Quantifying Cytomegalovirus-Specific T Cell Responses Prior to Hospital Discharge Allows Risk Stratification for Late Viral Reactivation after Allogenic Hematopoietic Cell Transplantation

Abstract

Background Cytomegalovirus (CMV) infection remains a significant complication after allogenic hematopoietic cell transplantation (HCT). In the first three months after HCT, close monitoring of patients associated with prophylactic or preemptive administration of ganciclovir effectively prevents CMV-related complications. However, once patients are discharged from cancer centers around day 100 after HCT their risk of late CMV reactivation remains high. Since virus-specific T cells are capable of controlling CMV replication, we assessed whether their quantification shortly before hospital discharge allows risk-stratification for late CMV reactivation. Methods Allogeneic HCT recipients of all ages at the Fred Hutchinson Cancer Research Center who were CMV-seropositive or had a CMV-seropositive donor and participated in a prospective immune monitoring study were analyzed if they received at least 50% of weekly scheduled viral load (VL) PCR tests between days 100 and 180 after transplantation. CMV-specific cellular immunity was quantified via interferon-gamma ELISpot assay from cryopreserved peripheral blood mononuclear cells (PBMC) that were collected around day 90 after HCT. Cells were stimulated with peptide pools spanning the entire IE-1 and pp65 proteins (15mers with 11 amino acid overlap). Univariate Cox regression analysis was performed to determine hazard ratios (HR) for developing CMV reactivation at ≥50 IU, ≥500, ≥1000 IU/mL using different threshold levels (10, 50, or 100 spot-forming units (SFU) per 200,000 PBMC) for IE-1 and pp65. Results 13 out of 69 patients (18.8%) in this cohort showed clinically significant viral reactivation between days 100 and 180 after HCT, defined as having a viral load ≥1000 IU/ml at least once during the observational period (this level is commonly used to start preemptive treatment in the late setting). The best separation of risk was achieved by using a threshold of 10 SFU for both antigens: Individuals with less than 10 SFU for IE-1 or pp65 had an increased risk for CMV reactivation (HR 8.49, 95% CI, p=0.04) compared to patients with at least 10 SFU for both antigens. Correspondingly, the cumulative incidence of CMV reactivation was significantly higher in the group of patients with Conclusions Quantification of CMV-specific cellular immune responses around day 100 can help identify patients that are at risk for developing clinically significant late CMV reactivation. Both IE-1 and pp65 are useful for predicting CMV reactivation. Further studies with larger cohorts should be performed to confirm the suggested threshold levels of immune protection.