Serum metabolomics reveals systemic metabolic alterations of Graves' disease before and after antithyroid drug treatment

Abstract

Objective: This study aimed to utilize untargeted metabolomics to analyze changes of serum metabolites before and after methimazole treatment in patients with Graves' disease (GD).

Methods: Total 40 GD patients and 30 healthy volunteers were recruited for the study. Metabolomics analysis was conducted using liquid chromatography-mass spectrometry.

Results: A total of 11,590 metabolites were measured. Compared to healthy controls, newly diagnosed untreated GD patients exhibited significant metabolic dysregulation, with 1,805 metabolites upregulated and 1,737 downregulated. After 1 year of methimazole treatment and normalization of thyroid hormone levels, 137 metabolites remained upregulated and 242 remained downregulated, suggesting incomplete metabolic recovery. Pathway enrichment analysis indicated significant alterations in tyrosine metabolism, biosynthesis of alkaloids derived from histidine and purine, and bile secretion pathways in untreated GD patients. In addition, pathways such as ABC transporters, folate biosynthesis, D-amino acid metabolism, anthranilate degradation, and purine metabolism remained significantly dysregulated after treatment. Correlation analysis revealed positive associations between docosatetraenoic acid and citric acid with free triiodothyronine and free thyroxine levels, between 4-hydroxyphenyllactic acid and free triiodothyronine, and between 13Z,16Z-docosadienoic acid and thyroid-stimulating hormone receptor antibody levels.

Conclusion: Significant alterations in plasma metabolomic profiles during the transition from hyperthyroidism to euthyroidism in GD patients were identified using untargeted metabolomics, highlighting persistent metabolic disruptions despite clinical recovery.

Keywords: Graves’ disease; biomarkers; metabolic pathway; untargeted metabolomics.

Figure 1

PCA of metabolite data in positive and negative ion modes for each comparison group. (A) PCA of positive ion mode metabolomic data for each comparison group. (B) PCA of negative ion mode metabolomic data for each comparison group. HC: healthy control, n = 30; GD: newly diagnosed untreated patients with Graves’ disease, n = 40; GD-MMI: GD patients at 1 year after commencement of methimazole treatment, n = 15.

Figure 2

Orthogonal partial least squares discriminant analysis (OPLS-DA) of metabolite data in positive and negative ion modes for each comparison group. (A) OPLS-DA of positive ion mode metabolic data from newly diagnosed untreated GD patients and HCs. GD, n = 40; HC, n = 30. (B) OPLS-DA of negative ion mode metabolic data from newly diagnosed untreated GD patients and HCs. GD, n = 40; HC, n = 30. (C) OPLS-DA of positive ion mode metabolomic data from 15 GD patients before and after 1 year of methimazole treatment. (D) OPLS-DA of negative ion mode metabolomic data from 15 GD patients before and after 1 year of methimazole treatment. HC: healthy control; GD: Graves’ disease; GD-MMI: GD patients at 1 year after commencement of methimazole treatment.

Figure 3

Permutation test plots of metabolite data in positive and negative ion modes for each comparison group. (A) Permutation test plots of positive ion mode metabolic data from newly diagnosed untreated GD patients and HCs. GD, n = 40; HC, n = 30. (B) Permutation test plots of negative ion mode metabolic data from newly diagnosed untreated GD patients and HCs. GD, n = 40; HC, n = 30. (C) Permutation test plots of positive ion mode metabolomic data from 15 GD patients before and after 1 year of methimazole treatment. (D) Permutation test plots of negative ion mode metabolomic data from 15 GD patients before and after 1 year of methimazole treatment. HC: healthy control; GD: Graves’ disease; GD-MMI: GD patients at 1 year after commencement of methimazole treatment.

Figure 4

Volcano plots of metabolite data in positive and negative ion modes for each comparison group. (A) Volcano plots of positive ion mode metabolic data from newly diagnosed untreated GD patients and HCs. GD, n = 40; HC, n = 30. (B) Volcano plots of negative ion mode metabolic data from newly diagnosed untreated GD patients and HCs. GD, n = 40; HC, n = 30. (C) Volcano plots of positive ion mode metabolomic data from 15 GD patients before and after 1 year of methimazole treatment. (D) Volcano plots of negative ion mode metabolomic data from 15 GD patients before and after 1 year of methimazole treatment. HC: healthy control; GD: Graves’ disease; GD-MMI: GD patients at 1 year after commencement of methimazole treatment.