Graves’ Disease Gut Microbiota: Distinct Microbial Signatures and Clinical Associations Original paper

What was studied?

This original research article examined alterations in the gut microbiota composition of patients with Graves’ disease (GD) compared to healthy controls using 16S rRNA gene sequencing. The primary aim was to characterize the gut microbial communities in GD, assess differences in bacterial diversity and abundance, and explore correlations between specific microbiota and clinical parameters of GD. The study also evaluated the potential of microbiome profiles to serve as discriminative biomarkers for distinguishing GD patients from healthy individuals, thereby contributing to a better understanding of the gut-thyroid axis in autoimmune thyroid disease.

Who was studied?

The study enrolled 55 patients with clinically diagnosed Graves’ disease and 48 age-, sex-, and BMI-matched healthy controls. All GD patients were recruited from the Division of Endocrinology and Metabolism at Chang Gung Memorial Hospital, Taiwan, and had been previously diagnosed and treated with anti-thyroid drugs (propylthiouracil, methimazole, or carbimazole), with an average follow-up of over 45 months. The control group was selected from a health screening center and had no history of thyroid disease, with normal thyroid-related laboratory values. Exclusion criteria included pregnancy, gastrointestinal disorders, concurrent autoimmune diseases, recent antibiotic or probiotic use, hormonal or herbal drug intake, gastrointestinal surgery, and strict vegetarianism. Fecal samples were collected from all participants for gut microbiota analysis.

Most important findings

The study found no statistically significant difference in overall gut microbial richness or diversity between GD patients and healthy controls, as measured by standard alpha diversity indices. However, beta diversity metrics—including principal coordinate analysis (PCoA) and partial least squares-discriminant analysis (PLS-DA)—demonstrated a clear separation in overall microbiota composition between the two groups (ANOSIM, p <0.001). At the phylum level, GD patients exhibited a decreased abundance of Firmicutes and an increased abundance of Bacteroidetes and Actinobacteria compared to controls. Key microbial signatures in GD included increased levels of the families Prevotellaceae and Veillonellaceae, and the genera Prevotella_9, Parabacteroides, Collinsella, and Actinomyces_odontolyticus. In contrast, healthy controls had higher abundances of Lachnospiraceae, Ruminococcaceae, and the genera Faecalibacterium, Lachnospira, and Lachnospiraceae NK4A136. Notably, 22 bacterial taxa showed statistically significant differences between groups; 18 were increased and 4 decreased in GD patients. Several of these taxa, especially those enriched in GD, were positively correlated with GD-associated clinical parameters such as thyroperoxidase antibody (TPOAb) and free thyroxine (FT4) levels, and negatively correlated with thyroid-stimulating hormone (TSH). A machine learning model based on the top 15 discriminant taxa achieved an area under the ROC curve (AUC) of 0.825, indicating strong potential for microbiota-based discrimination of GD status.

Key implications

This study demonstrates that Graves’ disease is associated with a distinct gut microbiota signature, despite similar overall bacterial diversity compared to healthy controls. The shift towards increased Bacteroidetes and Actinobacteria and decreased Firmicutes, as well as the enrichment of specific pro-inflammatory and immunomodulatory taxa (such as Prevotella_9 and Veillonellaceae), supports the hypothesis that gut dysbiosis may contribute to GD pathogenesis via immune modulation. The strong correlation between key microbial taxa and clinical indicators of GD suggests that these bacteria may play a role in disease activity or progression. Moreover, the successful discrimination of GD patients based on gut microbiota profiles suggests potential for developing non-invasive microbial biomarkers for GD diagnosis or monitoring. However, as this was a cross-sectional, single-center study, causality remains undetermined, and larger multi-omics studies are needed to clarify mechanistic pathways and explore therapeutic interventions targeting the gut microbiome in GD.