Proteomic classification of primary carcinomas using matrix-assisted laser desorption/ionization (MALDI) imaging

Background Correct histopathological diagnosis is mandatory for proper patient management. At present diagnosis relies on histology, histochemistry and immunohistochemistry or in some cases molecular methods. Since, in many cases only a limited amount of tissue is available- methods which require only on slide and provide a maximum amount of information are desirable. MALDI-imaging meets these requirements and provides information about numerous peptides and their spatial distribution. The aim of this study is to identify a proteomic signature able to discriminate between four tumor types using MALDI imaging on formalin-fixed paraffin-embedded (FFPE) tissue microarrays (TMAs). Methods We investigated eight TMAs consisting of needle core biopsies from 49 breast, 80 pancreatic, 73 lung, and 91 colon tumor biopsies. Each TMA section was stained (H&E) and examined by a pathologist that digitally marked cancer regions. 5μm thick sections were in-situ trypsin digested (0.1 μg/μl) and sprayed with alphacyano- 4-hydroxycinnamic acid matrix solution (7 mg/ml in 50/50 acetonitrile/0.5%TFA) using an ImagePrep sprayer (Bruker Daltonik). MALDI imaging data were acquired at a spatial resolution of 100 μm using an Autoflex Speed TOF/TOF system. Each dataset was analyzed and imaged with FlexImaging (Bruker Daltonik) and SCiLS Lab (SCiLS GmbH) software. Four TMA spectra datasets, each containing a specific tumor type, were used to train the linear discriminant analysis (LDA) model for classification analysis. Another subset of spectra from four different TMAs was used to validate the LDA model. Results A set of 300 most intense peaks were evaluated between samples by LDA model, which classified all different tumor types presented in this study. Specifically, patients with breast, colonic, lung, and pancreatic tumor were correctly classified in 91.36, 96.74, 91.75, and 74.48% of cases respectively. This classification model was successfully applied to liver metastases of the four tumor types. Conclusions Our data confirm that MALDI imaging represents a valuable method to distinguish different tumor types with great potential in diagnostics.