Serum Metabolomic Profiling of Patients with Non-Infectious Uveitis

Abstract

The activities of various metabolic pathways can influence the pathogeneses of autoimmune diseases, and intrinsic metabolites can potentially be used to diagnose diseases. However, the metabolomic analysis of patients with uveitis has not yet been conducted. Here, we profiled the serum metabolomes of patients with three major forms of uveitis (Behҫet's disease (BD), sarcoidosis, and Vogt-Koyanagi-Harada disease (VKH)) to identify potential biomarkers. This study included 19 BD, 20 sarcoidosis, and 15 VKH patients alongside 16 healthy control subjects. The metabolite concentrations in their sera were quantified using liquid chromatography with time-of-flight mass spectrometry. The discriminative abilities of quantified metabolites were evaluated by four comparisons: control vs. three diseases, and each disease vs. the other two diseases (such as sarcoidosis vs. BD + VKH). Among 78 quantified metabolites, 24 kinds of metabolites showed significant differences in these comparisons. Four multiple logistic regression models were developed and validated. The area under the receiver operating characteristic (ROC) curve (AUC) in the model to discriminate disease groups from control was 0.72. The AUC of the other models to discriminate sarcoidosis, BD, and VKH from the other two diseases were 0.84, 0.83, and 0.73, respectively. This study provides potential diagnostic abilities of sarcoidosis, BD, and VKH using routinely available serum samples that can be collected with minimal invasiveness.

Keywords: Behҫet’s disease; Vogt-Koyanagi-Harada disease; biomarker; liquid chromatography-mass spectrometry; metabolomics; sarcoidosis; serum.

Figure 1

Comparative design and multiple logistic regression (MLR) model development. (A) Four patterns of comparison include (a) Control vs (sarcoidosis + BD + VKH), (b) sarcoidosis vs (BD + VKH), (c) BD vs (sarcoidosis + VKH), and (d) VKH vs (BD + sarcoidosis). (B) The development and validation of an MLR model, which starts from the metabolite concentration matrix (sample × metabolites), selecting metabolites showing the significant differences, further selecting independent minimum metabolites by stepwise feature selection, and development of an MLR model. The developed model was validated by cross-validation (CV) and resampling methods. Abbreviations: Behҫet’s disease, BD; Vogt-Koyanagi-Harada disease, VKH. The discrimination abilities of the metabolites were evaluated using both univariate and multivariate analyses. Half of the minimum concentration was substituted for the data under the lower limit of quantification. Differences in metabolite concentrations between two groups were evaluated by the Mann-Whitney U test. The design of the comparison is depicted in Figure 1A. The discrimination ability of uveitis including all three diseases from controls were evaluated (a). To access the diagnostic ability among three diseases, the discrimination abilities of sarcoidosis from BD + VKH (b), BD from sarcoidosis + VKH (c), and VKH from BD + sarcoidosis (d) were also evaluated.

Figure 2

Heatmap of quantified metabolites by LC-TOFMS. Metabolite concentrations were transferred to Z-score for each metabolite. The averaged value for each group (C, sarcoidosis, BD, and VKH) were coloured in the blue-white-red scheme. Below the heatmap, black boxes are shown for the significantly different metabolites in the comparison of (a)–(d) in Figure 1. Pathways to which each metabolite belong are shown using the color box below the metabolite name. Clustering was conducted using Pearson correlation and prominent clusters are labelled (1)–(5). Abbreviations: Control, C; Behҫet’s disease, BD; Vogt-Koyanagi-Harada disease, VKH.

Figure 3

Discrimination ability of four MLR models. (A–D) Receiver operating characteristic (ROC) curves of the MLR models for comparisons (a–d) in Figure 1. (E–H) The distribution of the predicted probability of (A–D). Abbreviations: Control, C; Behҫet’s disease, BD; Vogt-Koyanagi-Harada disease, VKH.