MSI-based mapping strategies in tumour-heterogeneity

Abstract

Since the early 2000s, considerable innovations in MS technology and associated gene sequencing systems have enabled the "-omics" revolution. The data collected from multiple omics research can be combined to gain a better understanding of cancer's biological activity. Breast and ovarian cancer are among the most common cancers worldwide in women. Despite significant advances in diagnosis, treatment, and subtype identification, breast cancer remains the world's second leading cause of cancer-related deaths in women, with ovarian cancer ranking fifth. Tumour heterogeneity is a significant hurdle in cancer patient prognosis, response to therapy, and metastasis. As such, heterogeneity is one of the most significant and clinically relevant areas of cancer research nowadays. Metabolic reprogramming is a hallmark of malignancy that has been widely acknowledged in recent literature. Metabolic heterogeneity in tumours poses a challenge in developing therapies that exploit metabolic vulnerabilities. Consequently, it is crucial to approach tumour heterogeneity with an unlabeled yet spatially specific read-out of metabolic and genetic information. The advantage of DESI-MSI technology originates from its untargeted nature, which allows for the investigation of thousands of component distributions, at a micrometre scale, in a single experiment. Most notably, using a DESI-MSI clustering approach could potentially offer novel insights into metabolism, providing a method to characterise metabolically distinct sub-regions and subsequently delineate the underlying genetic drivers through genomic analyses. Hence, in this study, we aim to map the inter-and intra-tumour metabolic heterogeneity in breast and ovarian cancer by integrating multimodal MSI-based mapping strategies, comprising DESI and MALDI, with IMC (Imaging Mass Cytometry) analysis of the tumour section, using CyTOF, and high- throughput genetic characterisation of metabolically-distinct regions by transcriptomics. The multimodal analysis workflow was initially performed using sequential breast cancer Patient-Derived Xenografts (PDX) models and was expanded on primary tumour sections. Moreover, a newly developed DESI-MSI friendly, hydroxypropyl-methylcellulose and polyvinylpyrrolidone (HPMC/PVP) hydrogel-based embedding was successfully established to allow simultaneous preparation and analysis of numerous fresh frozen core-size biopsies in the same Tissue Microarray (TMA) block for the investigation of tumour heterogeneity. Additionally, a single section strategy was combined with DESI-MSI coupled to Laser Capture Microdissection (LCM) application to integrate gene expression analysis and Liquid Chromatography-Mass Spectrometry (LC-MS) on the same tissue segment. The developed single section methodology was then tested with multi-region collected ovarian tumours. DESI-MSI-guided spatial transcriptomics was performed for co-registration of different omics datasets on the same regions of interest (ROIs). This co-registration of various omics could unravel possible interactions between distinct metabolic profiles and specific genetic drivers that can lead to intra-tumour heterogeneity. Linking all these findings from MSI-based or guided various strategies allows for a transition from a qualitative approach to a conceptual understanding of the architecture of multiple molecular networks responsible for cellular metabolism in tumour heterogeneity.