Immunohistochemical Study of Bladder Cancer Molecular Subtypes and Their Association with PD-L1 Expression

Simple Summary The aim of our study was to stratify bladder cancer patients into their molecular subtypes using a simple and inexpensive immunohistochemical algorithm and further provide any associations with PD-L1 expression. Given the fact that there is a universal lack of predictive biomarkers for immunotherapy, we suggest the possibility of stratifying patients into likely-responders and likely-not-responders to anti-PD-L1 therapy, based on their bladder cancer molecular subtypes. The significant heterogeneity in clinical outcomes among patients with bladder cancer has highlighted the existence of different biological subtypes of muscle-invasive bladder cancer (MIBC) and non-muscle-invasive bladder cancer (NMIBC). Meanwhile, immune checkpoint proteins and their interference with tumor-related immune-evasive strategies has led to the development of several immunotherapeutic drugs targeting programmed death-1 (PD-1) or programmed death ligand-1 (PD-L1). However, the lack of any known biomarker that could predict responses to immunotherapy has led to a more agnostic therapeutic approach. Here, we present a study conducted in 77 bladder cancer (BC) patients (n = 77), ranging from stages pTa to pT2. Tumor specimens were resected via transurethral resection of bladder tumor (TURBT) and consistuted of 24 low-grade (LG) and 53 high-grade (HG) tumors. Patients' tumors were then categorized into molecular subtypes, via immunohistochemistry (CK5/6 and GATA3). Furthermore, all tumor specimens were stained with anti-PD-L1 and demonstrated significant correlations with basal immunophenotype, stage pT2 and HG tumors. As such, we attempted to stratify patients into groups of likely-responders and likely-not-responders to immunotherapy with anti-PD-L1, based on their molecular phenotype. Finally, in acknowledging the fact that there is a universal lack of biomarkers associated with predicting BC response to immunotherapeutic drugs, we tested all tumors for deficiency of mismatch repair proteins (MMR).

Automated summary

Our study aimed to stratify bladder cancer patients into molecular subtypes using a simple and inexpensive immunohistochemical algorithm, and investigate the associations with PD-L1 expression. The heterogeneity in clinical outcomes of bladder cancer has highlighted different biological subtypes. Immunotherapy drugs targeting PD-1/PD-L1 have been developed, but there is a lack of predictive biomarkers. We categorized patients' tumors into molecular subtypes and attempted to predict their response to anti-PD-L1 therapy based on their molecular phenotype. The deficiency of mismatch repair proteins (MMR) was also examined.