Document text
Summary of three economic analyses of the use of
20-valent pneumococcal conjugate vaccine (PCV20)
in children in the United States
Diepreye V. Ayabina
Division of Bacterial Diseases
CDC/NCIRDAdvisory Committee on Immunization Practices
June 22, 2023
Acknowledgements
•This presentation summarizes work conducted by three modeling teams
•Tulane -CDC team
•Charles Stoecker (Tulane University), Miwako Kobayashi (CDC), Andrew Leidner (CDC), Diepreye
Ayabina
•Pfizer team
•Mark Rozenbaum , Liping Huang, Alejandro Cane, Erica Chilson, Adriano Arguedas, Maria J. Tort,
Raymond Farkouh , Vincenza Snow
•Merck team
•Min Huang, Jessica Weaver, Elamin Elbasha
Views and opinions expressed in this presentation are the authors and do not necessarily represent the
views and opinions of the Centers for Disease Control and Prevention.2
Conflicts of interest statement
•Diepreye Ayabina: None.
•Tulane -CDC team: None.
•Pfizer team:
•Pfizer manufactures the PCV13 and PCV20 vaccines.
•Merck team:
•Merck manufactures the PCV15 and PPSV23 vaccines
3
Terminology
Abbreviation Full term/Meaning
AOM Acute otitis media
CMC Chronic medical conditions but not immunocompromised1
IC Immunocompromising conditions2
ICER Incremental cost -effectiveness ratio
IPD Invasive pneumococcal disease
PCV13 13-valent pneumococcal conjugate vaccine
PCV15 15-valent pneumococcal conjugate vaccine
PCV20 20-valent pneumococcal conjugate vaccine
PPSV23 Pneumococcal polysaccharide vaccine (23 serotypes)
QALYs Quality -adjusted life -years
1Includes chronic heart, lung, and liver diseases, diabetes.
2Includes chronic renal failure, nephrotic syndrome, congenital immunodeficiency, congenital or acquired asplenia, or splenic dysfunction, diseases associated with treatment of immunosuppressive drugs
or radiation therapy such as Hodgkin disease, leukemia, lymphoma, malignant neoplasm and solid organ transplant, HIV and sick le cell disease. 4
Outline
•Introduction
•Policy question #1: (Routine use of PCV20)
•Model description
•Estimated health outcomes and cost results
•Overview of cost -effectiveness results
•Model assumptions driving differences + sensitivity analysis
•Policy question #2: (Use of PCV20 in with underlying medical conditions that increase the risk of
pneumococcal disease)
•Estimated health outcomes and cost results
•Overview of cost -effectiveness results
•Model assumptions driving differences + sensitivity analysis
•Summary
5
Introduction
•This presentation describes three cost -effectiveness models that have been used to examine the costs and
benefits of including PCV20 as an option in the childhood immunization schedule developed.
•Tulane -CDC, Pfizer, and Merck
•Results of this presentation focus on routine use of PCV20 among <2 -year -olds, and among 2 -18-year -olds
with underlying conditions.
•Assessments of a supplemental PCV20 dose in children who completed the PCV series with PCV13 or PCV15 are not included
•All three reports went through the CDC economic review following the ACIP Guidance for Health Economics
Studies
•Completion of the economic review does not confer any explicit or implied approval of the model
6
Policy question 1
Routine use of PCV20 in children
Policy question #1
Should PCV 20berecommended as an option for pneumococcal
conjugate vaccination according tocurrently recommended dosing and
schedules, for children aged <2 years inthe United States?
𝐶𝑜𝑠𝑡𝑠𝐼𝑛𝑡𝑒𝑟𝑣𝑒𝑛𝑡𝑖𝑜𝑛 −𝐶𝑜𝑠𝑡𝑠 𝐶𝑜𝑚𝑝𝑎𝑟𝑎𝑡𝑜𝑟= 𝑰𝒏𝒄𝒓𝒆𝒎𝒆𝒏𝒕𝒂𝒍𝒄𝒐𝒔𝒕 −𝒆𝒇𝒇𝒆𝒄𝒕𝒊𝒗𝒆𝒏𝒆𝒔𝒔𝒓𝒂𝒕𝒊𝒐(𝑰𝑪𝑬𝑹)𝑂𝑢𝑡𝑐𝑜𝑚𝑒𝑠 𝐼𝑛𝑡𝑒𝑟𝑣𝑒𝑛𝑡𝑖𝑜𝑛 −𝑂𝑢𝑡𝑐𝑜𝑚𝑒𝑠 𝐶𝑜𝑚𝑝𝑎𝑟𝑎𝑡𝑜𝑟Intervention (dosing) Comparator (dosing)
PCV20 (3+1) PCV13 (3+1) /PCV15 (3+1)
8
Description of models
Policy question #1
Model characteristics Tulane -CDC Merck Pfizer
Cohort type Single birth cohort Single birth cohort Multi -cohort
Analytic model time frame 17 years 100 years 10 years
Include indirect effects Yes Yes ( only to IPD incidence) Yes
Base case perspective Societal Societal Societal
Currency year 2022 $ US 2022 $ US 2022 $ US
aVaccine price PCV13: $184
PCV15: $183
PCV20: $213PCV13: $192
PCV15: $189
PCV20: $213PCV13: $182
PCV15: $184
PCV20: $204
Other vaccine costs per dose Admin: $30; travel: $36 Admin: $19 Admin: $18
aAll three models used blended vaccine prices (vaccine price was weighted average of private and public market prices, weights wer e private and public market shares, which
varied across the models).
9
Health outcomes and cost resultsa
PCV20 vs PCV15, base case
PCV20 vs PCV15
Outcomes and cost Tulane -CDC Merck++Pfizer
Health outcomesIPD cases -360 -400 -30,000*
Pneumonia cases -13,000 -2,300 -1,400,000*
AOM cases -150,000 -201,000 -3,000,000*
Deaths -160 -28 -9,700*
QALYs 1,900 1,400 280,000
Costs ($ millions)Cost of disease + sequelae -210 -180 -12,000
Vaccine costs 440 330 2,100
Total cost 230 150 -10,000
10aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
All other values were discounted at 3%.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15.
++Pfizer base case results involve vaccination of 10 cohorts of children, results from single cohort shown in supplemental slid e.
PCV20 vs PCV15
Outcomes and cost Tulane -CDC Merck++Pfizer
Health outcomesIPD cases -360 -400 -30,000*
Pneumonia cases -13,000 -2,300 -1,400,000*
AOM cases -150,000 -201,000 -3,000,000*
Deaths -160 -28 -9,700*
QALYs 1,900 1,400 280,000
Costs ($ millions)Cost of disease + sequelae -210 -180 -12,000
Vaccine costs 440 330 2,100
Total cost 230 150 -10,000
11aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
All other values were discounted at 3%.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15
++Pfizer base case results involve vaccination of 10 cohorts of children, results from single cohort shown in supplemental slid e.Health outcomes and cost resultsa
PCV20 vs PCV15, base case
PCV20 vs PCV15
Outcomes and cost Tulane -CDC Merck++Pfizer
Health outcomesIPD cases -360 -400 -30,000*
Pneumonia cases -13,000 -2,300 -1,400,000*
AOM cases -150,000 -201,000 -3,000,000*
Deaths -160 -28 -9,700*
QALYs 1,900 1,400 280,000
Costs ($ millions)Cost of disease + sequelae -210 -180 -12,000
Vaccine costs 440 330 2,100
Total cost 230 150 -10,000
12aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
All other values were discounted at 3%.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15.
++Pfizer base case results involve vaccination of 10 cohorts of children, results from single cohort shown in supplemental slid e.Health outcomes and cost results
PCV20 vs PCV15, base casea
PCV20 vs PCV15
Outcomes and cost Tulane -CDC Merck++Pfizer
Health outcomesIPD cases -360 -400 -30,000*
Pneumonia cases -13,000 -2,300 -1,400,000*
AOM cases -150,000 -201,000 -3,000,000*
Deaths -160 -28 -9,700*
QALYs 1,900 1,400 280,000
Costs ($ millions)Cost of disease + sequelae -210 -180 -12,000
Vaccine costs 440 330 2,100
Total cost 230 150 -10,000
13aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
All other values were discounted at 3%.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15.
++Pfizer base case results involve vaccination of 10 cohorts of children, results from single cohort shown in supplemental slid e.Health outcomes and cost resultsa
PCV20 vs PCV15, base case
PCV20 vs PCV15
Outcomes and cost Tulane -CDC Merck++Pfizer
Health outcomesIPD cases -360 -400 -30,000*
Pneumonia cases -13,000 -2,300 -1,400,000*
AOM cases -150,000 -201,000 -3,000,000*
Deaths -160 -28 -9,700*
QALYs 1,900 1,400 280,000
Costs ($ millions)Cost of disease + sequelae -210 -180 -12,000
Vaccine costs 440 330 2,100
Total cost 230 150 -10,000
14aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
All other values were discounted at 3%.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15
++Pfizer base case results involve vaccination of 10 cohorts of children, results from single cohort shown in supplemental slid e.Health outcomes and cost resultsa
PCV20 vs PCV15, base case
Model assumptions
*Indirect effects of pediatric PCV20 vaccination on adult pneumonia incidence
ModelDisease
outcomes
impacted by
indirect effectsDetails on the indirect effects methodologyAdjust for
PCV20 use
among 65+Data years
for indirect
effect on
pneumoniaApproximate
% of PCV20
only type
pneumoniaApproximate
reduction in
PCV20 -only
pneumonia
per cohort
Merck •IPD•Assume 8% reduction per year, with a max of
33% achieved in year 4 and maintained until the
end of the cohortNo NA (IPD considered only)
•Higher indirect effects as a result of PCV20 use would lead to PCV20 being more cost -effective when compared to other PCV vaccin es.
15
Model assumptions
*Indirect effects of pediatric PCV20 vaccination on adult pneumonia incidence
ModelDisease
outcomes
impacted by
indirect effectsDetails on the indirect effects methodologyAdjust for
PCV20 use
among 65+Data years
for indirect
effect on
pneumoniaApproximate
% of PCV20
only type
pneumoniaApproximate
reduction in
PCV20 -only
pneumonia
per cohort
Merck •IPD•Assume 8% reduction per year, with a max of
33% achieved in year 4 and maintained until the
end of the cohortNo NA (IPD considered only)
Tulane -
CDC•IPD
•Pneumonia
•AOM•Based on regression model estimates
•Estimated impact per vaccinated cohort
•Indirect effect benefit attributed to birth cohort
of children who receive PCV20Yes 2013 -2014a 0.6 to 2.4% 3.2%
Pfizer•IPD
•Pneumonia
•AOM•Based on observed reduction in all -cause
pneumonia Percentage of PCV20 type disease
Gradual increase in indirect effect over 10 yearsYes2008/2009 to
2018aGreater than
1 to 7%0.5 to 3.4%b
*Higher indirect effects as a result of PCV20 use would lead to PCV20 being more cost effective when compared to other PCV vac cines. The Tulane -CDC and Pfizer models both accounted for the direct protection in all
adults.
aThe Tulane -CDC model estimated a monthly reduction in disease due to indirect effects, using data from 2013 and the first half of 2 014 (Pilishvili 2018a; Pilishvili et al. 2018b). The Pfizer model used a total reduction in
disease from 2008/2009 to 2018 (Tong et al. 2018, Table 3).
b.These are approximations based on calculations by the presenting author, based on each cohort in the Pfizer model contributing to i ndirect effects for 5 years. 16
Model assumptions
*Indirect effects of pediatric PCV20 vaccination on adult pneumonia incidence
ModelDisease
outcomes
impacted by
indirect effectsDetails on the indirect effects methodologyAdjust for
PCV20 use
among 65+Data years
for indirect
effect on
pneumoniaApproximate
% of PCV20
only type
pneumoniaApproximate
reduction in
PCV20 -only
pneumonia
per cohort
Merck •IPD•Assume 8% reduction per year, with a max of
33% achieved in year 4 and maintained until the
end of the cohortNo NA (IPD considered only)
Tulane -
CDC•IPD
•Pneumonia
•AOM•Based on regression model estimates
•Estimated impact per vaccinated cohort
•Indirect effect benefit attributed to birth cohort
of children who receive PCV20Yes 2013 -2014a 0.6 to 2.4% 3.2%
Pfizer•IPD
•Pneumonia
•AOM•Based on observed reduction in all -cause
pneumonia Percentage of PCV20 type disease
Gradual increase in indirect effect over 10 yearsYes2008/2009 to
2018aGreater than
1 to 7%0.5 to 3.4%b
*Higher indirect effects as a result of PCV20 use would lead to PCV20 being more cost effective when compared to other PCV vac cines. The Tulane -CDC and Pfizer models both accounted for the direct protection in all
adults.
aThe Tulane -CDC model estimated a monthly reduction in disease due to indirect effects, using data from 2013 and the first half o f 2014 (Pilishvili 2018a; Pilishvili et al. 2018b). The Pfizer model used a total reduction in
disease from 2008/2009 to 2018 (Tong et al. 2018, Table 3).
b.These are approximations based on calculations by the presenting author, based on each cohort in the Pfizer model contributing to i ndirect effects for 5 years. 17
Cost effectiveness results: base case
PCV20 vs PCV13
aCost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.Intervention Comparator Model Cost per QALY gained
PCV20 PCV13 Tulane -CDC 57,000
MerckaCost -saving
PfizeraCost -saving
PCV20 PCV15Tulane -CDC 125,000
Merck 105,003
PfizeraCost -saving
18
Cost effectiveness results: base case
PCV20 vs PCV15
aCost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.Intervention Comparator Model Cost per QALY gained
PCV20 PCV13 Tulane -CDC 57,000
MerckaCost -saving
PfizeraCost -saving
PCV20 PCV15Tulane -CDC 125,000
Merck 105,003
PfizeraCost -saving
19
Sensitivity analysis (Policy question #1)
Cost -effectiveness of routine vaccination with PCV20 for children
Intervention Comparison ModelCost per QALY gained
Base case Rangeb
PCV20 PCV13 Tulane -CDC 57,000 NA
Merck Cost -saving NA
Pfizer Cost -saving All cost -saving
PCV20 PCV15 Tulane -CDC 125,000 [31,000 210,000 ]
Merck 105,000 [8,000 dominateda]
Pfizer Cost -saving [Cost -saving 60,000 ]
aIn one scenario in the Merck model, PCV15 was cost -saving relative to PCV20 (or PCV20 was dominated by PCV15), this scenario as sumed re -emergence of IPD associated with certain
serotypes that PCV20 missed the trial primary endpoint.
b The ranges presented are obtained from the scenario analyses and probabilistic sensitivity analysis.
Sensitivity analysis (Policy question #1)
Cost -effectiveness of routine vaccination with PCV20 for children
Intervention Comparison ModelCost per QALY gained
Base case Rangeb
PCV20 PCV13 Tulane -CDC 57,000 NA
Merck Cost -saving NA
Pfizer Cost -saving All cost -saving
PCV20 PCV15 Tulane -CDC 125,000 [31,000 210,000 ]
Merck 105,000 [8,000 dominateda]
Pfizer Cost -saving [Cost -saving 60,000 ]
aIn one scenario in the Merck model, PCV15 was cost -saving relative to PCV20 (or PCV20 was dominated by PCV15), this scenario as sumed re -emergence of IPD associated with certain
serotypes that PCV20 missed the trial primary endpoint.
b The ranges presented are obtained from the scenario analyses and probabilistic sensitivity analysis.
Policy question 2
Use of PCV20 in children with underlying medical conditions that increase the risk of pneumococcal disease
Policy question #2
Should PCV20without PPSV23 berecommended as an option for pneumococcal vaccination
according tocurrently recommended dosing and schedules, for U.S. children 2 –18years with
underlying medical conditions that increase the risk ofpneumococcal disease (CMC/IC)?
•2.1. What isthe cost effectiveness ofPCV20 alone vsPCV13/ 15+PPSV 23series among CMC/IC?
•2.2. What isthe cost effectiveness ofadding PPSV 23toa PCV 20series among CMC/IC?
𝐶𝑜𝑠𝑡𝑠𝐼𝑛𝑡𝑒𝑟𝑣𝑒𝑛𝑡𝑖𝑜𝑛 −𝐶𝑜𝑠𝑡𝑠 𝐶𝑜𝑚𝑝𝑎𝑟𝑎𝑡𝑜𝑟= 𝐼𝑛𝑐𝑟𝑒𝑚𝑒𝑛𝑡𝑎𝑙 𝑐𝑜𝑠𝑡 𝑒𝑓𝑓𝑒𝑐𝑡𝑖𝑣𝑒𝑛𝑒𝑠𝑠 𝑟𝑎𝑡𝑖𝑜(𝐼𝐶𝐸𝑅)𝑂𝑢𝑡𝑐𝑜𝑚𝑒𝑠 𝐼𝑛𝑡𝑒𝑟𝑣𝑒𝑛𝑡𝑖𝑜𝑛 −𝑂𝑢𝑡𝑐𝑜𝑚𝑒𝑠 𝐶𝑜𝑚𝑝𝑎𝑟𝑎𝑡𝑜𝑟Comparator (dosing) Intervention (dosing) Population
PCV13/15(3+1) + PPSV23 (1) PCV20 (3+1) CMC
PCV13/15(3+1) + PPSV23 (1+1) PCV20 (3+1) IC
PCV20(3+1) + PPSV23 (1) PCV20 (3+1) CMC
PCV20(3+1) + PPSV23 (1+1) PCV20 (3+1) IC
23
Policy question 2.1
PCV20 alone vs PCV13/15+PPSV23 series among CMC/IC
Health outcomes and costs
PCV20 alone vs PCV15+PPSV23 in CMC, base case
Outcomes and costPCV20 vs PCV15 + PPSV23
Tulane -CDC Merck Pfizera
Health
outcomesIPD cases -26 -15 -5*
Pneumonia cases -780 -260 -1,100*
AOM cases -30,000 -28,000 -4,900*
Deaths -2 -1 -5*
QALYs 200 140 120
Costs
($ millions)Vaccine costs -1 7 5
Cost of disease + sequelae -32 -23 -18
Total cost -33 -16 -12
aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15+PPSV23.
a.The Pfizer model does not directly assess policy question 2 but compares the receipt of PCV20 with and without PPSV23 to the receipt of PPSV23 on six -year -
old children with a history of PCV13 vaccination.
Health outcomes and costs
PCV20 alone vs PCV15+PPSV23 in CMC, base case
Outcomes and costPCV20 vs PCV15 + PPSV23
Tulane -CDC Merck Pfizera
Health
outcomesIPD cases -26 -15 -5*
Pneumonia cases -780 -260 -1,100*
AOM cases -30,000 -28,000 -4,900*
Deaths -2 -1 -5*
QALYs 200 140 120
Costs
($ millions)Vaccine costs -1 7 5
Cost of disease + sequelae -32 -23 -18
Total cost -33 -16 -12
aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15+PPSV23.
a.The Pfizer model does not directly assess policy question 2 but compares the receipt of PCV20 with and without PPSV23 to the receipt of PPSV23 on six -year -old
children with a history of PCV13 vaccination.
Health outcomes and costs
PCV20 alone vs PCV15+PPSV23 in CMC, base case
Outcomes and costPCV20 vs PCV15 + PPSV23
Tulane -CDC Merck Pfizera
Health
outcomesIPD cases -26 -15 -5*
Pneumonia cases -780 -260 -1,100*
AOM cases -30,000 -28,000 -4,900*
Deaths -2 -1 -5*
QALYs 200 140 120
Costs
($ millions)Vaccine costs -1 7 5
Cost of disease + sequelae -32 -23 -18
Total cost -33 -16 -12
In these assessments,
all three models
estimate higher health
and lower total costsaThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15+PPSV23.
a.The Pfizer model does not directly assess policy question 2 but compares the receipt of PCV20 with and without PPSV23 to the receipt of PPSV23 on six -year -old
children with a history of PCV13 vaccination.
Cost -effectiveness results, policy question
base case
Risk group Intervention Comparator ModelCost per
QALY
CMCPCV20PCV13 + PPSV23 Tulane -CDC
&
Merck Cost -saving PCV15 +PPSV23
PCV13 +PCV20 PCV13 + PPSV23 Pfizera
ICPCV20PCV13 + PPSV23 +PPSV23 Tulane -CDC
&
Merck Cost -saving PCV15 + PPSV23 + PPSV23
PCV13 +PCV20 PCV13 + PPSV23 + PPSV23 Pfizera
aPfizer model had a different set of comparisons: in CMC: PCV13 (3+1) + PCV20 vs. PCV13 (3+1)+ PPSV23; in IC: PCV13 (3+1) + PCV20 vs. PCV13 (3+1)+ PPSV23 + PPSV23 ,because it started at age
6.6.28
Sensitivity analyses policy question #2.1
PCV20 alone vs PCV13/15+PPSV23
Cost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.
*Aside from the base case, no additional scenarios were conducted for this comparison.
aInthe scenarios where the Merck model assumed a re -emergence of IPD associated with certain serotypes that PCV20 missed the trial primary endpoint, PCV15 was found to be cost -saving relative to PCV20 (or
PCV20 was dominated by PCV15).
bPfizer model had a different set of comparisons: in CMC: PCV13 + PCV20 vs. PCV13(3+1)+ PPSV23; in IC: PCV13 + PCV20 vs. PCV13(3+1) + PPSV23 ,because it started at age 6.Risk
groupIntervention Comparator ModelRanges including sensitivity
analyses ($/QALY)
CMCPCV20 PCV13 + PPSV23Tulane -CDC*Cost -saving
Merck*Cost -saving
PCV13 + PCV20 PCV13 + PPSV23 Pfizer Cost -savingb
ICPCV20 PCV13 + PPSV23 + PPSV23Tulane -CDC*Cost -saving
Merck Cost -saving
PCV13 + PCV20 PCV13 + PPSV23 + PPSV23 Pfizer Cost -savingb
CMC
PCV20PCV15 + PPSV23Tulane -CDC Cost -saving to $19,000
Merck Cost -saving to dominateda
IC PCV15 + PPSV23 + PPSV23Tulane -CDC Cost -saving
Merck Cost -saving to dominateda
29
Sensitivity analyses policy question #2.1
PCV20 alone vs PCV13/15+PPSV23
Risk
groupIntervention Comparator ModelRanges including sensitivity
analyses ($/QALY)
CMCPCV20 PCV13 + PPSV23Tulane -CDC*Cost -saving
Merck*Cost -saving
PCV13 + PCV20 PCV13 + PPSV23 Pfizer Cost -savingb
ICPCV20 PCV13 + PPSV23 + PPSV23Tulane -CDC*Cost -saving
Merck Cost -saving
PCV13 + PCV20 PCV13 + PPSV23 + PPSV23 Pfizer Cost -savingb
CMC
PCV20PCV15 + PPSV23Tulane -CDC Cost -saving to $19,000
Merck Cost -saving to dominateda
IC PCV15 + PPSV23 + PPSV23Tulane -CDC Cost -saving
Merck Cost -saving to dominateda
30Cost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.
*Aside from the base case, no additional scenarios were conducted for this comparison.
aInthe scenarios where the Merck model assumed a re -emergence of IPD associated with certain serotypes that PCV20 missed the trial primary endpoint, PCV15 was found to be cost -saving relative to PCV20 (or
PCV20 was dominated by PCV15).
bPfizer model had a different set of comparisons: in CMC: PCV13 + PCV20 vs. PCV13(3+1)+ PPSV23; in IC: PCV13 + PCV20 vs. PCV13(3+1) + PPSV23 ,because it started at age 6.
Policy question 2.2
PCV20 +PPSV23 series vs PCV20 alone among CMC/IC
Risk
groupIntervention Comparator ModelCost per
QALY
CMCPCV20 + PPSV23 PCV20Tulane -CDC $4 million
Merck $1.9 milliona
PCV13 + PCV20 + PPSV23 PCV13+ PCV20 Pfizerb $6 million
ICPCV20 + PPSV23 +PPSV23 PCV20Tulane -CDC $690,000
Merck $204,000a
PCV13 + PCV20 + PPSV23 +
PPSV23PCV13 + PCV20 Pfizerb $535,000
aValues calculated by presenter using results in Merck technical report.
bPfizer model had a different set of comparisons: in CMC: PCV13 + PCV20 vs. PCV13(3+1)+ PPSV23; in IC: PCV13 + PCV20 vs. PCV13(3+1) + PPSV23 ,because it started at age 6. 32Cost -effectiveness results, policy question #2.2
PCV20+ PPSV23 vs PCV20 alone, base case
Risk
groupIntervention Comparator ModelCost per
QALY
CMCPCV20 + PPSV23 PCV20Tulane -CDC $4 million
Merck $1.9 milliona
PCV13 + PCV20 + PPSV23 PCV13+ PCV20 Pfizerb $6 million
ICPCV20 + PPSV23 +PPSV23 PCV20Tulane -CDC $690,000
Merck $204,000a
PCV13 + PCV20 + PPSV23 +
PPSV23PCV13 + PCV20 Pfizerb $535,000
aValues calculated by presenter using results in Merck technical report.
bPfizer model had a different set of comparisons: in CMC: PCV13 + PCV20 vs. PCV13(3+1)+ PPSV23; in IC: PCV13 + PCV20 vs. PCV13(3+1) + PPSV23 ,because it started at age 6. 33Cost -effectiveness results, policy question #2.2
PCV20+ PPSV23 vs PCV20 alone, base case
Summary
Summary
35Cost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.•Policy question #1
•Vaccination with PCV20 is expected to improve health outcomes
compared to PCV13 or PCV15.
•Compared to PCV13 ,the Pfizer and Merck models estimated
that PCV20 would be cost -saving, whereas the Tulane -CDC model
estimated a cost per QALY of $57,000.
•Compared to PCV15 , the Pfizer model estimated that PCV20
would be cost -saving, whereas the Tulane -CDC model and the
Merck model estimated a cost per QALY of just over $100,000.
Summary
36•Policy question #2.1
•Compared to PCV13/PCV15+PPSV23 series, PCV20 alone in CMC/IC
was estimated to be cost -saving in all but two scenarios investigated
•Policy question #2.2
•Compared to PCV20 alone , PCV20 +PPSV23 series in CM/IC was estimated to
cost between $204,000 to $7 million dollars for each QALY gained
Cost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.
Summary
37Cost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.
bAdditional details on these model inputs provided in supplemental slides.•Main differences across models appear to be related to indirect effects
•bOther important factors: model structure, Vaccine effectiveness, QALY
losses due to disease
Thank you!Contributors
•Miwako Kobayashi
•Andrew Leidner
Supplemental slides
39
References
1. Stoecker C, Hampton LM, Link -Gelles R, Messonnier ML, Zhou F, Moore MR. Cost -effectiveness of using 2 vs 3 primary doses of 13 -valent
pneumococcal conjugate vaccine. Pediatrics. 2013;132(2):e324 -e32.
2. Stoecker, Charles, Miwako Kobayashi, Almea Matanock, Bo -Hyun Cho, and Tamara Pilishvili. "Cost -effectiveness of continuing
pneumococcal conjugate vaccination at age 65 in the context of indirect effects from the childhood immunization program." Vaccine 38,
no. 7 (2020): 1770 -1777.
3. Tong S, Amand C, Kieffer A, Kyaw MH. Trends in healthcare utilization and costs associated with pneumonia in the United States during
2008 –2014. BMC Health Services Research. 2018 Dec;18(1):1 -8.
4. Pfizer. CDC ABCs data (Data on File)
5. Stoecker, Charles, Lee M. Hampton, Ruth Link -Gelles , Mark L. Messonnier, Fangjun Zhou, and Matthew R. Moore. "Cost -effectiveness of
using 2 vs 3 primary doses of 13 -valent pneumococcal conjugate vaccine." Pediatrics 132, no. 2 (2013): e324 -e332.
6. Pilishvili T. 2018a. “Estimating PCV13 direct and indirect effects on IPD among adults ≥65 years”. Presented at the meeting o f the Advisory
Committee on Immunization Practices, February 22, 2018. Atlant a, Georgia.
7. Pilishvili T, Girke R, Xing W, Farley M, Schaffner W, Thomas A, Reingold A, Harrison L, Lynfield R, Zansky S, Petit S, Barnes M, Bareta J, Beal
B, Shang N, Whitney C. 2018b. “Changes in invasive pneumococcal disease (IPD) among adults following 6 year s of 13 -valent
pneumococcal conjugate vaccine use in the U.S. ” Presented at the International Symposium on Pneumococci and Pneumococcal Diseases,
2018. Melbourne, Australia .
8. Prasad, Namrata, Charles Stoecker, Wei Xing, Bo -Hyun Cho, Andrew J. Leidner, and Miwako Kobayashi. "Public health impact and cos t-
effectiveness of 15 -valent pneumococcal conjugate vaccine use among the pediatric population of the United States." Vaccine 41, no. 18
(2023): 2914 -2921.
Health outcomes and cost resultsa
PCV20 vs PCV15, base case
PCV20 vs PCV15 PCV20 vs PCV13
Outcomes and cost Tulane -CDC Merck++Pfizer
Health outcomesIPD cases -360 -400 -430*
Pneumonia cases -13,000 -2,300 -25,000*
AOM cases -150,000 -201,000 -238,000*
Total deaths -160 -28 -93*
QALYs 1,900 1,400 5,600
Costs ($ millions)Cost of disease + sequelae -210 -180 NA
Vaccine costs 440 330 NA
Total cost 230 150 NA
41aThese are discounted values (rounded up to 2 significant figures) for the complete horizon for the base case of each model
*Outcomes not reported as discounted values.
All other values were discounted at 3%.
Negative values indicate averted cases/deaths or reduced costs of PCV20, as compared to PCV15.
++Pfizer scenario analyses with single cohort and no indirect effects applied, and comparison is between PCV20 and PCV13
Model assumptions: Selected base case assumptionsa
Policy question #1
Model input Tulane -CDC Merck Pfizer
Vaccine effectiveness •lower VE for ST3 and
ST19F across all
vaccines.
•the serotype specific
VEsin PCV20 were the
same as the serotype
specific VEs in PCV15
and PCV13 .•serotype specific VE
was lower in PCV20 for
six
serotypes (3,12F,1,4,2
3F,9V) than it was in
PCV15.•the serotype specific
VEsin PCV20 were the
same as the serotype
specific VEs in PCV15
and PCV13 .
QALY loss due to
hospitalized pneumonia in
adults0.0105 0.071 0.13
Model structure Single cohort (17 years) Single cohort (100 years) Multiple cohorts (10
years)
aFrom the review, in addition to indirect effects, these assumptions appear to be the important in terms of determining differences between model results for policy question #1.
For added context, a QALY losses of 0.0016, 0.071 and 0.130 could be considered as representing a 32 -day hospitalization 59 -day hospitalization, respectively, where 20% health -related quality
of life is experienced for the duration of hospitalization. 42
Description of models
Policy question #2
Model characteristics Tulane -CDC MerckaPfizer
Cohort type Single birth cohort Single birth cohort Single -cohort (6 -year -olds)
•~90 % have history of PCV13
Indirect effects Yes (vaccinated CMC/IC contribute
to indirect effect benefits)In scenarios only (not in base case) Yes (vaccinated CMC/IC experience
reduced incidence)b
Vaccine coverage PCV (3rddose): 94 %
PCV(4thdose): 82 %
cPPSV23(1stdose): 78 %
dPPSV23(2nddose): 64 %PCV (3+1): 93 %
PPSV23(1stdose): 64 %
PPSV23(2nddose): 53 %PCV20: 20%
PPSV23: 20%
eWeighted
vaccine price ($)PPSV23: 85 PPSV23: 91 PPSV23: 85
aThe Pfizer model does not directly assess policy question 2, but compares the receipt of PCV20 with and without PPSV23 to the rec eipt of PPSV23 on children with a history of PCV13 vaccination
bIncidence among 6+ year olds was reduced after year 2, assuming routine use of PCV20 would occur, so VPD burden declines from routine use.
cVaccine coverage of first PPSV23 dose is a percentage of those who got 4 doses of PCVs.
dCoverage of second PPSV23 dose is a percentage of those who got the first PPSV23 dose.
eAllthree models used blended vaccine prices (combined using private and public market prices share weights which varied across t hemodels). 43
Summary
44Policy QuestionBase case estimates ($/QALY)
*Range across all
models Tulane -CDC Merck Pfizer
1. PCV20 vs
PCV13/PCV1557,000 to 125,000 Cost -saving to 105,000 Cost -saving Cost -saving to
dominateda
2.1 PCV20 alone (in
CMC/IC)Cost -saving Cost -saving to
dominateda
2.2 PCV20+PPSV23 vs
PCV20 (in CMC/IC)690,000 to 4
million204,000 to 1.9 million 535,000 to 6 million 204,000 to 7 million
Cost -saving means lower costs and improved health outcomes in the intervention, as compared to the comparator.
*The ranges presented are obtained from the scenario analyses and probabilistic sensitivity analyses across all models
aThe Merck model found PCV15 was cost -saving relative to PCV20 (or PCV20 was dominated by PCV15) in scenarios where they assumed a r esurgence of specific PCV20 -only serotype IPD
disease.
bAdditional details on these model inputs provided in supplemental slides.•Main differences across models appear to be related to indirect effects
•bOther important factors: model structure, Vaccine effectiveness, QALY losses due to disease