Methods: CMV R+ adults with hematologic malignancies who received peripheral blood CD34+ selected allografts after myeloablative conditioning regimens from March 1, 2010 through May 2013 were monitored prospectively by CMV PCR at least weekly from day (d) +14 through +100 and at least once every 2 weeks through d +180. Preemptive therapy was initiated for CMV viremia, defined as ≥ 1 positive PCR value. There was no CMV prophylaxis during the study period. Follow up was through May 31, 2014. Time-dependent Cox-proportional hazard model of univariate and multivariate stepwise selection analyses were performed to identify predictors of CMV viremia.
Results: Of 113 CMV R+ (median age 57.6 years, 22.5-73.0), 60(53%) had acute leukemia, 23 (20%) myelodysplastic syndrome, 18 (16%) multiple myeloma and 12 (11%) other. Donors were 48 (42%) matched-related (MRD), 45 (40%) matched unrelated (MUD), or 20 (18%) mismatched; 68 (60%) CMV D+ or 45 (40%) D-. The cumulative incidence of CMV viremia at 6 months was 79%. Time to onset of viremia was median 26 days (13-83) after SCT, and time to resolution of viremia a median 30.5 days from onset of viremia. The incidence of CMV viremia was higher in R+/D+ compared to R+/D- (90.5% versus 62.2%; P<0.0001). Median time to onset and to resolution for R+/D+ vs R+/D- were not significantly different. In multivariable analysis, R+/D- pts were at lower risk for CMV viremia compared to R+/D+ (HR: 0.57, 95% CI: 0.36-0.90; P=0.017). In contrast, multiple myeloma (HR: 2.22, 95% CI: 1.24-3.98; P=0.008) was associated with increased risk compared with acute leukemia. Having a mismatched donor was also associated with increased risk (HR: 1.78, 95% CI: 0.98-3.23; P=0.058) compared with MRD and (HR: 2.30, 95% CI: 1.26-4.21; P=0.007) with MUD.
Conclusions: 1) Among CMV R+ TCD recipients, the incidence of CMV viremia was 79%. 2) In multivariate analysis, CMV D- with decreased risk for CMV viremia compared to D+. 3) Diagnosis of multiple myeloma and allografts from mismatched donor was associated with increased risk. 4) Further studies are needed to assess the effect of CMV donor serostatus in transplant outcomes in TCD. If validated in larger cohorts, our findings have implications for donor selection and targeted strategies for CMV prevention.