Avaliação da expressão do GSK3-B, do PD-L1, e do infiltrado linfocítico CD8-positivo em carcinomas uroteliais primários da bexiga

Abstract

Bladder cancer is the ninth most common neoplasm in the world, being malignant more frequent in males. Considering the importance of personalized diagnosis and treatment for bladder cancer, it may be relevant to evaluate correlations between the clinical behavior with the immune infiltrates and other predictive biomarkers. The use of digital pathology in the diagnosis of BC is important due to the use of digital tools that allow a more detailed analysis, improving the quality and efficiency of the diagnosis. The growth of digital pathology and whole slide imaging have created the opportunity to extract more information from histological samples through image analysis Objective: To evaluate the immunoexpression of markers GSK3-β, PD-L1 and CD8 positive cell density in urothelial carcinoma correlating with tumor grade and overall survival. Methodology: This is a retrospective, cross-sectional, observational study. 140 consecutive cases from a referral service were included. After confirmation of the histological diagnosis, the samples were used in the construction of the Tissue Microarray (TMA) and submitted to immunohistochemistry reactions with the markers GSK-3β (clone 27C10), CD8 (m7103) and PDL-1 (22c3) in the manual and/or automated use of the Roche and Dako platforms. The expression were evaluated by two pathologists blindly, and by the open source software called Qupath®. The analysis parameters for CD8 were the presence of positive cells per mm²; GSK3 was evaluated for intensity and percentage of staining to calculate the H-score and PD-L1 was used for analysis of the tumor proportion score (TPS) 0-100. Results: Of 140 samples from the of transurethral resection or total cystectomies (88%), the male sex was predominant (77.9%), patients were older 70 years (66.4%). Immunoexpression of GSK-3β (91%) was present in most cases, as well as PD-L1 (62.9%). The CD8 cell density ranged from 0 to 538 cells/mm². (p00 data-correlations,0002 between PD-L1 and PD-L1). The use of digital pathology software showed excellent correlation with the tumor grade made by the pathologist and by Qupath (Kappa agreement of 0.73) and interclass correlation parameter (ICC) of 0.851. PD-L1 positivity had a 5.81-fold risk of recurrence and high-grade tumors increased the risk of recurrence by 9.09. Conclusion: These data infer that despite the tumor heterogeneity and the microenvironment, the expression of these biomarkers showed to be promising against the therapeutic approach of urothelial cancer when evaluated the presence of the immune infiltrate (CD8) and the 16 immunoexpression of the markers GSK-3β and PD-L1. The use of QuPath as an automated scoring tool can be used to validate the analysis and reading of the PD-L1 and CD8 biomarkers.