Evaluation of Varicella Zoster Prophylaxis after Allogeneic Hematopoietic Cell Transplantation Using Valacyclovir Followed By Vaccination

Rationale: Varicella zoster virus (VZV) disease occurs frequently post allo-HCT. While cutaneous disease is often complicated by debilitating post-herpetic neuralgia (PHN), visceral or CNS disease can be fatal. Studies have demonstrated that VZV disease is reduced during prophylaxis with acyclovir; however, patients frequently develop VZV disease after discontinuation of acyclovir. Recently, the varicella vaccine has been found to be safe and immunogenic post allo-HCT. The optimal prophylactic strategy for VZV disease post allo-HCT has not been established. Here we present a retrospective single-center study, comparing pre-2007 prophylaxis, which consisted of less than 2 years of post allo-HCT acyclovir, with post-2007 prophylaxis, which consisted of 2 years of valacyclovir prophylaxis followed by live attenuated varicella vaccination.

Objective: To evaluate the hypothesis that the cohort treated with the pre-2007 strategy ("old strategy") had a higher cumulative incidence of VZV disease and PHN compared to the cohort treated with the post-2007 strategy ("new strategy").

Methodology: Charts of patients undergoing allo-HCT in Calgary between January 2004 and June 2011 were reviewed. VZV disease was defined as clinical zoster or visceral/CNS disease confirmed by immunostain or PCR. PHN was defined as pain in the affected dermatome persisting more than 3 months after the onset of the rash. Cumulative incidence of VZV disease and PHN were compared using competing risks regression (Fine-Gray), treating relapse, graft failure, second malignancy and death as competing risks.

Results: 482 patients underwent first allo-HCT in Calgary during the review period. 21 were excluded due to inadequate data. 119 were treated with the new strategy, 116 with the old strategy and 226 experienced a competing event while on prophylaxis. Some patients in the old strategy group received late VZV vaccination (at >2 years). The old strategy was associated with a trend towards higher cumulative incidence of VZV disease (sHR 1.57, 95% CI 0.95-2.62, p=0.08) and with significantly higher cumulative incidence of PHN (sHR 9.94, 95% CI 1.24-80.05, p=0.03). 3 episodes of VZV disease in the old strategy group and 7 episodes in the new strategy group were associated with deviation from protocol (most associated with medication noncompliance). When events associated with protocol deviation were excluded, the old strategy led to a significantly higher cumulative incidence of VZV disease (HR 2.40, 95% CI 1.32-4.38, p=0.004) and PHN was not observed in the new group (p= <0.01). Vaccination was safe; however 10% of patients developed immediate or late cutaneous VZV disease. PHN was not observed in the patients who developed the VZV disease post vaccination.

Conclusions: The new strategy was associated with a trend towards less VZV disease and significantly less PHN. Noncompliance was a significant contributor to VZV disease in the new strategy group. If compliance is ensured, the new strategy significantly reduces VZV disease and eliminates PHN.