Abstract
Purpose :
Reduced nuclear expression of BAP-1 is associated with a high risk for metastasis in uveal melanoma. Manual assessment of the expression level may face issues with interobserver reproducibility. This could be improved with digital image analysis (DIA).
Methods :
36 enucleated eyes from St. Erik Eye Hospital, Sweden (years 1984-1985) and Emory Eye Center, USA (years 2009-2017) were included and stained with BAP-1. Retrospective data on patient and tumor characteristics, gene expression class and survival was retrieved. 16 patients were randomized to a training cohort and 20 patients to a validation cohort. The tumor sections from both cohorts were digitally scanned and automatically scored for percentage of BAP-1 positive tumor cells with the QuPath Bioimage analysis software.
Results :
Interobserver concordance was 75 % (Cohen's kappa 0.52) with the manual method and 88-94 % with DIA (Cohen's kappa 0.75-0.88). Positive and negative predictive values for metastasis were 83 % and 100 % with DIA, 82 % and 89 % with the manual method and 78 % and 88 % with gene expression class 2. In binary logistic regression, manual and DIA of BAP-1, but not gene expression class 2, was associated with metastasis. Metastasis-free survival was significantly shorter with low BAP-1 expression as defined by DIA (Log-Rank p=0.009) and gene expression class 2 (Log-Rank p=0.049), but not manual BAP-1 (Log-Rank p=0.105).
Conclusions :
DIA of BAP-1 immunohistochemistry is a competitive alternative to manual assessment as well as gene expression profiling in prognostication of enucleated specimens with uveal melanoma.
This abstract was presented at the 2019 ARVO Annual Meeting, held in Vancouver, Canada, April 28 - May 2, 2019.