Employing semi-quantitative methods to classify breast cancer markers (ER, PR, HER2, Ki67) expression, using the RT-qPCR based APIS Breast Cancer Subtyping Kit.

e13091 Background: Accurate determination of breast cancer (BC) biomarker status is crucial for guiding patient management decisions. Histopathologists commonly use semi-quantitative scoring systems to convert subjective observations of immunohistochemistry (IHC) marker expression into quantitative data, allowing a more detailed categorisation of marker status. The APIS Breast Cancer Subtyping Kit (BCSK) is an in vitro diagnostic test, detecting the relative expression of seven mRNA target genes ( ESR1, PGR, ERBB2, MKI67, CCNA2, PCNA, and KIF23) from invasive BC tissue. The APIS BCSK reports a positive/negative result for each biomarker alongside a molecular classification. Here, we demonstrate that by implementing additional RNA expression cut-off values, it is possible to generate a semi-quantitative result by further stratifying target expression into negative, low, medium, and high status using the APIS BCSK. Methods: 368 formalin-fixed paraffin-embedded (FFPE) samples of BC cases were used, obtained by core needle biopsy or resection. IHC scores were correlated with copy numbers as determined by an independently validated digital PCR (dPCR) assay and were used to set copy number cut-off values corresponding to IHC classification. The generated dPCR cut-off values were used to evaluate the corresponding ∆Ct values (RNA expression results derived from the APIS BCSK). For each target, three ΔCt cut-offs were established, facilitating classification into negative, low, medium, and high expression. Results: The semi-quantitative approach successfully classified target expression levels ( ESR1, PGR, ERBB2, MKI67). Overlap in ΔCt values was observed in central IHC categories for all targets, likely due to tumor heterogeneity and differing measurement methods for IHC and RT-qPCR. Despite these discrepancies, the derived cut-off values define a semi-quantitative scale for each target (Table). Conclusions: This study has established a methodology that utilises target copy number values and IHC protein quantification to establish ΔCt cut-off points. For each target we have defined four distinct groups as negative, low, medium, and high expression. This approach provides a semi-quantitative scale for evaluating each target with the APIS BCSK and serves as a valuable tool for classifying and understanding target expression levels. Table: see text