Home/Issues/Issue 05/Study

Mouse experiments, alongside a separate observational cohort of people Publication date verified at source

Dietary intervention through bacterial-derived butyrate elicits anti-tumor activity and increases anti-PD-1 response

AI narration, generated on first listen
Journal
Gut Microbes
Authors
Benlaifaoui M
Institution
Centre de Recherche du CHUM, Montreal
Published
15 July 2026
Source
PMID 42454784 · DOI 10.1080/19490976.2026.2699457
Design
Controlled mouse tumour experiments with butyrate substitution, anti-PD-1 combination and CD8 dependency testing, plus a separate observational cohort. Tumour bearing mice, and adults with advanced non small cell lung cancer on anti-PD-1.
Sample
Mice for every causal experiment, plus a separate observational group of 117 adults with advanced non small cell lung cancer receiving anti-PD-1.

What butyrate is

Butyrate is a short chain fatty acid that gut bacteria produce when they ferment certain fibres, including inulin. It feeds the cells lining the colon and can also influence immune cells elsewhere in the body.

Why they ran it

Most people with non small cell lung cancer eventually progress on immune checkpoint inhibitors, and gut bacteria are now recognised as a major determinant of how well those drugs work, with diet the most direct and scalable way to reshape the gut microbiome.

The authors state that how fibre intake modulates the gut microbiome, and more importantly the response to checkpoint inhibitors in patients with this cancer, remains unknown, and they note that oncology guidelines currently say there is no evidence to recommend any diet during active treatment.

Drawn from the paper's introduction.

The causal work here was done entirely in mice. Mice were put on one of three diets for two weeks, a standard chow, a low fermentable diet built on 10 percent cellulose, or a high fermentable diet built on 10 percent inulin, then given tumour cells and treated with either anti-PD-1 or a control antibody. The team tracked tumour growth, sequenced the mice's gut bacteria, measured short chain fatty acids in mouse stool, and profiled immune cells inside the tumours. A separate arm surveyed the diets of 117 people with advanced non small cell lung cancer.

In the mice, inulin slowed tumour growth compared with both the standard and cellulose diets, and it boosted anti-PD-1 activity on top of that, across groups of 10 to 20 mice each. Inulin fed mice had more acetate, butyrate and propionate in their stool than cellulose fed mice, and more CD8 T cells inside their tumours. Butyrate given on its own shrank tumours, and stripping out CD8 T cells abolished that effect while stripping out CD4 T cells did not.

The human side did not follow. Among 117 patients on anti-PD-1, median age 72 with 14 months of follow up, fibre intake was not associated with overall survival, hazard ratio 1.01 with an interval of 0.93 to 1.09, or with progression free survival, 1.01 with an interval of 0.95 to 1.08. Only PD-L1 expression predicted outcome. Median intake was 15.6 grams a day, ranging from 7.1 to 31, leaving most of the group below the fibre thresholds used in earlier immunotherapy studies. Fibre did track with more butyrate producing bacteria, but not with living longer.

The numbers

undefinedundefined
undefinedundefined
undefinedundefined
undefinedundefined
undefinedundefined

Why this might happen

Shown in this study The study measured this step directly.

In mice, the authors report that an inulin enriched diet reshapes the gut microbiome and raises faecal short chain fatty acids, while butyrate given on its own acts further downstream, producing anti tumour effects without major changes in microbiome composition.

Also in mice, butyrate improved the checkpoint inhibitor response in a CD8 T cell dependent way, with transcriptional remodelling of tumour infiltrating CD8 T cells covering proliferation, trafficking and metabolic pathways.

The authors state plainly that these mechanisms still require further validation, and that their data cannot determine where the CCR9 positive CXCR3 positive T cells came from.

Drawn from the paper's discussion and results, in mice unless stated otherwise.

What this does not show

  • Fibre intake did not help the patients in this study. In the 117 person cohort the adjusted hazard ratio for fibre intake was 1.01 for overall survival and 1.01 for progression free survival, with both confidence intervals sitting across 1.0.
  • The tumour shrinkage happened in mice, not humans. Inulin and butyrate were tested in three mouse tumour models. No person in this paper was given inulin or butyrate.
  • The survival models did not include everyone. The adjusted models were run on complete cases only, leaving 105 patients for overall survival and 106 for progression free survival out of the 117 surveyed.
  • Most of the cohort ate little fibre to begin with. Median intake was 15.6 grams a day, and 59 to 76 percent of patients fell below the thresholds used in earlier work, which the authors say may have limited their ability to detect any association.

Where this leaves us

In mice, it is now more likely that a fermentable fibre diet and bacterial butyrate each slow tumour growth and add to anti PD-1 activity, and that the butyrate effect runs through CD8 T cells. That extends earlier preclinical fibre work rather than contradicting it. In humans nothing causal moved at all, and the observational part refines expectations downward, because fibre intake did not track survival in this cohort.

The causal part is mice, plainly and only mice: female C57BL/6 mice carrying implanted tumours, plus some in vitro work on human T cells from donors. The human part is a separate observational group of 117 adults with advanced non small cell lung cancer receiving anti PD-1, and it is correlational, with no diet assigned and no outcome benefit found.

The authors have already started a randomised clinical trial in patients with non small cell lung cancer on checkpoint inhibitors, comparing standard therapy with a dietitian supervised nutrition intervention that aims to raise total fibre intake above 25 g per day.

Caveats worth holding

  • Every causal claim in this paper rests on mice, and mouse tumour models have a long record of results that do not carry over to people.
  • The human portion is observational, so even the microbiome associations it did find cannot establish cause or direction.
  • Diet in the human cohort came from a 47 item food frequency questionnaire, which captures habits roughly rather than precisely.
  • The authors note that no clinical trial has yet tested butyrate supplementation alongside checkpoint immunotherapy in people.

Newsletter

Each issue by email, when the newsletter launches. Leaving your address puts you on the list, nothing is sent yet.