Trial of Pneumococcal Vaccine Schedules in Ho Chi Minh City, Vietnam

Overview

Pneumococcus is a group of bacteria that can cause pneumonia, meningitis and other diseases. These bacteria normally live in the nose of humans and are spread from person to person by touching or sneezing. There are vaccines available to protect against infection with these bacteria, and pneumococcus is currently the leading vaccine-preventable cause of death in young children. In countries where pneumococcal vaccine (PCV) has been introduced, there has been a big impact on the amount of disease caused by these bacteria. However, many countries, especially developing countries, are yet to introduce PCV as part of their routine immunizations. Currently a total of four doses of PCV is recommended, and the main barrier to vaccine introduction is cost. This study aims to identify a vaccination schedule to make PCV more effective and affordable for Vietnam and other developing countries. This study has two distinct purposes: 1) to compare different dosage schedules of PCV and 2) to compare different PCV vaccines. 1. Schedules of Synflorix (PCV10) involving a three, two or one dose PCV primary series and two booster options will be compared. Comparisons will be made firstly in terms of measures of immunity to the vaccine, and secondly in terms of the effect of vaccination on the carriage of bacteria in the nose. 2. The responses to PCV10 and Prevenar-13 (PCV13) will be compared, in the schedule most likely to be considered for global use. Again, comparisons will be made in terms of measures of immunity and effect on carriage in the nose. Infants aged two months will be randomly assigned to one of six study groups and will provide up to four blood samples for analysis of the measures of immunity and five nose swabs for analysis of carriage of bacteria. Infants will be followed up 8-9 times until the age of 24 months. An additional control group will be recruited at 18 months of age and also followed up until 24 months of age. The results of this study will be used to facilitate decision making, at global and national levels, regarding introduction of PCV.

Introduction The overall purpose of this study is to investigate simplified childhood vaccination schedules that are more appropriate for developing country use. This study is specifically designed to address two independent questions within a single study: 1. What is the optimal schedule for provision of EPI vaccines with the incorporation of PCV10? Schedules involving a three, two or one dose PCV10 primary series will be evaluated, timed around options for simplification of the Expanded Programme of Immunization (EPI) schedule for developing countries. A simplified schedule with the pneumococcal booster dose brought forward closer to the peak incidence of disease is likely both to increase compliance and vaccine effectiveness. A booster at nine months of age coincides with the usual time for administration of measles vaccine, whereas a booster at six months of age would provide earlier protection and may enable a further abbreviated 1+1 pneumococcal vaccination schedule. 2. How do the responses to PCV vaccination with PCV10 or PCV13 compare? PCV10 and PCV13 are the two PCVs available through the Advanced Market Commitment (AMC) mechanism, a mechanism that provides funds for vaccine introduction into developing countries. However, there have been no studies to date directly comparing these two vaccines. Directly comparing these two vaccines will provide useful information to countries considering introduction of PCV. There are important differences between these vaccines. PCV10 includes ten pneumococcal serotypes and PCV13 includes thirteen. PCV10 uses a non-typeable Haemophilus influenzae (NTHi) Protein D carrier, which may confer protection against H. influenzae, and PCV13 uses a CRM197 carrier. It is of interest to know whether these vaccines differ either in their immunogenicity or their impact on nasopharyngeal (NP) carriage. Design Infants will be randomized to one of six study arms (A-F). All infant participants receive four doses of Infanrix-hexa (DTaP-Hib-HBV-IPV) and at least two doses of PCV. The PCV schedules to be evaluated are: a 3+1 PCV10 schedule at 2, 3, 4 and 9 months of age (Arm A); a 3+0 PCV10 schedule at 2, 3 and 4 months of age (Arm B); a 2+1 PCV10 schedule at 2, 4 and 9 months of age (Arm C); a 1+1 PCV10 schedule at 2 and 6 months of age (Arm D); a 2+1 PCV13 schedule at 2, 4 and 9 months of age (Arm E). Arm F, the control group, receives two doses of PCV10 at 18 and 24 months of age. An additional control group (Arm G) will be recruited at 18 months of age and will receive Infanrix-hexa at 18 months of age and a single dose of PCV10 at 24 months of age. Reactogenicity will be assessed following all vaccination visits through the use of diary cards. Participants from arms A-E will provide six NP swabs for analysis of the NP carriage outcomes, at 2, 6, 9, 12, 18 and 24 months of age; and will provide four blood samples over the course of the trial for analysis of vaccine responses. Blood 1 will be taken four weeks post-primary series; Blood 2 will be taken pre-booster (arms A, C, D and E) or at 9 months of age (subset of arm B); and Blood 3 will be taken four weeks post-booster (arms A, C, D and E) or at 10 months of age (arm B). An additional blood sample will be taken at: 18 months of age for a subset of arms A, B, C, D and E; 2 months of age for a subset of arm A; 6 months of age for a subset of arms B and C; 9 months of age for a subset of arm D; or 3 months of age for a subset of arm E. Participants from the control arms will provide NP swabs at 2, 6, 9, 12, 18 and 24 months of age (arm F) or at 18 and 24 months of age (arm G), and will provide blood samples at 18 (Blood 4), 19 (Blood X) and 24 (Blood Y) months of age. Objectives 1. What is the optimal schedule for provision of EPI vaccines with the incorporation of PCV10? The primary objective is to compare a 2+1 schedule at 2, 4 and 9 months of age with a 3+1 schedule at 2, 3, 4 and 9 months of age, with a primary outcome of the immunogenicity of PCV10, four weeks post-primary series (Arm C vs. Arm A+B). Secondary objectives are to investigate an experimental 1+1 schedule at 2 and 6 months of age (Arm D vs. Arm A+B and Arm D vs. Arm C), and to assess the impact of a booster dose on carriage (Arm A vs. Arm F and Arm A vs. Arm B). 2. How do the responses to PCV vaccination with PCV10 or PCV13 compare? The primary objective is to compare a PCV13 schedule at 2, 4 and 9 months of age with a PCV10 schedule at 2, 3, 4 and 9 months of age, with a primary outcome of the immunogenicity of PCV, four weeks post-primary series (Arm E vs. Arm A+B). Secondary objectives are to compare a PCV13 schedule at 2, 4 and 9 months of age with a PCV10 schedule at 2, 4 and 9 months of age (Arm E vs. Arm C), and to compare the responses to a single dose of PCV13 or PCV10 (Arm E vs. Arm D). Other objectives are: to examine the decline in pneumococcal antibody levels over time (Arm B); to describe the serotype profile of transferred maternal pneumococcal antibodies (Arm A); and to describe the early rates of carriage (Arms A-F); to evaluate a single dose of PCV10 at 18 months of age (Arm F); and to evaluate the immunogenicity of Infanrix-hexa at 18 months of age in children who have received three doses of Infanrix-hexa or three doses of Quinvaxem (DTwP-Hib-HBV) in infancy. Sample Size The proposed infant sample size is 1200 with an allocation ratio of 3:3:5:4:5:4, resulting in arm sizes of: A=150, B=150, C=250, D=200, E=250, F=200. Sample size calculations were based on the primary outcomes for each of the two study questions: the post-primary series immunogenicity comparing 1) a two dose (Arm C) and three dose (Arm A+B) PCV10 primary series and 2) a two dose PCV13 (Arm E) and three dose PCV10 (Arm A+B) primary series. A non-inferiority margin of 10% difference in absolute risk (Arm A+B minus Arm C or Arm E), as used by regulatory authorities, is deemed clinically significant. The Farrington-Manning (1990) method was used for the sample size/power estimation, assuming one-sided 5% type I error. If the alternative hypotheses of non-inferiority are accepted for at least 7 out of 10 serotypes, overall non-inferiority will be declared. The power for testing individual serotype hypotheses was calculated using PASS Software 2002. The power for rejecting the overall null hypothesis was estimated by simulation, using a tailor-made simulation program written for implementation in Stata with 10,000 replications. A sample size of 1200 results in \>99% power for rejecting the overall null hypothesis for each of the two study questions, allowing for 5% loss to follow-up at four weeks post-primary series. An additional 200 participants aged 18 months (Arm G) will be recruited at the same time as participants from Arms A-F reach 18 months of age, bringing the total sample size of the trial to 1400 participants.