Identification of aging-related biomarkers for intervertebral disc degeneration in whole blood samples based on bioinformatics and machine learning.

INTRODUCTION

Aging is characterized by gradual structural and functional changes in the body over time, with intervertebral disc degeneration (IVDD) representing a key manifestation of spinal aging and a major contributor to low back pain (LBP).

METHODS

This study utilized bioinformatics and machine learning approaches to identify aging-related biomarkers associated with IVDD in whole blood samples. By analyzing GEO datasets alongside aging-related databases such as GeneCards, HAGR, and AgeAnno, we identified 15 aging-related differentially expressed genes (AIDEGs). Correlation and immune infiltration analyses were conducted on these AIDEGs, and diagnostic models were developed using WGCNA, logistic regression, random forest, support vector machine, k-nearest neighbors, and LASSO regression to identify key genes.

RESULTS

Among these, FCGR1A, CBS, and FASLG emerged as significant biomarkers with strong predictive capabilities for IVDD. Further exploration of biological pathways involving AIDEGs provided insights into their potential roles in IVDD pathogenesis. To further validate these findings, we collected human blood specimens and conducted in vitro experiments. ELISA assays confirmed that CBS and FASLG are crucial biomarkers of IVDD, with distinct expression patterns in patients with moderate versus severe degeneration.

DISCUSSION: These results highlight the diagnostic potential of AIDEGs and provide a new perspective for early intervention and treatment strategies in IVDD.