Abstract :
[en] Introduction of hematopoietic stem cell transplantation (HSCT) as an immunotherapy provided a new option to improve survival of patients affected by haematological malignancies and congenital immunodeficiency syndrome. However, acute graft-versus-host disease (aGVHD) remains a major life-threatening complication after HSCT, limiting its application. As no plasma biomarkers available in clinical laboratory for aGVHD diagnosis, observation of clinical manifestations and histological examination of organ biopsies constitute the current diagnostic approach. Thus, to optimize management of aGVHD and reduced therapy-related toxicity, an early specific, rapid and non invasive diagnosis is needed. As proteomic approaches are useful tools for the rapid screening of protein content in complex samples, it makes them attractive for the discovery of new disease biomarkers. In the present work, we used up-to-date proteomic approaches with the aim to find new plasma biomarkers for the diagnosis and the early detection of acute GVHD. To deal with the high dynamic range of plasma protein concentrations, different sample preparation methods were firstly investigated. A method based on combinatorial peptide ligand affinity beads that allows detection of more information with good reproducibility, was selected. In addition, to extract a maximum of information from patient plasma samples we used three complementary proteomic approaches, namely 2D-DIGE, SELDI-TOF-MS and 2D-LC-MSE. The differentially expressed proteins between patients with and without aGVHD indicate a significant increase of the inflammation response and disturbance in the coagulation cascade. Interestingly, the early variation of these proteins 15 days before aGVHD diagnosis suggests the detection of the disease before symptoms appearance. In addition, we found that measurement of IL-10 levels, in the first month after HSCT allows the prediction of subsequent aGVHD onset while it is not the case for IL-7 and IL-15 levels. Finally, multivariate analysis provided a combination of biomarkers comprising fibrinogen, fragment of fibrinogen beta chain, SAA, prothrombin fragments, apolipoprotein A1 and hepcidin that can optimally distinguish controls and aGVHD samples (AUC 94.7). In conclusion, a combination of proteomic approaches allows us to discover a new panel of biomarkers that may help for the diagnosis of aGVHD.
