMID: 24679541

    1. I share your doubt with respect to any simple causal relationship but I’m open to the notion that the composition of gut or oral microbiomes might contribute to conditions which favour tumour growth and survival, possibly through activation of the innate immune system or subtle interference in immune system cross-talk, resulting in sustained “pro-tumour” inflammation.

      The tumour microenvironment does appear to be influenced by systemic factors, including exercise and nutrition, so it’s not unimaginable that the metabolic effects of a particular gut flora composition might somehow promote (or constrain) tumour growth. But, even if this proves to be the case, I think we’re a long way off from incorporating directed manipulation of the gut microbiome in cancer treatment.

      Nutritional effects on T-cell immunometabolism. Cohen, S et al. Eur. J. Immunol 2017. 47(2); 1521-4141.

      Exercise-dependent regulation of the tumour microenvironment. Koelwyn, GJ et al. Nature Rev Cancer October 2017 17: 620-632.

  12. Dr David Taylor says:

    On a tangential note, about half of all the cases of Hodgkin’s lymphoma, including my own, appear to be due to Epstein–Barr virus. I contracted Glandular Fever at 66, to much hilarity from friends and family (its often called teenage kissing disease), but the humour faded when I was diagnosed two years later with early onset Hodgkins.

  13. Barry says:

    The association need not imply causality. Any successful intracellular pathogen must damp down host immune response. That immunosuppression (which may be local to the gut?) would be enough to explain higher incidence of colon cancers.

  14. Vader says:

    The spam is strong with this one.

  15. MFernflower says:

    I guess my idea of using minocycline as an anticancer drug was not that far-fetched afterall?

    Interesting read indeed!

  16. bacillus says:

    H. pylori, HPV and HCV are all officially classified as carcinogens. The vaccine against HPV is working remarkably well at preventing cervical cancer. A hep-C vaccine should be expected to do the same for hepatic carcinoma, and getting rid of H. pylori has significantly reduced gastro-duodenal ulcers and gastric cancer. However, the latter colonises the stomachs of ~50% of the global human population, so other co-factors are also in play. The F. nucleatum story is already old news, with solid evidence of it participating in colon cancers. If it is a major contributor, then a vaccine against it could help lower colon cancer rates. The ability of the normal flora to induce everything from type I diabetes, to MS to reactive arthritis and diseases of the heart has been scrutinized at one time or another, but with few if any clear smoking guns. Of course there are thousands of microbes in the normal flora and the vast majority of them are unculturable and haven’t been studied much beyond identifying their existence by various genetic approaches. Therefore, the possibility that dysbiosis of the normal flora underpins other cancers should not be lightly dismissed.

  17. David says:

    Does this mean stomach cancer and ulcers are linked or share precursors? Wasn’t H pylori linked to ulcers?

  18. WeltM says:

    Just run into a paper on Toxoplasma gonddi affecting monocytes hypermotility from JBC. Another aspect of bacteria and cancer?

Comments are closed.