AUTHOR=Alborghetti Marcos Rodrigo , Correa Maria Elvira Pizzigatti , Whangbo Jennifer , Shi Xu , Aricetti Juliana Aparecida , Silva Andreia Aparecida da , Miranda Eliana Cristina Martins , Sforca Mauricio Luis , Caldana Camila , Gerszten Robert E. , Ritz Jerome , Zeri Ana Carolina de Mattos 

TITLE=Clinical Metabolomics Identifies Blood Serum Branched Chain Amino Acids as Potential Predictive Biomarkers for Chronic Graft vs. Host Disease

JOURNAL=Frontiers in Oncology

VOLUME=Volume 9 - 2019

YEAR=2019

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2019.00141

DOI=10.3389/fonc.2019.00141

ISSN=2234-943X

ABSTRACT=Allogeneic hematopoietic stem cell transplantation procedure, the only curative therapy for many types of hematological cancers, is increasing and graft-versus-host disease (GVHD) is the main cause of morbidity and mortality after transplantation. Currently, GVHD diagnosis is clinically performed. Whereas biomarkers panels have been developed for acute GVHD (aGVHD), there is lack of information about the chronic form (cGVHD). Using nuclear magnetic resonance (NMR) and gas chromatography coupled to time-of-flight (GC-TOF) mass spectrometry, this study prospectively evaluated the serum metabolome of 18 Brazilian patients who had undergone allogeneic hematopoietic stem cell transplantation (HSCT). We identified and quantified 63 metabolites and performed the metabolomic profile on days -10, day 0, day +10 and day +100, in reference to day of transplantation. Patients did not present aGVHD or cGVHD clinical symptoms at sampling times. From 18 patients analyzed, 6 developed cGVHD. Branched chain amino acids (BCAA) leucine and isoleucine were reduced and the sulfur-containing metabolite (cystine) was increased, at day +10 and day +100. Area under receiver operating characteristics (ROC) curves were higher than 0.79. These findings were validated, by liquid chromatography coupled to tandem mass spectrometry (LC-MS/MS), in a 49 North American patients at day +100 for BCAA, but not cystine. Our results highlight the importance of multi-temporal and multivariate biomarkers panels for predicting and understanding cGVHD.