Introduction:
Breast cancer (BC) is the most common malignancy among women worldwide. Although observational studies have linked carnitine-related metabolites (CRMs) to BC, causal inference has been limited by confounding and reverse causality. This study utilized Mendelian randomization (MR) analysis to investigate the potential causal link between CRMs and BC.

Material and methods:
MR analysis was conducted using the Cancer Genome Atlas-BC and CRM-related datasets to explore the causal relationship between CRMs and BC. Variants were screened based on criteria encompassing genome-wide significance (p < 5×10⁻⁸), independence (r² < 0.001, kb=10000), and sufficient strength (F-statistic > 10). Various MR techniques, including Inverse-Variance Weighted, MR-Egger, simple mode, weighted mode, and weighted median approaches, were applied to assess these causal associations. Single-cell RNA sequencing (scRNA-seq) was employed to investigate the expression and biological functions of biomarkers linked to metabolites.

Results:
The MR analysis revealed a significant causal relationship between elevated levels of octanoylcarnitine and decanoylcarnitine and an increased risk of BC. Functional enrichment analysis identified 12 candidate genes associated with these metabolites, which are involved in fatty acid β-oxidation (FAO) pathways. ScRNA-seq analysis revealed eight distinct cell subpopulations, with macrophages exhibiting the highest intercellular communication. Six biomarkers were identified as potential contributors to BC development: ACADM, FNIP2, RAPGEF2, RABGGTB, PPID, and ST6GALNAC3.

Conclusions:
This study identifies octanoylcarnitine and decanoylcarnitine as significant causal risk factors for BC, linked to dysregulated FAO. Integration of single-cell transcriptomics revealed six CRM-associated biomarkers and their dynamic expression within tumor microenvironments. These findings provide novel candidates for BC diagnosis and personalized therapy, warranting further validation.