RA treatment responders show unique differences in gut microbiome



The gut microbiome, previously shown to have an association with rheumatoid arthritis, may also provide signals of a patient’s disease prognosis, researchers at the Mayo Clinic have reported.

Dr. Jaeyun Sung, a computational biologist with Mayo Clinic’s Center for Individualized Medicine in Rochester, Minn.

Dr. Jaeyun Sung

“We found that the gut microbiome is linked to whether patients with RA improve in their clinical symptoms or not,” cosenior author Jaeyun Sung, PhD, said in an interview. “We found features of the gut microbiome that linked to improvement, and we also put those features in a machine-learning model that can actually predict improvement at a follow-up visit.” Dr. Sung is a computational biologist with Mayo Clinic’s Center for Individualized Medicine in Rochester, Minn.

The retrospective, observational cohort study included 32 patients diagnosed with RA between 1988 and 2014. The researchers performed meta-genome shotgun sequencing on 64 stool samples kept in a biobank and collected at two separate visits 6-12 months apart. Dr. Sung and colleagues observed significantly different microbiome traits between patients who eventually showed minimally clinically important improvement and those who didn’t.

Computer illustration of the human digestive system and a close-up view of bacteria found in the intestine. KATERYNA KON/Science SourceSource

Computer illustration of the human digestive system and a close-up view of bacteria found in the intestine.

The study also provided a proof of concept for using machine-learning technology to analyze the gut microbiome to predict the course of the disease, Dr. Sung said.

Cosenior author John M. Davis III, MD, a clinical rheumatologist and rheumatology research chair of the Mayo Clinic, noted that their own previous study had confirmed dysbiosis in people with RA when compared with controls. “We had some preliminary insight that it may be linked to some extent to the disease state and maybe treatments,” Dr. Davis said. “So that led us to hypothesize that there may be an association between the gut microbiome and response to treatment or disease activity over time.”

Dr. John M. Davcis III, a clinical rheumatologist and rheumatology research chair of Mayo Clinic, Rochester, Minn.

Dr. John M. Davis III

The study found that age was the dominant factor in determining variations in the gut microbiome composition, but the next prevailing factor was minimum clinically important improvement status, which 12 of the 32 study participants achieved at their follow-up visits. At baseline, all patients were on some type of treatment – either biologic or conventional disease-modifying antirheumatic drugs (DMARDs, 46.9% and 87.5%, respectively), or prednisone (46.9%).

Gut microbiome composition

The patients who achieved minimum clinically important improvement had an average decline in Clinical Disease Activity Index of 16.7 units (standard deviation, 12.8) versus a gain of 5.7 (SD, 8.9) in the remaining patients. The study found higher species-level alpha diversity and richness and higher beta diversity in the group that achieved minimum clinically important improvement, compared with those who did not.

They identified six microbial taxa as higher in abundance in the improved patients: Negativicutes (class); Selenomonadales (order); Prevotellaceae (family); Coprococcus (genus); Bacteroides sp. 3_1_19 (species); and Bilophila sp. 4_1_30 (species). In the patients who showed no improvement, Eubacterium sp. 3_1_31 (species) was found to be higher (P < .05). They also found 15 metabolic pathways that were differently abundant between the two groups at baseline.

Two things make this study different from other studies of the gut microbiome in RA, Dr. Sung said: It didn’t have a control group, only RA patients, and it didn’t evaluate a specific drug in RA patients.

“We’re thinking beyond just drug or treatment, independent of prior treatment, independent of prior clinical measurements, independent of age, sex, and other factors, can we predict RA response just using the gut microbiome alone?” Dr. Sung said. “Is there an association between clinical improvement and the gut microbiome?”

The study also showed that the microbiome may be a modifiable target for RA, Dr. Davis said.

“This research is attractive because it may complement medical treatment for RA if we can identify dietary modifications,” he said. “Still, there’s the question if probiotics or prebiotics can influence the gut. Can we modify the gut microbiome to further ameliorate the disease state? That’s something I think is an open question that’s specifically called out in our paper.”

This study included patients with long-term disease, but the group’s ongoing research is focusing on patients with earlier-stage RA, Dr. Davis said. “The next steps have to be in validating [the findings] in additional and external populations and looking at patients with very early disease where a lot of the decision-making is very active and happening in real time.”

Dr. James T. Rosenbaum, an ophthalmologist and rheumatologist at Oregon Health & Science University in Portland

Dr. James T. Rosenbaum

James T. Rosenbaum, MD, an ophthalmologist and rheumatologist at Oregon Health & Science University, Portland, acknowledged that this is the first study in RA to find an effect on the gut microbiome using the minimum clinically important improvement endpoint.

“It also raises a ‘chicken-egg’ dilemma,” Dr. Rosenbaum said in an interview. “Did the patients improve and then their microbiome changed, or was the microbiome the first change that led to the clinical improvement? If the latter is correct, we potentially could alter the microbiome, for example, by diet, to treat rheumatic disease.”

He noted that studies with fecal transplants for ulcerative colitis support the therapeutic potential of microbiome modification. “But,” he added, “we are still a long way from putting this in practice.”

Dr. Claudia Mauri, a professor of immunology at University College London

Dr. Claudia Mauri

“The outcome is promising,” Claudia Mauri, PhD, a professor of immunology at University College London, said of the study. “Obviously, if this can be repeated in a very large cohort of patients, it would give us the possibility to be able to, based on the composition of the microbiota, to predict who is going to respond to treatment or not.”

She noted that, while RA has a broader array of available treatments than other autoimmune diseases, some RA patients don’t respond their first biologic treatment. “If from the outset we would be able to see who may not respond based on the microbiota, we may prepare physicians to better target these patients by, for example, offering them an alternative second biologic agent,” Dr. Mauri said.

Dr. Davis reported receiving research grants from Pfizer. Dr. Sung and other study coauthors have no financial relationships to disclose. Dr. Rosenbaum reported that the National Institutes of Health supports his research. Dr. Mauri has no financial relationships to disclose.

Next Article: