Global Alignment of Immunization Safety Assessment in Pregnancy - The GAIA project

: Immunization in pregnancy provides a promising contribution to globally reducing neonatal and under-five childhood mortality and morbidity. Thorough assessment of benefits and risks for the primarily healthy pregnant women and their unborn babies is required. The GAIA project was formed in response to the call of the World Health Organization for a globally concerted approach to actively monitor the safety of vaccines and immunization in pregnancy programs. GAIA aims to improve the quality of outcome data from clinical vaccine trials in pregnant women with a specific focus on the needs and requirements for safety monitoring in LMIC. In the first year of the project, a large and functional network of experts was created. The first outputs include a guidance document for clinical trials of immunization in pregnancy, a basic data collection guide, ten case definitions of key obstetric and neonatal health outcomes, an ontology of key terms and a map of pertinent disease codes. The GAIA Network is designed as an open and growing forum for professionals sharing the GAIA vision and aim. Based on the initial achievements, tools and services are developed to support investigators and strengthen immunization in pregnancy programs with specific focus on LMIC. Abstract Immunization in pregnancy provides a promising contribution to globally r educing neonatal and under-five childhood mortality and morbidity . Thorough assessment of benefits and risks for the primarily healthy pregnant women and their unborn babies is required. The GAIA project was formed in response to the call of the World Health Organization for a globally concerted approach to actively monitor the safety of vaccines and immunization in pregnancy programs . GAIA aims to improve the quality of outcome data from clinical vaccine trials in pregnant women with a specific focus on the needs and requirements for safety monitoring in LMIC. In the first year of the project, a large and functional network of experts was created. The first outputs include a guidance document for clinical trials of immunization in pregnancy, a basic data collection guide, ten case definitions of key obstetric and neonatal health outcomes, an ontology of key terms and a map of pertinent disease codes. The GAIA Network is designed as an open and growing forum for professionals sharing the GAIA vision and aim. Based on the initial achievements, tools and services are developed to support investigators and strengthen immunization in pregnancy programs with specific focus on LMIC.

Abstract: Immunization in pregnancy provides a promising contribution to globally reducing neonatal and under-five childhood mortality and morbidity. Thorough assessment of benefits and risks for the primarily healthy pregnant women and their unborn babies is required. The GAIA project was formed in response to the call of the World Health Organization for a globally concerted approach to actively monitor the safety of vaccines and immunization in pregnancy programs. GAIA aims to improve the quality of outcome data from clinical vaccine trials in pregnant women with a specific focus on the needs and requirements for safety monitoring in LMIC. In the first year of the project, a large and functional network of experts was created. The first outputs include a guidance document for clinical trials of immunization in pregnancy, a basic data collection guide, ten case definitions of key obstetric and neonatal health outcomes, an ontology of key terms and a map of pertinent disease codes.
The GAIA Network is designed as an open and growing forum for professionals sharing the GAIA vision and aim. Based on the initial achievements, tools and services are developed to support investigators and strengthen immunization in pregnancy programs with specific focus on LMIC.
concerted approach to actively monitor the safety of vaccines and immunization in pregnancy programs. GAIA aims to improve the quality of outcome data from clinical vaccine trials in pregnant women with a specific focus on the needs and requirements for safety monitoring in LMIC.
In the first year of the project, a large and functional network of experts was created. The first outputs include a guidance document for clinical trials of immunization in pregnancy, a basic data collection guide, ten case definitions of key obstetric and neonatal health outcomes, an ontology of key terms and a map of pertinent disease codes.
The GAIA Network is designed as an open and growing forum for professionals sharing the GAIA vision and aim. Based on the initial achievements, tools and services are developed to support investigators and strengthen immunization in pregnancy programs with specific focus on LMIC.

Introduction
Reducing neonatal and under-five childhood mortality and morbidity is a target of the Health Sustainable Development Goal [1]. Immunization in pregnancy provides a promising contribution to achieving this goal [2]. Whilst immunizing pregnant women against tetanus has been practiced for decades, new strategies such as antenatal influenza and pertussis vaccination are now being systematically evaluated and are recommended by WHO [3,4,5. ]Additional promising vaccines are in development for global use in pregnancy such as group B streptococcal (GBS) and respiratory syncytial virus (RSV) vaccines [6,7,8] Introduction of these vaccines comes with tremendous potential benefit, specifically for women and children living in low and middle income countries (LMIC) due to the higher perinatal and infant mortality rates in this setting. However, there is much at stake when it comes to immunization of pregnant women. The safety of any product given to primarily healthy mothers and their unborn babies receives intense professional and public scrutiny. First, this is because two lives can be directly affected during a time of vulnerability, yet both are also likely to benefit from the prevention of serious infections Second, safety concerns are not exclusive to maternal vaccination programs alone since similar concerns are raised with the use of the same vaccines in routine childhood and adult immunization programs (e. g. , influenza and pertussis vaccines). In turn, a vaccine found to be safe for pregnant women and neonates would likely be well accepted by the general public for other target groups. Thus, the potential beneficial and harmful effects of immunization in pregnancy and its ethical implications are augmented by the ramifications on routine pediatric and adult immunization programs. Therefore, thorough assessment of the safety of vaccines during pregnancy is required given the potential for numerous confounding events associated with pregnancy itself in women and the fetus, and in the neonatal period.
Particularly challenging for monitoring and communicating the benefits and risks of immunization programs in pregnancy is that several common health outcomes may be perceived as both a measure of benefit and risk and this assessment may change over time. For example, immunization may decrease the stillbirth rate if a vaccine decreases infections that lead to stillbirth. However, stillbirths will still occur in pregnancies and may also be perceived and reported as adverse events following immunization. Particularly early during program introduction, the impact of reducing mortality due to immunization may not be detectable on the population level while pregnancy complications, such as stillbirth, are registered. This makes early benefitrisk analyses challenging and may compromise the viability of an immunization program independently of any causal relationship between the complication and immunization.
Therefore, product or program specific safety issues need to be identified to appropriately assess the benefit-risk profile of these vaccines and their implementation programs and to protect the target population from unintended harm.
On the other hand, unfounded public or professional concerns can jeopardize beneficial vaccine programs and need to be rapidly refuted based on rigorous and credible science and globally coordinated decision-making and communication.
Addressing these issues requires more than communication strategies. It requires active monitoring and research to enable confident communication with high quality data. As important safety concerns tend to be serious but rare health events, their investigation requires a harmonized approach and needs to be based on large sample sizes to provide satisfactory statistical confidence of risk estimates and to enable comparison of multiple populations and programs. This is best addressed by close global collaboration based on a harmonized approach [9,10,11].
The general need for a globally concerted approach to actively monitor the safety of vaccines and programs of immunization is recognized by the WHO Global Vaccine Safety Blueprint, the strategic plan of the Global Vaccine Safety Initiative [12]. A recent WHO consultation specifically identified the currently fragmented research, the current lack of data comparability as well as the need to improve the quality of safety data to inform decision making and system strengthening [13].

Aim and objectives and first outcomes
The GAIA project aims to improve the outcome data quality from clinical vaccine trials in pregnant women with a specific focus on the needs and requirements for safety monitoring in LMIC. GAIA addresses three main objectives to achieving this aim. First, to improve comparability of safety data across products, programs, and populations for effective and efficient strengthening of immunization programs in pregnant women. Second, to optimize the value of local investigations by global harmonization of methods. Third, to promote scientific progress by increasing analytic power and options through globally concerted approaches.
In a first step, the GAIA project has established an open and dynamic network of professionals concerned with monitoring the safety of immunization in pregnancy.
Together, compiling a shared terminology and developing case definitions for selected obstetric and neonatal outcomes create a common understanding of the outcomes monitored. Consensus guidance is developed for harmonized safety monitoring in clinical trials, and tools are created for effective and efficient data collection, synthesis and pooling, with a focus on LMIC needs and requirements.

Network
The development of a global standard requires the engagement of a large number of stakeholders (e. g. regulatory agencies, public health organizations, academic institutions and health care providers) and experts who will collaborate on a voluntary basis on the development, review and validation of the standards and tools through an iterative process in the framework of multiple streamlined working groups with a Biological Standardization) collaborate in a carefully designed governance structure to coordinate and guide the activities of the network partners [14].
The GAIA project leverages the unique accrual of expertise in its project partners and the wider network and is designed to achieve its aim by capitalizing on existing methods and infrastructures. In the following sections, we outline the methods and first outcomes of the GAIA project.

Standardized case definitions for key outcomes
WHO and the Brighton Collaboration (BC) held a consultancy of key stakeholders in July 24-25, 2014 in Geneva to review current practice and advice on a strategic direction towards a harmonized approach for monitoring immunization in pregnancy programs [13]. This meeting highlighted the current lack of harmonization and the  [17]. To achieve the need of developing many definitions in a short timeframe, the Brighton Collaboration standard process was expanded to enable "batch production" of definitions. Two task forces comprising expertise from public health institutes, regulatory authorities, academic and patient care organizations and vaccine manufacturers were created, to simultaneously develop ten neonatal and ten obstetric outcome definitions, respectively, in dedicated  [18,19,20,21,22,23,24,25,26,27]. A subsequent set of 10 prioritized case definitions is in development ( Table 2)

Guidelines and Data Collection Matrix
The WHO and the Brighton Collaboration (BC) consultancy also recognized the need for guidance on basic data collection, analysis and presentation of vaccine safety data. This is specifically, because no such global consensus guidelines exist to meet the need of concerted safety monitoring throughout the life cycle of vaccines or for global access in rapidly emerging immunization in pregnancy programs.
The GAIA project develops guidelines for harmonized data collection, analysis and presentation according to the Brighton Collaboration standard process 16

Tools
To further promote a common understanding and shared language, a list of over 3000 terms comprising obstetric and neonatal outcomes (e. g. stillbirth) and enabling terms (e. g. prematurity) is structured in an ontology catalog demonstrating their hierarchical and conceptual dependencies enriched by synonyms and disease concept descriptions. This is of particular use for the development of multilingual data collection forms. This effort builds on the existing products and expertise at the Enterprise Vocabulary Services at the National Cancer Institute also in collaboration with the National Children's Study and the Eunice Kennedy Shriver National Institute of Child Health and Human Development at NIH and will be made available as an dedicated, user friendly, searchable database on the NIH website.
The case definitions with the glossary and ontology of terms have enabled creation of a map of disease codes that can be used for retrieval of data on specific outcomes from electronic health care databases (e. g. , ICD9, ICD10, MedDRA, WHOArt, and READ). The mapped disease codes will be made available via the same GAIA terminology database described above. A systematic literature search on existing observational studies around pertussis and influenza maternal immunization safety was coordinated by Erasmus Medical Center and will reflect disease codes and algorithms that have been used to extract data from electronic health care databases and prepares for further expansion of the GAIA work into observational settings.
The existing Automated Brighton Collaboration Case Classification tool (ABC tool) will be expanded to include the newly developed set of case definitions for automated classification of reported events into their level of diagnostic certainty. The rule based tool classifies the information and also prompts the investigator to the type of information that should be collected (on follow-up) for a given case to meet the highest possible level of diagnostic certainty [30]. All tools will be described in more detail in an upcoming special issue of Vaccine.

Dissemination
The GAIA project aims to serve different stakeholders in the field of immunization in pregnancy. To achieve this aim, GAIA is promoting review, use and recommendation of its outcomes by key stakeholders in the field of global vaccine safety research and monitoring including national and international public health and regulatory organizations as well as vaccine manufacturers. Similarly, the wider scientific community and health care professionals are invited to review outcomes early in the process, provide comments and utilize the shared network, standards and tools. This effort is facilitated by engagement and dialogue with stakeholder organizations, presentations and workshops at scientific conferences, regular newsletters and publication in the scientific literature.
All outputs are made available through the GAIA website [ 31 ]. Immunization in pregnancy is an evolving field, and adaptation of standards and tools to specific from regulatory authorities, public health institutes, academia and industry to discuss new safety data from immunization in pregnancy studies, to identify converging stakeholder needs and requirements for high quality data, to review GAIA standards and tools for safety monitoring and to build consensus on the best practice guidance for monitoring vaccine safety in pregnancy in the light of current experience with a focus on LMIC.

Summary and next steps
The GAIA project has established a large and functional network of experts and a purpose infrastructure around first outputs, which may serve as a platform for continued collaborative improvement of the quality of data generated for strengthening programs of immunization in pregnancy with specific focus on LMIC.
The immediate next outputs will be the next set of eleven case definitions and the finalized online services and tools.
Capitalizing on the initial achievements of the GAIA project and following the recommendations of the consensus conference, GAIA could also be effectively utilized as a platform for capacity building in LMIC, specifically for monitoring the safety of immunization in pregnancy. Such capacity building could include the development of specific training modules for data safety monitoring boards (DSMBs) and National Immunization Technical Advisory Groups (NITAGs). Investing in a globally concerted approach will give added value to the individual studies/investments and strengthen a multinational platform for immunization in pregnancy.
With increasing implementation of immunization in pregnancy programs and research in LMIC, innovative approaches to validate the implementation of classic research methods as well as novel study designs and benefit-risk monitoring frameworks will be needed as critical next elements of a global active safety monitoring infrastructure, which will ultimately allow rapid evaluation and response to safety signals or concerns related to products and programs for maternal immunization.

Acknowledgements
The GAIA project is supported by the Bill & Melinda Gates Foundation, grant number OPP1119788. The authors are grateful for the extraordinary effort by the many volunteering professional colleagues in the various working groups and the many additional participants providing excellent contributions peer review or in person consultations. An updated list of GAIA Network partners and participants is available at: http://gaia-consortium. net.