Successful modification of a pp65 antigenemia-based early treatment strategy for prevention of cytomegalovirus disease in allogeneic marrow transplant recipients.

We have previously reported that an intermittent pp65 antigenemiaguided early treatment strategy based on quantitative antigenemia was less effective in preventing cytomegalovirus (CMV) disease before day 100 than ganciclovir administered from engraftment until day 100 (ie, 14.1% v 2.7%, P 5 .002).1 We hypothesized that the high incidence of CMV disease before day 100 in the antigenemia-ganciclovir group was caused by the delay of treatment until high-grade antigenemia ($3 positive cells in 2 slides) and by the discontinuation of ganciclovir based on a negative antigen test.1 This hypothesis was supported by the finding that patients with severe graft-versus-host disease (GVHD) had rapid progression from low-grade antigenemia to disease and the observation that 5.3% of patients developed CMV pneumonia shortly after discontinuation of ganciclovir.1 Therefore, we advocated treating antigenemia at any level in patients with severe GVHD and continuing ganciclovir until day 100 after transplant.1 The projected incidence of CMV disease before day 100 with this modified strategy was calculated at 5.2%. We have recently implemented such a strategy among 102 consecutive allogeneic CMV seropositive allogeneic marrow transplant recipients. Patients were tested weekly for antigenemia by indirect immunofluorescence methods (CMV Brite; Biotest Diagnostic Corp, Denville, NJ) and ganciclovir was administered for antigenemia at any level and continued until day 100 after transplant (5 mg/kg every 12 hours for 7 days followed by 5 mg/kg/d 6 days per week). Of note, the risk profile of this patient cohort was comparable to that in the randomized trial (Table 1); in particular, the proportion of patients receiving marrow from an unrelated or HLA mismatched donors (Table 1) as well as GVHD prophylaxis regimens (data not shown) were similar. Among engrafted patients, 76% developed CMV antigenemia and were started on ganciclovir. The incidence of CMV disease before day 100 was 3.9% (pneumonia, N 5 1; gastrointestinal disease and pneumonia, N 5 1; gastrointestinal disease, N 5 2), which is similar to that observed with ganciclovir prophylaxis and significantly lower than that observed with the original antigenemia strategy (Table 1). Two patients had antigenemia detected coincident with or after diagnosis of disease, while the other two patients progressed to disease despite preceding antigenemia and initiation of preemptive ganciclovir. These results also appear to be similar to those recently reported with a polymerase chain reaction (PCR)-guided early treatment strategy in which ganciclovir was initiated when CMV-DNA was detected in two consecutive positive blood specimens.2 In that study, the incidence of CMV disease before day 100 was in 7.7% of engrafted patients who were seropositive or had a seropositive donor.2 The incidence of late CMV disease with the present strategy was higher than with the intermittent antigenemia-guided strategy, but somewhat lower than with ganciclovir prophylaxis (Table 1). These results are consistent with the hypothesis that higher use of ganciclovir before day 100 is associated with a delay in CMV-specific immune reconstitution, which may lead to a higher risk of late CMV disease.1,3 In our previous trial, there were significantly more invasive fungal infections in recipients of ganciclovir prophylaxis.1 The present strategy of antigenemia-guided early treatment was also associated with a low incidence of invasive fungal disease (Table 1), although more patients received ganciclovir for a longer period of time, which resulted in an 80% increase in use of ganciclovir compared to our previous antigenemia-guided strategy, which was associated with median use of ganciclovir of 22 days (compared to a median of 56 days with ganciclovir prophylaxis).1 Similar to the original trial,1 the differences in fungal infections could not be explained by differences in the incidence or severity of neutropenia. The results of this follow-up cohort as well as a subgroup analysis of the original randomized trial suggest that antigenemia-guided early therapy can be successfully implemented in unrelated marrow transplant recipients. For both the primary (CMV disease and neutropenia) and the secondary endpoints (bacterial and fungal infections and survival) there was no evidence that unrelated transplant recipients were at higher risk than the matched-related patients. Also, there was no statistically significant difference in survival at any time after transplant between the antigenemia-guided therapy groups and the ganciclovir prophylaxis group when unrelated transplant recipients were analyzed separately (data not shown). We acknowledge the fact that neither the randomized trial nor the analysis of the follow-up cohort were powered to detect specific differences for any of the outcomes in these subgroups. It appears that the strategy of antigenemia-guided early treatment in which low levels of CMV pp65 antigen in dextran-separated peripheral blood leukocytes are used as the guide to initiation of ganciclovir therapy is highly effective in preventing CMV disease without an Table 1. Comparison CMV Prevention Strategies