Treatment of inflammatory bowel diseases (IBD) with anti-tumor necrosis factor (anti-TNF) drugs restores diversity to the intestinal microbiome, researchers report in the November issue of Gastroenterology. The study associates metabolic interactions among luminal bacteria with outcomes of therapy.
Altered interactions between the mucosal immune system and intestinal microbiota contribute to development of IBD. It is not clear how inhibitors of cytokines, such as anti-TNF drugs, affect the intestinal microbiome. Konrad Aden et al investigated the effects of anti-TNF agents on gut microbe community structure and function in a longitudinal study of patients with IBD.
In a prospective study of 2 cohorts of patients with IBD starting new treatment with anti-TNF agents, Aden et al collected fecal samples before treatment (baseline) and then at 2, 6, and 14 or 30 weeks after treatment began. They also collected samples from healthy individuals or patients with rheumatic diseases receiving treatment with anti-TNF drugs, as controls. The authors then correlated features of the fecal microbiome and patterns of metabolites with outcomes of treatment (see figure).
Aden et al found that baseline microbial communities differed significantly between fecal samples from healthy individuals and patients with rheumatic disease or IBD. However, microbial species richness and phylodiversity increased in patients with IBD, toward that of healthy individuals, after therapy began, and reached statistical significance at week 30, compared with baseline. This shift was not observed in patients receiving anti-TNF therapy for rheumatic diseases.
Aden et al identified 14 phylotypes that differed significantly between untreated patients with IBD and healthy controls. When patients with IBD began anti-TNF treatment, all of these phylotypes normalized to those of healthy controls. For example, fecal samples of patients with IBD had reduced abundance of Coprococcus and Roseburia inulinivorans at baseline, compared to healthy controls. However, these increased in abundance with time of treatment, and there was no significant difference between healthy controls and patients with IBD at week 30. Aden et al state that it is noteworthy that phylotypes whose abundance changed significantly in patients with IBD toward the direction of healthy subjects (Coprococcus and Roseburia inulinivorans) are producers of short-chain fatty acids.
In silico modeling of microbial metabolism indicated a strong disruption of ecosystem function in the intestinal microbes of patients with IBD or rheumatic diseases. These were partly restored by anti-TNF therapy.
To determine whether the changes observed in fecal microbiomes could be attributed to inhibition of TNF or successful remission in general, the authors studied samples from an additional 23 patients with IBD who underwent first-time therapy with either anti-TNF (infliximab) or anti-α4β7 integrin (vedolizumab). In silico predictions of bacterial interactions indicated that differences in the intestinal microbial ecology depended on patients’ remission status, and were found in patients who responded to anti-TNF or vedolizumab.
Aden et al investigated stool metabolomes of 9 patients with IBD at baseline and after 14 weeks of anti-TNF treatment. To determine which metabolites were characteristic of baseline, nonremission, and remission groups, they used an orthogonal partial least-squares discriminant analysis model. Within a screening set of 50 metabolites, the authors identified 4 microbial metabolites that could discriminate among baseline, nonremission, and remission groups. Whereas the baseline group was characterized by increased levels of 3-indolepropionic acid and L-tyrosine, the remission group could be identified by increased 3-hydroxyphenylacetic acid and the nonremission group by increased pyruvic acid.
Overall, creatinine distinguished untreated (baseline) patients from those treated with anti-TNF (week 14), irrespective of remission status. Interestingly butyric acid was significantly increased in only the stool metabolome of patients in remission during anti-TNF therapy. In contrast, 3-methyl-thiopropionic acid, methyl 2-(methylthio)acetate was increased specifically in samples from patients with IBD who had not achieved remission at week 14.
The authors conclude that anti-TNF treatment restores intestinal microbial diversity in patients with IBD, whereas in patients with rheumatic diseases, these changes are less pronounced and transient. These findings indicate the functional links between the intestinal microbiota and the therapeutic effects of TNF inhibitors. Further studies are needed to determine whether this microbial metabolic interaction network can be used to improve the effects of anti-TNF therapy or identify patients most likely to respond to treatment.