Background: Nepalese infants receive ten-valent pneumococcal conjugate vaccine (PCV10) with a 1 month interval between priming doses for programmatic reasons. We aimed to investigate whether immune responses to PCV10 serotypes were non-inferior if the second priming dose of PCV10 was delivered at a 1 month interval as opposed to a 2 month interval.
Methods: We did an open-label, randomised, parallel group trial in healthy Nepalese infants aged 40-60 days at Patan Hospital, Kathmandu, Nepal. Children were eligible for inclusion if they were healthy, were born at more than or equal to 37 weeks' gestation, were residing in Kathmandu, and had not had any previous vaccinations other than BCG, and oral polio vaccine. Participants were randomly assigned (1:1) by means of a computer-generated list with randomly varying permuted block sizes accessed through a validated web-based interface, to receive PCV10 either at 6 weeks and 10 weeks of age (6 + 10 group) or at 6 weeks and 14 weeks of age (6 + 14 group), with both groups receiving a booster at 9 months of age. Laboratory staff, masked to study intervention, analysed serum samples for antibodies against PCV10 serotypes by ELISA. The primary outcome was to determine whether the 6 + 10 schedule was non-inferior to the 6 + 14 schedule at 9 months of age, on the basis of the proportion of infants with serotype-specific IgG greater than or equal to 0·35 μg/mL. Non-inferiority was established with a 10% margin, and the primary endpoint was measured in a modified intention-to-treat population, which included only participants who successfully had a blood sample collected. This trial is registered at ClinicalTrials.gov, number NCT02385513.
Findings: Between Aug 21, 2015, and April 4, 2016, 304 Nepalese children were randomly assigned to either the 6 + 10 group (n=152) or the 6 + 14 group (n=152). At 9 months of age, the 6 + 10 schedule was non-inferior for serotype 5 (79 [55·2%] of 143 vs 78 [53·4%] of 146, difference 1·82% [95% CI -9·6 to 13·25], p=0·021), serotype 9V (66 [46·1%] of 143 vs 55 [37·6%] of 146, difference 8·48% [-2·84 to 19·8], p=0·001), serotype 14 (110 [77·4%] of 142 vs 110 [74·8%] of 147, difference 2·63% [-7·27 to 12·54], p=0·006), and serotype 19F (135 [95%] of 142 vs 146 [100%] of 146, difference -4·92% [-9·86 to 0], p=0·022). At the same timepoint, non-inferiority was not shown for serotype 1 (36 [25·1%] of 143 vs 42 [28·5%] of 147, difference -3·39% [95% CI -13·56 to 6·77], p=0·102), serotype 4 (70 [48·9%] of 143 vs 87 [59·1%] of 147, difference -10·23% [-21·64 to 1·18], p=0·516), serotype 6B (96 [67·1%] of 143 vs 114 [77·5%] of 147, difference -10·41% [-20·65 to -0·18], p=0·532), serotype 7F (99 [69·2%] of 143 vs 109 [74·1%] of 147, difference -4·91% [-15·26 to 5·42], p=0·168), serotype 18C (89 [62·2%] of 143 vs 114 [77·5%] of 147, difference -15·31% [-25·78 to -4·83], p=0·840), and serotype 23F (37 [25·8%] of 143 vs 41 [27·8%] of 147, difference -2·01% [-12·19 to 8·16], p=0·062). After the booster dose, at 10 months of age, there were no significant differences in immunogenicity (proportion of children with antibody greater than or equal to 0.35 μg/mL) for any of the ten serotypes, when comparing the two schedules. Serious adverse events occurred in 32 participants, 11 (7%) of 152 in the 6 + 10 group and 21 (14%) of 152 in the 6 + 14 group.
Interpretation: The 6 week, 14 week, and 9 month schedule should be implemented where possible. However, post-booster responses, which are thought to drive herd immunity, were similar in the two schedules. Therefore, the 6 week, 10 week, and 9 month schedule is an alternative that can be used when logistically necessary, and is expected to provide herd protection.
Funding: Gavi, the Vaccine Alliance.
Copyright © 2019 Elsevier Ltd. All rights reserved.