Complement and Graft-versus-host Disease

The study will be performed in allografted patients with myeloablative conditioning for an haematological malignancy from 3 adult transplant units. Patients will be followed for at least12 months after transplantation and blood samples drawn before conditioning and once a week for 12 weeks after transplantation to analyze: * serum concentration of Complement factors (C3, C4, B factor), Complement regulatory proteins (C1-inhibitor, I and H Factors) and analysis of the surface expression of Complement regulatory molecules such as CD46, CD55 and CD59. * serum inflammatory cytokine levels In addition, patients with clinical signs of gut GVHD will be explored by gastrointestinal endoscopy to perform gut biopsies. C5b9 deposure will be then analyzed by immunohistochemistry on GVHD lesions. Activation of complement system will be defined by a decrease of complement factor levels of 50% and values under lower physiological limits. The clinical evolution and the inflammatory cytokine profile of patients with such an activation profile will be compared to that of those without complement activation. A data base containing biological and clinical data will be established. Biological results will be correlated to post-transplant clinical events, in particular the occurrence of gut GVHD but also non relapse mortality and overall survival by adapted statistical tests (comparison of percentages by Chi-2 of Pearson, comparison of survival curves by logrank, multivariate analysis by logistic regression test or cox model). The number of required patients will be established by comparison of the percentage of gut GVHD in the patients with or without complement activation. Based on our preliminary results, we hypothesize that 2/3 patients will not have complement activation among whose 20% will develop acute gut GVHD. We expect an increase of acute gut GVHD up to 60% of the patients with complement activation that would represent 1/3 of the cohort. With a bilateral alpha risk of 5% and a power of 80%, the number of required patients is 23 in the activated group and 46 in the non activated group, thus a total of 69 patients.