OBJECTIVE: To identify the potential biomarkers of interstitial cystitis/painful bladder syndrome (IC), a chronic syndrome of bladder-centric pain with unknown etiology that has an adverse impact on quality of life, we analyzed the urine and serum metabolomes of a cohort of IC patients and non-disease controls (NC). METHODS: Home collection of serum and urine samples was obtained from 19 IC and 20 NC females in the Veterans Affairs (VA) Health Care System. IC was diagnosed independently by thorough review of medical records using established criteria. Biostatistics and bioinformatics analyses, including univariate analysis, unsupervised clustering, random forest analysis, and metabolite set enrichment analysis (MSEA), were then utilized to identify potential IC biomarkers. RESULTS: Metabolomics profiling revealed distinct expression patterns between NC and IC. Random forest analysis of urine samples suggested discriminators specific to IC; these include phenylalanine, purine, 5-oxoproline, and 5-hydroxyindoleacetic acid. When these urinary metabolomics-based analytes were combined into a single model, the AUC was 0.92, suggesting strong potential clinical value as a diagnostic signature. Serum-based metabolomics did not provide potential IC discriminators. CONCLUSION: Analysis of serum and urine revealed that women with IC have distinct metabolomes, highlighting key metabolic pathways that may provide insight into the pathophysiology of IC. The findings from this pilot study suggest that integrated analyses of urinary metabolites, purine, phenylalanine, 5-oxoproline, and 5-HIAA, can lead to promising IC biomarkers for pathophysiology of IC. Validation of these results using a larger dataset is currently underway.
OBJECTIVE: To identify the potential biomarkers of interstitial cystitis/painful bladder syndrome (IC), a chronic syndrome of bladder-centric pain with unknown etiology that has an adverse impact on quality of life, we analyzed the urine and serum metabolomes of a cohort of IC patients and non-disease controls (NC). METHODS: Home collection of serum and urine samples was obtained from 19 IC and 20 NC females in the Veterans Affairs (VA) Health Care System. IC was diagnosed independently by thorough review of medical records using established criteria. Biostatistics and bioinformatics analyses, including univariate analysis, unsupervised clustering, random forest analysis, and metabolite set enrichment analysis (MSEA), were then utilized to identify potential IC biomarkers. RESULTS: Metabolomics profiling revealed distinct expression patterns between NC and IC. Random forest analysis of urine samples suggested discriminators specific to IC; these include phenylalanine, purine, 5-oxoproline, and 5-hydroxyindoleacetic acid. When these urinary metabolomics-based analytes were combined into a single model, the AUC was 0.92, suggesting strong potential clinical value as a diagnostic signature. Serum-based metabolomics did not provide potential IC discriminators. CONCLUSION: Analysis of serum and urine revealed that women with IC have distinct metabolomes, highlighting key metabolic pathways that may provide insight into the pathophysiology of IC. The findings from this pilot study suggest that integrated analyses of urinary metabolites, purine, phenylalanine, 5-oxoproline, and 5-HIAA, can lead to promising IC biomarkers for pathophysiology of IC. Validation of these results using a larger dataset is currently underway.