Imaging Mass Spectrometry for the Classification of Melanoma Based on BRAF/NRAS Mutational Status.

Mutations of the oncogenes v-raf murine sarcoma viral oncogene homolog B1 (BRAF) and neuroblastoma RAS viral oncogene homolog (NRAS) are the most frequent genetic alterations in melanoma and are mutually exclusive. BRAF V600 mutations are predictive for response to the two BRAF inhibitors vemurafenib and dabrafenib and the mitogen-activated protein kinase kinase (MEK) inhibitor trametinib. However, inter- and intra-tumoral heterogeneity and the development of acquired resistance to BRAF inhibitors have important clinical implications. Here, we investigated and compared the molecular profile of BRAF and NRAS mutated and wildtype melanoma patients' tissue samples using imaging mass spectrometry-based proteomic technology, to identify specific molecular signatures associated with the respective tumors. SCiLSLab and R-statistical software were used to classify peptide profiles using linear discriminant analysis and support vector machine models optimized with two internal cross-validation methods (leave-one-out, k-fold). Classification models showed molecular differences between BRAF and NRAS mutated melanoma, and identification of both was possible with an accuracy of 87-89% and 76-79%, depending on the respective classification method applied. In addition, differential expression of some predictive proteins, such as histones or glyceraldehyde-3-phosphate-dehydrogenase, correlated with BRAF or NRAS mutation status. Overall, these findings provide a new molecular method to classify melanoma patients carrying BRAF and NRAS mutations and help provide a broader view of the molecular characteristics of these patients that may help understand the signaling pathways and interactions involving the altered genes.

Multicenter Evaluation of Tissue Classification by Matrix-Assisted Laser Desorption/Ionization Mass Spectrometry Imaging.

Many studies have demonstrated that tissue phenotyping (tissue typing) based on mass spectrometric imaging data is possible; however, comprehensive studies assessing variation and classifier transferability are largely lacking. This study evaluated the generalization of tissue classification based on Matrix Assisted Laser Desorption/Ionization (MALDI) mass spectrometric imaging (MSI) across measurements performed at different sites. Sections of a tissue microarray (TMA) consisting of different formalin-fixed and paraffin-embedded (FFPE) human tissue samples from different tumor entities (leiomyoma, seminoma, mantle cell lymphoma, melanoma, breast cancer, and squamous cell carcinoma of the lung) were prepared and measured by MALDI-MSI at different sites using a standard protocol (SOP). Technical variation was deliberately introduced on two separate measurements via a different sample preparation protocol and a MALDI Time of Flight mass spectrometer that was not tuned to optimal performance. Using standard data preprocessing, a classification accuracy of 91.4% per pixel was achieved for intrasite classifications. When applying a leave-one-site-out cross-validation strategy, accuracy per pixel over sites was 78.6% for the SOP-compliant data sets and as low as 36.1% for the mistuned instrument data set. Data preprocessing designed to remove technical variation while retaining biological information substantially increased classification accuracy for all data sets with SOP-compliant data sets improved to 94.3%. In particular, classification accuracy of the mistuned instrument data set improved to 81.3% and from 67.0% to 87.8% per pixel for the non-SOP-compliant data set. We demonstrate that MALDI-MSI-based tissue classification is possible across sites when applying histological annotation and an optimized data preprocessing pipeline to improve generalization of classifications over technical variation and increasing overall robustness.

Cross-Normalization of MALDI Mass Spectrometry Imaging Data Improves Site-to-Site Reproducibility.

Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) is an established tool for the investigation of formalin-fixed paraffin-embedded (FFPE) tissue samples and shows a high potential for applications in clinical research and histopathological tissue classification. However, the applicability of this method to serial clinical and pharmacological studies is often hampered by inevitable technical variation and limited reproducibility. We present a novel spectral cross-normalization algorithm that differs from the existing normalization methods in two aspects: (a) it is based on estimating the full statistical distribution of spectral intensities and (b) it involves applying a non-linear, mass-dependent intensity transformation to align this distribution with a reference distribution. This method is combined with a model-driven resampling step that is specifically designed for data from MALDI imaging of tryptic peptides. This method was performed on two sets of tissue samples: a single human teratoma sample and a collection of five tissue microarrays (TMAs) of breast and ovarian tumor tissue samples (N = 241 patients). The MALDI MSI data was acquired in two labs using multiple protocols, allowing us to investigate different inter-lab and cross-protocol scenarios, thus covering a wide range of technical variations. Our results suggest that the proposed cross-normalization significantly reduces such batch effects not only in inter-sample and inter-lab comparisons but also in cross-protocol scenarios. This demonstrates the feasibility of cross-normalization and joint data analysis even under conditions where preparation and acquisition protocols themselves are subject to variation.

Using the Chemical Noise Background in MALDI Mass Spectrometry Imaging for Mass Alignment and Calibration.

Matrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) is an established tool for the investigation of formalin fixed paraffin embedded (FFPE) tissue samples and shows a high potential for applications in clinical research and histopathological diagnosis. The applicability and accuracy of this method, however, heavily depends on the quality of the acquired data, and in particular mass misalignment in axial time-of-flight (TOF) MSI continues to be a serious issue. We present a mass alignment and recalibration method that is specifically designed to operate on MALDI peptide imaging data. The proposed method exploits statistical properties of the characteristic chemical noise background observed in peptide imaging experiments. By comparing these properties to a theoretical peptide mass model, the effective mass shift of each spectrum is estimated and corrected. The method was evaluated on a cohort of 31 FFPE tissue samples, pursuing a statistical validation approach to estimate both the reduction of relative misalignment, as well as the increase in absolute mass accuracy. Our results suggest that a relative mass precision of approximately 5 ppm and an absolute accuracy of approximately 20 ppm are achievable using our method.

Deep learning for tumor classification in imaging mass spectrometry.

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. Since mass spectra exhibit certain structural similarities to image data, deep learning may offer a promising strategy for classification of IMS data as it has been successfully applied to image classification. 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 algorithmically challenging tumor classification tasks and compared to a baseline approach. Competitiveness of the proposed methods is 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 tasks. Thus, this study may serve as a starting point for further development of deep learning approaches in IMS classification tasks. Availability and implementation: https://gitlab.informatik.uni-bremen.de/digipath/Deep_Learning_for_Tumor_Classification_in_IMS. Contact: jbehrmann@uni-bremen.de or christianetmann@uni-bremen.de. Supplementary information: Supplementary data are available at Bioinformatics online.

Accelerated pre-senile systemic amyloidosis in PACAP knockout mice - a protective role of PACAP in age-related degenerative processes.

Dysregulation of neuropeptides may play an important role in aging-induced impairments. Among them, pituitary adenylate cyclase-activating polypeptide (PACAP) is a potent cytoprotective peptide that provides an endogenous control against a variety of tissue-damaging stimuli. We hypothesized that the progressive decline of PACAP throughout life and the well-known general cytoprotective effects of PACAP lead to age-related pathophysiological changes in PACAP deficiency, supported by the increased vulnerability to various stressors of animals partially or totally lacking PACAP. Using young and aging CD1 PACAP knockout (KO) and wild type (WT) mice, we demonstrated pre-senile amyloidosis in young PACAP KO animals and showed that senile amyloidosis appeared accelerated, more generalized, more severe, and affected more individuals. Histopathology showed age-related systemic amyloidosis with mainly kidney, spleen, liver, skin, thyroid, intestinal, tracheal, and esophageal involvement. Mass spectrometry-based proteomic analysis, reconfirmed with immunohistochemistry, revealed that apolipoprotein-AIV was the main amyloid protein in the deposits together with several accompanying proteins. Although the local amyloidogenic protein expression was disturbed in KO animals, no difference was found in laboratory lipid parameters, suggesting a complex pathway leading to increased age-related degeneration with amyloid deposits in the absence of PACAP. In spite of no marked inflammatory histological changes or blood test parameters, we detected a disturbed cytokine profile that possibly creates a pro-inflammatory milieu favoring amyloid deposition. In summary, here we describe accelerated systemic senile amyloidosis in PACAP gene-deficient mice, which might indicate an early aging phenomenon in this mouse strain. Thus, PACAP KO mice could serve as a model of accelerated aging with human relevance. © 2018 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland.

Proteomics in Pathology.

Proteomic approaches are of growing importance in the biologist's toolbox. It greatly benefited from past and recent advances in sampling, chemical processing, mass spectrometry (MS) instrumentation, and data processing. MS-based analysis of proteins is now in the process of being translated in pathology for objective diagnoses. In this viewpoint, we present the workflows that we think are the most promising for applications in pathology. We also comment what we think are prerequisites for a successful translational implementation.

Reliable Entity Subtyping in Non-small Cell Lung Cancer by Matrix-assisted Laser Desorption/Ionization Imaging Mass Spectrometry on Formalin-fixed Paraffin-embedded Tissue Specimens.

Histopathological subtyping of non-small cell lung cancer (NSCLC) into adenocarcinoma (ADC), and squamous cell carcinoma (SqCC) is of utmost relevance for treatment stratification. However, current immunohistochemistry (IHC) based typing approaches on biopsies are imperfect, therefore novel analytical methods for reliable subtyping are needed. We analyzed formalin-fixed paraffin-embedded tissue cores of NSCLC by Matrix-assisted laser desorption/ionization (MALDI) imaging on tissue microarrays to identify and validate discriminating MALDI imaging profiles for NSCLC subtyping. 110 ADC and 98 SqCC were used to train a Linear Discriminant Analysis (LDA) model. Results were validated on a separate set of 58 ADC and 60 SqCC. Selected differentially expressed proteins were identified by tandem mass spectrometry and validated by IHC. The LDA classification model incorporated 339 m/z values. In the validation cohort, in 117 cases (99.1%) MALDI classification on tissue cores was in accordance with the pathological diagnosis made on resection specimen. Overall, three cases in the combined cohorts were discordant, after reevaluation two were initially misclassified by pathology whereas one was classified incorrectly by MALDI. Identification of differentially expressed peptides detected well-known IHC discriminators (CK5, CK7), but also less well known differentially expressed proteins (CK15, HSP27). In conclusion, MALDI imaging on NSCLC tissue cores as small biopsy equivalents is capable to discriminate lung ADC and SqCC with a very high accuracy. In addition, replacing multislide IHC by an one-slide MALDI approach may also save tissue for subsequent predictive molecular testing. We therefore advocate to pursue routine diagnostic implementation strategies for MALDI imaging in solid tumor typing.

A new classification method for MALDI imaging mass spectrometry data acquired on formalin-fixed paraffin-embedded tissue samples.

Matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI IMS) shows a high potential for applications in histopathological diagnosis, and in particular for supporting tumor typing and subtyping. The development of such applications requires the extraction of spectral fingerprints that are relevant for the given tissue and the identification of biomarkers associated with these spectral patterns. We propose a novel data analysis method based on the extraction of characteristic spectral patterns (CSPs) that allow automated generation of classification models for spectral data. Formalin-fixed paraffin embedded (FFPE) tissue samples from N=445 patients assembled on 12 tissue microarrays were analyzed. The method was applied to discriminate primary lung and pancreatic cancer, as well as adenocarcinoma and squamous cell carcinoma of the lung. A classification accuracy of 100% and 82.8%, resp., could be achieved on core level, assessed by cross-validation. The method outperformed the more conventional classification method based on the extraction of individual m/z values in the first application, while achieving a comparable accuracy in the second. LC-MS/MS peptide identification demonstrated that the spectral features present in selected CSPs correspond to peptides relevant for the respective classification. This article is part of a Special Issue entitled: MALDI Imaging, edited by Dr. Corinna Henkel and Prof. Peter Hoffmann.

Typing of colon and lung adenocarcinoma by high throughput imaging mass spectrometry.

In advanced tumor stages, diagnosis is frequently made from metastatic tumor tissue. In some cases, the identification of the tumor of origin may be difficult by histology alone. In this setting, immunohistochemical and molecular biological methods are often required. In a subset of tumors definite diagnosis cannot be achieved. Thus, additional new diagnostic methods are required for precise tumor subtyping. Mass spectrometric methods are of special interest for the discrimination of different tumor types. We investigated whether it is possible to discern adenocarcinomas of colon and lung using high-throughput imaging mass spectrometry on formalin-fixed paraffin-embedded tissue microarrays. 101 primary adenocarcinoma of the colon and 91 primary adenocarcinoma of the lung were used to train a Linear Discriminant Analysis model. Results were validated on an independent set of 116 colonic and 75 lung adenocarcinomas. In the validation cohort 109 of 116 patients with colonic and 67 of 75 patients with lung adenocarcinomas were correctly classified. The ability to define proteomic profiles capable to discern different tumor types promises a valuable tool in cancer diagnostics and might complement current approaches. This article is part of a Special Issue entitled: MALDI Imaging, edited by Dr. Corinna Henkel and Prof. Peter Hoffmann.

Imaging mass spectrometry analysis of renal amyloidosis biopsies reveals protein co-localization with amyloid deposits.

Amyloidosis is a heterogeneous group of protein misfolding diseases characterized by deposition of amyloid proteins. The kidney is frequently affected, especially by immunoglobulin light chain (AL) and serum amyloid A (SAA) amyloidosis as the most common subgroups. Current diagnosis relies on histopathological examination, Congo red staining, or electron microscopy. Subtyping is done by immunohistochemistry; however, commercially available antibodies lack specificity. The purpose of this study was to identify and map amyloid proteins in formalin-fixed paraffin-embedded tissue sections using matrix-assisted laser desorption/ionization imaging mass spectrometry (MALDI IMS) coupled with liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis in an integrated workflow. Renal amyloidosis and non-amyloidosis biopsies were processed for histological and MS analysis. Mass spectra corresponding to the congophilic areas were directly linked to the histological and MS images for correlation studies. Peptides for SAA and AL were detected by MALDI IMS associated to Congo red-positive areas. Sequence determination of amyloid peptides by LC-MS/MS analysis provided protein distribution and identification. Serum amyloid P component, apolipoprotein E, and vitronectin proteins were identified in both AA and AL amyloidosis, showing a strong correlation with Congo red-positive regions. Our findings highlight the utility of MALDI IMS as a new method to type amyloidosis in histopathological routine material and characterize amyloid-associated proteins that may provide insights into the pathogenetic process of amyloid formation.

Investigation of neutrophilic peptides in periprosthetic tissue by matrix-assisted laser desorption ionisation time-of-flight imaging mass spectrometry.

PURPOSE: The accurate diagnosis of periprosthetic joint infection (PJI) relies on clinical investigation, laboratory parameters, radiological methods, sterile joint aspiration for synovial fluid leucocyte count and microbiological analysis and tissue sampling for histopathology. Due to the limits in specificity and sensitivity of these methods, molecular techniques and new biomarkers were introduced into the diagnostic procedure. Histological examination is related to the amount of neutrophils in the periprosthetic tissue in frozen sections and formalin-fixed paraffin embedded material (FFPE). However, the threshold of neutrophils per defined area of tissue among various studies is very inconsistent. METHODS: We have applied matrix-assisted laser desorption ionisation time-of-flight imaging mass spectrometry (MALDI IMS) to a total of 32 periprosthetic tissue samples of patients with PJI to detect peptides associated with areas of neutrophil infiltration. RESULTS: Specific peaks associated with a high amount of neutrophils were detected. Of these m/z peaks, four could be assigned to predictive neutrophil molecules. These peptides include annexin A1, calgizzarin (S100A11), calgranulin C (S100A12) and histone H2A. By MALDI IMS, these peptides could be shown to be co-localised with the infiltration of neutrophils in the immediate vicinity of the periprosthetic interface, whereas more distant areas did not show neutrophil invasion or infection-related peptides. CONCLUSIONS: MALDI IMS is a new method allowing identification of neutrophil peptides in periprosthetic tissues and may be a surrogate for counting neutrophils as an objective parameter for PJI.

Imaging mass spectrometry to discriminate breast from pancreatic cancer metastasis in formalin-fixed paraffin-embedded tissues.

Diagnosis of the origin of metastasis is mandatory for adequate therapy. In the past, classification of tumors was based on histology (morphological expression of a complex protein pattern), while supportive immunohistochemical investigation relied only on few "tumor specific" proteins. At present, histopathological diagnosis is based on clinical information, morphology, immunohistochemistry, and may include molecular methods. This process is complex, expensive, requires an experienced pathologist and may be time consuming. Currently, proteomic methods have been introduced in various clinical disciplines. MALDI imaging MS combines detection of numerous proteins with morphological features, and seems to be the ideal tool for objective and fast histopathological tumor classification. To study a special tumor type and to identify predictive patterns that could discriminate metastatic breast from pancreatic carcinoma MALDI imaging MS was applied to multitissue paraffin blocks. A statistical classification model was created using a training set of primary carcinoma biopsies. This model was validated on two testing sets of different breast and pancreatic carcinoma specimens. We could discern breast from pancreatic primary tumors with an overall accuracy of 83.38%, a sensitivity of 85.95% and a specificity of 76.96%. Furthermore, breast and pancreatic liver metastases were tested and classified correctly.

Biological injection therapy with leukocyte-poor platelet-rich plasma induces cellular alterations, enhancement of lubricin, and inflammatory downregulation in vivo in human knees: A controlled, prospective human clinical trial based on mass spectrometry imaging analysis.

Objective: To investigate the in vivo biological effects of leukocyte-poor platelet-rich plasma (LpPRP) treatment in human synovial layer to establish the cellular basis for a prolonged clinical improvement. Methods: Synovial tissues (n = 367) were prospectively collected from patients undergoing arthroscopic surgery. Autologous-conditioned plasma, LpPRP, was injected into the knees of 163 patients 1-7 days before surgery to reduce operative trauma and inflammation, and to induce the onset of regeneration. A total of 204 patients did not receive any injection. All samples were analyzed by mass spectrometry imaging. Data analysis was evaluated by clustering, classification, and investigation of predictive peptides. Peptide identification was done by tandem mass spectrometry and database matching. Results: Data analysis revealed two major clusters belonging to LpPRP-treated (LpPRP-1) and untreated (LpPRP-0) patients. Classification analysis showed a discrimination accuracy of 82%-90%. We identified discriminating peptides for CD45 and CD29 receptors (receptor-type tyrosine-protein phosphatase C and integrin beta 1), indicating an enhancement of musculoskeletal stem cells, as well as an enhancement of lubricin, collagen alpha-1-(I) chain, and interleukin-receptor-17-E, dampening the inflammatory reaction in the LpPRP-1 group following LpPRP injection. Conclusions: We could demonstrate for the first time that injection therapy using "autologic-conditioned biologics" may lead to cellular changes in the synovial membrane that might explain the reported prolonged beneficial clinical effects. Here, we show in vivo cellular changes, possibly based on muscular skeletal stem cell alterations, in the synovial layer. The gliding capacities of joints might be improved by enhancing of lubricin, anti-inflammation by activation of interleukin-17 receptor E, and reduction of the inflammatory process by blocking interleukin-17.

A Comparison of Different Sample Processing Protocols for MALDI Imaging Mass Spectrometry Analysis of Formalin-Fixed Multiple Myeloma Cells.

Sample processing of formalin-fixed specimens constitutes a major challenge in molecular profiling efforts. Pre-analytical factors such as fixative temperature, dehydration, and embedding media affect downstream analysis, generating data dependent on technical processing rather than disease state. In this study, we investigated two different sample processing methods, including the use of the cytospin sample preparation and automated sample processing apparatuses for proteomic analysis of multiple myeloma (MM) cell lines using imaging mass spectrometry (IMS). In addition, two sample-embedding instruments using different reagents and processing times were considered. Three MM cell lines fixed in 4% paraformaldehyde were either directly centrifuged onto glass slides using cytospin preparation techniques or processed to create paraffin-embedded specimens with an automatic tissue processor, and further cut onto glass slides for IMS analysis. The number of peaks obtained from paraffin-embedded samples was comparable between the two different sample processing instruments. Interestingly, spectra profiles showed enhanced ion yield in cytospin compared to paraffin-embedded samples along with high reproducibility compared to the sample replicate.

Cardiac and skeletal muscle expression of mutant ß-myosin heavy chains, degree of functional impairment and phenotypic heterogeneity in hypertrophic cardiomyopathy.

Several mutations in distinct genes, all coding for sarcomeric proteins, have been reported in unrelated kindreds with familial hypertrophic cardiomyopathy (FHC). We have identified nine individuals from three families harboring two distinct mutations in one copy of the ß-myosin heavy chain (ß-MHC) gene. In this study, the expression of the mutant ß-myosin protein isoform, isolated from slow-twitch fibers of skeletal muscle, was demonstrated by Northern and Western blot analysis; this myosin showed a decreased in vitro motility activity and produced a lower actin-activated ATPase activity. Isometric tension, measured in single slow-twitch fibers isolated from the affected individuals, also showed a significant decrease. The degree of impairment of ß-myosin function, as well as the loss in isometric tension development, were strictly dependent on the amount of the isoform transcribed from the mutated allele. Interestingly, a strong correlation was also demonstrated between mutant ß-myosin content and clinical features of FHC. On the other hand, we were unable to detect any correlation between mutant ß-myosin expression and degree of cardiac hypertrophy, thereby strengthening the hypothesis that hypertrophy, one of the hallmarks of FHC, might not necessarily be related to the clinical evolution of this disease. These findings lend support to the notion that additional factors rather than the mutated gene may play a pathogenetic role in cardiac wall thickening, whereas the prognosis appears to be strongly related to the amount of mutant protein.

Detection of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) including Variant Analysis by Mass Spectrometry in Placental Tissue.

Among neonates, tested positive for SARS-CoV-2, the majority of infections occur through postpartum transmission. Only few reports describe intrauterine or intrapartum SARS-CoV-2 infections in newborns. To understand the route of transmission, detection of the virus or virus nucleic acid in the placenta and amniotic tissue are of special interest. Current methods to detect SARS-CoV-2 in placental tissue are immunohistochemistry, electron microscopy, in-situ hybridization, polymerase chain reaction (PCR) and next-generation sequencing. Recently, we described an alternative method for the detection of viral ribonucleic acid (RNA), by combination of reverse transcriptase-PCR and mass spectrometry (MS) in oropharyngeal and oral swabs. In this report, we could detect SARS-CoV-2 in formal-fixed and paraffin-embedded (FFPE) placental and amniotic tissue by multiplex RT-PCR MS. Additionally, we could identify the British variant (B.1.1.7) of the virus in this tissue by the same methodology. Combination of RT-PCR with MS is a fast and easy method to detect SARS-CoV-2 viral RNA, including specific variants in FFPE tissue.

Robust subtyping of non-small cell lung cancer whole sections through MALDI mass spectrometry imaging.

Subtyping of the most common non-small cell lung cancer (NSCLC) tumor types adenocarcinoma (ADC) and squamous cell carcinoma (SqCC) is still a challenge in the clinical routine and a correct diagnosis is crucial for an adequate therapy selection. Matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) has shown potential for NSCLC subtyping but is subject to strong technical variability and has only been applied to tissue samples assembled in tissue microarrays (TMAs). To our knowledge, a successful transfer of a classifier from TMAs to whole sections, which are generated in the standard clinical routine, has not been presented in the literature as of yet. We introduce a classification algorithm using extensive preprocessing and a classifier (either a neural network or a linear discriminant analysis (LDA)) to robustly classify whole sections of ADC and SqCC lung tissue. The classifiers were trained on TMAs and validated and tested on whole sections. Vital for a successful application on whole sections is the extensive preprocessing and the use of whole sections for hyperparameter selection. The classification system with the neural network/LDA results in 99.0%/98.3% test accuracy on spectra level and 100.0%/100.0% test accuracy on whole section level, respectively, and, therefore, provides a powerful tool to support the pathologist's decision making process. The presented method is a step further towards a clinical application of MALDI MSI and artificial intelligence for subtyping of NSCLC tissue sections.

Multimodal Lung Cancer Subtyping Using Deep Learning Neural Networks on Whole Slide Tissue Images and MALDI MSI.

Artificial intelligence (AI) has shown potential for facilitating the detection and classification of tumors. In patients with non-small cell lung cancer, distinguishing between the most common subtypes, adenocarcinoma (ADC) and squamous cell carcinoma (SqCC), is crucial for the development of an effective treatment plan. This task, however, may still present challenges in clinical routine. We propose a two-modality, AI-based classification algorithm to detect and subtype tumor areas, which combines information from matrix-assisted laser desorption/ionization (MALDI) mass spectrometry imaging (MSI) data and digital microscopy whole slide images (WSIs) of lung tissue sections. The method consists of first detecting areas with high tumor cell content by performing a segmentation of the hematoxylin and eosin-stained (H&E-stained) WSIs, and subsequently classifying the tumor areas based on the corresponding MALDI MSI data. We trained the algorithm on six tissue microarrays (TMAs) with tumor samples from N = 232 patients and used 14 additional whole sections for validation and model selection. Classification accuracy was evaluated on a test dataset with another 16 whole sections. The algorithm accurately detected and classified tumor areas, yielding a test accuracy of 94.7% on spectrum level, and correctly classified 15 of 16 test sections. When an additional quality control criterion was introduced, a 100% test accuracy was achieved on sections that passed the quality control (14 of 16). The presented method provides a step further towards the inclusion of AI and MALDI MSI data into clinical routine and has the potential to reduce the pathologist's work load. A careful analysis of the results revealed specific challenges to be considered when training neural networks on data from lung cancer tissue.

Imaging Mass Spectrometry-Based Proteomic Analysis to Differentiate Melanocytic Nevi and Malignant Melanoma.

The discrimination of malignant melanoma from benign nevi may be difficult in some cases. For this reason, immunohistological and molecular techniques are included in the differential diagnostic toolbox for these lesions. These methods are time consuming when applied subsequently and, in some cases, no definitive diagnosis can be made. We studied both lesions by imaging mass spectrometry (IMS) in a large cohort (n = 203) to determine a different proteomic profile between cutaneous melanomas and melanocytic nevi. Sample preparation and instrument setting were tested to obtain optimal results in term of data quality and reproducibility. A proteomic signature was found by linear discriminant analysis to discern malignant melanoma from benign nevus (n = 113) with an overall accuracy of >98%. The prediction model was tested in an independent set (n = 90) reaching an overall accuracy of 93% in classifying melanoma from nevi. Statistical analysis of the IMS data revealed mass-to-charge ratio (m/z) peaks which varied significantly (Area under the receiver operating characteristic curve > 0.7) between the two tissue types. To our knowledge, this is the largest IMS study of cutaneous melanoma and nevi performed up to now. Our findings clearly show that discrimination of melanocytic nevi from melanoma is possible by IMS.