Treatment with Granulocyte Colony-Stimulating Factor after Allogeneic Transplantation Increases the Risk of Hepatic Veno-Occlusive Disease and Death: A Retrospective Analysis of 440 Cases at a Single Center.

Hepatic veno-occlusive disease (VOD) is one of the most serious complications of hematopoietic stem cell transplantation (HSCT). Various factors have been identified as increasing the risk of hepatic VOD; history of liver dysfunction, elevated AST level, transplant from an HLA-mismatched or unrelated donor, and abdominal irradiation, but few of them have been associated with a significantly increased risk. A recent study showed a newly association between the use of granulocyte colony-stimulating factor (G-CSF) and subsequent higher incidence of VOD. The present analysis was performed to elucidate the clinical effects of G-CSF on the incidence of hepatic VOD and survival after HSCT in a large group of patients at a single institution. We retrospectively reviewed 440 patients (264 males, 176 females; median age 35, range 0–67) with hematological diseases who underwent allogeneic transplantation between September 1986 and December 2003. G-CSF was given basically according to the primary diseases and type of transplant, and a total of 230 patients with lymphoid malignancy received G-CSF shortly after transplantation, while G-CSF was not administered in remaining 210 patients with myeloid malignancy. A diagnosis of VOD was made according to the McDonald’s criteria and 47 patients (11.0%) were eventually diagnosed with VOD. Hepatic VOD occurred in 33 of 230 patients (14%) in the patients with G-CSF versus 14 of 210 patients (6%) in the non G-CSF group. Several possible risk factors based on the previous studies, including the use of G-CSF, were counted on an initial univariate analysis and cumulated significant factors further analyzed for their potential value for VOD development in multivariate analysis. By univariate analysis, the HLA disparity, performance status before transplantation, febrile episode during conditioning therapy and the use of G-CSF were significantly related to the incidence of VOD. A multivariate analysis with logistic regression again identified the use of G-CSF as the factor that significantly influenced the occurrence of VOD, in addition to the former three independent variables. Our analysis of risk factors in this series confirmed some other published studies but differed for other factors. Surprisingly,the administration of G-CSF may have an adverse impact on the subsequent survival after transplantation. Patient survival probability was statistically different (p=0.0063) between two groups, as analyzed by the Kaplan-Meier plot: the ten year survival rate was 50±4% in the G-CSF group and 64±4% in the non G-CSF group (p=0.0063). There were a total of 98 deaths in the G-CSF group compared with 70 deaths in the non G-CSF group. Although relapse and GVHD were the most common cause of death in both groups, death due to hepatic failure/VOD/TMA was observed 10% in the G-CSF group versus 4% in the non G-CSF group. Our data based on retrospective analysis should be interpreted with some caution, but they have suggested that G-CSF might be involved in VOD development and also adversely affected the outcome. Although underlying mechanism how G-CSF adversely effect on survival should be multifactorial, it is possible that G-CSF could activate the coagulation system and therefore induce thrombotic events including VOD. Thus, this putative mechanism of G-CSF towards a prothrombotic tendency warrants a further clarification, and the use of G-CSF should be carefully assessed in transplant recipients.