Optimizing letermovir timing: Day +5 to +8 post haplo-HSCT minimizes EBV dnaemia without compromising CMV prophylaxis

Abstract

Abstract Background The era of Letermovir (LTV) has reshaped CMV prophylaxis and treatment following allogeneic HSCT, with its efficacy in reducing CMV infection well established. However, emerging evidence suggests an association between LTV exposure and increased risk of EBV DNAemia. In clinical practice, the timing of LTV initiation varies widely, and optimal scheduling remains undefined. This study aimed to evaluate the impact of different LTV initiation times on the cumulative incidence of CMV and EBV DNAemia, as well as overall survival (OS), to provide evidence for the optimization of LTV prophylaxis timing in clinical practice. Methods This study included 256 patients (pts) aged ≥14 years with hematologic disorders who underwent their first ATG/G-CSF-based haplo-HSCT at Hebei Yanda Lu Daopei Hospital or Beijing Lu Daopei Hospital between May 2022 and January 2024. All pts received LTV prophylaxis for CMV infection within 100 days post-transplant. Pts who received LTV for the treatment of CMV DNAemia or initiated LTV outside the +1 to +30 day window post-HSCT were excluded. Based on the timing of LTV initiation, pts were stratified into 4 cohorts: cohort 1 (78 pts，days +1 to +4), cohort 2 (92 pts，days +5 to +8), cohort 3 (49 pts，days +9 to +14), and cohort 4 (37 pts，days +15 to +30). All patients were followed through January 1, 2025. Results A total of 256 pts were included, comprising 12 with aplastic anemia (AA), 183 with acute myeloid leukemia (AML), and 61 with acute lymphoblastic leukemia (ALL). The cohort consisted of 150 males and 106 females, with a median age of 35 years (range, 14–71). Among pts with malignant hematologic diseases, 207 of 244 (84.8%) were in complete remission (CR) at the time of transplantation, while 37 of 244 (15.2%) were in non-remission or partial remission (NR/PR). Notably, cohort 3 had a higher proportion of patients with NR/PR status (26.5%) and measurable residual disease (MRD) positivity (50%) prior to transplantation, whereas cohort 2 had the highest use of FK506-based graft-versus-host disease (GVHD) prophylaxis (11.96%) among all groups. No statistically significant differences were observed in baseline characteristics—including sex, age, conditioning regimen, neutrophil engraftment time, and others—among the 4 cohorts. The median follow-up duration was 16.0 months (range, 1.5–27.3). Across the entire cohort, the cumulative incidence of CMV DNAemia by day +180 was 22.05% (95% CI, 17.50–27.79%), and that of EBV DNAemia was 12.90% (95% CI, 9.38–17.74%). The estimated 2-year OS for the whole population was 80.08% (95% CI, 74.94–85.23%). Stratified by the timing of LTV initiation, cohort 2 demonstrated the most favorable outcomes. By day +180 post-transplant, the cumulative incidence of CMV DNAemia was 17.39% (95% CI, 11.14–27.15%) in cohort 2, compared with 20.70% (95% CI, 13.38–32.03%) in cohort 1, 35.95% (95% CI, 24.56–52.63%) in cohort 3, and 18.92% (95% CI, 9.71–36.86%) in cohort 4; no statistically significant differences were observed among the groups (P > 0.05). However, cohort 2 had a significantly lower incidence of EBV DNAemia at 6.52% (95% CI, 3.01–14.14%) compared with cohort 1 (16.67%, 95% CI, 10.15–27.38%), cohort 3 (14.29%, 95% CI, 7.20–28.36%), and cohort 4 (18.92%, 95% CI, 9.71–36.86%), with all comparisons reaching statistical significance (P < 0.05). The estimated 2-year OS was also highest in cohort 2 at 87.99% (95% CI, 81.33–94.65%), followed by cohort 4 at 83.78% (95% CI, 71.91–95.66%), cohort 1 at 76.26% (95% CI, 66.11–86.41%), and cohort 3 at 67.99% (95% CI, 53.62–82.35%). The inferior OS in cohort 3 may be partly attributed to a higher proportion of patients with non-remission or partial remission status at the time of transplant (26.5%, 13/49). Conclusion Initiating LTV prophylaxis between days +5 to +8 post haplo-HSCT (cohort 2) was associated with a comparable cumulative incidence of CMV DNAemia, the lowest incidence of EBV DNAemia, and the highest 2-year OS. These findings suggest that day +5 to +8 may represent the optimal window for initiating LTV prophylaxis to balance effective CMV prevention while minimizing the risk of EBV reactivation. Further prospective validation is needed.