VZV Foundation - A Decade Of Leadership In The Fight Against VZV Infections

	Oral Presentation Abstracts: 25D [25D] PREDICTION OF POSTVACCINATION VARICELLA BREAKTHROUGH RATES BASED ON THE VARICELLA ANTIBODY DETERMINED BY gp-ELISA 6 WEEKS POSTVACCINATION I.S.F. Chan (1), S. Li (1), J.F. Heyse (1), H. Matthews (1), C.Y. Chan (2), S.J.R. Vessey (2), B.J. Kuter (2), and J.C. Sadoff (2) Departments of (1) Biostatistics and (2) Vaccine Infectious Disease, Merck Research Laboratories, West Point, PA, USA. Objective: To develop a statistical model describing the relationship between the varicella antibody response 6 weeks after vaccination and the postvaccination varicella breakthrough infection rate. Methods: The varicella antibody response determined by gpELISA 6 weeks postvaccination with Oka/Merck varicella vaccine has been shown to be inversely correlated with cumulative varicella breakthrough infection rates in a cohort of 1,087 children who were followed for 7 years postvaccination. Several accelerated failure time models developed to predict breakthrough rates on the basis of 6 week postvaccination varicella antibody responses were evaluated for 'goodness of fit' to these long-term follow up data. The applicability of the final model was assessed by comparing model predicted and observed breakthrough rates in other cohorts of vaccinees where long-term follow-up was available. The model was used to predict breakthrough rates based on the 6 week postvaccination varicella antibody responses in subjects vaccinated with either Oka/Merck varicella vaccine or Varilrix Ô (SB Bio (Oka-RIT)). Results: The final statistical model selected was a log-normal accelerated failure time model with varicella antibody titer 6 weeks postvaccination and age at vaccination as covariates. Based on this model, Oka/Merck varicella vaccine at doses between 3,000 and 50,000 PFU was predicted to have an efficacy of ~88% compared with a predicted efficacy of ~82% for Varilrix Ô when administered at the release dose range. Conclusions: A statistical model can be used to predict varicella breakthrough rates on the basis of varicella antibody responses 6 weeks postvaccination. Corresponding Author: Dr. S.J.R. Vessey, Associate Director, Clinical Vaccine Research, Merck & Co., Inc.; Phone: (610) 397-3770, Fax: (610) 397-3371, Email: rupert_vessey@merck.com

[25D]

PREDICTION OF POSTVACCINATION VARICELLA BREAKTHROUGH RATES BASED ON THE VARICELLA ANTIBODY DETERMINED BY gp-ELISA 6 WEEKS POSTVACCINATION

I.S.F. Chan (1), S. Li (1), J.F. Heyse (1), H. Matthews (1), C.Y. Chan (2), S.J.R. Vessey (2), B.J. Kuter (2), and J.C. Sadoff (2)
Departments of (1) Biostatistics and (2) Vaccine Infectious Disease, Merck Research Laboratories, West Point, PA, USA.

Objective:
To develop a statistical model describing the relationship between the varicella antibody response 6 weeks after vaccination and the postvaccination varicella breakthrough infection rate.

Methods:
The varicella antibody response determined by gpELISA 6 weeks postvaccination with Oka/Merck varicella vaccine has been shown to be inversely correlated with cumulative varicella breakthrough infection rates in a cohort of 1,087 children who were followed for 7 years postvaccination. Several accelerated failure time models developed to predict breakthrough rates on the basis of 6 week postvaccination varicella antibody responses were evaluated for 'goodness of fit' to these long-term follow up data. The applicability of the final model was assessed by comparing model predicted and observed breakthrough rates in other cohorts of vaccinees where long-term follow-up was available. The model was used to predict breakthrough rates based on the 6 week postvaccination varicella antibody responses in subjects vaccinated with either Oka/Merck varicella vaccine or Varilrix Ô (SB Bio (Oka-RIT)).

Results:
The final statistical model selected was a log-normal accelerated failure time model with varicella antibody titer 6 weeks postvaccination and age at vaccination as covariates. Based on this model, Oka/Merck varicella vaccine at doses between 3,000 and 50,000 PFU was predicted to have an efficacy of ~88% compared with a predicted efficacy of ~82% for Varilrix Ô when administered at the release dose range.

Conclusions:
A statistical model can be used to predict varicella breakthrough rates on the basis of varicella antibody responses 6 weeks postvaccination.

Corresponding Author: Dr. S.J.R. Vessey, Associate Director, Clinical Vaccine Research, Merck & Co., Inc.; Phone: (610) 397-3770, Fax: (610) 397-3371, Email: rupert_vessey@merck.com