Changing the Gut Microbiome to Improve the Efficacy of Immunotherapy

Drs. Diwakar Davar and Ben Boursi discuss the role of the gut microbiome in the outcome of cancer immunotherapy and the prevention of immunotherapy-related adverse events, as well as compelling research on nutritional interventions to improve response to immune checkpoint inhibitors. TRANSCRIPT   Dr. Diwakar Davar: Hello, and welcome to the ASCO Daily News Podcast. I'm your guest host, Dr. Diwakar Davar. I'm an associate professor of medicine and the clinical director of the Melanoma and Skin Cancer Program at the University of Pittsburgh's Hillman Cancer Center.   Researchers have shown that microorganisms in the gut can impact the effectiveness of immunogenic chemotherapy for patients with cancer. Although microbial therapies for cancer are still at a very early stage of clinical development, compelling research in recent years has shown that changing the gut microbiome can help improve outcomes in patients receiving treatments for cancer enduring immune checkpoint inhibition.  My guest today is Dr. Ben Boursi, a GI medical oncologist at the Sheba Medical Center at Tel Aviv University in Israel. Dr. Boursi is also an adjunct professor at the Center for Clinical Epidemiology and Biostatistics at the University of Pennsylvania. He joins me today to discuss his pivotal research on the role of the gut microbiome in mediating its effects on immunotherapy. And again, I want to highlight that we're recording this on October 9th, and as you may well know, many recent events over the last couple of days have happened in Israel, and so Dr. Boursi has joined us at a very difficult time. So, we're very grateful for him taking time out of his suddenly very busy schedule to join us at a time that is fraught for all. You'll find our disclosures in the transcript of this episode. You'll also find the disclosures of all guests on the podcast at asco.org/DNpod. Ben, it's great to have you on the podcast today. Thank you for being here at such a difficult time, sharing what will, I think, be a great episode.  Dr. Ben Boursi: Thanks for having me, Diwakar. Dr. Diwakar Davar: Ben, the gut microbiome and its role in terms of mediating effects and side effects of cancer immunotherapy has gotten a lot of interest recently. You've done some fundamental work in this space. Why don't you briefly summarize for the audience, firstly, what is the gut microbiome and what are the major themes in relation to cancer immunotherapy?  Dr. Ben Boursi: Well, the microbiome is the ecosystem of microorganisms, bacteria, phages, fungi, that are crucial for immunologic, metabolic and hormonal homeostasis of the host. In the last decade, we began to understand the central role of the gut and tumor microbiome in tumorigenesis, metastasis, treatment efficacy and toxicities, and in 2022, polymorphic microbiomes became one of the hallmarks of cancer, in addition to previous hallmarks that focused mainly at the cellular/genetic levels. The initial studies in mice models showed that therapeutic efficacy of immunotherapy depends on both the presence and composition of the microbiota (In germ-free or antibiotic treated mice, immunotherapy is ineffective), and following these studies, three observational studies in human patients showed that the gut microbiome can predict response to immunotherapy and that response to immunotherapy could be transferred to germ-free mice by fecal microbiota transplantation (FMT) from responding patients.  These studies helped us to define three main research questions regarding the possible role of microbial modulation in cancer treatment. First, can microbial modulation overcome resistance to immunotherapy, both primary and secondary resistance? And this question was the focus of the initial proof of concept studies. Second, can microbial modulation improve response to immunotherapy in treatment-naive patients? And third, can microbial modulation prevent or treat immune related adverse events? The initial positive results of clinical trials also led to additional questions. For example, can microbial modulation induce anti-tumor immune response even in non-immunogenic tumors? And it is important to note that there are many ways to modulate the microbiota, but so far, the only reliable way that showed positive results is fecal microbiota transplantation that allows the transfer of the entire microbiota both in terms of composition and relative abundance. Dr. Diwakar Davar: That’s great. Essentially with the trials that I think the data sets that you're referencing of course, are papers by Jennifer Wargo, Thomas Gajewski, and Lawrence Zitvogel, looking at the role of gut microbiota in several different cancers, primarily immune checkpoint sensitive tumors such as melanoma, non-small cell lung cancer and kidney cancer. And then the work from several different groups showing that essentially proof of concept experiments can be done to try to change this, certainly preclinically, and now we know that that can be done clinically.   So, I guess the failure rates of immunotherapy in some patients are quite high. And we know that the microbial composition can change the likelihood to respond to immunotherapy based on all these trials. And actually, even going back to 2015, we had two seminal papers that looked at the role of CTLA-4 blockade as well. But subsequently, many years after that, 7 years after 2015, and certainly 3 years after 2018, when the three observational PD-1 papers were published, there were 2 pivotal trials in PD-1 advanced or refractory melanoma. They demonstrated that changing the gut microbiome can reprogram the immune system to attack tumors. So, there were 2 separate trials, both published the same issue of Science. One trial was led by your group at Sheba, and another one's led by us, the University of Pittsburgh. Why don't you summarize both studies for our audience. Dr. Ben Boursi: So, both studies were Phase I clinical trials of FMT in metastatic melanoma patients who failed immunotherapy. Recipients were metastatic melanoma patients that progressed on at least one line of anti PD-1 and in BRAF mutated patients, BRAF inhibitors as well. Donors in the Sheba study were metastatic melanoma patients with durable complete responses to immunotherapy for at least one year, and in the Pittsburgh study, you also included patients with durable partial responses of more than two years as donors. It is important to note that each fecal transplant in both studies was composed of a single donor. Prior to transplantation, we performed a microbiome depletion phase using a combination of two antibiotics, vancomycin and neomycin. The goal of this phase was to assist in engraftment (by avoiding colonization-resistance by recipient bacteria) and to “reset” the immune system, which may remind some people of the logic behind bone marrow transplantation. In the Pittsburgh study, there was no bacterial eradication with antibiotics, mainly because of studies showing that response to immunotherapy is lower following antibiotic treatment.  Both studies performed FMT through colonoscopy. At Sheba, we also performed maintenance FMT using capsules in order to keep the donor's microbial composition. After the initial FMT, both studies reintroduced the same immunotherapy in which the patient progressed in the past. Clinically, we have seen a 30% response rate with durable, complete and partial responses, and in the Pittsburgh study, there was a 20% response rate and 40% disease control rate. Both studies showed following FMT, immune response in the gut and in the tumor, and tumors that were immune deserts prior to FMT became infiltrated with lymphocytes. Interestingly, in our study, there were no moderate to severe immune related adverse events following FMT and reintroduction of immunotherapy. And this is despite the fact that five of the patients had significant side effects during previous rounds of the same immunotherapy.  Dr. Diwakar Davar: So essentially, in these very early proof of concept studies, what I think is pretty remarkable is that obviously the sample sizes were very small, but remarkably, patients that appeared to respond, responded in a setting in which they were not expected to respond. So, the probability of a patient responding to attempt at giving PD-1 in patients who were PD-1 relapse refractory is on the order of about 7%, based on an FDA analysis by Viva et al. And here, two separate studies, two independent studies, investigators had not known that each paper was being published, remarkably similar results clearly demonstrating that this is perhaps one of the best pieces of evidence to suggest that microbiome modulation may actually truly be effective in reversing PD-1 refractoriness.  More recently, our colleague Dr. Bertrand Routy at University of Montreal has done a proof of concept trial in evaluating the use of healthy donor fecal microbiota transplant in addition to anti PD-1 monotherapy in PD-1 naive metastatic melanoma. In this study, published in Nature Medicine a few weeks ago, his group reported an objective response rate of 65%. What are your thoughts about this study? And specifically, what are your thoughts about some of the pharmacodynamic and translational results that were demonstrated?  Dr. Ben Boursi: This is a very interesting question, because in both the Sheba and the University of Pittsburgh studies we chose responding patients as donors. We thought that by using these patients, we provide beneficial bacteria that enhance responses to immunotherapy through several mechanisms (molecular mimicry, immunomodulatory bacterial metabolites, modulation of immune checkpoint expression, and much more), and here in the Routy paper, the researchers used FMT from healthy donors without any selection for specific beneficial bacteria, and they demonstrated a similar effect on overall response rate. So maybe FMT works actually through reducing colonization by deleterious bacteria? Another question that we should ask is whether we need to choose donors differently when we use microbial modulation in treatment resistant patients compared to treatment-naive patients? Moreover, a previous meta-analysis of FMT studies across indications that was conducted by the group of Dr. Nicola Segata, demonstrated that recipients with better engraftment were more likely to experience clinical benefit, and that increased engraftment was mainly observed in individuals receiving FMTs through multiple routes, colonoscopy and capsules, as well as recipients that received antibiotics prior to FMT. But in Routy’s trial, they not only used healthy donors, they performed bacterial cleansing only prior to FMT instead of bacterial eradication with antibiotics, and used FMTs through colonoscopy only, and they didn't give maintenance FMT. Of course, such an approach is much more feasible in the clinical setting and is relevant for designing future clinical trials.  Dr. Diwakar Davar: So, many differences, relatively few similarities, but I guess one interesting point is that of engraftment, which is that in your paper, our paper, and certainly in Bertrand's paper, it is very interesting that engraftment appears to be a key pharmacodynamic biomarker of microbiome modulation. And certainly, the analogy that you used earlier, which is that it's very similar to what happens in a stem cell transplant, which is that if there's no take, there's probably not going to be any effect. So that's very interesting that engraftment is emerging as a key PD biomarker of essentially the success of any kind of microbiome modulation across multiple different settings.  Now, we've heard of certainly defined microbial consortia, of cultivated species, as an alternative gut microbiome modulation strategy that balances the benefits of the ecological complexity of FMT with the scalability and practicality of probiotics. Do you think we are ready to design consortia?  Dr. Ben Boursi: So to date there are several probiotics that use a single bacteria and several microbial consortia that were evaluated in clinical trials, and as you mentioned, they may offer more tractable solutions for widespread clinical use. If we begin with the single bacteria probiotics, two phase 2 clinical trials found that administration of the butyrate producing probiotic clostridium butyricum 588 (CBM588) to immunotherapy naive patients with metastatic renal cell carcinoma led to markedly better immunotherapy responses, although the probiotic had a minimal effect on the composition of the microbiota, and the control arm of the trial responded worse than expected. In addition, in preclinical studies, probiotic strains of lactobacillus and bifidobacterium have been shown to enhance immune control of transplanted tumors and to augment anti PD-1 activity. However, a clinical trial in patients with metastatic melanoma found that the use of lactobacillus or bifidobacterium probiotics was associated with reduced microbiota diversity and worse responses to anti PD-1.  So here the conclusion is that when we try to design probiotics, we should not focus only on the composition since other factors, such as the relative abundance also matter. Too much of a beneficial bacterial species may potentially be worse than having a balanced and diverse microbiota. For example, a recent study of patients with non-small cell lung cancer receiving immunotherapy found that patients with a detectable Akkermansia muciniphila in their gut microbiota (this is a beneficial bacteria) responded well to treatment, but those with relative abundance of Akkermansia muciniphila greater than 5% responded worse than patients lacking Akkermansia, and this is due to the mucolytic effect of the bacteria. So, the use of rationally designed consortia may be better than a single probiotic strain.  And there are currently 3 main microbial consortia that are being evaluated: the SER-401, a bacterial consortium enriched with clostridium, led in a randomized controlled trial to reduced response to immunotherapy compared to placebo control in first line metastatic melanoma patients, potentially due to a confounding effect of a vancomycin pretreatment; MET4 is a 30 bacteria consortium that was shown to be safe and to alter the gut microbiota and serum metabolome of immunotherapy naive patients. Here, the initial study was underpowered to determine the effect on treatment efficacy; And finally, VE800 is an immunotherapy enhancing 11-bacterial consortium that is currently being evaluated in phase 1 and 2 clinical trials, and we are looking forward to see the results with this agent. Dr. Diwakar Davar: So I guess where we are right now is that social design is clearly difficult because of all the reasons you've mentioned. The SER-401 data and the MET4-IO trials certainly give us pause for thought. Certainly, no pharmacodynamic changes that were seen with SER-401, MET4-IO did result in pharmacodynamic shifts metagenomically, but neither trial was positive. And certainly, the VE800 trial, which has been ongoing now for several years, and the lack of publicly reported data certainly doesn't suggest that there's a huge efficacy signal. So consortias, at least at this point, certainly do not appear to be having a significant effect, though we don't know what might happen in the future. Data from multiple groups has shown that gut microbial composition influences the development of immune related adverse events (irAEs) in both PD-1 and combination PD-1 and CTLA-4 treated patients. Unsurprisingly, as a result, there have been attempts made at evaluating the role of fecal microbiota transplants to treat refractory immune related adverse events and very specifically immune checkpoint associated colitis or IMC. So, Dr. Yinghong Wang, who is the chair of the Immunotherapy Toxicity Working Group at the University of Texas MD Anderson Cancer Center has been very prominent in this space, and in a recent paper published in Science Translational Medicine, which is a follow up paper to her early work in Nature Medicine, she reported that HDFMT, healthy donor fecal transplantation, was very efficacious in feeding early refractory immune checkpoint colitis. So, what are your thoughts on this approach and how important is this space and where else might it be efficacious?  Dr. Ben Boursi: When I talked about the Sheba clinical study, I mentioned the possible role for microbiota modulation in the prevention of immunotherapy related adverse events in general, not only colitis. But the study by Dr. Yinghong showed that FMT can actually treat immune-related colitis refractory to steroids and anti-TNF. Now, this approach is probably relevant not only for immune related colitis, but also to other immune related adverse events. We can define certain bacterial species that may be associated with different immune related events. For example, streptococci can be associated with immune related arthritis. And maybe in the future we won't need to use FMT, but we will rather be able to target these specific immunogenic strains by narrow spectrum antibiotics or phages. The main challenge would be to develop microbiotic targeting interventions that reduce immune related adverse events without compromising therapeutic efficacy.   Now, is microbial modulation relevant only for toxicity from immune checkpoint inhibitors? So, the answer is ‘no’. We know mainly from animal models of hematopoietic cell transplantation, CAR T, and immune agonist antibodies that antibiotic-treated or germ-free mice have markedly reduced immunotoxicity, such as graft versus host disease, cytokine release syndromes, and more. It is also worth mentioning that microbial modulation is relevant not only for reducing toxicity from immunotherapy, but also from chemotherapy and other anticancer modalities. And the best example is the gastrointestinal toxicity of irinotecan that is mediated by the bacterial beta-glucuronidase. And here the targeting may even be a bit less complex. Dr. Diwakar Davar: So, what we take away from that is that starting with actually your paper originally, and papers to be produced, immune-related adverse events can be prevented using microbiome modulation with FMT, and Dr. Wang's data suggesting that eventually FMT can be used to eradicate highly refractive colitis, again, this is important to keep in mind that this approach is not yet FDA-approved. It's being done under IND. It's not currently something that is a certain standard of care. One interesting area of drug development is that there's a French microbiome company named MaaT Pharma where they have an agent that is a very interestingly a pooled microbiome product from multiple different donors. Again, the trials in both Israel and Pittsburgh used individual donors. This is a pooled donor construct. The lead candidate is actually graft versus host disease. The trial is the ARES trial, A-R-E-S, as in the Roman god of war. This trial is actually ongoing in Europe, and I believe there's some effort to try to see whether or not it's going to be a trial that can be done in the United States as well. So, at this point in time, again, we don't know whether or not there are any developmental approaches from a pharmaceutical company in the United States, but certainly this is definitely an area of interest.  So microbial therapies are still relatively early. It's going to be interesting to see how the advanced field of nutritional interventions provide an appealing method for modulating the gut microbiome due to the excellent safety profile, cost effectiveness and noninvasiveness. And certainly, if you are what you eat and your bacteria are what they eat, which goes down to our diet, there's enough rationale to believe that certain nutritional interventions can have an effect via the intermedial gut microbiota modulation. Holistic dietary changes and or supplementation specific nutrients such as prebiotics could therefore be utilized to specifically shape the population of beneficial microbes and shift the immune microbiota landscape. Now, we have seen in data published by several of our colleagues that in patients with cancer, high fiber intake is associated with greater microbial diversity, greater abundance in fiber fermenting microbes such as members of the Ruminococcaceae family, and these are all associated with the response to checkpoint inhibitor therapy. So, what do you think about nutritional interventions? Do you want us to briefly summarize data regarding nutritional data and where it stands in cancer at his time? And can you speculate as to how effective this might be in the context of patients with cancer? Dr. Ben Boursi: So, let's begin with diet. A growing number of clinical and preclinical studies suggest that specific dietary interventions such as a high fiber diet can not only improve response to immune checkpoint blockers, but also reduce immunotoxicity such as graft versus host disease. And there are many other diets that are being tested such as ketogenic diets and intermittent fasting. And the effects of diet may be mediated by both microbiota-dependent and microbiota-independent mechanisms. The limitation of this approach is that changes to the microbiota induced by diet are generally quite variable between patients and can depend on an individual's microbiota prior to intervention. And patient compliance is also a concern, particularly in the very strict diets.  Now, regarding high fiber diets, several large cohorts of melanoma patients from the US, Australia, and the Netherlands demonstrated how a high fiber diet modulates the microbiome and results in a better response to immunotherapy, better progression-free survival. Additional studies that were presented at AACR in 2023 showed that high fiber dietary interventions, in which patients received a fiber-enriched diet for six weeks, was feasible and that the high fiber diet resulted in a rapid shift in the gut microbiota toward fiber-responsive short chain fatty acid-producing taxa and a shift of the metabolome, with increase in the short chain fatty acid acetate, Omega-3, Omega-6, polyunsaturated fatty acid, and tryptophan metabolites. Prebiotics can also promote the growth of beneficial microbial species in the gut by providing targeted nutrition. And one example of a prebiotic that was shown to enhance immunotherapy efficacy in mouse models is castalagin, which is isolated from the camu-camu berry. Castalagin directly binds the outer membrane of ruminococci and promotes their growth, which has been shown to increase the CD8-positive T-cell activity and anti-PD-1 efficacy. Now, since prebiotics rely on the presence of beneficial taxa already in the host microbiota, symbiotics, which refers to the administration of the appropriate prebiotic and probiotic together, may prove in the future to be more effective than using either separately. Dr. Diwakar Davar: Certainly, these dietary interventions can be very exciting and certainly we do know of several colleagues who are doing these diet interventions, though compliance with any kind of dietary intervention may be a challenge that decides how effective such an approach is going to be. So microbial therapies in general are still at a relatively early stage of development. And it'll be exciting to see how they advance. What approaches are you excited about? What is on your radar?  Dr. Ben Boursi: There are many exciting works that are currently ongoing, and to emphasize just a few: there are many clinical trials in immunogenic tumors, in addition to melanoma, for example, renal cell carcinoma, and non-small cell lung cancer, that also evaluate different modulation protocols. We should remember that one size does not fit all, and different tumors have different microbiomes. We have a project in collaboration with MD Anderson in MSI-high patients with exciting initial results. Another study that was initiated at Sheba is using microbial modulation in order to improve TIL therapy (to overcome resistance to TIL and T-cell exhaustion). There are also studies that try to change the pharmaco-microbiome, for example, to eradicate bacteria that inactivates the chemotherapy agent, gemcitabine, in pancreatic cancer patients. And there are groups that try to identify recipients that will respond to microbial modulation and to generate better donor-recipient matching algorithms. There are already signatures like TOPOSCORE that was presented at ASCO 2023 that try to predict response to immunotherapies through the ratio between harmful and beneficial bacteria. Now, there’s also more basic science work, for example, bacterial engineering. There was a wonderful study from the Fischbach group in Stanford that demonstrated how Staphylococcus epidermidis engineered to express melanoma tumor antigens was able to generate a systemic tumor-specific response in mice models when applied topically; functional imaging of the microbiome, for example, FDG uptake in the colon can reflect microbial diversity and response to immunotherapy; works that characterizes other microbiomes such as the urinary and skin microbiomes, and their interaction with the gut microbiome; and studies of the nonbacterial component of the microbiome, mainly phages and fungi. But for me, the most important word should probably be collaboration, because without joining forces internationally, we won't be able to understand the human metaorganism, the variations according to geography, ethnicity, lifestyle, diets, and much more in the microbiome. And this is crucial in order to really understand the complex tumor ecological niche within the human host. Dr. Diwakar Davar: I think one of the key points that you just mentioned is collaboration. That's going to be very, very critical as we move this forward for many reasons, including the unexpected impact of geography upon the composition of the gut microbiome in work that has been published by many groups, but also including ours in a paper that we published about a year ago now.  So, Dr. Boursi, thank you for your great work in this area. Thank you for sharing your insights with us today on the ASCO Daily News Podcast. This is a very difficult time for all of you and your colleagues in Israel, and we thank you so much for taking such a great deal of time out of your busy workday to spend some time with us.  Dr. Ben Boursi: Thank you very much. Dr. Diwakar Davar: Thank you to all our listeners today. This is a very exciting area. This is an area where we are discovering more every day than we knew just up until the day prior. You will find the links to the studies that were discussed today in the transcript of this episode. Finally, if you value the insights that you hear on the ASCO Daily News Podcast, please take the time to rate, review, and subscribe wherever you get your podcast.  Disclaimer:  The purpose of this podcast is to educate and to inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions. Guests on this podcast express their own opinions, experience, and conclusions. Guests' statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.   Follow today’s speakers:   Dr. Diwakar Davar  Dr. Ben Boursi   Follow ASCO on social media:   @ASCO on Twitter    ASCO on Facebook    ASCO on LinkedIn      Disclosures:   Dr. Diwakar Davar:     Honoraria: Merck, Tesaro, Array BioPharma, Immunocore, Instil Bio, Vedanta Biosciences    Consulting or Advisory Role: Instil Bio, Vedanta Biosciences    Consulting or Advisory Role (Immediate family member): Shionogi    Research Funding: Merck, Checkmate Pharmaceuticals, CellSight Technologies, GSK, Merck, Arvus Biosciences, Arcus Biosciences    Research Funding (Inst.): Zucero Therapeutics    Patents, Royalties, Other Intellectual Property: Application No.: 63/124,231 Title: COMPOSITIONS AND METHODS FOR TREATING CANCER Applicant: University of Pittsburgh–Of the Commonwealth System of Higher Education Inventors: Diwakar Davar Filing Date: December 11, 2020 Country: United States MCC Reference: 10504-059PV1 Your Reference: 05545; and Application No.: 63/208,719 Enteric Microbiotype Signatures of Immune-related Adverse Events and Response in Relation to Anti-PD-1 Immunotherapy       Dr. Ben Boursi: No relationships to disclose.