Standard Duration of Varicella Zoster Virus Prophylaxis May be Inadequate to Prevent Viral Reactivation in Patients with Chronic Lymphocytic Leukemia Treated with Front-Line Bendamustine Plus Rituximab

Background: Bendamustine plus rituximab (BR) is an effective front-line treatment for chronic lymphocytic leukemia (CLL). Immunosuppression induced from disease- and/or treatment-related factors can frequently result in infectious complications, including varicella zoster virus (VZV) reactivation. Antiviral prophylaxis is not universally prescribed for these patients and the duration of treatment is not well defined. We therefore retrospectively reviewed the incidence of VZV reactivation and the impact of antiviral prophylaxis on this complication in CLL patients treated with front-line BR.

Methods: We reviewed charts of 82 consecutive CLL patients with adequate clinical follow-up who were treated a median of 5 cycles with front-line BR at the Cleveland Clinic and University Hospitals of Cleveland from December 2005 to February 2015. Patients were grouped as having received antiviral prophylaxis or not. Baseline characteristics were compared between groups using the Chi-square test or Wilcoxon rank sum test. VZV reactivation from the start of front-line therapy was estimated using the cumulative incidence method and compared using the Gray test. Univariable prognostic factors for VZV reactivation were identified with Fine and Gray regression. The number of patients who reactivated was too small for multivariable analysis. All analyses were done using SAS software (SAS Institute Inc., Cary, NC, USA). All statistical tests were two-sided and P < 0.05 was used to indicate statistical significance.

Results: The median age of patients at the time of BR treatment was 69 years (range 44-93 years). 71% were male. Median follow-up was 1.5 years. 23 patients (28%) received antiviral prophylaxis, 20 acyclovir and 3 valacyclovir. The median duration of antiviral prophylaxis was 8.8 months. Patients who received antiviral prophylaxis were more likely to have had a history of VZV than those who did not receive prophylaxis (17% vs. 0%, P = 0.001). Additionally, patients who received antiviral prophylaxis were more likely to have received prior Zostavax compared to those who did not receive prophylaxis (22% vs. 5%, P = 0.022). All other baseline demographics including age, gender, ECOG PS, Rai stage, and cytogenetics were similar between the groups. 7 patients (8.5%) reactivated VZV following front-line BR therapy. Of the 7, 4 received antiviral prophylaxis with acyclovir and 3 did not receive prophylaxis (P = 0.20). Only one patient who reactivated VZV did so after receiving second line therapy. None of the patients who reactivated had a prior history of zoster activation. Notably, patients who received antiviral prophylaxis and reactivated VZV did so more than 6 months from the time of BR initiation, with reactivation occurring greater than 3 years after the start of front-line BR. [Figure 1]

Conclusions: CLL patients who receive front-line BR are at risk for VZV reactivation regardless of antiviral prophylaxis. History of VZV did not predict the likelihood of viral reactivation. Patients who received prophylaxis and experienced VZV reactivation did so more than 6 months from treatment initiation. These data suggest that when given to CLL patients receiving front-line BR, a standard 6-8 month regimen of VZV prophylaxis is insufficient. Rather, prolonged prophylaxis extending years after treatment completion may be more appropriate to prevent VZV reactivation. Further prospective studies are required to further explore this benefit.

Figure 1: Incidence of VZV reactivation over time in CLL patients treated with front-line BR who received antiviral prophylaxis.