Original Studies
Group A Streptococcal Carriage and Seroepidemiology in
Children up to 10 Years of Age in Australia
Helen S. Marshall, MD,*†‡ Peter Richmond, FRACP,§¶ Michael Nissen, MB BS,‖** Stephen Lambert, PhD,‖**
Robert Booy, MD,††‡‡ Graham Reynolds, FRACP,§§ Shite Sebastian, PhD,¶¶ Michael Pride, PhD,¶¶
Kathrin U. Jansen, PhD,¶¶ Annaliesa S. Anderson, PhD,¶¶ and Ingrid L. Scully, PhD¶¶
Background: Group A streptococci (GAS) and other β-hemolytic
­streptococci (BHS) cause pharyngitis, severe invasive disease and seri-
ous nonsuppurative sequelae including rheumatic heart disease and post
streptococcal glomerulonephritis. The aim of this study was to assess
­carriage rates and anti-streptococcal C5a peptidase (anti-SCP) IgG levels
and identify epidemiologic factors related to carriage or seropositivity in
Australian children.
Methods: A throat swab and blood sample were collected for microbio-
logical and serological analysis (anti-SCP IgG) in 542 healthy children aged
0–10 years. Sequence analysis of the SCP gene was performed. Serological
analysis used a competitive Luminex Immunoassay designed to preferen-
tially detect functional antibody.
Results: GAS-positive culture prevalence in throat swabs was 5.0% (range
0–10%), with the highest rate in 5 and 9 years old children. The rate of
non-GAS BHS carriage was low (<1%). The scp gene was present in all 22
isolates evaluated. As age of child increased, the rate of carriage increased;
odds ratio, 1.14 (1.00, 1.29); P = 0.50. Geometric mean anti-SCP titers
increased with each age-band from 2 to 7 years, then plateaued. Age, geo-
graphic location and number of children within the household were signifi-
cantly associated with the presence of anti-SCP antibodies.
Conclusions: Children are exposed to GAS and other BHS at a young age,
which is important for determining the target age for vaccination to protect
before the period of risk.
Accepted for publication February 12, 2015.
From the *Vaccinology and Immunology Research Trials Unit, Women’s and
Children’s Hospital, †School of Paediatrics and Reproductive Health, ‡Rob-
inson Research Institute, University of Adelaide, North Adelaide, South
Australia, Australia; §School of Paediatrics and Child Health, University
of Western Australia, ¶Vaccine Trials Group, Wesfarmers Centre of Vac-
cines and Infectious Diseases, Telethon Kids Institute, Subiaco, Western
Australia, Australia; ‖Queensland Paediatric Infectious Diseases Laboratory,
Queensland Children’s Medical Research Institute, **Royal Children’s Hos-
pital, University of Queensland, Herston, Brisbane, Queensland, Australia;
††National Centre for Immunisation Research and Surveillance of Vaccine
Preventable Diseases, The Children’s Hospital at Westmead, Westmead,
‡‡Marie Bashir Institute, The University of Sydney, Sydney, New South
Wales, Australia; §§Department of Paediatrics and Child Health, ANU
­Medical School, The Canberra Hospital, Canberra, Australian Capital Terri-
tory, Australia; and ¶¶Pfizer Vaccine Research, Pearl River, New York.
Michael Nissen, MB BS is currently at GlaxosmithKline and University of
Queensland, Herston, Brisbane, Queensland, Australia.
This study did not require registration on a clinical trials registry as it did not
meet the WHO ICMJE definition for registration.
H.S.M., P.R., M.N., R.B. and G.R. declare their institutions received funding
from Wyeth (now Pfizer) to complete the work disclosed in this manu-
script. H.S.M. is a member of the Australian Technical Advisory Group on
Immunisation and the Therapeutic Goods Administration Australian Influ-
enza Vaccine Committee. H.S.M.’s institution (Women’s and Children’s
Hospital) has received research grants from GlaxoSmithKline, Novartis,
Sanofi Pasteur and Pfizer for independent investigator led studies and on
occasion travel support for H.S.M. to present research findings. She has
previously been a member of vaccine advisory boards for GlaxoSmith-
Kline and Novartis. H.S.M. has not accepted or received any personal
payment. R.B.’s institution has received funding from CSL, Hoffmann–La
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.
ISSN: 0891-3668/15/3408-0831
DOI: 10.1097/INF.0000000000000745
The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015 www.pidj.com | 831
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Key Words: Group A streptococcus, carriage, anti-streptococcal C5a pepti-
dase gene, GAS vaccines, acute rheumatic fever, children
(Pediatr Infect Dis J 2015;34:831–838)
G roup A Streptococcus (GAS) and occasionally other
β-hemolytic Streptococcus species (BHS) are the etiologic
agents of the common childhood infections, streptococcal phar-
yngitis and impetigo. GAS and other BHS [which include Group
B Streptococcus (GBS)] occasionally cause more severe disease,
such as septicemia, pneumonia, scarlet fever, necrotizing fascii-
tis, toxic shock syndrome and immune-mediated sequelae, such
as glomerulonephritis and acute rheumatic fever.1,2 Over the past
3 decades, there appears to have been a resurgence of severe inva-
sive GAS infections, such as streptococcal toxic shock syndrome
and necrotizing fasciitis with high-case fatality rates at 36% and
24%, respectively.3 The reason for this is unknown but may partly
be because of emergence of more virulent strains.4–6
Globally, there are over 100 million prevalent cases of pyo-
derma and over 600 million new cases of GAS pharyngitis each
year.2–7 The burden of disease with regard to pharyngitis impacts
both developed and developing countries equally. However, the
Roche, Sanofi, GlaxoSmithKline group of companies, Novartis, Baxter
and Pfizer to conduct sponsored research, educational grants or to attend
and present at scientific meetings. R.B. also received honorarium for
delivering educational presentations. Any funding received is directed
to a research account at The Children’s Hospital at Westmead and is not
personally accepted by R.B. P.R. has received institutional funding for
investigator-initiated research from GlaxoSmithKline Biologicals, Novar-
tis, Pfizer and Merck and received travel support from Pfizer and Baxter to
present study data at international meetings. M.N. previously directed the
Queensland Paediatric Infectious Diseases laboratory that has performed
the Meningococcal Antigen Testing System assay on Australian isolates
causing invasive meningococcal disease on the behalf of Novartis. M.N.
is now an employee of GlaxoSmithKline. He has received travel support
from GSK and Pfizer for conference attendance and presentation of data of
independent research at international meetings; honoraria from bioCSL,
Novartis and Pfizer for educational lectures; institutional funding for
investigator initiated research from Abbott Australasia as well as been an
principal investigator on vaccine and epidemiological studies sponsored
by a range of vaccine manufacturers, and in this role has received support
for conference attendance, presentation of data and membership of vac-
cine advisory boards. M.N. is the current Chair of the Australian National
Verification Committee for Measles Eradication and a past member of
ATAGI. S.L. and G.R. report no other conflicts of interest. S.S., M.P.,
K.U.J., A.S.A. and I.S. are employed by Pfizer (Vaccine Research) and as
such may own Pfizer shares. Pfizer was the funding source for the study
conduct. The microbiological and serological analyses were conducted by
Pfizer employees. The statistical analysis and modelling of predictors was
conducted by the non-industry investigators. No funding was provided for
the costs associated with the development and the publishing of the pres-
ent manuscript, which was drafted and completed by the first author with
contribution from all authors.
Address for correspondence: Helen S. Marshall, Vaccinology and Immunology
Research Trials Unit, Women’s and Children’s Hospital, 72 King William
Rd, North Adelaide, 5006 South Australia, Australia. E-mail: Helen.mar-
shall@adelaide.edu.au.
Marshall et al The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015
burden of serious complications or sequelae is concentrated in chil-
dren in developing countries and in indigenous children in devel-
oped countries.7–12
Development of a broadly effective GAS vaccine has been
challenging. There is considerable strain diversity within and
between countries, and the epidemiology of GAS is dynamic, with
shifts in serotype prevalence within populations.11,13,14 The GAS M
protein is a major virulence factor and target for vaccine develop-
ment.15,16 However, the diversity of M types in circulation around
the world makes this vaccine approach difficult to enable sufficient
coverage. Limited data indicate that unlike what is observed in
high-income countries in North America and Europe, there appear
to be wide diversity with no dominant M types in the low-income
and middle-income countries of Africa, Asia and the Pacific.17 The
26-valent M protein-based vaccine, which contains M peptides
from the most common serotypes in North America, is likely to
provide high coverage in high-income countries, but poor coverage
in Africa and the Pacific and only intermediate coverage in Asia and
the Middle East.17,18 In addition, although M proteins are present
in non-GBS BHS [Group C Streptococcus (GCS)/Group G Strep-
tococcus (GGS)], they are diverse, rendering a broadly protective
M-based approach challenging.19–21 Therefore, other approaches
for development of a broadly protective vaccine have been con-
sidered.22 One of a number of promising vaccine candidates is the
recombinant streptococcal C5a peptidase (SCP) antigen,23 which is
conserved across BHS, and has shown preclinical efficacy against
GAS and GBS.24,25
This study aimed to acquire nationally representative epide-
miological data on GAS in Australia to scope the appropriate age
for vaccination with a potential candidate vaccine to prevent GAS
infection in children. Our objectives were to describe anti-SCP
serum IgG antibodies in relation to age and BHS carriage status in
children aged 0 to 10 years and to determine any epidemiological
factors associated with carriage or seropositivity to SCP antigen to
inform GAS vaccine development.
METHODS
Study Population
Healthy children aged 0–10 years were recruited from 5
pediatric research centers in capital cities in Australia [Adelaide,
South Australia (SA), Perth, Western Australia (WA), Brisbane,
Queensland (Qld), Sydney, New South Wales (NSW), Canberra,
Australian Capital Territory (ACT)] between March 2007 and
November 2009. The study aimed to recruit 550 children, with
each center enrolling 10 children from each year of age from 0 to
10 years, with no more than 70% of the participants being male
or female. Exclusion criteria included a prior history of culture
confirmed invasive streptococcal disease, rheumatic heart disease,
glomerulonephritis, confirmed streptococcal pharyngitis or impe-
tigo within preceding month, antibiotics in the previous 3 days and
known/suspected immunodeficiency condition.
Children were randomly recruited from the community,
or children undergoing minor surgical procedures (excluding ear,
nose and throat procedures) were recruited from a day surgical unit.
Informed consent was obtained from parents, and demographic and
medical history details were collected.
Collection and Testing of Pharyngeal Swabs
Pharyngeal swabs were collected from all enrolled partici-
pants and cultured for GAS and other BHS in the microbiology
departments of participating hospitals following a standardized
protocol agreed by all investigators. To collect pharyngeal sam-
ples, a dry flocculated swab was wiped across the post pharyngeal
832  |  www.pidj.com
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
wall and tonsillar region by a trained research nurse or doctor. The
throat swabs were immediately streaked onto horse blood agar
plates within the study center microbiology departments, which
were then incubated at 37°C overnight and examined for the pres-
ence of colonies with zones of hemolysis typical of BHS. More
than one colony was checked for the presence of BHS, unless all
colonies appeared identical. The identity of β-haem colonies was
confirmed using the Phadebact Streptococcus Test (MKL Diag-
nostics, Stockholm, Sweden). The Phadebact tests use the coag-
glutination technique for confirmation of streptococcal group.
Antibodies, raised in rabbit and specific against groups A, B, C,
D, F and G streptococci, are bound to Protein A on the surface on
nonviable staphylococci. When a sample containing streptococci
belonging to one of these groups is mixed with its reagent, the spe-
cific antigens on the surface of the streptococci bind to the corre-
sponding specific antibodies. A coagglutination lattice is formed,
visible to the naked eye.
Phadebact Streptococcus tests, Strep A test, Strep B test,
Strep D test and Strep F test are intended for the identification of
BHS belonging to groups A, B, C, D, F and G.
Any positive culture result was reported to the treating phy-
sician for follow-up. BHS isolates were tested by Pfizer for the
presence of and sequence of the scp gene. Streptococcal DNA was
prepared using the Qiagen DNeasy Blood and Tissue Kit (Valen-
cia, CA) with the addition of 10 mg/mL lysozyme and 100 μg/mL
mutanolysin to the lysis buffer. A ~3.5-kb scp polymerase chain
reaction amplicon that spans the protease through the Fn3 domain
of the open reading frame was generated using the bacterial DNA
template, oligonucleotide primers 5′-CAAATACTGTGACAGAA-
GACACTCCTG-3′ and 5′-TTTGTTTTAAAGATATGA GCTAC-
TAATCCCAAG-3′, PrimeSTAR HS DNA polymerase (TaKaRa
Bio, Mountain View, CA) and 30 amplification cycles (98°C × 10
seconds/57°C × 15 seconds/72°C × 4 minutes). Polymerase chain
reaction products were purified using Agencourt AMPure XP
beads (Beckman Coulter, Brea, CA). DNA sequence was gener-
ated using the Big Dye Terminator V3.1 sequencing kit and the
ABI 3730 sequence analyzer (Applied Biosystems, Carlsbad, CA).
The deduced amino acid sequence of SCP (residues 89–1029) was
aligned using MegAlign (DNASTAR Lasergene software).
Blood Sample Collection for Serology
Blood samples were collected, and anti-C5a peptidase anti-
body concentrations were measured using an SCP competitive
Luminex Immunoassay (cLIA) at Pfizer laboratories. The single-
plex SCP cLIA is a serology assay capable of detecting antibodies
to the SCP protein present in serum samples. This assay is based
on Luminex technology platform and utilizes spectrally unique
carboxylated microsphere coated with the SCP protein. Briefly,
the SCP-coated microspheres are incubated overnight at 4°C with
appropriately diluted serum samples, controls and reference stand-
ard serum. A phycoerythrin labeled mouse monoclonal antibody
specific to SCP is then added to the microsphere/serum mixture,
and following a 2-hour incubation at ambient temperature, the rea-
gents are transferred to an opaque filter plate, where the unbound
components are washed using vacuum through the filter plate. The
fluorescent protein coupled to the SCP monoclonal antibody allows
for the measurement of antibody bound to the antigen (SCP) coated
microspheres by the Bio-Plex reader. Signals are expressed as
median fluorescence intensities and read against a reference stand-
ard. This is a competitive assay, and the magnitude of the fluores-
cent phycoerythrin signal is inversely proportional to the amount of
Ag-specific antibodies in the sample. For cLIA, the median fluores-
cence intensity data for unknown test sera and controls were gener-
ated and converted to units per milliliter using 4-parameter logistic
© 2015 Wolters Kluwer Health, Inc. All rights reserved.
The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015 GAS Epidemiology

regression plots of the standard reference sera with assigned anti-
body titers (10,000 U/mL) using SAS 9.1 software (SAS Institute,
Cary, NC). The quantitative limits of the SCP cLIA ranged from
50.4 to 5895.5 U/mL.26
Statistical Analysis
Presence of potential risk factors for GAS was sought
including a history of prematurity, household tobacco smoke expo-
sure, number of children in the household and previous diagnosis of
streptococcal infection in a household member. A sample size of 30
participants per age-band was estimated to be sufficient to evaluate
the serum antibody mean and variance. As the study was descrip-
tive in nature, no formal sample size estimation was conducted.
With 5 centers participating and 50 participants enrolled in each
age-band, the sample size was sufficient to describe and compare
anti-SCP serum IgG antibodies in children in relation to age and
GAS carriage status. Immunogenicity and carriage data analysis
were performed for descriptive purposes. Comparisons between
groups are based on the point estimates and corresponding 95%
confidence intervals (CIs) for each individual age group.
For the immunogenicity analysis, all participants who had
at least 1 anti-SCP IgG measurement were included in the immu-
nogenicity analysis population. For the carriage analysis, all par-
ticipants who had at least 1 carriage evaluation were included in
the carriage analysis population. Carriage variables on throat swab
testing included (1) GAS-positive, (2) BHS-positive (any BHS;
GAS, GCS, GGS), (3) Non-GAS BHS-positive (GCS, GGS) and
(4) BHS-negative.
Categorical variables were compared between subgroups by
using χ2 or Fisher’s exact test as appropriate. Univariate and multi-
variable logistic regression was used to assess potential risk factors
for anti-SCP IgG seropositivity and univariate logistic regression
for GAS carriage. Associations were expressed using odds ratios
(OR) and 95% CIs.
Reverse cumulative distribution curves for anti-SCP con-
centrations were determined. Statistical analyses were performed
using SAS Version 9.3 (SAS Institute Inc., Cary, NC).27
This study was approved by the respective human research
ethics committees of the participating centers.
RESULTS
Total Study Population
A total of 542 healthy children aged 1 month to 10 years 5
months were enrolled in the study with 542 evaluable for carriage
(pharyngeal sample available) and 517 evaluable for both carriage
and serological analysis (pharyngeal swab and blood sample availa-
ble). The majority (82%) of children were Caucasian, with 4.0% of
Aboriginal or Torres Strait Islander (Aboriginal) origin (Table 1).
One center (Brisbane) enrolled a significantly higher proportion of
Aboriginal participants (10.9%; 12 of 110) than the remaining four
centers (10 of 422; 2.4%; P = 0.001)). Demographic factors includ-
ing gender, age, history of prematurity and passive smoking were
consistent across centers.
BHS isolation
Twenty-seven children (5.0%) had GAS isolated from their
pharynx with a median age of 6 years (interquartile range: 4, 9). Non-
GAS BHS was identified in 6 children. GAS-positive culture preva-
lence ranged from 0% to 12%, with the highest values in 5 (9.6%) and
9 (12.2%) year-old children (Table 2). Positive cultures were found in
each season with a predominance in spring (n = 13: summer, n = 5;
autumn, n = 8; winter, n = 1). There was some variability in enrolment
during each season at each center, although samples were collected
all year round. The only age-band showing no evidence of GAS car-
riage was the 1 year olds; however, 2 children in this age-band had
GGS identified. In total, 3 GGS isolates (from SA, WA, ACT) and 2
GCS isolates (from ACT) were identified. The carrier prevalence of
BHS was 5.7% (n = 31 of 542), with 5.0% carrying GAS (n = 27 of
542), 0.4% carrying GCS, 0.5% carrying GGS and no GBS isolates
identified in this cohort (Table 2). The highest GAS carriage preva-
lence occurred in SA (8.2%) and ACT (8.9%) with the lowest carriage
prevalence of isolates identified in children enrolled in NSW (1.8%).
All isolates received, and from which DNA was obtainable were
sequenced. Isolates (n = 20 GAS, 2 GCS) were characterized for the
distribution and diversity of SCP. All isolates contained the gene, and
the minimum sequence identity between isolates was 98%. Eleven
unique alleles were detected; there was no correlation between geo-
graphic location and SCP sequence, although numbers were small.
Risk Factors for GAS carriage
Season and age were significantly associated with GAS car-
riage on univariate analysis (Table 3). The odds of GAS carriage
were 9.5 times higher in spring than winter (P = 0.031). As age
of child increased, the rate of carriage increased; OR, 1.14 (1.00,
1.29); P = 0.05. There was a trend apparent for increased risk of
GAS carriage associated with parental smoking inside the house
[OR, 2.74 (0.89, 8.46); P = 0.078].
Prematurity
Fifty-nine infants (10.9%) were premature with a gesta-
tional age of 25–36 weeks, slightly above the proportion of preterm

TABLE 1.  Participant Demographic Characteristics and BHS Carriage Status at Each Center

Parameter Adelaide Brisbane Perth Canberra Sydney Total

Number enrolled 110 110 110 101 111 542

Median age (interquartile range) 5 (2–8) 5 (2–8) 5 (2–7) 5 (2–8) 5 (2–8) 5 (2–8)
Gender: male, % 61.8 55.5 50.9 61.4 54.9 56.8
Ethnicity: Aboriginal/ 3.6 10.9* 2.7 1.0 1.8 4.1
Torres Strait Islander, %
History of prematurity, % 9.1 15.5 10.1 7.9 18.9 12.4
GAS isolated, % 8.2 2.7 3.6 8.9 1.8 5.0
Any BHS isolated. % 9.1 4.6 4.6 10.9 1.8 6.1
No. children in household; 2.4 (1.2) 2.5 (1.2) 2.7 (1.3) 2.3 (0.9) 2.4 (1.1) 2.4 (1.2)
mean (SD)
Previous streptococcal infection 7.3 20.0* 10.9 11.9 3.6* 10.7
in household member, %
Household member smokes inside
At least 1 smoker in household, % 36.7 29.1 26.4 42.6 33.3 33.3
*Statistically significant difference in proportion compared with other 4 centers.

© 2015 Wolters Kluwer Health, Inc. All rights reserved. www.pidj.com  |  833

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Marshall et al The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015

TABLE 2.  Distribution of Isolates by Age of Participant (Years)

No BHS Isolate
Identified,
Age (yr) n GAS, N (%) Non GAS BHS, N (%) N (%)

<1 48 1 (2.1) 0 (0.0) 47 (97.9)

1 to <2 51 0 (0.0) 2 (GGS) (3.9) 49 (96.1)
2 to <3 49 2 (4.1) 1 (GGS) (2.0) 46 (93.9)
3 to <4 49 2 (4.1) 0 (0.0) 47 (95.9)
4 to <5 48 3 (6.3) 0 (0.0) 45 (93.8)
5 to <6 52 5 (9.6) 0 (0.0) 47 (90.4)
6 to <7 53 1 (1.9) 1 (GGS) (1.9) 51 (96.2)
7 to <8 46 2 (4.3) 0 (0.0) 44 (95.7)
8 to <9 49 3 (6.1) 0 (0.0) 46 (93.9)
9 to <10 49 6 (12.2) 1 (GCS) (2.0) 42 (85.7)
10 to <11 44 2 (4.5) 1 (GCS) (2.3) 41 (93.2)
Unk 4 0 0 4 (100.0)
All 542 27 (5.0) 6 (1.1) 509 (93.9)

TABLE 3.  Risk Factors Associated with GAS Carriage in Children

Variable OR Lower 95% CI Upper 95% CI P Value

Number of children in household* 1.10 0.80 1.51 0. 552

Gender female vs. male 0.90 0.41 1.98 0.794
Aboriginal or Torres Strait Islander — — — —
Peoples†
Age category ≤3 vs. > 3 years 0.39 0.14 1.03 0.058
Age in years* 1.14 1.00 1.29 0.050
Previous diagnosis household 0.63 0.15 2.76 0.547
member yes vs. no
Inside smoking yes vs. no 2.74 0.89 8.46 0.078
Born premature yes vs. no 0.56 0.13 2. 44 0.443
Season‡ 0.175
 Autumn vs. winter 6.56 0.81 53.17 0.078
 Spring vs. winter 9.52 1.23 73.73 0.031
 Summer vs. winter 8.79 1.01 76.72 0.049
*Fitted as a continuous variable in the logistic regression model.
†Model did not converge because of no ATSI children with GAS.
‡Global P value comparing all categories quoted in first row.

infants delivered in Australia (8%).28 Although prematurity was not
significantly associated (Fisher’s exact P = 0.760) with a higher
likelihood of GAS carriage, non-GAS BHS was isolated from a
greater proportion of premature children (3 of 59) than non-prema-
ture children (3 of 466; Fisher’s exact P = 0.020). Because of the
small number of children positive for GAS carriage (n = 25), pre-
maturity could not be included in the multivariate logistic regres-
sion model.
likely to have detectable SCP antibodies than children aged >3 years
(OR: 0.15; 95% CI: 0.10, 0.23, P < 0.001), and each additional child
within the household increased the odds of the enrolled participant
having detectable SCP antibodies (OR: 1.51; 95% CI: 1.21, 1.88;
P < 0.001). Children from Qld and WA were approximately twice
as likely to be seropositive compared with children from NSW,
even when adjusted for age and the number of children within
household (Table 5).

SCP Seropositivity Carriage and Anti-SCP Antibody Titers

Of the 517 children enrolled with immunogenicity data
available, 338 (65.4%) had measurable (above the lower limit of
detection) anti-SCP antibodies (Fig. 1). Geometric mean anti-SCP
titers increased with each age-band from 2 years (96 U/mL) to 7
years of age (1339 U/mL) and then plateaued (Fig. 1). Anti-SCP
titers during the first year are likely reflected maternal antibody
(Fig. 2).
Univariate logistic regression identified age, geographic
location and the number of children within the household as vari-
ables significantly associated with the presence of anti-SCP anti-
bodies (Table 4). Significant univariate predictors of seropositiv-
ity were included in a final logistic regression multivariable model
(Table 5). Age category and number of children were independent
predictors for the presence of anti-SCP antibodies in Australian
children aged 0–10 years of age. Children aged ≤3 years were less
Anti-SCP titers were significantly higher (P < 0.001) in chil-
dren with GAS-positive throat cultures (2215.8 U/ml) compared
with those with GAS-negative throat cultures (269.9 U/mL; Fig. 3).
DISCUSSION
This study provides important information on GAS car-
riage, identification of the SCP gene in isolates and seroepide-
miology data in children, with evidence of age related trends in
anti-SCP antibody and a strong association between GAS carriage
and antibody titre. Our study identified that healthy children who
carry GAS have significantly higher anti-SCP titers than GAS-­
negative children, and that carriage isolates consistently contain the
SCP gene. This finding raises the possibility that anti-SCP may be
behaving either as an acute phase reactant or potentially like antist-
reptolysin O, as an indicator of recent infection with the possibility

834  |  www.pidj.com © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015 GAS Epidemiology

TABLE 4.  Risk Factors Associated with GAS Seropostitivity

Variable Odds Ratio Lower 95% CI Upper 95% CI P Value

Number of children in household* 1.74 1.42 2.12 <0.001

Gender female vs. male 0.99 0.68 1.42 0.934
Aboriginal or Torres Strait Islander 1.84 0.67 5.08 0.237
Peoples
Age category ≤3 vs. >3 years 0.14 0.09 0.21 <0.001
Age in years† — — — —
Previous diagnosis in a household 1.48 0.80 2.77 0.214
member yes vs. no
Inside smoking yes vs. no 1.72 0.76 3.89 0.196
Born premature yes vs. no 0.91 0.53 1.55 0.705
Season‡ 0.314
 Autumn vs. winter 0.76 0.46 1.26 0.291
 Spring vs. winter 1.07 0.65 1.76 0.797
 Summer vs. winter 0.69 0.38 1.27 0.229
Site‡ 0.039
 Adelaide vs. Sydney 1.15 0.67 1.99 0.616
 Brisbane vs. Sydney 2.0 1.13 3.53 0.018
 Perth vs. Sydney 1.88 1.07 3.32 0.029
 Canberra vs. Sydney 1.03 0.58 1.84 0.922
*Fitted as a continuous variable in the logistic regression model.
†Model assumptions violated when age in years treated as a continuous predictor, so results not presented.
‡Global P value comparing all categories quoted in first row.

FIGURE 1.  Anti-SCP cLIA titers

and culture results stratified by
age (40 normal adults shown as
comparator, US data). Anti-SCP titers
were determined by cLIA. Titers
from individuals with GAS-positive
cultures are indicated in red, and
titers from individuals with non-GAS
BHS-positive cultures are indicated in
green. Geometric mean titers (GMT)
of individuals without BHS-positive
cultures are indicated.

that anti-SCP antibody may protect against infection after GAS
exposure. If anti-SCP antibody does protect against infection after
GAS exposure, it would lend support for the inclusion of SCP and/
or other GAS antigens as components of an investigational vaccine.
GAS carriage rates in our study (0–10% in each age group) were
similar to carriage rates in healthy children in other countries, such
as Croatia (6%) and Iran (11%).29,30 A previous study in 1 Australia
location (Victoria) reported carriage rates between 8% and 16%
(seasonal) in a total of 852 children aged 3–12 years. In that study,
24% of families experienced an episode of GAS pharyngitis each
year.31 Carriage of GAS, as expected, was higher than carriage of
other BHS as in other studies.29
Children 0–11 months showed lower (maternal) antibody
titers with anti-SCP titers increasing through the age-bands and
plateauing between 7 and 10 years of age. A trend was observed for
increasing carriage as age-band increased, which is consistent with
our understanding of acquisition of GAS infection or carriage.29 The
implications and clinical understanding of this carrier state remain
poorly understood. Previous studies have shown similar low levels
of SCP antibodies in children compared with adults. It is likely that
several encounters with the streptococcal antigen are required for
adequate protection; however, the threshold of ­protective antibody
is unknown.23
The presence of SCP neutralizing antibody facilitates chem-
otaxis and preserves the C5a-mediated inflammatory response,

© 2015 Wolters Kluwer Health, Inc. All rights reserved. www.pidj.com  |  835

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Marshall et al The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015

FIGURE 2.  Anti-SCP antibody titers and

positive GAS or BHS culture results from 1
to 12 months of age. Anti-SCP titers were
determined by cLIA. Titers from individuals
with GAS-positive cultures are indicated in
red, and titers from individuals with non-
GAS BHS-positive cultures are indicated
in green. Geometric mean titers (GMT) of
individuals without BHS-positive cultures
are indicated.

TABLE 5.  Multivariate Analysis for GAS Seropositivity

Variable Odds Ratio Lower 95% CI Upper 95% CI P Value

Number of children in household* 1.51 1.21 1.88 <0.001

Age category ≤3 vs. > 3 years 0.15 0.10 0.23 <0.001
Site† 0.052
 Adelaide vs. Sydney 1.24 0.66 2.33 0.497
 Brisbane vs. Sydney 2.37 1.23 4.56 0.010
 Perth vs. Sydney 2.01 1.04 3.88 0.037
 Canberra vs. Sydney 1.19 0.61 2.32 0.608
*Fitted as a continuous variable in the logistic regression model.
†Global P value comparing all categories quoted in first row.

thereby preventing establishment of infection.23 The cLIA has
been developed to preferentially identify functional antibody
through competitive binding to relevant epitopes, as opposed to
enzyme-linked immunosorbent assay, which measures both func-
tional antibody and nonfunctional low-affinity antibody.32 Further
opsonophagocytic investigation is required to correlate the cLIA
results with functional assays, such as and adhesion assays, as well
as protection against carriage or disease.
GAS-positive children had significantly higher anti-SCP
titers than GAS-negative children, suggesting carriage could be an
immunizing event. Although this finding supports a serum antibody
response to carriage acquisition, this finding is difficult to interpret
as our study was limited in its cross-sectional design with testing
at a single time point. It is still unclear as to whether anti-SCP anti-
bodies protect against carriage. Johnson et al33 conducted a longi-
tudinal cohort study and reported an increase in antistreptolysin O
antibody titers at the time of acquisition of the GAS in the carrier
state; however, the titers remained stable over the carriage period
and did not increase compared with active GAS infection.
We identified that predictors of SCP seropositivity include
age, geographic location and the number of children within the
household. This is the first study to our knowledge to demonstrate
that children with a history of prematurity are at increased risk of
acquisition of Non-GAS BHS. Our isolate sample size was small,
so this may represent a chance finding but nonetheless one that
deserves further investigation. Children with a history of prema-
turity are at increased risk of other infections, and our finding may
indicate that they are particularly at risk of disease caused by Non-
GAS BHS.
Differences in carriage rates in different locations may be
because of variations in case ascertainment or recruitment methods
at enrolling centers; however, our findings are supported by another
study that identified region as a factor associated with differences
in carriage rates.34 Several previous studies have shown intersite
and intrasite variation in rates of invasive GAS infection rates,35–37
which can be partly attributed to differences found in circulating
GAS strains and in the population’s susceptibility to these strains.3
Variations may also be because of center-to-center differences in
the prevalence of risk factors for GAS; however; in our study; this
is unlikely as age distribution, gender, ethnicity, prematurity and
smoking status of parents were similar between the centers. Our
finding that carriage rates, as for GAS infection rates,38,39 differ by
season, is supported by other epidemiological studies.31,40 Increased
isolate detection in spring and summer may reflect increased inci-
dence of GAS related infections, such as impetigo during these
seasons. Laboratory methods may also have contributed to under

836  |  www.pidj.com © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015 GAS Epidemiology

FIGURE 3.  Anti-SCP titers in GAS-positive and GAS-negative children. Anti-SCP titers were determined by cLIA and stratified
by GAS-positive or GAS-negative culture status. The geometric mean titre (GMT) of each group is indicated by an X. Anti-
SCP titers were significantly higher in GAS-positive subjects than in GAS-negative subjects.

detection of GAS. Although horse blood is used by many laborato-
ries for GAS isolation, it may not be the most optimal medium for
identification of GAS particularly when carriage is being investi-
gated and numbers of bacteria are likely to be low. For a preventa-
tive strategy to be effective against GAS, ideally, children should
be immunized before initial exposure. However, a single infection
is unlikely to induce broadly protective immunity in children, and
there are potential benefits in immunization post exposure with the
possibility of an anamnestic response to a GAS vaccine. Our study
indicates that the onset of acquisition of GAS through nasopharyn-
geal carriage or active disease commences in the second year of
life. This suggests immunization to prevent GAS infection should
ideally commence in the second year of life or earlier, despite the
higher incidence of pharyngitis in older children, to immunize
before the period of risk.
O’Loughlin et al3 estimated that a 26-valent GAS M pro-
tein-based vaccine could prevent 40–50% of cases and 50–60%
of deaths because of invasive GAS infections among children and
the elderly in the US. Development of a broadly protective vac-
cine particularly for indigenous communities where strain diversity
may be different to those of developed countries has been slow.31
Data from a trial of a new 30-valent M-protein-based GAS vac-
cine have shown that antibodies produced to vaccine serotypes are
cross-reactive with non-vaccine serotypes of GAS suggesting a
broadly protective vaccine may be achievable.41,42 The large number
of M protein variants likely required for a broadly protective vac-
cine has led many researchers to seek alternate strategies for iden-
tifying antigens to include in a GAS and/or BHS vaccine. These
approaches have used bioinformatic and proteomic approaches to
identify highly conserved surface antigens to target. These antigens
are then validated by epidemiologic studies, such as this one, and by
protection in small animal models. SCP is the second most highly
expressed protein on the bacterial cell surface (after M protein) and
is highly conserved both within GAS and across the BHS family of
species.24,25 The high level of conservation was replicated here, as
all 22 isolates tested, possessed the gene and sequence was >98%
identical. The low genetic variability of SCP means that a single
variant can be used in a vaccine. Additional surface antigens with
similar characteristics could be added to SCP to form a multivalent
vaccine that targets multiple virulence factors of the organism.
Ideally, a GAS vaccine would provide long-term protec-
tion against both carriage and active infection with broad coverage
against all circulating strains, particularly, those associated with
clinically significant disease and life-threatening complications in
children.17 However, opinion varies on this, as carriage may provide
natural boosting. These considerations are important in deciding
the scheduling of a GAS vaccine program.
ACKNOWLEDGMENTS
The authors thank all the families that took part in this study.
Helen Marshall is supported by an NHMRC Career Development
Fellowship (1016272). Stephen Lambert is supported by an NHMRC
Early Career Fellowship (1036231) and a Mid-career Fellowship
from the Queensland Children’s Medical Research Institute and the
Children’s Hospital Foundation Queensland (50025). The authors
acknowledge Dr. Ying Ying Liew and Mr. Andrew Lawrence (micro-
biological assessment) for assistance with study processes, Dr. Paul
Liberator and Mrs. Michelle Clarke for help with preparation of
the manuscript and Mr. Tom Sullivan for statistical assistance. The
authors thank Drs. Karen Prosser, Ruth Thornton and Anthony Keil
for their contribution to study conduct and microbiology assays in
Perth. This research study was an Investigator led and designed study
with funding provided by Wyeth (subsequently Pfizer) to cover study
costs. Swabs/blood samples were collected and sent to Wyeth (Pfizer)
for analysis to detect antibodies and/or β-hemolytic streptococcus
isolates. Pfizer performed serology and correlated the serology with
GAS and non-GAS BHS detection as shown in Figures 1–3.
REFERENCES
1. Carapetis JR, Currie BJ, Kaplan EL. Epidemiology and prevention of group
A streptococcal infections: acute respiratory tract infections, skin infec-
tions, and their sequelae at the close of the twentieth century. Clin Infect
Dis. 1999;28:205–210.

© 2015 Wolters Kluwer Health, Inc. All rights reserved. www.pidj.com  |  837

Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Marshall et al The Pediatric Infectious Disease Journal  •  Volume 34, Number 8, August 2015
2. Steer AC, Danchin MH, Carapetis JR. Group A streptococcal infections in
children. J Paediatr Child Health. 2007;43:203–213.
3. O’Loughlin RE, Roberson A, Cieslak PR, et al. The epidemiology of inva-
sive group A streptococcal infection and potential vaccine implications:
United States, 2000–2004. Clin Infect Dis. 2007;45:853–862.
4. Stevens DL. The flesh-eating bacterium: what’s next? J Infect Dis. 1999;179
(Suppl 2):S366–S374.
5. O’Grady KA, Kelpie L, Andrews RM, et al. The epidemiology of inva-
sive group A streptococcal disease in Victoria, Australia. Med J Aust.
2007;186:565–569.
6. Carapetis J, Robins-Browne R, Martin D, et al. Increasing severity of inva-
sive group A streptococcal disease in Australia: clinical and molecular epi-
demiological features and identification of a new virulent M-nontypeable
clone. Clin Infect Dis. 1995;21:1220–1227.
7. Carapetis JR, Steer AC, Mulholland EK, et al. The global burden of group A
streptococcal diseases. Lancet Infect Dis 2005;5:685–694.
8. Streeton CL, Hanna JN, Messer RD, et al. An epidemic of acute post-strep-
tococcal glomerulonephritis among aboriginal children. J Paediatr Child
Health. 1995;31:245–248.
9. Marshall CS, Cheng AC, Markey PG, et al. Acute post-streptococcal glomer-
ulonephritis in the Northern Territory of Australia: a review of 16 years data
and comparison with the literature. Am J Trop Med Hyg. 2011;85:703–710.
10. Steer AC, Jenney AJ, Oppedisano F, et al. High burden of invasive beta-
haemolytic streptococcal infections in Fiji. Epidemiol Infect. 2008;136:621–627.
11. Towers RJ, Carapetis JR, Currie BJ, et al. Extensive diversity of Streptococcus
pyogenes in a remote human population reflects global-scale transmission
rather than localised diversification. PLoS ONE. 2013;8: e73851.
12. Carapetis JR, Connors C, Yarmirr D, et al. Success of a scabies control
program in an Australian aboriginal community. Pediatr Infect Dis J.
1997;16:494–499.
13. Kaplan EL, Wotton JT, Johnson DR. Dynamic epidemiology of group A strep-
tococcal serotypes associated with pharyngitis. Lancet. 2001;358:1334–1337.
14. Steer AC, Batzloff MR, Mulholland K, et al. Group A streptococcal vac-
cines: facts versus fantasy. Curr Opin Infect Dis. 2009;22:544–552.
15. McNeil SA, Halperin SA, Langley JM, et al. Safety and immunogenicity
of 26-valent group a streptococcus vaccine in healthy adult volunteers. Clin
Infect Dis. 2005;41:1114–1122.
16. Hu MC, Walls MA, Stroop SD, et al. Immunogenicity of a 26-valent group
A streptococcal vaccine. Infect Immun. 2002;70:2171–2177.
17. Dale JB, Fischetti VA, Carapetis JR, et al. Group A streptococcal vaccines:
paving a path for accelerated development. Vaccine. 2013;31 (Suppl 2):
B216–B222.
18. Steer AC, Law I, Matatolu L, et al. Global emm type distribution of group A
streptococci: systematic review and implications for vaccine development.
Lancet Infect Dis. 2009;9:611–616.
19. Menon T, Lloyd C, Malathy B, et al. emm type diversity of beta-haemolytic
streptococci recovered in Chennai, India. J Med Microbiol. 2008;57(Pt
4):540–542.
20. Bisno AL, Collins CM, Turner JC. M proteins of group C streptococci isolated
from patients with acute pharyngitis. J Clin Microbiol. 1996;34:2511–2515.
21. Collins CM, Kimura A, Bisno AL. Group G streptococcal M protein exhib-
its structural features analogous to those of class I M protein of group A
streptococci. Infect Immun. 1992;60:3689–3696.
22. Richardson LJ, Towers RJ, Cheng AC, et al. Diversity of emm sequence
types in group A beta-haemolytic streptococci in two remote northern ter-
ritory indigenous communities: implications for vaccine development.
Vaccine. 2010;28:5301–5305.
838  |  www.pidj.com
Copyright © 2015 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
23. Shet A, Kaplan EL, Johnson DR, et al. Immune response to group A strep-
tococcal C5a peptidase in children: implications for vaccine development.
J Infect Dis. 2003;188:809–817.
24. Park HS, Cleary PP. Active and passive intranasal immunizations with strep-
tococcal surface protein C5a peptidase prevent infection of murine nasal
mucosa-associated lymphoid tissue, a functional homologue of human ton-
sils. Infect Immun. 2005;73:7878–7886.
25. Cheng Q, Debol S, Lam H, et al. Immunization with C5a peptidase or
peptidase-type III polysaccharide conjugate vaccines enhances clear-
ance of group B Streptococci from lungs of infected mice. Infect Immun.
2002;70:6409–6415.
26. Smith JF, Kowalski R, Esser MT, et al. Evolution of type-specific immu-
noassays to evaluate the functional immune response to Gardasil®. Hum
Vaccin 2008;4:134–142.
27. SAS Institute Inc. SAS OnlineDoc, 9.1.3. Cary, NC: SAS Institute Inc.; 2004.
28. Commonwealth of Australia. Australian Social Trends 2007; Australia’s
babies. The Australian Bureau of Statistics. Available at: http://www.abs.
gov.au/AUSSTATS/abs@.nsf/Latestproducts/04FEBEF9C81FE6BACA257
32C002077A2?opendocument Accessed December 13, 2014
29. Begovac J, Bobinac E, Benic B, et al. Asymptomatic pharyngeal carriage of
beta-haemolytic streptococci and streptococcal pharyngitis among patients
at an urban hospital in Croatia. Eur J Epidemiol. 1993;9:405–410.
30. Fazeli MR, Ghaemi E, Tabarraei A, et al. Group A Streptococcal serotypes
isolated from healthy schoolchildren in Iran. Eur J Clin Microbiol Infect
Dis. 2003;22:475–478.
31. Danchin MH, Rogers S, Selvaraj G, et al. The burden of group A strep-
tococcal pharyngitis in Melbourne families. Indian J Med Res. 2004;119
(Suppl):144–147.
32. Hawkins J, Kodali S, Matsuka YV, et al. A recombinant clumping factor
A-containing vaccine induces functional antibodies to Staphylococcus
aureus that are not observed after natural exposure. Clin Vaccine Immunol.
2012;19:1641–1650.
33. Johnson DR, Kurlan R, Leckman J, et al. The human immune response to
streptococcal extracellular antigens: clinical, diagnostic and potential patho-
genetic implications. Clin Infect Dis 2010;50:481–490.
34. Gür E, Akkus S, Arvas A, et al. Prevalence of positive throat cultures for
group A beta-hemolytic streptococci among school children in Istanbul.
Indian Pediatrics. 2002;39:569–573.
35. O’Brien KL, Beall B, Barrett NL, et al. Epidemiology of invasive group
a streptococcus disease in the United States, 1995–1999. Clin Infect Dis.
2002;35:268–276.
36. Tyrrell GJ, Lovgren M, Kress B, et al. Invasive group A streptococcal disease
in Alberta, Canada (2000 to 2002). J Clin Microbiol. 2005;43:1678–1683.
37. Lamagni TL, Efstratiou A, Vuopio-Varkila J, et al. The epidemiology of
severe Streptococcus pyogenes associated disease in Europe. Euro Surveill.
2005;10:179–184.
38. Shaikh N, Leonard E, Martin JM. Prevalence of streptococcal pharyngi-
tis and streptococcal carriage in children: a meta-analysis. Pediatrics.
2010;126:e557–e564.
39. Gunnarsson RK, Holm SE, Söderström M. The prevalence of beta-haemo-
lytic streptococci in throat specimens from healthy children and adults.
Implications for the clinical value of throat cultures. Scand J Prim Health
Care. 1997;15:149–155.
40. Gerber MA. Comparison of throat cultures and rapid strep tests for diagno-
sis of streptococcal pharyngitis. Pediatr Infect Dis J. 1989;8:820–824.
41. Dale JB, Penfound TA, Chiang EY, et al. New 30-valent M protein-based
vaccine evokes cross-opsonic antibodies against non-vaccine serotypes of
group A streptococci. Vaccine. 2011;29:8175–8178.
42. Dale JB. Current status of group A streptococcal vaccine development. Adv
Exp Med Biol. 2008;609:53–63.
© 2015 Wolters Kluwer Health, Inc. All rights reserved.