Immunization, not vaccination: monoclonal antibodies for infant RSV prevention and the US vaccines for children program

Abstract

In the US, RSV imposes significant burdens on infants, households, and the health system. Yet the only licensed immunization is accessible to only certain risk groups comprising 2% of the infant population, leaving the remaining 98% unprotected. An effective immunization for all infants is a significant public health priority. One possible solution is the FDA-approved monoclonal antibody nirsevimab, which recent evidence suggests is safe and effective in preventing RSV in all infants, and which is currently being considered for inclusion in the pediatric immunization schedule and the federal Vaccines for Children (VFC) program. But the question arises whether passive immunization products like nirsevimab ought to be eligible for the VFC, which nominally and traditionally centers on vaccines providing active immunity. Addressing this is urgent because VFC inclusion will be decided on imminently. I argue there are strong policy grounds, i.e., reasons grounded in the ultimate health system goals of maximizing population health or social welfare subject to resource constraints, not to exclude passive immunization from VFC eligibility. Active and passive immunizations both provide adaptive immunity and can therefore produce qualitatively similar effects on risks of infection, disease, and transmission; on disease severity and duration; and on health, welfare, and health resource use. The distinction between active and passive immunization does not intrinsically matter since what matters for the attainment of health system goals is the extent of immunity conferred, not whether immunity is active or passive. Nor can passivity be considered a useful proxy for conferring a lesser extent of immunity, since no such proxy is needed (existing valuation methods can cope with variations in product attributes), and it is a poor proxy (passive immunizations can be better for individuals with impaired immune systems and can have comparable effectiveness durations and economic value as vaccines).

Keywords:

• Respiratory syncytial virus (RSV)
• infants
• passive immunization
• monoclonal antibodies
• nirsevimab
• vaccination
• active immunization
• advisory committee on immunization practices (ACIP)
• vaccines for children program (VFC)

1. The burdens of infant RSV

In the US, infant respiratory syncytial virus infection (RSV) is an important public health problem. It is the leading cause of infant hospitalization¹, and infects about two-thirds of infants by their first birthday and about 90% by their second². Before the second birthday, about 2% of infants might be hospitalized³, 21% might go to an emergency room, and 18% might be brought to a pediatrician because of RSV⁴. Economic burdens are considerable. Treatment costs have been estimated at $34,132 and $3869 per inpatient and outpatient case respectively⁵; and at about $710 million annually in aggregate, or about $187 per birth⁶. Out-of-pocket costs have been estimated at $299, and indirect costs at $1921–3873, per infant⁷.

Infection is usually mild⁸ but can also result in more severe episodes⁹ involving hospitalizations⁵ and mortality risks¹⁰ from bronchiolitis and pneumonia¹¹. Risk of severe RSV is higher among infants with risk factors like prematurity or congenital heart disease, but most severe cases (about 80%¹²) occur among otherwise healthy infants¹³. Risks are especially high immediately after birth: over half and three-fourths of hospitalizations occur within the first three and six months of life respectively¹⁴. RSV infections are mostly seasonal, rising through the fall, peaking during winter, and ending during the spring, though patterns vary by year and geography¹⁵ and have been disrupted by Covid¹⁶.

RSV risks are affected by socio-economic factors like household income¹⁷, race and ethnicity, and Medicaid status¹. RSV imposes burdens not just on infants and the health system, but also on parents and caregivers whose mental health¹⁸, family life¹⁸, and finances suffer⁷. RSV burdens can be long-term: health care utilization may rise for as long as five years after infection¹⁹. Infant RSV may contribute to adolescent wheezing and asthma²⁰, which come with their own health and economic burdens (including lost work and schooling)²¹.

2. On the need for RSV immunization for all infants

Significant RSV burdens in otherwise healthy infants in the first year of life makes protecting all infants (not just those with risk factors) during that year an important public health goal^20,22. But there was, until very recently, only one FDA-approved immunization, the monocolonal antibody (mAb) palivizumab, and it is indicated only for infants who are preterm or have heart or lung disease²³, which constitute perhaps only 2% of infants²⁴, leaving the remaining 98% unprotected.

There are three potential immunization-related strategies for RSV prevention in all infants: infant vaccination, maternal vaccination (MV), and mAbs²⁵. The human immune system has two components²⁶. The innate system relies on non-specific mechanisms of action such as physical barriers and inflammation that apply indiscriminately to many pathogens. The adaptive immune system, in contrast, offers pathogen-specific protection. It does so through the workings of antibodies, specialized proteins that recognize and bind to specific molecular features (epitopes of antigens) of specific pathogens, thereby inactivating the pathogen, interfering with its workings, and marking it for destruction by other parts of the immune system²⁷.

Infant vaccination, like all vaccination and past infections, confers active immunity, that is, immunity resulting from the work of antibodies generated by the individual’s own immune system, which has been taught to recognize the pathogen by vaccines or past infection²⁸. However, infant vaccination is unlikely to be effective until the infant immune system has sufficiently matured to actively generate its own antibodies, which may happen only six months after birth²⁹. Thus protection during the first six months of life likely benefits from passive immunity, where the antibodies come from an external source²⁸, through either MV or mAbs. MV stimulates antibody production by the maternal immune system (providing active immunity to the mother) and transplacental transfer of such antibodies to the infant (providing passive immunity to the infant). With mAbs, the dose itself contains the antibodies³⁰.

On 17 July 2023, the FDA approved the long-acting mAb, nirsevimab, for RSV prevention in all infants³¹, paving the way for its availability during the 2023–2024 RSV season. Recently concluded trials showed nirsevimab to have a similar safety profile to palivizumab³², and about 80% efficacy against medically attended RSV-related lower respiratory tract infections in infants born at term or pre-term during their first RSV season³³. (FDA approval for the MV, RSVpreF, is expected in the next few weeks.³¹)

Upon FDA approval of a vaccine or “related agent” for controlling disease, the Advisory Committee on Immunization Practices (ACIP)—an independent group of medical and public health experts–develops recommendations for its use³⁴. To inform those recommendations, ACIP conducts a health technology assessment (HTA) of the product that considers “disease epidemiology and burden of disease, vaccine safety, vaccine efficacy and effectiveness, the quality of evidence reviewed, economic analyses, and implementation issues”³⁴. (ACIP does not refer to its assessment as an HTA though it satisfies the general definition of one³⁵.) Such HTA culminates in a decision to either recommend the product for use in a target population, not recommend that product for use in such population, or recommend it for use on an individual basis based on shared clinical decision-making³⁶. If the CDC accepts an ACIP use recommendation, it becomes the official CDC recommendation³⁴ and is codified in the US Immunization Schedule³⁷. CDC-recommended immunizations must be covered by health insurers and employer-sponsored health plans established after the Affordable Care Act (ACA). CDC recommendations are typically considered the standard of care for immunization across the country and a reference-point for immunization coverage decisions globally³⁸.

ACIP is also authorized by law to determine which vaccines will be made available through the federal Vaccines for Children (VFC) program, which provides free access to such vaccines for Medicaid-insured, American Indian and Alaska Natives, underinsured, and uninsured children³⁹. ACIP codifies its decisions on vaccine inclusions through VFC Resolutions, which are “separate[] from” and “may not necessarily match” ACIP/CDC recommendations⁴⁰, but which in practice overlap significantly with them^40,41, suggesting VFC Resolutions are significantly informed by HTA findings. ACIP has scheduled votes for an ACIP use recommendation and for VFC inclusion of nirsevimab on 3 August 2023⁴².

3. Immunization, not vaccination

An issue that arises, though, is that nirsevimab confers passive immunity, while the VFC is, as is evident from its name, nominally and traditionally focused on vaccines, which confer active immunity. This raises the question: should passive immunization products, solely by virtue of their passivity, be excluded from VFC eligibility given the program’s nominal and traditional focus on active immunization? This question is timely, indeed urgent, given an imminent ACIP decision in time for the fall 2023 RSV season.

Two clarifications: First, this question is not about specific products like nirsevimab, but rather passive immunization products as a class. The issue therefore does not hinge on the specificities of either nirsevimab or infant RSV, but encompasses other passive immunization products like MV and palivizumab for infant RSV prevention, and monoclonal antibodies (like a potential future mAb for Covid prevention in high-risk children⁴³) and immune globulin products (like the Hepatitis B immune globulin (HBIG)⁴¹) more generally. I discuss nirsevimab and infant RSV prevention to exemplify the issue but not to delimit its scope.

Second and more importantly, the issue is not about inclusion, but rather eligibility for inclusion: is passivity in and of itself disqualifying? Vaccine inclusion in the VFC depends on a favorable HTA outcome. If passive immunizations were ineligible, then they could not be included even with favorable HTA outcomes. And if they were eligible, they would be subject to the same criteria and processes for VFC inclusion as vaccines.

These clarifications imply the issue is not: should nirsevimab be included in the VFC? Such inclusion depends on how well it performs relative to HTA criteria, and I do not assess such performance. The issue is: if nirsevimab or any other passive immunization performed favorably in an HTA, should it nevertheless be disqualified from VFC inclusion just because it is passive? The issue is only consequential for passive immunizations with favorable HTA outcomes, for if outcomes were unfavorable, ineligibility wouldn’t matter since these immunizations shouldn’t be included anyway. Thus, we should for the sake of argument assume a passive immunization that performs sufficiently favorably in an HTA that, were it a vaccine, it would be included in the VFC. We should then ask: should we nevertheless exclude it?

3.1. Policy grounds

I do not attempt a full answer to this question, for this depends on at least one consideration I do not address: whether there are legal constraints to such eligibility⁴⁴. Instead, I offer policy grounds for non-exclusion, that is, grounds related to the fundamental goals of the health system. Conditional on there being no legal constraints, these policy grounds provide an affirmative reason for non-exclusion.

There is debate regarding the highest-level goal or metric of success of the health system⁴⁵. One ethical view, extra-welfarism, says the health system should maximize (or be assessed by the extent to which it succeeds in maximizing) population health subject to resource- and other constraints. The most important alternative view, welfarism, replaces population health with social welfare as the quantity to be maximized, where welfare reflects not just health but also other non-health goods like economic living standards. These views differ in policy implications: if policy A raises both health and household income (e.g., through reductions in disease-related out-of-pocket costs and lost earnings), while policy B has smaller health benefits but larger income benefits, then the health-centric view would prioritize A while the welfare-centric view could prioritize B if the income gains are large enough.

I do not take a stand here on which ethical view is correct, but I take health and welfare to be the main candidates for the intrinsic or ultimate values served by the health system. However, since the health system is resource constrained, an immunization product’s net value to society reflects its impact on not just health and welfare but also on resource utilization within the health sector, government, and society in general (“resources”): the more modest its resource demands, the more is left over for producing health and welfare by other means. Thus, we should ultimately assess immunization products with respect to their impacts on three quantities: health, welfare, and resources. (Some believe that equity is important so that, for example, the health and well-being of the worse off should count more when maximizing population health or well-being⁴⁶. My argument generalizes to accommodate this: simply replace all subsequent mentions of health or well-being with “equity-weighted” health or well-being and the argument follows as before.)

Any aspect of a product that is relevant (irrelevant) to its impact on these three quantities is policy relevant (irrelevant). To build policy grounds for the non-exclusion of passive immunization from the VFC program, I first describe the policy-relevant similarities between active and passive immunization, then proceed to argue that the difference between active and passive immunization is policy irrelevant.

3.2. On the policy-relevant similarities between active and passive immunization

3.2.1. Adaptive immunity, immunity effects, and health, welfare, and resource impacts

The extent of adaptive immunity conferred by immunization has multiple quantitative dimensions. These include the magnitudes of the reduction in the risk of infection, the risk of disease (disease is symptomatic while infection need not be), the severity and duration of disease from breakthrough infections, the risk of onward transmission of breakthrough infections to others, and the duration of these reductions⁴⁷.

It is through these multidimensional quantitative immunity effects, along with their costs, that immunizations affect health, welfare, and resources. For example, the magnitudes of RSV-related improvements in infant and caregiver health, reductions in hospitalization costs, and short- and long-run household income gains depend on the magnitudes of the reduction in the risk, severity, and duration of RSV disease.

This is the fundamental policy-relevant similarity between active and passive immunization: they both confer adaptive, that is, pathogen-specific antibody-mediated, immunity. They therefore both in principle have the same types of multidimensional immunity effects, that is, reductions in infection, disease, and transmission (see, e.g., effectiveness evidence for nirsevimab³³, palivizumab⁴⁸, RSVpreF⁴⁹, and HBIG⁵⁰). And they are therefore both in principle capable of the same types of effects on health, welfare, and resources.

Of course, non-immunization-based prevention technologies (e.g., handwashing, masks, gloves, physical isolation) can also reduce infection and disease risk. But these are in many cases poor substitutes for immunity-based protection: infant RSV burdens, for example, currently persist despite the general availability of these other prevention technologies. And of course, while active and passive immunization can in principle have the same types of effects, the quantitative magnitudes of these effects vary from case to case, as I discuss later.

3.2.2. Health system and government requirements

There are other policy relevant similarities. Population-wide provision of passive immunization imposes the exact same requirements on health systems and governments as vaccinations, and the systems that satisfy these requirements for vaccination can straightforwardly do so for passive immunization. These requirements include the conduct of regulatory activities aimed at ensuring product safety and efficacy; the conduct of health technology assessments (HTA) by National Immunization Technical Advisory Groups (NITAGs) such as ACIP; the formulation of immunization schedules, guidelines, and recommendations; a transparent and well-justified decision-framework for payer reimbursement decisions; appropriate governance, organization, and management practices; ensuring an adequate health workforce and strong supply chains; a capacity to provide high quality health services with high geographical penetration within which routine immunizations can be provided; information systems to track immunization progress; securing sustainable financing; efforts to build community trust and confidence in immunization; and efforts to facilitate healthy investment in R&D into novel immunization products.

3.2.3. Ethical issues

Active and passive immunizations face the same ethical issues. These include the need to respect and balance the four bioethical principles of autonomy (i.e., the right of individuals or parents to choose for themselves or their children), non-maleficence (i.e., doing no harm, which among other things requires minimizing risks of adverse events), beneficence (i.e., the obligation to promote the patient’s and general public’s health and well-being, which can weigh in favor of recommending, incentivizing, and on occasion mandating immunization), and justice (which encompasses equity and fairness in access to immunization)⁵¹. Whether some immunization product or policy succeeds or fails to satisfy some ethical requirement doesn’t depend on whether it involves active or passive immunity.

3.3. The active-passive distinction is policy-irrelevant

3.3.1. The active-passive distinction doesn’t intrinsically matter

Recall that a product’s immunity effect is quantitative and multidimensional. Recall too that a product’s impact on health, welfare, and resources is wholly mediated by its cost and immunity effect. These imply that two hypothetical immunization products that are identical with respect to costs and immunity effect will have identical impacts on health, welfare, and resources, even if one of them happens to provide active immunity and the other passive. Given the choice between such hypothetical products, our ultimate health system goals would leave us indifferent between them. We might say that what ultimately matters is immunization, not vaccination, or that the active-passive distinction doesn’t intrinsically matter.

3.3.2. The active-passive distinction doesn’t derivatively matter

The intrinsic irrelevance of the active-passive distinction implies that the only reason we might care about the distinction is if it correlates somehow with the magnitudes of immunity or cost and therefore with the magnitudes of health, welfare, and resource impacts. If, for example, it just so happens that passive immunization products tend to have smaller immunity effects or higher costs than vaccines, then the former will tend to have less attractive impacts on health, welfare, and resources than the latter. The active versus passive distinction may therefore matter derivatively, in the sense that we may use passivity as a proxy for having lower net value (i.e., health or welfare benefits net of resource costs). Two possible version of this proxy view are that passive products tend to confer shorter durations of immunity²⁸ or sometimes have much higher prices⁵² than vaccines, and therefore may have less net value.

There are two problems with the proxy view.

3.3.2.1. No proxy needed

First, we only need a proxy if our HTA methods are incapable of directly accounting for product variation in the drivers of net value: immunity effects and costs. That is, the fully spelled out argument in favor of the proxy view must be something like: if we care about duration of effectiveness and costs, if our HTA methods are incapable of accounting for product variations in the duration of effectiveness and costs, and if passive products have shorter durations of effectiveness and higher costs than active ones, then we may try to compensate for the inability of our HTA methods to account for effectiveness duration and costs by screening out passive products from the HTA process.

But the need for a proxy disappears if our evaluation methods are perfectly capable of directly accounting for product variation in cost, immunity effects, and impacts on health, welfare, and resources. And indeed, such is the case. ACIP guidelines allow the use of cost-effectiveness, cost-utility, and cost-benefit analysis (CEA, CUA, and CBA respectively) to assess the economic value of immunizations⁵³. Such analyses can readily generate net value indicators (like the incremental cost-effectiveness ratio of CEA and CUA, or the benefit-cost ratio of CBA) that reflect products’ costs, immunity effects, and health-, welfare-, and resource impacts. Indeed, ACIP has already been considering CEAs and CUAs of nirsevimab that account for its duration of efficacy, health impacts, resource impacts (including costs per dose and averted treatment costs), and productivity impacts⁵⁴. Since the above methods can directly accommodate all the drivers of net value, proxies for such drivers are unnecessary.

3.3.2.2. Poor proxy

The second problem is that passivity can be a poor proxy for having lower net value: passive immunizations can be better for individuals with impaired immune systems and can have comparable durations of effectiveness and net value as vaccines. Nirsevimab exemplifies these. The high risks of RSV infection in the first days, weeks, and months of life imply the importance of immediate immunity. But passive immunity compares favorably to active immunity in two respects. First, passive immunity gives immediate protection⁵⁵, while active immunity typically takes weeks to develop²⁸ leaving infants exposed to infection risk in the interim. Second, the immaturity of the infant immune system in the first few months of life implies that its immune response to vaccination may be “suboptimal and short-lived”²⁵. This implies that passive immunity may be more likely than infant vaccines to confer the needed antibodies during the period of immune system immaturity in the first few months of life. Beyond infant RSV, passive immunization may have an advantage relative to vaccination for some patients with weakened immune systems incapable of mounting an optimal immune response to vaccination (e.g., in some children undergoing immunosuppressive chemotherapy^56,57.

Evidence suggests nirsevimab provides protection against RSV for at least the length of an RSV season⁵⁸, which puts it on par with annual seasonal influenza vaccination which often also provides protection for a single influenza season⁵⁹, which is of roughly equal length⁶⁰. The CDC estimates an incremental cost-effectiveness ratio (ICER) for nirsevimab of $102,805 per quality-adjusted life year (QALY)⁵⁴, which compares favorably to that, for example, of the ACIP-recommended Meningococcal ACWY vaccine for adolescents, which exceeds $200,000/QALY⁶¹. Other passive immunizations with evidence of good value are palivizumab for RSV prevention in high-risk infants (90% of ICER estimates from the US, Canada, and other countries are below $50,000/QALY)²³ and HBIG for Hepatitis B prevention in infants with carrier mothers (evidence from China suggesting this is cost saving when added to universal infant vaccination against Hepatitis B)⁶².

4. Conclusion

The health system’s goal is to maximize population health or social welfare subject to resource constraints. Active and passive immunity have the same types of immunity effects, and therefore the same types of health, welfare, and resource impacts. They impose the same types of requirements on health systems and governments and raise the same ethical issues. While active and passive immunization differ at the biological level, this difference is neither intrinsically nor derivatively relevant to achieving health system goals. These provide us with policy grounds not to deny passive immunization products VFC eligibility.

I do not claim nirsevimab should be VFC-eligible, since this depends on legal considerations I ignore. Nor do I claim nirsevimab should be included in the VFC, for this depends on quantitative assessments of value-related attributes like cost-effectiveness, public health impact, and health system requirements, which I do not perform. I conclude only that absent legal constraints, ultimate health system goals suggest not excluding passive immunization products like nirsevimab from VFC eligibility. I end by noting that although I discuss these issues in the US setting, my reasoning generalizes to all health systems globally facing the issue of including passive immunization into their immunization schedules: there are strong policy reasons for subjecting passive and active immunizations to the same criteria and processes in inclusion decisions.

Transparency

Author contributions

All authorship roles were filled by JPS.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

The Editor in Chief helped with adjudicating the final decision on this paper.

Acknowledgements

No assistance in the preparation of this article is to be declared.

Declaration of financial/other interests

JPS is an economist at Data for Decisions, LLC (DfD) in Waltham, MA, USA, and a Research Associate at the Harvard T.H. Chan School of Public Health (HCSPH) in Boston, MA, USA. He has received financial support for his other work from vaccine- and medical device manufacturers, the World Health Organization, the Bill and Melinda Gates Foundation, and other organizations. He has received such support through contracts between these funders and DfD and HCSPH where he is employed, and through independent consulting relationships.

Additional information

Funding

This work was funded by Sanofi through a contract with Data for Decisions, LLC, where JP Sevilla is an employee.