Analytical Sensitivity and HER2-Low: To Measure or Not to Measure, That Is the Question

Analytical sensitivity is a fundamental property of all biomarker assays, yet its relationship to the diagnostic and clinical sensitivity of immunohistochemistry biomarkers has only recently begun to be defined. Nationwide Dutch comparison of HER2 immunohistochemistry performance reported a significant correlation between the lower limit of detection (LOD) and staining intensity in low-expressing cores, but no association between LOD and registry-derived HER2-low prevalence. This editorial explains why that null finding should not be interpreted as evidence that analytical sensitivity is unimportant for diagnostic or clinical sensitivity. The following 3 methodological constraints limit the ability of this design to reveal such an association: (1) restriction of LOD values to a narrow high-sensitivity band, (2) ecological correlation of laboratory-level LOD with aggregated case-level categorical scores, and (3) substantial misclassification inherent in the broad, only moderately reproducible HER2-low endpoint. Prior studies measuring LOD and dynamic range show that LOD materially affects low-level HER2 measurement and remains central to fit-for-purpose HER2-low testing.