Vaccine xxx (xxxx) xxx

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

Efficacy, safety and immunogenicity of a pneumococcal protein-based

vaccine co-administered with 13-valent pneumococcal conjugate
vaccine against acute otitis media in young children: A phase IIb
randomized study
Laura L. Hammitt a,⇑, James C. Campbell a, Dorota Borys b, Robert C. Weatherholtz a, Raymond Reid a,
Novalene Goklish a, Lawrence H. Moulton a, Magali Traskine b, Yue Song c, Kristien Swinnen b,
Mathuram Santosham a, Katherine L. O’Brien a
a
Center for American Indian Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States
b
GSK, Wavre, Belgium
c
XPE Pharma & Science c/o GSK, Wavre, Belgium

a r t i c l e i n f o a b s t r a c t

Article history: Background: Native American populations experience a substantial burden of pneumococcal disease
Received 11 June 2019 despite use of highly effective pneumococcal conjugate vaccines (PCVs). Protein-based pneumococcal
Received in revised form 18 September vaccines may extend protection beyond the serotype-specific protection elicited by PCVs.
2019
Methods: In this phase IIb, double-blind, controlled trial, 6–12 weeks-old Native American infants ran-
Accepted 20 September 2019
Available online xxxx
domized 1:1, received either a protein-based pneumococcal vaccine (dPly/PhtD) containing pneumolysin
toxoid (dPly, 10 mg) and pneumococcal histidine triad protein D (PhtD, 10 mg) or placebo, administered
along with 13-valent PCV (PCV13) at ages 2, 4, 6 and 12–15 months. Other pediatric vaccines were given
Keywords:
Streptococcus pneumoniae
per the routine immunization schedule. We assessed vaccine efficacy (VE) against acute otitis media
Immunogenicity (AOM) and acute lower respiratory tract infection (ALRI) endpoints. Immunogenicity, reactogenicity
Reactogenicity and unsolicited adverse events were assessed in a sub-cohort and serious adverse events were assessed
Acute otitis media in all children.
Vaccines Results: 1803 infants were randomized (900 dPly/PhtD; 903 Control). VE against all episodes of American
Native American Academy of Pediatrics (AAP)-defined AOM was 3.8% (95% confidence interval:
11.4, 16.9). Point esti-
mates of VE against other AOM outcomes ranged between 2.9% (
9.5, 14.0) and 5.2% (
8.0, 16.8).
Point estimates of VE against ALRI outcomes ranged between
4.4% (
39.2, 21.8) and 2.0% (
18.3,
18.8). Point estimates of VE tended to be higher against first than all episodes but the confidence intervals
included zero. dPly/PhtD vaccine was immunogenic and had an acceptable reactogenicity and safety pro-
file after primary and booster vaccination in Native American infants.
Conclusions: The dPly/PhtD vaccine was immunogenic and well tolerated, however, incremental efficacy
in preventing AAP-AOM over PCV13 was not demonstrated.
Clinical trials registration: NCT01545375 (www.clinicaltrials.gov)
Ó 2019 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://
creativecommons.org/licenses/by/4.0/).

Abbreviations: AAP, American Academy of Pediatrics; ABS, active bacterial surveillance; AE, adverse event; ALRI, acute lower respiratory tract infection; AOM, acute otitis
media; ATP, according-to-protocol; CI, confidence interval; dPly, pneumolysin toxoid; ELISA, enzyme linked immunosorbent assay; EL.U, ELISA units; GMC, geometric mean
concentration; HCP, health-care provider; IHS, Indian Health Service; IPD, invasive pneumococcal disease; LLOQ, lower limit of quantification; MA, medically-attended;
mATP, modified according-to-protocol; PCV, pneumococcal conjugate vaccine; PCV13, 13-valent pneumococcal conjugate vaccine; PHiD-CV, pneumococcal non-typeable
Haemophilus influenzae protein D conjugate vaccine; PhtD, pneumococcal histidine triad protein D; Ply, pneumolysin; SAE, serious adverse event; SAS, Statistical Analysis
System; TVC, total vaccinated cohort; ULOQ, upper limit of quantification; VE, vaccine efficacy.
⇑ Corresponding author at: Johns Hopkins Bloomberg School of Public Health, 415 N. Washington St., 4th Floor, Baltimore, MD 21231, United States.
E-mail address: [email protected] (L.L. Hammitt).

https://doi.org/10.1016/j.vaccine.2019.09.076
0264-410X/Ó 2019 The Authors. Published by Elsevier Ltd.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
2 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx

1. Introduction immunogenicity of dPly/PhtD antigens and safety/reactogenicity

of the dPly/PhtD vaccine.
Routine use of licensed pneumococcal conjugate vaccines
(PCVs) has led to a substantial reduction in invasive pneumococcal 2. Material and methods
disease (IPD) among vaccinated and unvaccinated individuals [1].
However, currently available vaccines protect against only a frac- We conducted this study between May 2012 and July 2016 at
tion of the more than 90 pneumococcal serotypes. Streptococcus five sites on the Navajo Nation and White Mountain Apache tribal
pneumoniae remains an important cause of IPD, pneumonia, and lands in the southwest United States. Health care is administered
acute otitis media (AOM). In 2015, there were an estimated 3.7 free of charge to tribal members through the Indian Health Service
million episodes of severe pneumococcal disease globally, in a set- (IHS). The institutional review boards of the Johns Hopkins Bloom-
ting where 129 countries had included PCV as part of the infant berg School of Public Health, the Navajo Nation, the Phoenix Area
vaccination programs [1,2]. Prior to the use of PCVs, the rate of Indian Health Service, and the Tribal Council of the White Moun-
IPD among Native Americans was one of the highest in the world tain Apache Tribe approved the study. The study was conducted
[3–6]. While use of PCVs has greatly reduced the burden of IPD, in accordance with Good Clinical Practice guidelines and the prin-
disparities persist. Age-group specific IPD incidence among Native ciples of the Declaration of Helsinki. Written informed consent was
Americans in the southwest United States remains 2–4-fold higher obtained from a parent or legally-authorized representative of each
than the general US population; the majority of this residual pneu- infant before enrollment. An independent data monitoring com-
mococcal disease is caused by serotypes not covered by currently mittee was charged with oversight of participant safety. This study
available pneumococcal vaccines [7–9]. was registered at www.clinicaltrials.gov (NCT01545375). A proto-
Protein antigens that are conserved across pneumococcal sero- col summary is available at www.gsk-studyregister.com (study ID:
types may offer an alternative to serotype-restricted vaccination 115597).
strategies [10]. Two such proteins are pneumolysin (Ply) – a cyto- Eligible infants aged 6–12 weeks were randomized 1:1 to
lytic protein that contributes to the host inflammatory response, receive dPly/PhtD vaccine or placebo (Control group) at ages 2, 4,
facilitates colonization and plays a role in acute lung injury – and 6 and 12–15 months, each co-administered with PCV13 (Fig. 1).
pneumococcal histidine triad protein D (PhtD), which has a role Inclusion and exclusion criteria are listed in Supplementary Text.
in ion homeostasis and pneumococcal virulence [11–13]. Alone A subset of children was enrolled into the immunogenicity/reacto-
or in combination Ply and PhtD have demonstrated protection genicity sub-cohort. Randomization was stratified by site with a
against pneumococcal disease or carriage in in vitro and animal block size of six and treatment allocation at the investigator site
studies and were selected by GSK for clinical development for dPly/PhtD was performed using a central randomization system
[12,14–18]. Vaccines containing dPly (a pneumolysin toxoid), PhtD on the internet. All study participants, investigators, funding staff,
and 10 serotype-specific polysaccharide conjugates of the pneu- and monitoring staff were masked to drug allocation. dPly/PhtD
mococcal non-typeable Haemophilus influenzae protein D conjugate vaccine and placebo were supplied to the site in identical-looking
vaccine (PHiD-CV, Synflorix, GSK) were found to be immunogenic and -labelled vials in coded kits and stored at 2–8 °C until the time
and well-tolerated in children and infants; however, efficacy of dosing. The investigational pneumococcal vaccine (dPly/PhtD)
against pneumococcal nasopharyngeal carriage prevalence, den- contained dPly and PhtD at 10 mg each, adsorbed on aluminum
sity, acquisition or clearance was not observed [19–22]. We report phosphate (vaccine adjuvant; aluminum content 500 mg). The pla-
results of a phase IIb randomized controlled trial that assessed the cebo contained 500 mg aluminum phosphate. The placebo was
incremental efficacy (over 13-valent PCV [PCV13], Prevenar 13/Pre- visually indistinguishable from the dPly/PhtD vaccine. Study pro-
vnar 13, Pfizer) of an investigational pneumococcal protein-based duct (vaccine or placebo) was administered intramuscularly into
vaccine, dPly/PhtD, against AOM and acute lower respiratory tract the right thigh (primary vaccination) or deltoid (booster vaccina-
infection (ALRI) in Native American infants. We also report tion). Co-administered PCV13 was given intramuscularly on the

Invesgaonal arm (dPly/PhtD + PCV13)

Randomizaon
1:1
Control arm (AlPO4 Placebo + PCV13)

Age (months) 2 4 6 7 12–15 13–16 18–22 24–27

Study vaccinaon

Total Vaccinated Efficacy follow-up (ATP)

Cohort (TVC) AOM/ALRI Efficacy follow-up (TVC)
AE reporng SAE reporng

Blood sampling
Immunogenicity/
reactogenicity Solicited AE reporng, Day 0 – Day 3
Sub-cohort
AE reporng
Unsolicited AE reporng, Day 0 – Day 30

Fig. 1. Study design. dPly/PhtD, pneumolysin toxoid/pneumococcal histidine-triad protein D; PCV13, 13-valent pneumococcal conjugate vaccine; AOM, acute otitis media;
ALRI, acute lower respiratory tract infection; ATP, according-to-protocol; AE, adverse event; SAE, serious adverse event.
Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx 3

opposite side. All other vaccines were given according to the rec- 2.2. Study procedures
ommended childhood immunization schedule in the United States,
including influenza vaccine for children 6 months old which was Active surveillance for any medical event was conducted by
allowed during the influenza seasons, as per Advisory Committee study staff at study facilities and other facilities providing care in
on Immunization Practices recommendations [23]. the study communities starting from administration of the first
dose through study end for each participant to identify SAEs and
2.1. Study endpoints episodes of AOM or ALRI. Immunogenicity and unsolicited adverse
events (AEs) occurring within one month following administration
We assessed dPly/PhtD vaccine efficacy (VE) against all epi- of each dose of study vaccine were assessed in the immunogenic-
sodes of clinically diagnosed AOM meeting the 2004 American ity/reactogenicity sub-cohort (N = 400). Solicited local and general
Academy of Pediatrics criteria (AAP-AOM [2004]; primary objec- AEs were documented on a diary card during 4 days post-
tive) and other AOM and ALRI endpoints (Table 1) [24]. In addition, vaccination by the parent/guardian of each participant in the
we report on the following secondary objectives: the safety of the immunogenicity/reactogenicity sub-cohort. Participants in this
dPly/PhtD vaccine in all participants in terms of serious adverse sub-cohort also had blood collected at five timepoints (prior to
events (SAEs) and the immunogenicity and reactogenicity of the dose 1, one month post-dose 3, prior to the booster vaccination,
dPly/PhtD vaccine in a sub-cohort of participants. Results of other one month and 12 months post-booster vaccination). Samples
objectives (i.e., immunogenicity of co-administered pneumococcal were centrifuged and serum was stored at
20 °C or colder until
and Hib vaccines, and bacterial nasopharyngeal carriage) will be assayed for anti-Ply and anti-PhtD antibodies by GSK using an in-
described elsewhere. house enzyme linked immunosorbent assay (ELISA) [25].

Table 1
Case definitions for efficacy endpoints.

AAP, American Academy of Pediatrics; MA, medically-attended; ALRI, acute lower respiratory tract infection; HCP, health-care provider.

Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
4 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx

2.3. Sample size and statistical analysis VE was estimated using the generalized Cox regression model,
by using time-to-event; the point estimates were not adjusted
The primary objective was to demonstrate that the VE induced for multiplicity.
by the dPly/PhtD vaccine against clinical AOM diagnosed and ver- The statistical analyses were performed using the Statistical
ified using AAP criteria was greater than 0%, as compared to the Analysis System (SAS) in the SAS Drug Development environment.
Control group. Sample size calculations were based on the assump-
tion that, among AOM, approximately 60% is bacterial, of which
40% is pneumococcal [26–28]. True VE against pneumococcal 3. Results
AOM was estimated to be similar to the observed PCV VE against
vaccine-type pneumococcal AOM (approximately 50–60%) Of the 1803 infants in the TVC, 900 were in the Investigational
[26,29–31]. Based on 1800 enrolled participants (i.e., 1537 pro- group (dPly/PhtD + PCV13) and 903 in the Control group (Placebo +
jected evaluable participants), the study had 97.2% power to PCV13) (Fig. 2). 808 infants in the Investigational group and 831 in
demonstrate AOM VE >0% based on a true VE of 17%, an incidence the Control group were included in the mATP cohort for efficacy.
rate in the control group of 0.6 episodes per child-year, an over- Demographic characteristics and duration of follow-up were simi-
dispersion of 1.4 and a one-sided test, nominal type I error of lar in the two groups (Supplemental Tables 1 and 2). In the mATP
17.8%. The interim analysis is presented in the Supplementary cohort for efficacy, 485 AAP-AOM events were detected in the dPly/
Text. PhtD group and 518 in the Control group, giving an incidence of
The primary efficacy analysis was based on (i) the occurrence of 0.43 and 0.44 episodes/child-year, respectively (Table 2); the pro-
the primary endpoint (i.e., AAP-AOM 2004) anytime from two portion of participants experiencing 3 or more episodes of AAP-
weeks after the administration of the third primary dose of the AOM was 5.1% in the dPly/PhtD group and 4.8% in the Control
study vaccine up to the final study visit or the time of censoring group. 5.7% and 5.9% of children experienced recurrent AOM diag-
(e.g., for participants with non-compliance with the protocol) or nosed by a health-care provider (HCP) in the dPly/PhtD and Control
up to the time of withdrawal (e.g., if a participant withdrew before groups, respectively (Supplemental Table 3). For both groups, the
completing the study) and (ii) the modified according-to-protocol peak incidence of AOM occurred at 6–12 months of life. There were
(mATP) cohort for efficacy analysis which included all participants 163 medically-attended (MA)-ALRI events in the dPly/PhtD group
from the ATP cohort for efficacy analysis and participants for and 165 in the Control group, giving an incidence of 0.14
whom non-compliance with the vaccination intervals during pri- episodes/child-year in each group (Table 3).
mary vaccination constituted the only elimination criterion from
ATP cohort for efficacy analysis. The other cohorts analyzed are
described in Supplementary Text. 3.1. Vaccine efficacy
Time-to-occurrence of primary endpoint events during the
defined efficacy follow-up period was compared between groups Incremental efficacy of dPly/PhtD in preventing AAP-AOM
by estimating VE and its 95% confidence interval (CI) using the (2004) over PCV13 was not demonstrated (Table 2). In the mATP
Anderson & Gill model, with a robust sandwich estimator for vari- cohort for VE, VE against all AAP-AOM episodes was 3.8% (95%
ance matrix considered as a generalization of the Cox proportional CI:
11.4, 16.9). VE against the first AAP-AOM episode was 11.3%
hazard model, taking into account recurrent events [32]. VE was (
3.4, 23.9). VE against all episodes of draining AOM, pneumococ-
defined as (1
hazard ratio)  100%. cal draining AOM, and non-pneumococcal draining AOM ranged
Antibody geometric mean concentrations (GMCs) and percent- from –32.2% (
298.3, 56.1) to 17.8% (
57.0, 57.0) (Supplemental
age of infants with antibody concentration above pre-specified Table 4). VE against all episodes of MA-ALRI, MA-ALRI with fever,
cut-offs (12 ELISA units[EL.U]/mL for Ply and 17 EL.U/mL for PhtD) and MA-HCP-ALRI with fever ranged between
4.4% (
39.2,
were assessed. For the purpose of GMC calculation, antibody con- 21.8) and 2.0% (
18.3, 18.8) (Table 3). VE against first episodes of
centrations below the lower limit of quantification (LLOQ) of the clinical AOM or ALRI tended to be higher compared to VE against
assay were given an arbitrary value of half the LLOQ and those all episodes (Tables 2 and 3). This finding was more pronounced
above the upper limit of quantification (ULOQ) were assigned a in a post-hoc analysis restricted to first episodes within the first
value equivalent to the ULOQ. year of life (Fig. 3; Supplemental Figs. 2 and 3; Supplemental
Because acquired maternal antibodies may inhibit the infant Table 5).
immune response to primary immunization, we assessed as Exploratory analyses of VE against various AOM and ALRI end-
exploratory analyses the relationship between pre- and post- points among participants with low (27498 EL.U/mL for Ply;
immunization anti-Ply and anti-PhtD antibody levels using scatter 3739.5 EL.U/mL for PhtD) and high (>27498 EL.U/mL for Ply;
plots [33]. Further, the impact of post-primary series anti-Ply and >3739.5 EL.U/mL for PhtD) post-primary anti-protein antibody
anti-PhtD antibody levels on VE was assessed by dividing partici- concentrations revealed a trend for higher efficacy against AOM
pants into ‘‘low” and ‘‘high” categories based on the median anti- endpoints with higher post-primary anti-Ply antibody levels. For
body concentration. ALRI endpoints, VE seemed to be higher with lower post-primary
Influenza disease and vaccination may impact the incidence of anti-Ply antibody levels but VE point estimates remained positive
AOM and ALRI, and may consequently bias estimates of VE against for both low and high antibody ranges (Supplemental Table 6).
these diseases [34,35]. Therefore, we also assessed the effect of VE against AOM and ALRI endpoints appeared in the same range
seasonality and influenza vaccination in a post-hoc exploratory for low and high post-primary anti-PhtD antibody levels (Supple-
analysis. The influenza season was defined from November 30 to mental Table 7).
May 31 and, in any given season, VE against AOM and ALRI end- AOM incidence followed a seasonal pattern with peaks in win-
points was assessed in the total vaccinated cohort (TVC) and mATP ter seasons and tended to be lower in the last season of the study,
cohort for efficacy. For each participant, VE follow-up duration for reflecting the older age of the study population (Supplemental
a given season was defined as shown in Supplemental Fig. 1. VE Fig. 4). No consistent impact of seasonal influenza vaccination
was computed within each subgroup (seasonal influenza- was observed on dPly/PhtD efficacy against AOM and ALRI end-
vaccinated or seasonal influenza-unvaccinated) to assess the effi- points across the four winter seasons encompassed during the
cacy of dPly/PhtD versus the Control group. study (Supplemental Fig. 5).

Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx 5

Fig. 2. Trial profile. ATP, according-to-protocol; PCV13, 13-valent pneumococcal conjugate vaccine. Note that children may have several criteria for elimination from ATP
cohorts and only the primary reason is provided here. aOne participant was censored before vaccine dose 3 and did not contribute to the efficacy analysis from two weeks
after the administration of dose 3.

3.2. Reactogenicity and safety LLOQ of the assay at all timepoints. Antibody GMCs for both anti-
gens were higher in the dPly/PhtD group than the Control group
Pain and irritability were the most frequently reported solicited one month post-primary and post-booster vaccination and
local and general symptoms, respectively, after primary and boos- increased after primary and booster vaccination compared to base-
ter vaccination (Table 4). Percentage of doses followed by at least line (Fig. 4). Twelve months post-booster vaccination, anti-Ply anti-
one unsolicited and grade 3 unsolicited AEs were within similar body GMCs in the dPly/PhtD group waned but remained higher
ranges between groups. SAEs were reported from 229/900 compared to the Control group; while anti-PhtD antibody GMCs
(25.4%) dPly/PhtD and 232/903 (25.7%) Control children (Supple- were within the same range for both groups. In exploratory analy-
mental Table 8). Vaccination-related SAEs were reported from ses, we did not detect a clear relationship between pre- and post-
two children in the dPly/PhtD group (pyrexia plus diarrhea and primary vaccination anti-Ply and anti-PhtD antibody concentration
pyrexia) and one in the Control group (febrile convulsion); no fatal (Supplemental Fig. 6).
SAEs were reported. IPD was reported from five children in the
dPly/PhtD group and one in the Control group, all due to non-
4. Discussion
PCV13 types.
Currently available pneumococcal vaccines do not include anti-
3.3. Immunogenicity gens against all pneumococcal serotypes. Vaccines that offer
serotype-independent protection, such as those targeting con-
In the ATP immunogenicity cohort (N = 124/group), all children served proteins that are common to most or all pneumococcal
had anti-Ply and anti-PhtD antibody concentrations above the strains, could substantially reduce global morbidity and mortality.
Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
6 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx

Table 2
Vaccine efficacy of dPly/PhtD vaccine against acute otitis media outcomes by cohort.

Modified ATP cohort for efficacy ATP cohort for efficacy TVC
Follow-up from 2 weeks after dose 3 Follow-up from 2 weeks after dose 3 Follow-up from dose 1
dPly/PhtD Control VE, % dPly/PhtD Control VE, % dPly/PhtD Control VE, %
group group (95% CI) group group (95% CI) group group (95% CI)
N = 808 N = 831 N = 731 N = 758 N = 900 N = 903
All AOM episodes
AAP-AOM 485a 518 3.8* 455 487 3.1 641 680 4.1
(0.43)b (0.44) (
11.4, 16.9) (0.44) (0.45) (
12.6, 16.7) (0.40) (0.42) (
9.6, 16.0)
Modified AAP-AOM 648 702 5.2 607 659 4.5 853 913 4.9
(0.57) (0.60) (
8.0, 16.8) (0.58) (0.61) (
9.2, 16.5) (0.54) (0.57) (
7.1, 15.5)
HCP-AOM 774 819 2.9 722 766 2.3 1019 1065 2.6
(0.68) (0.70) (
9.5, 14.0) (0.69) (0.71) (
10.8, 13.8) (0.64) (0.66) (
8.7, 12.7)
First AOM episode
AAP-AOM 309 346 11.3 289 324 9.9 382 405 6.3
(0.37) (0.42) (
3.4, 23.9) (0.38) (0.43) (
5.5, 23.1) (0.33) (0.36) (
7.8, 18.5)
Modified AAP-AOM 380 425 11.2 356 398 9.7 459 486 5.9
(0.50) (0.57) (
2.0, 22.6) (0.52) (0.59) (
4.1, 21.8) (0.44) (0.47) (
6.9, 17.2)
HCP-AOM 423 463 8.9 389 429 8.7 509 530 4.2
(0.59) (0.66) (
3.9, 20.2) (0.60) (0.67) (
4.8, 20.4) (0.52) (0.54) (
8.2, 15.2)

dPly/PhtD Group = dPly/PhtD vaccine co-administered with PCV13 at 2, 4, 6 and 12–15 months of age.
Control Group = Placebo co-administered with PCV13 at 2, 4, 6 and 12–15 months of age.
dPly/PhtD, pneumolysin toxoid/pneumococcal histidine-triad protein D; ATP, according-to-protocol; TVC, total vaccinated cohort; CI, confidence interval; VE, vaccine
efficacy; N, number of participants; AAP, American Academy of Pediatrics; AOM, acute otitis media; HCP, health-care provider; PCV13, 13-valent pneumococcal conjugate
vaccine.
a
Number of episodes.
b
Incidence (episodes/child-year).
*Result for primary objective.

Table 3
Vaccine efficacy against acute lower respiratory tract infection outcomes by cohort.

Modified ATP cohort for efficacy ATP cohort for efficacy TVC
Follow-up from 2 weeks after dose 3 Follow-up from 2 weeks after dose 3 Follow-up from dose 1
dPly/PhtD Control VE, % dPly/PhtD Control VE, % dPly/PhtD Control VE, %
group group (95% CI) group group (95% CI) group group (95% CI)
N = 808 N = 831 N = 731 N = 758 N = 900 N = 903
All ALRI episodes
MA-ALRI 163a 165
1.5 153 152
4.4 236 236
1.7
(0.14)b (0.14) (
32.6, 22.4) (0.15) (0.14) (
37.7, 20.9) (0.15) (0.15) (
28.1, 19.3)
MA-ALRI with fever 125 123
4.4 116 113
6.4 169 162
6.1
(0.11) (0.11) (
39.2, 21.8) (0.11) (0.11) (
43.8, 21.3) (0.11) (0.10) (
36.7, 17.6)
MA-HCP-ALRI with fever 289 303 2.0 266 279 1.2 401 422 3.4
(0.25) (0.26) (
18.3, 18.8) (0.26) (0.26) (
20.3, 18.8) (0.25) (0.26) (
14.4, 18.3)
First ALRI episode
MA-ALRI 121 135 9.3 113 126 8.3 163 175 5.7
(0.12) (0.13) (
15.9, 29.0) (0.12) (0.13) (
18.2, 28.9) (0.12) (0.12) (16.7, 23.9)
MA-ALRI with fever 99 106 4.9 91 98 4.6 127 134 4.0
(0.09) (0.10) (
25.1, 27.7) (0.09) (0.10) (
26.9, 28.3) (0.09) (0.09) (
22.4, 24.7)
MA-HCP- ALRI with fever 207 226 8.2 188 211 10.0 267 286 6.4
(0.22) (0.24) (
10.9, 24.0) (0.22) (0.24) (
9.5, 26.1) (0.21) (0.22) (
10.6, 20.8)

dPly/PhtD Group = dPly/PhtD vaccine co-administered with PCV13 at 2, 4, 6 and 12–15 months of age.
Control Group = Placebo co-administered with PCV13 at 2, 4, 6 and 12–15 months of age.
dPly/PhtD, pneumolysin toxoid/pneumococcal histidine-triad protein D; ATP, according-to-protocol; TVC, total vaccinated cohort; N, number of participants; CI, confidence
interval; VE, vaccine efficacy; MA, medically-attended; ALRI, acute lower respiratory tract infection; HCP, health-care provider; PCV13, 13-valent pneumococcal conjugate
vaccine.
a
Number of episodes.
b
Incidence (episodes/child-year).

In this study, dPly/PhtD vaccine was immunogenic and had an designed to evaluate an effect of pneumococcal proteins against
acceptable reactogenicity and safety profile after primary and disease in humans as a primary endpoint. Among infants in The
booster vaccination when co-administered with PCV13; however, Gambia study, the inclusion of pneumococcal proteins plus 10
we did not observe significant efficacy against any of the pre- serotype-specific polysaccharide conjugates (PHiD-CV/dPly/PhtD),
defined AOM or ALRI endpoints. administered at 2, 3, 4 months or 2, 4, 9 months of age did not
In pre-clinical studies, Ply and PhtD, administered either sepa- reduce the prevalence or density of pneumococcal carriage beyond
rately or in combination, provided protection against pneumococ- the effect of PHiD-CV [20]. The reasons for the differences in out-
cal carriage and disease, including pneumonia in macaques comes between animal models and human clinical trials are
[12,16,17,36]. Several other studies have investigated the impact unclear. It has been suggested that the absence of an effect on car-
of protein-based or whole cell vaccines against colonization end- riage in infants in The Gambia may be a consequence of the acqui-
points in humans, but to our knowledge, this was the first study sition of carriage early in life, along with other risk factors for high
Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx 7

1
dPly/PhtD Group Control Group
0.9

0.8

Cumulave hazard for first episode

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 5 10 15 20 25
Time (month) since 2 weeks aer the administraon of dose 3
Number at risk
dPly/PhtD 808 627 515 479 9
Control 831 620 509 454 9

1
dPly/PhtD Group Control Group
0.9
Cumulave hazard for first episode

0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0
0 5 10 15 20 25
Time (month) since 2 weeks aer the administraon of dose 3
Number at risk
dPly/PhtD 808 749 672 641 19
Control 831 749 681 646 14

Fig. 3. Cumulative hazard curves for the first occurrence of AAP-AOM (top panel) and MA-ALRI (bottom panel), from 2 weeks after the administration of dose 3
(modified ATP cohort for efficacy). dPly/PhtD = dPly/PhtD vaccine co-administered with PCV13 at 2, 4, 6 and 12–15 months of age. Control = Placebo co-administered with
PCV13 at 2, 4, 6 and 12–15 months of age. AAP, American Academy of Pediatrics; AOM, acute otitis media; MA, medically-attended; ALRI, acute lower respiratory tract
infection; ATP, according-to-protocol; dPly/PhtD, pneumolysin toxoid/pneumococcal histidine-triad protein D; PCV13, 13-valent pneumococcal conjugate vaccine.

density carriage; however, populations with a high force of infec- month post-dose 3 was observed compared to baseline; whereas
tion and a variety of risk factors are precisely where new vaccines antibody GMCs in the Control group declined. Levels of anti-PhtD
are needed most. antibody were similar at 12 months of age and, although there
In spite of the lack of demonstrated efficacy we observed a was an increase in antibody GMCs one month-post booster in
robust serum antibody response to the protein antigens. Antibody dPly/PhtD recipients compared to the Control group, levels were
levels to the Ply component were substantially higher in the dPly/ similar again by 24-months of age. The pattern of anti-protein anti-
PhtD group compared to the Control group at all timepoints. Anti- body responses was similar to findings in infants in The Gambia
Ply antibody concentrations were frequently beyond the assay and in Europe, although the GMCs were substantially higher in
ULOQ following the booster dose. Because values above the assay the Native American infants at baseline (pre-vaccination) and
ULOQ were assigned as the ULOQ value, the anti-Ply antibody post-primary vaccination compared to European infants who
GMCs for dPly/PhtD recipients are underestimated. We were received dPly/PhtD containing vaccine at 2, 3, 4 and 12–15 months
unable to quantify the functional response of anti-Ply antibodies and Gambian infants who were vaccinated at 2, 3, 4 months
(i.e., the inhibition of Ply hemolysis activity) because of issues with [20,22]. The high baseline anti-protein antibody levels in this study
assay stability. For anti-PhtD, an increase in antibody GMCs one presumably result from exposure at a young age or from maternal

Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
8 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx

Table 4
Percentages of doses (with 95% confidence interval) followed by solicited or unsolicited adverse events (AEs) – immunogenicity/reactogenicity sub-cohort (total vaccinated
cohort).

3-dose primary vaccination phase Booster dose

dPly/PhtD group Control group dPly/PhtD group Control group
Solicited AEs at dPly/PhtD vaccine or AlPO4 Placebo injection site (days 0–3)
N = 551 N = 553 N = 156 N = 153
Pain 69.3 66.2 64.1 54.2
(65.3, 73.2) (62.1, 70.1) (56.0, 71.6) (46.0, 62.3)
Grade 3 14.3 17.9 15.4 11.8
(11.5, 17.5) (14.8, 21.4) (10.1, 22.0) (7.1, 18.0)
Redness 31.6 31.1 37.8 39.2
(27.7, 35.6) (27.3, 35.1) (30.2, 45.9) (31.4, 47.4)
>30 mm 0.2 0.0 0.0 0.0
(0.0, 1.0) (0.0, 0.7) (0.0, 2.3) (0.0, 2.4)
Swelling 22.7 16.1 25.0 24.2
(19.3, 26.4) (13.1, 19.4) (18.4, 32.6) (17.6, 31.8)
>30 mm 0.2 0.0 0.0 0.0
(0.0, 1.0) (0.0, 0.7) (0.0, 2.3) (0.0, 2.4)
General solicited AEs (days 0–3)
N = 551 N = 554 N = 156 N = 154
Drowsiness 45.0 46.9 41.7 42.2
(40.8, 49.3) (42.7, 51.2) (33.8, 49.8) (34.3, 50.4)
Grade 3 6.5 3.4 8.3 7.8
(4.6, 8.9) (2.1, 5.3) (4.5, 13.8) (4.1, 13.2)
Irritability 64.4 62.3 62.2 55.8
(60.3, 68.4) (58.1, 66.3) (54.1, 69.8) (47.6, 63.8)
Grade 3 9.4 10.8 17.3 10.4
(7.1, 12.2) (8.4, 13.7) (11.7, 24.2) (6.1, 16.3)
Loss of appetite 24.9 23.6 27.6 (20.7, 35.3) 26.6
(21.3, 28.7) (20.2, 27.4) (19.8, 34.3)
Grade 3 2.5 1.8 3.8 (1.4, 8.2) 4.5
(1.4, 4.2) (0.9, 3.3) (1.8, 9.1)
Fevera 7.6 10.6 5.1 9.7
(5.5, 10.2) (8.2, 13.5) (2.2, 9.9) (5.6, 15.6)
>40.0 °C 0.0 0.2 0.0 0.0
(0.0, 0.7) (0.0, 1.0) (0.0, 2.3) (0.0, 2.4)
Unsolicited AEsb (days 0–30)
N* = 572 N* = 579 N* = 178 N* = 174
Any 29.4 28.0 29.8 27.0
(25.7, 33.3) (24.4, 31.8) (23.2, 37.1) (20.6, 34.3)
Grade 3 3.1 3. 5.6 2.9
(1.9, 4.9) (1.9, 4.9) (2.7, 10.1) (0.9, 6.6)

dPly/PhtD Group = dPly/PhtD vaccine co-administered with PCV13 at 2, 4, 6 and 12–15 months of age.
Control Group = Placebo co-administered with PCV13 at 2, 4, 6 and 12–15 months of age.
dPly/PhtD, pneumolysin toxoid/pneumococcal histidine-triad protein D; N/N*, number of documented/administered doses; grade 3, crying when limb was moved/limb
spontaneously painful (pain), preventing normal everyday activity (drowsiness, unsolicited AEs), crying inconsolably/preventing normal everyday activity (irritability), not
eating at all (loss of appetite); PCV13, 13-valent pneumococcal conjugate vaccine.
a
38.0 °C (regardless of the route of measurement; axillary temperature presented here).
b
The most frequently reported unsolicited AEs were upper respiratory tract infection, viral infection, pyrexia, cough and rhinorrhea post-primary vaccination, and pyrexia
and upper respiratory tract infection post-booster.

antibody transfer in utero, both of which are plausible given the of GSK’s dPly or PhtD antigens on AOM. Other studies have shown
high burden of pneumococcal carriage and disease in this popula- potential protection by pneumococcal proteins, including PhtD and
tion [8,37,38]. In a post-hoc analysis, we did not find any evidence Ply, against AOM in a murine model; however, these findings
to suggest that pre-vaccination antibody interfered with dPly/PhtD might not translate to humans [40]. Additionally, differences in
immunogenicity. the production method might lead to antigenic differences that
It is important to acknowledge that this study measured serum could affect efficacy. The study was powered to detect an incre-
but not mucosal antibodies. In a prospective study of 100 healthy mental VE of 17% or greater against the syndrome of AOM; this
infants 6–24 months of age, higher mucosal antibody levels to PhtD may have been too high a bar given that the dPly/PhtD vaccine
and dPly correlated with reduced risk of development of pneumo- was being co-administered with PCV13. The reported VE of
coccal AOM in children, although no significant difference was licensed PCVs against clinical AOM assessed in clinical trials ranged
found in mucosal antibody levels to PhtD and dPly between chil- up to a 15% reduction, but was non-statistically significant [41]; we
dren with and without pneumococcal carriage [39]. Results were used a slightly higher estimated VE given the serotype-
not stratified by episode number so it is unclear if these findings independent nature of the dPly/PhtD vaccine. At the time the study
were similar between first and subsequent episodes of AOM. While was designed, the available data suggested that around a quarter of
the study presented herein did not measure mucosal antibodies – a AOM was pneumococcal; however, the burden of pneumococcal
step that should be considered in future studies of pneumococcal AOM, particularly amongst cases of recurrent AOM, has continued
protein vaccines – the serum antibody levels intimate that mucosal to decline in the PCV era and more recent data suggest an increas-
levels were also likely high among dPly/PhtD vaccine recipients. ing role of other pathogens, particularly H. influenzae [42,43]. Accu-
There are several possible reasons that efficacy was not rately diagnosing AOM can be challenging and overdiagnosis of
observed, despite very good serum antibody responses to the AOM is common; inclusion of events that were not truly AOM
dPly/PhtD vaccine. Pre-clinical studies did not examine the effect would reduce the ability to detect efficacy. We attempted to
Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx 9

100000

GMC (95% CI) log10

10000

1000

100

10

1
Baseline 1M post-primary Pre-booster 1M post-booster 12M post-booster
(6-12 weeks) (7 months) (12-15 months) (13-16 months) (24-27 months)

dPly/PhtD Group (N=116) Control Group (N=120)

100000
GMC (95% CI) log10

10000
1000
100
10
1
Baseline 1M post-primary Pre-booster 1M post-booster 12M post-booster
(6-12 weeks) (7 months) (12-15 months) (13-16 months) (24-27 months)

dPly/PhtD Group (N=116) Control Group (N=120)

Fig. 4. Antibody geometric mean concentrations (GMCs) against pneumococcal Ply (top panel) and PhtD (bottom panel) proteins (ATP cohort for immunogenicity).
dPly/PhtD Group = dPly/PhtD vaccine co-administered with PCV13 at 2, 4, 6 and 12–15 months of age. Control Group = Placebo co-administered with PCV13 at 2, 4, 6 and 12–
15 months of age. dPly/PhtD, pneumolysin toxoid/pneumococcal histidine-triad protein D; ATP, according-to-protocol; CI, confidence interval; M, month; N, maximum
number of participants with available results; PCV13, 13-valent pneumococcal conjugate vaccine.

minimize incorrect AOM diagnoses by training providers on the dPly/PhtD vaccine beyond that of PCV13, which was more pro-
diagnosis of AOM through in-person sessions at the start of the nounced for MA-ALRI. Compared to the Control group, dPly/PhtD
study and online refresher modules during the study, and by using vaccine recipients had higher antibody GMCs to PhtD at 7 months
a standardized template for capture of AOM outcomes that of age, but this difference was no longer evident by 1 year of age.
included all relevant elements of the history and exam [44]. During PhtD plays a role in adherence of S. pneumoniae to human nasopha-
the course of the study, the AAP published updated criteria for ryngeal epithelial cells [46]. It is possible that the trend toward
diagnosis of AOM, which were intended to more clearly differenti- greater protection among dPly/PhtD vaccine recipients against first
ate AOM from otitis media with effusion. Because the study was AOM or ALRI episodes in the first year of life relates to the notable
ongoing, no changes to the study objectives were made apart from difference in anti-PhtD levels, and that this clinical benefit was no
collection of new data to allow assessment of efficacy according to longer evident when levels became more similar to those of the
the updated definition (AAP-AOM [2013]) [45]. VE against episodes Control group. However, the lack of consistent and statistically sig-
of AAP-AOM was similar using the 2004 and 2013 definitions; nificant findings limits our ability to make conclusions. Another
however, this analysis was limited because some details of the possible reason that efficacy may have been greater against first
AOM events could not be collected retrospectively. Reassuringly, episodes relates to the role of biofilms. Although not assessed in
the incidence rate of AOM in this study (0.43–0.44 episodes/ this study, biofilms are typically associated with recurrent infec-
child-year) was similar to that documented by tympanocentesis tions and allow bacteria to resist host immune responses [47]. Dis-
among children aged 6–36 months in Rochester, New York with persed biofilm bacteria have increased virulence; animal models
AAP-AOM [2013] in the PCV era (0.38 episodes/child-year during that evaluate the impact of anti-protein antibodies on disease
the first year of life and 0.48 episodes/child-year during the second using broth-grown bacteria, as opposed to dispersed biofilm bacte-
year of life) [42]. Nevertheless, AOM may still have been mis- or ria, may not accurately reflect pathogenesis in humans.
overdiagnosed, which could have resulted in the study being Because influenza vaccination and season might impact the
underpowered and biased toward the null hypothesis. In addition, incidences of AOM and ALRI [34,35], we also conducted an explora-
culture of middle ear fluid to assess the etiology of draining AOM tory analysis to understand their potential impact on dPly/PhtD
was done at the discretion of the clinical provider and not as a efficacy. This post-hoc analysis did not show a consistent impact
study procedure. of influenza vaccination on the VE of dPly/PhtD against AOM and
Efficacy tended to be higher against first AOM and ALRI epi- ALRI. The small number of children in the subsets and hence the
sodes compared to subsequent episodes. This finding, along with low number of AOM and ALRI cases recorded are a limitation for
the observed antibody profiles over the course of the study, this analysis.
prompted us to conduct a post-hoc analysis in the mATP cohort The frequency of solicited and unsolicited AEs was similar in
of VE against first AOM and first MA-ALRI events at age dPly/PhtD and Control participants after primary and booster vac-
<12 months. There was a suggestion of incremental VE of cination. IPD events, which were identified through the surveil-

Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
10 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx

lance for SAEs since they were not disease outcome events in the ical care providers at facilities serving study participants. We also
protocol, were more common in the dPly/PhtD group (n = 5) com- appreciate the time and contribution of the Independent Data
pared to the Control group (n = 1), although this was not a statis- Monitoring Committee. We thank the clinical and serological labo-
tically significant difference. As part of a long-standing ratory teams of the GSK group of companies for their contribution
population-based Active Bacterial Surveillance (ABS) system for to this study, in particular Asparuh Gardev (XPE Pharma & Science
IPD at the study sites, two additional cases of IPD in participants c/o GSK) for clinical development team, Sonia Schoonbroodt and
from the Control group were identified: one in a participant who Sophie Eugène for clinical laboratory team, Patricia Lommel, Nancy
completed the study before IPD occurred at 31 months of age, and François and Florence Lemahieu (Keyrus Biopharma c/o GSK) for
one in a participant that had withdrawn from the study prior to statistical input, Anne Schuind and Arshad Amanullah for input
the occurrence of IPD at 25 months of age. Because of these cir- in the influenza vaccination analysis, Kevin Carrick, Katleen van
cumstances the events were not included in the clinical study Hoefs (Keyrus Biopharma c/o GSK), Maaria Soila, and Liesbet De
database although they were reported to ABS. The study was Cock for study monitoring and management, Valérie Balosso, Marie
not powered to detect a difference in IPD outcomes and the Devèze (Keyrus Biopharma c/o GSK), and Aurélia Le Prince for
work-up for possible invasive bacterial infection was done at study data management, Liliana Manciu for protocol writing, and
the discretion of the child’s provider, rather than systematically Domenica Majorino (Modis c/o GSK) for clinical report writing.
as part of the study. Stéphanie Deroo (Modis c/o GSK) provided editorial assistance by
In conclusion, this study showed that vaccination with dPly and collating and incorporating authors0 comments and revisions to
PhtD antigens was well-tolerated and immunogenic but did not the manuscript and formatting the manuscript to journal
show additional protection to children against AOM or ALRI. The guidelines.
interactions between immunity, carriage, and progression to dis-
ease are complex and poorly understood. Future studies should
Authors0 contributions
aim to improve our understanding of the effect of anti-protein
antibodies on pathogenesis of pneumococcal disease.
All authors have substantially contributed to the submitted
work. L. H. codesigned and performed the study, collected, ana-
5. Trademark statement lyzed, and interpreted the data, and wrote the manuscript. J. C.,
R. W., and M. S. codesigned and performed the study, and collected,
Synflorix is a trademark of the GSK group of companies. Prevenar analyzed, and interpreted the data. D. B., K. S., and K. O’B. code-
13/Prevnar 13 is a trademark of Pfizer, Inc. signed the study and analyzed and interpreted the data. R. R. per-
formed the study and collected, analyzed, and interpreted the data.
Funding N. G. performed the study and collected the data. M. T., Y. S., and L.
M. analyzed and interpreted the data. All authors reviewed and
This study was funded by GlaxoSmithKline Biologicals SA. edited the manuscript, and approved its submission.
All authors attest they meet the ICMJE criteria for authorship.
Declaration of Competing Interest
Appendix A. Supplementary material
The authors declare the following financial interests/personal
relationships which may be considered as potential competing Supplementary data to this article can be found online at
interests: [Potential Conflict of interest: L.L. Hammitt, J.C. Camp- https://doi.org/10.1016/j.vaccine.2019.09.076.
bell, R.C. Weatherholtz, R. Reid, M. Santosham and K.L. O’Brien
report institutional research grants from the GSK group of compa-
References
nies, during the conduct of the study. L.L. Hammitt, R.C. Weather-
holtz and K.L. O’Brien also report institutional grants from Pfizer, [1] Wahl B, O’Brien KL, Greenbaum A, et al. Burden of Streptococcus pneumoniae
Novavax and Merck outside the submitted work. M. Santosham also and Haemophilus influenzae type b disease in children in the era of conjugate
reports an institutional research grant for Rota Council from the vaccines: global, regional, and national estimates for 2000–15. Lancet Glob
Health 2018;6(7):e744–57.
GSK group of companies. K.L. O’Brien also reports to be an external [2] IVAC. VIEW-hub. 2017; http://view-hub.org/. Accessed July 25, 2017.
scientific advisor for Pfizer, Sanofi Pasteur, Merck and the GSK [3] Cortese MM, Wolff M, Almeido-Hill J, Reid R, Ketcham J, Santosham M. High
group of companies. L.H. Moulton reports personal fees from a incidence rates of invasive pneumococcal disease in the White Mountain
Apache population. Arch Intern Med 1992;152(11):2277–82.
Merck Scientific Advisory Board outside the submitted work. N.
[4] Davidson M, Parkinson AJ, Bulkow LR, Fitzgerald MA, Peters HV, Parks DJ. The
Goklish has no conflict of interest to disclose. M. Traskine, K. Swin- epidemiology of invasive pneumococcal disease in Alaska, 1986–1990–ethnic
nen and D. Borys are employees of the GSK group of companies, and differences and opportunities for prevention. J Infect Dis 1994;170(2):368–76.
[5] O’Brien KL, Santosham M. Potential impact of conjugate pneumococcal
K. Swinnen and D. Borys hold shares of the GSK group of companies.
vaccines on pediatric pneumococcal diseases. Am J Epidemiol 2004;159
Y. Song worked for XPE Pharma & Science as a consultant for the (7):634–44.
GSK group of companies.] [6] Watt JP, O’Brien KL, Benin AL, et al. Invasive pneumococcal disease among
Navajo adults, 1989–1998. Clin Infect Dis 2004;38(4):496–501.
[7] Bruce MG, Singleton R, Bulkow L, et al. Impact of the 13-valent pneumococcal
Acknowledgements conjugate vaccine (PCV13) on invasive pneumococcal disease and carriage in
Alaska. Vaccine 2015;33(38):4813–9.
[8] Weatherholtz R, Grant L, Douglass G, et al. Impact of pneumococcal conjugate
We are deeply grateful to the parents of the Native American vaccines (PCV) on invasive pneumococcal disease among American Indians
children who participated in the study and the research program living on the Navajo reservation. Melbourne, Australia: 11th International
assistants, study nurses and study physicians for their dedicated Symposium on Pnumococci and Pneumococcal Diseases; 2018.
[9] Weatherholtz R, Millar EV, Moulton LH, et al. Invasive pneumococcal disease a
efforts on this trial. This study would not have been possible with-
decade after pneumococcal conjugate vaccine use in an American Indian
out the collaboration of the IHS and tribally-operated hospitals and population at high risk for disease. Clin Infect Dis 2010;50(9):1238–46.
clinics and the guidance and oversight of the Institutional Review [10] Alderson MR. Status of research and development of pediatric vaccines for
Boards of Johns Hopkins Bloomberg School of Public Health, the Streptococcus pneumoniae. Vaccine 2016;34(26):2959–61.
[11] Eijkelkamp BA, Pederick VG, Plumptre CD, et al. The first histidine triad motif
Navajo Nation and the Phoenix Area IHS. We thank Dr. Alejandro of PhtD Is critical for zinc homeostasis in Streptococcus pneumoniae. Infect
Hoberman for providing training on the assessment of AOM to clin- Immun 2016;84(2):407–15.
Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076
 L.L. Hammitt et al. / Vaccine xxx (xxxx) xxx 11

[12] Hermand P, Vandercammen A, Mertens E, et al. Preclinical evaluation of a [28] Casey JR, Adlowitz DG, Pichichero ME. New patterns in the otopathogens
chemically detoxified pneumolysin as pneumococcal vaccine antigen. Hum causing acute otitis media six to eight years after introduction of
Vaccin Immunother 2017;13(1):220–8. pneumococcal conjugate vaccine. Pediatr Infect Dis J 2010;29(4):304–9.
[13] Rioux S, Neyt C, Di Paolo E, et al. Transcriptional regulation, occurrence and [29] Eskola J, Kilpi T, Palmu A, et al. Efficacy of a pneumococcal conjugate vaccine
putative role of the Pht family of Streptococcus pneumoniae. Microbiology against acute otitis media. N Engl J Med 2001;344(6):403–9.
2011;157(Pt 2):336–48. [30] O’Brien KL, David AB, Chandran A, et al. Randomized, controlled trial efficacy of
[14] Adamou JE, Heinrichs JH, Erwin AL, et al. Identification and characterization of pneumococcal conjugate vaccine against otitis media among Navajo and
a novel family of pneumococcal proteins that are protective against sepsis. White Mountain Apache infants. Pediatr Infect Dis J 2008;27(1):71–3.
Infect Immun 2001;69(2):949–58. [31] Tregnaghi MW, Saez-Llorens X, Lopez P, et al. Efficacy of pneumococcal
[15] Alexander JE, Lock RA, Peeters CC, et al. Immunization of mice with nontypable Haemophilus influenzae protein D conjugate vaccine (PHiD-CV) in
pneumolysin toxoid confers a significant degree of protection against at young Latin American children: a double-blind randomized controlled trial.
least nine serotypes of Streptococcus pneumoniae. Infect Immun 1994;62 PLoS Med 2014;11(6):e1001657.
(12):5683–8. [32] Kelly PJ, Lim LL. Survival analysis for recurrent event data: an application to
[16] Denoel P, Godfroid F, Hermand P, Verlant V, Poolman J. Combined protective childhood infectious diseases. Stat Med 2000;19(1):13–33.
effects of anti-PhtD and anti-pneumococcal polysaccharides. Vaccine 2011;29 [33] Edwards KM. Maternal antibodies and infant immune responses to vaccines.
(38):6451–3. Vaccine 2015;33(47):6469–72.
[17] Denoel P, Philipp MT, Doyle L, Martin D, Carletti G, Poolman JT. A protein- [34] Flannery B, Chung JR, Belongia EA, et al. Interim estimates of 2017–18 seasonal
based pneumococcal vaccine protects rhesus macaques from pneumonia after influenza vaccine effectiveness - United States, February 2018. MMWR Morb
experimental infection with Streptococcus pneumoniae. Vaccine 2011;29 Mortal Wkly Rep 2018;67(6):180–5.
(33):5495–501. [35] Norhayati MN, Ho JJ, Azman MY. Influenza vaccines for preventing acute otitis
[18] Kaur R, Surendran N, Ochs M, Pichichero ME. Human antibodies to PhtD, PcpA, media in infants and children. Cochrane Database Syst Rev 2017;10:
and Ply reduce adherence to human lung epithelial cells and murine CD010089.
nasopharyngeal colonization by Streptococcus pneumoniae. Infect Immun [36] Salha D, Szeto J, Myers L, et al. Neutralizing antibodies elicited by a novel
2014;82(12):5069–75. detoxified pneumolysin derivative, PlyD1, provide protection against both
[19] Odutola A, Ota MO, Ogundare EO, et al. Reactogenicity, safety and pneumococcal infection and lung injury. Infect Immun 2012;80(6):2212–20.
immunogenicity of a protein-based pneumococcal vaccine in Gambian [37] Grant LR, Hammitt LL, O’Brien SE, et al. Impact of the 13-valent pneumococcal
children aged 2–4 years: a phase II randomized study. Hum Vaccin conjugate vaccine on pneumococcal carriage among American Indians. Pediatr
Immunother 2016;12(2):393–402. Infect Dis J 2016;35(8):907–14.
[20] Odutola A, Ota MOC, Antonio M, et al. Efficacy of a novel, protein-based [38] Mendy AL, Agbla SC, Odutola AA, et al. Kinetics of antibodies against
pneumococcal vaccine against nasopharyngeal carriage of Streptococcus pneumococcal proteins and their relationship to nasopharyngeal carriage in
pneumoniae in infants: A phase 2, randomized, controlled, observer-blind the first two months of life. PLoS ONE 2017;12(10):e0185824.
study. Vaccine 2017;35(19):2531–42. [39] Xu Q, Casey JR, Almudevar A, Pichichero ME. Correlation of higher antibody
[21] Prymula R, Pazdiora P, Traskine M, Ruggeberg JU, Borys D. Safety and levels to pneumococcal proteins with protection from pneumococcal acute
immunogenicity of an investigational vaccine containing two common otitis media but not protection from nasopharyngeal colonization in young
pneumococcal proteins in toddlers: a phase II randomized clinical trial. children. Clin Microbiol Infect 2017;23(7):487.e1–6.
Vaccine 2014;32(25):3025–34. [40] Xu Q, Pryharski K, Pichichero ME. Trivalent pneumococcal protein vaccine
[22] Prymula R, Szenborn L, Silfverdal SA, et al. Safety, reactogenicity and protects against experimental acute otitis media caused by Streptococcus
immunogenicity of two investigational pneumococcal protein-based pneumoniae in an infant murine model. Vaccine 2017;35(2):337–44.
vaccines: Results from a randomized phase II study in infants. Vaccine [41] Fortanier AC, Venekamp RP, Boonacker CW, et al. Pneumococcal conjugate
2017;35(35 Pt B):4603–11. vaccines for preventing acute otitis media in children. Cochrane Database Syst
[23] Centers for Disease Control and Prevention. Prevention and control of Rev 2019;5. CD001480.
influenza with vaccines: recommendations of the advisory committee on [42] Kaur R, Morris M, Pichichero ME. Epidemiology of acute otitis media in the
immunization practices (ACIP) 2011. MMWR Morb Mortal Wkly Rep 2011;60 postpneumococcal conjugate vaccine era. Pediatrics 2017;140(3):e20170101.
(33):1128–32. [43] Levy C, Varon E, Ouldali N, et al. Bacterial causes of otitis media with
[24] American Academy of Pediatrics Subcommittee on Management of Acute spontaneous perforation of the tympanic membrane in the era of 13 valent
Otitis Media. Diagnosis and management of acute otitis media. Pediatrics pneumococcal conjugate vaccine. PLoS ONE 2019;14(2):e0211712.
2004;113(5):1451–65. [44] Kaleida PH, Ploof DL, Kurs-Lasky M, et al. Mastering diagnostic skills:
[25] Leroux-Roels G, Maes C, De Boever F, Traskine M, Ruggeberg JU, Borys D. enhancing proficiency in otitis media, a model for diagnostic skills training.
Safety, reactogenicity and immunogenicity of a novel pneumococcal protein- Pediatrics 2009;124(4):e714–20.
based vaccine in adults: a phase I/II randomized clinical study. Vaccine [45] Lieberthal AS, Carroll AE, Chonmaitree T, et al. The diagnosis and management
2014;32(50):6838–46. of acute otitis media. Pediatrics 2013;131(3):e964–99.
[26] Prymula R, Peeters P, Chrobok V, et al. Pneumococcal capsular polysaccharides [46] Khan MN, Pichichero ME. Vaccine candidates PhtD and PhtE of Streptococcus
conjugated to protein D for prevention of acute otitis media caused by both pneumoniae are adhesins that elicit functional antibodies in humans. Vaccine
Streptococcus pneumoniae and non-typable Haemophilus influenzae: a 2012;30(18):2900–7.
randomised double-blind efficacy study. Lancet 2006;367(9512):740–8. [47] Bergenfelz C, Hakansson AP. Streptococcus pneumoniae otitis media
[27] Van Dyke MK, Pircon JY, Cohen R, et al. Etiology of acute otitis media in pathogenesis and how it informs our understanding of vaccine strategies.
children less than 5 years of age: a pooled analysis of 10 similarly designed Curr Otorhinolaryngol Rep 2017;5(2):115–24.
observational studies. Pediatr Infect Dis J 2017;36(3):274–81.

Please cite this article as: L. L. Hammitt, J. C. Campbell, D. Borys et al., Efficacy, safety and immunogenicity of a pneumococcal protein-based vaccine co-
administered with 13-valent pneumococcal conjugate vaccine against acute otitis media in young children: A phase IIb randomized study, Vaccine, https://
doi.org/10.1016/j.vaccine.2019.09.076