Abstract
The status of human epidermal growth factor receptor 2 (HER2) is a critical determinant of breast cancer progression and outcome, but the role of glycosylation in modulating HER2-related pathways remains poorly understood. Here, we examined the relationship between HER2 and the glycosyltransferases ST3GAL1 and GCNT3, combining data mining with experimental and clinicopathological validation. In silico analyses across public BC cohorts showed that higher expression of ST3GAL1, GCNT3, and HER2 was associated with reduced survival. Immunohistochemistry on invasive ductal carcinoma specimens (n = 25) demonstrated increased ST3GAL1 and GCNT3 in advanced stages/grades. Cross-platform correlation analyses revealed a positive association between ST3GAL1 and HER2, whereas GCNT3 showed an inverse association with HER2. Functional assays in HER2-negative cell lines (MCF7, MDA-MB-231, MDA-MB-435) and the HER2-positive line SKBR3 indicated that GCNT3 supports migratory capacity and clonogenicity, consistent with an oncogenic role independent of HER2 status. Gene set enrichment pointed to upregulation of MUC1 and β-catenin; tissue validation of MUC1, β-catenin, and Cyclin D1 confirmed their clinicopathological relevance, with HER2 expression inversely correlated with β-catenin and Cyclin D1. Collectively, these findings suggested a model in which GCNT3-driven O-glycosylation might remodel the MUC1/β-catenin/Cyclin D1 axis in BC, adding a glyco-regulatory layer to HER2-linked pathobiology. From a diagnostic perspective, ST3GAL1 (positive with HER2) and GCNT3 (inverse with HER2) could be considered as candidate biomarkers that might complement HER2 assessment for risk stratification. This work suggests a plausible mechanistic and clinicopathologic foundation for incorporating glycosylation markers into precision pathology workflows in breast cancer. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12885-026-15821-w.