H-score and ERBB2 mRNA refine identification of HER2-low and HER2-ultralow breast cancer

Abstract Background Accurate HER2 assessment is essential for breast cancer (BC) treatment, as it directs targeted therapy decisions and predicts patient prognosis. Even though immunohistochemistry (IHC) is widely used, manual scoring is susceptible to variability. This study was conducted to evaluate HER2 expression in HER2-negative breast cancer patients treated with neoadjuvant chemotherapy (NAC), using H-score and ERBB2 mRNA quantification. Methods We conducted a retrospective cohort study in breast cancer patients treated with NAC between 2017 and 2023. H-score was evaluated in 506 patients. For gene expression analyses, only tumors with available H-score and Ct ≤ 35 for all candidate reference genes were included, resulting in 173 evaluable cases for ERBB2 mRNA quantification. HER2-zero tumors were further stratified into HER2-null (H-score = 0) and HER2-ultralow (H-score 1–120) groups. Results Among all 506 HER2-negative tumors, 54.6% were classified as HER2-low and 45.4% as HER2-zero. Within the HER2-zero subgroup, 47.4% showed measurable HER2 expression (HER2-ultralow). ERBB2 mRNA levels were significantly higher in HER2-low tumors compared to HER2-zero. Among HER2-zero, HER2-ultralow tumors had higher ERBB2 expression than HER2-null. Higher H-scores were associated with increased ERBB2 transcript expression, according to Spearman correlation coefficient ( p < 0.05). HER2-null tumors were more frequently triple-negative. Conclusions Quantitative assessment by H-score and ERBB2 transcript analysis reveals heterogeneity within HER2-zero tumors, identifying a HER2-ultralow population with potentially distinct clinical behavior. These findings support the integration of continuous HER2 expression metrics into clinical decision-making and trial design, particularly in the era of HER2-targeted antibody–drug conjugates.