Pilot Multi-Omic Analysis of Human Bile from Benign and Malignant Biliary Strictures: A Machine-Learning Approach
Jesús M Urman, José M Herranz, Iker Uriarte, María Rullán, Daniel Oyón, Belén González, Ignacio Fernandez-Urién, Juan Carrascosa, Federico Bolado, Lucía Zabalza, María Arechederra, Gloria Alvarez-Sola, Leticia Colyn, María U Latasa, Leonor Puchades-Carrasco, Antonio Pineda-Lucena, María J Iraburu, Marta Iruarrizaga-Lejarreta, Cristina Alonso, Bruno Sangro, Ana Purroy, Isabel Gil, Lorena Carmona, Francisco Javier Cubero, María L Martínez-Chantar, Jesús M Banales, Marta R Romero, Rocio I R Macias, Maria J Monte, Jose J G Marín, Juan J Vila, Fernando J Corrales, Carmen Berasain, Maite G Fernández-Barrena, Matías A Avila
Cholangiocarcinoma (CCA) and pancreatic adenocarcinoma (PDAC) may lead to the development of extrahepatic obstructive cholestasis. However, biliary stenoses can also be caused by benign conditions, and the identification of their etiology still remains a clinical challenge. We performed metabolomic and proteomic analyses of bile from patients with benign (n = 36) and malignant conditions, CCA (n = 36) or PDAC (n = 57), undergoing endoscopic retrograde cholangiopancreatography with the aim of characterizing bile composition in biliopancreatic disease and identifying biomarkers for the differential diagnosis of biliary strictures.
Comprehensive analyses of lipids, bile acids and small molecules were carried out using mass spectrometry (MS) and nuclear magnetic resonance spectroscopy (1H-NMR) in all patients. MS analysis of bile proteome was performed in five patients per group. We implemented artificial intelligence tools for the selection of biomarkers and algorithms with predictive capacity.
Our machine-learning pipeline included the generation of synthetic data with properties of real data, the selection of potential biomarkers (metabolites or proteins) and their analysis with neural networks (NN). Selected biomarkers were then validated with real data. We identified panels of lipids (n = 10) and proteins (n = 5) that when analyzed with NN algorithms discriminated between patients with and without cancer with an unprecedented accuracy.
CITA DEL ARTÍCULO Cancers (Basel). 2020 Jun 21;12(6):1644. doi: 10.3390/cancers12061644.