Background: Osteoarthritis (OA) is a degenerative joint disorder influenced by genetic, molecular, and environmental factors. Programmed cell death (PCD) pathways, including apoptosis, pyroptosis, necroptosis, ferroptosis, and autophagy, are linked to cartilage degradation, but their role in OA pathogenesis remains unclear.

Methods: Based on a large-scale GWAS database, this study employs a two-sample Mendelian randomization (MR) framework, integrating genomic data from 14 genes related to PCD at three levels (DNA methylation, gene expression, and protein abundance) to reveal causal relationships between these genes and OA. The MR analysis utilizes QTLs (mQTL, eQTL, and pQTL) as instrumental variables and employs five regression models (MR-Egger regression, Random-Effects Inverse Variance Weighted, Weighted Median, Weighted Mode, and Simple Mode) to assess causal effects. Furthermore, the reliability of causal inference is strengthened through FDR multiple testing correction, Steiger test, and colocalization analysis. Multi-omics evidence is integrated to identify key PCD genes causally related to OA. Finally, enrichment analysis, PPI analysis, and OA-related transcriptome analysis are used to explore the biological mechanisms of these key PCD genes.

Findings: Through MR analysis, we ultimately identified 103 PCD-related CpG sites, 170 PCD-related gene expressions, and 53 PCD-related protein levels that have significant causal relationships with OA. Multi-omics integration pinpointed 2 Tier 1 genes (CASP10, CASP3) and 14 Tier 2 genes (e.g., FGR, GAPDH). Validation across three cohorts confirmed causal associations for CASP10, GAPDH, PARK7, and others. Enrichment analysis implicated these genes in critical biological processes, such as neuronal apoptosis, protease binding, and the MAPK signaling pathway. Protein-protein interaction (PPI) network analysis identified CASP3 (Degree = 9) and CASP10 (Degree = 4) as central hubs, suggesting they may play a central role in the pathophysiological mechanisms of OA and could serve as potential therapeutic targets for OA. Transcriptome analysis confirmed MR findings. Tier 1 gene CASP3 was significantly upregulated in OA patients (log2FC = 1.30, adjusted P < 0.05), and CASP10 showed non-significant upregulation. Tier 2 genes (GAPDH, CD14, CHMP2B, GM2A, ITGAM) also showed significant changes (P < 0.05) consistent with MR results.

Interpretation: This study provides a multi-omic framework for understanding the role of PCD in OA, providing insights into potential PCD-targeted therapies.