In our study, we used a novel methodology to define seroinfection from immunogenicity data to compare the relative efficacy of PCVs in preventing infection. Our results using individual-level data from a global meta-analysis provide the first estimates of the comparative protection afforded by different pneumococcal vaccines and show that for many serotypes, carriage events are less common after PCV13 than PCV10, likely due to a higher antibody response. In addition, we quantify the relationship between the immune response to vaccination and protection against infection, measured serologically, and show that higher antibody responses in infants are associated with greater protection from infection.
The observed heterogenicity in immune responses was unexpected. We assumed that if one vaccine is able to induce more antibody than another, then it would do so with some degree of consistency across all trials. However, this was not what was observed. Comparisons of the same vaccines in different studies gave widely varying estimates, and although we have reported the summary GMR estimates in our immunogenicity meta-analyses, the large degree of between-study heterogeneity in these models means these overall estimates are difficult to interpret. In some settings, PCV13 performed better; yet, in others, PCV10 was the more immunogenic vaccine. Although there was no single study-level factor that could be identified that might explain the variation in estimates, only three candidate factors could be considered (location, schedule and co-administered vaccines) and data reporting on co-administered vaccines were not always comprehensive. The assays used have been WHO standardised and unlikely to cause this variation, and additionally, only studies directly comparing the two vaccines were included.
Of note, comparisons between vaccines from the same manufacturer (PCV13 vs. PCV7) were more consistent than comparisons between vaccines from different manufacturers. Immune interference (‘bystander effects’) has been noted when vaccines with similar components are co-administered,108 and this may affect the responses to one vaccine over another. It is interesting that 18C and 19F were serotypes that showed a very large degree of between-study heterogeneity. These two serotypes in PCV10 have different carrier proteins (18C is conjugated to TT and 19F is conjugated to diphtheria toxoid) and may be more susceptible than other serotypes to the co-administration of vaccines containing tetanus or diphtheria components. An additional potential confounder that is unmeasured in these studies is the degree of exposure to circulating serotypes of pneumococcus in each setting, which also has the potential to influence the immune response to vaccines.
These diverse immunogenicity findings from studies of the same vaccines raise the question of whether such differences in immunogenicity lead to meaningful differences in protection. If so, it may be important to know which vaccine performs better in which setting and further investigation into the predictors of the immune response to vaccines may be warranted. We addressed this question by modelling the relationship between seroefficacy estimates and immunogenicity comparisons (GMRs), analysed at the trial level across all serotypes and studies. This method capitalises on the observed between-study heterogenicity rather than being hindered by it. In our model, vaccines with higher antibody levels were also those with greater protection against subclinical infections in general. A vaccine with twice the antibody production was predicted to halve the rate at which carriage occurred.
Licensure of new vaccines is based on non-inferiority comparisons with current vaccines and the proportion of antibody responses above the agreed threshold as a minimum requirement. Once a vaccine meets this ‘at-least-as-good-as’ immunogenicity criteria, it has previously not been clear whether exceeding it is of benefit, and the WHO position paper states ‘It is unknown whether a lower serotype-specific GMC of antibody indicates less efficacy’.3 Our results show that lower protection against subclinical infection does indeed follow from lower antibody production and that two vaccines that produce a similar level of antibody will provide similar levels of protection, even if they are from different manufacturers.
The implications of these findings are of greatest importance when a new vaccine roll-out is being considered. Lower antibody production or lower seroefficacy for one vaccine product does not necessarily imply limited effectiveness against invasive pneumococcal diseases when considering vaccines such as PCV10 and PCV13 which are highly effective vaccines in many settings. Instead, lower antibody responses lead to less rapidly observed indirect protection after implementation into a national programme as a smaller proportion of transmission events are blocked by the vaccine. This is evident in the mathematical modelling in Chapter 4 which showed less rapid decreases in the number of cases of invasive disease when introducing PCV10 compared with PCV13.
For serotypes where protective impact has not been observed (serotype 3), new vaccines with substantially higher antibody responses may be needed. A Phase II clinical trial of PCV15 compared with PCV13 reported almost twice the antibody level for serotype 3 at 28 days post-primary series for PCV15 (GMR 1.93, 95% CI 1.71 to 2.18).82 Based on our modelled association between GMR and RR, the RR of seroinfection with PCV15 versus PCV13 would be 0.48 (95% CI 0.21 to 0.87). Previously reported vaccine effectiveness estimates against nasopharyngeal carriage of serotype 3 include –27% (95% CI –180 to 44) and 1% (95% CI –106 to 52),30,109 and these translate to point estimates of 39% (95% CI –16% to 66%) and 52% (95% CI 9% to 79%) vaccine effectiveness against carriage of this serotype with PCV15 based on the relationship: (VE(pcv15) = (1 − RR(pcv15 vs pcv13) × (1 − VE(pcv13)/100%)) × 100%).
Implications for practice
This evidence of differences in serotype-specific protection can be incorporated into cost-effectiveness models used to compare vaccine products.16 Cost-effectiveness studies have highlighted the lack of direct evidence of comparative efficacy of different PCVs, resulting in previous cost-effectiveness models that ignore serotype-specific differences and assume equivalent efficacy for all serotypes covered by different PCVs.110–112 Our study fills this evidence gap and allows researchers and policy-makers to use more accurate vaccine-specific models in decision-making.
Our cost-effectiveness analysis of a hypothetical scenario showed that introducing infant PCV13 was predicted to avert a higher burden of pneumococcal disease compared to PCV10. This would have realised a small saving of £13 million discounted over 24 years.
When considering the introduction of new pneumococcal vaccines into the routine immunisation schedule, we recommend that differences in antibody responses for different vaccines be considered in modelling scenarios as higher antibody responses result in reduced transmission and greater impact on invasive diseases. Vaccine-specific threshold prices can then be determined for cost-effective vaccines. Our analysis showed that due to its higher efficacy against some serotypes, a higher threshold price per dose could be paid for PCV13 while remaining cost-effective. Seroefficacy estimates can also be determined for new pneumococcal vaccines and could contribute to licensing or policy decisions in the future.
Strength and limitations
Seroefficacy analyses need to be restricted to serotypes contained in both vaccines. Comparing a vaccinated cohort to a cohort that is unvaccinated, or receives a vaccine that does not contain the serotype of interest, will result in biased estimates as the immune response after exposure to a pathogen will differ in children whose immune system is primed for that pathogen, when compared with a naïve population. For this reason, we restricted our seroefficacy analysis to shared serotypes between vaccines. While seroinfection is most likely an indicator of nasopharyngeal carriage, it may also represent cases of asymptomatic bacteraemia.
Our analyses are based on a large set of studies conducted in infants for the most commonly used vaccines and make use of data originally collected for a different purpose than ours. We used the time points available in these studies. However, the time points one might use if designing a study for the purpose of calculating seroefficacy may differ and would likely include an additional time point 6–9 months after the booster dose. Without this time point, we are extrapolating pre-booster efficacy to post-booster time periods, and the impact of this assumption is unknown.
The mathematical and economic models used were based on outputs from our NMA models which contained significant heterogeneity. The potential for bias when using inconsistent data for modelling scenarios in this situation is hard to quantify but needs to be considered.
Public and patient involvement
The Oxford Vaccine Group public and patient involvement (PPI) group was involved at the design stage in the development of the plain language summary for the submitted grant proposal. Due to the nature of the project being a reanalysis of data from previously completely studies, there was no involvement of the PPI group in the data collection or analysis. Dissemination of results was discussed with the PPI lead, but as the conclusions and recommendations are relevant only to policy-makers and academics and as decision on vaccine product choice and purchases are made by the Department of Health for the whole of England, not by individuals, the PPI lead felt that further PPI review for the project would not be a good use of the PPI volunteer time.
Equality, diversity and inclusivity
The research team for the project was small but reflected contributions from under-represented groups. In particular, a large proportion of the key contributors were female, including first, second and senior authors, including the principal investigator. Junior members of the team are fully acknowledged with authorship on the final report and all publications arising from the project.
Data from individual studies included in the systematic review were from a wide range of countries and regions.
Conclusions
In conclusion, we estimated serotype-specific difference in both seroefficacy and immunogenicity between PCV10 and PCV13. Higher IgG antibody levels confer better protection against seroinfection. This methodology can be further used to compare novel high-valent PCVs and to inform cost-effectiveness models.
