A Pan-Subtype Fatty Acid Metabolic Signature Reveals Universal Prognostic Stratification in Breast Cancer

Breast cancer exhibits profound molecular heterogeneity. Each subtype is characterized by distinct biological features, therapeutic responses, and prognostic outcomes. However, a unifying feature across most breast cancer subtypes is an adipocyte-enriched microenvironment. Dysregulated lipid metabolism drives tumor progression across subtypes.

We developed a universal prognostic model independent of molecular classification based on fatty acid metabolism-related genes.

Prognostic fatty acid metabolism-related genes were identified in the training set using univariate Cox regression. LASSO regression refined these genes to construct the fatty acid metabolic signature for prognosis (FAMOUS), enabling risk stratification. FAMOUS's prognostic utility was validated in the testing set and external cohort, and across molecular subtypes and therapies.

FAMOUS comprises 15 genes. High FAMOUS scores independently predicted poor overall survival in the training set (HR = 4.97, 95% CI: 3.17-7.79; p < 0.001), testing set (HR = 7.39, 95% CI: 2.19-24.97; p = 0.001), and GSE72245 (HR = 7.97, 95% CI: 3.39-18.75; p < 0.001), after adjusting for molecular subtype, stage, and age. It stratified risk across Luminal A, Luminal B, HER2+, and TNBC subtypes and retained prognostic value in patients receiving chemotherapy, radiotherapy, endocrine therapy, or trastuzumab. High FAMOUS scores correlated with immune-suppressive microenvironments, marked by downregulated immune-related pathways and altered immune cell infiltration (e.g., reduced CD8+ T cells, enriched M2 macrophages), suggesting implications for immunotherapy patient stratification.

FAMOUS is a novel, pan-subtype prognostic tool for breast cancer, transcending molecular classification and therapeutic modalities.