Abstract
Purpose :
Retinoblastoma (RB) tumor microenvironment is a multifaceted interplay between the tumor cells, immune cells, stromal cells, extracellular matrix and the non-cellular component. Herein, we aim to analyze the immunohistochemical expression of hypoxia related markers in the enucleated eyes.
Methods :
A retrospective observational study was performed on clinically diagnosed and enucleated globes of RB which were further categorized into low-risk group without tumor invasion into uvea or optic nerve or sclera or extra scleral tissue (n=5) and high-risk group with tumor invasion into uvea or optic nerve or sclera or extrascleral tissue (n=6) based on histopathology features. Formalin fixed paraffin embedded sections were stained with RB1 protein, hypoxia inducible factors like HIF-1, VEGF, TGF-beta along with the exosome markers CD9, CD81 and TSG101. Protein expression pattern, intensity and location in the sections were correlated with tumor pattern, differentiation, tumor invasion into adjacent structures and pathological TNM staging.
Results :
High-risk group were poorly differentiated; showed increased and heterogeneous expression of HIF-1 (83%) and VEGF (50%) while TSG101 showed strong and homogenous expression (83%) in tumor and invasive areas. TGF- b (16%), CD9 (16%) and CD81(16%) showed minimal expression. RB-1 protein expression was retained in 5/6 cases.
Low-risk group were better differentiated and showed increased, homogeneous expression of HIF-1 (60%) and VEGF (80%). There was no expression of TGF-b. Exosome markers TSG101 and CD9 showed homogenous expression in 80% and 60% cases respectively. CD 81 and TGF- b did not show any expression. RB-1 protein expression was retained in 4/5 cases.
Conclusions :
Both high-risk and low-risk groups showed increased expression of hypoxia inducing factors HIF-1 and VEGF along with exosome specific marker TSG101 suggesting a correlation. Tetraspanins CD9, CD 81 showed variable expression in both the groups. The clinical significance needs to be further explored.
This abstract was presented at the 2024 ARVO Annual Meeting, held in Seattle, WA, May 5-9, 2024.