Endocrine Abstracts

Background: Following IHC of CALR, H-scoring was performed by a pathologist. H-scoring was validated by MIF of the same tissue, wherein random-forest machine learning (ML) classifiers were employed to classify cells. ML classifiers were trained to distinguish between pNET cells and tumor stroma using approximately 30% of cells for the respective cell population of interest in each TMA core. Pearson’s correlations were used to evaluate the relationship between H-scoring an...