MALDI-TOF analysis of blood serum proteome can predict the presence of monoclonal gammopathy of undetermined significance

• Published in: PloS one Vol. 13; no. 8; p. e0201793
• Main Authors: Barceló, Francisca, Gomila, Rosa, de Paul, Ivan, Gili, Xavier, Segura, Jaume, Pérez-Montaña, Albert, Jimenez-Marco, Teresa, Sampol, Antonia, Portugal, José
• Format: Journal Article
• Language: English
• Published: United States Public Library of Science 02.08.2018; Public Library of Science (PLoS)
• Subjects: Adult; Aged; Aged, 80 and over; Benign monoclonal gammopathy; Bioinformatics; Biological markers; Biology and Life Sciences; Biomarkers; Biomarkers - blood; Blood serum; Bone marrow; Classification; Complications and side effects; Computer and Information Sciences; Diagnosis; Engineering and Technology; Female; Gene expression; Hematologic Tests - methods; Humans; Learning algorithms; Male; Mass spectrometry; Mass spectroscopy; Mathematical models; Medicine and Health Sciences; Middle Aged; Model accuracy; Monoclonal gammopathies; Monoclonal Gammopathy of Undetermined Significance - blood; Multiple myeloma; Patients; Performance prediction; Physical Sciences; Prediction models; Predictive control; Proteins; Proteome; Proteomes; Proteomics; Quality Control; Research and Analysis Methods; Risk factors; Scientific imaging; Sensitivity and Specificity; Serum - metabolism; Spectrometry, Mass, Matrix-Assisted Laser Desorption-Ionization - methods; Support Vector Machine; Support vector machines; Young Adult; Majorca; New York; United States > US; Spain
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0201793

Monoclonal gammopathy of undetermined significance (MGUS) is a plasma cell dyscrasia that can progress to malignant multiple myeloma (MM). Specific molecular biomarkers to classify the MGUS status and discriminate the initial asymptomatic phase of MM have not been identified. We examined the serum peptidome profile of MGUS patients and healthy volunteers using MALDI-TOF mass spectrometry and developed a predictive model for classifying serum samples. The predictive model was built using a support vector machine (SVM) supervised learning method tuned by applying a 20-fold cross-validation scheme. Predicting class labels in a blinded test set containing randomly selected MGUS and healthy control serum samples validated the model. The generalization performance of the predictive model was evaluated by a double cross-validation method that showed 88% average model accuracy, 89% average sensitivity and 86% average specificity. Our model, which classifies unknown serum samples as belonging to either MGUS patients or healthy individuals, can be applied to clinical diagnosis.