Preemptive valganciclovir for cytomegalovirus infection in hematological patients

Abstract

To the Editor We read with interest the ¢ndings by Busca et al. (1) published in the Journal, adding to similar recent ¢ndings supporting the e⁄cacy of preemptive valganciclovir in allogeneic hematopoietic stem cell transplant (HSCT) recipients (2, 3). We add here our single-center preliminary experience in hematological patients. Before November 2005 we usually treated cytomegalovirus (CMV) reactivation (higher than 1500 copies/mL) with intravenous ganciclovir. During a 24 -month period starting in November 2005, all our patients undergoing HSCTor aggressive chemotherapy were monitored with weekly real-time quantitative polymerase chain reaction (PCR) assay (Cobas Amplicor CMV Monitor test, Roche Diagnostics, Mannheim, Germany; sensitivity5 600 CMV DNAcopies/mL) on peripheral blood during treatment and in the ¢rst 3 months after discharge. Thirty-seven patients with CMV viremia higher than 1500 copies/mL (at a median of 45 from end of chemotherapy or HSCT) were identi¢ed and treated for 3 weeks with preemptive oral valganciclovir 450 mg b.i.d. (n5 5) or 900 mg b.i.d. (n5 32) according to renal function. No patient had concurrent infections or was receiving steroids at the time of reactivation. No reduction in immunosuppression was performed in allogeneic HSCT recipients at the time of CMVreactivation. All but 2 patientswho reactivated CMVafter aggressive chemotherapy (n513: 7 myeloablative and 6 nonmyelobalative), or autologous (n5 5: 4 high-dose melphalan and 1 BEAM) or allogeneic (n519) HSCTshowed weekly drops in CMV levels and ¢nally cleared viremia at the ¢nal control on day 21 day since beginning of treatment. Both patients who failed the endpoint were allogeneic HSCT recipients with initial loads lower than 5000 copies/mL who had been treated with valganciclovir 900mgb.i.d.; they had persistent low-level viremia (4000 copies/mL) for at least 2 months despite salvage treatment with intravenous foscarnet. Only 3 patients, all allogeneic HSCTrecipients, developed neutropenia (absolute neutrophil counto500/mL) while on valganciclovir, which could be explained by concurrent immunosuppressive drugs. The median initial viral load in our cohort was 3900 copies/mL, which is a relatively low level and could have contributed to the high e⁄cacy of valganciclovir.W|th the use of such a low threshold, we did not observe any case of symptomatic CMVdisease. Recent data support that the e⁄cacy of valganciclovir 450mg b.i.d. is identical to 900mg b.i.d. in allogeneic HSCT recipients (90% clearance of antigenemia on day 121) (4), and this is likely truewhenever initial CMV load remains quite low, allowing proper treatment of patients with renal failure. In reduced-intensity conditioning HSCT recipients without gastrointestinal graft-versus-host disease (GvHD), the r 2008 Wiley Periodicals, Inc.