Counting EBV and T cells to predict PTLD

Epstein-Barr virus (EBV)–induced lymphoproliferative disease (PTLD) is a common and often fatal complication of transplantation and immunocompromise. Treatments for PTLD are more effective if initiated early, and high levels of EBV DNA in the peripheral blood of these individuals are a sensitive but rather nonspecific predictor of the disorder (Rowe et al, Transpl Infect Dis. 2001;3:79-87). In this issue, Meij and colleagues (page 4290) describe 25 recipients of partially T-cell–depleted stem cell transplants (SCTs) who developed a high virus load. None of 10 patients who received rituximab developed PTLD, in contrast with 9 of the remaining 15 patients who had equivalent EBV DNA levels but who were not so treated. Of these 9, only 6 responded to rituximab alone, and 2 died.

But since all current treatments for PTLD impart some risk, it would be valuable to have a predictive test that was more specific than simply measuring total EBV DNA in peripheral blood. Meij et al found that the predictive value of a high plasma virus load can be increased from 40% to 100% by counting EBV-specific CD8⁺ T cells. They used MHC class I tetramers that detected T cells specific for EBV in patients with 5 common HLA alleles, allowing them to track immunity in 61 of 100 stem cell transplant recipients. All patients who developed high levels of EBV DNA prior to discernible recovery of EBV-specific T cells subsequently developed PTLD. This result further supports the belief that PTLD could be avoided if the recovery of EBV-specific T cells is accelerated. Cytokine polymorphisms and NK-cell KIR receptor compatibility are proving influential in hematopoietic recovery after transplantation (Davies et al, Blood. 2002;100:3825-3827; and Cavet et al, Blood. 1999;94:3941-3946). If these factors also influence susceptibility to viral infection, manipulation of the posttransplantation cytokine environment or a change in criteria for donor selection may ultimately reduce the risk of this complication.