Model Design & Inputs

• An age-structured and serotype-specific deterministiccompartmental model whichincorporates seasonality, host-vector interactions and 4 denguevirus serotypes immunologicalinteractions was adopted5.

• The model was calibrated andvalidated with Malaysia specificepidemiological data and vaccineefficacy data from phase-III efficacystudies6.

• Two vaccination strategies(Table1), were simulated toevaluate the vaccine cost-effectiveness.

• Malaysia is currently experiencing the most serious dengueepidemic in its history, with the number of dengue cases 5-7times higher than the median between 2011-20151.

• On April 15th 2016, WHO-SAGE on immunization recommendedthe use of dengue vaccine in countries with high endemicity2.

• The first dengue vaccine, Dengvaxia® by Sanofi Pasteur, has beenregistered in 4 countries.

• With the rapid advancement of dengue vaccine developmentand projected availability by this year, a Malaysia-specificeconomic evaluation of dengue vaccine is needed urgently toinform decision makers on the available options for controllingand preventing dengue disease with vaccines.

5/28/2016

Both strategies showed a consistent reduction on dengue cases (40%) anddengue-related mortality (43%) compared to no vaccination.

Acknowledgement: This study was funded by Sanofi-Aventis Singapore Pte. Ltd.

SCHOOL OF

PHARMACEUTICAL

SCIENCES

EXPLORING THE COST-EFFECTIVE THRESHOLD PRICE OF DENGUE

VACCINATION PROGRAMS IN MALAYSIA: A VALUE-BASED PRICING

ASSESSMENTHui Yee YEO1, Asrul Akmal SHAFIE1, Laurent COUDEVILLE2, Lucas STEINBERG3, Balvinder-Singh GILL4, Rohani JAHIS4

1Universiti Sains Malaysia, Penang, Malaysia, 2Sanofi Pasteur, Lyon, France, 3Sanofi Pasteur, Petaling Jaya, Malaysia, 4Disease Control Division, Ministry of Health, Malaysia

BACKGROUND

OBJECTIVES1. To assess the potential health and economic impact of a 3-dose

tetravalent dengue vaccine in Malaysia under differentvaccination strategies.

2. To determine the cost-effective threshold price of the denguevaccine for different vaccination strategies.

METHODS

Model parametersVaccination strategy 1:

NW R9C17

Vaccination strategy 2:

THS R9C17References

Geographical location Nationwide 6 hotspot districts* Expert opinion

Routine + catch-up age cohort 9 years old + 10 – 17 years old Expert opinion

Catch-up duration 1 year Expert opinion

Vaccine coverage + compliance 95% + 90%Malaysia statistics,

Expert opinion

Underreporting factor – hospitalized 1.7 [3]

Underreporting factor – ambulatory 3.79 [3]

Vaccine wastage 20% Expert opinion

Discount rate 3% MPG**

Vaccine protection duration 10 years Assumption

DALY loss per dengue case – hospitalized 0.031 [4]

DALY loss per dengue case – ambulatory 0.01 [4]

Treatment cost / hospitalized dengue case US$ 672.51 [3]

Treatment cost / ambulatory dengue case US$ 326.51 [3]

Malaysia GDP/capita, 2013 US$ 10,456.89 IMF report 2014

* Kuala Lumpur, Petaling, Klang, Hulu Langat, Gombak and Putrajaya.

** MPG = Malaysia Pharmacoeconomics Guideline

Table 1: Model inputs and simulation parameters

Analysis Perspective

• Malaysia public provider.

Time Horizon• 10 years (2016 – 2025)

Outcomes / Impacts

• Dengue cases, dengue-related mortality,years-of-life lost (YLL), disability-adjusted-life-year (DALY), dengue treatment cost,and highly cost-effective threshold price(<1x Malaysia GDP per capita). All costswere expressed in 2013 US$.

Sensitivity Analyses

• Probabilistic sensitivity analysis (PSA) was performed by 1,000 Monte Carlosampling from the joint distribution of plausible ranges of values on denguetreatment cost per case (±20%), vaccine administration cost (±20%), vaccinecoverage (90-100%), vaccine wastage (±10%), vaccine protection duration (5-30years), annual endemicity (1-1000), vaccine efficacy (1-100), and relativeefficacy against infection (0-1).

• One-way deterministic sensitivity analysis (DSA) was performed on the highlycost-effective threshold price for the following parameter uncertainties:discount rate (0-5%), vaccine protection duration (5-30 years), model timehorizon (5-30 years), catch-up vaccine coverage (±30%), hospitalizedunderreporting factor (1.0-2.45), ambulatory underreporting factor (1.0-10.38),dengue treatment cost per case (±20%), vaccine administration cost (±20%),and vaccine wastage (±10%).

RESULTS OBJECTIVE 1

RESULTS OBJECTIVE 2

CONCLUSIONS

The results of this study were robust to a wide range of sensitivity. Our analysis demonstrated that introducing a dengue vaccinationprogram in Malaysia has the potential to substantially reduce the disease and economic burden (40% and 43% respectively),especially if it is introduced during current dengue epidemic. Nationwide vaccination showed a bigger overall impact compared totargeted hotspot vaccination. Our study suggested that dengue vaccination is a potentially good investment if the purchaser couldnegotiate a price at or below the highly cost-effective threshold price, i.e. US$ 12.60 for nationwide vaccination and US$ 34.29 fortargeted hotspot vaccination. We hope this study would inform the decision makers on the fair vaccine price and that it could be amedium for negotiations between stakeholders when considering the incorporation of the dengue vaccine into the NationalImmunization Program.

Table 2: Cumulative vaccination impact on dengue cases & dengue related

mortality of both NW and THS vaccination over 10 years (2016-2025)

Dengue Cases Dengue Related Mortality

NW no vaccination2,123,995

(95% CI: 1,828,532 – 2,544,115)

2,289

(95% CI: 1,972 – 2,742)

NW R9C17 vaccination1,261,000

(95% CI: 888,176 – 1,697,554)

1,308

(95% CI: 914 – 1,785)Reduction - 858,559 - 976

Dengue Cases Dengue Related Mortality

THS no vaccination904,259

(95% CI: 778,470 – 1,083,117)

973

(95% CI: 838 – 1,165)

THS R9C17 vaccination536,860

(95% CI: 378,135 – 722,720)

556

(95% CI: 388 – 759)Reduction - 365,510 - 415

Figure 1: Vaccination impact on health consequences over year 2016-2025

46,964 (95%CI: 40,812-55,568) 29,517

(95%CI: 20,633-40,161) 19,967 (95%CI: 17,350-23,627)

12,549 (95%CI: 8,771-17,074)

68,010 (95%CI: 59,010-80,415)

43,439(95%CI: 30,401-59,185)

28,926 (95%CI: 25,098-34,201)

18,475(95%CI: 12,931-25,172)

-

20,000

40,000

60,000

80,000

100,000

120,000

NW No Vaccination NW R9C17 Vaccination THS No Vaccination THS R9C17 Vaccination

YEA

RS

YLL DALY

Both NW and THS vaccination strategies exhibited a reduction of year-of-life lost (YLL) and disability-adjusted-life-year (DALY) compared to novaccination.

Figure 2: Vaccination impact on on dengue treatment cost (ambulatory and

hospitalized) and total dengue management cost

The model predicted that both NW and THS vaccination strategies wouldreduce the total dengue treatment cost and total dengue managementcost over 10 years (2016-2025).

487.61

275.94

763.55

287.33

155.87

581.30

207.67

117.31

324.98

122.37

66.27

213.53

0

100

200

300

400

500

600

700

800

Dengue Treatment Cost (Ambulatory) Dengue Treatment Cost (Hospitalized) Total Dengue Management Cost

US$

MIL

LIO

NS

NW No Vaccination

NW R9C17 Vaccination

THS No Vaccination

THS R9C17 Vaccination

44%(95% CI: 35 – 53)

41%(95% CI: 33 – 50)

34%(95% CI: 27 – 42)

44%(95% CI: 35 – 53)

41%(95% CI: 33 – 50)

24%(95% CI: 18 – 30) References:

1. World Health Organization. Dengue Situation Updates. [5/4/16];http://www.wpro.who.int/emerging_diseases/dengue_biweekly_20160405.pdf?ua=12. http://www.who.int/immunization/research/development/dengue_vaccines/en/3. Shepard, D.S., et al., Use of multiple data sources to estimate the economic cost of dengue illness in Malaysia. Am J Trop Med Hyg, 2012. 87(5): p. 796-805.4. Salomon, J.A., et al., Healthy life expectancy for 187 countries, 1990–2010: a systematic analysis for the Global Burden Disease Study 2010. The Lancet. 380(9859): p. 2144-2162.5. Coudeville, L. and G.P. Garnett, Transmission Dynamics of the Four Dengue Serotypes in Southern Vietnam and the Potential Impact of Vaccination. PLoS ONE, 2012. 7(12): p.

e51244.6. Shafie AA, et al., A Dynamic Transmission Model For Dengue Virus Infection In Malaysia. [Unpublished] 2015.

Figure 3: Acceptability curve on highly cost-effective threshold price (<1x GDP/capita) of the PSA from public provider perspective

0

10

20

30

40

50

60

70

80

90

100

0 10 20 30 40 50 60 70

Pro

bab

ility

Vac

cin

e B

ein

g H

igh

ly C

ost

-Ef

fect

ive

, %

Vaccine Threshold Price per Dose, US$

NW R9C17 THS R9C17

• The price threshold for NW vaccination strategy (US$ 12.60,95%CI: US$ 8.11 – 18.70) is found to be lower than the THSvaccination strategy (US$ 34.29, 95%CI: US$ 23.60 – 48.83).

• At a price of US$ 10.00, the THS vaccination is 80% highly cost-effective.

• At a price of US$ 25.00, the NW vaccination is 95% highly cost-effective.

Figure 4a & 4b: One-way DSA on highly cost-effective threshold price (<1x GDP/capita) for both NW and THS vaccination strategies from public provider

perspective over 10 years

The threshold price from both vaccination strategies were most sensitive to the ambulatory underreporting factor, vaccine protection duration andhospitalized underreporting factor. It is least sensitive to the vaccine administration cost and vaccine coverage for catch-up cohort.

14.3

14.95

12.43

16.45

19.27

20.92

10

10.67

8.79

14.2

13.51

16.03

12

12.73

14.76

18.8

19.87

27.15

5 10 15 20 25 30

Vaccine coverage rate for catch up (90%/100%)

Vaccine administration cost (80%/120%)

Dengue treatment cost (80%/120%)

Vaccine wastage (10%/30%)

Discount rate (0%/5%)

Model time horizon (5/30 years)

Underreporting factors - hospitalized (1/2.45)

Vaccine protection duration (5/30 years)

Underreporting factors - ambulatory (1/10.38)

Cost-Effectiveness Threshold Price/dose, US$

Upper Lower

38.3

38.85

33.91

43.34

49.95

53.79

28.08

29.81

25.32

38.06

37.41

42.35

32.91

37.44

39.37

48.87

51.36

68.76

20 25 30 35 40 45 50 55 60 65 70

Vaccine coverage rate for catch up (90%/100%)

Vaccine administration cost (80%/120%)

Total dengue treatment cost (80%/120%)

Vaccine wastage (10%/30%)

Discount rate (0%/5%)

Model time horizon (5/30 years)

Underreporting factors - hospitalized (1/2.45)

Vaccine protection duration (5/30 years)

Underreporting factors - ambulatory (1/10.38)

Cost-Effectiveness Threshold Price/dose, US$

Upper Lower

4a 4b

Contact: huiyeeyeo@yahoo.com