Why do some people respond to cancer therapy while others do not?
Martin Hornshaw, PhD
Director, Scientific Marketing -- Metabolon, Inc.
Respond or not respond, that is indeed the therapeutic question.
Why do some people respond to a cancer therapy and some do not? While cancer is a genetic disease, response to therapy is not as systematic or clear-cut as simply matching a mutation to treatment. Other factors are clearly involved. Recent work in both mice and humans suggests that one answer may involve the gut.
What might the difference be that enables a positive response to treatment? One possibility is that patients' gut microbiomes (the community of microorganisms living in the gut) might differ and in a way that affects their response to therapy. Indeed, cancer patients do have differences in their microbiomes that correlate with responses to immunotherapy as recently described in two papers in Science. A third paper also looks at microbiome differences in cancer patients, but the authors performed global metabolomics to explain mechanistically what might be different between those who respond and those who do not respond to treatment. The work in these papers suggests the gut microbiome as a possible route to enhancing cancer immunotherapy.
Matters of the gut
Some general conclusions from two of the studies were that there were significant differences in the diversity and composition of the gut microbiome of patients who responded well to treatment (responders) compared with those who did not (non-responders) and that taking antibiotics during treatment negatively affects outcomes. Wargo et al.1 identified that systemic and anti-tumor immunity were enhanced in responders who had the "favorable" gut microbiome and in germ-free mice with cancer who had received fecal transplants from responders. They further observed impaired immune response in mice that had received fecal transplants from non-responders. Perhaps a better way of looking at this is that resistance to treatment was caused by an abnormal gut microbiome composition, which likely affected the ability to mount an immune response.
The second team led by Laurence Zitvogel2 found that patients on antibiotics, which disrupt the gut microbiome, did not live as long. They also transferred the gut microbiome by fecal transplant from responders to germ-free mice with cancer. These mice fared better on treatment than corresponding mice who received fecal transplants from patients who did not respond well to therapy. Zitvogel's team took things a step further by studying oral supplementation with bacteria they had noted were enriched in responders, such as Akkermansia muciniphila. The Akkermansia 'treatment' resulted in an enhanced immune response into the tumor beds, the normal tissue where the tumors make their horrible home.
Of note is that the first study was in melanoma patients and the second in patients with epithelial tumors such as lung, kidney and bladder cancers. We can conclude that the composition of your gut microbiome seems to be incredibly important as to whether you respond to immunotherapy for at least several types of cancer.
What was not studied in these two very significant papers was the metabolome. Notably, the metabolome has become a valuable tool for understanding microbiome-associated phenotypes across a host of areas. How the microbiome, through the metabolome, might relate to clinical outcomes in cancer immunotherapy brings me to another study.
Frankel et al.3 described some intriguing, but preliminary, observations in 39 metastatic melanoma patients who either responded favorably to treatment or progressed to disease.
Cancer checkmate? Not quite, checkpoint.
As with the first two papers described above, this study sought to investigate the clinical responses to immune checkpoint therapy (ICT) and understand why ICT, which stimulates the immune system, sometimes show great efficacy, but only sometimes. Associations of clinical responses (responder/non-responder) to the different treatments with gut microbiota taxonomy profiles, gut metabolite levels and patient dietary and antibiotic histories were evaluated. Each patient had metagenomic shotgun sequencing performed on fecal specimens collected prior to treatment. Aliquots of the same fecal samples were also analyzed by global metabolomics.
Differences in microbiomes and metabolites and response to checkpoint inhibitors
In terms of differences in microbiome, there were subtle differences between responders and those who progressed. All ICT responders had a microbiome enriched with the same species, Bacteroides caccae, for example. There were also gut microbiome differences in terms of enriched bacterial species between the different regimens of treatment. The study was too small to address the impact of antibiotics.
In addition to differences in gut microbiomes, the researchers found significant differences in responders' metabolites. By far, the greatest difference was in a plant xenobiotic 15:2 anacardic acid observed in the fecal samples, which is an alkyl derivative of the plant hormone salicylic acid. In ICT responders, it was greatly elevated compared to those who suffered progressive disease.
Anacardic acid is not known to be a bacterial metabolite, and as far as we know it comes from plant-based products. It is interesting that five of the six patients with the highest levels of anacardic acid reported consuming cashews, which are known to produce anacardic acid (mango does, as well). It has been noted that anacardic acids stimulate immune cells, neutrophils and macrophages, which may then go on to recruit T-cells to tumor metastases and thus enhance ICT. This gives one plausible mechanism for why some patients respond to ICT and others do not. Of course, there could be more than one mechanism.
We have a way to go, but these results are intriguing.
It is obvious that these results warrant further investigation, not just in terms of repeat studies, but also as to a possible mechanism for gut microbiome action in stimulating the immune system to attack cancer cells. In addition, the surprising and unexpected finding of highly elevated levels of anacardic acid in responder gut samples merits study as a potential therapeutic intervention.
Global metabolomics should be performed with the goal of identifying and quantifying a wide range of metabolites derived from host metabolism, the microbiome, diet and so on. If the researchers3 had only performed targeted assays of known microbial metabolites, such as secondary bile acids or short chain fatty acids, this discovery would not have been made.
Interestingly, Wargo et al. did attempt an indirect analysis of metabolism using whole genome sequencing data of the gut microbiome metagenome1. They inferred from that analysis that there were likely differences between responder and non-responder microbiomes metabolically. However, they did not perform direct analyses of the global metabolome of the gut, which would have been the best way to assess differences in the metabolomes of responders and non-responders.
Similarities & differences
While there were similarities, for example the two studies involving melanoma1,3 both identified that Faecalibacterium were enriched in some responders, all three studies identified different bacteria as being enriched in the gut microbiomes of responders. Why? What does this mean? Is there commonality in the metabolomic output of these different bacteria, or are they working differently in terms of an immune-stimulatory effect? Does any healthy gut microbiome prevent dysfunction of the gut barrier that might then prevent immunosuppression? Are there many or few bacterial routes to positive impact on cancer treatment of the gut microbiome?
There are many interesting and important questions to answer and routes to positive microbiome health to be explored. A metabolomics approach that studies a breadth of metabolism generally gives very strong insight into fundamental biology. This will likely uncover clues as to what is similar and what is unique about the gut metabolism and immune-response of these cancer patients, which may one day lead to improved treatments. I am extremely encouraged that we may soon see more effective cancer treatments based on a combination of immunotherapy plus microbial and/or metabolite supplements.
1. V. Gopalakrishnan et al. Gut microbiome modulates response to anti-PD-1 immunotherapy in melanoma patients Science (2017) epub Nov 2 DOI: 10.1126/science.aan4236
2. B Routy et al. Gut microbiome influences efficacy of PD-1-based immunotherapy against epithelial tumors Science (2017) epub Nov 2 DOI: 10.1126/science.aan3706
3. AE Frankel et al. Metagenomic Shotgun Sequencing and Unbiased Metabolomic Profiling Identify Specific Human Gut Microbiota and Metabolites Associated with Immune Therapy Efficiency in Melanoma Patients. Neoplasia (2017) 19, 848-855