Authors: David C. LaFon
Categories: Pneumococcal, Streptococcus pneumoniae, immunoassays, epidemiology, clinical trials, vaccine response
Source: Human Vaccines & Immunotherapeutics
Authors: David C. LaFon
Pneumococcal conjugate vaccines (PCVs) have resulted in a dramatic reduction in vaccine-type disease in both children and adults. Several expanded PCVs have recently been approved and incorporated into immunization programs for children and adults, with numerous additional vaccine formulations in development or undergoing evaluation in clinical trials. This concise review provides an introduction to the evolving approaches to vaccine evaluation used in the decades since the initial studies of PCV7, followed by a contemporary assessment of challenges and future directions for studies of efficacy, immunogenicity, and real-world effectiveness.
Following their implementation in national immunization programs beginning in the early 2000s, pneumococcal conjugate vaccines (PCVs) represented a major advance in the prevention of pneumococcal disease in children and adults. While the pneumococcal polysaccharide vaccine (PPV23) had previously been licensed for prevention of invasive pneumococcal disease (IPD; infection in the bloodstream, meningeal space, or other usually sterile sites) in adults),^1^ PPV23 results in a T-cell independent response without immune memory, and is poorly immunogenic in infants and young children.^2,3^ PCVs, which contain pneumococcal capsular polysaccharide linked to a carrier protein, result in a T-cell dependent immune response with induction of immune memory and were developed to address these limitations of PPV23.^4^
The 7-valent PCV (Prevnar; Wyeth/Pfizer) was licensed and approved in the United States (US) based on randomized clinical trials in the late 1990s that demonstrated efficacy in prevention of IPD and otitis media in infants and young children.^5^ Assays that measured serotype-specific pneumococcal IgG levels and antibody function (opsonophagocytic responses) were pivotal for assessment of immunogenicity in these studies.^1,6^ The World Health Organization pneumococcal enzyme-linked immunosorbent assay (WHO-ELISA)^7^ and multiplexed opsonophagocytosis assay (MOPA)^8^ were developed to meet the demands for reliable and reproducible measurement of pneumococcal antibody levels and function, respectively.
In the years following introduction of PCV7, there was a sharp decline in vaccine-type IPD among vaccinated infants.^9^ There was also a reduction in vaccine-type pneumococcal disease in unvaccinated adults, indicating a strong indirect effect as a result of decreased circulation of PCV7 serotypes in children.^10^ However, this reduction in vaccine serotypes corresponded with a rise in the prevalence of non-vaccine serotypes in both nasopharyngeal carriage and disease, a phenomenon referred to as “serotype replacement.^11^ ” These effects provided impetus for the development of new PCVs containing increasing numbers of serotypes. PCV10 (Synflorix; GSK) and PCV13 (Prevnar-13; Pfizer) were licensed based on demonstration of noninferior immune responses compared to PCV7.^12^ This “immunobridging” approach was more recently used for the evaluation and licensure of PCV15 (Vaxneuvance; Merck)^13^ and PCV20 (Prevnar-20; Pfizer)^14^ in pediatric populations. While PCVs had been used in infant immunization programs in high-income countries for decades, many countries had, until recently, lacked access to affordable PCVs.^15^ Assessment of immunogenicity facilitated the development and approval of a lower-cost 10-valent PCV produced by Serum Institute of India (SII-PCV10; Pneumosil)^16^ that has expanded global access to PCVs.
PCV implementation in adult vaccination programs occurred later than inclusion in pediatric schedules in the US, and has not been adopted as widely on a global scale. In 2012, the US Centers for Disease Control (CDC) first recommended PCV13 in series with PPV23 for use in adults with immunocompromising or other selected high-risk conditions.^17^ Several years later, this recommendation was expanded to include all adults aged ≥65 years,^18^ based on the results of a large randomized controlled trial demonstrating efficacy in preventing vaccine-type community-acquired pneumonia (CAP).^19^ Over the following years, this recommendation was updated for conditional use based on shared decision-making,^20^ until supplanted by updated guidelines, the most recent of which advise use of a PCV among adults ≥50 years old as well as younger adults with chronic medical conditions.^21^ Currently approved PCVs in adults include PCV15, PCV20, and most recently PCV21 (Capvaxive; Merck). These PCVs were investigated and approved based on noninferior immunogenicity for serotypes shared with previously licensed vaccines.^21^ Among the latest generation of PCVs, PCV21 is unique in that it was the first that was developed for adults, including eight serotypes that are not present in other PCVs.^21^
PCVs represent a remarkable success story in the prevention of pneumococcal disease. An increasing number of vaccines targeting ever-expanding numbers of serotypes demands a careful consideration of how efficacy and effectiveness will be evaluated for current and investigational PCVs. This review provides a concise discussion of these issues, concluding with future directions for vaccine-related studies.
The original efficacy studies for PCV7 in infants and children involved large randomized controlled trials, including a study by the Northern California Kaiser Permanente Vaccine Study Center Group that enrolled 37,868 children.^5^ The study took place across numerous clinical sites and took nearly 3 years to enroll, with additional follow-up time required for efficacy outcome ascertainment. The vaccine demonstrated excellent efficacy in preventing vaccine-type IPD, as well as some efficacy for prevention of otitis media.
Enrollment and longitudinal follow-up of large numbers of participants in clinical efficacy trials presented a barrier to trials evaluating clinical efficacy for subsequent vaccine formulations. In addition, the sharp reduction in vaccine-type IPD in the years following PCV7 implementation^9^ further complicated the issue of evaluating clinical efficacy for expanded PCVs due to decreasing incidence of the primary outcome measure (IPD) for serotypes common to both new and existing vaccines. The clear efficacy and effectiveness of PCV7 also precluded randomized trials that involved withholding PCVs, based on ethical issues arising from a lack of equipoise regarding the likelihood of benefit for prevention of pneumococcal disease prevention in children. For these reasons, PCV13 was licensed and approved based on studies that demonstrated noninferior immunogenicity for serotypes shared with PCV7.^12^ Subsequent PCVs were evaluated for use in children by bridging immunogenicity data in a similar fashion, and repeated efficacy studies for prevention of IPD or other pneumococcal diseases have not been performed for each new vaccine.^14^
The immunoassays used for initial evaluation of PCV7 have been instrumental in this ‘immunobridging’ approach. Pooled immunogenicity and outcomes data from trials of PCV7 were used to estimate a serotype-specific IgG level of 0.35 µg/mL as corresponding with protection from vaccine-type IPD in children.^6^ While this cutoff has been used across all serotypes as a surrogate for protection in subsequent pediatric clinical trials,^22^ it is important to note that this estimate may not correspond with protection for all serotypes (beyond PCV7 serotypes that were studied) or from nasopharyngeal colonization, otitis media, or pneumonia. The WHO ELISA was a third-generation assay developed to improve upon the specificity of previous versions, and enable reproducible quantification of serotype-specific pneumococcal IgG levels by different laboratories.^23^ While WHO ELISA provided specificity and reproducibility for PCV7 evaluation, it became impractical for studies of subsequent PCVs. WHO ELISA is cumbersome to perform, requiring a separate assay run for each serotype tested. As the PCVs have expanded to include greater numbers of serotypes, the analytical burden and volume of serum required (which can be a particular issue for infants) for WHO ELISA became increasingly problematic.^4^ The immunogenicity of subsequent PCVs has been evaluated using multiplexed assays (e.g. electrochemiluminescence or multiplexed bead array assays) that can measure numerous serotypes in a single assay run.^24^
While measurement of serotype-specific IgG levels remains the primary means of evaluating the immunogenicity of PCVs, studies of antibody responses following the introduction of PCV7 indicated that measurement of IgG levels alone may not be sufficient to indicate protection from pneumococcal infections. PCV7, which contains serotype 19F, leads to production of IgG levels ≥0.35 µg/mL against 19A, a related serotype that is not included in the vaccine.^25^ Despite the presence of IgG levels that exceeded the putative threshold for protection, measurement of the functional capacity of serotype 19A antibodies using an opsonophagocytosis assay (OPA) indicated that they were ineffective in killing bacteria and thus unlikely to result in protection from pneumococcal infections in vivo. Since these studies were performed, OPA has been included as a requisite secondary endpoint for pediatric PCV trials.^14^ A protocol for the multiplexed opsonophagocytosis assay (MOPA) has been accepted internationally for pneumococcal vaccine evaluation and is available at www.vaccine.uab.edu.
The Community-Acquired Pneumonia Immunization Trial in Adults (CAPITA),^19^ which led to the initial US CDC recommendation for PCV13 use in older adults,^18^ was conducted roughly a decade after initial use of PCV7 in infants. CAPITA enrolled 84,496 adults aged ≥65 years in a randomized, double-blind placebo-controlled trial evaluating the efficacy of PCV13 for prevention of vaccine-type CAP. Notably, the study utilized serotype-specific urine antigen detection (SSUAD), a novel assay designed to overcome the limitations of conventional diagnostic approaches for non-bacteremic pneumococcal CAP. While increasingly uncommon due to the success of PCVs, ascertainment of vaccine-type IPD as an outcome in efficacy studies is relatively straightforward since cases are typically severe enough to require hospitalization and result in recovery of bacteria from a usually sterile site to facilitate clear identification of the causative serotype. In contrast, serotype-specific diagnosis of non-bacteremic pneumococcal CAP is highly challenging since infections are often treated outside of the hospital and lack a sensitive, confirmatory diagnostic test. Conventional microbiological cultures are affected by early use of broad-spectrum antibiotics as well as frequent contamination with other oropharyngeal and respiratory flora. Likewise, assays that detect pneumococcal cell wall polysaccharide in the urine (e.g. BinaxNOW) have variable sensitivity, cannot indicate causative serotype, and may be falsely positive in children with heavy nasopharyngeal colonization.^26,27^ To address these limitations, the CAPITA study utilized SSUAD, a multiplexed assay that detects pneumococcal capsular polysaccharide excreted into in the urine of patients with CAP.
Subsequent PCVs (PCV15, PCV20, and PCV21) were evaluated based on bridging of immunogenicity data,^21^ with several key differences and caveats compared to the approaches used in children. Older adults may be susceptible to pneumococcal infections despite the presence of ‘normal’ serum IgG levels, indicating a discordance between antibody levels and function.^28,29^ Since functional antibody responses may serve as a better indicator of protection, OPA responses have been accepted as the primary immunogenicity endpoint in adult PCV trials.^30^
Lack of evidence-based cutoffs that correspond with protection from infections also presents challenges for use of pneumococcal IgG levels in adult studies. While PCV7 trials provided a protective cutoff for pneumococcal IgG levels in children, serotype-specific thresholds in adults have been more elusive.^31^ While clinicians who are evaluating patients for suspected antibody deficiency syndromes may utilize the pediatric cutoff 0.35 µg/mL or alternatively 1.3 µg/mL (which was initially derived using arbitrary cutoffs and extrapolated from now-obsolete assays),^32^ there are not serotype-specific pneumococcal IgG cutoffs that have been validated for use as surrogates of protection in adult vaccine studies.
Following introduction of a novel PCV into a national immunization program, ongoing surveillance programs have provided key information regarding the real-world effectiveness of the vaccine. Following introduction of PCV7 (which includes serotypes 4, 6B, 9 V, 14, 18C, 19F, and 23F) in infants, there was a sharp and sustained reduction in vaccine-type IPD among vaccinated infants (direct effect),^9^ which was followed by a reduction in vaccine-type infections in unvaccinated adults (indirect effect).^10^ However, these effects have been less consistent across the additional serotypes unique to PCV13 (1, 3, 5, 6A, 7F, 19A) following its widespread use in infants and children, and pneumococcal surveillance programs have indicated that several PCV13 serotypes have persisted in causing infections in adults. Serotype 3 has a distinct mechanism for capsule production that can be resistant to immune clearance. Serotype 3 is less immunogenic relative to other vaccine serotypes, and remains a major cause of pneumonia and IPD for children and adults.^33^ Despite inclusion in both PCV7 and PCV13, serotype 4 strains containing a capsular variant have been implicated in outbreaks of IPD in adults experiencing homelessness.^34^ Host factors that impair antibody production and/or antibody-mediated bacterial clearance may also be involved with the persistence of these serotypes in high-risk individuals. Investigation of the mechanisms that underlie the continued circulation and pathogenicity of these serotypes can be leveraged in the design of future PCVs, and will inform best practices for vaccination in specific populations.
Surveillance programs to monitor serotype-specific epidemiology of pneumococcal infections are also critical for monitoring the dynamic effects of serotype replacement, and for characterizing geographic and temporal heterogeneity in serotype distribution. To estimate the overall effect of a PCV on pneumococcal disease, it is essential to account for corresponding increases in non-vaccine serotypes. Effectiveness studies in adult populations must also consider the indirect effect resulting from pediatric PCV programs, which could attenuate the anticipated direct effects of vaccinating adults. Prevalent serotypes and overall rates of carriage and disease exhibit marked variation by geographic location, living environment, environmental exposures, and other external factors.^35^ For these reasons, national and regional (and sometimes even local) surveillance programs are required to guide best practices for vaccination and vaccine development.
Several additional factors have arisen over the past several years that could further complicate assessment of pneumococcal epidemiology and the real-world efficacy of PCVs. The starkest disruption to the overall trends in pneumococcal epidemiology occurred during the coronavirus disease 2019 (COVID-19) pandemic. As a result of social distancing and other pandemic-related precautions, the usual seasonal fluctuation in pneumococcal disease was abolished, with very few cases of IPD overall.^36^ In the United Kingdom, this presented challenges in evaluating the effects of a reduced dosing schedule for PCV13 in infants that was implemented just prior to the pandemic.^33^ The introduction of several different PCVs in relatively short succession will make it challenging to disentangle the efficacy of individual vaccines for common serotypes, and there may be different priorities for efficacy studies across PCV formulations. While PCV21 was developed to target distinct serotypes causing pneumococcal infections in adults, its approval and recommendation were based on immunogenicity data.^21^ Assessment of real-world effectiveness is necessary to determine whether there are advantages to this approach to adult vaccination. On the other hand, PCV20 utilized a more conventional approach by adding emerging serotypes to those covered by previous PCVs, and efficacy studies should therefore focus on the indirect effect for these additional serotypes, while monitoring circulation of serotypes that remain prevalent despite inclusion in prior vaccine formulations.
An increasingly complex landscape for PCV evaluation will require new approaches to assessing efficacy and effectiveness. Future clinical trials could target enrollment of subjects with specific high-risk conditions in order to enrich for outcomes of interest, increasing the power of a study for a given sample size and thus enhancing feasibility. Regarding outcomes, it may be necessary to consider clinical endpoints considered beyond IPD. Cardiovascular events,^37^ functional/mobility limitation,^38^ cognitive impairment,^39^ and other sequelae have been associated with pneumonia and pneumococcal infections, and might be particularly germane for older adults.
Further investigation of immunoassays and implementation of novel diagnostic approaches can also facilitate ongoing efficacy and effectiveness studies. Determining serotype-specific antibody responses (particularly using OPA) that correspond with protection from infection in adults would be valuable to establish cutoffs as surrogates for adult clinical trials. Beyond its role in efficacy studies, expanding use of SSUAD for identification of noninvasive vaccine-type infections in pneumococcal surveillance could provide valuable insight into real-world effectiveness.