Paul T. Heath Paediatric Infectious Diseases Research Group

& Vaccine Institute,

St George’s, University of London

Perinatal GBS –

Time to change strategy

or wait for new developments?

3rd Swansea Perinatal

Symposium 2017

Pathophysiology of neonatal GBS infection Early onset (0-6d) = 70%* vs. Late onset (7-90d) = 30% (* varies according to IAP use)

EO = vertical transmission: ingestion/aspiration, contact in birth canal

20-25% women are colonized

50% of babies born to these mothers are colonized

1-2% of babies develop invasive disease

LO = vertical or horizontal transmission: parents, health care workers, equipment, breast milk, etc.

UK GBS incidence < 3 months of age

0

50

100

150

200

250

300

350

0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90

Fre

qu

en

cy

Day of presentation

2000-2001

2014-2015

Lancet 2004;363(9405):292-4 O`Sullivan et al. (unpublished)

EO: 90% on day 1: 98% signs < 12 hrs age

Year Total Incidence / 1000 live births

(95% CI)

EO Incidence (<7 days of age)

LO Incidence (7-90 days of age)

2014 0.95 (0.88-1.00)

0.57 (0.52-0.62)

0.37 (0.33-0.41)

2000 0.72 (0.66-0.78)

0.48 (0.43-0.53)

0.24 (0.21-0.28)

Bacteraemias 0-2 days of age England & Wales 2006-8

(Muller-Pebody, et al. ADC 2010)

bacteria number cumulative

Group B strep 477 40%

Other strep sp. 142 12%

E. coli 137 13%

S. aureus 75 8%

E. faecalis 37 4%

Enterobacteriaceae 35 3%

H. influenzae 34 6%

S. pneumoniae 32 3%

Bacterial Infection • Total 273 cases where a bacterial pathogen specified

0102030405060708090

87

19 13

38 37 34

19 21

5

Depani et al. PIDJ 2010

The Contribution of Infections to Neonatal

Deaths in England and Wales 2003-5: 768 infection related deaths (pathogen in 339)

2015: UK EOGBS case fatality rates by gestational age

47%

9%

3%

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

50%

EOD (UK)

<28 wk

28-36

>37

%

CFR EOGBS: 5.2% (10.6% in 2000-1, p=0.01)

Bacteria All 1st

month

2nd

month

3rd

month

GBS 50% 58% 47% 24%

E.coli 13% 15% 12% 11%

SPn 9% 6% 7% 29%

Nm 4% 2% 15% 24%

Bacterial meningitis < 3 months of age (2011, UK)

Arch Dis Child Fetal Neonatal Ed 2001;84:F85-9

Clin Infect Dis. 2014;59(10):e150-7

GBS meningitis in young infants

GBS:

86%

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

<2mo

Pe

rcen

tag

e o

f to

tal

cas

es

GBS:

78%

USA 2003-07 UK & ROI 2010-11 Cases < 2 months of age

NEJM 2011;

364:

2016-2025

Clin Infect Dis

2014;15;

59(10):50-7

GBS meningitis morbidity UK

•98 with GBS meningitis (1985 - 1987); follow-up at 5 y:

– 13% severe disability

– 17% moderate disability

– 18% mild disability

BMJ 2001;323:533-6

USA

•43 with GBS meningitis (1998 – 2006); follow-up at 6.8y (3–12):

– 19% severe impairment

– 25% mild-to-moderate impairment

Pediatrics 2012;130;e8

~ 44 - 50% have disability at 5-7 years of age

Prevention of neonatal GBS infection

Early onset (0-6d) = 70%* vs. Late onset (7-90d) = 30% (* varies according to IAP use)

EO = vertical transmission: ingestion/aspiration, contact in birth canal

20-25% women are colonized

50% of babies born to these mothers are colonized

1-2% of babies develop invasive disease

LO = vertical or horizontal transmission: parents, health care workers, equipment, breast milk, etc.

• 4 trials: 852 women

• 3 trials (500 women): IAP versus no treatment – IAP reduced incidence of EOGBS (risk ratio (RR) 0.17, 95%

CI 0.04 to 0.74, NNT to benefit 25 (14-100)

– IAP did not significantly reduce incidence of all cause mortality, GBS mortality or mortality from other infections

• 1 trial (352 women): ampicillin vs penicillin – no significant difference in neonatal or maternal

outcomes

• “IAP appeared to reduce EOGBS, but this result may well be due to bias as we found a high risk of bias for one or more key domains in the study methodology and execution”

• “There is lack of evidence from well designed and conducted trials to recommend IAP to reduce neonatal EOGBS”

Schrag SJ, Verani JR. Vaccine 2012

Strategies for IAP

How to identify which mothers should receive IAP?

– Screening vaginal/rectal culture-based strategy

– Risk factor based strategy (MMWR 1996;45:1-24)

– Rapid test in labour: real time PCR

– Combination strategies • presence of risk factors PLUS +ve culture or +ve PCR

(BJOG 2005;112:820-6; current UK trial)

strategy risk based culture based

Australia 2016 x x

Belgium 2003 x

Canada 2013 x

France 2001 x

Japan 2011 x

Hong Kong x

Italy x

Netherlands x

Poland x

USA 2010 x

UK 2012 x

Spain 2012 x

Switzerland x

New Zealand 2013 x

Denmark x

South Africa x

Risk factors:

Previous infant who had invasive

GBS disease

GBS bacteruria during this

pregnancy

Yes

Give

intrapartum

penicillin

Collect rectal and vaginal

swab for GBS culture at 35-37

weeks gestation**

GBS +ve

Offer intrapartum

penicillin

Not done,

incomplete, or

results unknown

Risk factors:

Intrapartum

temperature

38C

Membrane

rupture 18

hours

Delivery at <

37 weeks

gestation

yes

Give

intrapartum

penicillin

GBS –ve

No intrapartum prophylaxis needed

*Methods critical! low vaginal/rectal

swab, selective media.

CDC algorithm for prevention of early-onset GBS –

screening based approach

CDC algorithm for prevention of early-onset GBS –

risk factor based approach

Yes

Give intrapartum penicillin

No

No intrapartum prophylaxis needed

Are any of the following risk factors

present?

Previously delivered infant who

had invasive GBS disease

GBS bacteruria during this

pregnancy

Delivery at < 37 weeks gestation

Duration of ruptured

membranes 18 hours

Intrapartum temperature 38C

efficacy of risk based vs screening based strategies?

• no randomised controlled trials

• U.S. retrospective cohort study – >600,000 births

– RR EOD following screening based IAP

= 0.46 (95% CI 0.36 – 0.60) vs risk based IAP

N Engl J Med, 2002. 347(4): p. 233-9

Experience in countries with risk-based strategies

• New Zealand

– 0.5/1000, 1998–1999 to 0.23, 2009–2011

[Darlow et al., 2014]

• Denmark [Andersen et al 2004]

• Netherlands

– incidence rate increased

– 0.11/1000, 1987 to 0.19, 2011 [Bekker et al., 2014]

• UK

– incidence rate increased

– 0.48/1000 (0.43-0.53), 2000-2001 to 0.57 (0.52-0.62), 2014-15 • NB. mortality decreased, incidence < 2500g decreased

[O’Sullivan et al, 2016]

Why may risk-based strategies be less effective?

• not all women whose babies develop EOGBS will have risk factors • variable compliance with guidelines

– RCOG national audit 2015: “there is considerable variation in the extent to which protocols in these units reflected the national (RCOG) guidance”

• those with risk factors may not receive antibiotics in a timely fashion – 89% of women identified based on GBS carriage received IAP vs 61% based

on risk factors N Engl J Med, 2002. 347(4): p. 233-9

• UK 2015: – 44% of EOGBS cases with risk factors received IAP – median time of IAP administration was 2h before birth – 12 different antibiotic combinations were used

• New Zealand 2009-11: – 55% of cases had RF, 31% of whom received IAP

Risk factor based IAP in UK

Risk factor Number (%)

Known GBS carriage *~^ 39 (9.1%)

Previous baby with iGBS *~^ 2 (0.5%)

GBS bacteriuria *~^ 18 (4%)

Maternal fever *~^ 83 (19%)

Suspected chorioamnionitis *~^ 133 (31%)

Prelabour ROM in prems ~ 49 (11%)

Prolonged ROM >18h in prems ~^ 41 (10%)

Premature ^ 94 (22%)

Prolonged ROM >18h ^ 136 (32%)

UK national surveillance 2015:

35% had ≥1 RCOG risk factors*

41% had ≥1 NICE risk factors ~

59% had ≥1 CDC risk factors ^

% born at term with NO risk

factors: 41% CDC - 63%

(RCOG)

Why may risk-based strategies be less effective?

• not all women whose babies develop EOGBS will have risk factors

• variable compliance with guidelines – RCOG national audit 2015: “there is considerable variation in the extent to

which protocols in these units reflected the national (RCOG) guidance”

• those with risk factors may not receive antibiotics in a timely fashion – 89% of women identified based on GBS carriage received IAP vs 61% based

on risk factors N Engl J Med, 2002. 347(4): p. 233-9

– UK 2015: • 44% of EOGBS cases with risk factors received IAP • median time of IAP administration was 2h before birth • 12 different antibiotic combinations were used

– New Zealand 2009-11: • 55% of cases had RF, 31% of whom received IAP

Test accuracy

• test accuracy and natural transitions in carriage status are not well defined in the literature

• retrospective analyses of screening have found a number of cases where the mothers had a negative screening test [Puopolo et al., 2005] [Van Dyke et al., 2009]

– false negative or natural transition between screening and delivery?

Why may culture based IAP be less effective? not all women whose babies develop EOGBS will have

had a positive culture at 35-37 weeks!

Why may culture based strategies be less effective?

• high incidence of new acquisition and loss of colonization over pregnancy

– based on screening at 31–35 weeks:

• 29% of those colonized at 37+ would not have received IAP

• 13% would have received IAP but would not have been colonized at 37+ weeks

• failures of communication between laboratories and obstetric units

• failures in administering IAP

PLoS One. 2014 Jun 30;9(6)

Timing of antenatal swab

Early Human Development 2007;83:149-56

An intrapartum screening strategy?

• technical and logistic challenges

– women need to arrive in maternity units at an earlier stage of labour

– penicillin allergy: no mechanism for antibiotic susceptibility testing (vs. culture-based)

– operator issues relating to the use of rapid testing technologies in a busy setting

Rapid tests in labour

• 1st generation: identification of the GBS group specific antigen (latex agglutination, enzyme linked immune-sorbent or optical-immunotechnology or DNA-hybridization)

• good specificity (>95%), low sensitivity (33–65%)

Rapid tests in labour

• In 2000, Bergeron et al: PCR test

• Over last decade, significant advances in PCR technologies & new detection platforms for bacterial identification: new rapid real-time PCR assays for GBS detection

– vs. enriched GBS cultures:

• sensitivity & specificity 62.5%–100% & 84.6–100%

• PPV & NPV 65–100% and 92.3–100%

Performance of Nucleic Acid Amplification tests (NAAT) compared with enriched and non-enriched group B

streptococcus culture

Rapid tests in labour: desirable features

– Sensitivity and specificity >90% and >95% respectively

– Fully automated processing with integrated internal controls, full traceability of the results and minimum maintenance

– Easy to perform and interpret results by delivery staff with a minimum of training

– Short turnaround time (< 1 hour)

– Availability 24 hours a day, 7 days a week

• Improvement: detection of mutations a/w resistance to clindamycin

Increase in IAP exposure:

USA

NEJM 2009;360:2626-36

IAP use: the possible negative aspects

• antimicrobial resistance (AMR) – International efforts to encourage antimicrobial

stewardship to conserve effectiveness of antibiotics – resistance to clindamycin is now common (UK: 3-17%) – resistance to penicillin is very rare [Chu et al., 2007] [Kimura et al., 2008]

• long term effects on infant gut flora – obesity, diabetes, ……

• increase in culture negative GBS sepsis or non-GBS sepsis

• maternal anaphylaxis – UKOSS 2012-2015: 37 cases = 1.6/100,000 (50% a/w

antibiotics) (www.npeu.ox.ac.uk)

IAP use: the possible negative aspects

Impact on infant gut flora

• Bifidobacterium – compared to control infants, significantly lower mean colonisation in 6-7 day old infants whose mothers were treated with IAP for GBS prevention

[Aloisio et al. 2014] [Corvaglia et al. 2016]

• Lactobacillus – lower transmission in mother-infant pairs treated with IAP for GBS prevention compared to controls immediately after birth & at 30 days [Keski-Nisula et al. 2013] [Arboleya et al 2015]

Lancet Infect Dis. 2014;14(11):1083-1089

Pediatrics 2016;138(6):e20162013

Neonatal EO E coli sepsis: trends in incidence and antimicrobial resistance in the era of IAP

N Engl J Med 2002; 347(4): 240–247

IAP is not a risk factor for EO Ecoli sepsis

0 = no screening, 1 = swab based screening, 2 = risk based screening, 3 = vaccination

JAMA 1993;270:1442-8

UK rates

Total costs of GBS by prevention strategy

Prevention of EOGBS in UK: modelling a culture-based strategy

700,000 women offered screening test at 35-37 weeks

140,000 women will be colonized & offered IAP

…to prevent 330 babies* with invasive disease

i.e. 0.25% of those born to colonized women

2000 women screened to prevent 1 case of EOGBS *550 EOGBS; 40% have risk factors & offered IAP

Perception’s a funny thing

• In countries with low(er) GBS incidence, concerns related to costs, logistics, medicalisation of labour and drawbacks associated with exposure to IAP are not positively balanced by the perceptions of effectiveness in implementing a culture-based screening-based policy…..

What do pregnant women want?

• Does culture-based screening enhance or override choice, increase or decrease the medicalisation of pregnancy & labour or improve or worsen the quality of a woman’s experience of childbirth?

Opportunistic, risk-based or universal: choosing an appropriate screening strategy

for Group B Strep in pregnancy?

a clear and simple national screening strategy is better than none--or several….

consistent adherence to a strategy is the most effective way of minimising the risk of EOGBSD, be it culture-based or risk-based..

national guidelines should include clear audit criteria and an enhanced audit program

RCOG GREENTOP UPDATED GUIDELINES OUT TO CONSULTATION

research is still required to fill current gaps in the evidence base

Research needs

• Exploring (qualitative) factors that affect adherence to guidelines

• Defining better markers of risk

• clinical / demographic / genetic risk factors, biomarkers, pathogen-related factors

• Validate and assess utility of new rapid tests in labour (inc. new targets)

Perinatal GBS –

Time to change strategy

or wait for new developments?

3rd Swansea Perinatal

Symposium 2017

AND

Opportunistic, risk-based or universal:

choosing an appropriate screening strategy

for Group B Strep in pregnancy?

or vaccination to prevent GBS?

Vaccination against GBS - rationale

• In 1930s Lancefield demonstrated that protection against GBS infection in mice could be achieved using CPS-specific rabbit sera

(J Exp Med 1938;67:25-40)

• Low levels of naturally occurring maternal antibody to CPS correlate with susceptibility of the neonate to GBS disease

(N Engl J Med 1976;294:753-6)

•Vaccine

targets:

– capsular

polysaccharide

(CPS)

– surface

proteins

Currently available strategies (IAP - based) are satisfactory (in some countries)…but short-term…. and not realistic for many countries → as with other significant causes of childhood bacterial meningitis, vaccination provides best prospect for prevention

Development Status of GBS Vaccine Candidates

Heath PT. Vaccine. 2016 Jun 3;34(26):2876-9.

Newborn antibody concentrations 49–79% of maternal levels

0

2.5

5

7.5

10

GMC

(ug/ml)

30-3

2 w

k

deli

very

cord

in

fan

t 1 m

on

th

infa

nt

2 m

on

ths

GBS conjugate vaccines in

pregnant women

Licensure of a GBS conjugate vaccine?

• Large-scale efficacy trial in pregnant women – >60,000 pregnant women (using invasive disease as endpoint)

OR

• Serological correlates of protection

– 90% reduction when maternal III IgG > 0.5 ug/ml – 88% reduction when Ia IgG >5 ug/ml

J Infect Dis 2001;184:1022-8

+ extensive post marketing surveillance re effectiveness /safety i.e. UK MenC & MenB models

JAMA 2008;299:2056-2065

Pregnant

women:

0.29 / 1000

in 1993

0.23

in 1998* (P<0.03)

*NEJM 2000;342(1):15-20

Pregnancy associated GBS

Pregnancy associated GBS ABCs 1999-2005

• 0.12 /1000 live births; median age 28y

• where pregnancy outcome known:

– 61% spontaneous abortion or stillborn

– 5% live-born infants with clinical infections

– 4% induced abortions

– 30% infants without apparent illness JAMA 2008;299:2056-2065

There are also “known unknowns”…….

• Contribution of GBS to prematurity?

• Contribution of GBS to stillbirths?

O`Sullivan et al. (unpublished)

Online survey results McQuaid, F. et al (1013 women)

• 2/3 of women aged 18-44yrs don’t know what GBS is

• However they are open to the idea of antenatal vaccines

…and taking part in vaccine studies in pregnancy

0 10 20 30 40 50 60 70 80 90 100

As part of a research study, previouslytested in 500 pregnant women

As part of a research study, previouslytested in 5000 pregnant women

Licensed, not specificallyrecommended

Licensed and routinely recommended

Likely Don't know Unlikely% respondents

How likely

would you be

willing to

receive a GBS

vaccine during

pregnancy if it

were…?

79%

52%

43%

32%

0 20 40 60 80 100

As part of a research study,previously tested in 500

pregnant women

As part of a research study,previously tested in 5000

pregnant women

Licensed, not specificallyrecommended

Licensed and routinelyrecommended

Likely

Don't know

Unlikely

% respondents

UK National Screening Committee

• rationale:

– the distribution of disease: preterm babies are unlikely to be affected by screening

– limitations in the evidence relating to the effectiveness of IAP

– screening test accuracy and performance

– concern about the expansion of the use of prophylactic antibiotic into a predominantly low risk population

• A lack of data on long term outcomes from EOGBS is likely to remain a limitation in similar assessments until more comprehensive epidemiological data on EOGBS related morbidity is available.

GBS: UK research questions What is the optimal strategy for administering IAP in UK women at risk of EO GBS?

enhanced risk-based strategy

rapid, better targeted swab-based strategy

What is the burden of GBS-related stillbirths & prematurity in the UK?

How can we expedite licensure & implementation of an effective antenatal vaccine in the UK?

support development of serological correlates of protection through collection of case / control

neonatal & antenatal sera and standardisation assay methods

establish trial network to support early phase immunogenicity (“bridging”) & safety studies

prioritise antenatal GBS vaccines in JCVI / DG “horizon scanning” & preparatory work

optimise surveillance

develop UK infrastructure for vaccine effectiveness & safety evaluation

• Establish: reference ELISA (& functional assay) protocols, standardised

reference serum collection, reference ranges for ST-specific antibody

• Prospective studies in diverse settings to validate current serological correlates

• Randomised vaccine trials in pregnant women:

– safety, immunogenicity, placental transfer, persistence, colonisation,

interference, breast milk

– & in different populations (to allow bridging)

• An efficacy trial with clearly defined endpoints encompassing EO and LO

disease (culture proven and clinical), colonisation, stillbirths, prematurity and

asphyxia would close many of the knowledge gaps…..

• Vaccine trials in HIV infected pregnant women: Immunogenicity, placental

transfer, persistence (with different schedules)

• Modelling: a population specific, optimised vaccination strategy

• Establish: robust surveillance of GBS strain population: pre & post vaccination

Research Gaps in immunology / biology: a summary….

Global GBS serotype distribution, 1980-2011

Lancet 2012 Vol 379, February 11