Imaging mass spectrometry (IMS) to discriminate colon from lung adenocarcinoma using formalin-fixed paraffin-embedded (FFPE) tissue

Background There is an increasing need of novel diagnostic and prognostic biomarkers that discriminate between primary tumors. In this study, we specifically addressed the question whether it is possible to discern colon from lung carcinomas at the site of origin using high-throughput imaging mass spectrometry (IMS) technology on formalin-fixed paraffin-embedded (FFPE) tissue microarrays (TMAs). The ability to define proteomic profiles capable of such tumor typing makes IMS a valuable tool in cancer diagnostics and might complement currently used approaches. Methods TMAs were collected from patients with primary colon and lung adenocarcinoma. For each TMA specimen, one section was used for hematoxylin/eosin staining and another section was mounted onto an ITO-coated glass slide for IMS analysis. Sections were on-tissue digested with trypsin (0.1 μg/μL), and coated with CHCA matrix solution (7 mg/ml). IMS data acquisition was performed with an Autoflex Speed TOF/TOF mass spectrometer (Bruker Daltonik GmbH) at spatial resolution of 100 μm. A classification model was generated and validated with a linear discriminant analysis (LDA) algorithm model using SCiLS Lab software. Results The LDA classification model discerned colon from lung primary carcinoma with a sensitivity of 99.17% and a specificity of 80.31%. Furthermore, the classification model was applied to colon adenocarcinoma lung metastases. High similarity of protein expression profiles between the primary colon tumors and the colon metastases profiles were consistent with histopathological diagnosis. Conclusions Differentiation between colon and lung adenocarcinoma is possible on FFPE tissues by IMS. This expression map analysis may further provide a basis for subtyping metastasis from colon and lung carcinoma.