Arachidonic acid metabolism-based stratification reveals immune microenvironment heterogeneity and defines a prognostic signature in triple-negative breast cancer

Arachidonic acid (AA) metabolic reprogramming has emerged as an important driver of tumor progression and immune escape. However, how AA metabolism–related genes (AAMRGs) contribute to immune microenvironment heterogeneity and clinical outcomes in triple-negative breast cancer (TNBC) remains poorly defined. Transcriptomic and clinical data for TNBC were obtained from TCGA, GEO (GSE58812, GSE13565, GSE91061), and IMvigor210 databases. AAMRGs were screened using GeneCards. Differential expression, GO/KEGG/GSEA/GSVA enrichment analyses were performed to explore the biological functions of AAMRGs. Consistency clustering was used to classify TNBC into subtypes. A 10-gene prognostic signature was constructed using LASSO regression and validated in GEO datasets. Immune infiltration was analyzed using CIBERSORT/Xcell algorithms, while immune therapy response and drug sensitivity were evaluated using TIDE/TIP and OncoPredict. IHC and qRT-PCR/Western blot were used to validate core gene expression in clinical TNBC samples and cell lines. A total of 34 core AAMRGs were identified, classifying TNBC into two distinct molecular subtypes with significant prognostic differences (AMS-SM1 and AMS-SM2). AMS-SM1 was associated with better prognosis (p < 0.05), enriched in M1 macrophage infiltration, and had a 38% immune therapy response rate, while AMS-SM2 exhibited poorer prognosis, with significant infiltration of B cells, M2 macrophages, and cancer-associated fibroblasts (CAFs), and a 20% immune therapy response rate. The 10-gene signature showed high predictive accuracy (3-year, 5-year, and 7-year AUC: 0.966, 0.862, 0.858) in the TCGA cohort and was significantly associated with immune infiltration, therapy response, and drug sensitivity. GFUS was identified as a core gene, with elevated expression in both TNBC tissue and TNBC cell lines. We established an AAMRG-based subtype framework and risk score for TNBC, linking AA metabolic activity to distinct immune microenvironment patterns and highlighting GFUS as a prominent prognostic-associated gene.