Natural History of Epstein-Barr Virus Replication and Viral Load Dynamics after Alemtuzumab-Based Allogeneic Stem Cell Transplantation.
Marzolini MAV., Wilson AJ., Sanchez E., Carpenter B., Chakraverty R., Hough R., Kottaridis P., Morris EC., Thomson KJ., Peggs KS.
Epstein-Barr virus (EBV) load monitoring after allogeneic hematopoietic stem cell transplantation (HSCT) enables earlier detection of EBV replication and often serves as a trigger for preemptive therapies aimed at reducing EBV-related diseases. Our institutional strategy is to treat patients with clinical signs of EBV-related disease accompanied by a rising viral load, rather than to intervene based solely on viral load. This affords an opportunity to study the natural history of EBV replication and to assess whether our strategy reduces overtreatment without compromising outcomes. The objectives of the present study were to assess the natural history of untreated EBV replication in patients who underwent an alemtuzumab-based allogeneic HSCT and to examine whether our clinical strategy reduced overtreatment without compromising patient outcomes. In this retrospective single-center observational study of 515 consecutive patients (age ≥18 years) undergoing T cell-depleted allogeneic HSCT incorporating alemtuzumab, patients underwent surveillance monitoring for EBV by quantitative PCR in the peripheral blood at least weekly up to 100 days post-transplantation and longer if they remained on immunosuppressive therapy. The cumulative incidence of EBV detection and EBV-related disease were assessed. Among the 515 patients, 192 had EBV DNA detectable on ≥1 occasion, with a cumulative incidence of 35.8% (31.8% to 40.4%), although this remained below the limit of quantification in 93 patients. The median time to first detection was 89.5 days (range, 0 to 2254 days). The incidence was higher in recipients of sibling donor transplants (45.4% versus 30%; P = .00021) compared with recipients of unrelated donor transplants. Twenty patients developed EBV-related disease (cumulative incidence, 3.9%). Two patients had immunosuppression reduction alone, 18 received rituximab, and 5 required additional therapies. Five patients died from post-transplantation lymphoproliferative disorder, all of whom had received rituximab. The positive predictive value of EBV load for disease was higher in the unrelated donor cohort but remained <75% regardless of EBV threshold (57.1% to 72.7%). The cumulative incidence of EBV-related disease in our study (3.9%) is comparable to that reported in other studies incorporating alemtuzumab, and our clinical strategy reduced overtreatment in this patient population. PCR-based surveillance strategies have limitations, as reflected in the relatively low sensitivity of the assay coupled with the low positive predictive value, which may influence the potential choice of a threshold for preemptive intervention. We conclude that it remains unclear whether treatment based on a rising EBV viral load alone provides superior overall results to treatment based on the development of clinical signs of EBV-related disease in the context of a rising viral load.