Validation of artificial intelligence (AI) based pathology solution in scoring of biomarker expression in breast cancer and its impact on clinical decision making in breast cancer therapy

This is a prospective cross-sectional study that will be conducted on cases of breast carcinoma. Glass slides prepared from blocks containing tumour tissue would be stained by immunohistochemistry for ER, PR, and HER2/neu expressions and Ki 67. The IHC slides of ER, PR, HER2 neu and Ki 67 will be de-identified and coded. 4 pathologists will independently score ER, PR, Her 2/neu and Ki 67 expression.

The results of manual scoring will be compared with that of artificial intelligence (AI) based scoring. The scores of IHC expression of ER, PR, Her 2 and ki67 obtained manually from pathologists and AI-based algorithm will be provided to the blinded medical oncologist who will decide and chart out the treatment plan separately for both manual score and the AI-based score based on these biomarkers expression scores. The treatment plan for both arms will be compared to study the impact of the AI-based solutions on clinical decision making and planning of further treatment for a patient with breast cancer.