Document text
Economics of Preventing Respiratory Syncytial
Virus Disease among US Infants by Maternal
Vaccination Prior to Birth
A SUMMARY REPORT COMPARING MODELS FROM:
Pfizer AND University of Michigan and CDC
Ismael R. Ortega -Sanchez, PhD
NCIRD/CDC
ACIP Meeting, September 22, 2023
1Disclaimer : The findings and conclusions in this report are those of the authors and do not necessarily represent the views of
the Centers for Disease Control and Prevention.
National Center for Immunization & Respiratory Diseases
Conflict of interest
•Pfizer model : Amy Law et al., [complete authors list and affiliations, upon
request ]
•Pfizer manufactures RSVpreF vaccine
•Policy Analysis Inc. (Boston, MA, US) was funded by Pfizer
•UM-CDC model : David W Hutton et al. from University of Michigan,
Ismael R Ortega -Sanchez et al. from CDC [complete authors list and affiliations,
upon request ]
•All authors: No conflicts of interest
2
Economic analysis
Policy question: Should Pfizer RSVpreF vaccine be recommended for pregnant mothers to be given
during 32 through 36 weeks gestation to prevent RSV lower respiratory tract infection in infants?
Question : Is vaccinating pregnant mothers prior to birth to protect infants against RSV disease cost-
effective ?
Comparator Intervention
Base -case scenario: What is the incremental cost-effectiveness of vaccinating pregnant mothers 32 -36wGA and
≥2 weeks prior to birth with RSVPreF vaccine relative to “No vaccination”?
3Unvaccinated
Mothers +
Standard of care (SoC)
for InfantsVaccination of
pregnant mothers
prior to birth
Standard of Care (SoC) = Palivizumab only for infants eligible as per AAP recommendations, and no immunization for all other pr e-term and full term infants
Focus on key features for model comparison
•Modeling approach
•Targeted population(s)
•Perspective (healthcare vs. societal)
•Intervention strategies and comparators
•Inputs for RSV disease burden, vaccine efficacy, and costs
•Incidence of RSV disease, rates of outcomes
•Direct and indirect costs of RSV disease
•Intervention: efficacy, duration of protection, safety and program costs
•Assumptions
•Strong, influential assumptions
4
5Modeling design and assumptions
Pfizer UM-CDC
Static analytical decision -making models ✔ ✔
Sensitivity analyses (and probabilistic simulation) ✔(✔) ✔
Hypothetical population: All pregnant mothers, all year round ✔ ✔
Time Frame: First year after birth ✔ ✔
Analytic Horizon: One year (for temporary disability) and Life
Expectancy (for premature infant mortality)✔
✔✔
✔
Discount rate: 3% ✔ ✔
Year of economic outcomes measured: 2022 ✔ ✔
Societal perspective (and healthcare perspective) ✔(✔) ✔(✔)
6Inputs and main outcomes
Prevention of:
•RSV LRTI ED/OC visits
•RSV LRTI hospitalizations
•RSV-associated deaths
QALYs saved
$/QALY saved
Number needed to
vaccinate (NNV) to avert an:
•RSV LRTI hospitalization
•RSV-associated deathPfizer UM-CDC
✔ ✔
✔ ✔
✔ ✔
✔ ✔
✔ ✔
✔ ✔
✔ ✔
HCRU = health care resource use, ED= Emergency department, OC= outpatient clinic, LRTI= Lower respiratory tract infection,
QALY= quality -adjusted life year
$11 $33 $68 $311
$13 $104
$- $50 $100 $150 $200 $250 $300 $350
Outpatient ED Inpatient ICU Inpatient
DayICU DayCost per Event Averted ( thousands )UM-CDC: Base case estimates for maternal vaccination,
Vaccination Window (VW) 32 -36wGA, vaccine cost $295/dose
7Cost per type of health outcome prevented (in thousands ) Summary outcomes Base -Case
$/QALY gained $400,304
$/RSV -associated ED/OC visit
averted $32,652 / $11,337
$/RSV -associated
LRTI hospitalization averted$68,423
$/RSV -assoc. death averted>$71.5Million
NNV to avert an RSV -associated
ED/ OC patient115 / 40
NNV to avert an RSV -associated
LRTI hospitalization242
NNV to avert an RSV -associated
death241,989
Assuming 50% uptake in vaccinated group
VW = vaccination window
wGA = weeks gestational age
LRTI= lower respiratory tract infection
OC = office clinic for outpatient care
ED = emergency department
NNV = Number needed to vaccinate
Pfizer model : Base case estimates for maternal vaccination,
VW 32 -36wGA, vaccine cost $295/dose & PSA
8Summary outcomes Base -Case
$/QALY gained $84,690
$/RSV -associated ED/OC visit
averted$6,145/ $2,101
$/RSV -associated LRTI
hospitalization averted$14,932
$/RSV -associated death averted>$7.7Million
NNV to avert an RSV -associated
ED/OC patient66 / 22
NNV to avert an RSV -associated
LRTI hospitalization159
NNV to avert an RSV -associated
death82,243
Probabilistic sensitivity analysis (PSA)87.1%
65.1%
34.7%
Difference in total QALYsDifference in total costs ( in millions )
Base -case Assuming 54.9% uptake in vaccinated group and 93.2% born ≥2 weeks after adm.
VW = vaccination window
wGA = weeks gestational age
LRTI= lower respiratory tract infection
OC = office clinic
ED = emergency department
NNV = Number needed to vaccinate
WTP = Willingness to pay
Pfizer and UM-CDC models comparison: base -case
selected outcome ratios for maternal vaccination
9UM-CDC model
Price per dose $295
VW=32 -36wGA
Year -round
administrationPfizer model
Price per dose $295
VW=32 -36wGA
Year -round
administration
$ / QALY gained $400,304 $84,690
$ / RSV LRTI hospitalization averted $68,423 $14,932
$ / Death averted among RSV LRTI hospitalized infants >$71.5Million >$7.7Million
NNV to prevent a
RSV LRTI associated hospitalization 242 159
Death among RSV LRTI hospitalized infants 241,989 82,243
Assuming 50% (UM -CDC) and 54.9% (Pfizer) uptake in vaccinated group and 100% (UM -CDC) to 93.2% (Pfizer) born ≥2 weeks after administration.
VW = Vaccination window
LRTI= lower respiratory tract infection
NNV = Number needed to vaccinate
Note: Both models vaccination window = 32 -36wGA only ( wGA = weeks gestational age)
$0 $100 $200 $300 $400 $500 $600 $700 $800
Probability of Prematurity
RSV QALYS Lost
RSVpreF vaccine cost/dose
Disease-specific inpatient costs (per inpatient case)
Vaccination Efficacy, Hospitalized RSV LRTI through 180 days
RSV-related QALYs lost Outpatient Child
Vaccination Efficacy, RSV MA-LRTI through 180 days
Proportion of RSV infections with an LRTI diagnosis…
RSV-related QALYs lost Outpatient Caregiver
RSV-related QALYs lost ED ChildIncremental Cost -Effectiveness Ratio ( in thousands )
Low
HighUM-CDC model : One-way Sensitivity Analyses
Base case: $ 400,304/ QALY saved, vaccine price $ 295/dose
10MA= Medically -attended
LRTI= Lower respiratory tract infection
QALY= Quality adjusted life year$1.3
Million
Base -case $400,304
Pfizer model : One-way Sensitivity Analyses
Base case: $84, 690/QALY saved, Vaccine cost $295/dose
11CFR= Case fatality ratio
DoP = Duration of protection
VE= Vaccine efficacy
PT = Preterm
FT = Full termRSV-H = RSV -associated hospitalization
RSV-ED = RSV -associated emergency department care
RSV-OC = RSV -associated outpatient care15) VE lower bound
14) RSV-H medical care cost lower bound
13) US healthcare system perspective
12) DoP = 6 months
11) RSV-H CFR lower bound (overall = 0.1%)
10) Vaccine uptake 32 -36 wGA (uniform)
9) Disutilities form Glasser (RSV -H) and
Reigner (RSV -ED, RSV -OC)
8) Uptake = 100%
7)Base case
6) VE for late PT = for FT
5) Palivizumab scenario
4) DoP = 12 months
3) RSV-H CFR upper bound (overall=0.53%)
2) RSV-H medical care cost upper bound
1) VE upper bound
Base -case $84,690
Sensitivity / Scenario AnalysisCost per QALY
Pfizer and UM-CDC models comparison:
Selected influential inputs
12•RSV-hospitalization risk and RSV seasonality
•Vaccine efficacy
•Duration of protection and waning
•RSV Case fatality rate
•Medical cost of RSV hospitalization, ED and Outpatient care
•Vaccine associated adverse events
•Quality of life lost by patients and caregivers
Pfizer and UM-CDC: comparison of base -case risk of
RSV-related hospitalization by age and RSV Seasonality
13Risk of RSV Hospitalization : Pfizer and UM-CDC: Based on laboratory -confirmed RSV-associated hospitalization by age in months from New Vaccine Surveillance
Network (NVSN) data for children under 2 years of age (December 2016 to September 2020). Risk estimates are based only on RSV cases that manifest as LRTI.
RSV Seasonality: Pfizer :based on Midgley et al. J Infect Dis 2017. (data based in NRVSS) https://pubmed.ncbi.nlm.nih.gov/28859428/#full -view -affiliation -1
UM-CDC: based on National Respiratory and Enteric Virus Surveillance System (NREVSS) (2015 -2019)0%5%10%15%20%25%30%
Apr May Jun Jul Aug Sep Oct Nov Dec Jan Feb MarFraction of Annual Infections Pfizer
UM-CDC
00.0050.010.0150.020.0250.030.035
0 1 2 3 4 5 6 7 8 9 10 11Risk of RSV LRTI hospitalization
Age (in months)
Pfizer and UM-CDC:Differences in initial vaccine
efficacy
14NOTE: None of Pfizer Phase 3 (i.e., MATISSE) endpoint definitions overlapped ideally with the case definition used for the US burden data.
a. CDC: Average between efficacy for full term and preterm reported from Phase 3 were used. VE for RSV -LRTI hospitalization is an average over months 0-6 reported
in Phase 3 trial , and RSV -positive MA -LRTI for VE against RSV -LRTI in the ED and outpatient. Kampmann et al New England Journal of Medicine . 2023 Apr
b. Pfizer: Efficacy against severe RSV -positive MA -LRTI was used as a proxy for VE against RSV -LRTI requiring hospitalization, and efficacy against RSV -positive MA -LRTI
was used as a proxy for VE against RSV -LRTI treated in the ED. Pfizer: Kampmann et al New England Journal of Medicine . 2023 Apr
c. VE for late preterm infants was assumed to be 83.3% of corresponding values for full term infants. VE for URTI was assumed equal to VE for MA LRTI in ED care
d. Based on overall respiratory tract efficacy from phase 3 trial ( Kampmann et al New England Journal of Medicine . 2023 Apr )UM-CDCPfizer
Full Term Late Preterm
Initial vaccine efficacy against
RSV LRTI hospitalization (%)Month 0 = 81.0
Average Month 0-6 = 56.8 a88.1
Severe MA
RSV-LTRIb73.4
Severe MA
RSV-LRTI b, c
Initial vaccine efficacy against
Medically attended RSV associated LRTI for ED
and Outpatient care (%)Month 0 = 73.0
Average Month 0 -6 = 51.3 a47.6b39.7 c
For Scenario Analysis only : Initial efficacy
against RSV URTI treated in Outpatient care (%)37.9 d47.6b39.7 c
0%10%20%30%40%50%60%70%80%90%100%
0 3 6 9 12Efficacy
Month
Against medically-attended RSV-associated LRTI
Against hospitalizationaverage 6 -month efficacy =
efficacy against hospitalization
or medically -attended RSV -
associated LRTI
Zero
efficacy
The pink -shaded areas denote a higher level of uncertainty of the waning assumption beyond available phase 3 data
UM-CDC: Average between efficacy for full term and preterm were used for DoP for RSV -LRTI hospitalization over months 0 -6 and RSV -positive MA -LRTI for DoP
against RSV -LRTI in the ED and outpatient also over months 0 -6. Both r eported in Phase 3 trial , Kampmann et al New England Journal of Medicine . 2023 AprPfizer and UM-CDC: Assumption on duration of
protection ( DoP)
15Pfizer UM-CDC
VE against ED or OC VE against hospitalization4 trial
endpoints
81.0
73.0
16UM-CDC Pfizer
Case fatality rate (CFR) among RSV -
hospitalized infants <12 months of age0.10%
(0.04% -0.20%)a0.10% (full term)
0.80% (all preterm) b
Medical costs per RSV hospitalization$11,487
($4,804 -$86,646)cAverage: $20,483 d
$13,171 –$33,876 (full term)
$19, 415 –$51,343 (late preterm)
Medical costs per RSV ED visit$563
($544 -$581)cAverage: $1,840 d
$1,620 –$2,520 (full term)
$1,787 –$2,779 (late preterm)
Medical costs per RSV outpatient visit$82
($46 -$118)cAverage: $348 d
$292 –$730 (full term)
$328 –$823 (late preterm)
a. RSV mortality per hospitalization: 0.10% (range 0.04 -0.20%) in 0 -5months, 0.10% (range 0.04 -0.20%) in 6 -11months.
b. Case fatality due to RSV -Hospital in full term infants per 100 cases = 0.1 (based on Li et al CEA -RSV in children 2022). I n pre -term infants CFR per 100= 0.8 (assumes 16.3% of all RSV -Hosp
are among preterm)
c. Adapted from Bowser et al., J Infect Dis . 2022 Aug 15; 226(Suppl 2): S225 –S235 (A systematic review using studies from 2014 -2021. Cost updated to 2022 using the GPD de flator)
d. Source: Pfizer data on file. Costs in the base -case varied by age and term at birth. Weighted average cost among full and late preterm infants in commercially insured and Medicaid
populationsPfizer and UM-CDC models comparison:
Differences in key inputs
Pfizer and UM-CDC: Vaccine -associated adverse
events
17UM-CDC Pfizer Source
Rate of injection site reaction0.41
(0.38 –0.44)0.41
(0.38 –0.44}Pfizer Phase III Trial
Probability of healthcare visit, given
injection site reaction0.02
(0.015 –0.025 )0.02
(0.015 –0.025 )Curran, 2020
Cost of outpatient visit$367.76
(23.15 –1,758)$367.76
(23.15 –1,758)(Deluca, 2023) also in Pfizer CEA
technical report (August 2023)
Recipient time, physician office for
injection site reaction (hours)2
(1 –3)n/r Assumption
Hypothetical serious adverse event0.000001
(0 -0.0002)n/rBase: Prosser, 2006
High: 95% CI Phase 3 data for
RSV adult vaccines
Potential risk increase of prematurity0.0%
(0 –2%)n/rPfizer Phase III Trial
Kampmann et al New England
Journal of Medicine . 2023 Apr
Note: Values in bold are used for the base -case scenario. Range values in parenthesis for sensitivity analyses
n/r = not reported
Pfizer and UM-CDC: Quality of life lost ( in days )
by RSV LRTI outcome for patients & caregivers
18UM-CDC* Pfizer**
Outpatient: Child3.1
(1.8 -16.6)2.22
Outpatient: Caregiver1.5
(0 -9.1)1.5
ED: Child4.9
(2.9 –16.6)2.22
ED: Caregiver2.5
(0 -9.1)2.5
Hospitalized: Child6.2
(3.7 –26.5)5.7
Hospitalized: Caregiver2.4
(0 –13.6)2.4
Scenario Analysis : Vaccine -related increased
risk of prematurity ***11
(0 –438)n/r
* Base case values are from EGlasser et al 2022, lower values are based in Regnier et al 2013, higher values are based in JIVE (unpublished data)
**Patient values are base in Roy 2013, Caregiver values are based on Hutton. Economic Analysis of Nirsevimab in Pediatric Populations. ACIP; February 23, 2023.
Note: Values in bold are used for the base -case scenario. Range values in parenthesis for sensitivity analyses
*** Sources: Werner, et al 2015. Petrini et al, 2008, Hirvonen et al, 2014, Crump et al, 2021, Darcy -Mahoney et al, 2016, Carroll et al, 2009, Payakachat et al, 2014
n/r = not reported
19UM-CDC
UM-CDC model Base -case$400,304
ACost of RSV -LRTI hospitalization: $20,000 or $50,000 (i.e., 85% to 450% increase, base -case
cost= $11,487)$350,500 -$174,987
BUM-CDC model with same VE duration of protection as Pfizer : 6 months slowly, linearly
declining efficacy and declining faster after month 6 reaching 0% at month 9$286,769
CCFR among RSV LRTI hospitalization: 1% (base -case = 0.1%)$122,539
DCombining A and B vs. Combining A, B and C$233,736 -$52,108
EIncrease in the risk of prematurity in one-or two -percent points (base -case risk of prematurity
after vaccination = 0% increase )$874,609 -$1.3 Million
FVaccine administration timing: September -January vs. February -July (base -case, year -round) <$200,000 -$Millions
GVaccine cost = $50/dose or $500/dose (base -case vaccine cost = $295/dose) $65,304 -$680,609
HVE for URTI = 37.9% (Base -case = 0% VE against URTI) $279,490Pfizer and UM-CDC models comparison:
$/QALY for selected scenarios in UM-CDC model
20Pfizer
Pfizer model base -case $84,690
VE for late preterm infants assumed same as for full term infants a$73,404
Vaccine Efficacy : 80% or 120% of base case values $155,834 -$43,813
Trial -based VE over 6 months, then 0% VEor linear waning to 0% VE at 12 months$111,473 -$61,925
Overall CFR among RSV hospitalized: 0.1% or 0.53% b$102,431 -$52,000
Costs of RSV -Hospitalization 80% or 120% of base -case)$118,625 -$50,755
Vaccination window: 32 -36 wGA only ( uniform: 20% in each week ) c$91,036
Maternal vaccine and palivizumab in <1y infants vs.palivizumab only for prevention of RSV$66,796
a.Base -case assumes for late pre -term a VE =83% of the VE for full term (VE = Vaccine efficacy)
b.CFR among RSV hospitalized: 0.1% (RR of RSV -H death = 1 for full term and preterm infants). For CFR 0.53% (using CFR values fro m full term = 0.3% and preterm = 1.7%)
c.wGA = weeks gestational age Pfizer and UM-CDC models comparison:
$/QALY for selected scenarios in Pfizer model (I)
21Pfizer
APfizer model with selected UM-CDC Inputs $343,000
BPfizer model with selected UM-CDC Inputs except for Pfizer initial VE
assumptions$265,000
CPfizer model with selected UM-CDC Inputs except for Pfizer initial VE
assumptions and Pfizer DoP $173,000
DPfizer model with selected UM-CDC Inputs except for Pfizer initial VE
assumptions, Pfizer DoP and Pfizer Medical costs$83,000
EPfizer model with selected UM-CDC Inputs except for Pfizer initial VE
assumptions, Pfizer DoP , Pfizer Medical costs and Palivizumab use$67,000
DoP = Duration of protection
VE = vaccine efficacyPfizer and UM-CDC models comparison:
$/QALY for selected scenarios in Pfizer model (II)
Limitations
22•Factors not considered that may result in overestimating the ICER
(underestimating the cost -effectiveness) of maternal vaccination
•In base -case: both models assumed
•No protection against URTI
•No benefits of vaccination for vaccinated pregnant women
•No out -of-pocket cost accrued by caregivers during infants RSV illness
•Neither model included RSV -related costs incurred after discharge from an
RSV-associated hospitalization or emergency department visit:
•Productivity losses incurred by caregivers after discharge
•Both models assumed no indirect effects of vaccination (i.e., no protection
against RSV transmission)
Conclusion
23•Differences in key inputs among Pfizer and UM-CDC models explain differences in results:
•Initial vaccine efficacy and assumptions about protection waning
•Medical costs
•Quality of life associated with RSV LRTI outcomes for patient and caregivers
•Vaccine related adverse events
•In addition, the UM-CDC also identified two important factors that could drive the results
•Hypothetically severe vaccine -associated adverse events
•Timing of vaccination to RSV season
•Base -case in both models:
•Maternal vaccination would significantly reduce RSV disease burden and costs in infants
•Data from clinical trials used in both models support the reduction in RSV disease and
associated costs
•Economic value of vaccinating pregnant people to protect infants could increase costs
•Reasonable vaccine price and duration of protection combined with careful design of
seasonal interventions would determine the cost-effectiveness value of routine
vaccination of pregnant people during the 32 -36wGA
Acknowledgements
From NCIRD/CDC
•Jefferson Jones
•Mila M. Prill
•Katherine E. Fleming -Dutra
•Jamison Pike
•Andrew Leidner
•Meredith McMorrow
Also:
•Maternal/Pediatric RSV working group members
24
End of Summary