Impact of Chronic Graft Versus Host Disease in the Overall Survival and Disease Free Survival of Patients with Hematologic Disorders Submitted to Allogeneic Bone Marrow Transplantation

Abstract

The management of the chronic graft versus host disease (C-GVHD) in the allogeneic hematopoietic stem cell transplantation (allo-HSCT) is still challenging and does not follow a consensus. This disease can manifest as auto-immune phenomena, often affecting multiple organs, and it is considered as a determinant factor for higher post transplant mortality and detrimental quality of life, although conversely associated with a strong graft versus tumor effect which determines a beneficial biological response in patients with haematological pathologies. The aim of this study was to determine the impact of the C-GVHD in the overall survival (OS) and disease free survival (DFS) in patients with hematologic pathologies who underwent allo-HSCT with HLA identical donors in the BMT Hospitals Sao Paulo and Santa Marcelina, in Sao Paulo, Brazil. We performed a retrospective study of historical cohort including 233 allo-HSC transplants, 151 patients aged 9 to 63 years old (median 31) with survival superior to 100 days, from August 1993 to December 2004. The classification of DECH-C followed the criteria of Seattle (Shulman et al. 1980) revised by Lee et al (2003). Our series was composed of 97 patients with chronic myeloid leukemia (CML), 54 with acute leukemias (AL), 42 with marrow failure (MF) and 40 with lymphoproliferative diseases (LPD). Forty one percent of the patients exhibited the advanced disease at HSCT time and the main source of HSC was bone marrow (n=174, 74.7%). The diagnosis of C-GVHD was performed in 166 procedures (71.2%). Seventy one patients (42.8%) were classified as C-GVHD quiescent, 69 (41.6%) as de novo, and 25 (15.1%) as progressive. The extensive form occurred in 95 cases (57.2%) and the limited form in 70 cases (42.2%). The mean OS for our cohort was 34.5 months (m) and the mean DFS was 29.5 m. The mean OS of patients with C-GVHD was 42 m, significantly higher than than 33.8m OS of patients without C-GVHD (p=0.05). Similarly, the DFS was 38.6m for C-GVHD patients, against 22.6m of DFS for patients without C-GVHD (p=0.0001). The OS of those patients with C-GVHD de novo (47m) and quiescent (44.8m) were superior than the OS of patients without C-GVHD (33.8%) or with the progressive form of C-GVHD (20.8m), p=0.002. The same effect was observed for the DFS in the de novo form (42.6m) and quiescent form (41.4m) of C-GVHD, with a DFS twice superior than the progressive form (19.8m) and than the C-GVHD absent (22.6m), p=0.001. When compared with cases without C-GVHD, the clinical severity of the C-GVHD (limited or extensive) showed a strong correlation with a higher DFS. Patients with limited C-GVHD had a DFS of 38.3m and the extensive form a DFS of 38.9m, significantly superior than the DFS of patients without C-GVHD with a DFS of 22.6m (p=0.002). Hence, this study shows the existence of a correlation between the presence of C-GVHD and higher OS and DFS in patients with haematological disorders submitted to allo-HCT. The diagnosis of progressive C-GVHD was associated to a deleterious effect over the outcome of allo-HCT, therefore inactivating the benefits of the graft versus tumour effect observed in the clinical presentations de novo and quiescent of C-GVHD.