Background/Purpose:  Despite significant advances in the therapeutics of inflammatory arthritides, methotrexate (MTX) remains the mainstay of treatment for rheumatoid arthritis (RA) and related conditions worldwide. However, it has a hihgly variable inter-individual bioavailabity (ranging from 20 to 80%) and there is currently no mechanism to predict efficacy. One of the fundamental roles of the intestinal microbiome is to metabolize xenobiotics and synthetic drugs. Here we characterize the effects of oral methotrexate on the gut community composition of patients with RA and the potential role of baseline human microbiota in predicting response to MTX therapy.

Methods: Demographic characteristics, drug use and disease activity scores-28 (DAS28) were recorded from new-onset rheumatoid arthritis (NORA) patients (n=33). For each participant, fecal samples were collected at baseline and at pre-established intervals for at least 3 months after initiation of oral methotrexate (range 3-48 months). 16S rDNA was extracted per protocol (MoBio, USA) and amplicons targeting the hypervariable V4 region were sequenced using 454 and MiSeq (Illumina) platforms to define the microbiota composition. The obtained 16S rRNA sequences were analyzed using the Quantitative Insights into Microbial Ecology (QIIME) pipeline. Taxonomic relative abundance at all hierarchical levels was determined to establish baseline microbiota composition prior to MTX initiation. Two-tailed Wilcoxon non-parametric test was applied to identify significant microbiota taxonomic changes that occur after MTX therapy. The False Discovery Rate (FDR) approach was applied to adjust for multiple hypothesis testing. Changes with a P<0.05 and FDR<0.2 were considered significant.Spearman correlations between baseline relative composition of intestinal microbiota and clinical response to MTX at each time point were also applied.

Results: To quantify microbiota similarities among fecal samples we used unweighted UniFrac and hierarchical clustering. Samples from MTX-treated patients clustered with their respective baseline samples, indicating that the gut microbiota is stable with inter-individual taxonomic differences maintained for at least 6 months. NORA patients receiving MTX developed minimal changes over time. Intriguingly, however, baseline microbiome signatures in these patients predicted clinical response to MTX at 3 and 6 months, including the overabundance of unclassified Coriobacteriaceae (r=-0.756; P <0.01) and a Coprococcus-related OTU (r=-0.755; P =0.022).

Conclusion: Although oral methotrexate does not induce significant changes in the over all structure of the human intestinal microbiota of NORA patients, the abundance of several taxa at baseline correlate with a significant improvement in clinical disease activity 3 and 6 months into therapy. Whether specific intestinal commensals can modulate the pharmacokinetics and bioavailability of methotrexate (and other DMARDs), remains to be elucidated. Better understanding of MTX pharamcomicrobiomics will be necessary to achieve precision medicine strategies in RA and related conditions.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/pharmacomicrobiomics-of-methotrexate-baseline-intestinal-microbiota-correlates-with-therapeutic-response/