Pilot Multi-Omic Analysis of Human Bile from Benign and Malignant Biliary Strictures: A Machine-Learning Approach

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.
This research was funded by: Instituto de Salud Carlos III (ISCIII) co-financed by Fondo Europeo de Desarrollo Regional (FEDER) Una manera de hacer Europa, grant numbers: PI16/01126 (M.A.A.), PI19/00819 (M.J.M. and J.J.G.M.), PI15/01132, PI18/01075 and Miguel Servet Program CON14/00129 (J.M.B.); Fundación Científica de la Asociación Española Contra el Cáncer (AECC Scientific Foundation), grant name: Rare Cancers 2017 (J.M.U., M.L.M., J.M.B., M.J.M., R.I.R.M., M.G.F.-B., C.B., M.A.A.); Gobierno de Navarra Salud, grant number 58/17 (J.M.U., M.A.A.); La Caixa Foundation, grant name: HEPACARE (C.B., M.A.A.); AMMF The Cholangiocarcinoma Charity, UK, grant number: 2018/117 (F.J.C. and M.A.A.); PSC Partners US, PSC Supports UK, grant number 06119JB (J.M.B.); Horizon 2020 (H2020) ESCALON project, grant number H2020-SC1-BHC-2018–2020 (J.M.B.); BIOEF (Basque Foundation for Innovation and Health Research: EiTB Maratoia, grant numbers BIO15/CA/016/BD (J.M.B.) and BIO15/CA/011 (M.A.A.). Department of Health of the Basque Country, grant number 2017111010 (J.M.B.). La Caixa Foundation, grant number: LCF/PR/HP17/52190004 (M.L.M.), Mineco-Feder, grant number SAF2017-87301-R (M.L.M.), Fundación BBVA grant name: Ayudas a Equipos de Investigación Científica Umbrella 2018 (M.L.M.). MCIU, grant number: Severo Ochoa Excellence Accreditation SEV-2016-0644 (M.L.M.). Part of the equipment used in this work was co-funded by the Generalitat Valenciana and European Regional Development Fund (FEDER) funds (PO FEDER of Comunitat Valenciana 2014–2020). Gobierno de Navarra fellowship to L.C. (Leticia Colyn); AECC post-doctoral fellowship to M.A.; Ramón y Cajal Program contracts RYC-2014-15242 and RYC2018-024475-1 to F.J.C. and M.G.F.-B., respectively. The generous support from: Fundación Eugenio Rodríguez Pascual, Fundación Echébano, Fundación Mario Losantos, Fundación M Torres and Mr. Eduardo Avila are acknowledged. The CNB-CSIC Proteomics Unit belongs to ProteoRed, PRB3-ISCIII, supported by grant PT17/0019/0001 (F.J.C.). Comunidad de Madrid Grant B2017/BMD-3817 (F.J.C.).