Metabolomic differentiation of benign vs malignant pulmonary nodules with high specificity via high-resolution mass spectrometry analysis of patient sera

An end-to-end deep learning method for mass spectrometry data analysis to reveal disease-specific metabolic profiles

Untargeted metabolomic analysis using mass spectrometry provides comprehensive metabolic profiling, but its medical application faces challenges of complex data processing, high inter-batch variability, and unidentified metabolites. Here, we present DeepMSProfiler, an explainable deep-learning-based method, enabling end-to-end analysis on raw metabolic signals with output of high accuracy and reliability. Using cross-hospital 859 human serum samples from lung adenocarcinoma, benign lung nodules, and healthy individuals, DeepMSProfiler successfully differentiates the metabolomic profiles of different groups (AUC 0.99) and detects early-stage lung adenocarcinoma (accuracy 0.961). Model flow and ablation experiments demonstrate that DeepMSProfiler overcomes inter-hospital variability and effects of unknown metabolites signals. Our ensemble strategy removes background-category phenomena in multi-classification deep-learning models, and the novel interpretability enables direct access to disease-related metabolite-protein networks. Further applying to lipid metabolomic data unveils correlations of important metabolites and proteins. Overall, DeepMSProfiler offers a straightforward and reliable method for disease diagnosis and mechanism discovery, enhancing its broad applicability.

High-Performance Metabolic Profiling of High-Risk Thyroid Nodules by ZrMOF Hybrids

Thyroid nodules (TNs) have emerged as the most prevalent endocrine disorder in China. Fine-needle aspiration (FNA) remains the standard diagnostic method for assessing TN malignancy, although a majority of FNA results indicate benign conditions. Balancing diagnostic accuracy while mitigating overdiagnosis in patients with benign nodules poses a significant clinical challenge. Precise, noninvasive, and high-throughput screening methods for high-risk TN diagnosis are highly desired but remain less explored. Developing such approaches can improve the accuracy of noninvasive methods like ultrasound imaging and reduce overdiagnosis of benign nodule patients caused by invasive procedures. Herein, we investigate the application of gold-doped zirconium-based metal-organic framework (ZrMOF/Au) nanostructures for metabolic profiling of thyroid diseases. This approach enables the efficient extraction of urine metabolite fingerprints with high throughput, low background noise, and reproducibility. Utilizing partial least-squares discriminant analysis and four machine learning models, including neural network (NN), random forest (RF), logistic regression (LR), and support vector machine (SVM), we achieved an enhanced diagnostic accuracy (98.6%) for discriminating thyroid cancer (TC) from low-risk TNs by using a diagnostic panel. Through the analysis of metabolic differences, potential pathway changes between benign nodule and malignancy are identified. This work explores the potential of rapid thyroid disease screening using the ZrMOF/Au-assisted LDI-MS platform, providing a potential method for noninvasive screening of thyroid malignant tumors. Integrating this approach with imaging technologies such as ultrasound can enhance the reliability of noninvasive diagnostic methods for malignant tumor screening, helping to prevent unnecessary invasive procedures and reducing the risk of overdiagnosis and overtreatment in patients with benign nodules.

Ultrasensitive PD-L1-Expressing Exosome Immunosensors Based on a Chemiluminescent Nickel-Cobalt Hydroxide Nanoflower for Diagnosis and Classification of Lung Adenocarcinoma

Programmed death ligand-1 (PD-L1)-expressing exosomes are considered a potential marker for diagnosis and classification of lung adenocarcinoma (LUAD). There is an urgent need to develop highly sensitive and accurate chemiluminescence (CL) immunosensors for the detection of PD-L1-expressing exosomes. Herein, N-(4-aminobutyl)-N-ethylisopropanol-functionalized nickel-cobalt hydroxide (NiCo-DH-AA) with a hollow nanoflower structure as a highly efficient CL nanoprobe was synthesized using gold nanoparticles as a "bridge". The resulting NiCo-DH-AA exhibited a strong and stable CL emission, which was ascribed to the exceptional catalytic capability and large specific surface area of NiCo-DH, along with the capacity of AuNPs to facilitate free radical generation. On this basis, an ultrasensitive sandwich CL immunosensor for the detection of PD-L1-expressing exosomes was constructed by using PD-L1 antibody-modified NiCo-DH-AA as an effective signal probe and rabbit anti-CD63 protein polyclonal antibody-modified carboxylated magnetic bead as a capture platform. The immunosensor demonstrated outstanding analytical performance with a wide detection range of 4.75 × 103-4.75 × 108 particles/mL and a low detection limit of 7.76 × 102 particles/mL, which was over 2 orders of magnitude lower than the reported CL method for detecting PD-L1-expressing exosomes. Importantly, it was able to differentiate well not only between healthy persons and LUAD patients (100% specificity and 87.5% sensitivity) but also between patients with minimally invasive adenocarcinoma and invasive adenocarcinoma (92.3% specificity and 52.6% sensitivity). Therefore, this study not only presents an ultrasensitive and accurate diagnostic method for LUAD but also offers a novel, simple, and noninvasive approach for the classification of LUAD.

Identification and validation of serum metabolite biomarkers for endometrial cancer diagnosis

Endometrial cancer (EC) stands as the most prevalent gynecological tumor in women worldwide. Notably, differentiation diagnosis of abnormity detected by ultrasound findings (e.g., thickened endometrium or mass in the uterine cavity) is essential and remains challenging in clinical practice. Herein, we identified a metabolic biomarker panel for differentiation diagnosis of EC using machine learning of high-performance serum metabolic fingerprints (SMFs) and validated the biological function. We first recorded the high-performance SMFs of 191 EC and 204 Non-EC subjects via particle-enhanced laser desorption/ionization mass spectrometry (PELDI-MS). Then, we achieved an area-under-the-curve (AUC) of 0.957-0.968 for EC diagnosis through machine learning of high-performance SMFs, outperforming the clinical biomarker of cancer antigen 125 (CA-125, AUC of 0.610-0.684, p < 0.05). Finally, we identified a metabolic biomarker panel of glutamine, glucose, and cholesterol linoleate with an AUC of 0.901-0.902 and validated the biological function in vitro. Therefore, our work would facilitate the development of novel diagnostic biomarkers for EC in clinics.