A Cost-Effectiveness Analysis
of Alternative Human
papillomavirus (HPV)
Vaccination Strategies
Elamin H. Elbasha
Merck Research Laboratories, USA
Presentation outline
HPV infection and disease
HPV vaccines
Merck model
Public health impact
Economic impact
Summary and conclusions
2
HPV infection
HPV is small, non-enveloped, encapsulated, double-stranded DNA
virus
HPV encodes two structural proteins
L1 codes for major capsid protein
L2 codes for minor capsid proteins
Enormous HPV diversity
More than 100 HPV genotypes
More than 40 types infect ano-genital tract
At least 13 high-risk types cause cervical cancer
Ubiquitous
Lifetime Risk of HPV infection up to 70% among sexually active
Major risk factor for HPV acquisition: number of sexual partners
3
4
HPV infection life cycle
Few months to few years
5
Up to 20 years
Goodman A., Wilbur D. C. Case 32-2003 — A 37-Year-Old Woman with Atypical Squamous
Cells on a Papanicolaou Smear. N Engl J Med 2003; 349:1555-1564
Conditions associated with HPV types
16, 18, 6,11
HPV 16, 18
Estimated attributable %
– Cervical cancer
– High grade cervical abnormalities
– Low grade cervical abnormalities
70 %
50 %
30 %
– Anal cancer
– Vulva / Vagina / Penile
– Head and neck cancers
~70 %
~40 %
~3-12 %
HPV 6, 11
– Low grade cervical abnormalities
– Genital warts
– Recurrent respiratory papillomatosis (RRP)
10 %
90 %
90 %
Clifford, BJ Ca 2003; Munoz Int J Cancer 2004; Brown J Clin Micro 1993; Carter Cancer
Res 2001; Clifford Cancer Epi Biomarkers Prev 2005; Gissman Proc Natl Acad Science
6
1983; Kreimer Cancer Epidemiol Biomarkers Prev 2005
By end of presentation, 16 women
would die from cervical cancer
Second most common cancer among women
274,000 deaths from cervical cancer in 2002
Globocan 2002
7
Immunologic Basis for HPV vaccines
L1 HPV major capsid protein self-assembles into empty
virus-like particles (VLPs)
In animal models of papillomavirus infection using
species-specific VLPs
Vaccination results in protection from infection and disease
Efficacy associated with development of neutralizing antibodies
Transfer of serum from vaccinated to unvaccinated animals
transfers protective efficacy
Protection is prophylactic, not likely to be therapeutic
Protection is likely to be type-specific
8
HPV vaccines
Prepared from virus-like particles (non-infectious)
Produced by recombinant technology
Do not contain any live biological product or DNA
GARDASIL® [prophylactic quadrivalent HPV
(6,11,16,18) vaccine] licensed in U.S. & other
countries
First vaccine to prevent cervical cancer, precancerous genital
lesions, and genital warts
Series of three injections over a six-month period
Safe and highly efficacious
CERVARIX® [prophylactic bivalent HPV (16,18) vaccine]
in final stages of clinical testing
9
Research questions
What are the epidemiologic consequences of HPV
vaccination?
What is the sensitivity of vaccine health impact (HPV,
CIN, cervical cancer, genital warts) to:
vaccine characteristics (e.g., duration of protection)?
vaccination strategies (females and males, females-only, catchup, etc.)?
What is the cost-effectiveness of programs using a
quadrivalent HPV (6/11/16/18) vaccine?
10
Methods
Direct and indirect ‘herd immunity’ effects of
vaccination
Describe transmission of the virus and resulting disease in a
population
Assess impact of vaccine on vaccinees and their contacts
An integrated disease transmission model and costutility analysis
Demographic model
Behavioral model
HPV infection and disease models
Economic model
US healthcare system data and perspective
Assumes existing screening practices
11
Death
Death
Transfer
diagram, no vaccine
compartments
Immune, 16/18 types
1
Z klib
Death
Infection 6/11/16/18
Susceptible
Xklib+Sklib
Infection 6/11
Waning
Immunity
Coinfected
12
Yklib
Clearance 6/11/16/18
Immune, all 4 types
12
Z klib
Death
Infected, 6/11 types
Death
Clearance 6/11
Clearance 6/11
Immune, 6/11 types
2
klib
Y
Death
Infected, 6/11 types
2
U klib
Clearance 16/18
New Entrants
Infection 6/11
Clearance 6/11
Clearance 16/18
Infection 16/18
Infected, 16/18 types
1
Yklib
Death
Z
12
2
klib
Death
Infection 16/18
Infected, 16/18 types
1
U klib
Death
 W diagram, vaccine  Q
Transfer
compartments
1
klib
ki
1
 ki1  ki1 Wklib
Infected, HR types
1
Wklib
2
ki Pklib
1
klib
ki
Immune, HR types
1
Qklib
1
ki2 kli2 Qklib
Infected, LR types
2
Pklib
1
 ki2 kli2 Wklib
12
 kiWklib
  V
1
ki
Bklbkl0b
1
kli klib
12
ki1 ki2 kli
X klib
Vaccinated
Vklib
Coinfected
12
Wklib
12
 ki1  ki1  ki12Wklib
12
 ki12 ki12 ki12Wklib
kiVklib
kiVklib
 ki2 kli2 Vklib
Infected, LR types
Immune, both types
12
Qklib
12
 ki Z klib
12
 ki2  ki2  ki12Wklib
1
2
 ki1 kli
Wklib
2
 ki2  ki2Wklib
Immune, LR types
2
klib
2
klib
W
2
 kiWklib
2
 ki2  ki2 Pklib
Q
13
2
 ki Qklib
1
2
ki1 kli
Qklib
1
 ki1  ki1 Pklib
Infected, HR types
1
Pklib
1
ki Pklib
 ICIN
Transfer diagram, CIN
compartments
r1
Infected
Yh,Uh,Ph
h
 f 1Yflib
Undetected CIN1
CIN1hlib
h
1lib
1ib CIN1hlib
Detected CIN1
DCIN1hlib
(1  1 )1 DCIN1hlib
 1i CIN1hlib
h
 f 3Y flib
h
 f 2Y flib
Treated & Infected
ICIN1hlib
 1i DCIN1hlib
(1  2 ) 2 DCIN 2hlib
Undetected CIN3
CIN 3hlib
 2i CIN 2hlib
Undetected CIN2
CIN 2hlib
 2ibCIN 2hlib
Detected CIN2
DCIN 2hlib
Treated & Infected
ICIN 2hlib
 r 2 ICIN1hlib
 3i CIN 3hlib
 2  2 DCIN 2hlib
 11 DCIN1hlib
 3ibCIN 3hlib
 r 3 ICIN 3hlib
Invasive Cancer
h
CCslib
 3i DCIN3hlib
Detected CIN3
DCIN 3hlib
 33 DCIN3hlib
Treated & Cured
TCIN slib
14
(1  3 )3 DCIN3hlib
Treated & Infected
ICIN3hlib
Vaccine characteristics: data
and assumptions
Vaccine take (% of vaccinees with vaccine effect)
HPV 16/18 100%, HPV 6/11 100%
Vaccine degree of protection
HPV 16/18, HPV 6/11: against infection 90% (CI:74100)
HPV 16/18, HPV 6/11: against disease 100% (CI:87100)
Vaccine duration of protection
HPV 16/18, HPV 6/11: 10 years to lifetime
Breakthrough infections
Infectiousness and clearance same as natural infections
15
Vaccination strategies
Description
Definition
A. Routine 12-year-old
females
Vaccinate females before reaching age 12
B. Routine 12-year-old
females and males
Vaccinate females and males before
reaching age 12
C. 12-year-old females + 12–
24-year-old females catch-up
Strategy A + a temporary catch-up program
targeting 12–24-year-old females
D. 12-year-old females and
males + 12–24-year-old
females catch-up
Strategy B + a temporary catch-up program
targeting 12–24-year-old females
E. 12-year-old females and
males + 12–24-year-old
females and males catch-up
Strategy B + a temporary catch-up program
targeting 12–24-year-old females and males
16
Vaccination penetration rates:
assumptions
Routine 12-year olds
increase vaccine penetration linearly from 0% in Year 0 to 70% in
Year 5 and after
Catch-up 1224-year olds
All cohorts (1224): increase vaccine penetration linearly from 0%
in Year 0 to 50% in Year 5
Program stops after 5 years
17
Impact of vaccination strategies diagnosed HPV
16/18-related cervical cancer incidence, females
(12+y), lifelong duration
5
No Vaccination
12-yo females
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
12-yo females&males+females and males catch up
Incidence per 100,000 .
4
3
2
1
0
0
10
20
30
40
50
Time in Years
18
60
70
80
90
100
Impact of vaccination strategies diagnosed
HPV 16/18-related CIN 2/3 incidence- females
(12+y) lifelong duration
Incidence per 100,000 .
100
No Vaccination
12-yo females
75
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
12-yo females&males+females and males catch up
50
25
0
0
10
20
30
40
50
Time in Years
19
60
70
80
90
100
Impact of vaccination strategies diagnosed
HPV 6/11/16/18-related CIN 1 incidence - females
(12+y) lifelong duration
Incidence per 100,000 .
30
No Vaccination
12-yo females
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
20
12-yo females&males+females and males catch up
10
0
0
10
20
30
40
50
Time in Years
20
60
70
80
90
100
Impact of vaccination strategies diagnosed HPV
6/11-related genital warts incidence - females
(12+y) lifelong duration
160
No Vaccination
12-yo females
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
12-yo females&males+females and males catch up
Incidence per 100,000 .
140
120
100
80
60
40
20
0
0
10
20
30
40
50
Time in Years
21
60
70
80
90
100
Impact of vaccination strategies diagnosed HPV
6/11-related genital warts incidence - males (12+y)
lifelong duration of protection
160
No Vaccination
12-yo females
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
12-yo females&males+females and males catch up
Incidence per 100,000 .
140
120
100
80
60
40
20
0
0
10
20
30
40
50
Time in Years
22
60
70
80
90
100
Cumulative quality-adjusted life
years
T
QALYs a   [ qualityi * healthstateia (t )]e
0
i
23
t
Cumulative costs
T
Costa   [Vaccinatea (t )  Screena (t )  Treat a (t )]e
0
24
t
Cost-effectiveness analysis of HPV
vaccination strategies*
Discounted total
Strategy
Incremental
Costs
QALYs
Costs
QALYs
$/QALYs**
No vaccination
72,659,302
2,698,711
––
––
––
12-year-old females
74,042,990
2,699,178
1,383,687
467
$2,964
12-year-old females and males
78,707,825
2,699,327
4,664,835
149
dominated
12-year-old females + 1224-yearold females catch up
74,815,667
2,699,343
3,892,159
16
$4,666
12-year-old females and males +
1224-year-old females catch up
79,746,357
2,699,461
4,930,690
118
$41,803
12-year-old females and males +
1224-year-old females and males
catch up
81,761,210
2,699,506
2,014,853
45
$45,056
*Assumes
cost of vaccination series is $360 and duration
of protection is lifelong.
25
with the preceding non-dominated strategy.
**Compared
Sensitivity analysis: Impact of vaccination strategies
diagnosed HPV 16/18-related CIN 2/3 incidencefemales (12+y) 10-years duration vs. lifetime
Incidence per 100,000 .
100
75
No Vaccination
12-yo females
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
12-yo females&males+females and males catch up
12-yo females&catch up (lifelong)
50
25
0
0
10
20
30
40
50
Time in Years
26
60
70
80
90
100
Impact of Vaccination Strategy
Cervical Cancer Incidence - Females (12–85y)
Lifelong duration, 50% coverage
Incidence per 100,000 .
5
No Vaccination
12-yo females
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
12-yo females&males+females and males catch up
4
3
2
1
0
0
10
20
30
40
50
Time in Years
27
60
70
80
90
100
Impact of Vaccination Strategy
Cervical Cancer Incidence - Females (12–85y)
Lifelong duration, 90% coverage
Incidence per 100,000 .
5
No Vaccination
12-yo females
12-yo females+females catch up
12-yo females&males
12-yo females&males+females catch up
12-yo females&males+females and males catch up
4
3
2
1
0
0
10
20
30
40
50
Time in Years
28
60
70
80
90
100
Sensitivity analyses: Incremental cost-effectiveness
ratio ($/QALY) vs. duration of protection & cost
Vaccination costs
Input range/Program
$300
$500
12-year-old females + 1224-year-old
females catch up
2,422
9,900
12-year-old females & males + 1224year-old females & males catch up
36,161
65,810
12-year-old females + 1224-year-old
females catch up
16,194
32,619
12-year-old females & males + 1224year-old females & males catch up
44,562
79,115
Vaccine duration of protection:
lifelong
Vaccine duration of protection: 10Years
29
Sensitivity analyses:
Incremental cost-effectiveness ratio ($/QALY) vs.
vaccine coverage and cost
Vaccination costs
Input range/Program
$300
$500
2,056
9,271
28,845
53,479
2,925
10,739
82,241
142,830
Vaccine coverage: 50%
12-year-old females + 1224-yearold females catch up
12-year-old females & males +
1224-year-old females & males
catch up
Vaccine coverage: 90%
12-year-old females + 1224-yearold females catch up
12-year-old females & males +
1224-year-old females & males
catch up
30
Limitations & outstanding
research questions
Vaccine characteristics (e.g., duration of protection) are
influential
Need more and better epidemiologic and natural history of
disease data to support model
Need to analyze the impact on other important HPV-related
diseases such as vulvar and vaginal neoplasias and cancers,
recurrent respiratory papillomatosis
Need to reflect the indirect costs of HPV-related disease
Need to model HPV types interaction/cross protection
If screening practices change, the model can reflect the shifting
impact of vaccination
31
Summary
A prophylactic quadrivalent HPV vaccine can
substantially reduce the incidence of cervical cancer,
CIN, and genital warts
Catch up vaccination can provide earlier and greater
reductions in HPV-related disease
Vaccinating males and females before age 12
combined with a temporary 1224-year olds catch up
program can be cost-effective and efficiently added to
current screening programs
32
Acknowledgement
Erik J. Dasbach, PhD
Ralph P. Insinga, PhD
Merck Research Laboratories, USA
33