Identification of Raman spectroscopic markers for the characterization of normal and adenocarcinomatous colonic tissues

Raman spectroscopy has been recognised as a valuable analytical tool in biological and medical research. This technique allows probing molecular vibrations of samples without external labels or extensive preparation. This non-destructive optical technique can provide rapid and objective and reproducible measurements of sample biochemistry and identify variations that occur between healthy and diseased tissues. In fact, biochemical changes within tissue may either initiate disease or occur as a result of the disease process. The qualitative analysis of such changes provides important clues in the search for a specific diagnosis and the quantitative analysis of biochemical abnormalities is important in measuring the extent of the disease process, designing therapy and evaluating the efficacy of treatment. In this paper, we discuss one medical application of near-infrared Raman microspectroscopic imaging as a diagnostic tool to investigate, ex vivo, the changes between normal and adenocarcinomatous human colonic tissues. Multivariate statistical analysis was applied on these measured data to identify the molecular composition and distribution of lipids, proteins, mucus and collagens in normal and malignant tissue. Unsupervised hierarchical cluster analysis shows two unsupervised distinct clusters that were assigned to normal and adenocarcinomatous in accordance with conventional histopathological examination. The spectral images allowed good correlation between pseudo-color Raman and histopathological features. (C) 2009 Elsevier Ireland Ltd. All rights reserved.