Predictors for persistent cytomegalovirus reactivation after T-cell-depleted allogeneic hematopoietic stem cell transplantation.

UNLABELLED

Cytomegalovirus (CMV) reactivation occurs in up to 60% of CMV-seropositive recipients after allogeneic hematopoietic stem cell transplantation (HSCT). The incidence of CMV disease among T-cell-depleted HSCT patients has been reported from 5-15%. The incidence of reactivation refractory to antivirals in this population is not well studied.

METHODS

In this retrospective study we characterized the outcome of CMV reactivation in a cohort of 255 adult and pediatric patients who underwent T-cell-depleted HSCT at Memorial Sloan-Kettering Cancer Center from September 1999 through August 2004. CMV infection was monitored by the pp65 antigenemia assay (CMV Ag). Persistent reactivation was defined as antigenemia positivity >21 days on antiviral therapy.

RESULTS

Of 118 CMV-seropositive recipients, 69 (58.4%) had reactivated CMV. Twenty of 69 (29%) developed persistent reactivation at first episode of reactivation, and 7 (10%) in subsequent episode. All patients with persistent reactivation received >/=2 antivirals and CMV hyperimmune globulin; 45% received combination antiviral therapy. The median duration of persistent reactivation was 98 days, range 31-256 days. In multivariate analysis, maximum CMV Ag >25 cells/slide was associated with persistent reactivation (odds ratio 16.2%, 95% confidence interval 4-64, P<0.0001). CMV disease occurred in 6/27 (22%) patients with persistent reactivation. Patients with persistent reactivation had lower CD4(+) and CD8(+) lymphocyte counts compared with those with non-persistent reactivation at day +90 post HSCT (P=0.01 and 0.02, respectively).

CONCLUSIONS

Persistent reactivation occurred in 39% of T-cell-depleted HSCT despite treatment with currently available antivirals. Maximum CMV Ag >25 cells/slide was associated with persistent CMV reactivation. More effective treatment modalities are needed for this high-risk population to reduce CMV-associated morbidity and mortality.