Overcoming challenges in predictive biomarker assessment in breast and head & neck cancer
Overcoming challenges in predictive biomarker assessment in breast and head & neck cancer
This thesis of Maaike-Anna Hempenius focuses on improving patient selection for new cancer therapies by optimising the assessment of predictive biomarkers in pathology. The research focuses on two important biomarkers in breast cancer and head and neck cancer: PD-L1 and HER2. These biomarkers are typically assessed using immunohistochemistry (IHC), a technique that visualises proteins in tumour tissue.
This thesis investigated how reliable these tests are in routine diagnostic practice and to what extent differences between laboratories and testing methods may lead to variation in results. Nationwide analyses showed that PD-L1 testing in breast cancer and head and neck cancer in the Netherlands is largely performed according to national guidelines. However, considerable variation exists between laboratories in the proportion of tumours reported as positive. This variation may influence the selection of patients for immunotherapy.
In addition, several PD-L1 and HER2 testing methods were compared. These studies demonstrated that different IHC assays and antibodies vary in sensitivity, which means that patients may be classified differently depending on the test used. Particularly in tumours with very low HER2 expression, substantial differences between assays were observed. This may lead to differences in the selection of patients for targeted therapy.
Finally, this thesis shows that novel approaches, such as quantitative IHC calibrators and digital image analysis, may contribute to a more standardised and objective assessment of biomarkers. Improved standardisation of biomarker diagnostics is essential to ensure that patients receive the most appropriate treatment.