A multi-omics approach to investigate the complex interplay between muscle-invasive bladder cancer and the host immune response

Abstract

Muscle-invasive bladder cancer (MIBC) prognosis is mainly assessed by clinical cancer stage which is codified using the Tumour-Node-Metastasis (TNM) staging system. However, recent studies have demonstrated that disease progression and thus prognosis is profoundly influenced by the immune context of the tumour microenvironment. Multiplex immunofluorescence was applied on MIBC tissue sections to capture whole slide images and quantify potential prognostic markers related to lymphocytes, macrophages, PD-L1 and tumour buds. Two independent machine learning-based methodologies were implemented and the resulting prognostic models reported that: (i) tumour budding was the most significant feature (HR=2.59, P=0.0091) for the stratification of non-metastatic patients into high or low risk of disease specific death, and (ii) the combination of image, clinical, and spatial features stratified MIBC patients into two risk groups with high statistical significance (P<1E⁻⁰⁵) and greater accuracy than the current clinical gold standard, the TNM staging system . To provide further insights into the tumour-immune microenvironment, spatially resolved differential expression of immunologically relevant proteins was quantified across entire MIBC tissues using a 31-plex spatial profiling platform. Significant alterations in the expression of proteins were identified within different compartments of the tissue related to tumour core, tumour buds, stroma and tumour infiltrating lymphocytes showing that this technology has the capability to capture immunological signatures if applied in a larger heterogeneous sample population. Lastly, to delve into the molecular causes of immune evasion by cancer cells, extracellular vesicles (EVs) were isolated by differential ultra- centrifugation from conditioned media of PD-L1 and PD-L1 knockout human bladder carcinoma cells. Co-culture assays demonstrated that EVs derived from PD-L1 bladder carcinoma cells can impair immune functions by reducing CD8 T-cell proliferation. In addition, 210 EV proteins were identified by proteomic analysis, including two newly identified proteins which are not present in known exosome databases.

Date of Award	1 Dec 2021
Original language	English
Awarding Institution	University of St Andrews
Supervisor	David James Harrison (Supervisor) & Peter David Caie (Supervisor)