Introduction: Human cytomegalovirus infection is still a major complication after pediatric bone marrow transplantation and could be fatal in some cases. The toxicity of the drug in dividing transplanted haematopoietic cells combined with the suppression of cell growth caused by the virus remains a major problem in managing human cytomegalovirus infection. Methods: The aim of the current in vitro study was to evaluate the effect of the intensity (1-20 mg/l) and duration (1, 2, 7 or 14 days) of ganciclovir exposure on toxicity in B lymphoblastoid cells (using cell counting and viability measurements). Results: A correlation was found between the dose of ganciclovir exposure and a decrease in total cell number when duration exceeded 2 days (r 2 =0.92 and 0.93 after 7 and 14 days, respectively). High levels (20 mg/l) of ganciclovir were not more toxic than lowest levels (1 mg/l) for the shortest durations of ganciclovir exposure (1 and 2 days). Moreover, 50% cytotoxic concentrations markedly decreased with the duration of ganciclovir exposure (374-3 mg/l from 1 to 14 days respectively) after 14 days of culture. Conclusions: This in vitro study demonstrated for the first time that ganciclovir exhibited an in vitro durationdependent toxicity on haematopoietic-derived cells when in vivo doses of the drug were used.Human cytomegalovirus (HCMV) remains a major problem after bone marrow transplantation (BMT) and can be fatal in some cases [1][2][3][4][5][6]. Ganciclovir (GCV) is considered the standard treatment for HCMV after paediatric BMT [1,7]; however, because of a lack of paediatric recommendations, GCV therapy is still based on adult schedules and adapted to patient weight. The major problem with GCV is haematological toxicity [8,9] and myelosuppressive effects [9]. In fact, due to the resemblance of GCV to guanosine, this nucleoside analogue is likely to have adverse effects on mammalian DNA synthesis in addition to its effect on viral replication [10]. GCV is particularly toxic to rapidly dividing populations of blood cells and their precursors [10]. This side effect represents an in vivo risk factor for post-transplantation death [8], opportunistic infection [2] and tardive HCMV infection because of high immunosuppression [7]. The duration of treatment is a risk factor for this potentially fatal situation [2,7]. Despite the clinical use of GCV for 20 years, few studies have determined the toxicity of GCV on haematopoietic cells [9][10][11][12] and no studies have tested both dose and duration effects on cell toxicity using B lymphoblastoid cell lines (BLCLs), which exhibit a high similarity to the target cells for GCV toxicity in vivo. This led us to study the effect of GCV exposure on the growth of BLCLs. The present study is a preliminary report from a project aimed at evaluating the antiviral effect of shorter treatments at higher doses, which is supposed to be less toxic for bone marrow.