Exosomes in bodily fluids are a highly stable resource of disease biomarkers

Biomarkers are measurable indicators of a biological state. As our understanding of diseases meliorates, it is generally accepted that early diagnosis renders the best chance to cure a disease. In the context of proteomics, the discovery phase of identifying bonafide biomarkers and the ensuing validation phase involving large cohort of patient samples are impeded by the complexity of bodily fluid samples. High abundant proteins found in blood plasma make it difficult for the detection of low abundant proteins that may be potential biomarkers. Extracellular vesicles (EVs) have reignited interest in the field of biomarker discovery. EVs contain a tissue-type signature wherein a rich cargo of proteins and RNA are selectively packaged. In addition, as EVs are membranous structures, the luminal contents are protected from degradation by extracellular proteases and are highly stable in storage conditions. Interestingly, an appealing feature of EV-based biomarker analysis is the significant reduction in the sample complexity compared to whole bodily fluids. With these prescribed attributes, which are the rate-limiting factors of traditional biomarker analysis, there is immense potential for the use of EVs for biomarker detection in clinical settings. This review will discuss the current issues with biomarker analysis and the potential use of EVs as reservoirs of disease biomarkers.     

Application of quantitative proteomics technologies to the biomarker discovery pipeline for multiple sclerosis

Multiple sclerosis is an inflammatory-mediated demyelinating disorder most prevalent in young Caucasian adults. The various clinical manifestations of the disease present several challenges in the clinic in terms of diagnosis, monitoring disease progression and response to treatment. Advances in MS-based proteomic technologies have revolutionized the field of biomarker research and paved the way for the identification and validation of disease-specific markers. This review focuses on the novel candidates discovered by the application of quantitative proteomics to relevant disease-affected tissues in both the human context and within the animal model of the disease known as experimental autoimmune encephalomyelitis. The role of targeted MS approaches for biomarker validation studies, such as multiple reaction monitoring will also be discussed.     

Urinary Proteomics and Precision Medicine for Chronic Kidney Disease: Current Status and Future Perspectives

Precision medicine is since long an ongoing refinement of classical medicine, integrating improved and more detailed pathophysiological understanding with rapid technological advances. In the heterogenous area of chronic kidney disease there seems to be a high potential for the improvement in treatment and prognosis for several causes, with new technologies under development, that are yet to be introduced in clinical practice. As in other medical disciplines, investigation of abundant peptide patterns (proteomics) has gained recent interest. Especially relevant for kidney disease, urinary proteomics may provide both improved diagnosis and, as reviewed here, also holds promise for personalized treatment in the future. So far, capillary electrophoresis coupled to mass spectrometry (CE‐MS) is the most widely applied technique, and in addition to several cross‐sectional and cohort studies, there is even an ongoing randomized controlled trial that will soon report on the concept used as a method of personalizing treatment. In addition, there is hope that urinary proteomics can turn into a “liquid biopsy,” replacing the invasive diagnostic procedure. The next couple of years will provide more answers on the topic.     

Identification of Apolipoprotein AI as a serum biomarker of chronic kidney disease in liver transplant recipients, using proteomic techniques

Chronic kidney disease (CKD) is a common complication post‐orthotopic liver transplantation (OLT). Development of CKD is detected by monitoring serum urea and creatinine, however disease can occasionally be at an advanced stage before they become abnormal. Therefore, more accurate parameters are required. In order to identify novel biomarkers of CKD, serum was obtained from 47 OLT recipients with CKD (glomerular filtration rate <60 mL/min) and 23 with normal renal function (glomerular filtration rate >90 mL/min). Using the proteomic technique SELDI‐TOF‐MS, three protein biomarkers (55.6 kDa, 9.5 kDa and 11.4 kDa) were identified that, together, could stratify patients into cases or controls with a sensitivity and specificity of 93.6 and 91.3%, respectively. The area under the curve was 0.94. The primary splitter of the groups at 55.6 kDa was an alternative version of a molecule at 27.8 kDa, which was subsequently identified by 1‐D SDS‐PAGE and LC‐ESI‐MS/MS to be Apolipoprotein AI. Protein expression was shown to be reduced in CKD, by both ELISA (p = 0.057) and Western blot analysis (p = 0.003). Apolipoprotein AI is a novel, accurate marker of CKD post‐OLT. It does require further validation in a large, more diverse patient population but could potentially improve detection of CKD.     

An explainable deep learning model for prediction of early-stage chronic kidney disease

Chronic kidney disease (CKD) is a major public health concern with rising prevalence and huge costs associated with dialysis and transplantation. Early prediction of CKD can reduce the patient's risk of CKD progression to end-stage kidney failure. Artificial intelligence offers more intelligent and expert healthcare services in disease diagnosis. In this work, a deep learning model is built using deep neural networks (DNN) with an adaptive moment estimation optimization function to predict early-stage CKD. The health care applications require interpretability over the predictions of the black-box model to build conviction towards the model's prediction. Hence, the predictions of the DNN-CKD model are explained by the local interpretable model-agnostic explainer (LIME). The diagnostic patient data is trained on five layered DNN with three hidden layers. Over the unseen data, the DNN-CKD model yields an accuracy of 98.75% and a roc_auc score of 98.86% in detecting CKD risk. The explanation revealed by the LIME algorithm echoes the influence of each feature on the prediction made by the DNN-CKD model over the given CKD data. With its interpretability and accuracy, the proposed system may effectively help medical experts in the early diagnosis of CKD.