Role of transcript-based HER2 quantification in classification and determining therapeutic eligibility.

e15195 Background: Immunohistochemistry (IHC) is widely used to determine human epidermal growth factor receptor 2 (HER2) expression for tumor classification and treatment eligibility; however, its semi-quantitative and observer-dependent nature may underestimate biologically relevant HER2 activity. The clinical efficacy of trastuzumab deruxtecan (T-DXd, Enhertu) in HER2-low and ultralow tumors has challenged IHC-based stratification, prompting the need for more objective biomarkers such as bulk mRNA sequencing. Methods: We analyzed ERBB2 (HER2) transcript expression in three cancers—breast, colorectal, and osteosarcoma—classified as HER2-low or negative by IHC. Normalized RNA-sequencing data were benchmarked against matched normal tissues from the corresponding organ of origin. In addition, we included a coverage analysis to confirm mature mRNA expression and to reveal transcript variant distribution. In addition, we carried out an independent analysis on the data of The Metastatic Breast Cancer Project dataset to assess the frequency of HER2-low and HER2-negative tumors with aberrant HER2 expression on mRNA level. Results: All tumors demonstrated elevated HER2 transcription, with two exceeding the full expression range of normal tissues and one reaching the 89th percentile. Coverage analysis confirmed the presence of mature, spliced HER2 transcripts, and isoform analysis identified upstream 5′ untranslated-region (UTR) variants as well as the Δ16 splice variant associated with constitutive receptor activation. Independent analysis of the Metastatic Breast Cancer Project dataset similarly revealed that substantial subsets of HER2-low and HER2-negative tumors exceed the 95th and even the 99th percentiles of normal breast expression. Conclusions: These findings demonstrate that transcript-level profiling provides an objective and sensitive measure of clinically meaningful HER2 activity that may be missed by IHC, offering a rational framework for personalized treatment selection and potentially expanding therapeutic options for patients otherwise deemed ineligible for targeted therapy.