Matrix-assisted laser desorption/ionization (MALDI) time-of-flight (TOF) imaging mass spectrometry (IMS) is an evolving technique in cancer diagnostics and combines the advantages of mass spectrometry (proteomics), detection of numerous molecules, and spatial resolution in histological tissue sections and cytological preparations. This method allows the detection of proteins, peptides, lipids, carbohydrates or glycoconjugates and small molecules.Formalin-fixed paraffin-embedded tissue can also be investigated by IMS, thus, this method seems to be an ideal tool for cancer diagnostics and biomarker discovery. It may add information to the identification of tumor margins and tumor heterogeneity. The technique allows tumor typing, especially identification of the tumor of origin in metastatic tissue, as well as grading and may provide prognostic information. IMS is a valuable method for the identification of biomarkers and can complement histology, immunohistology and molecular pathology in various fields of histopathological diagnostics, especially with regard to identification and grading of tumors.

Motivation: Tumor classification using Imaging Mass Spectrometry (IMS) data has a high potential for future applications in pathology. Due to the complexity and size of the data, automated feature extraction and classification steps are required to fully process the data. Deep learning offers an approach to learn feature extraction and classification combined in a single model. Commonly these steps are handled separately in IMS data analysis, hence deep learning offers an alternative strategy worthwhile to explore. Results: Methodologically, we propose an adapted architecture based on deep convolutional networks to handle the characteristics of mass spectrometry data, as well as a strategy to interpret the learned model in the spectral domain based on a sensitivity analysis. The proposed methods are evaluated on two challenging tumor classification tasks and compared to a baseline approach. Competitiveness of the proposed methods are shown on both tasks by studying the performance via cross-validation. Moreover, the learned models are analyzed by the proposed sensitivity analysis revealing biologically plausible effects as well as confounding factors of the considered task. Thus, this study may serve as a starting point for further development of deep learning approaches in IMS classification tasks. Source Code: https://gitlab.informatik.uni-bremen.de/digipath/Deep Learning for Tumor Classification in IMS Data: https://seafile.zfn.uni-bremen.de/d/85c915784e/

Purpose To facilitate the transition of MALDI–MS Imaging (MALDI–MSI) from basic science to clinical application, it is necessary to analyze formalin‐fixed paraffin‐embedded (FFPE) tissues. The aim is to improve in situ tryptic digestion for MALDI–MSI of FFPE samples and determine if similar results would be reproducible if obtained from different sites. Experimental Design FFPE tissues (mouse intestine, human ovarian teratoma, tissue microarray of tumor entities sampled from three different sites) are prepared for MALDI–MSI. Samples are coated with trypsin using an automated sprayer then incubated using deliquescence to maintain a stable humid environment. After digestion, samples are sprayed with CHCA using the same spraying device and analyzed with a rapifleX MALDI Tissuetyper at 50 µm spatial resolution. Data are analyzed using flexImaging, SCiLS, and R. Results Trypsin application and digestion are identified as sources of variation and loss of spatial resolution in the MALDI–MSI of FFPE samples. Using the described workflow, it is possible to discriminate discrete histological features in different tissues and enabled different sites to generate images of similar quality when assessed by spatial segmentation and PCA. Conclusions and Clinical Relevance Spatial resolution and site‐to‐site reproducibility can be maintained by adhering to a standardized MALDI–MSI workflow.

No abstract

Objective. MRL-FasConclusion. Taken together, our findings suggest that IL-18-producing TECs may directly be involved in the pathogenesis of lupus nephritis.