Molecular assessment of surgical-resection margins of gastric cancer by mass-spectrometric imaging.

摘要: Surgical resection is the main curative option for gastrointestinal cancers. The extent of cancer commonly assessed during surgery by pathologic evaluation (frozen sections of) tissue at resected specimen margin(s) to verify whether present. We compare this method an alternative procedure, desorption electrospray ionization mass spectrometric imaging (DESI-MSI), 62 banked human cancerous and normal gastric-tissue samples. In DESI-MSI, microdroplets strike sample, resulting splash enters a spectrometer, statistical analysis, here, Lasso (which stands least absolute shrinkage selection operator which multiclass logistic regression with L1 penalty), applied classify tissues based on molecular information obtained directly from DESI-MSI. methodology developed 28 frozen training samples clear histopathologic diagnosis showed overall accuracy value 98% 12,480 pixels evaluated in cross-validation (CV), 97% when completely independent set was tested. By applying additional spatial smoothing technique, both CV 99% compared histological diagnoses. To test our clinical use, we it total 21 tissue-margin prospectively nine gastric-cancer patients. results suggest that DESI-MSI/Lasso may be valuable routine intraoperative assessment margins surgery.