The rapidly changing landscape of respiratory syncytial virus prophylaxis

Abstract

The introduction of nirsevimab (a respiratory syncytial virus [RSV] monoclonal antibody that can protect for minimum 5 months with a single dose) and RSV maternal vaccines to protect young infants has the potential to dramatically decrease RSV hospitalizations in Canada. However, there remain many unanswered questions before optimal use of these products can be assured.

Introduction

Traditional teaching is that respiratory syncytial virus (RSV) infects almost all children by 24 months of age (1). However, in most countries, including Canada, only very small numbers of children had RSV infection during the first year of the COVID-19 pandemic, likely due to public health measures in place at that time.

In Canada, RSV typically circulates for 4–5 months from November through March, with the onset of RSV season generally being later in more northern jurisdictions (2). For unclear reasons, every second year, there tends to be a “bad” RSV season with an increased number of hospitalizations (3). After having practically no 2020–2021 RSV season in Canada, the next two seasons started earlier than usual. The duration of the season was typical in most jurisdictions, but longer than usual in some, including the Prairie provinces in 2021–2022 when RSV hospitalizations occurred over a period of 8 months (4).

With their first RSV infection, approximately 70% of children are asymptomatic or have an uncomplicated upper respiratory tract infection (URTI). The other 30% have lower respiratory tract infection (LRTI), mainly mild bronchiolitis. A recent prospective study conducted in Argentina, Bangladesh, Canada, Finland, Honduras, South Africa, Thailand, and the United States reported RSV hospitalization rates of 1.86% at 0–5 months of age, 0.26% at 6–11 months, and 0.27% at 12–23 months (5); 78% of hospitalizations occurred before 6 months of age. During 2017–2022, surveillance data from 13 pediatric tertiary care centres in the Canadian Immunization Monitoring Program Active (IMPACT) showed a somewhat smaller proportion (50%) of admissions with RSV occurred before 6 months of age (6). However, admissions where RSV was not the primary diagnosis may have been more common in children older than 5 months in this study. RSV is the leading cause of hospitalization worldwide during the first year of life, typically for bronchiolitis but occasionally for pneumonia or apnea (7). The main risk factors for hospitalization among young children include age less than 6 weeks, prematurity, chronic lung disease, hemodynamically significant congenital heart disease, and immunocompromise (2), but the vast majority of admissions occur in previously healthy children. Death due to RSV is rare in Canada (8). in contrast to influenza infection, extra-pulmonary manifestations are uncommon with RSV.

Despite there being only two subgroups of RSV (A and B), infection recurs throughout our life span. Passive maternal antibodies provide some protection (9), yet unlike with other respiratory viruses, hospitalization with RSV is common in early infancy. Antibodies from natural infection do not necessarily prevent reinfection, but decrease the severity such that it is extremely rare for children to have two RSV hospitalizations during the same RSV season (10). Only children with chronic medical conditions are likely to have RSV hospitalizations in more than one RSV season, or to be hospitalized for RSV beyond 24 months of age. However, during the 2022-2023 RSV season, the median admission age in British Columbia was higher than before the COVID-19 pandemic, potentially because the 2020-2021 RSV season did not happen (11).

RSV again becomes a problem for the elderly, manifesting as pneumonia, asthma, or exacerbations of underlying pulmonary or cardiac conditions (12). The incidence of hospitalizations appears to be lower than for influenza (13,14), but estimates are compromised by a lack of RSV testing in the elderly.

Monoclonal antibodies

The monoclonal antibody palivizumab (AstraZeneca) has been used in high-risk children in Canada for over 20 years. In the original randomized controlled trials, efficacy for prevention of hospitalization ranged from 39% (95% CI 20% to 58%) for children with chronic lung disease to 78% (95% CI 66% to 90%) for healthy preterm infants (2). The only severe adverse event linked to palivizumab is anaphylaxis, which is rare and readily managed. However, palivizumab is costly (approximately $1500 per dose for a 7 kg child) and requires frequent dosing. Although the product monograph calls for five monthly doses (15), some Canadian centres give maximum four doses as the purported therapeutic level is exceeded for more than 30 days once a child has received multiple doses (2,15); success has been described with abbreviated regimens (16,18). The indications for use vary widely across provinces (18); however, the National Advisory Committee on Immunization recently published updated recommendations (20).

Nirsevimab (AstraZeneca and Sanofi) is a monoclonal antibody authorized for use in Canada in April 2023. Like palivizumab, nirsevimab binds to the RSV fusion glycoprotein, but its half-life and target affinity are much greater; one dose is protective for at least 150 days. Efficacy for prevention of hospitalization using the dose proposed for licensing (50 mg if <5 kg and 100 mg if 5 kg or greater) was 87% (95% CI 54% to 96%) in infants 29–34 weeks GA and 77% (95% CI 49% to 89%) in infants born after 34 weeks gestational age (GA) (data presented at ReSViNET, Lisbon 22-24 Feb) (21) (Figure 1). Nirsevimab is estimated to prevent 79.5% (95% CI 65.9% to 87.7%) of outpatient visits for RSV LRTI in healthy infants born at full term or preterm (≥29 weeks GA) (22), an outcome never reported for palivizumab. It is not known whether nirsevimab prevents RSV infection or just lessens the severity. For the moment, there do not appear to be important safety concerns. There is an ongoing randomized controlled trial of the efficacy of nirsevimab compared to palivizumab for infants currently eligible for the latter in the United States (GA < 29 weeks, chronic lung disease, hemodynamically significant congenital heart disease) (23). In the meantime, there are pharmacokinetic data showing that nirsevimab levels are comparable in healthy infants and those with high-risk conditions (22), which may be sufficient to convince some jurisdictions to replace palivizumab with nirsevimab. The initial list price will be $952 per dose in 2023 (personal communication – Sanofi).

Figure 1:

Nirsevimab clinical trial efficacy results (150 days follow-up)

Modified from Mueller et al; presented at RSVVW’23, February 22–24, 2023

wGA = weeks gestational age; MA = medically attended; RSV = respiratory syncytial virus; LRTI = lower respiratory tract infection; RRR = relative risk reduction

Studies estimate that RSV care costs a mean of US $187 per infant born in the United States (24). If the cost of nirsevimab approximated this, universal use would be cost neutral or even cost saving. However, the cost will be higher, and decisions will need to be made about eligibility. An economic analysis presented at the US Advisory Committee on Immunization Practices (ACIP) meeting in February 2023 estimated that at a nirsevimab cost of US $300, it would cost approximately US $100,000 to gain one additional quality-adjusted life year (QALY) with universal use in children <8 months of age at the start of RSV season and those born during the season (excluding children eligible for palivizumab). This fell to US $59,000 if nirsevimab replaced palivizumab and fell further if one assumed that outpatient visits due to upper respiratory tract infections would decrease (an outcome not assessed yet in nirsevimab trials) (25). A second economic model (from Sanofi), applied to the same population but with a 1-month shorter RSV season, estimated approximately US $70,000 per QALY, assuming a nirsevimab price of US $500 (26).

Clesrovimab (Merck) is another long-acting monoclonal antibody currently in phase 3 trials anticipated to finish in 2024 (27), so there is hope that there will be competition in the marketplace.

The variability in guidelines for use of palivizumab across Canada (19) leads to potential inequities. Since longer acting less expensive monoclonal antibodies offer the potential for universal use, one would hope that soon there will be uniform monoclonal antibody programs across the country. One could imagine a program where all infants born during RSV season were offered a monoclonal antibody during their birth hospitalization, with a catch-up program for infants born prior to RSV season. One could provide these catch-up doses at well-baby check-ups, but this would miss infants not brought to such visits (who may be at higher risk for severe RSV as they tend to be of lower socioeconomic status), plus antibody levels may wane by the end of RSV season if nirsevimab is given too far in advance. Therefore, a mass catch-up campaign run by public health at the beginning of RSV season should also be considered. Even if the goal is simply to prevent RSV in the first 6 months of life, almost every infant is likely to require a dose since there are usually only about 7 months between RSV seasons.

Determining the start and end of RSV season is vital for optimizing the timing of administration of monoclonal antibodies. As previously mentioned, RSV seasons in Canada have been less predictable recently. Therefore, there is a need for use of local laboratory data (28) and potentially wastewater testing for RSV (29) to make decisions rather than relying on fixed start and stop dates.

Maternal vaccine

Infants born to women immunized against RSV during the third trimester are born with higher levels of RSV neutralizing antibodies than those born to unimmunized women, due to transplacental transfer of antibodies (30). The first RSV maternal immunization phase 3 trial used a F (fusion) protein nanoparticle vaccine (Novavax) and prevented about 40% of RSV hospitalizations in the first 6 months of life, proving that passive maternal antibodies are of benefit, though failing to meet the prespecified success criterion for efficacy (31). It then became apparent that the Pre-F conformation of the protein was a far better vaccine target (32). In the phase 3 trial of an unadjuvanted bivalent (RSV subgroups A and B) Pre-F protein subunit vaccine (Pfizer) given at 24–36 weeks GA, efficacy was 57% up to day 90 of life, 47% day 91–180 (calculated by us as vaccine efficacy = 1 – relative risk = (risk_placebo – risk_vaccination)/risk_placebo, excluding children who already had RSV infection) and 12% day 181–360 for medically attended LRTI (defined as respiratory rate ≥60 if less than 2 months old and ≥50 if older, oxygen saturation < 95%, or chest wall indrawing). Efficacy for severe LRTI (defined as respiratory rate ≥70 if less than 2 months old and ≥60 if older, oxygen saturation <93%, need for high flow nasal cannula or mechanical ventilation, unresponsive state or ICU admission for more than 4 hours) was 82% up to day 90 and 56% day 91–180 (calculated by us). Efficacy for prevention of hospitalization was 68% up to day 90 and 31.5% day 91–180 (calculated by us), while there were more hospitalized infants in the maternal immunization group than in the placebo group day 181–360 (19 of 3476 versus 13 of 3436) (33,34). Confidence intervals are wide for all these estimates. There is a potential concern about a non-statistically significant increased risk of preterm delivery (preterm births in RSVPreF recipients (5.7%) compared to those who received placebo (4.7%), RR 1.19 [95% CI 0.97 to 1.45]), but no differences in neonatal death. The available data are insufficient to establish or exclude a causal relationship between preterm birth and receipt of the RSVPreF vaccine. An unadjuvanted Pre-F3 protein subunit vaccine (GSK) was also studied in pregnancy (35), but its development was stopped because of a statistically significant imbalance in the proportions of preterm births and neonatal deaths between the vaccine and the placebo groups in the phase 3 trial (36).

The bivalent RSVPreF vaccine (Pfizer) is likely to be licensed for use in pregnant persons by Health Canada during early 2024. Unanswered questions are:

• (A)What GA is optimal for administration, given that one wants the highest possible antibody levels at delivery, but delivery spans 22–42 weeks GA with infants born preterm being at the highest risk of severe RSV infection? The US Food & Drug Administration (FDA) has approved the use of RSVPreF in pregnant individuals at 32 through 36 weeks gestational age, to mitigate the potential risk of preterm birth. https://www.fda.gov/news-events/press-announcements/fda-approves-first-vaccine-pregnant-individuals-prevent-rsv-infants. Is it an accurate assumption that it takes approximately 2 weeks for maternal antibodies to rise sufficiently to protect the infant?
• (B)Can RSVPreF vaccine be co-administered with influenza, COVID-19, and pertussis vaccines during pregnancy?
• (C)How high will the uptake be? In 2021, uptake during pregnancy in Canada was only 65% for pertussis vaccine and 53% for influenza vaccine (37).
• (D)Universal maternal immunization would be the simplest program. However, can we make the program more cost-effective by not immunizing women whose infants are likely to be more than 6 months old for most of their first RSV season, potentially giving nirsevimab if their infants are born preterm and discharged during RSV season?
• (E)Along those same lines, what is the optimal way to combine use of nirsevimab with maternal RSV vaccine? Providing both for one infant would be costly. The safety and efficacy of exposing an infant to both is yet to be studied. Should nirsevimab use be limited to infants born to women who did not receive vaccine at the optimal time in pregnancy, perhaps having an indication for infants with underlying conditions that increase their risk of RSV hospitalization?
• (F)Should nirsevimab ever be used in the second year of life? Current NACI guidelines only recommend second-season palivizumab for children with chronic lung disease of prematurity although it may be considered for heart transplant candidates or recipients, or those with severe immunodeficiency (20).

Infant vaccines

A formalin-inactivated vaccine studied in infants and young children in the 1960s increased the severity of infections, presumably due to ‘immune enhancement,’ leading to two deaths (38) and a decade long moratorium on RSV vaccine development. However, a recent review article listed 16 vaccines in phase 1 or 2 trials for infants, with the authors expressing optimism that the future may include live attenuated intranasal vaccines offered to infants not (or no longer) protected by maternal vaccines or monoclonal antibodies (39). As postulated for conjugated pneumococcal (40) and influenza vaccines (41), widespread protection of children may result in “herd immunity” and prevent disease in the elderly.

In conclusion, we optimistically anticipate that over the next decade, new preventive strategies will substantially reduce the burden of severe RSV disease.

Funding Statement

Funding: No funding was received for this work.

Acknowledgements:

The authors would like to acknowledge Chelsea Caya who performed the statistics.

Contributors:

Writing – Original draft, JL Robinson; Writing – Review & Editing, J Papenburg;

Ethics Approval:

N/A

Informed Consent:

N/A

Registry and the Registration No. of the Study/Trial:

N/A

Data Accessibility:

All data will not be made publicly available. Researchers who require access to the study data can contact the corresponding author for further information.

Funding:

No funding was received for this work.

Disclosures:

J Robinson has no conflicts of interest related to this topic. J Papenburg reports grants for research from Merck and MedImmune, honoraria for speaking from AstraZeneca, and consulting fees from Merck.

Peer Review:

This manuscript has been peer reviewed.

Animal Studies:

N/A