Epstein-Barr Virus Reactivation and Post-Transplant Lymphoproliferative Disorders after Allo-HSCT in the Era of Letermovir for Cytomegalovirus Prophylaxis

Introduction: Letermovir is an antiviral agent that significantly decreases the reactivation of cytomegalovirus (CMV) after allogeneic hematopoietic stem cell transplantation (allo-HSCT), but its impact on other virus infections following allo-HSCT, especially Epstein-Barr virus (EBV), remains uncertain.

Methods: In this study, we conducted a retrospective cohort study of 593 patients who underwent allo-HSCT, including 298 receiving letermovir (letermovir group) and 295 without letermovir (control group) as CMV primary prophylaxis.We compared the cumulative incidences (CI) of CMV reactivation, clinically significant EBV reactivation (cs-EBVr), and post-transplant lymphoproliferative disorders (PTLD) between the two groups. We also identified the risk factors for EBV reactivation and PTLD using the Cox proportional hazards model in patients with letermovir prophylaxis. Furthermore, we quantified the anti-virus capacity using FlowSpot assay, and evaluated the immune reconstitution (IR) in the first three months after transplant with multi-parameter flow cytometry (MFC).

Results: For the entire cohort, the median follow-up time was 354 (range: 80–1112) days, and the median duration of letermovir administration for the letermovir group was 100 (range: 30-327) days. Our data confirmed that letermovir had significant efficacy in preventing CMV reactivation with a 1-year CI of 31.21% in the letermovir group compared to 76.95% in the control group (P < 0.0001). The baseline characteristics of the two groups were comparable. Compared to the control group, there was no significant difference in the 1-year CI of cs-EBVr in the letermovir group (57.1% vs. 62.7%, P=0.12). However, cs-EBVr patients with letermovir prophylaxis exhibited lower EBV DNA titers in plasma than those in the control group (median[range]: 2345 [557-695000] vs. 1645 [579-91400] IU/mL, P = 0.011). Besides, the 1-year CI of PTLD was higher in the letermovir group compared to the control group (7.05% vs. 1.70%, P=0.0015). The median time from HSCT to PTLD diagnosis of the two groups was similar (median[range]: 62[50-300] vs. 69.5[21-452] days, P=0.414). Antecedent EBV viremia was observed in 96.30% of PTLD patients (26/27). Multivariate analysis identified letermovir as an independent risk factor for PTLD (hazard ratio [95% confident interval]: 4.71 [1.78-12.50], P < 0.001). To investigate the immune characteristics related to PTLD in the letermovir group, we quantified the antiviral capacity of peripheral blood mononuclear cells (PBMCs) within the first 5 months post-transplant by assessing their ability to release IFN-γ using flowspot assay for 115 patients. Besides, monitoring of lymphocyte subsets by MFC revealed that letermovir altered the early reconstitution trajectory compared to the control group. Specifically, we found that letermovir delayed virus-specific immune recovery, contributing to the occurrence of PTLD. Delayed IR within the first 3 months, particularly the interaction of CD8⁺ T and NK cells, was observed in PTLD patients receiving letermovir prophylaxis. The deficiency of CD8⁺ naïve T (Tn, CD8⁺CD45RA⁺CD27⁺) cells in PTLD patients in the letermovir group was observed in the first 3 months after allo-HSCT. There was a significant reduction in CD8⁺ terminally differentiated effector memory T cells (TemRA, CD8⁺CD45RA⁺CD27^-) in PTLD patients. In addition, the reconstitution of NK cells was delayed in the letermovir group. Correlation analysis also revealed a negative association between NK and CD8⁺ T subsets in the first month in PTLD patients. The interplay between NK and CD8⁺ T cells is pivotal for anti-virus process. NKG2A⁺ NK cells are deficient in anti-EBV capacity, and reversely regulate CD8⁺ TEMRA cells, and the latter is the major cell population in anti-virus pathology. In the early stage after allo-HSCT, NKG2A⁺ NK cells were highly proliferated in PTLD patients in the letermovir group, which may contribute to the unique PTLD pathophysiology in the letermovir era.

Conclusion: Letermovir prophylaxis did not increase the risk of cs-EBVr in allo-HSCT recipients; however, it was associated with a higher incidence of PTLD. Our findings emphasized the significance of CD8⁺ T and NK cell interaction in the early IR, crucial in shaping the outcomes of EBV reactivation. Further studies focusing on IR dynamics are warranted to elucidate the epidemiology and kinetics of EBV in the letermovir era.