Gut Microbiome 'May Hold Key to Immunotherapy Outcomes'

Liam Davenport

June 26, 2018

BARCELONA, Spain — The microbial environment in the intestines, aka the microbiome, not only may affect an individual's risk of developing and surviving colorectal cancer but may also have a dramatic impact on the efficacy of anticancer immunotherapies, an issue that could be tackled with advanced probiotics, suggests a leading expert.

Presenting a summary of the latest evidence here at the 2018 World Congress on Gastrointestinal Cancer (WCGC), Herbert Tilg, MD, professor of internal medicine at Medical University Innsbruck, Austria, told the audience that the gut microbiome "is an exciting area of research" and one that is "very dynamic."

He said that the evidence shows that "inflammation drives cancer and this sequence might also involve certain bacteria," adding: "It is very clear now, and we have enough data in hand, to show that colorectal cancer is associated with profound alterations in the gut microbiota."

Tilg added that it is also "very evident now, and this is a major link of the gastrointestinal tract to probably many cancers throughout the body, that the gut microbiome affects the efficacy of systemic treatment with immune checkpoint inhibitors."

Reacting to the talk here at the meeting, Eric van Cutsem, MD, PhD, from University Hospital Gasthuisberg, Leuven, Belgium, and co-chair of the Congress, said that while it is still "very early days," the data that Tilg presented are "very relevant" and that the microbiome "is going to be important for the future."

He told Medscape Medical News that although the evidence suggests the microbiome is linked to the sidedness of colorectal cancer and has a prognostic effect, "what we cannot yet link is to see whether that is the only explanation."

He nevertheless believes that "it is going to be part of the puzzle."

In terms of the impact of the microbiome on outcomes with immunotherapy, van Cutsem pointed out, "Of course, responses to the different immunotherapy agents are always multifactorial."

"But having seen some of the preclinical data presented here and some other data, there is probably an important role for the microbiome...not only in the response to treatment but also in carcinogenesis."

Half a Kilo of Bacteria

Tilg began his presentation by pointing out that the microbiota in the human gut consists of 100 trillion bacteria, with a biomass of half a kilogram.

Within the intestine, the microbiota is segregated and compartmentalized, and its composition is relatively stable over time, with differences seen primarily at the species and family level.

Most of the microbiota is bacterial, with over 500 species cultured so far, although Tilg pointed out that more than 50% of the bacteria identified in stool samples are unculturable.

Nevertheless, the microbiota has an "astonishing" degree of specificity to each individual, with its make-up so specific as to be characterizable as a microbial fingerprint.

Despite this, changes in diet can lead to rapid alterations in the composition of the microbiota, with an animal-based diet associated with overgrowth of bile-resistant bacterial strains, such as Aristides, Bilophila, and Bacteroides.

This, Tilg said, underlines the link between the consumption of animal fats, increases in bile acids, and the propagation of pro-inflammatory species.

Looking at the interplay the inflammation, the microbiome, and gastrointestinal cancer, he pointed to the protein lipocalin 2, which prevents bacterial iron acquisition, plays a critical role in neutrophil function, modulates macrophage biology, and regulates cell death, among other functions.

To examine the role of lipocalin 2 in inflammatory bowel disease, Moschen et al studied a mouse model of colitis, finding that the mice that produced the protein remained healthy while those that did not developed massive inflammation.

Moreover, the mice lacking lipocalin 2 spontaneously developed right-sided colorectal tumors, which were responsive to antibiotic therapy.

The mice also had an altered microbial ecology in their intestines. Specifically, Alistipes and Robinsoniella species were overrepresented; these two species evoked colitis and right-sided tumors when orally transmitted to other mice.

Colorectal Cancer

Taking these observations further, a review by Brennan et al showed that each stage in the development of colorectal cancer is accompanied not only by hyperplasia and immune cell infiltration but also by microbial proliferation in the tumor environment.

This includes increases in the prevalence of Streptococcus gallolyticus, colibactin-producing (pks+) Escherichia coli, enterotoxigenic Bacteroides fragilis (EBTF), and Fusobacterium nucleatum.

Dejea et al also demonstrated that the microbiota is organized differently between different forms of colorectal cancer, with right-sided cancers having a greater biofilm depth and diversity than tumors in other parts of the colon.

In addition, patients with familial adenomatous polyposis have been found to have colon biofilm that harbors bacteria with tumorigenic activity, including pks+ E coli and EBTF.

Mouse models of colorectal cancer revealed that co-colonization by those two bacteria is associated with significant increases in colon tumor onset compared with colonization by either alone, as well as with a substantial increase in mortality.

Seeking to translate these findings into humans, Feng et al studied 55 healthy controls, 42 patients with advanced adenoma, and 41 patients with colorectal cancer, identifying an increase in various Bacteroides species and E coli in colorectal cancer, with several also correlated with levels of the inflammatory C-reactive protein.

The research also showed that a high intake of red meat relative to fruit and vegetables was linked to the outgrowth of bacteria potentially associated with a "more hostile" intestinal environment.

Furthermore, an attempt by Yu et al to identify noninvasive biomarkers for colorectal cancer from the fecal microbiome in patients and controls identified two novel bacterial strains associated with the disease, as well as 20 gene markers that distinguished between patients and controls.

In another study, Nakatsu and colleagues found that it was not only the diversity of the intestinal bacterial population that increased with colorectal cancer; in addition, 22 viral taxa were altered between patients and controls.

They also showed that virome dysbiosis was associated with both early- and late-stage colorectal cancer and that a combination of four taxonomic markers was linked to reduced survival.


Another area where the impact of the micobiota has been felt is the use of immunotherapy for cancers.

While drugs such as ipilumumab that target cytotoxic T-lymphocyte antigen 4 (CTLA-4) have achieved remarkable successes in recent years, Vétizou et al showed that CTLA-4 blockade had no impact on tumor growth in antibiotic-treated or germ-free mice.

More pertinently, they demonstrated that immunotherapy with CLTA-4 blockers induced intestinal enrichment of several Bacteroides species, which Tilg said underlines the "key role" of these bacteria in the immunomodulatory effects of these drugs.

Going further, he asked whether the microbiome can be used to predict responses to immunotherapy.

Recent data by Routy et al published this year revealed that the efficacy of anti–programmed cell death 1 (PD-1) immunotherapy was compromised in both mouse models and patients with cancer by the use of antibiotics, with worsened survival.

Crucially, metagenomic analysis of fecal samples predicted responses to PD-1 immunotherapy at 3 months, with numerous different bacterial species differentially enriched between responders and nonresponders.

The results also showed a significant stepwise decrease in the percentage of patients with Akkermansia bacteria in their microbiome from partial responders to patients with stable disease to those with progressive disease.

These findings were underlined by Gopalakrishnan et al, who demonstrated that the gut microbiome modulated the response to anti–PD-1 immunotherapy in patients with melanoma, with a high abundance of Faecalibacterium prausnitzii associated with better treatment responses and survival.

In another 2018 study, Matson et al found a significant association between commensal microbial composition and clinical response, with bacterial species such as Bifidobacterium longum, Collinsella aerofaciens, and Enterococcus faecium more abundant in responders.

Tilg said that the message from these three papers is that "the gut microbiota obviously, crucially, affects responses to immunotherapy."

The gut microbiota obviously, crucially, affects responses to immunotherapy. Dr Herbert Tilg


He added that this "axis" between the gut microbiota and antitumor activity "obviously must be of major relevance, and of course the conclusion might be that 'designed' probiotics using Akkemansia could be potentially relevant in cancer treatment".

van Cutsem declares grant/research support from Amgen, Boehringer Ingelheim, Celgene, Ibsen, Lilly, Merck, MSD, Novartis, Roche, and Servier; consulting for Bristol-Meyers Squibb; and speakers' bureau for Bayer, Bristol-Meyers Squibb, Celgene, Lilly, Novartis, and Servier. Tilg has disclosed no relevant financial relationships.

2018 World Congress on Gastrointestinal Cancer (WCGC). Abstract LBA-001. Presented June 20, 2018.

For more from Medscape Oncology, follow us on Twitter:  @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.