MALDI imaging mass spectrometry reveals multiple clinically relevant masses in colorectal cancer using large-scale tissue microarrays

For identification of clinically relevant masses to predict status, grade, relapse and prognosis of colorectal cancer we applied MALDI imaging mass spectrometry (IMS) to a tissue micro array (TMA) containing formalin-fixed and paraffin-embedded tissue samples from 349 patients. Analysis of our MALDI-IMS data revealed 27 different m/z signals associated with epithelial structures. Comparison of these signals showed significant association with status, grade and Ki-67 labeling index. 15 out of 27 IMS signals revealed a significant association with survival. For 7 signals (m/z 654, 776, 788, 904, 944, 975, and 1013) the absence, and for 8 signals (m/z 643, 678, 836, 886, 898, 1095, 1459, and 1477) the presence was associated with decreased life expectancy, including 5 masses (m/z 788, 836, 904, 944, and 1013) that provided prognostic information independently from the established prognosticators pT and pN. Combination of these 5 masses resulted in a 3-step classifier that provided prognostic information superior to univariate analysis. In addition, a total of 19 masses were associated with tumor stage, grade, metastasis, and cell proliferation. Our data demonstrate the suitability of combining IMS and large-scale TMAs to simultaneously identify and validate clinically useful molecular marker.