Abstract
Clinical manifestations of respiratory syncytial virus (RSV) infection vary widely from mild, self-limiting illness to severe life-threatening disease. There are gaps in knowledge of biomarkers to objectively define severe disease and predict clinical outcomes.
A systematic search was performed, 1945–March 2019 in databases Ovid Medline, Embase, Global health, Scopus, and Web of Science. Risk of bias was assessed using the Cochrane tool.
A total of 25 132 abstracts were screened and studies were assessed for quality, risk of bias, and extracted data; 111 studies met the inclusion criteria. RSV severity was correlated with antibody titers, reduced T and B cells, dysregulated innate immunity, neutrophil mobilization to the lungs and blood, decreased Th1 response, and Th2 weighted shift. Microbial exposures in respiratory tract may contribute to neutrophil mobilization to the lungs of the infants with severe RSV compared with mild RSV disease.
Although a wide range of biomarkers have been associated with RSV disease severity, robust validated biomarkers are lacking. This review illustrates the broad heterogeneity of study designs and high variability in the definition of severe RSV disease. Prospective studies are required to validate biomarkers. Additional research investigating epigenetics, metabolomics, and microbiome holds promise for novel biomarkers.
Respiratory syncytial virus (RSV) is the leading cause of lower respiratory tract infection (LRTI) in children younger than 5 years and was responsible for around 33.1 million infections globally in 2015 in this age group [1]. Despite significant efforts over the past decades to develop a better understanding of RSV, there remain significant gaps, including identification of robust molecular markers as predictive tools or biological correlates of disease severity in infants. Host characteristics and microbial exposures during early life may have long-term consequences by altering the immune response and making an infant more susceptible to severe disease. Based on a systematic literature review, we describe the host and microbial factors that have been reported in association with RSV disease severity characteristics, including humoral and cellular immunity, cytokine/chemokine response, genetics, transcriptome, epigenetics, and respiratory microbiome.
METHODS
Search Strategy and Selection Criteria
A systematic literature review was performed using a combination of search terms (“human respiratory syncytial virus,” “respiratory syncytial virus,” or “RSV” in human studies) in the Ovid Medline, Embase, Global health, Scopus, and Web of Science databases including studies from 1945 to March 2019. No restriction on study design, language, or publication were initially applied. Additional articles were identified by scanning of reference lists of identified citations. The inclusion and exclusion criteria (eligibility criteria) of studies are listed in Table 1. Data selection was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [2]. The PRISMA flow diagram is shown in Supplementary Figure 1.
Eligibility Criteria
| Inclusion | Exclusion |
|---|---|
| Human RSV studies Severity of RSV infection assessed Biological marker investigated Relation explored between biomarker and severity of RSV infection Studies written in English, French, Spanish, Italian, or Portuguese Studies including subjects aged younger than 18 years | Studies in animal models Studies in cell lines Studies in adults Studies focused on treatment, diagnostics, or epidemiology of RSV infection Studies without a clear definition of disease severity Studies without a definitive RSV diagnosis Studies on antibodies or viral characteristics Literature reviews Studies written in any language other than those mentioned in the inclusion criteria |
| Inclusion | Exclusion |
|---|---|
| Human RSV studies Severity of RSV infection assessed Biological marker investigated Relation explored between biomarker and severity of RSV infection Studies written in English, French, Spanish, Italian, or Portuguese Studies including subjects aged younger than 18 years | Studies in animal models Studies in cell lines Studies in adults Studies focused on treatment, diagnostics, or epidemiology of RSV infection Studies without a clear definition of disease severity Studies without a definitive RSV diagnosis Studies on antibodies or viral characteristics Literature reviews Studies written in any language other than those mentioned in the inclusion criteria |
Abbreviation: RSV, respiratory syncytial virus.
Eligibility Criteria
| Inclusion | Exclusion |
|---|---|
| Human RSV studies Severity of RSV infection assessed Biological marker investigated Relation explored between biomarker and severity of RSV infection Studies written in English, French, Spanish, Italian, or Portuguese Studies including subjects aged younger than 18 years | Studies in animal models Studies in cell lines Studies in adults Studies focused on treatment, diagnostics, or epidemiology of RSV infection Studies without a clear definition of disease severity Studies without a definitive RSV diagnosis Studies on antibodies or viral characteristics Literature reviews Studies written in any language other than those mentioned in the inclusion criteria |
| Inclusion | Exclusion |
|---|---|
| Human RSV studies Severity of RSV infection assessed Biological marker investigated Relation explored between biomarker and severity of RSV infection Studies written in English, French, Spanish, Italian, or Portuguese Studies including subjects aged younger than 18 years | Studies in animal models Studies in cell lines Studies in adults Studies focused on treatment, diagnostics, or epidemiology of RSV infection Studies without a clear definition of disease severity Studies without a definitive RSV diagnosis Studies on antibodies or viral characteristics Literature reviews Studies written in any language other than those mentioned in the inclusion criteria |
Abbreviation: RSV, respiratory syncytial virus.
Definitions
RSV infection was identified definitively on the basis of a diagnostic test of body fluid including polymerase chain reaction (PCR), viral culture, or antigen test. Biomarkers were defined as any traceable biological parameter/substance that was measurable. Age group inclusion in this study was based on the World Health Organization (WHO) definition for children; namely, studies including subjects aged 18 years or older were excluded. The control group was defined as a population without RSV infection, without respiratory infection associated with other pathogens, or infants without respiratory diseases.
Quality of Evidence
Each included study was reviewed by 2 review authors (D. Ö., S. B. D., C. M., G.-L. L., S. J., J. B.) using the Cochrane tool for quality assessment using grade guidelines [3]. The review was registered with PROSPERO (registration number: CRD42019119615) and conducted according to PRISMA guidelines. The risk of bias result is shown in Supplementary Figure 2.
RSV Severity Classifications
Throughout the different articles surveyed, an RSV case is defined as “severe” based on 1 or more clinical parameters, for example, the “duration of” and/or the “need of” 1 or more of the following parameters: hospitalization [4, 5], oxygen supplementation [6, 7], and mechanical ventilation [8, 9]. RSV severity scores, such as Wood’s clinical asthma score (M-WCAS) [10] or a modified scoring system [11], incorporate different clinical end points. Most of these severity scoring systems have, however, been designed for evaluation by medical professionals in a hospital setting where mostly severe disease is observed, but not by parents or to assess mild disease in a community (outpatient) setting. The ReSVinet scale has been proposed to be useful and reliable in pediatric infectious diseases, either recorded by pediatricians or parents [12].
The focus of scoring systems towards more severe disease symptoms poses challenges for biomarker research on diversity in RSV disease severity. Indeed, differences in the classifications of RSV severity may lead to conflicting results. For instance, when infants with LRTI were graded for severity, IL-8 concentrations in blood has been positively associated by a number of studies [10, 13–16]. However, a few studies did not confirm the association with RSV severity parameters, even though increased IL-8 was observed between control and RSV patients [4, 17]. This difference in the outcome may potentially be attributed to the diversity in RSV severity parameters and to the definition of mild, moderate, or severe RSV disease in different studies. A universally applied scoring system would be beneficial to investigate biomarkers related with RSV disease severity.
In the Supplementary Tables of this review, we comprehensively summarize the literature findings, that is associations (positive, negative, or no association when applicable) of the biomarkers and the severity parameters, enabling the reader to balance and judge the underlying evidence. When an association was noted by more than 1 evidence/study, we listed the corresponding biomarker as a potential biomarker for RSV severity, regardless of whether the same biomarker led to conflicting data in other studies due to study design limitations.
RESULTS
Risk Factors for Severe RSV Disease in Infants
Prematurity
Infants who are born preterm are 3 times more likely to be hospitalized upon RSV infection than are term infants [18]. This may be attributed to differential immune profiles of infants who are born preterm, for example, lower neutrophil proportions [19] or reduced TLR4 surface protein and mRNA expression in monocytes [20]. The reduced TLR4 mRNA expression has been correlated with decreased interleukin-1β (IL-1β), IL-6, and tumor necrosis factor- α (TNF-α) production in the whole blood ex vivo. Decreased cytokine release leading to suppressed innate immunity may create susceptibility to infections of the prematurely born infant [20]. In RSV disease, differential TLR4 mRNA expression has been also shown between preterm versus term infants with RSV bronchiolitis [21]. Conversely, in preterm infants with RSV bronchiolitis, blood neutrophil TLR4 mRNA expression was higher compared to term-born infants with RSV bronchiolitis despite reduced surface protein levels in blood and bronchoalveolar lavage, indicating possible impaired TLR4-dependent pathway in preterm infants [21].
Although being born prematurely has been described as an important risk factor for severe RSV, it has been estimated that approximately 80% of RSV-related hospitalizations occur in previously healthy, term-born infants [18]. Genetic background and demographics, such as young age and male sex, and external exposures including maternal smoking, absence of breastfeeding, having siblings, and crowding, have also been identified as risk factors for RSV-related bronchiolitis [22].
Effect of Age
Infants younger than 2 months more frequently suffer from severe RSV infection, comprising 44% of RSV-related hospitalizations [18]. During the first few months of life, maternal antibodies play a role in protection of infants from bacterial and viral exposures, although this protection is only partial. For instance, infants younger than 3 months were found to display the highest rate of positivity for maternal RSV-specific IgG antibody; however, the avidity of IgG was found to be low compared with older infants [23].
Age-related differences were found also at the transcriptional level in RSV-infected infants. Mejias et al showed that transcriptional profiles of RSV-infected younger infants (< 6 months) have reduced expression of genes related to innate and adaptive immunity when compared with age-matched controls, versus equally ill older infants (6–24 months) when compared with age-matched controls, indicating overall suppressed immunity in younger infants [24].
In RSV disease, infants’ immunity is reported to have a limited T helper 1 (Th1) antibacterial and antiviral response, which is an important host defense system [25]. A Treg and T helper 2 (Th2) skewed response upon RSV infection in infants is assumed to contribute to disease severity and limit recovery [25]. This time window may also predispose the infant to environmental exposures such as permitting the colonization by commensals in the intestine and respiratory tract. Consequently, Th2 skewed immune response upon RSV infection is particularly important to emphasize when studying the long-term consequences, such as allergy and asthma, of severe RSV infection during infancy.
Host Genetics as a Potential Predictive Biomarker
To investigate the effect of host genetics in RSV hospitalization, Thomsen et al conducted a study with more than 12 000 pairs of twins [26]. They concluded that host genetics contributed up to 16% of the risk for RSV-related hospitalization during infancy, but their study design did not allow identification of the underlying genes [26].
The underlying genetic studies in this systematic review applied a targeted, gene-specific approach. Complementary unbiased genome-wide (ie, whole-genome sequencing) studies hold the potential to identify novel genetic biomarkers of severe RSV disease susceptibility. Of note, 1 reported genome-wide association study in the RSV field focused on predisposition to RSV bronchiolitis rather than the association with RSV disease severity (and without assessing disease severity parameters) [27]. Here we list potential biomarkers of RSV disease severity assessed by the targeted, gene-specific approach.
Specific genetic polymorphisms in several genes were identified to be associated with predisposition to severe RSV disease [28]. In this review, significant associations of genetic variations with RSV severity were reported in genes involved in Th1 immune response (eg, IFNG) [29, 30], cytokines associated with the recruitment of neutrophils (eg, CXCL8) [31] and other proinflammatory or anti-inflammatory cytokines (eg, IL1RL1, IL6, and IL10) [30, 32, 33], surfactant proteins (eg, SFTPA2 and SFTPD) [34, 35], RSV receptors on respiratory epithelium (eg, TLR4 and CX3CR1) [7, 36, 37], cannabinoid receptor (ie, CNR2) [5], and vitamin D receptor (ie, GC) [38]. The list of specific polymorphisms in each gene and associated amino acid and codon change is shown in Supplementary Table 1.
The most biologically plausible evidence from independent studies investigating genetic associations and RSV severity were shown with polymorphisms in IFNG [29, 30], IL10 [30, 33], and TLR4 [37] genes.
Biomarkers for Severe RSV Disease
Investigating Biomarkers for Severe RSV Disease in the Respiratory Tract
The epithelium in the lung is coated with a layer of viscous mucus covering cilia of the epithelial cells containing mucins. Mucins not only act as a barrier between environmental exposures and the lung epithelium but also regulate the immune response and activate signal transduction pathways [39]. In transcriptomics analysis, RSV severity was found to be negatively correlated with TSPAN8 (encoding a cell surface protein TSPAN-8), MUC13 (encoding to an epithelial cell surface protein MUC-13), and MSP (immunoglobulin binding factor) [40]. Additionally, RSV severity was found to be positively correlated with CCL7 mRNA levels (a chemokine attracting macrophages and degrading components of extracellular matrix) [40].
RSV-F and RSV-G proteins have been shown to interact with epithelial cell receptors (eg, TLR4 and CX3CR1) and this interaction is related to RSV disease severity. For instance, Zhivaki et al showed that RSV infects neonatal-specific regulatory B cells via interaction with the RSV G protein and CX3CR1 receptor on the cell, promoting IL-10 production and downregulating Th1 immunity [41]. As shown in Supplementary Table 1, single nucleotide polymorphisms in genes encoding RSV receptors (TLR4 and CX3CR1) on respiratory epithelium have been reported [7, 36, 37]. This may have transcriptional and translational consequences, leading to changes in host immune response upon RSV infection. For instance, lower CX3CR1 mRNA levels have been reported in infants with prolonged wheezing [42].
RSV infection has been associated with dysregulated innate immunity and neutrophil infiltration [43]. Consistent with this, mRNA expression of nasopharyngeal samples from infants have been characterized by alterations in recruitment of neutrophils. A positive correlation between the mRNA levels of CXCL8 (a neutrophilic chemotactic factor) and RSV disease severity (infants requiring oxygen or mechanical ventilation) has been observed [44]. Although increased concentrations of IL-8 were linked with RSV disease severity [45–47], no significant differences in IL-8 concentrations were found between RSV-induced upper respiratory tract infection (URTI) and bronchiolitis in nasopharyngeal secretions [48, 49]. These contradictory results might be attributed to the differences in RSV severity classifications. Although neutrophils aid in clearing the infected cells and inhibiting viral replication, they release reactive oxygen species, which potentially also damage the lung epithelium and may predispose infants for subsequent susceptibility to allergy and asthma.
Decreased interferon-γ (IFN-γ) cytokine levels have been found in respiratory samples from infants with severe RSV disease [6, 9, 50–53]. Similarly, a negative correlation between IFN-γ mRNA levels and RSV disease severity has been reported in nasopharyngeal samples [54], indicating a suppressed type II IFN (IFN-γ) response. A positive association between IL-4:IFN-γ ratio in respiratory samples and gene expression levels has been shown [51], indicating a Th2 shift in the immune response in severe RSV disease. All cytokines associated with RSV disease severity (or those with conflicting evidence for their role) in respiratory samples are listed in Supplementary Table 2.
In nasopharyngeal samples, the association with RSV severity has been found to be negative for IFN-γ [6, 9, 50–53], and positive for IL-1β [45, 47], IL-6 [14, 17, 45, 47], IL-8 [45–47], and macrophage inflammatory protein-1β (MIP-1β) [17, 45] cytokines. RSV disease severity has been associated with decreased Th1 response and neutrophil recruitment.
Respiratory Microbiome
The microbiome refers to the genetic materials of microorganisms in a specific environment (referred to as microbiota). Although still limited, there is growing evidence that suggests an association between the nasopharyngeal microbiome and RSV disease severity and predisposition to severe RSV disease during infancy (Supplementary Table 3) [55].
In a study conducted in Western Australia, it was reported that in children up 2 years old the nasopharyngeal microbiome is composed predominantly of 6 bacterial genera: Moraxella (40.1%), Streptococcus (13.3%), Corynebacterium (12.1%), Alloiococcus (11.1%), Haemophilus (8.6%), and Staphylococcus (4.2%) [56]. An abundance of Moraxella, Streptococcus, and Haemophilus was positively associated with LRTI when compared to URTI in acute respiratory infections, including RSV infected infants after adjustments for confounders [56, 57]. The data also show that after adjustment to the detected virus, Streptococcus, Haemophilus, and Moraxella microbiome profile groups remained significantly associated with the respiratory symptoms, indicating that both bacteria and viruses may independently contribute to disease [56]. Jiang et al tested nasal aspirates from 608 subjects (< 2 years old) with bronchiolitis and observed that when pathogenic bacteria were present (eg, Streptococcus pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, or Staphylococcus aureus), the percentage of neutrophils was higher and the length of hospital stay was longer (4 days vs 3 days) [58]. In this study, infection with RSV and Haemophilus influenzae was found to be an independent risk factor for longer duration of hospitalization [58]. Consistent with this observation, Suárez-Arrabal et al also observed that infants whose nasopharyngeal microbiome was predominantly colonized with gram-negative bacteria (Moraxella catarrhalis, Haemophilus influenzae) needed a longer duration of oxygen supplementation, had higher plasma IL-6 and IL-8 (but not TNF) levels and higher neutrophil counts when compared with infants whose nasopharyngeal microbiome was predominantly colonized with gram-positive bacteria (Staphylococcus aureus, Streptococcus pneumoniae, β-hemolytic Streptococcus) [59]. Consistently, the presence of Haemophilus was associated with increased IL-8 chemokine levels, previously also related to RSV disease severity and neutrophil recruitment [60]. In a study by de Steenhuijsen Piters et al, an Haemophilus influenzae and Streptococcus dominated microbiota was associated with RSV infection and related hospitalization in infants younger than 2 years [61]. Although no correlation was found between IFN-related genes and the nasal microbiome, Haemophilus influenzae and Streptococcus dominated microbiota were associated with increased mRNA expression of Toll-like receptor linked genes and by neutrophil and macrophage activation [61]. On the other hand, an abundance of Staphylococcus aureus was negatively associated with RSV infection and related hospitalization [61]. Brealey et al correlated the detection of Streptococcus pneumoniae with RSV disease severity assessed by the RSV clinical disease severity score (Woods clinical asthma score) in infants younger than 2 years [62].
Overall, for nasal microbiome data, the association with RSV severity has been found to be positive with an abundance of Moraxella catarrhalis [56, 57, 59], Haemophilus influenzae [56–61], and Streptococcus pneumoniae [56, 57, 61, 62].
Biomarkers of Severe RSV in Blood
Biomarkers in blood not only serve as a surrogate for studying lung immunity and biomarkers associated with severe RSV infection in the lung but also provide information about the systemic immune response triggered by the infection.
Studies of cytokine response have shown conflicting evidence, probably due to marked heterogeneity in study design and sample size. Although the data suggest a predominantly decreased IFN-γ production in nasal samples, in blood the data are conflicting with either positive association [63], negative association [13, 64–66], or a lack of association in several studies (Supplementary Table 4). Although conflicting evidence exists, some data indicate positive associations of other pro- and anti-inflammatory cytokines in blood and RSV disease severity, such as IL-6 [4, 11, 14, 15, 47] and IL-8 [10, 13–16], and negative association of IL-4 [13, 63], IL-12, and RSV disease severity [8, 65], that is cytokines involved in differentiating T cells into Th2 or Th1 cells. Other cytokines in blood that have been associated with RSV disease severity (or those with conflicting evidence for their role) are listed in Supplementary Table 4.
Whole blood transcriptomics may provide more robust information on blood cell immunity in severe RSV disease. Mejias et al demonstrated that investigating the transcriptome can assist in discriminating the severity of RSV disease by translating gene sets into a biologically relevant context, such as postulating changes in cellular populations in relations to RSV severity [24]. For instance, RSV severity is associated with the immune dysregulation, such as overexpression of neutrophil-related gene sets, and inflammation and interferon genes, and suppression of T and B cell-related gene sets [24].
Supplementary Figure 3 lists 18 differentially expressed genes that overlap in 2 studies comparing mild and severe RSV disease [67, 68]. The role and function of these overlapping genes is described in the literature and their identification is consistent with the finding of neutrophil recruitment in severe RSV disease [24]. For instance, MMP8 gene expression has been identified in multiple studies and may regulate neutrophil recruitment from the periphery to the lung [67–69]. Also, elevated CXCL8 mRNA transcription [44, 67] and increased IL-8 cytokine production observed in blood or respiratory samples [10, 13–16, 45–47] provide supporting evidence for enhanced neutrophil recruitment.These results hold the potential for future diagnostic applications determining severe RSV patient groups [70]. In blood samples, the association with RSV severity has been found to be associated with imbalanced Th1 and Th2 cytokine response. Additionally, RNA expression of genes involved in neutrophil recruitment (eg, MMP8, CXCL8, and MPO) were upregulated [44, 67–69].
Effect of Humoral Immunity
Antibodies mediate protection from infection through binding and neutralization of RSV, therefore they are potentially an important component in protection from severe RSV infection.
Lower avidity of RSV-specific IgG antibodies was found in RSV-infected infants when compared with healthy controls in the same age group (<3 months) [71]. However, RSV illness severity was not correlated with several serological characteristics (ie, RSV-IgG antibody titers, avidity of RSV-IgG, virus neutralizing capacity, titers against pre- and postfusion F or G protein ectodomains, and the prefusion F antigenic site Ø) [71]. Although high concentrations of maternal RSV-neutralizing antibodies were shown to protect against RSV hospitalization before 6 months of age, no correlations were found with the severity of illness amongst the infants who were hospitalized [72]. Similarly, no significant difference was observed in RSV antibody concentration as tested by an RSV antibody neutralization assay when compared between RSV-induced URTI and LRTI (< 6 months) [73]. In infants older than 2 years, it was found that the avidity of RSV-specific IgG was lower in infants with RSV-induced LRTI than those with RSV-induced URTI, indicating the protective role of high avidity of RSV-induced IgG in this age group [23].
Reported protective and deleterious role of humoral immunity factors in RSV disease severity are found to be as follows. B-lymphocyte stimulating factors, such as APRIL (proliferation-inducing ligand), and IgA and IgM antibodies (but not IgG), were associated with better oxygen saturation, indicating a protective role in RSV disease [74]. Capella et al reported a negative correlation between the concentrations of pre-F and G antibodies (but not post-F antibodies) and clinical severity score in infants younger than 2 years of age [75]. Deleterious effects of IgE have been reported in infants with pneumonia, LRTI, and higher degree of hypoxia [76–78].
In the literature, protective effects of RSV-specific IgG [23], IgA, IgM [74], pre-F and G antibodies [75], and deleterious effect of IgE antibodies [76–78] were reported. The conflicting data may likely be attributed to the small sample size of the included studies, considering other risk factors are also involved (eg, gestational age, birthweight, maternal smoking, siblings, and day care). Larger studies are needed to find a difference in outcome.
CONCLUSIONS AND NEXT STEPS
The main findings of this review are graphically summarized in Figure 1. Severe RSV disease is associated with a suppression of T cells, B cells, and cytotoxic natural killer (NK) cells, dysregulated innate immunity, and neutrophil mobilization to the respiratory tract and blood. Neutrophil infiltration to the lung might be mediated through upregulation of MMP8 and CXCL8 mRNA expression and increased IL-8 cytokine production. Also, peripheral OLFM4 mRNA expression, as a marker of innate immunity, has been reported as a biomarker to discriminate between mild and severe RSV disease in infants.
Graphical abstract of the main findings of the systematic literature review on RSV disease severity. The figure distinguishes the proposed biomarkers of severe RSV disease (in comparison to milder disease forms) that can be detected in blood, in respiratory samples, and in host DNA. The biomarkers selected for this figure are supported by at least 2 publication reporting similar results and extrapolated from minimal conflicted evidence. Generated using BioRender. Abbreviations: IFN-γ, interferon-γ; IL, interleukin; MIP-1β, macrophage inflammatory protein-1β; nBreg, neonatal-specific regulatory B cell; NK, natural killer; RSV, respiratory syncytial virus; Th, T helper cell.
Downregulation of IFN-γ cytokine in respiratory samples, suppression of T cells, and Th2-skewed response have been associated with severe RSV disease. The Th2-skewed immune response may permit microbiota colonization in the lung. Airway microbiome may be affected by, or lead to, recruitment of neutrophils to the lung.
Although published data are conflicting, humoral immunity may also affect the susceptibility to severe infection. Pre-F and G antibodies [75] and RSV-specific IgG [23] may have a protective effect and IgE levels have been reported to increase the risk of RSV severity.
Early reports are available on investigations of epigenetic alterations, such as DNA methylation [45, 46] and miRNA [79], and metabolomic biomarkers [80], in relation to RSV severity. As more data will become available, investigations on these end points may hold promise for future biomarker research.
It should be noted that the review included biomarkers associated with severe RSV disease but the robustness of the conclusions may be limited because of large heterogeneity amongst included studies (eg, definition of severe RSV cases, low number of studies per biomarker, and varying significance of biomarker data within studies). Therefore, caution is required before drawing definitive conclusions. Future additional large prospective trials investigating RSV disease severity, combined with biomarker analysis strategies that also include novel and high-dimensional readouts are needed in order to deliver and validate robust biomarkers for disease severity that can bring value to clinical practice.
Supplementary Data
Supplementary materials are available at The Journal of Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.
Notes
RESCEU investigators. Harish Nair, Harry Campbell (University of Edinburgh); Peter Openshaw (Imperial College London); Philippe Beutels (Universiteit Antwerpen); Louis Bont (University Medical Centre Utrecht); Andrew Pollard (University of Oxford); Eva Molero (Synapse); Federico Martinon-Torres (Servicio Galego de Saude); Terho Heikkinen (Turku University Central Hospital); Adam Meijer (National Institute for Public Health and the Environment); Thea Kølsen Fischer (Statens Serum Institut); Maarten van den Berge (Academisch Ziekenhuis Groningen); Carlo Giaquinto (Fondazione PENTA for the Treatment and Care of Children with HIV-ONLUS); Clarisse Demont, Scott Gallichan (Sanofi Pasteur); Philip Dormitzer (Pfizer); Amanda Leach (GlaxoSmithKline); Laura Dillon (AstraZeneca); Jeroen Aerssens (Janssen Pharmaceutica); Brian Rosen (Novavax).
Acknowledgments. We thank Debby Bogaert, Linong Zhang, and Ruth Karron for reviewing the earlier versions of the manuscript.
Disclaimer. The views expressed in this article do not necessarily represent the views of UK Department of Health and Social Care (DHSC), Joint Committee on Vaccination and Immunization (JCVI), National Institute for Health Research (NIHR), or World Health Organization (WHO).
Financial support. This work was supported by Innovative Medicines Initiative 2 Joint Undertaking (grant number 116019) with support from the European Union Horizon 2020 Research and Innovation Programme and European Federation of Pharmaceutical Industries and Associations (EFPIA). RESCEU is a joint research and innovation programme supported by EFPIA http://www.imi.europa.eu.
Supplement sponsorship. This supplement is sponsored by RESCEU (REspiratory Syncytial Virus Consortium in EUrope).
Potential conflicts of interest. D. Ö. and J. A. are employees of Janssen Pharmaceutica NV and receive salary. S. B. D. acts on behalf of St George’s, University of London as an Investigator on studies sponsored and/or funded by vaccine manufacturers including Janssen and Medimmune. He receives no personal financial benefit for this work. A. J. P. is Chair of the DHSC JCVI and the European Medicines Agency scientific advisory group on vaccines, and is a member of the WHO Strategic Advisory Group of Experts. A. J. P. is an NIHR Senior Investigator. All other authors report no potential conflicts.
All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.
Presented in part: European Society of Paediatric Infectious Diseases, Madrid, Spain, 23–27 May 2017; and European Society of Paediatric Infectious Diseases, Malmo, Sweden, 28 May–2 June 2018.
References
Author notes
D. O. and S. B. D. contributed equally.
Members of the study group are listed at the end of the text.
