Gut Enzymes Fingered in Some 5-ASA Treatment Failures

Neil Osterweil

January 30, 2023

AURORA, COLO. – The therapeutic action of 5-aminosalicylic acid (5-ASA), one of the most frequently prescribed drugs for inflammatory bowel disease (IBD), can be defeated by enzymes that reside in the very gut that the drug is designed to treat.

"What we found is two gut microbial acetyltransferase families that were previously unknown to be participating in drug metabolism that directly inactivate the drug 5-ASA. It seems that in turn, having a subset of these microbial acetyltransferases is prospectively linked with treatment failure, and could potentially explain why some of these patients of ours fail on the drug," Raaj S. Mehta, MD, MPH, a postdoctoral fellow at Massachusetts General Hospital in Boston, said at the annual Crohn's & Colitis Congress®, a partnership of the Crohn's & Colitis Foundation and the American Gastroenterological Association.

More than half of all patients with IBD treated with 5-ASA either lose their response to the drug or never respond to it at all, including some of his own patients, Dr. Mehta said.

There is an urgent need for a way to predict which patients will be likely to respond to 5-ASA and other drugs to treat IBD, he said.

The same old story

In the early 1990s, investigators at St. Radboud Hospital, Nijmegen, the Netherlands, studied cultured feces from patients with IBD treated with 5-ASA, and found in some patients that the drug was metabolized into N-acetyl-5-ASA. In an earlier, double-blind comparison trial in patients with idiopathic proctitis, the same investigators found N-acetyl-5-ASA to be "no better than placebo."

"But prior to our work, we didn't know which specific bacteria or enzymes performed this conversion ... of the drug, and we didn't know if having these enzymes in your intestines or colon could explain why people are at risk for failing on 5-ASA," Dr. Mehta said.

New evidence

Dr. Mehta and his colleagues first turned to the Human Microbiome Project 2, a cohort of 132 persons with IBD followed for 1 year each, with the goal of generating molecular profiles of host and microbial activity over time.

The patients provided stool samples about every 2 weeks, as well as blood and biopsy specimens, and reported details on their use of medications.

The investigators generated metagenomic, metatranscriptomic, genomic, and metabolomic profiles from the data, and then narrowed their focus to 45 participants who used 5-ASA and 34 who did not.

They found that "5-ASA has a major impact on the fecal metabolome," with significant increases in fecal drug levels of both 5-ASA and the inactive metabolites, as well as more than 2,000 other metabolites.

Looking at the gemomics of gut microbiota, the investigators identified gene clusters in two superfamilies of enzymes, thiolases and acyl CoA N-acyltransferases. They identified 12 candidates.

To bolster their findings, they then expressed one gene from each superfamily in Escherichia coli and purified the protein. When they cocultured it with acetylCoA and 5-ASA, there was a greater than 25% conversion of the drug within 1 hour.

They also found that microbial thiolases appear to step outside of their normal roles to inactivate 5-ASA in a manner similar to that of an N-acetyltrasferase not found in persons with IBD.

Clinical relevance

To see whether their findings had clinical implications, the investigators conducted a case-cohort study nested within the Human Microbiome Project 2 cohort. They saw that, after adjusting for age, sex, IBD type, smoking, and N-acetyltransferase (NAT2) phenotype, 4 of the 12 acetyltranfserase candidates were associated with a roughly threefold increase in steroid use, suggesting that 5-ASA treatment had failed the patients.

"So then to take it one step further, we turned to the SPARC IBD cohort," Dr. Mehta said.

SPARC IBD is an ongoing prospective cohort of patients who provide stool samples and detailed medication and symptom data.

They identified 208 cohort members who were on 5-ASA, were free of steroids at baseline, and who had fecal metabolomic data available. In this group, there were 60 cases of new corticosteroid prescriptions after about 8 months of follow-up.

The authors found that having three or four of the suspect acetyltransferases in gut microbiota was associated with a an overall odds ratio for 5-ASA treatment failure of 3.12 (95% confidence interval, 1.41-6.89).

"Taken together, I think this advances the idea of using the microbiome for personalized medicine in IBD," Dr. Mehta said.

"Right now it's an ideal outcome for a patient with [ulcerative colitis] to retain a robust remission on 5-ASA alone," commented session moderator Michael J. Rosen, MD, MSCI, a pediatric gastroenterologist at Stanford University Medical Center in Palo Alto, Calif., who was not involved in the study.

Asked in an interview whether the findings would be likely to change clinical practice, Dr. Rosen replied that "I think it's fairly early stage, but I think it's wonderful that they sort of rediscovered this older data and are modernizing it to understand why [5-ASA] may not work for some patients. It certainly seems like it might be a tractable approach to use the microbiome to personalize therapy and potentially increase the effectiveness of 5-ASA."

The study was supported by grants from Pfizer, the National Institutes of Health, American College of Gastroenterology, and the Crohn's & Colitis Foundation. Dr. Mehta disclosed that his team has filed a provisional patent application related to the work. Dr. Rosen reported no relevant conflict of interest.

This article originally appeared on, part of the Medscape Professional Network.


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.