GBS vaccines in the UK: a round table discussion

Epidemiology and Surveillance

Burden of invasive GBS disease in the UK – Rebecca Guy

The first presentation was given by Rebecca Guy, principal scientist in epidemiology at the UK Health Security Agency (UKHSA) outlining the current burden of invasive GBS (iGBS) disease in the UK. Several data sources were used, including the UKHSA routine lab surveillance data of GBS isolated from sterile sites in England (SGSS), data from the Staphylococcal and Streptococcal Reference Service (SSRS), information from the NHS demographic batch services (DBS) from the NHS Spine and hospital episode statistics (HES). Some of the data presented were unpublished.

iGBS disease incidence trends in England from 2017 to 2021 remain broadly stable across this period at around 5 per 100,000 population in all ages. Incidence trends consistently differ between age groups; the highest incidence rate is in under 1-year olds at 90/100,000 population, followed by the over 75-year-olds at 14/100,000 population.

Infant iGBS infection rates have increased markedly between 1996 to 2021, for both EO and LO disease. Infant disease rates reached 0.95/1000 livebirths in 2019. There was a 40% increase in EOD between 2012 and 2020 (0.37 and 0.53/1000 livebirths respectively). During the two periods of enhanced surveillance in the UK and Ireland in 2000/01 and 2014/15, there was a comparable increase in incidence, suggesting slight under-ascertainment by routine surveillance data.

In infant iGBS from 2010 to 2021, capsular serotypes from England, Wales and Northern Ireland (NI) continue to differ between EOD and LOD. EOD demonstrates more serotype diversity, with type III still predominant but declining alongside decreased serotype Ia and increased IV in recent years. Looking cumulatively at infant iGBS isolate serotypes from sterile sites 2017 to 2021, 73% of LOD are due to serotype III but just 42% of EOD. 95% of infant disease isolates from 2017 to 2021 would be covered by the hexavalent vaccine currently in clinical trials. It is noted that this data represents the subset of cases referred to SSRS. Improvements to microbiology surveillance would be beneficial prior to a vaccine being launched.

Antimicrobial resistance (AMR) is increasing in England in iGBS cases. In 2021, 40% of isolates from invasive disease in under 1-year olds had erythromycin, and 36% clindamycin resistance. Despite the change in IAP guidance in 2017 reducing use of clindamycin, this trend has continued. Improved surveillance is important ahead of vaccine introduction to assess whether trends are of emerging resistance or expansion of serotypes.

Reporting on infant iGBS infections by ethnicity in England, 2019 to 2020, the highest disease rates are in Black infants (1.1/1000 livebirths) and lower in white (0.81/1000 livebirths) and mixed ethnicity infants (0.6/1000 livebirths). The infection rate in Black infants is therefore 36% higher than in white infants. There is also significant variability between Asian infant subgroups. The same findings apply to maternal ethnic variation in iGBS disease; iGBS is highest overall in the mothers of Black infants (0.69/1000 livebirths) at more than twice the rate of white mothers (0.29/1000 livebirths). iGBS in England by deprivation decile from 2015 to 2021, for all ages, demonstrates a consistent trend indicating cases are reported more frequently from deprived areas. One previous study with UKHSA at a London hospital showed differentials in carriage rates between ethnicities²⁷ which may provide part of the explanation but further work is needed to understand the whole picture and is being undertaken in a partnership between University of Nottingham and UKHSA.

The group discussed UKHSA’s genomic work on GBS looking at potential linkages between cases. Although LOGBS has traditionally been thought of as individually acquired community cases, the genomic work showed that 1 in 12 isolates were linked in hospital clusters and so it can also be thought of as a hospital acquired infection (HAI).

The Scottish perspective on current GBS surveillance – Lynne Rush

Lynne Rush gave the Scottish perspective. There is not currently enhanced disease surveillance for GBS in Scotland. Isolates are received via the Scottish microbiology reference laboratories, but these are not linked to clinical data. This would need to be addressed in a systematic way, preferably in line with the rest of the UK. Public Health Scotland (PHS) have good connections with the MatNeo Data Hub in partnership with the Scottish Perinatal Network, Healthcare Improvement Scotland, Scottish Government and National Records Scotland which will be a valuable framework to work with going forwards.

In terms of vaccine coverage data, historically antenatal vaccine uptake has been challenging to record but Scotland has recently been through the Vaccine Transformation Programme leading to major changes including use of the new Scottish Linked Pregnancy and Baby Dataset (SLiPBD), the new Vaccine Management Tool and changes to how vaccines in pregnancy are captured.

PHS has a robust system for monitoring vaccine safety, with weekly adverse events reported from the NHS board of immunisation co-ordinators discussed weekly at a vaccine safety team meeting. In terms of cost-effectiveness studies, the Scottish birth cohort is approximately 50,000 – a sample size too low to power a stand-alone study for Scotland and therefore being part of a UK-wide trial would be welcomed.

Proposed surveillance of iGBS disease in British and Irish infants <90 days – Konstantinos Karampatsas

A presentation by Konstantinos Karampatsas, clinical research fellow at SGUL, introduced a submitted proposal for a new enhanced surveillance study looking at iGBS disease in infants < 3 months of age in the UK and Republic of Ireland. The most recent enhanced surveillance for GBS in the UK was the British Paediatric Surveillance Unit (BPSU) study in 2014-15.¹⁴ 856 infants were identified over 13 months giving an incidence of 0.94 cases/1000 livebirths and 53 infants died meaning a case fatality rate (CFR) of 6.2%. Serotype III was the most prominent serotype in both EOD and LOD (60% of cases) followed by serotype Ia (17% of cases). One of the most important conclusions from this study was that the incidence of GBS had significantly increased from the previous BPSU study in 2000 for both EOD and LOD, which means that the burden of EOGBS has increased despite the national guidelines introducing a risk-based policy for IAP.

The need for a new surveillance study was highlighted: it will provide baseline information ahead of both a test-based screening and vaccine programme to accurately assess their impact. It will describe the current prevalence of clinical risk factors e.g., prematurity; and will help to identify whether at-risk groups will be covered by a vaccine during pregnancy. The study will also describe any changes in the incidence of iGBS and assess the effect of the implementation of the current IAP guidelines. Additionally, it will help describe serotypes circulating to bolster confidence that the vaccines in development will offer adequate cover.

The proposed study design uses the BPSU network in collaboration with microbiology reference laboratories and public health agencies in the UK and Ireland. The team will also collaborate with the UK Obstetric Surveillance System (UKOSS), raising awareness amongst obstetricians and midwives and helping to capture critical data around pregnancy and birth. The team are aiming to consult with PPI organisations to promote transparency and accountability as a non-consent model will be used.

The definition of cases at present is GBS identified via positive culture or polymerase-chain reaction (PCR) test from a normally sterile site. Engaging with hospital laboratories will be key to the success of this project, especially given that only 45% of all isolates are currently sent to the reference laboratories in the UK. The BPSU also recommended an additional timepoint in the study with a follow up at 2 years of age to assess long-term morbidity of iGBS.

A Phase 1 application was accepted in December 2022 and the Phase 2 application is being updated before approval, with a plan to meet shortly with all laboratories and agencies to discuss logistics. The study will commence within 12 months of phase 2 approval.

Epidemiology and Surveillance: Knowledge Gaps Identified and Next Steps

Group discussion identified epidemiology and surveillance knowledge gaps in the context of a GBS vaccine programme for the UK. Table 1 details these knowledge gaps as well as next steps to begin to address these.

Progress Towards a GBS Vaccine in Pregnancy

GBS Vaccines – Paul Heath

A presentation from Paul Heath, Professor of Paediatric Infectious Diseases at SGUL provided an update on GBS vaccine development. An effective GBS vaccine could prevent an estimated global 127,000 EOGBS cases (UR: 63,300 to 248,000), 87,300 LOGBS cases (UR: 38,100 to 209, 000) averting 31,100 infant deaths (UR: 14,400 to 66, 400) and 17,900 cases of moderate and severe neurodisability (UR 6,380 to 49,900).²⁰

There are two main GBS vaccine targets – the capsule and it’s 10 different serotypes, and several conserved outer membrane proteins targets. The global GBS capsular serotype distribution is important because one of two current leading vaccine candidates is a serotype-specific vaccine, covering the six most frequent serotypes. Serotype replacement is therefore a theoretical possibility, as has been seen with pneumococcal conjugate vaccines, although perhaps less likely for GBS.

To date, vaccine candidates have been developed and trialled by multiple companies but the two most advanced currently are by Pfizer and MinervaX. The Pfizer vaccine is a six-valent polysaccharide protein conjugate vaccine¹⁸ whilst MinervaX have an Alpha-like protein (Alp) based vaccine.¹⁹ Both manufacturers are in phase II trials in pregnant people and are currently planning phase III trials, grappling with the issues around route to licensure: primarily, large sample size requirement.

A 2021 study of the polysaccharide conjugate vaccine demonstrated immunogenicity and persistence against 6 serotypes in non-pregnant adults as part of a dose-finding study.³⁰ This study ultimately chose a 20mcg per serotype dose, finding the aluminium phosphate adjuvant not necessary for immunogenicity. The protein vaccine is based on the Alp proteins which are conserved across nearly all strains and is being tested in pregnant participants. It uses an aluminium hydroxide adjuvant.

Some of the issues to consider ahead of a vaccine programme are: reactogenicity and immunogenicity of co-administered vaccines in pregnancy (for example dTaP/IPV, COVID-19, RSV, influenza and tetanus), optimal timing of vaccine(s) in pregnancy as placental transfer is maximal in the 3^rd trimester yet premature infants experience the highest incidence of GBS, attitudes towards antenatal vaccination. Other issues include impact on colonisation and the persistence of antibodies in infants and also in adults into future pregnancies: will this vaccine be given in every pregnancy? Finally, any interference with primary immunisations must be evaluated, given that pertussis and polio vaccines in pregnancy can affect infant responses.³¹

Serocorrelates as the basis for conditional licensure of vaccines in the UK – Kirsty Le Doare

A presentation from Kirsty Le Doare, Professor of Vaccinology and Immunology at SGUL, gave an overview of the serocorrelates route to GBS vaccine licensure. There are three potential approaches to vaccine licensure currently; the first, the classic approach, is a phase III trial. The second approach is an immunological endpoints trial embedded within a classic phase III trial. The third option is conditional approval based on an immunological endpoint followed by a phase IV clinical effectiveness trial. At present, the second two options are the most likely routes for a GBS vaccine. To demonstrate efficacy, a larger sample size is needed than would generally be practical for a phase III trial in pregnancy.²² Even in a high disease burden country, with a disease incidence of 1.0-2.0/1000 livebirths, for a 1:1 randomised, controlled GBS clinical vaccine efficacy trial, 30-62,000 participants would need to be enrolled to detect vaccine efficacy of 80%.²³

A basis for licensure based on serocorrelates has been used previously for pneumococcal vaccines where an aggregate immunoglobulin G (IgG) serocorrelate value (the amount of antibody that protects an infant from invasive disease) that covered all serotypes was developed based on three clinical efficacy trials and agreed by consensus to be 0.35 μg/ml.³² Importantly, the way forward for licensure was having a standardised assay. A similar approach was taken for meningococcal vaccines where serocorrelates moved the field forward. The premise behind the correlate of protection (COP) is that the birthing-parent has antibodies to GBS post-vaccination or post-exposure to GBS. IgG is passed via the placenta to the fetus and if passed in sufficient amounts, the birthing-parent and baby are protected against disease in either pregnant adult, fetus or infant.

Carol Baker has worked extensively on GBS COP. Baker has spent much time on potential IgG threshold values; her studies from the early 1990’s looked at a tetanus toxoid conjugate vaccine against two of the common GBS serotypes in the USA, serotype Ia and Ib.³³ These studies found that the percentage of people with IgG >1 μg /ml, when examining vaccine versus placebo, was much higher in those who were vaccinated with >90% of non-pregnant adults achieving a titre of >1 μg /ml after vaccination. Studies in the UK examining antibody levels post-natural infection found that infants with antibody concentrations of >2 μg/ml were more likely to be protected against iGBS disease when compared to infants with lower antibody concentrations.³⁴ These results have been replicated in a seroepidemiological study from Europe which examined colonised versus non-colonised women and found that those colonised had higher antibody concentrations for all the major serotypes.³⁵ There have subsequently been several modelling studies in recent years including further work from Baker³⁶ which took a Bayesian approach to predict how much antibody is sufficient, with the overall combined estimate that antibody concentrations >1 μg /ml had an estimated 90% protection against serotypes Ia and III, and 70% protection against serotype IV. COP have also been examined and proposed for the AlpN antibodies; relevant for the protein-based vaccine that is now in Phase II trials. These studies predict an estimated 90% reduction in disease with concentrations of 0.4μg/ml of Rib-N IgG and Alp1-N IgG concentrations of 0.11 μg/ml.³⁷

One factor advancing this area more recently has been the development of the international GBS assay standardisation consortium, which SGUL leads. The three objectives of the consortium are:

A 2021 International Symposium on Streptococcus agalactiae Disease (ISSAD) report presented data showing good assay correlation between all laboratories which has now enabled larger studies to go forward. For example, the iGBS3 study in the UK, a study in the USA examining 150 dried blood cards with 3:1 controls from prospective surveillance in newborns performed via the CDC, a large trial under way in South Africa and Uganda and additional work via SGUL (the EDCTP PREPARE study) collecting isolates from around Europe and African countries. All these studies will contribute to defining the COP that can be used for conditional licensure.

Potential cost effectiveness of GBS vaccines – what more do we need to know? – Professor Caroline Trotter

Professor Caroline Trotter, Epidemiologist and Health Economist at the University of Cambridge and Imperial College London presented on the cost-effectiveness of a GBS vaccine in the UK. A cost-effectiveness analysis modelling study was performed in 2018²⁶ which concluded that a GBS vaccine in pregnancy would be cost effective, even at a relatively high estimated threshold price. The model estimated cost-effectiveness at £54 per maternal GBS vaccine dose at £20,000/QALY or £71 per dose at £30,000/QALY. There were no barriers shown to cost-effectiveness in this study.

Data from this and other previous studies^20,26 give a strong starting point for the Joint Committee on Vaccination and Immunisation (JCVI) to make robust recommendations to the Department of Health (DOH). A significant difference in cost-effectiveness since 2018 is unlikely. Updating the working economic model for the UK with new epidemiological, surveillance or cost study data will be helpful. Inputs to update include: infant disease incidence and CFR, expected rates of sequelae, costs of acute and long-term care, litigation costs, vaccine details as known and other important outcomes such as stillbirth, preterm birth reduction and maternal sepsis and death prevention.

An element which has changed since 2018 is a more crowded pregnancy vaccine space leading to increased competition. In the previous paper,²⁶ inclusion of maternal deaths and in particular stillbirths, increased the vaccine threshold price. These were not included in the base case as maternal deaths are very uncommon and stillbirths are not technically included within the National Institute for Health and Care Excellence (NICE) framework for cost-effectiveness analysis. Stillbirth was also demonstrated as an important factor in the global cost-effectiveness analysis.²⁰ The economic handling of stillbirths may therefore be an avenue for further research; the UK could put forward both cost-effectiveness arguments for comparison, with and without value of life years lost to stillbirths, using standard life expectancy. Although relatively small in number, inclusion of intrapartum GBS related stillbirth in the model would likely impact the overall cost-effectiveness.

GBS vaccines for maternal immunisation to protect against invasive disease in early infancy: Regulatory clinical considerations – Mair Powell

A presentation was given by Mair Powell, Senior Clinical Assessor at the Healthcare Products Regulatory Authority (HRA) in Ireland, chair of the European Medicines Agency (EMA) Vaccine Working Party, member of the EMA Scientific Advice Working Party and member of the EMA Infectious Diseases Working Party.

The factors affecting time to first licensure for a vaccine were outlined: firstly, what will be the basis for demonstrating or inferring efficacy. Pre-licensure efficacy trials may not be needed if: 1. there is an agreed immune COP, or an agreed threshold value to enable prediction of protection from an immune response, or 2. an efficacious, licensed vaccine already exists, with a prior efficacy/effectiveness trial, and the new candidate can be shown to elicit similar immune responses to allow immunobridging. If neither of the above scenarios exist, an efficacy trial will be required if this is feasible.

If an efficacy trial is not feasible, perhaps in the case of a rare, sporadic, episodic or small outbreak of disease then several approaches can be considered. These include: deriving a threshold level of immune response for protection, from a non-clinical model that can be applied to humans, or deriving a threshold value from natural protection, using serological data and disease surveillance. Lastly, human challenge studies can be considered, where appropriate.

Another consideration is the size of the pre-licensure safety database; generally, for a new vaccine, this will be in excess of 3000 persons exposed to the final vaccine formulation and dose regimen. However, this number is negotiable taking into consideration factors such as the vaccine content and platform. Risk management plans and periodic safety update reports are now mandatory in the UK, EU and US. Specific safety issues may need to be addressed either via a large, pre-licensure safety database, and/or a post authorisation safety surveillance (PASS) study.

Considering effectiveness studies, it is recognised that these are unlikely to be conducted by the marketing authorisation holders (MAH) themselves. It is requested that the MAH engages with relevant public health bodies, working with them to develop a plan for a vaccine effectiveness study.

In summary, the data requirement for vaccine licensure and post-licensure is determined on a case-by-case basis. EU regulators have agreed that a pre-licensure efficacy study for a GBS vaccine may be waived if a sponsor, academic body, consortium or otherwise can propose a putative COP or threshold value. Different approaches to identifying putative COPs or threshold values are being taken to date by sponsors, considering both EOD and LOD as well as individual serotypes.

Summary of current GBS vaccine pre-licensure considerations – Paul Heath

If a putative COP can be established to allow inference of vaccine protection, this might suffice for initial approval followed by a post-approval vaccine effectiveness study. Such a study might include secondary endpoints in addition to a primary endpoint based on invasive infant disease.

A pre-licensure efficacy study is probably not achievable as a conventional disease-endpoint, phase III, placebo-controlled RCT. This is because relatively low disease incidence mandates a large sample size to evidence efficacy, participant recruitment during pregnancy means follow up of at least 6-12 months with an 8-12 month enrolment period and at least 200 study sites would likely be required. This study would take several years and significant investment to complete, resulting in a delay to vaccine availability. It is possible that one of the two vaccine manufacturers might pursue a large phase III efficacy trial as this remains the gold standard, but for the other, the approach would likely be a conditional licensure approach.

A conditional licensure approach, based on a putative COP, will still require large numbers of pregnant participants, but is less costly than a pre-licensure, phase III clinical trial. Examples of vaccines approved via a COP route include the Meningococcal C, Meningococcal B and Meningococcal ACWY vaccine.

Progress Towards a Vaccine: Knowledge Gaps Identified and Next Steps

Group discussion identified further knowledge gaps in the area of progression towards a GBS vaccine programme in the UK. Table 2 details these knowledge gaps as well as next steps to begin to address these.

Advocacy and Engagement with Stakeholders

A GBS vaccine in pregnancy programme: the midwifery perspective – Vanessa Greening and Emma Eccleston

A presentation given by Vanessa Greening and Emma Eccleston, Research Midwives at SGUL Vaccine Institute, outlined some of the current barriers midwives might face in supporting pregnant people around GBS and GBS vaccines. Midwives may not always discuss GBS with pregnant people due to lack of time, staffing resource and confidence. Although very aware of the burden GBS places on maternity and paediatric services, midwives may lack specific knowledge of GBS incidence, mortality and morbidity rates.

Effective counselling on vaccines in pregnancy can be challenging; people may be hesitant discussing vaccines and when midwives lack confidence due to inadequate depth of training, their concerns may not be addressed. A study in 2013 asked London midwives their views on vaccination in pregnancy via an anonymous, online survey.³⁸ Just 25% of 266 respondents felt adequately prepared for this role and 69% of respondents agreed with the policy of vaccination against influenza in pregnancy.

Interestingly, many midwives themselves may be vaccine hesitant. UKHSA data for the 2021/22 seasonal influenza vaccine offer to frontline healthcare workers (HCWs) showed 75% uptake for nurses at St George’s Hospital,³⁹ but only 40% uptake amongst midwives at the same trust (local, unpublished data). Similarly, the 2013 survey study found seasonal influenza vaccine uptake rates in midwives of 43%, with non-uptake reasons including doubts about necessity, safety and effectiveness.³⁸

What do parents need to know now? by Group B Strep Support – Jane Plumb

Jane Plumb, Chief Executive Officer and Founder of Group B Strep Support (GBSS) charity reported on an unbranded survey GBSS conducted in April 2021, partnering with Bounty, to determine what information families already had about GBS in pregnant people and newborns. This was circulated online; the majority of respondents self-reported as pregnant people or having a child younger than 2 years of age.

3500 responses were obtained, and the results were summarised. 87% of respondents had heard of GBS, compared with a previous GBSS survey via a pregnancy and birth magazine when only 10% of respondents had. 54% of respondents had heard about GBS through their own personal experiences or via friends and only 14% from their midwives, with others citing pregnancy books, leaflets, websites, and social media. This demonstrates how information on GBS may come from non-academic, non-medical sources.

Importantly, 73% of respondents wanted more information on GBS, responding that they had not had ‘enough’ or ‘any’ information on the disease. Since 2017, the Royal College of Obstetrics and Gynaecology’s (RCOG) recommendation has been that all pregnant people should be provided with information on GBS; evidently, this target is not being met and more work needs to be done to provide access to resources on this infectious disease.

Almost all (99%) respondents felt that pregnant people should be informed about GBS by their doctor or midwife during pregnancy, 93% thought the UK should introduce a national screening programme for GBS in pregnancy (like other developed countries), selecting ‘extremely’ or ‘very important’ for screening policies and practices to be put in place. Finally, 90% of respondents answered they would be ‘very’ or ‘fairly’ likely to accept a GBS vaccine during pregnancy.

More recently, GBSS and charity partners conducted a repeat survey, advertised to those of child-bearing age; 321 responses were received at the time of this presentation in 2023 and data are both in progress and unpublished. When asked how important they felt different aspects of a GBS vaccine were, key concerns from respondents were protection of the baby against GBS infection and any side effects for the baby.

What will GBS3 tell us and how do we use these results when thinking about GBS vaccines – Kate Walker

Kate Walker, Professor of Obstetrics at Nottingham and Co-Chief Investigator of the GBS3 trial, gave an overview of the trials current progress. The GBS3 trial is a two-arm, parallel, cluster randomised trial of GBS testing strategies with an economic and acceptability evaluation. The sample size comprises 320,000 participants. Sites are randomised to one of two arms: routine testing or a risk-factor based strategy (usual care), with the routine testing sites further sub-randomised to either testing via intrapartum rapid PCR tests for GBS or via antenatal enriched culture medium (ECM) testing at 35-37 weeks gestation. Data is collected for 12 months with no individual data collected, except for a subset of 8000 participants. Aside from verbal consent for any swab taken, there is no written informed consent for the trial. The primary outcome is all-cause early neonatal sepsis. Data collection will continue until end of March 2024 with results expected Summer 2025.

The GBS3 trial results will remain relevant even with a GBS vaccine available as there will always be a requirement for GBS testing, for example for those who are unable to receive or access a vaccine. It was noted that during the COVID-19 pandemic, vaccine uptake in pregnancy was around 60% with just 30% uptake amongst Black pregnant people and 38% amongst the most deprived groups.

What have we learnt from vaccination in pregnancy with COVID-19 vaccines? – Asma Khalil

Asma Khalil, Professor of Fetal Maternal Medicine at SGUL presented on lessons learned from rapid rollout of a new vaccine in pregnancy during the COVID-19 pandemic. Approval of the first COVID-19 vaccine by the MHRA was in December 2020 but the JCVI did not recommend offering a COVID-19 vaccine in pregnancy until April 2021. Initial uptake was low, with figures at 2.8%. RCOG issued strong recommendations to offer vaccination in pregnancy in July 2021 followed by an intense campaign by RCOG and the government. Uptake improved following this campaign rising to 60% by January 2022.

Looking at COVID-19 vaccine uptake and hesitancy in pregnancy multiple studies made similar findings. An SGUL study published in July 2021²⁸ showed that approximately one in three pregnant people at St George’s hospital were accepting a COVID-19 vaccine in pregnancy. This SGUL study also examined the determinants of vaccine uptake. Multivariate regression analysis showed that there were three independent predictors of vaccine uptake: African-Caribbean ethnicity with an odds ratio (OR) of 0.27 (p value=0.044); those in the fifth (lowest) Indices of Multiple Deprivation (IMD) quintile had an OR 0.1 (p=0.003); and those with pre-gestational diabetes were 10.5 times more likely to get the vaccine (p=0.014). The ethnic group and socioeconomic disparities were demonstrated again in surveillance data from both England²⁹ and Scotland.⁴⁰ The role of healthcare professionals is an important determinant of vaccine uptake, particularly the advice given by midwives and obstetricians.

One of the key issues in vaccine implementation was addressing any safety issues quickly for pregnant individuals, particularly when a side effect becomes associated with a vaccine. From clinical practice experience, safety of the COVID-19 vaccine was the predominant concern to pregnant people, rather than efficacy.

Despite the RCOG’s efforts calling for pregnant people to be included in randomised-controlled trials (RCTs), it proved to be extremely difficult given that large numbers were needed, safety data was not yet available and sponsors were not always willing to recruit pregnant participants. There have been three RCTs examining the COVID-19 vaccine in pregnancy, none of which have reported data yet. An approach in future that doesn’t insist upon RCTs for vaccination in pregnancy is vital for timely rollout of antenatal vaccines.

Advocacy and Engagement with Stakeholders: Knowledge Gaps Identified and Next Steps

Group discussion identified knowledge gaps in advocacy and education for a GBS vaccine programme in the UK. Table 3 details these knowledge gaps as well as next steps to begin to address these.

Considerations for GBS vaccine Phase IV Studies in the UK

Post-licensure vaccine effectiveness assessment, based on observational studies and the use of real-world evidence, requires high-quality disease surveillance systems to be in place with linkage to pregnancy vaccination systems. There are different ways in which these studies might be done and certain vaccine implementation strategies could allow for concurrent effectiveness data collection, including cluster or stepped wedge randomisation design to evaluate implementation. Endpoints of effectiveness and vaccine safety can be studied in various ways including case-control or passive cohort studies. Clinical effectiveness will need to be evaluated, with iGBS <90 days of age, both EOD and LOD, including bacteraemia and meningitis, as the endpoints. The implementation of the vaccine may lead to the discovery of other endpoints that GBS is contributing to, using a vaccine probe approach. This could yield a wide range of potential information regarding the role of GBS in invasive maternal disease and other maternal infections such as urinary tract infection or post-partum sepsis, as well as other neonatal infections (arthritis/cellulitis/culture negative sepsis), all-cause neonatal infections, stillbirths, prematurity and low birthweight, maternal and neonatal GBS colonisation, maternal and infant antibiotic use, serotype or strain replacement.

UKHSA Overview of Potential Designs for a GBS Vaccine Phase IV Study

Nick Andrews first outlined the role and responsibilities of UKHSA when introducing a new vaccine in the UK. UKHSA have a legal mandate to undertake surveillance without patient consent with responsibility for evaluating several areas after vaccine introduction:

This is achieved using a variety of data sources including laboratory data from SGSS, vaccine registers, GP databases, hospital and mortality data, enhanced surveillance via questionnaires to GPs may be used (especially effective for rare diseases), and case note review. UKHSA also considers the vaccine schedule and the need for additional immunogenicity evaluation in trials, for example with the National Immunisation Schedule Evaluation Consortium (NISEC).

GBS vaccine phase IV studies may be able to take similar approaches to previous pathogens. Some examples of previous phase IV studies carried out in the UK post-vaccine introduction are outlined in Table 4.

It was noted that the approach used for COVID-19 vaccine in pregnancy may have most relevance to GBS. Linkage to NIMS was very effective for these studies. As a national register, NIMS removed the need to try to access GP vaccination data, which is time-consuming. Replicating this approach will depend on whether GBS vaccines will be recorded in NIMS or via GP records, and, if into GP records first, whether a national extraction from these records can be done that allows linkage. This may be dependent on the vaccine delivery setting, for example vaccination via GP versus maternity services.

Potential phase IV study designs and relevant data sources for GBS include:

Without access to a vaccine register, the case-coverage approach or a small cohort study could be used, both necessitating contacting GPs for vaccination status, which is less efficient. Infant/mother linkage data would still be required. Other endpoints could be evaluated using these proposed study designs, for example stillbirth, within a vaccine probe study.

Considering the possibility of effectiveness data collection alongside vaccine implementation, for example stepped wedge or cluster randomisation implementation design, it was noted that these methodologies are infrequently used due to more complex feasibility and ethical considerations. Previous antenatal vaccines introduced into the UK, for example the pertussis vaccine, have had highly protective effects, making these types of observational studies more useful. For a vaccine with lower effectiveness, these studies become challenging. This would also be true of a vaccine probe study; better results are yielded at higher effectiveness levels. At lower effectiveness, the benefits of cluster randomised or stepped wedge implementation design might be more apparent. It was noted that a GBS vaccine is anticipated to be highly effective against the serotypes and strains included but with monitoring for serotype or strain replacement.

Further discussion and development of plans for phase IV studies is needed. Many of these will be primarily carried out at national level by UKHSA within their remit for evaluation of vaccine implementation. In addition to these large-scale effectiveness and safety studies, there are other research questions better suited to smaller scale studies than by public health bodies. SGUL is already involved in several areas of work to progress GBS vaccine implementation in the UK, both at pre-licensure and post-licensure stage. This work is in close collaboration with charity partners, such as GBSS, regulators including the MHRA, government departments and industry partners. This multi-agency approach is vital for an effective and accessible vaccine programme. Some of these areas of work, current and planned, are outlined below:

Pre-vaccine implementation:

Post-vaccine implementation:

List of Meeting Attendees

Chairs:

Kirsty Le Doare – Professor of Vaccinology and Immunology, St George’s University, Centre for Neonatal and Paediatric Infection

Paul Heath – Professor of Paediatric Infectious Diseases, St George’s University, Centre for Neonatal and Paediatric Infection

Attendees (A-Z):

Arlene Reynolds – Senior Professional Adviser in Public Health in Scottish Government, Previously Public Health Scotland

Asma Khalil – Professor of Fetal Maternal Medicine at St George’s University, London

Caroline Trotter – Epidemiologist, Cambridge and Imperial College Universities

Cecilia Hultin – Senior Research Nurse St George’s University Vaccine Institute

Cheryl Battersby – Consultant Neonatologist at Chelsea and Westminster Hospital, NIHR Clinician Scientist

Claire Sharkey – Clinical Research Fellow at St George’s University Vaccine Institute

Claire Wright – Head of Evidence and Policy, Meningitis Research Foundation

Colin Brown – Director of Clinical Emerging Infection at UKHSA

Debbie King – Research Lead, Vaccines at Wellcome Trust

Elaine Devine – Director of Communications and Advocacy, Meningitis Research Foundation

Elizabeth Hayter – Clinical Trials Co-Ordinator St George’s University Vaccine Institute

Emma Eccleston – Research Midwife at St George’s University Vaccine Institute

Eva Galiza – Senior Clinical Research Fellow, St George’s University, Centre for Neonatal and Paediatric Infection

Helen Campbell – Clinical Scientist, Immunisations Team at UKHSA

Jane Plumb – Chief Executive of Group B Strep Support (GBSS)

Juliana Coelho – National Infection Service: Respiratory and Vaccine Preventable Bacteria Reference Unit (NIS:RVPBRU), UKHSA

Kate Walker – Professor of Obstetrics at Nottingham University, Co-Chief Investigator of GBS3 Trial

Konstantinos Karampatsas – Clinical Research Fellow, St George’s University, Centre for Neonatal and Paediatric Infection

Laura McDonald – Health Care Scientist, Immunisation team, Public Health Scotland

Lauren Wallis – Communications/Engagement Officer at St George’s University, Centre for Neonatal and Paediatric Infection

Lynne Rush – Public Health trainee, Immunisation team at Public Health Scotland

Mair Powell – Senior Clinical Assessor at the Healthcare Products Regulatory Authority, Ireland,

Merryn Voysey – Associate Professor of Statistics in Vaccinology, Oxford Vaccine Group

Michelle Falconer – Nurse Consultant, Vaccine and Immunisations, Public Health Scotland

Natasha Thorn – Clinical Research Fellow at St George’s University, Centre for Neonatal and Paediatric Infection

Nick Andrews – Statistician at UKHSA

Oliver Plumb – Advocacy and Information Manager, Group B Strep Support (GBSS)

Phil Steer – Chair of Medical Advisory Committee, Group B Strep Support (GBSS)

Rebecca L Guy – Epidemiologist at UKHSA

Sam Knight – Helpline & Information Officer at Group B Strep Support (GBSS)

Shamez Ladhani – Paediatrician and UKHSA Epidemiologist

Theresa Lamagni – Epidemiologist, Healthcare Acquired Infections and Antimicrobial Resistance, UKHSA

Vanessa Greening - Research Midwife, St George’s University Vaccine Institute

Xinxue Liu – Statistician, Oxford Vaccine Group