2021 journal article

Spatially resolved metabolomic characterization of muscle invasive bladder cancer by mass spectrometry imaging

METABOLOMICS, 17(8).

By: A. Tu n, N. Said * & D. Muddiman n 

co-author countries: United States of America πŸ‡ΊπŸ‡Έ
author keywords: Bladder cancer; IR-MALDESI mass spectrometry imaging; Metabolomics; Lipidomics; Intratumor heterogeneity
MeSH headings : Feasibility Studies; Humans; Metabolomics / methods; Muscle, Smooth / metabolism; Muscle, Smooth / pathology; Spectrometry, Mass, Matrix-Assisted Laser Desorption-Ionization / methods; Urinary Bladder Neoplasms / diagnostic imaging; Urinary Bladder Neoplasms / metabolism; Urinary Bladder Neoplasms / pathology
Source: Web Of Science
Added: August 2, 2021

Muscle invasive bladder cancer (MIBC) is an advanced stage of bladder cancer which poses a severe threat to life. Cancer development is usually accompanied by remarkable alterations in cell metabolism, and hence deep insights into MIBC at the metabolomic level can facilitate the understanding of the biochemical mechanisms involved in the cancer development and progression.In this proof-of-concept study, the optimal cutting temperature (OCT)-embedded MIBC samples were first washed with pure water to remove the polymer compounds which could cause severe signal suppression during mass spectrometry. Further, the tissue sections were analyzed by infrared matrix-assisted laser desorption electrospray ionization mass spectrometry imaging (IR-MALDESI MSI), providing an overview on the spatially resolved metabolomic profiles.The MSI data enabled the discrimination between not only the cancerous and normal tissues, but also the subregions within a tissue section associated with different disease states. Using t-Distributed Stochastic Neighbor Embedding (t-SNE), the hyperdimensional MSI data was mapped into a two-dimensional space to visualize the spectral similarity, providing evidence that metabolomic alterations might have occurred outside the histopathological tumor border. Least absolute shrinkage and selection operator (LASSO) was further employed to classify sample pathology in a pixel-wise manner, yielding excellent prediction sensitivity and specificity up to 96% based on the statistically characteristic spectral features.The results demonstrate great promise of IR-MALDESI MSI to identify molecular changes derived from cancer and unveil tumor heterogeneity, which can potentially promote the discovery of clinically relevant biomarkers and allow for applications in precision medicine.