Differences in commensal microbiome influence how we respond to cancer treatment


Fig 1A. Comparison of the 62 OTUs with varying abundance in responders (R) and nonresponders (NR). Red corresponds to a higher abundance while blue corresponds to low abundance.

DOI: 10.1126/science.aao3290

Melanoma is the deadliest form of skin cancer; the American Cancer Society estimates that in the United States alone, over 178,000 will be diagnosed with melanoma and close to 10,000 will die in 2018 (1). While melanoma is curable if caught early on, once it metastasizes, patients have a 16% five-year survival rate and treatments are limited (2)(3). Immunotherapy using PD-1 inhibitors, which consists of antibodies against PD-1, is commonly used. PD-1, which stands for programmed cell death 1, is expressed by cancer cells and known to be implicated in their escape of immune destruction (4)(5). Anti-PD-1 immunotherapy is used clinically to treat patients with metastatic melanoma; however, the efficacy of the treatment is limited to a subset of patients who exhibit T cell response in the tumor microenvironment prior to immunotherapy (6). Thus, current research is oriented toward exploring factors involved in T cell infiltration. Interestingly, studies point towards a link between the commensal microbiome and response to immunotherapy, which occurs by modulation of the inflammatory process contributing to cancer and cancer therapy (7). Building off of this, Matson et al from the University of Chicago explored how microbiome composition might impact patient response to anti-PD-1 immunotherapy.

In Biochemistry 441, we covered a paper on the human gut microbiome and the impact of different diets on diversity over subsequent generations. The paper we read studied how the microbiome has changed over the years but didn’t assess how this might affect human health. The Matson et al researchers study how the microbiome impacts health. Additionally, the paper we read in BCM 441 stressed biodiversity, and this paper provides possible evidence towards the importance of having the “right” kind of diversity.

In their paper, Matson et al explored a possible relationship between commensal microbiome composition and response to anti-PD-1-based immunotherapy. They first retrieved stool samples from patients who were responsive (R) and nonresponsive (NR) to anti-PD-1 treatment. A comparison of the microbiome composition revealed 39 operational taxonomic units (OTUs) more abundant in R and 23 more abundant in NR (Fig 1A). Using screening methods, such as BLAST and qPCR, they narrowed these numbers down to 8 and 2 respectively. In order to investigate the possible correlation between the microbiome and response to immunotherapy, they injected fecal material from R and NR cohorts into germ-free mice. They found that, following melanoma implantation, the mice segregated into two phenotypes: a slower growing tumor and faster growing tumor group. To test whether this difference was due to host immunity, the researchers performed an Enzyme-Linked Immunospot (ELISPOT) of ex vivo SIY-stimulated splenocytes. SIY is an epitope found on the surface of the melanoma cell line used in this paper and can be recognized by immune cells (9). ELISPOT assays are similar in principle to ELISAs and are most often used to monitor immune response. By measuring response to SIY via ELISPOT, the authors found that the R mice exhibited a greater number of activated T cells. Additional analysis of the tumor microenvironment revealed they were mostly SIY-specific cytotoxic T cells, which kill cancer cells, not regulatory T cells, which suppress immune response. Finally, qPCR of the fecal DNA from the mice confirmed that they exhibited the same pattern of enrichment as seen in patients. The authors concluded that response to anti-PD-1 treatment was indeed influenced by composition of commensal microbiota.

These results will likely impact how we treat patients, not only with melanomas, but other forms of cancer as well. Immunotherapy relies heavily on a patient’s ability to generate an immune response, and the microbiome composition could be used as a model to predict this response (10). While the authors recognize that additional factors may affect this, their results provide a powerful addition to combination therapy, which in itself is a cornerstone of cancer therapy. In the future, therapies might combine the knowledge that certain bacteria provide higher response to treatment with the treatment itself. This knowledge may also be used to assess a patient’s response to a treatment and therefore impact whether they receive one therapy over another. Similarly, the David Solit lab has research paralleling that of Matson et al. Dr. Solit’s lab performs genomic sequencing on individuals who respond favorably to a cancer treatment. After finding what differs between these individuals and nonresponsive patients, they decide whether one therapy over another may be more effective. Matson et al follow a new direction in cancer research, which involves research into treatments that are tailored to the specific needs of the individual.

Future studies might involve exploring the mechanism(s) behind how certain bacteria confer a better immune response than others. Research into this subject is limited; however, it has been suggested that certain drugs might alter the composition of the microbiota, which then induces movement of certain bacteria species into secondary lymphoid organs where they stimulate the generation of specific types of T cells (8). While intriguing, whether this is the case with anti-PD-1 therapy is unknown yet unlikely; the drug from the study previously mentioned involved an alkylating agent which differs significantly from an antibody. Research into the mechanism behind how antibodies in particular can alter the microbiome remains to be explored.



  1. Key Statistics for Melanoma Skin Cancer. (Jan 2018). American Cancer Society. Retrieved from https://www.cancer.org/cancer/melanoma-skin-cancer/about/key-statistics.html
  2. Smyth EC, Carvajal RD. (NA). Treatment of Metastatic Melanoma A New World Opens. Skin Cancer Foundation. Retrieved from https://www.skincancer.org/skin-cancer-information/melanoma/melanoma-treatments/treatment-of-metastatic-melanoma
  3. Jazirehi AR, Lim A, Dinh T. (2016). PD-1 inhibition and treatment of advanced melanoma-role of pembrolizumab. American Journal of Cancer Research. 6(10): 2117-2128
  4. Dolan DE, Gupta S. (Jul 2014). PD-1 Pathway Inhibitors: Changing the Landscape of Cancer Immunotherapy. Cancer Control. 21(3), 231-237
  5. Dana-Farber Cancer Institute. (Dec 2017). What is the Science of PD-1 and Immunotherapy? Retrieved from http://blog.dana-farber.org/insight/2015/05/the-science-of-pd-1-and-immunotherapy/
  6. Pembrolizumad. (May 2016). The Melanoma Research Alliance. Retrieved from https://www.curemelanoma.org/patient-eng/melanoma-treatment/immunotherapy/pembrolizumab-keytruda-/
  7. Lida et al. (Nov 2013). Commensal Bacteria Control Cancer Response to Therapy by Modulating the Tumor Microenvironment. Science AAAS. 342(6161) : 967-970. DOI: 10.1126/science.1240527
  8. Viaud et al. (Nov 2013). The Intestinal Microbiota Modulates the Anticancer Immune Effects of Cyclophosphamide. Science AAS. 342(6161): 971-976. DOI: 10.1126/science.1240537
  9. Kline et al. (Mar 2012). Cellular and molecular requirements for rejection of B16 melanoma in the setting of regulatory T cell depletion and homeostatic proliferation. J Immunology. 188(6): 2630-2642. DOI: 10.4049/jimmunol.1100845
  10. Characiejus et al. (Feb 2011). Prediction of Response in Cancer Immunotherapy. Anticancer Research. 31(2): 639-647.

17 Replies to “Differences in commensal microbiome influence how we respond to cancer treatment”

  1. Hi Lily,
    This is such an unexpected link between the gut microbiota and cancer, as most of the gut microbiota-related diseases we saw in class had to do with metabolic disease. It sounds to me like the authors of this paper established a link between the two diseases but didn’t go into the more specific biochemical pathways that make up this link. I’m wondering if the responsive cell line had a different composition of bacteria that in some way impacted the proliferative metabolism of the melanoma cancer cells. In other words, if the melanoma cancer tissue exhibits the Warburg effect and having the “right” gut microbiota composition results in a slower growing tumor and/or responsiveness to anti-PD-1 treatment due to an interference in proliferation mode of metabolism. Thoughts on this hypothesis?

    1. Hi Suzi,
      Thanks for your question! The melanoma cancer tissue most likely exhibits the Warburg effect, as most cancer cells perform aerobic glycolysis. The observation that certain microbiota cause slower growing tumors is linked to the presence of active T cells in the tumor microenvironment. The authors didn’t look into the specifics of what each type of bacteria does and how they might influence the increased number of T cells. I imagine this is something that would be looked into in a future mechanistic study.

  2. Hi Lily! This article was surprisingly niche, and your explanation of what exactly the authors were doing was very helpful. Do you think future work would be better spent on how we can foster the growth of these types of microbiota inside patients, or elucidating exactly how this change is created in T cells? Regardless, I’m a big fan of this new era of personalized medicine, and I’m eager to see exactly how we can influence the gut microbiome to work for us. Like Dr. Colabroy has stated in lecture, this will likely require a much better method of culturing these bacteria outside of the gut. Would such a model be helpful in this type of treatment, where other tissue types and organs may be significantly involved?

    1. Hi Andrew,
      Great question! I think that research into how the bacteria cause this change would definitely be worthwhile, since the results of such a study could potentially have application to other diseases as well. If we could learn what it is about these bacteria that causes the immune system to respond better, I think this would constitute a big step forward in understanding the dynamics of the gut microbiome. But, like you said, it would be extremely difficult to test due to the challenge of cultivating the gut microbiome outside the human body, where it could be better studied. The literature indicates that mice seem to be the most popular model organisms for this purpose (1). As for cultivating the bacteria outside of the gut, I think that since the authors are looking at them in conjunction with T cell response, a whole model organism is probably the most helpful in this case.
      1) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3639412/

  3. As I was reading I began to think about the talk we saw at LVHN, and was wondering if you think we should be taking into account not only what mutation any kind of cancer presents but also the condition of other systems in the body as well, such as the gut microbiota mentioned here. Since cancer is a disease of our own cells, I feel like it would make sense to take into account everything that is going on in the body in order to find the best treatment.

    1. Hi Brandon,
      You bring up a really good point, and yes, I think that doctors are now starting to look at patients as individuals when coming up with a treatment. As Andrew mentioned in his comment, we’re seeing a shift towards a more personalized form of medicine. This is great, for cancer especially, since everyone responds differently to treatment. I definitely think that in the future, it’s possible we’ll see tests for the presence of certain gut bacteria as well for mutations in genes.

  4. Hi Lily, I especially enjoyed reading this post because it’s exciting to see how microbiota research is being pushed forward. It seems like many substantial breakthroughs are on the verge in this field in terms of biochemical mechanisms, and right now researchers are still establishing the direct implications of the right or wrong microbiota diversity. It seems like PD-1 inhibitors are only effective in melanoma patients if the patient’s T-cells are attempting to proliferate and mount an immune response against the tumor. Otherwise, inhibiting PD-1 which destroys T-cells would be of no avail. It was also mentioned that microbiota diversity can cause certain bacterial species to inhabit lymphatic organs to stimulate T-cell attack of the cancer. Do you think that patient supplementation with probiotics with the right types of microbiota can significantly alter PD-1 treatment in metastatic melanoma patients?

    1. Hi Brandon,
      Thank you for the question! From their research, it would appear that the presence of certain types of bacteria gives the host a greater immune response and consequently a better response to anti PD-1 treatment. The researchers took the bacteria present in human responders and non-responders and transplanted them into mice. They were able to see the same pattern of response in the mice, with bacteria from human responders giving the mice a better response. I think that these experiments provide a preliminary answer that yes, supplementation with a certain type of bacteria would be beneficial.

  5. Hey Lilly, you did such a great job summarizing this paper and putting it into context with the class. I really enjoyed how you compared the work of Matson et al and Dr. Solit from LVHN. I have been really enjoying reading these papers about the microbiome, it appears that it has more roles the more we study it. With this being said, the authors do not suggest a mechanism that favors a positive reaction to the immunotherapy. Do you think that these microbiota are able to influence pathways like the Wnt pathway or any other cancer related pathway? Additionally, do you know if there are any other papers that cite an beneficial immunological effect on the host of a healthy person?

    1. Hi Calvin,
      Thank you! Those are interesting question you bring up. For your first question, researchers found that there may be regulation of the Wnt pathway by the microbiota that could change with differences in microbiota composition (1). Additional research suggests there might be cross talk between the microbiota and various important pathways, one of which is the Wnt pathway (the others are Notch and TLR signaling) (2). It’s entirely possible that the bacteria from this study influence this pathway, especially since one effect of Wnt signaling is the development of effector T cell (3). As for your second question, from this research, there appears to be a link between immune response and presence of certain types of bacteria. I don’t know of any papers in particular, but my thought is that in a person benefiting from a “responder” microbiome, the immune response is much better and may possibly work to eradicate tumors that in another person would go on to become cancerous.
      1) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4049928/
      2) https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5440531/
      3) https://www.ncbi.nlm.nih.gov/pubmed/18617885

  6. Lily – our class discussion about the importance of microbiota was very interesting to learn about. It remarkable the microbiota can impact so a myriad of biological processes from metabolism to tumor proliferation. You talk about an alkylating agent that is can possibly a step towards using these findings as a therapeutic. How exactly can alkylation work better than an antibody?

    1. Hi Jess,
      Great question! What alkylating agents do is damage the DNA of a cell, preventing it from dividing further. Since cancer cells divide much faster than normal cells, this makes them a good target for this type of treatment. Antibodies, on the other hand, target a specific antigen, in cancer therapy, this would mean one found on a cancer cell. Since immunotherapies and chemotherapies both operate via different mechanisms, I’m not sure that they would interact with the microbiome the same way. What might work in terms of anti-PD-1 therapy, would be to first treat with an alkylating agent (to stimulate T cell generation) and then with anti-PD-1 antibody (which requires T cell response).

  7. This was an interesting read. The impact of biodiversity on immunoresponse, and not just digestive and metabolic breakdown is interesting and poses the question of how intertwinned our health and the health of our microbiota are. As far as practical applications to treatment, does this imply fecal transplants of responsive bacteria may be the future? While unappealing, this seems a rather trivial aspect of what could be a high impact treatment.

    1. Hi Will,
      This research would certainly point toward a future where fecal transplants are not uncommon. If we think about fecal transplants alone as a treatment for a disease, then yes, this does seem rather trivial. However, I think that most likely what we’ll see is fecal transplants combined with another form of treatment. Whether we start using them as preventative treatments is another question, and I’m not sure of that answer. I can see them being an option, such as a vaccination, and people either choosing to have one or not. What do you think?

  8. Hi Lily! Thanks for sharing this a really interesting article. We always hope to perpetuate any new ideas to treat cancer in the hopes of one day curing it completely, so this article helps us get closer and closer. You mentioned that they tested the microbiota from patients who were responsive to anti-PD-1-treatment and unresponsive and then implanted the fecal matter in mice. They determined that the responsive mice exhibited a greater number of T cells and there is a link between anti-PD-1-treatment and composition of microbiota. With any cancer treatment, there are harmful side effects. I can potentially see a lot of the microbiota in the gut possibly dying off or not dividing at a fast-enough rate from drugs that may be used in immunotherapy. Do you think there could be a way to counteract this? If the gut microbiota are important to generate a response, is there a way to not harm the microbiota while committing to the process of killing cancer cells, which microbiota seem to help do?

    1. Hi Endonita,
      That’s a great point, and I think an issue that needs to be addressed. In fact, previous research into PD-1 and cancer found that antibiotic consumption was associated with poor response to treatment (1). I think that poor response is an issue in a lot of diseases where antibiotics are involved and what happens is patients tend to take fecal transplants after treatment. I can’t think of any ways that cancer treatments involving antibiotics not harm microbiota, and it may just be an inevitable process in cancer therapy. I imagine the easiest way to get around that would seem to be taking a supplement after treatment, or possibly finding treatments without antibiotics. What it probably boils down to is which one is more effective in the long run. In other words, do we need the microbiota more or the cancer treatment more? Maybe something we might see if first taking the antibiotic treatment, then a fecal transplant, then the next form of treatment.

  9. Hi lily, I really liked your choice of the paper and how you were able to link this paper to class. I was just wondering how the microbiota changed the effectiveness of the interfere or amplify the drugs working together with peoples immune systems? I know you mentioned mechanisms but are there any specific that you found stood out in your research?

Leave a Reply

Your email address will not be published. Required fields are marked *