The gut microbome may influence responses to immune checkpoint inhibitors, based on results from two studies, and one of the investigators is now gearing up for the next step - evaluating in a clinical trial whether altering the microflora will actually improve responses.
In the first study, investigators carried out a series of experiments using fecal microbiome samples from patients with metastatic melanoma embarking on therapy with a PD-1 (programmed cell death protein 1) inhibitor.
“In melanoma patients, there were differential signals in the gut microbiome of responders versus nonresponders, and I think the clincher was when we transplanted fecal samples from responders to nonresponders in germ-free mice, essentially reconstituting the microbiome and showing that it equally affected the systemic immunity and antitumor immunity when we implanted tumors, as well as response to checkpoint blockade,” lead author , MD, MMSc, of the University of Texas MD Anderson Cancer Center in Houston, said in an interview.
Dr. Wargo and her colleagues first collected buccal and fecal microbiome samples from 112 patients with metastatic melanoma before they began therapy with a PD-1 inhibitor. After performing taxonomic profiling on all samples, they found that there was a clustering effect by response status in the gut microbiome, but not the oral microbiome, and because changes in the oral microbiome did not appear to be related to treatment response, they focused on the gut.
When Dr. Wargo and her colleagues studied the posttherapy microbiomes of 43 patients (30 responders and 13 nonresponders) according to Response Evaluation Criteria in Solid Tumors (RECIST 1.1), they found that the responders had a significantly higher degree of alpha diversity, a measure of species diversity within a specific environment, compared with nonresponders (P less than .01). In addition, responders had a relative abundance of Ruminococcaceae, commonly occurring gut microbes that break down complex carbohydrates, the investigators reported (Science. 2017 Nov. 2. ).
They found that patients whose microbiomes were diverse in general, and in particular were enriched with Faecalibacterium and Clostridiales species, were more likely to respond to immunotherapy with a PD-1 inhibitor and have a longer duration of progression-free survival. In contrast, patients whose microbiomes were more enriched with Bacteroidales species were more likely to be nonresponders.
To get a better understanding of the mechanisms whereby gut bacteria may influence response to PD-1 inhibitors, they performed metagenomic analysis on samples from 14 responders and 11 nonresponders, and found that responders had micro-organisms predominantly associated with anabolic functions that may support host immunity, whereas nonresponders had microbiomes where catabolic functions were more common.
The investigators next performed immune profiling, and found that both systemic immunity and local immunity in the tumor microenvironment in responders were associated with the aforementioned favorable gut microbiome.
The researchers then transplanted feces from the human donors into germ-free mice and then injected tumor cells into the mice, and found that tumor growth was significantly reduced, and response to PD-1 inhibition was significantly enhanced, in mice who received feces from responders.
“An obvious next step is to run a clinical trial to test the hypothesis that by modulating the microbiome, you can actually enhance responses to therapy,” Dr. Wargo said. Details of the clinical trial are still being worked out, but will likely involve fecal transfers and other mechanisms for modulating the microbiome in hopes of improving responses to PD-1 inhibitors.
“It’s going to be a very biomarker-heavy trial,” she said. “We’re going to look, certainly, for changes in the microbiome, and will also do a lot of profiling in the blood, the tumor, and in the microbiome to see if there are changes that occur by modulating that microbiome. Then of course we’ll look for differences in response rates in patients as well.”
Bacteria also affect epithelial cancers
In a separate study, also published in Science, investigators led by Bertrand Routy, MD, of the Gustave Roussy Cancer Institute in Villejuif, France, reported that patients with non–small cell lung cancer and urothelial carcinoma who had previously used systemic antibiotics had reduced survival when treated with a PD-1 inhibitor, compared with patients who had never taken antibiotics (Science. 2017 Nov. 2.
Analysis of the gut microbiome in these patients showed that higher levels of Akkermansia muciniphila were associated with the best clinical outcomes, with the species detectable in the microbiome of 69% of patients who had partial responses to anti–PD-1 therapy, and in 58% of those with stable disease. In contrast, the bacterium was detectable in only 34% of patients who experienced disease progression.
As in the experiments by Dr. Wargo and her associates, when the French investigators first treated mice with antibiotics and then gave them oral supplements containing the bacteria, the supplements restored response to PD-1 blockade,
“We conclude from the study that the gut microbiome markedly influences the outcome of PD-1 blockade in mice and patients,” Dr. Routy and his associates wrote.
They acknowledged that the mechanism whereby a common organism such as Akkermansia muciniphila might have an immunomodulatory effect is still unknown,
“Irrespective of these remaining questions, our findings suggest that the microbiome governs the cancer-immune set point of cancer-bearing individuals and offer[s] novel avenues for manipulating the gut ecosystem to circumvent primary resistance to [immune checkpoint inhibitors],” they wrote.
The study by Dr. Wargo and her colleagues was supported by contributions to the University of Texas MD Anderson Melanoma Moon Shots program. Dr. Wargo is supported by the Binational Science Foundation, Melanoma Research Alliance, Stand Up to Cancer, and the MDACC Melanoma Moon Shots Program. The work by Dr. Routy and his associates was supported by the Goustave Roussy Cancer Institute and McGill University. Coauthors were supported by the National Cancer Institute of France and other agencies and philanthropies.