Metabolic changes preceding bladder cancer occurrence among Korean men: a nested case-control study from the KCPS-II cohort

Background Bladder cancer (BLCA) research in Koreans is still lacking, especially in focusing on the prediction of BLCA. The current study aimed to discover metabolic signatures related to BLCA onset and confirm its potential as a biomarker. Methods We designed two nested case-control studies using...

Full description

Saved in:
Bibliographic Details
Published in:Cancer & Metabolism 2023, Vol.11 (1)
Main Authors: Han, Youngmin, Kim, Unchong, Jung, Keum Ji, Lee, Ji-Young, Lee, Kwangbae, Shin, Sang Yop, Kimm, Heejin, Jee, Sun Ha
Format: Report
Language:English
Subjects:
Analysis
Bladder cancer
Cancer
Lysine
Metabolites
Oncology, Experimental
Prevention
Tryptophan
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Background Bladder cancer (BLCA) research in Koreans is still lacking, especially in focusing on the prediction of BLCA. The current study aimed to discover metabolic signatures related to BLCA onset and confirm its potential as a biomarker. Methods We designed two nested case-control studies using Korean Cancer Prevention Study (KCPS)-II. Only males aged 35-69 were randomly selected and divided into two sets by recruitment organizations [set 1, BLCA (n = 35) vs. control (n = 35); set 2, BLCA (n = 31) vs. control (n = 31)]. Baseline serum samples were analyzed by non-targeted metabolomics profiling, and OPLS-DA and network analysis were performed. Calculated genetic risk score (GRS) for BLCA from all KCPS participants was utilized for interpreting metabolomics data. Results Critical metabolic signatures shown in the BLCA group were dysregulation of lysine metabolism and tryptophan-indole metabolism. Furthermore, the prediction model consisting of metabolites (lysine, tryptophan, indole, indoleacrylic acid, and indoleacetaldehyde) reflecting these metabolic signatures showed mighty BLCA predictive power (AUC: 0.959 [0.929-0.989]). The results of metabolic differences between GRS-high and GRS-low groups in BLCA indicated that the pathogenesis of BLCA is associated with a genetic predisposition. Besides, the predictive ability for BLCA on the model using GRS and five significant metabolites was powerful (AUC: 0.990 [0.980-1.000]). Conclusion Metabolic signatures shown in the present research may be closely associated with BLCA pathogenesis. Metabolites involved in these could be predictive biomarkers for BLCA. It could be utilized for early diagnosis, prognostic diagnosis, and therapeutic targets for BLCA. Keywords: Bladder cancer, Predictive biomarker, LC/MS metabolomics, Gaussian graphical model, Lysine metabolism, Tryptophan-indole metabolism
ISSN:2049-3002
2049-3002
DOI:10.1186/s40170-023-00324-0