Tutorial: multivariate statistical treatment of imaging data for clinical biomarker discovery

Area of Science:

• Oncology
• Analytical Chemistry
• Biotechnology

Background:

• Matrix-assisted laser desorption/ionization (MALDI) tissue imaging is a novel technology with significant potential in cancer research.
• MALDI imaging can identify molecular differences in cancerous versus healthy tissue, including metabolic, growth, and apoptotic processes.
• It offers insights into molecular differentiation beyond classical histological techniques, aiding in tumor characterization and progression prediction.

Purpose of the Study:

• To describe a workflow for the efficient and unambiguous interpretation of MALDI imaging data within a histopathological context.
• To enable histopathologists to leverage MALDI imaging for understanding tumor heterogeneity and therapeutic susceptibility.
• To facilitate the comparison of disease states between patients for biomarker discovery.

Main Methods:

• Utilizing MALDI tissue imaging to generate molecular images of tissue sections.
• Applying statistical preprocessing tools, specifically principal component analysis (PCA) and hierarchical clustering (HC), to MALDI imaging data.
• Integrating molecular images with high-resolution histological images for interpretation.

Main Results:

• MALDI imaging can reveal molecular heterogeneity within tumors, including variations in biochemical pathways and microenvironments.
• Statistical methods like PCA and HC allow for efficient and straightforward interpretation of complex MALDI imaging datasets.
• The described workflow facilitates the correlation of molecular data with histological context, overcoming interpretation challenges.

Conclusions:

• MALDI tissue imaging, when integrated with statistical analysis and histopathology, is a powerful tool for cancer research.
• This approach enhances the understanding of tumor molecular landscapes and aids in predicting therapeutic responses and disease progression.
• The workflow streamlines the analysis of large MALDI imaging datasets, accelerating biomarker discovery and clinical applications.