Streptococcus pneumoniae (pneumococcus) is the leading cause of bacterial pneumonia, otitis media, meningitis and sepsis, particularly affecting the elderly and young children. In 2000, it was estimated that pneumococcus was responsible for 14.5 million serious infections worldwide and over 800,000 deaths annually in children under 5 years of age.1
S. pneumoniae is comprised of at least 91 different serotypes (ST), which are defined by differences in surface polysaccharide capsules.2,3 Currently, several vaccines are licensed for the prevention of pneumococcal disease. These vaccines are either free pneumococcal polysaccharide [Pneumovax23 (PPV23)] or capsular pneumococcal polysaccharides (PnPs) conjugated to a carrier protein [SynflorixTM (PCV10) and Prevnar13 (PCV13)]. Widespread use of pneumococcal conjugate vaccines (PCVs) has been shown to be effective in reducing vaccine-type invasive pneumococcal disease (IPD) in both pediatrics and adults.4,5 However, there is evidence of increasing IPD due to pneumococcal ST not included in PCVs.6,7 To broaden protection against disease caused by emerging ST, an expanded 15-valent vaccine containing all 13 ST in Prevnar13 plus 2 additional ST (22F and 33F) was developed and is currently being evaluated in several pediatric and adult clinical trials.8–10
Following licensure of the first PCV (Prevnar; PCV7) in 2000, approval of investigational conjugate vaccines for pneumococcal disease is based on the demonstration of comparable safety and immunogenicity profiles to a licensed PCV. In infants, a WHO-accepted IgG threshold value of 0.35 µg/mL is the primary immunogenicity endpoint used for the comparison of the proportion of vaccine responders for shared ST between investigational and licensed PCVs.11 These anti-capsular polysaccharide antibodies protect the host by facilitating the opsonophagocytic killing of pneumococci. Opsonophagocytosis assay (OPA) titers have been shown to be a better surrogate marker of vaccine efficacy than IgG titers in some studies.12 During preclinical evaluation of new vaccine formulations, immunogenicity in animal models is frequently used to establish vaccine potency and screen for limited safety outcomes.13,14 Mouse models have been used to evaluate PCV15 immunization dose and route of immunization or compare the immunogenicity of different PCVs.15,16 The infant rhesus monkey model was used to study the immunogenicity of PCV15 and comparison to Prevnar13.17 However, there is no study directly comparing the immunogenicity of PCVs in animal models with that in human infants. Here, we report an immunogenicity evaluation of PCV15 in 3 animal models: IRM, NZWR and CD1 mice. The efficacy of PCV15 was also evaluated in a S. pneumoniae intraperitoneal (IP) challenge model in CD1 mice. Serotype-specific IgG antibody levels were evaluated using a pneumococcal electrochemiluminescent (Pn ECL) assay and functional antibody titers using a multiplexed opsonophagocytosis assay (MOPA) assay.18–20 To determine which animal model best predicted the immunogenicity of PCV15 in human infants, the correlation of the antibody titers between human infants and animal species was analyzed.
MATERIALS AND METHODS
Cell Lines and Reagents
HL-60 cells and RPMI 1640 medium were purchased from ATCC (Manassas, VA), fetal bovine sera from GE Healthcare Life Sciences (Logan, UT), Penicillin-Streptomycin (10,000 U/mL) and Glutamine (200 mM) from Invitrogen (Carlsbad, CA), Todd Hewitt yeast extract medium from Teknova (Hollister, CA), Sulfo-anti-rabbit IgG and MSD Sulfo-Tag NHS Ester kit from MSD (Rockville, MD), Goat anti-human IgG from Southern Biotech (Birmingham, AL) and Goat anti-mouse IgG from Jackson ImmunoResearch Labs (West Grove, PA). The anti-human IgG and anti-mouse IgG antibodies were labeled with a Sulfo tag using MSD protocols. N, N Dimethylformamide from EMD Millipore Corp. (Darmstadt, Germany).
Pneumococcal Conjugate Vaccine
Pneumococcal polysaccharide purification, conjugation, characterization and formulation were performed as described previously.15,21 Briefly, the investigational PCV was formulated in 20 mM histidine, 150 mM NaCl and 0.2% (w/v) polysorbate-20 using a mixture of 15 PnPs (1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, 22F, 23F and 33F) conjugated to cross-reactive material 197 (CRM197) and 250 µg/mL of aluminum phosphate adjuvant. Final vaccine material was filled into vials and stored at 2–8°C. A large volume of PCV15 was formulated and used across all preclinical studies reported herein to reduce the variability between animal models.
Animal Immunization and Challenge Study
All animal experiments were performed in strict accordance with the recommendations in the Guide for Care and Use of Laboratory Animals of the National Institutes of Health. The IRM immunization study was conducted at New Iberia Research Center (New Iberia, LA); the NZWR immunization study was conducted at Covance Inc. (Denver, PA); CD1 mouse immunization and challenge study was conducted at the Research Laboratories of Merck & Co., Inc., West Point, PA. All animal protocols were approved by the Institutional Animal Care and Use Committee at MRL and contract research labs.
Immunization and blood collection schedules of IRM, NZWR and CD1 mice are shown in Figure 1. Animal immunization schedules are considered a primary series. PCV15 vaccine containing 0.4 µg of polysaccharide for each serotype (0.8 µg for 6B) was administered intramuscularly into IRM, NZWR and mice (n = 17/group). Blood was collected in serum separator tubes (BD, Franklin Lakes, NJ). Sera were stored at −80°C until assayed.
For the mouse study, an additional group was immunized intramuscularly with aluminum phosphate adjuvant as a negative control and another group immunized intraperitoneally with 2 × 107 heat-killed S. pneumoniae ST 14 bacteria (HK-ST14) as a positive control for the ST14 challenge study. HK-ST14 bacteria were prepared by combining frozen culture aliquots and incubating in a 60°C water bath for 1 hour. The HK-ST14 preparation was confirmed to have no viable bacteria on blood agar plates.
Three weeks following the last immunization, mice were intraperitoneally challenged with ST14, stored frozen and diluted in Todd Hewitt broth on the day of challenge to a concentration of 1 × 106 colony forming unit (CFU) in a 0.5 mL volume. Inoculum concentration was confirmed by enumeration on blood agar plates. The mice were observed at least once daily for morbidity and mortality for 10 days. Moribund animals were euthanized. Log-rank (Mantel Cox) statistical test was used for analyses of survival curves.
PCV15 Safety and Immunogenicity Studies in Human Infants
PCV15 was evaluated in a Phase I/II randomized, double-blind, multicenter clinical study.22 Infants were given a 4-dose regimen of 4 different formulations of PCV15 at 2, 4, 6 and 12–15 months of age. A group of ~50 infants received a similar vaccine formulation as that used in the preclinical animal studies. Blood was collected ~30 days post-dose 3 (PD3), pre-dose 4, and 30 days post-dose 4 (PD4). Serotype-specific IgG geometric mean concentrations (GMCs) and OPA geometric mean titers (GMTs) measured in infants at 1 month PD3 (following the primary series) were compared with those measured in various animal species.
Electrochemiluminescent Assays and Data Analysis
Serotype-specific antibody responses were measured in 96-well ECL assays as described previously.15,18 IRM and human infant antibody titers were expressed as antibody concentrations (µg/mL) calculated from a standard curve using the serotype-specific IgG concentrations assigned to the human standard reference serum 007sp.23 Mouse and rabbit antibody levels were expressed as titers calculated as the reciprocal of the linearly interpolated dilution (1/dilution) corresponding to the cutoff value determined using positive control sera.
MOPA and Data Analysis
Pneumococcal MOPA were performed as described previously and can be found at www.vaccine.uab.edu.15,19 The OPA titers were calculated as the reciprocal of the serum dilution (1/dilution) with 50% killing compared with the average growth in the complement control (no serum control) wells using the Opsotiter 3 software owned by and licensed from UAB Research Foundation.
Spearman Correlation Analysis of the Immunogenicity Data
When evaluating the correlation between the functional OPA titers and binding IgG titers for each animal model and human infants, OPA GMTs were paired with IgG GMTs for each serotype. For each model, the paired GMTs of each serotype were used to assess the correlation (Spearman exact) and the P value (Monte Carlo exact) of no relationship between 2 assays using Stat-xact 4.0.1 software.
When comparing the IgG GMTs of each serotype between animal model and human infants, IgG GMTs were paired for each of the 15 ST for each comparison (IRM vs. human infants, NZWR vs. human infants, mice vs. human infants). The paired IgG GMTs of the 15 ST were used to assess the correlation (Spearman exact) and the P value (Monte Carlo exact) of no relationship using Stat-xact 4.0.1 software. Same correlation analysis was done for OPA GMTs of 15 ST for all 3 comparisons. Mouse serotype 7F OPA data were excluded for the analysis because the anti-7F MOPA assay of mouse sera failed the assay acceptance criteria.
Measurements of Serotype-specific IgG Titers in Sera From IRM, NZWR and CD1 Mice Immunized With PCV15
Anti-PnPs IgG titers of sera from all animal models were analyzed using multiplexed Pn ECL assay. The serotype-specific IgG GMTs were calculated and compared between time points if sera were analyzed individually.
Anti-PnPs IgG concentrations in IRM sera were calculated from a standard curve of human standard reference serum 007sp. Anti-PnPs IgG GMTs at PD1 increased from 1.2-fold (ST23F) to 48-fold (ST3) compared with pre-immune sera (Fig. 2A). The IgG concentrations further increased after administration of a second dose of PCV15 for all vaccine ST. At PD3, anti-PnPs IgG GMTs increased from 14-fold (ST6A) to 138-fold (ST4) compared with pre-immune sera. Anti-PnPs IgG GMCs were greater than the WHO-accepted threshold value of 0.35 µg/mL for all ST after 3 doses of PCV15.
Similar to IRM responses, anti-PnPs IgG in NZWR sera were induced after 1 dose of PCV15 for all ST. Anti-PnPs IgG GMTs increased from 3-fold (ST5) to 166-fold (ST18C) compared with pre-immune sera (Fig. 2B). After 2 doses of PCV15, anti-PnPs IgG GMTs increased from 10-fold (ST3) to 495-fold (ST19F) compared with pre-immune sera. Due to blood collection timing and volume, only pre-immune and PD3 (Day 35) sera were collected in mice. Anti-PnPs IgG titers in mouse sera increased significantly for all ST after 3 doses of PCV15 (Fig. 2C). The anti-PnPs IgG GMTs increased from 12-fold (ST33F) and 4000-fold (ST4) compared with pre-immune sera. These data demonstrated that PCV15 was immunogenic in all 3 animal species.
Measurements of Serotype-specific OPA Titers in Sera From IRM, NZWR and CD1 Mice Immunized With PCV15
Functional antibody titers of individual serum in each animal model after the final vaccine dose were also determined using MOPA (Fig. 3). Similar to IgG responses, OPA titers to the S. pneumoniae ST increased significantly compared with the pre-immune sera. OPA GMTs of day 70 (PD3) sera in IRM increased from 7-fold (ST3) to 550-fold (ST9V) compared with pre-immune sera (Fig. 3A). For NZWR sera, OPA GMTs of day 28 (PD2) sera increased from 24-fold (ST23F) to 1400-fold (ST6A) compared with pre-immune sera (Fig. 3B). OPA GMTs of day 35 (PD3) mouse sera increased from 1.4-fold (ST33F) to 660-fold (ST6A) compared with pre-immune sera. These data showed that each animal species generated functional antibody titers in response to PCV15, except for 7F in CD1 mice only. Serotype 7F OPA data from mice sera were excluded from this analysis because the 7F MOPA assay failed the acceptance criteria, which has also been reported previously.15
Protection of PCV15 Immunized CD1 Mice From ST14 Challenge
To evaluate PCV15 efficacy, CD1 mice were challenged with 1 × 106 CFU of S. pneumoniae ST14 at 3 weeks following the final immunization. More mice vaccinated with PCV15 were protected from challenge (P < 0.001) when compared with mice in the adjuvant control group (Fig. 4). More mice in the positive control HK-ST14 group were also protected from challenge (P < 0.001) relative to the adjuvant control. At the end of the challenge study, the PCV15 immunized group and the HK-ST14 immunized group had 100% and 94% survival rates, respectively. These data confirmed that the functional antibodies induced by PCV15 immunization can protect mice from IP challenge.
Correlation Analysis of Serotype-specific Antibody Level Measured in Human Infants and Those Measured in IRM, NZWR and CD1 Mice
We first analyzed the correlation between IgG and OPA GMTs measured in human infants and those measured in each animal model. There was good correlation between IgG and OPA GMTs in human infants (rs=0.57, P=0.03, Fig. 5A) and in NZWR (rs=0.60, P=0.02, Fig. 5C). However, there was no correlation between IgG and OPA GMTs in IRM (rs=0.09, P=0.75, Fig. 5B) and in mice (rs=0.37, P=0.20, Fig. 5D).
To determine which animal model best predicts the immunogenicity of PCV15 in human infants, the serotype-specific IgG or OPA GMTs of IRM, NZWR, and mice were compared with that of human infants immunized with PCV15. IgG GMTs did not correlate between human infants and any of the animal models (Fig. 6A,C,E). However, there was a good correlation between human infants and IRM for the OPA GMTs (rs=0.69, P=0.006, Fig. 6B). A medium correlation was observed for OPA GMTs between human infants and NZWR (rs=0.49, P=0.06, Fig. 6D), but no correlation in OPA GMTs was observed between human infants and mice (rs=0.04, P=0.89, Fig. 6F). These data indicate that OPA titers maybe a better surrogate biomarker than IgG titers in evaluating the correlation of vaccine-induced antibody responses between animal models and human infants.
Preclinical animal models are often used to evaluate the immunogenicity and safety of experimental pneumococcal vaccine formulations due to their utility in providing preliminary and meaningful indication about the performance of the vaccine.13–15,17,24 However, significant differences in immunogenicity exist among animal species and between animal models and humans.15 An animal model’s predictive value for vaccine efficacy depends on the robustness of the animal model and an accepted immune correlate of protection. For example, animal protection after vaccination with heterologous prototype virus-like particles predicted successful efficacy in clinical trials for human papillomavirus vaccine. However, the establishment of animal models’ predictive value of vaccine effectiveness in humans has a low success rate to date.25–27 Here, we compared the immunogenicity of an investigational next generation PCV comprised of 15 PnPs ST in 3 preclinical animal models. These preclinical data were then compared with Phase II human infant data to determine possible correlation in vaccine-induced serotype-specific antibody responses. To reduce the variability between studies, similar vaccine formulations were used across human and animal studies. Consistent with human data, PCV15 was found to be immunogenic to all 15 ST included in the vaccine in all animal models that were evaluated.22
In this study, IRMs vaccinated with PCV15 had levels of IgG GMCs >0.35 µg/mL after 1 dose for 4 of the 15 ST (1, 3, 4, 22F). This number increases to 13 of 15 ST after 2 doses. ST 23F and 33F required 3 immunizations to obtain this level of antibody. The boosting effect of PCV15 is consistent with our previous data,17 although the antibody concentration is lower after each dose in this study, which may be due to a shorter dosing interval (28 vs. 56 days interval). We have seen trends of lower IgG responses to 6A, 6B, 23F and 33F in IRM and mice compared with NZWR in these studies. The lower IgG responses to 23F in mice may be due to the lack of appropriate variable genes encoding 23F-specific antibody or 23F antigens are rapidly cleared by a mechanism involving recognition of 23F PnPs by lectin-like molecules.24,28 The mechanism of lower IgG response to 6A, 6B and 33F in IRM and mice is not well understood. We also observed significant variation in immunogenicity between ST for each animal species. The differences of immunogenicity between ST may be due to the modification of Th1/Th2 response to the carrier protein by different polysaccharides, which in turn affects the immune response to the polysaccharides.29 Variation of immunogenicity between ST and between animal models were also observed in the OPA responses. For example, compared with pre-immune sera, ST 3, 23F and 33F had the lowest increase in OPA titers in IRM, NZWR and mice, respectively.
Experimental pneumococcal disease has been investigated by use of different rodents and the rabbit, but the mouse represents the most commonly used animal model.14S. pneumoniae mouse challenge models were developed in our lab to study the protective effect of PCV15 immunization. Mouse challenge models are variable depending on the route of challenge, S. pneumoniae serotype and mouse strain. The challenge dose should be optimized for each S. pneumoniae serotype and each mouse strain. We have reported previously that PCV15 can protect Balb/c mice from 6B IP challenge.17 In this study, we developed a serotype 14 IP challenge model in CD1 mice and showed that PCV15 can protect CD1 mice from ST14 challenge. A previous study showed that serotype 14 was avirulent in Balb/c mice when injected IP with 106 CFU.30 The difference may be due to the different mouse strain (CD1) or serotype 14 isolate used in this study. We have also observed that different mouse strains and the age of the mice may require a different challenge inoculum concentration to be successful. One limitation of this challenge model is that IP infection is not the natural route of human pneumococcal infection. Additional mouse challenge models are being developed to evaluate the efficacy of PCV15.
The demonstration of a good correlation between serotype-specific IgG and OPA GMTs and which immune parameter (IgG or OPA) is a better surrogate of vaccine efficacy have been of great interest to the scientific community. Strong correlations between IgG and OPA have been observed in some studies of young children.31 However, poor correlations have been found in some clinical situations and populations.32 We have found better correlations between IgG and OPA in human infant and NZWR sera than in IRM and mice sera. The poor correlation between IgG and OPA may due to some binding antibodies which have no opsonophagocytic killing activity. This has been observed in elderly humans who often have high concentrations of nonfunctional anti-capsular polysaccharide antibodies.12 The discrepancy may also be due to the observation that PCVs induce anti-PnPs IgA and IgM antibodies which have OPA activity but are not detected in the Pn ECL assay.33 The target population for PCV15 is human infants and adults. In this study, we sought to identify a good correlation between preclinical animal models and human infants for IgG or OPA measured following immunization with PCV15, which could serve for preclinical evaluations of future PCVs and/or other non-PCVs. For the first time, we show that OPA in human infant sera correlate very well with that in infant rhesus macaque sera, followed by that in NZWR sera, but did not correlate with OPA measured in CD1 mouse sera. This may be due to the IRM’ closer genetic relationship to human. However, precaution must be taken when trying to extrapolate these observations to other PCV vaccines. The correlation analysis results may vary depending on, but not limited to, the ST included in the vaccines, the formulation of the vaccines, or whether the ratio of titers to a control group was used.
In summary, our data show that PCV15 is immunogenic in IRM, New Zealand white rabbits and CD1 mice; and infant rhesus macaque is the preferred animal model to predict the immunogenicity of PCV15 in human infants.
We would like to thank Covance Inc. (Denver, PA), New Iberia Research Center (New Iberia, LA), and MRL West Point Laboratory Animal Resources for the conducting the animal studies. We also thank Alison Pedley and Jianing Li for providing the ECL and OPA data from PCV15 Phase II clinical trials.
1. O’Brien KL, Wolfson LJ, Watt JP, et al; Hib and Pneumococcal Global Burden of Disease Study Team. Burden of disease caused by Streptococcus pneumoniae
in children younger than 5 years: global estimates. Lancet. 2009;374:893–902.
2. Geno KA, Gilbert GL, Song JY, et al. Pneumococcal capsules and their types: past, present, and future. Clin Microbiol Rev. 2018;28:29.
3. Park IH, Pritchard DG, Cartee R, et al. Discovery of a new capsular serotype (6C) within serogroup 6 of Streptococcus pneumoniae
. J Clin Microbiol. 2007;45:1225–1233.
4. Andrews NJ, Waight PA, Burbidge P, et al. Serotype-specific effectiveness and correlates of protection for the 13-valent pneumococcal conjugate vaccine
: a postlicensure indirect cohort study. Lancet Infect Dis. 2014;14:839–846.
5. Moore MR, Link-Gelles R, Schaffner W, et al. Effect of use of 13-valent pneumococcal conjugate vaccine
in children on invasive pneumococcal disease in children and adults in the USA: analysis of multisite, population-based surveillance. Lancet Infect Dis. 2015;15:301–309.
6. Waight PA, Andrews NJ, Ladhani SN, et al. Effect of the 13-valent pneumococcal conjugate vaccine
on invasive pneumococcal disease in England and Wales 4 years after its introduction: an observational cohort study. Lancet Infect Dis. 2015;15:535–543.
7. Pilishvili T, Lexau C, Farley MM, et al; Active Bacterial Core Surveillance/Emerging Infections Program Network. Sustained reductions in invasive pneumococcal disease in the era of conjugate vaccine. J Infect Dis. 2010;201:32–41.
8. Peterson JT, Stacey HL, MacNair JE, et al. Safety and immunogenicity
of 15-valent pneumococcal conjugate vaccine
compared to 13-valent pneumococcal conjugate vaccine
in adults >/=65 years of age previously vaccinated with 23-valent pneumococcal polysaccharide vaccine. Hum Vaccin Immunother. 2019;15:540–548.
9. Greenberg D, Hoover PA, Vesikari T, et al. Safety and immunogenicity
of 15-valent pneumococcal conjugate vaccine
(PCV15) in healthy infants. Vaccine. 2018;36:6883–6891.
10. Ermlich SJ, Andrews CP, Folkerth S, et al. Safety and immunogenicity
of 15-valent pneumococcal conjugate vaccine
in pneumococcal vaccine-naïve adults ≥50 years of age. Vaccine. 2018;36:6875–6882.
11. WHO. Recommendations to assure the quality, safety and efficacy of Pneumococcal conjugate vaccines. Replacement of TRS 927 2009. Available at: http://www.who.int/biologicals/publications/trs/areas/vaccines/pneumo/en/
12. Romero-Steiner S, Frasch CE, Carlone G, et al. Use of opsonophagocytosis for serological evaluation of pneumococcal vaccines. Clin Vaccine Immunol. 2006;13:165–169.
13. Steinhoff MC; Pneumococcal Vaccine Animal Model Consensus Group. Animal models
for protein pneumococcal vaccine evaluation: a summary. Vaccine. 2007;25:2465–2470.
14. Chiavolini D, Pozzi G, Ricci S. Animal models
of Streptococcus pneumoniae
disease. Clin Microbiol Rev2008;21:666–685.
15. Caro-Aguilar I, Indrawati L, Kaufhold RM, et al. Immunogenicity
differences of a 15-valent pneumococcal polysaccharide conjugate vaccine (PCV15) based on vaccine dose, route of immunization and mouse strain. Vaccine. 2017;35:865–872.
16. Park C, Kwon EY, Choi SM, et al. Comparative evaluation of a newly developed 13-valent pneumococcal conjugate vaccine
in a mouse model. Hum Vaccin Immunother. 2017;13:1169–1176.
17. Skinner JM, Indrawati L, Cannon J, et al. Pre-clinical evaluation of a 15-valent pneumococcal conjugate vaccine
(PCV15-CRM197) in an infant-rhesus monkey immunogenicity
model. Vaccine. 2011;29:8870–8876.
18. Marchese RD, Puchalski D, Miller P, et al. Optimization and validation of a multiplex, electrochemiluminescence-based detection assay for the quantitation of immunoglobulin G serotype-specific antipneumococcal antibodies in human serum. Clin Vaccine Immunol. 2009;16:387–396.
19. Burton RL, Nahm MH. Development and validation of a fourfold multiplexed opsonization assay (MOPA4) for pneumococcal antibodies. Clin Vaccine Immunol. 2006;13:1004–1009.
20. Burton RL, Nahm MH. Development of a fourfold multiplexed opsonophagocytosis assay
for pneumococcal antibodies against additional serotypes and discovery of serological subtypes in Streptococcus pneumoniae
serotype 20. Clin Vaccine Immunol. 2012;19:835–841.
21. He J, MacNair JE, Smith WJ, et al. Enhancing Immunogenicity
of Streptococcus Pneumoniae Polysaccharide-Protein Conjugates. 2018:USA; Merck Sharp & Dohme Corp: 65.
22. Rupp R, Hurley D, Grayson S, et al. A dose ranging study of 2 different formulations of 15-valent pneumococcal conjugate vaccine
(PCV15) in healthy infants. Hum Vaccin Immunother. 2019;15:549–559.
23. Goldblatt D, Plikaytis BD, Akkoyunlu M, et al. Establishment of a new human pneumococcal standard reference serum, 007sp. Clin Vaccine Immunol. 2011;18:1728–1736.
24. Alonso de Velasco E, Verheul AF, van Steijn AM, et al. Epitope specificity of rabbit immunoglobulin G (IgG) elicited by pneumococcal type 23F synthetic oligosaccharide- and native polysaccharide-protein conjugate vaccines: comparison with human anti-polysaccharide 23F IgG. Infection and immunity. 1994;62:799–808.
25. Golding H, Khurana S, Zaitseva M. What is the predictive value of animal models
for vaccine efficacy in humans? The importance of Bridging Studies and species-independent correlates of protection. Cold Spring Harb Perspect Biol. 2018;10;a028902.
26. Herati RS, Wherry EJ. What is the predictive value of animal models
for vaccine efficacy in humans? Consideration of atrategies to improve the value of animal models
. Cold Spring Harb Perspect Biol. 2018;10;a031583.
27. Jameson SC, Masopust D. What is the predictive value of animal models
for vaccine efficacy in humans? Reevaluating the potential of mouse models for the human immune system. Cold Spring Harb Perspect Biol. 2018;10;a029132.
28. McCool TL, Harding CV, Greenspan NS, et al. B- and T-cell immune responses to pneumococcal conjugate vaccines: divergence between carrier- and polysaccharide-specific immunogenicity
. Infect Immun. 1999;67:4862–4869.
29. Mawas F, Feavers IM, Corbel MJ. Serotype of Streptococcus pneumoniae
capsular polysaccharide can modify the Th1/Th2 cytokine profile and IgG subclass response to pneumococal-CRM(197) conjugate vaccines in a murine model. Vaccine. 2000;19:1159–1166.
30. Briles DE, Crain MJ, Gray BM, et al. Strong association between capsular type and virulence for mice among human isolates of Streptococcus pneumoniae
. Infect Immun. 1992;60:111–116.
31. Nahm MHR-SS. Functional Asssys for Pneumococcal Antibody. 2008.Washington, DC: ASM Press.
32. Song JY, Moseley MA, Burton RL, et al. Pneumococcal vaccine and opsonic pneumococcal antibody. J Infect Chemother. 2013;19:412–425.
33. Clutterbuck EA, Salt P, Oh S, et al. The kinetics and phenotype of the human B-cell response following immunization with a heptavalent pneumococcal-CRM conjugate vaccine. Immunology. 2006;119:328–337.