Modelling the population-level benefits and cost-effectiveness of cell-based quadrivalent influenza vaccine for children and adolescents aged 6 months to 17 years in the US

ABSTRACT

Background

Cell-based quadrivalent influenza vaccines (QIVc) can increase effectiveness against seasonal influenza by avoiding mismatch from egg adaption of vaccine viruses. This study evaluates the population-level cost-effectiveness and impacts on health outcomes of QIVc versus an egg-based vaccine (QIVe) in children aged 6 months to 17 years in the US.

Research Design and Methods

A dynamic age-structured susceptible-exposed-infected-recovered model was used to simulate influenza transmission in low and high incidence seasons for two scenarios: 1. QIVe for 6 months–17 year-olds, QIVc for 18–64 year-olds, and adjuvanted QIV (aQIV) for ≥ 65 year-olds, and 2. QIVc for 6 months–64 year-olds, and aQIV for ≥ 65 year-olds. Probabilistic sensitivity analysis was performed to account for uncertainty in parameter estimates. Cost-effectiveness was evaluated as incremental cost-effectiveness ratios (ICERs).

Results

Extension of QIVc to children resulted in 3–4% reductions in cases (1,656,271), hospitalizations (16,688), and deaths (2,126) at a population level in a high incidence season, and 65% reductions (cases: 2,856,384; hospitalizations: 31667; deaths: 4,163) in a low incidence season. Use of QIVc would be cost-saving, with ICERs of -$16,427/QALY and -$8,100/QALY from a payer perspective and -$22,669/QALY and -$15,015/QALY from a societal perspective, for low and high incidence seasons respectively. Cost savings were estimated at approximately $468 million and $1.366 billion for high and low incidence seasons, respectively.

Conclusion

Use of QIVc instead of QIVe in children > 6 months of age in the US would reduce the disease burden and be cost-saving from both a payer and societal perspective.

KEYWORDS:

• influenza vaccine
• cost-effectiveness
• children
• cell-based vaccines
• vaccine effectiveness

1. Introduction

Influenza is a highly contagious respiratory disease caused predominantly in humans by influenza viruses A and B [1]. While influenza affects individuals of all age groups, young children <5 years of age and older adults are at increased risk of severe disease from seasonal epidemics [2]. In children, influenza is associated with high rates of hospitalizations and complications. Neurologic complications such as seizures, encephalitis, and encephalopathy, are identified in around 10% of children hospitalized with laboratory-confirmed influenza [3,4]. A systematic review of influenza burden in Italy found that between 19.8% and 44.4% of pediatric patients with influenza experience complications, depending on the study setting, including 10.8%–13.9% who develop otitis media and 0.4%–8.1% who develop pneumonia [5].

Prior to the COVID-19 pandemic, influenza-related hospitalizations among young children living in the US ranged from 6,000 to 27,000 annually [6]. While pediatric mortality from influenza is uncommon (approximately 100–200 deaths per season in the US), an estimated 50% of deaths occur in patients without underlying high-risk medical conditions [7], and nearly all in children who are not fully vaccinated [8]. In the US, vaccination against influenza is recommended for all children from 6 months to <18 years [9]. Since the COVID-19 pandemic, influenza vaccine coverage rates in this age group have declined to 55.1%, ranging from 63.7% in children <4 years to 46.8% in 13–17 year-olds [10]. Vaccination has been shown to significantly reduce the risk of critical and life-threatening influenza in children, even in seasons when mismatch occurs with circulating strains [11]. As well as being at risk of severe complications, children are also key vectors in the transmission of influenza [12], therefore vaccination of this age group has both individual and community benefits in reducing the overall burden of influenza.

Influenza vaccine effectiveness varies annually depending on the degree of mismatch between vaccine and circulating viruses, as well as the timing of vaccination relative to peak virus circulation [13,14]. Varying effectiveness in recent seasons has been particularly relevant for the A/H3N2 strain, which has at least in part been linked to adaptive mutations developing during propagation of the vaccine virus in eggs [15,16]. Development of vaccine strains using cell culture methods avoids the process of egg-adaptation, and several studies have reported improved match between circulating and vaccine strains and enhanced vaccine effectiveness against the A/H3N2 strain compared with egg-based vaccines reported [17–21].

Cell-based quadrivalent influenza vaccine (QIVc) is approved in the US for persons >6 months of age and is recommended for use in children and adults <65 years of age [22]. In total, approximately 173–185 million doses of influenza vaccine were supplied in the US during the 2022–2023 influenza season, of which approximately 20% were QIVc [23]. Previously, we have demonstrated the potential cost-effectiveness of switching from egg-based quadrivalent influenza vaccine (QIVe) to QIVc in adults aged 18–64 years in the US [24]. In this analysis, we modeled the potential population-level health benefits and cost-effectiveness of extending the use of QIVc to include children aged 6 months to 17 years.

2. Patients and methods

2.1. Epidemiological model

A published dynamic age-structured susceptible-exposed-infected-recovered (SEIR) model was used to simulate transmission of the four circulating influenza strains [25]. Economic analysis was performed based on a decision tree model, with inputs from the infected and vaccinated infected compartments of the SEIR model (Supplementary Figure S1). The key model parameters are outlined in Table 1. As in the previous model, the contact matrix was based on age-stratified data in the US [32], and heterogeneity in the annual impact of influenza was represented by low and high incidence seasons, based on data from 2011–2012 and 2017–2018, respectively [33]. We estimated the transmission parameters by age-groups (beta_i) using a maximum likelihood procedure, where the number of symptomatic cases was considered to follow a Poisson distribution. We then used a non-linear optimizer algorithm (BOBYQA [34]) to efficiently explore the parameter space, with transmission parameters estimated in a low influenza incidence (season 2011–2012) and a high incidence context (season 2017–2018). Intermediary epidemiological scenarios were inferred from linear interpolation between these two extremes. Overall, 27% of individuals were assumed to be protected prior to vaccination due to past infections and/or vaccination [26,27]. Probability of symptomatic, hospitalized, and fatal infection varied by age group (6 months–4 years, 5–17 years, 18–49 years, 50–64 years, and ≥65 years) and risk status (low or high risk), based on published data from the US (Supplementary Table S1) [28]. Results were considered for the overall US population eligible for influenza vaccination (≥6 months of age), estimated at 326 million individuals.

Table 1. Key model parameters.

Item	Values	Reference
Proportion of individuals assumed protected prior to vaccination	27%	[26,27]
Probability of symptomatic infection^a	0.313–0.910	[28]
Probability of hospitalization^a	0.0006–0.0421	[28]
Probability of death^a	0.00001–0.01170	[28]
Vaccine effectiveness QIVe^b	31–48%	[29]
rVE QIVc to QIVe children	8.1%	[28]
rVE QIVc to QIVe adults	11.4%	[28]
Vaccine coverage^c	34–73%	[30]
Cost of GP consultation^d	$56.88–$1147.49	[28]
Cost of hospitalization per episode^d	$8513.45–$74,709.10	[28]
Loss of productivity costs^d	$14.84–106.34	[28]
Vaccine cost^e	$19.92–$63.00	[31]
QALY lost per hospitalization^d	0.013–0.076	[28]
QALY lost per non-fatal ILI^d	0.005–0.007	[28]
Discounting rate	3%	–

GP, general practitioner; ILI, influenza-like illness; QALY, quality-adjusted life year; QIVc, cell-based quadrivalent influenza vaccine; QIVe, egg-based quadrivalent influenza vaccine; rVE, relative vaccine effectiveness.

^aFor further details of estimates by age see Supplementary Table S1.

^bFor further details of estimates by age see Supplementary Table S2.

^cFor further details of estimates by age see Supplementary Table S3.

^dFor further details of estimates by age see Supplementary Table S4.

^eFor further details of estimates by age see Supplementary Table S5.

2.2. Scenarios

Two scenarios were evaluated in the model. Scenario 1 assumed the use of QIVe for children aged 6 months to 17 years, QIVc for adults aged 18–64 years, and an adjuvanted QIV (aQIV) for adults ≥65 years. Scenario 2 replaced the use of QIVe in children with QIVc for all individuals aged 6 months to 64 years, together with aQIV for adults ≥65 years. Vaccine effectiveness (VE) assumptions for QIVe were based on mean VE by age group for the 2011–2012 to 2019–2020 influenza seasons in the US (Supplementary Table S2) [29], with relative VE (rVE) of 8.1% and 11.4% for QIVc versus QIVe in children and adults, respectively [35]. Assumed vaccine coverage rates varied by age (Supplementary Table S3), based on data from the 2021–2022 season [30].

2.3. Economic model

Economic assumptions were based on published median costs per health outcome by age group and adjusted for inflation over time (Supplementary Table s4) [28]. Vaccine prices were based on 2022 list prices, with QIVe price based on a mean of available QIVe vaccines (Supplementary Table s5) [31]. As 2 doses of influenza vaccine are recommended for some children who have not been previously vaccinated, we assumed that 50% of the cohort less than age 9 years would require 2 doses, and adjusted the mean price per dose accordingly. Cost-effectiveness of the vaccine scenarios was evaluated from both a payer and societal perspective in terms of humanistic outcomes, i.e. quality-adjusted life years (QALY), and incremental cost-effectiveness ratios (ICER). The payer perspective included evaluation of the direct costs associated with outpatient consultations, hospitalization, vaccine, and vaccine administration costs. The societal perspective also included the indirect costs of loss of productivity using a human capital approach, which estimated the number of working days lost to illness multiplied by mean daily wage (see Nguyen et al 2023 for additional details) [36]. Discounting was applied at 3% of the life years gained. Although the US does not have a formal cost-effectiveness threshold, an ICER of <$50,000/QALY was assumed to be cost-effective, based on thresholds used in similar analyses [37].

2.4. Sensitivity analysis

Probabilistic sensitivity analysis (PSA) was performed to account for uncertainty in parameter estimates. For the simulation, 500 sets of parameters were randomly drawn from the distributions outlined in Supplementary Table S6, with different parameter ranges for the 6 month–4 years and 5–17 years age groups. Given the lack of published data on rVE of QIVc vs QIVe in children, we performed sensitivity analysis within the simulation, ranging rVE estimates from 0% to 20% above baseline estimates (i.e. 28.8% and 31.4% for children and adults, respectively). Simulations were performed separately for low and high incidence influenza seasons. In addition, a tornado sensitivity analysis was performed on the impact of key parameters on ICER, for extreme high and low values of parameters grouped across age and risk groups.

2.5. Software

The dynamic transmission model was developed in R 4.2.1 and C++ with a Shiny package interface, using Rcpp 1.0.9, RcppArmadillo 0.11.2.3.1, and RcppGSL 0.3.11. The nloptr package was used for model calibration.

3. Results

In a season with a high incidence of influenza, the use of QIVc in children would result in 4% reductions at a population level in the number of symptomatic cases and general practitioner visits, and 3% reductions in the number of hospital visits and deaths, compared with QIVe (Table 2). Overall 16,688 fewer individuals would require hospitalization and there would be 2,126 fewer deaths. In a low incidence influenza season, the relative impact of QIVc would be much higher than in a high incidence season, resulting in a 65% in reduction in all evaluated health outcomes compared with the use of QIVe. This translates to 31,667 fewer hospitalizations and 4,163 fewer deaths.

Table 2. Estimated number of symptomatic influenza cases, GP visits, hospital visits, and deaths at a population level in low and high incidence influenza seasons with the use of QIVe or QIVc in children aged 6 months to 17 years.

	Symptomatic cases	GP visits	Hospital visits	Deaths
High incidence
QIVe	41,536,030	17,359,448	566,533	80,063
QIVc	39,879,759	16,711,185	549,845	77,937
Absolute difference	−1,656,271	−648,263	−16,688	−2,126
Relative difference	4%	4%	3%	3%
Low incidence
QIVe	4,377,497	1,755,718	48,824	6,432
QIVc	1,521,113	611,731	17,157	2,269
Absolute difference	−2,856,384	−1,143,987	−31,667	−4,163
Relative difference	65%	65%	65%	65%

GP, general practitioner; QIVc, cell-based quadrivalent influenza vaccine; QIVe, egg-based quadrivalent influenza vaccine.

Estimates were based on a total eligible US population size of 326 million.

Despite the increased vaccine costs, use of QIVc in children would be cost-saving compared with QIVe, resulting in total cost reductions of approximately $468 million in a high incidence season and $1.366 billion in a low incidence season (Table 3). Total QALY lost would be reduced by 31,213 in a high incidence season, and 60,256 in a low incidence season (Supplementary Table S7). From a payer perspective, estimated ICERs were -$16,427/QALY and -$8,100/QALY for low and high incidence seasons respectively, while estimated ICERs from a societal perspective were -$22,669/QALY and -$15,015/QALY, respectively (Table 4).

Table 3. Estimated costs of GP visits, hospital visits, loss of productivity, and vaccination in low and high incidence influenza seasons with the use of QIVe or QIVc in children aged 6 months to 17 years. Costs are stated in USD.

	GP cost	Hospitalization cost	Loss of productivity cost	Vaccine cost	Total cost
High incidence
QIVe	$8,058,853,440	$18,028,597,298	$5,828,536,703	$5,738,957,442	$37,654,944,883
QIVc	$7,837,196,815	$17,562,347,553	$5,612,681,970	$6,174,054,040	$37,186,280,378
Absolute difference	-$221,656,625	-$466,249,745	-$215,854,733	$435,096,598	-$468,664,505
Low incidence
QIVe	$721,393,770	$1,480,183,662	$577,037,180	$5,738,957,442	$8,517,572,053
QIVc	$254,287,099	$522,347,058	$200,956,706	$6,174,054,040	$7,151,644,903
Absolute difference	-$467,106,671	-$957,836,604	-$376,080,474	$435,096,598	-$1,365,927,150

GP, general practitioner; QIVc, cell-based quadrivalent influenza vaccine; QIVe, egg-based quadrivalent influenza vaccine.

Table 4. Difference in total quality-adjusted life years (QALY) lost and incremental cost-effectiveness ratio (ICER) of QIVc versus QIVe from the payer and societal perspectives, in a low or high incidence influenza season.

Perspective	Total costs	Difference in total QALY lost	ICER
High incidence
Payer	-$252,809,772	−31,213	Dominant (-$8,100)
Societal	-$468,096,505	−31,213	Dominant (-$15,015)
Low incidence
Payer	-$989,846,477	−60,256	Dominant (-$16,427)
Societal	-$1,365,927,150	−60,256	Dominant (-$22,669)

ICER, incremental cost-effectiveness ratio; QALY, quality-adjusted life years; QIVc, cell-based quadrivalent influenza vaccine; QIVe, egg-based quadrivalent influenza vaccine.

Sensitivity analysis showed that varying parameters within expected ranges still resulted in cost-saving compared with the use of QIVe in children (Supplementary Figure S2). The largest impacts were seen in varying the vaccine price and rVE of QIVc versus QIVe in both low and high incidence seasons. From a payer perspective, 71% of simulations would be cost-saving (i.e. ICER < $0/QALY), and 96% would be cost effective, at an ICER acceptability threshold of $50,000/QALY. (Supplementary Figure S3a). From a societal perspective, 80% of simulB

4. Discussion

In this analysis, the use of QIVc in children <18 years of age in the US would be cost-effective compared with use of QIVe from both a societal and payer perspective, with total cost reductions of approximately $468 million in a high incidence season and $1.366 billion in a low incidence season. Replacing QIVe with QIVc in children would result in 16,688 fewer hospital visits and 2,126 fewer deaths in the overall US population in a high incidence influenza season, with relative decreases of 65% in a low incidence season.

Egg adaptation of vaccine viruses, particularly A/H3N2, can have substantial impacts on vaccine effectiveness. In a recent systematic review and expert consensus of the impact of egg adaptations on VE, it was estimated that egg adaptation reduces the match between vaccine and circulating strains by approximately 7–21%, and reduces VE by up to 16%, with the largest impacts on effectiveness against A/H3N2 in individuals <65 years of age [38]. A/H3N2 is a major cause of influenza-related hospitalizations, causing approximately twice as many hospital admissions as the other seasonal circulating A strain, A/H1N1pdm09 [39]. Removing the potential for mismatch due to egg adaptation by generating vaccine viruses without the use of eggs (e.g. using recombinant or cell culture-derived vaccines) would have significant impacts on healthcare resource use at a population level. The direct impact of QIVc would reduce the rates of influenza-related morbidity and mortality in children themselves. Additionally, as school-age children are one of the main vectors for influenza transmission [40], use of a cell-based influenza vaccine in this age group could also reduce the rate of severe influenza in vulnerable groups such as older adults. A study in Japan evaluating the impact of childhood vaccination against influenza estimated that vaccination of children prevented approximately 37,000 to 49,000 deaths per year, mostly in older adults, equating to around 1 death for every 420 children vaccinated [40]. Therefore use of vaccines with increased effectiveness have substantial impacts, both for the vaccinated individuals and at a public health level.

Our analysis indicates higher impacts of QIVc in a low incidence influenza season compared with a high incidence season. While this may initially seem counter-intuitive, this is likely related to the impact of vaccination on the basic reproduction rate (R0), which for seasonal influenza is around 1.3, but varies between seasons [41]. In a low incidence season, R0 is already closer to 1 than in a high incidence season, therefore vaccination can potentially reduce R0 to < 1, thus stopping spread of the virus. In a high incidence season, vaccination also reduces R0, but as the baseline R0 is higher this may not result in an R0 < 1. Therefore the epidemic would continue to grow, resulting in lower overall impacts on hospitalizations and mortality than seen in a low incidence season.

While QIVc is currently available in the US for individuals aged 6 months to 64 years, usage of QIVc is not as widespread as QIVe, comprising approximately 20% of the available influenza vaccines in the 2022–2023 season [23]. Our analysis indicates that extending the use of QIVc in children would be cost-saving and result in fewer negative health outcomes in both low and high incidence seasons. Sensitivity analysis indicated that the use of QIVc in children would always be cost-saving in a low incidence influenza season and would be cost-saving 80% of the time in a high incidence season. These findings echo those from previous analysis of the impact of using QIVc in adults aged 18–64 years in the US, where 95% of simulations were cost-saving compared with QIVe, with a median ICER ranging from -$12,200/QALY to -$1,800/QALY, based on the number of years where circulating and vaccine strains were mismatched within a 3-year time horizon [24]. In other countries, the use of QIVc has also been demonstrated to be cost-effective compared with egg-based vaccines. In Germany, replacing QIVe with QIVc in individuals aged ≥9 years was estimated to prevent 2,091 hospitalizations and 103 deaths annually, with an ICER of €2,285/QALY [42]. Similarly, in Spain, the use of QIVc in children and adults was estimated to reduce emergency room visits by 1,015 and hospitalizations by 88 per year, resulting in a cost saving of €3.4 million [43]. QIVc has also been shown to be cost-effective in Argentina and Canada, with higher benefits in seasons where there is mismatch between vaccine and circulating strains caused by egg adaption, where effectiveness of QIVe would be expected to be reduced [35,44].

As with all simulations, our analysis has a number of limitations. Firstly, we assumed that all adults aged 18–64 years would receive QIVc. While this is not currently the case in the US, given the lower VE of QIVe compared with QIVc in both the adult and pediatric populations, the use of QIVc only in the model likely results in an underestimate of the burden of influenza on healthcare resource utilization and thus the potential impact of QIVc use in children. Results have been previously published on the cost-effectiveness of QIVc in adults [44], therefore we restricted this analysis to children to demonstrate any incremental benefit of extending QIVc coverage in the pediatric population. As with all models, the results are limited by the input parameters and influenza incidence and vaccine effectiveness vary annually. In particular, few studies have published estimates of rVE of QIVc vs QIVe specifically in pediatric age groups. To mitigate this variation, we performed sensitivity analysis performed across a broad range of input parameter values which confirmed the overall results, indicating that robustness of the conclusions over a range of influenza scenarios. Finally, this analysis was performed based on US-based parameters; while estimates of cost-effectiveness may be different for other regions with different vaccine and healthcare costs, it is likely that use of QIVc in this age group will remain cost-effective across many high income countries.

5. Conclusions

In summary, the use of QIVc instead of QIVe in children aged 6 months to 17 years in the US would be cost-saving from both a payer and societal perspective. While the incremental benefits will vary between influenza seasons, sensitivity analysis indicated that QIVc would be cost-effective in > 95% of simulations, with reductions in cases, hospitalizations, and influenza-related mortality of up to 65%.

Declaration of interest

VH Nguyen and SI Pelton’s work for this study was funded by Seqirus U.S.A. Inc. J F Mould-Quevedo is an employee of Seqirus U.S.A. Inc., and a CSL shareholder. Seqirus has manufactured and marketed QIVc since the 2017–2018 influenza season. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or material discussed in the manuscript apart from those disclosed.

Reviewer disclosures

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

Author contributions

All authors substantially contributed to the conceptualization, methodology, and analysis of the study, together with the design, review and revision of the manuscript. Modelling was performed by Van Hung Nguyen.

Previous conference presentation

The data from in this manuscript have been previously presented at ID Week 2023: Mould-Quevedo, J.; Nguyen, V.H. The Value of the Influenza Cell-Based Vaccine in the Pediatric Population. A Dynamic Transmission Modelling Approach in the U.S. In Proceedings of the ID Week 2023, Boston, MA, U.S.A., 11–15 October 2023.

Acknowledgments

The authors would like to thank Dr J Engelmoer (Sula Communications BV, Utrecht, The Netherlands) for editorial assistance in preparation of this manuscript.

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/14760584.2023.2295014

Additional information

Funding

This manuscript was funded by CSL Seqirus Inc.