MODELLING HIV VACCINE EFFICACY AND EFFECTIVENESS

1
MODELLING HIV VACCINE EFFICACY AND EFFECTIVENESS

• UNAIDS, CDC, EMORY UNIVERSITY
• IRA LONGINI (EMORY)
• DAN BARTH-JONES (WAYNE STATE)
• MARTA ACKERS (CDC)
• TIM MASTRO (CDC)
• DALE HU (CDC)
• JOSE ESPARZA (UNAIDS)

2
Kenyan Collaboration
BOAZ CHELUGET (MINISTRY OF HEALTH) LARRY MARUM
(CDC)
3
Thai Collaboration
Dr. Wiwat Peerapatanapokin (MINISTRY OF HEALTH)
4
Objectives

• Optimal distribution of HIV vaccine
• Moderately effective
• Limited quantity
• Heterogeneous population
• Countries
• Thailand
• Kenya - example presented in this talk
• Brazil
• U.S.

5
Measures of Vaccine Efficacy

• (VES , VEP , VEI)
• VES Vaccine Effect on Susceptibility (classical
  phase III vaccine trials)

6
Measures of Vaccine Efficacy

• VEP Vaccine Effect on Progression (classical
  phase III vaccine trials)
• Clinical symptoms
• Viral load threshold
• CD4 cell count threshold
• Composite endpoints
• Viral load initiation of HAART

7
Measures of Vaccine Efficacy

• VEI Vaccine Effect on Infectiousness (partner
  augmented phase III vaccine trials, viral load as
  a surrogate?)
• Indirect Effects (community randomized vaccine
  trials, modeling)

8
Transmission

• Infection hazard rate
• Incidence contact rate trans. prob. preval.
  infect.
• ?(t) c p I(t)/n
• epidemic dynamics are expressed through I(t)/n
• Basic Reproductive Number
• R0 ? c p D
• if R0 ? 1, then transmission can not be
  maintained

9
Transmission with vaccination
Infection hazard rate to unvaccinated person
?0(t) c p (1 - f)(I0(t)/n0) f
(1-VEI)(I1(t)/n1) Infection hazard rate to
vaccinated person ?1(t) a (1-VES) ?0(t)
Reproductive Number if Rf ? 1, then
transmission can not be maintained
10
Simple differential equations
For unvaccinated dS0(t)/dt - ?0(t)
S0(t) dI0(t)/dt ?0(t) S0(t) - ? I0(t) For
the vaccinated dS1(t)/dt - ?1(t)
S1(t) dI0(t)/dt ?1(t) S1(t) - ? I1(t)
11
Threshold Results With Vaccination

• Reproductive number with the fraction f
  vaccinated
• Rf 1 - f VES VEI - VES VEI R0
• e.g., R0 2, VES 0.3, VEI 0.5
• When f 0.5, then R0.5 1.35
• When f 0.8, then R0.8 0.96

12
Critical Vaccination Fraction

• Critical vaccination fraction f
• Solution of Rf 1
• f R0 - 1/VES VEI - VES VEI R0
• e.g., f 0.769

13
f vs VES and VEI when R0 2
14
f vs VES and VEI when R0 2
15
Critical Vaccination Fraction f when R0 2
16
Project Epidemic Model
17
Project Epidemic Model
FHet
MHet
Low Risk
High Risk
Large system of differential equations
18
Simulation Scenario

• Total At Risk Population Size N100,000
• MHet, High 36,000 36
• Mhet, Low 8,600 9
• Fhet, High 14,000 14
• Fhet, Low 33,000 33
• Adolescents 8,400 8

19
Vaccination Process

• Fixed number of vaccine doses available per
• year
• Target monthly proportion of unvaccinated
• susceptibles in each group
• Priority of each group

20
Fitting The Model

• Monte Carlo Simulation 23 parameters
• Uniform and normal prior distributions
• Least squares fit
• Parameters
• 2 transmission probabilities
• 7 contact rates
• 10 risk change and timing parameters
• 4 staging and gender-specific multipliers

21
Fit of Model for High Risk Heterosexual Women1
Data, 2 Modeled
1.00
2
1
0.50
2
2
1
1
2
1
2
1
2
1
2
1
2
0.00
0.00
50.60
101.20
151.80
202.40
253.00
Months
Time 0 is January, 1980 Time range 1980-2001
22
Fit of Model for Low Risk Heterosexual Women 1
Data, 2 Modeled
2 Fhet Low HIV Prev 2
1
1.00
2
1
0.50
2
1
2
1
2
1
2
1
1
2
2
0.00
0.00
63.25
126.50
189.75
253.00
Months
Untitled
23
Fit of Model for High Risk Heterosexual Men 1
Data, 2 Modeled
1
1.00
2
1
0.50
2
1
2
1
2
1
2
1
2
2
1
1
2
0.00
0.00
50.60
101.20
151.80
202.40
253.00
Months
Untitled
24
BASELINE EPIDEMICPrevalence
1 Mhet Comb HIV Prev 2
2 Fhet High Prevalence 2
3 Fhet Low HIV Prev 2
4 HIV Prev in Risk Active Pop 2
1
0.50
2
3
4
2
2
1
2
0.25
3
4
4
1
4
3
2
1
3
4
1
2
1
3
4
2
1
3
3
4
0.00
0.00
60.00
120.00
180.00
240.00
Months
Group Prevalences in Comparison Pop
25
VES 0.3, VEI 0.5, VEP 0.5, TARGET 80
HIGH RISK ADULTS 60 Mo.
1 Mhet Comb HIV Prev
2 Fhet High HIV Prev
3 Fhet Low HIV Prev
4 HIV Prev in Risk Active Pop
1
0.50
2
3
4
1
2
0.25
3
4
2
2
2
1
1
4
4
1
4
2
1
3
3
3
4
2
1
3
3
4
0.00
0.00
60.00
120.00
180.00
240.00
Page 1
Months
Group Prevelances in Pop with Vacc Program
26
VES 0.3, VEI 0.5, VEP 0.5, TARGET 80 OF
ADOLECENTS
1 Mhet Comb HIV Prev
2 Fhet High HIV Prev
3 Fhet Low HIV Prev
4 HIV Prev in Risk Active Pop
1
0.50
2
3
4
2
2
1
2
0.25
3
4
4
1
4
3
2
1
3
4
1
2
1
3
4
2
1
3
3
4
0.00
0.00
60.00
120.00
180.00
240.00
Months
Group Prevelances in Pop with Vacc Program
27
TOTAL HIV INFECTIONS PREVENTEDVES 0.3, VEI
0.5, VEP 0.51 80 high risk adults, 2 80
adolescents
Tot HIV Inf Prevented
1
35000
1
1
17500
1
2
1
2
2
1
2
1
0
0.00
60.00
120.00
180.00
240.00
Months
28
TOTAL HIV INFECTIONS PREVENTED PER 1OO DOSESVES
0.3, VEI 0.5, VEP 0.51 80 high risk
adults, 2 80 adolescents
HIV Inf Obj
1
100
1
50
1
2
1
2
1
2
1
2
1
0
0.00
60.00
120.00
180.00
240.00
Months
29
TOTAL HIV INFECTIONS PREVENTEDVES 0.3, VEI
0.5, VEP 0.51 80 high risk adults 2 80
at risk adults
Tot HIV Inf Prevented
1
35000
2
1
1
17500
2
1
2
1
1
2
1
0
0.00
60.00
120.00
180.00
240.00
Months
30
TOTAL HIV INFECTIONS PREVENTEDVEI 0.5, VEP
0.5, target 80 high risk adults 1 VES 0.5,
VEI 0, 2 VES 0, VEI 0.5
Tot HIV Inf Prevented
35000
1
2
17500
1
2
2
1
1
2
0
0.00
60.00
120.00
240.00
180.00
Months
31
TOTAL HIV INFECTIONS PREVENTEDVES 0.3, VEI
0.5, VEP 0.5 target 80 high risk adultsVac.
time 1 beginning, 2 5 years, 3 10 years
Tot HIV Inf Prevented
1
40000
1
2
1
20000
3
1
2
3
1
2
1
2
3
3
1
0
0.00
60.00
120.00
180.00
240.00
Months
32
TOTAL HIV INFECTIONS PREVENTEDVEI 0.5, VEP
0.5, target 80 high risk adults 1 VES 0.0,
2 VES 0.3, 3 VES 0.5, 4 VES 0.8
Tot HIV Inf Prevented
1
40000
4
3
2
1
20000
4
1
3
2
1
4
3
2
1
1
2
3
4
1
0
0.00
60.00
120.00
180.00
240.00
Months
33
EFFECT OF BEHAVIOR CHANGE ON VES 0.3, VEI
0.5, VEP 0.5, target 80 high risk adults1
no increase in contact rate2 1.5 fold
increase, 3 2.0 fold increase
Tot HIV Inf Prevented
1
35000
1
2
1
2
1
1
2
2
3
1
0
3
3
1
-35000
0.00
60.00
120.00
180.00
240.00
Months
34
(No Transcript)
35
(No Transcript)
36
Uncertainty Analysis

• Distribution of 108 best fits from Monte Carlo
  selection on 23 parameters

37
Sensitivity analysis when VES 0.3 and VEI 0.6
cChildren only chChildren, Male Female
High cmChildren, Male High, Females feFemales fh
Female High flFemale Low hiMale Female
High mhMale High mlMale Low
38
SOME CONCLUSIONSFOR POPULATION MODELED

• We can get control the epidemic with a moderate
  efficacious vaccine if properly distributed
• We need to estimate VES, VEI, VEP and waning
  effects from properly designed and analyzed HIV
  vaccine trials
• We will need to build HAART into vaccine trial
  designs
• We will need to carry out community vaccine
  trials to estimate vaccine effectiveness

39
CONCLUSIONS CONTINUEDFor an African Population

• Cases prevented per 100 doses of vaccine
• Female high risk
• Total cases prevented
• Adult high risk
• Adult high risk adolescents

40
Simulation Scenario Thailand

• Total At Risk Population Size N100,000
• MHet, High 10,000 10
• Mhet, Low 40,000 40
• Fhet, Low 49,000 49
• CSWs 1,000 1
• MIDUs 500 0.5
• 2,200 Adolescents

41
BASELINE EPIDEMICPrevalence
42
VES 0.3, VEI 0.5, VEP 0.5, TARGET 80 OF
HIGH RISK ADULTS
43
VES 0.3, VEI 0.5, VEP 0.5, TARGET 80 OF
ADOLECENTS
44
TOTAL HIV INFECTIONS PREVENTEDVES 0.3, VEI
0.5, VEP 0.51 80 high risk adults, 2 80
adolescents
45
TOTAL HIV INFECTIONS PREVENTED PER 1OO DOSESVES
0.3, VEI 0.5, VEP 0.51 80 high risk
adults, 2 80 adolescents
46
(No Transcript)
47
(No Transcript)
48
CONCLUSIONS South East Asia

• Cases prevented per 100 doses of vaccine
• CSW and IDU
• Total cases prevented
• CSW, IDU, Adolescent, MHetHigh
• All the best strategies involve CSW and MHetHigh

49
The End
50
Prophylactic and Therapeutic Vaccines