Hassane M. Zarour, MD, discussed the research surrounding manipulation of the gut microbiome to improve response to immunotherapy in melanoma and other tumor types.
Hassane M. Zarour, MD
A new area of research in the treatment landscape of melanoma focuses on manipulating the gut microbiome to improve immunotherapy responses, particularly with antiPD-1 and anti–CTLA-4 antibodies.
Based on data seen in several mouse studies in Chicago and France, it is suggested that altering the microbiome may be necessary to promote responses to checkpoint inhibition. While the data are still new, several groups are looking at ways to further investigate this idea in different patient populations.
While this approach is being evaluated in non­small cell lung cancer and renal cancers, a group at the University of Pittsburgh, led by Hassane M. Zarour, MD, has designed a trial involving a subset of patients with melanoma. In the trial, patients will undergo a fecal microbiota transplant.
Prior to this trial, Zarour and his group collected stool samples from patients to identify cells commonly found in responders versus non-responders who received an antiPD-1 antibody for the treatment of melanoma.
In an interview withTargeted OncologyduringtheCRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference,Zarour, professor of Medicine and Immunology at the University of Pittsburgh, discussed the research surrounding manipulation of the gut microbiome.
TARGETED ONCOLOGY:Your presentation is on manipulating the gut microbiome to improve immunotherapy response in melanoma. Can you give me an overview of your talk?
Zarour:This is kind of a new area of research in the field of cancer immunology, and also for me. It really stems from preclinical findings that were made 3 years ago by a French group. Data has shown that in a mouse model, response to the new therapy of cancer, the so-called immune checkpoint blockade, can be regulated by the gut microbiota. In some [patients], it would promote response, and maybe in some others, it would impede response.
There was a paper that was published 3 years ago that identified some common cells that drive response for CTLA-4, in the case of the French group, or that drive response to PD-1 blockade, in the case of the Chicago group. I think this was very interesting, as we have always had a focus on the tumor intrinsic mechanism that deepens the response or what happens in the tumor microenvironment. This puts an emphasis on the tumor extrinsic outside the tumor microenvironment – which can have a strong effect.
When you see that many of the gut microbiota had an interplay with the immune system, and they would manipulate a response to potent immunotherapy to cancer, this is very fascinating. This is an angle we didn’t have until these papers came out. At this time, there’s many in the field that find it very intriguing to find out if this is true in humans.
At this time, at the University of Pittsburgh, we decided to collect stool. We are already banking blood and tumors, and now we are banking stool, so it is not easy to start with.
In patients, particularly with immune checkpoint blockade, and mainly the antiPD-1 antibody, we collect stool from both pre- and post-treatment to try and evaluate the gut microbiota in these patients and find if there is any correlation with response. We did that in collaboration with Giorgio Trinchieri, MD, of the National Institutes of Health, he’s a famous immunologist. He was assembled a team, and it was a great collaborative effort between 2 different institutions.
A number of groups have published some of the data about that, both the French group in nonsmall cell lung cancer and renal cancer, a group with Jennifer Wargo, MD, MMSC, of The University of Texas MD Anderson Cancer Center, on a small subset of melanoma patients, and again, Thomas Gajewski, MD, PhD, at the University of Chicago Medicine, in a small subset of melanoma patients. The published data look at the common cells that were highly present in the stool of responders versus non-responders, identifying a number of species which were going in the same direction as what was found in mice. However, there are a number of questions raised by this study, including that there are not a lot of commonalities in the studies. This is a question that we need to answer.
Coming back to our data, we not only needed to show correlation between responder and non-responder, but we also tried to compare our data with their data. The lack of commonality, what does it mean? In terms of interpreting these data, [we need to identify] the role of the many environmental factors that may impact the gut microbiome, like diet, treatment, genetics, or so many other factors that may play a role there before coming to any meaningful conclusion.
What should we do in terms of further pushing this area of investigation? In addition to metagenomic sequencing, [we will need to be] collecting all this useful information [such as] diet, treatment, and so on, experimenting in mice, and trying to find what is important. What is important may not be the most evident, but a group of cells may have evidence for the function of doing this. This is a real complex target and a very challenging question.
TARGETED ONCOLOGY:What is the unmet need you’re trying to address with this research?
Zarour:There are a lot of things missing. Basically, the PD-1 antibody has really been a game changer. We are at [a point where] 30% to 40% of melanoma patients respond to this. However, some still do not respond, so the combination of antiPD-1 and anti–CTLA-4 could be better. We don’t know in the long-term if the response will be any better, so we still need to wait, but the majority of patients need alternative treatment.
I think that now with a better understanding of the mechanism that impedes tumor T cell response in the tumor microenvironment, you can see if this is really adding anything that could play a role in the tumor microenvironment. This microbiome story could open the door to antibiotics, probiotics, diet intervention, etc. once we know what we want to achieve. You can see that already some groups are moving in some [of these] directions in terms of what we see based on these data.
We already have designed a trial which is a fecal microbiota transplant. The reason why we did that is 2-fold. The first is that at that time we didn’t think we had enough data to select a common group of cells. With the data I’m going to present going in the same direction, we still don’t know what to give, so it’s too soon to do that.
We had the idea of doing the fecal microbiota transplant. As you know, this has been done already in the clinic for people with some infection, such as colitis. Most of those patients are cured with 1 single administration of a fecal microbiota transplant from a normal donor. This is safe, and it can be implemented easily in the clinic. The main question is, what donor are you going to pick? We agree that this is a difficult question, so we have decided to take from a long-term complete responder who did extremely well for a long time after receiving an anti-PD-1 antibody. This was the first criteria. The second, which we thought was also important, was we initially hoped the meta-analytic data could really define our group of common cells that are responsible for causing therapy response and can refine the donor with this information, but the data does not go in this direction, so for now it is clinical outcome. Once we know better about the group of bacteria that are linked to response, maybe we can refine the selection of donors at this point.
TARGETED ONCOLOGY:What would you say is your main message about this topic?
Zarour:There are multiple things. The first, and what I think is the most important, is that giving antibiotics to patients who are receiving immunotherapy may not be as [effective]. You need to really think twice about if a patient needs antibiotics because you may no longer be deleterious to a specific therapy, so you need to be extra careful. I think that’s the main thing. In terms of diet, it is too soon to say, but in the future, we may have some data which is important. Hopefully that is more information that will come out in these studies.