Leveraging the Global Influenza Surveillance and Response System for global respiratory syncytial virus surveillance—opportunities and challenges

Abstract Background Respiratory syncytial virus (RSV)‐associated acute lower respiratory infection is a common cause for hospitalization and hospital deaths in young children globally. There is urgent need to generate evidence to inform immunization policies when RSV vaccines become available. The WHO piloted a RSV surveillance strategy that leverages the existing capacities of the Global Influenza Surveillance and Response System (GISRS) to better understand RSV seasonality, high‐risk groups, validate case definitions, and develop laboratory and surveillance standards for RSV. Methods The RSV sentinel surveillance strategy was piloted in 14 countries. Patients across all age groups presenting to sentinel hospitals and clinics were screened all year‐round using extended severe acute respiratory infection (SARI) and acute respiratory infection (ARI) case definitions for hospital and primary care settings, respectively. Respiratory specimens were tested for RSV at the National Influenza Centre (NIC) using standardized molecular diagnostics that had been validated by an External Quality Assurance program. The WHO FluMart data platform was adapted to receive case‐based RSV data and visualize interactive visualization outputs. Results Laboratory standards for detecting RSV by RT‐PCR were developed. A review assessed the feasibility and the low incremental costs for RSV surveillance. Several challenges were addressed related to case definitions, sampling strategies, the need to focus surveillance on young children, and the data required for burden estimation. Conclusions There was no evidence of any significant adverse impact on the functioning of GISRS which is primarily intended for virologic and epidemiological surveillance of influenza.


| INTRODUC TI ON
Respiratory syncytial virus (RSV) is an important cause of respiratory infections, with the most severe infections occurring in young children, older adults, and adults with chronic co-morbidities. 1,2 Acute lower respiratory infections (bronchiolitis and pneumonia) due to RSV are among the most common causes of hospital admission (estimated 3.2 million annually) and hospital deaths (estimated 59 600) in young children globally and are associated with a high hospitalization cost burden. 3,4 The overall number of deaths associated with RSV is less certain but has been estimated to be between 94 600 and 149 400 deaths in young children globally. 5 RSV disease in young children has been associated with persisting sequelae such as recurrent wheeze and asthma later in childhood. 6 The burden of severe disease and deaths from RSV in high-risk adults remains uncertain but initial estimates suggest that influenza virus is more common than RSV, but RSV infection has a higher risk of death than seasonal influenza in adults. 7 The RSV vaccine development landscape is evolving rapidly. There are currently several vaccine candidates and monoclonal antibodies in various stages of clinical trials. 8,9 The World Health Organization (WHO) Product Development for Vaccines Advisory Committee assessed RSV as the most promising new vaccine of potential major public health importance that may be ready for widespread implementation in the next 10 years. 10 A gap analysis indicates that most data on RSV disease symptoms, treatment, and acute sequelae are from high-income countries in contrast to the RSV burden distribution. RSV disease data from low-and middle-income countries (LMICs) are improving, but additional data and granularity are needed particularly on disease in the first 6 months of life. 11 The Gavi Vaccine Alliance approved RSV immunization products in its Vaccine Investment Strategy for 2021-25 contingent on availability of a licenced product, a SAGE recommendation, WHO pre-qualification, and meeting the financial assumptions used as the basis of the RSV investment case. 12 Given the importance of RSV as a significant cause of hospitalization and death in young children, it is likthat a cost-effective vaccine would be considered for inclusion in national immunization programs globally. The Global Influenza Surveillance and Response System (GISRS) coordinated by WHO and endorsed by national governments tests more than two million respiratory specimens annually to monitor the spread and evolution of influenza viruses through a network of about 150 well-established laboratories in 114 countries representing 91% of the world's population. The GISRS network is ideally placed as a platform for the introduction of a more systematic testing for RSV associated with respiratory illness as it already uses molecular diagnostics on respiratory specimens and conditioned to report regular and disciplined reporting to a global platform.
Countries in Latin America and the Caribbean region started testing for RSV by immunofluorescence. As RSV molecular diagnostics became widely available, many countries within the GISRS started testing for RSV and other respiratory viruses as a by-product of influenza surveillance using the WHO-recommended influenza-like illness (ILI), acute respiratory infection (ARI), and severe acute respiratory infection (SARI) case definitions (Table 1) 15 These countries were selected in consultation with the WHO regional offices based on existing and strong capacities in laboratory and epidemiological surveillance and willingness to participate in the pilot. The overall aim of the pilot was to investigate the suitability of the GISRS platform for RSV surveillance in a range of settings. This paper describes the strategy and discusses how various epidemiological and laboratory challenges were addressed.

| PILOTING THE S TR ATEGY FOR R S V SURVEILL AN CE
The overall strategy aimed to leverage and build on the existing capacities of the well-established GISRS to test for RSV without adversely affecting its vital role in influenza surveillance. 15 The con- the NICs on request. The WHO FluMart data reporting platform for influenza was adapted for countries to directly upload anonymized case-based RSV data using their own database structure and system without having to transform it. Interactive data visualization of trends, distributions, and other outputs were developed that could be stratified by country, year, period, age, hospital or outpatientbased surveillance, and presence or absence of fever. An external advisory group provided oversight to the pilot project.

| OPP ORTUNITIE S AND CHALLENG E S
The RSV surveillance strategy aims to leverage existing national resources of the GISRS without adversely impacting its established

| Opportunities presented by the pilot
The

| Epidemiology-related challenges
Five major epidemiological concerns were identified for using GISRS for RSV surveillance. First, an ILI, ARI, or SARI case definition that requires fever would have low sensitivity to detect RSV, as over half of young children with RSV infection present without fever. 17,18 Furthermore, non-respiratory presentations of RSV infection in very young infants are recognized (apnea, sepsis with no apparent cause) and a wider case definition would be needed in this age group. 19,20 The pilot addressed this concern by using extended SARI and ARI case definitions (which do not include fever) for hospital-based and community-based surveillance, respectively. In addition, the case definition for infants aged less than 6 months included apnea and sepsis from undetermined cause (Table 1). Second, the RSV disease burden is highest in children aged less than 2 years, 21

| Laboratory-related challenges
Three major laboratory-related concerns were considered. First, most NICs were already testing for RSV and other respiratory viruses using commercial or laboratory-developed molecular diagnostics, either monoplex or multiplex RT-PCR assays as well as antigenbased testing methods that varied in sensitivity and specificity.
Laboratories in Latin America were using direct or indirect immunofluorescence assay (IFA) to detect RSV, the sensitivity and specificity of which varies with the age of the patient, and type and quality of specimen. The sensitivity of IFA might be low for RSV detection especially in the off-season period. 26,27 All laboratories agreed to use real-time RT-PCR for RSV detection. Although most switched to the CDC monoplex RT-PCR assay, a few laboratories opted to continue using commercial or other laboratory-developed RT-PCR assays after validation against the CDC assay. Second, though important to collect information on the circulating RSV strain (A or B) to better understand their association with disease severity, 28-30 the decision to type RSV was left to the individual laboratories, depending upon their available resources and expertise. Third, the decision to conduct genome sequencing of RSV was deferred to a later phase to prioritize establishing a surveillance system for RSV detection during the pilot phase, and to better define the goals of virologic surveillance of RSV.

| D ISCUSS I ON
The overarching aim of the pilot was to build on the successful global influenza surveillance platform, GISRS, to obtain RSV surveillance data without having a negative impact on GISRS. The primary motivation was to establish systems to obtain data to inform policy and decision making for possible future RSV vaccine investment, especially in LMIC settings in which it may be unlikely that there will be other relevant sources of suitable RSV data for this purpose. A midterm review assessed that it was feasible to leverage GISRS for RSV surveillance with little incremental cost without any significant adverse impact on the functioning of GISRS for influenza surveillance.
We have outlined the major challenges faced, and the decisions made to tackle these in the development of the RSV surveillance pilot approach. The pilot collected data to monitor performance at each site and will be able to report on the degree of success achieved using the approach adopted. For example, the pilot will report on the relative performance of the case definitions for the identification of RSV and influenza cases in hospital and primary care settings and will monitor the success in recruitment of cases in the specific targeted high-risk groups. A limitation of the current pilot is that it did not specifically target pregnant women for RSV Enhanced data on respiratory support required during hospitalization will also assist in assessment of the health costs associated with RSV. These data will be important in health economical assessments of any intervention (such as vaccines). Equally, it will be important to raise awareness of the possibility of RSV infection in causing the respiratory disorders in older adults and those with chronic medical conditions.
Once RSV seasonality is better defined, it may be possible to limit site-specific surveillance to high transmission seasons for both RSV and influenza. Once RSV case definitions have been evaluated, the need for ongoing individual case data will be reconsidered.
Multiyear data from RSV surveillance will allow the study of evolutionary changes in RSV over time and in response to widespread vaccine introduction.
An important aspect of the pilot is the collection of surveillance data on RSV and influenza in both primary and secondary care settings. This is important as coinfection in causation of lower respiratory infections is increasingly recognized. 32 This joint surveillance will help improve knowledge of RSV and influenza global transmission patterns and will inform future considerations of the need for and feasibility of broader infectious disease surveillance platforms which could encompass several other important respiratory pathogens that can result in hospitalization.
A major advantage of using the existing GISRS platform is that GISRS has built close relationships with Ministries of Health and policymakers in individual countries. This is clearly signaled by the financial support to the RSV surveillance pilot provided by national GISRS platforms. The pilot served to strengthen these relationships by providing policymakers with information on RSV from which vaccine investment, healthcare planning decision, and cost-benefit analyses can be made in the future. We expect that these data raise awareness of RSV as one of the most common causes of hospitalization in young children globally and an underappreciated cause of disease in older adults. We will work closely with other stakeholders in each country to ensure that these data are presented to the appropriate audiences and respond to local comments and needs.
Another possible approach to long-term sustainability may be to integrate this surveillance within a broader (respiratory) infectious disease surveillance platform in secondary care that encompasses important infectious agents in a cost-effective manner. The longterm sustainability of RSV surveillance will depend on the success of relationships with the Expanded Program of Immunization and the Maternal, Neonatal and Child Health programs responsible for delivering RSV immunization products.
The pilot project should help define the minimum dataset and minimum surveillance effort required to maintain a functional and useful RSV surveillance, in a cost-efficient manner and without adversely impacting on essential influenza surveillance. Information obtained should also help in better understanding the relative importance of RSV as a cause of healthcare burden in various risk groups and the overlap between influenza and RSV viral infection globally. It is hoped that experience from this pilot will serve as a basis from which to plan future RSV surveillance activities. Future extensions of the pilot could include incorporating additional highrisk groups, such as pregnant women, and extending the global scale of RSV surveillance so that global transmission patterns and seasonality can be better studied.
Finally, the pilot phase did not track mortality or focus on severity of illness and did not follow up subjects for long-term sequelae such as wheeze or asthma, or conduct surveillance in pregnant women to assess the impact of RSV infection on pregnancy and birth outcomes, which could be a limitation or an opportunity to explore, when RSV surveillance is established using lessons learnt from the pilot phase.
In conclusion, challenges notwithstanding, the pilot project demonstrated the feasibility of leveraging the GISRS for RSV surveillance without any significant negative impact on influenza surveillance.