Strategies for managing multi-patient 3D mass spectrometry imaging data

Area of Science:

• Biomedical Research
• Analytical Chemistry
• Oncology

Background:

• Mass spectrometry imaging (MSI) is crucial for localizing metabolites and proteins in diseased tissues like tumors.
• Current MSI applications, including 2D and 3D imaging, can introduce sampling bias at sample or patient levels.
• Understanding sampling bias is essential for accurate biomarker discovery in complex diseases.

Purpose of the Study:

• To investigate the impact of sampling bias on sample representativeness.
• To assess the effect of sampling bias on the precision of biomarker discovery for bladder cancer histological grading using MSI.
• To evaluate the utility of 3D matrix-assisted laser desorption/ionization (MALDI) MSI for analyzing formalin-fixed paraffin-embedded (FFPE) bladder cancer tissues.

Main Methods:

• Formalin-fixed paraffin-embedded (FFPE) tissues from 14 bladder cancer patients were analyzed using 3D MALDI-MSI.
• Novel outlier detection routines were applied to 3D-MSI data, evaluating digestion efficacy and z-directed regression.
• Data preprocessing involved removing 20% of outlier data before analysis.

Main Results:

• An average of 33% of a tissue sample requires measurement to adequately cover biological variance.
• Outlier detection routines improved the reliability of 3D-MSI data analysis.
• Sampling bias significantly influences result variability, particularly in smaller patient cohorts.

Conclusions:

• 3D MALDI-MSI can be effectively applied to FFPE bladder cancer tissues, demonstrating reproducibility with optimized protocols.
• Addressing sampling bias is critical for precise biomarker discovery in multi-patient MSI studies.
• The presented data analysis workflow offers a pipeline for multi-patient 3D FFPE-MSI studies.