CHICAGO — Supplementation with fiber to stimulate short-chain fatty acid (SCFA) production in patients with rheumatoid arthritis (RA), or direct supplementation with the SCFA butyrate, appears to enhance patients’ response to methotrexate, according to three studies presented at American College of Rheumatology (ACR) 2025 Annual Meeting. Further, it may be possible to predict patient response to methotrexate based on gut microbial signatures, including presence of bacteria that produce butyrate, according to one of the studies, presented by Vinod K. Gupta, PhD, of the Mayo Clinic in Rochester, Minnesota. The results may partly help explain why about 20%-50% of patients with RA do not respond to methotrexate.

“The studies continue to demonstrate a role for the intestinal microbiome in methotrexate responsiveness,” Kristi Kuhn, MD, PhD, a professor of medicine at the University of Colorado Anschutz Medical Campus in Aurora, Colorado told Medscape Medical News. Although she was not involved in any of the studies, Kuhn has also been studying the relationship between gut microbiome and methotrexate response.

Data from Gupta’s study “suggest that there may be generalizable microbial signatures across geography to predict responders, allowing us to move closer to precision medicine,” while the findings from the two posters “further suggest that an underlying diet can modulate the responsiveness to methotrexate,” Kuhn said.

That said, data are insufficient to make specific recommendations on supplementations, Kuhn added, but the evidence base continues “to support eating a healthy diet rich in fibers — fruits, vegetables, whole grains — and sparing of processed foods. Thus, the Mediterranean diet continues to be our best recommendation.”

Gupta’s study involved 60 participants with early RA before any treatment. They were an average 51 years old, had an average BMI of 29, and included 72% women. The participants had an average Disease Activity Score in 28 joints based on C-reactive protein (DAS28-CRP) of 4.41 at baseline, and two thirds were positive for anti-citrullinated protein antibodies.

All the participants started 15 mg oral methotrexate and provided stool samples at baseline and after 3-3.5 months of treatment to undergo whole genome shotgun metagenomic sequencing. There were 37 responders and 23 nonresponders based on European Alliance of Associations for Rheumatology (EULAR) response criteria — a decrease of > 1.2 on the DAS28-CRP or a decrease of > 0.6 on the DAS28-CRP with a follow-up DAS28-CRP ≤ 5.1.

While no significant differences in alpha- and beta-diversity emerged between responders and nonresponders, significant differences did exist between the groups in prevalence of some of the five bacteria species the researchers examined. Responders had significantly more abundant Faecalibacillus intestinalis (P = .029) and had significantly less abundance of Erysipelatoclostridium ramosum (P = .023), which previous research has linked to pro-inflammatory phenotypes, suggesting the potential for impaired methotrexate efficacy. Responders had marginally less Clostridia bacterium (P = .048) and Eisenbergiella tayi (P = .049).

Further, while nonresponders had substantially greater prevalence of Eubacterium sp AF16-48, responders had much greater prevalence of Bacteroides fragilis. Gupta noted past findings showing that enrichment of Bacteroides species are linked to methotrexate responses in RA, and referenced another study showing no therapeutic effect from methotrexate in mice without B fragilis, which produce the SCFA butyrate. That latter study, however, showed that butyrate supplementation restored a methotrexate response in gut microbiota-deficient mice that was on par with the response seen when giving those mice B fragilis.

Gupta hypothesized that part of the mechanism driving the interaction between different gut microbial species and methotrexate response relates to the genes for the enzyme dihydrofolate reductase (DHFR), which was significantly less abundant in responders (P = .031). Lachnospiraceae bacterium is a microbial source of DHFR, and methotrexate can bind with DHFR from both human and bacterial sources.

“Higher abundance of microbiome-derived DHFR may reduce the bioavailability of methotrexate and lead to nonresponse,” Gupta said.

Gupta’s team then explored whether gut microbiome signatures could predict methotrexate response using machine learning with the internal leave-one-out cross validation technique and identified a model with an area under the curve of 0.83, a sensitivity of 89.2%, a specificity of 65.2%, and an overall accuracy of 80%. External validation on a different cohort provided an area under the curve of 0.77.

“Our machine-learning model suggests the potential of gut microbiome profiling as a personalized predictive tool for methotrexate outcomes,” Gupta said.

The association they identified between butyrate-producing B fragilis abundance and methotrexate response was concordant with the other two posters assessing two types of supplementation. The data from those posters “indicate that diet could help form a methotrexate-sensitive diet, although their work has yet to show the dietary effects on the microbiome that led to increased methotrexate-sensitivity,” Kuhn said. “Patients are interested in how their diet could improve disease control,” she continued. “Research such as these studies are helpful in addressing a large unmet need for patients: rigorous and validated data informing on how the diet can affect disease control.”

Role of SCFA Supplementation

In one of the posters, a proof-of-principle study involving participants with RA who took methotrexate and butyrate supplements, researchers found that those who had higher fecal concentration of SCFAs also had greater improvement in disease activity — but only if they were consuming high-quality diets, defined as consuming at least one component of a healthy diet, such as the Mediterranean diet, a vegetarian diet, or one high in fish and other seafood (or taking omega-3 supplements).

Kaci French, MD, and her colleagues at NYU Langone Health, New York City, noted in explaining the rationale for their study that past research has suggested omega-3 fatty acids may modulate inflammatory cytokines in RA while fiber-rich diets like Mediterranean and vegetarian diets have been linked to proliferation of bacteria that produce SCFAs, including butyrate.

“In turn, short-chain fatty acids have been shown to modulate the gut microbiome, enhance anti-inflammatory mediators in murine RA models, and be associated with decreased risk of progression to clinical RA in at-risk individuals,” they wrote. But because most studies on SCFA supplementation have been in animal models, they wanted to see how diet and butyrate supplementation interacted with treatment in patients with new-onset RA.

The study included eight patients taking only methotrexate and 17 taking methotrexate as well as 1 g of butyrate three times daily for 4 months. The average methotrexate dose was 15 mg. The participants were mostly White women (80%), including 40% who were Hispanic individuals, and they were an average 53 years old. They completed food frequency questionnaires at 1, 2, and 4 months, when clinical data and fecal samples were also collected. Methotrexate responders were classified as participants who entered remission, defined as a DAS28 < 2.6, or improved > 1.8 from baseline on the DAS28 based on erythrocyte sedimentation rate at 4 months.

Only three (37.5%) of the participants taking only methotrexate were responders. Meanwhile, data from the 11 responders and six nonresponders taking butyrate supplements revealed a positive correlation between improved DAS28 scores and increased fecal SCFA concentration that was statistically significant at 1 month (P = .001).

The researchers also observed in 12 patients that fecal acetate was elevated in those with higher-quality diets but decreased in those with the lowest quality diets (P = .013). Participants who ate better diets were significantly more likely to respond to methotrexate (P = .028) but only if they were also taking butyrate supplements. Diet quality did not make a difference in those taking methotrexate alone.

In addition to reinforcing the link between SCFAs and disease activity, “our results suggest that the change in short-chain fatty acids after butyrate supplementation may be a potential marker of methotrexate response in [patients with new-onset RA] and that butyrate supplementation may act in concert with high-quality diet to increase methotrexate efficacy,” they concluded.

Supplementing With Fiber Instead of Butyrate

A similar study conducted by a team led by Claire Immediato Daien, MD, PhD, of the University Hospital of Montpellier and the University of Montpellier, Montpellier, France, relied on the same understanding of fiber’s role in stimulating gut bacteria production of SCFAs like butyrate, but instead of testing butyrate supplementation, it examined the effect of dietary fiber supplementation. This slightly larger study, named SUPERFIBRES, was a randomized controlled trial with 49 participants who had active RA despite stable management with conventional synthetic disease-modifying antirheumatic drugs, so it differed in assessing patients with established RA instead of new-onset RA. The population was again predominantly women (85.7%) but a little older, an average 62 years old, with a mean DAS28 of 3.86 at baseline.

For 30 days, 26 participants took a daily supplement containing the prebiotic fiber inulin while the other 23 received a daily placebo. The groups were statistically similar in demographics and clinical characteristics at baseline, including a BMI of 24.5 in the fiber group and 26.6 in the placebo group.

A EULAR response occurred in 14 participants receiving daily inulin compared with five receiving placebo, resulting in nearly a fivefold higher likelihood of response after adjustment for compliance, baseline DAS28, age, BMI, and sex (odds ratio, 4.65; P = .03). There was no significant difference between the groups in rates of DAS28 low disease activity or DAS28 remission based solely on fiber supplementation, but similar to what French and colleagues found, a difference did emerge in those taking methotrexate. Participants taking methotrexate with fiber supplements saw a greater improvement on DAS28 (-1.0) than those taking methotrexate without a fiber supplement (-0.34; P = .01). No such effect appeared in patients not taking methotrexate.

The researchers also looked at circulating T helper 17 (Th17) cells and the Th17/Treg ratio, which contributes to RA. They saw a significant reduction in both measures in participants taking fiber supplements (P = .02) but no change in those receiving placebo. The researchers surmised that the improvement in EULAR response and Th17/Treg balance was driven by the enhanced efficacy of methotrexate in those taking inulin. “These findings establish the gut-immune axis as a novel therapeutic target and identify dietary fiber as a safe, inexpensive adjunct to methotrexate in rheumatoid arthritis,” they wrote.

None of the authors of the butyrate supplementation study had any disclosures, and that study was funded by the National Center for Advancing Translational Sciences, the National Institutes of Health, and an Arthritis Foundation Pilot Award. The University of Montpellier sponsored the SUPERFIBRES trial. Daien reported financial relationships with AbbVie/Abbott, Amgen, Bristol-Myers Squibb, Eli Lilly, Janssen, Merck, Novartis, Pfizer, and UCB. The gut microbiome signatures study was funded by the Mark E. and Mary A. Davis Initiative in Rheumatoid Arthritis Research and various Mayo Clinic entities; the authors did not note any disclosures. Kuhn reported consulting for Solarea Bio, UCB, and Spyre.

Tara Haelle is a science/health journalist based in Dallas.