Flavivirus Vaccination

organized by Fondation Mérieux

MEETING REPORT Version 1 CONFIDENTIAL

The meeting was held at Les Pensieres Conference Center in Veyrier du Lac, France

December 6- 8, 2010

The following report summarizes the information provided during this meeting based on abstracts and speaker’s lectures, procedure specifics for research investigation are not

detailed in this report. Report Issued Jan 8, 2011

Meeting Reporter: Valentina Picot

Disclaimer Information on this report was obtained from the lectures and abstracts given by the speakers as per scientific agenda on the Flavivirus Vaccination meeting held December 6-8, 2010 at “Les Pensieres” conference center in Veyrier du lac, France. All graphs, flow charts and images were obtained from the speaker’s presentations to facilitate the comprehension on the subject. The information on this report was authorized to be used for this meeting report writing a diferent form of vulgarization of this information might require further speaker’s authorization. The information provided does not constitute a manual or technical sheet on the subject, it might have omissions, we cannot assure its completeness or accuracy, and should not be used for the diagnosis or treatment of disease. Commercial products and prototypes are named and illustrated for information purposes only, no endorsement or recommendation by the Fondation Mérieux or those of the organization partner or the meeting reporter are implied or should be inferred. They do not necessarily represent the views of Fondation Mérieux and have not been formally disseminated and should not be construed to represent any agency determination or policy.

Contents

I. Meeting Background and Objectives

II. Summary of Presentation

Keynote lecture: Perspective in the development of vaccines against

Flaviviruses

Session 1: Infection with flaviviruses: disease burden, epidemiology Session 2: Molecular structure of flaviviruses; cellular receptors of flaviviruses; flaviviruses cell entry mechanisms Session 3: Pathogenesis and immune response; mechanisms underlying in vivo protection; neutralization and enhancement of infections Session 4: Vaccination against flaviviruses: vaccines in development

III. Annex 1: Meeting’s Program

Background and Objectives

Flaviviruses comprise a family of more than 70 single-stranded RNA viruses, the majority of which are transmitted to their hosts by vectors such as mosquitoes or ticks.

Flavivirions are spherical and composed of a lipid membrane surrounding a capsid (C) protein bound to a viral RNA (positive)-sense, which together form the nucleocapsid core. The surface proteins are the glycosylated envelope (E) and membrane (M) proteins. The M protein is a mature form of the pre-membrane (prM) protein, important for infectivity and pathogenicity. The E protein is responsible for viral attachment to cellular receptors and specific membrane fusion. The E protein is the major target for virus-neutralizing and hemagglutination-inhibiting antibodies.

A single serotype has been identified for Japanese encephalitis (JE) yellow fever (YF), and tick-borne encephalitis (TBE) virus, whereas 4 serotypes exist for Dengue virus, complicating its vaccine development. The prototype flavivirus is YF virus, which was first isolated in 1927 from “Asibi”, a resident of Ghana. Despite the effective vaccines available against YF, JE, and TBE since many years, the numbers of human cases have increased in the last 2 decades, mainly due to changes in climate conditions and increased travel and urbanization. These diseases have become a major public health problem in a number of countries once again. There are currently >200,000 cases and 30,000 death for YF, and it is believed that JE is responsible for more than 50,000 cases of encephalitis annually, with at least 10,000 deaths. Some 2.5 billion people – two fifths of the world's population – are now at risk for Dengue, and despite 50 years of efforts, there is no Dengue vaccine available on the market. WHO currently estimates that there may be 50 million Dengue infections worldwide every year. In 2007 alone, there were more than 890,000 reported cases of Dengue in the Americas, of which 26,000 cases were diagnosed as Dengue Hemorrhagic Fever (DHF). There are almost 100 asymptomatic infections for every reported flavivirus case, and the case fatality rates vary from 1 to 30% depending on the infecting flavivirus.

The pathogenesis of flaviviruses is not completely understood. The first steps of infection involve the interaction of the virus with the host cells. The virus transmitted to the human body by mosquito or tick bite proliferates locally and in regional lymph nodes, and enters the blood vessels causing short-lived viremia lasting 5 to 15 days. The infection may be asymptomatic or spread to visceral organs or the central nervous system, triggering innate and adaptive immune responses by the host with the outcome of recovery, neurological sequelea, or death, depending on the virus species and host factors which affect susceptibility to the infections.

There is no specific treatment for flavivirus infections and the viruses cannot be eradicated given their animal reservoir. Hence, vaccination is the most effective approach to disease control.

Currently, there are effective vaccines against TBE, JE and YF available for travellers who will spend time in endemic or epidemic regions, as well as the population residing in these areas. These vaccines have been shown to be successful in combating the disease and reducing the number of human cases around the world.

Farshad Guirakhoo, Ph.D.

The Meeting’s objectives :

1. To focus on epidemiology, antigenic and molecular structure, as well as on pathogenesis,

and host immune responses against flaviviruses;

2. To review the safety and effectiveness of current and future vaccines such as TBE, JE,

YF and Dengue.

Keynote lecture : Perspective in development of vac cines against flaviviruses Thomas MONATH, Kleiner Perkins Caufield & Byers and Adjunct Professor Harvard School of Public Health – USA Vaccine development has naturally followed the medical and veterinary public health importance of members of the virus genus. Some vaccines have been developed in response to dramatic and unpredicted disease emergences, such as Kyasanur Forest disease (1957), Rocio encephalitis (1975), and West Nile (1999), whereas others were developed in response to the recognition of the long-standing burden of endemic (and periodically epidemic) disease (e.g. yellow fever, Japanese encephalitis, tick-borne encephalitis, Dengue). The approach to vaccine development has evolved in concert with improvements in virological methods and with the advent of molecular definition of antigenic determinants and cloning techniques. Yellow fever was the first Flavivirus vaccine to be developed, and since the history of that development illustrates many important tenets of vaccinology, it will be briefly reviewed. The remarkable self-adjuvanting activation of innate immunity appears to explain the very high and durable adaptive response to this vaccine. Other classical live, attenuated Flavivirus vaccines have been developed, some successful, some not. The current focus for replicating vaccines is the use rationally designed viral vectors. Such vectors provide a plug-and-play approach that could be used to rapidly develop a vaccine against a new, emerging Flavivirus threat. A remaining challenge in the development of vaccine strategies against this group of viruses is a safe and effective vaccine against Dengue, which is complicated by the occurrence of four serotypes in the species and by the unique pathogenesis of enhanced disease on heterologous reinfection. With the introduction and deployment of each new effective vaccine, the risk:benefit equation ultimately shifts the emphasis from efficacy and disease control to safety and avoidance of risk. This is driving the development of some new vaccines with improved safety profiles. The following is a list of some of the pathogenic flaviviruses diseases and if existing vaccines,

Some of the key issues that facilitated the development of new vaccines play a different role upon diseases as follows:

As one can see Dengue has been through the years the elusive one for vaccine development.

A glance on the evolution up today of the live and inactivated Flavivirus vaccines

Today the following vaccines are under clinical trials, and as one can observed in the bottom table most are live vaccines due to that live vaccines are incredibly immunogenic.

Research is often confronted to the see saw dilemma of live vaccines between:

However the inmunogenecity triggered by live vaccines is of most importance for vaccine development, for instance the following example of a first generation JE vaccine:

One of the recurrent themes regarding Flaviviruses vaccine development is the viscerotropism and neurothropism and the importance of understanding these distinct

biological phenotypes in order to build vaccines, taking into account that the virus have shown different tropisms upon the animal models used. A summary in the following learning leasons Learnings No. 1 Residual neurotropism (vicerotropism) are a feature of all live vaccines

–Host and viral factors determine disease expression –There is a small percentage of the human population with genetically determined susceptibility to severe flavivirus infection

Immunogenicity and attenuation are correlated

–Reflected in dose response and nonclinical biomarkers A single dose of live vaccine is as immunogenic as multiple doses of inactivated antigen

–Adjuvant effect, innate immunity, replication/antigenic mass Second Generation Live Vaccines : Rational design enabled by

–infectious clone technology –an understanding of genome structure-function relationships –sequencing for QC

Two general approaches: –Site directed mutagenesis or deletion –Chimeric constructs –These approaches are often combined

Can take advantage of existing attenuated vaccines (e.g. YF, JE, DEN-2) –As ‘body parts ’ in the chimeric constructs –As guides to construction •e,.g. use of wild-type DEN strains as prM-E donors (experience with DEN-2 PDK-53 vaccine)

Chimeric Vaccine strategy –Heterologous vector

Vector can be an existing vaccine -Provides a benchmark phenotype for nonclinical and clinical evaluation.

Less interference between constructs with different donor gene serotypes -Heterologous T cell epitopes, fewer cross-reactive CTLs where these may be undesirable (e.g. Dengue).

Up today three viruses had been worked for vaccine vector genotypes: • YF 17D vector : 15 mutations • DEN2 16681 PDK-53 : 5 mutatations • JE SA14-14-2 : 3 mutations

Learnings No. 2 Construction of this rationally deisign viruses is relatively easy, assessing biology is not. Imperfect knowledge of

–Molecular determinants of virulence (virulence is multigenic) –Epitope composition is not very well known

Achieving the right balance of attenuation and immunogenicity is challenging. Genetic instability of RNA viruses remains an issue for live vaccines. YF 17D as a Vector

•Long history of use, approved in all countries •Powerful immunogen (innate immune activation, self-adjuvanting) •Single injection, low dose requirement •Rapid onset of immunity (10 days) •Durable immunity (¡Ölife -long) •No anti-vector immunity (prM-E replaced) •Rare SAEs, but steps could be taken to dial in additional safety features

Construction of Chimeric Virus

Learnings No. 3 Chimerization process per se attenuates virulence. Chimera of two empirically derived attenuated vaccines (SA14-14-2 and YF 17D) yielded a suitable candidate. E gene principal determinant of virulence/attenuation. –Insertion of highly attenuated prM-E from SA14-14-2 abrogated YF 17D neurotropism At least 3 E gene a.a. mutations produced neuroattenuation. At least 3 reversions required to restore neurovirulence.

Learnings No. 4 YF 17D serves as a benchmark for attenuation of new flavivirus vaccines –Quantitative measures in comparative studies of neurotropism

However, attenuation of viscerotropism is more difficult to assess –Monkey viremia levels used as a biomarker, but no data to show correlation with YEL-AVD

Out of the several experiences one can conclude that Chimerization of YF 17D by insertion of a heterologous prM-E sequence from a less hepatotropic virus (JE, Dengue) will reduce likelihood of serious adverse events. Learnings No. 4 A single dose of live chimeric vaccine can provide superior immunity to multiple doses of inactivated antigen/ Rapid, durable N antibody response/ T cell responses to both the donor (E gene) and backbone/ Strain differences in donor E gene modulate antigenicity and neutralization

–Other examples •YF 17D neutralization > YF Asibi •JE Beijing-1 > Nakayama •Den2 (American) > Den2 (Asian) •EDIII specific immunization –Implications for: •Selection of strains as vaccine candidates •Restricted epitope constructs (e.g. gene shuffling, EDIII) •Design of non-inferiority trials where two vaccines incorporate different strains

Regarding Dengue, there are a number of vaccines in development, which main candidates are live vaccines

Learnings No. 5 - Interference High rates of seroconversion and high antibody levels when a single dose of monovalent vaccine (titer 3-4 logs) given to seronegative recipients. When 4 vaccines mixed, antibody productionlower and some serotypes predominate while others are missed. Interference can be modified somewhat by increasing dose of less active components. Same set of observations were historically seen for OPV.

Mitigating Interference Live flavi vectors •Multiple doses (interval allowing subsidence of innate and cross-reactive immunity) •Balanced dose formulation •Separate anatomical sites •Heterologous pre-immunity (e.g.YF) •Shuffled E gene Other approaches •Single live vector tetravalent E or EDIII (Ad5, measles) •Single cycle vector, tetravalent or monovalent mixture •Inactivated viruses •Subunit antigens •DNA •Prime-boost strategies Multiple dose requirement Schedule issues specially with endemic market. Compliance, schedule different from EPI, potential for ADE

–Travelers and military require rapid immunization Original antigenic sin

–Secondary responses (on boosting) may be poor quality (cross-reactive) antibody –Vaccine should induce primary response to all serotypes on first dose

Next Generation Vaccines Major goals:

–Improved safety –Reduce interference (Dengue) –Short interval boosting –Set immune response to all four Dengue antigens on first dose –Durable response, strong T cell memory

Current approaches: –Single cycle flavivirus –Heterologous defective or live vectors (adeno, alpha, measles) –DNA launch (single round infectious particles) –Inactivated virus with appropriate adjuvant –Recombinant VLPs

Status: –Early stage, preclinical

Challenges for Next Gen Vaccines Clinical development : –Increasingly difficult after 2 nd gen vaccine(s) approved due to ethical issues (placebo controlled trials) and decreasing incidence at established sites Regulatory path : –Licensure based on non-inferiority (seroconversion, GMT) to licensed product (likely to be a live 2 nd gen vaccine) Showing marketing advantage, differentiation and label claims Next Generation Live Flavivirus Vaccines Attributes : Single cycle ‘pseudoinfectious ’virus or live heterologous vectors Potentially higher safety, no progressive infection In vivo expression of immunogenic subviral prM-E particles and NS1 with native conformation of epitopes •Memory, durability, Th1 responses should resemble live vaccines Questions : Sufficient antigenic mass and immunogenicity? Activation of innate immunity, durability? Interference? Leaning No. 6 All comes down to the immunogenecity triggered, many factors determine vaccine immunogenicity

• Attenuation, replication, antigenic mass • Anatomical/cell tropism, antigen duration • Antigenic structure, conformation • HLA and cytokine/cytokine receptor gene polymorphisms • Innate immunity and specific pathways activated • Previous immunity to related viruses and original antigenic sin • Interference and vaccine interactions

A fundamental question is : How much immunity is required? What is the level of antibodies corresponding to protection? Established for JE and TBE vaccines only (PRNT 50 ¡Ý10) Seroprotection for viscerotropic viruses (YF, DEN) may be 2 to 10-fold higher Dengue is problematic because of cross-reactive epitopes and difficulty determining homotypic and heterotypicresponses –Dengue infections occur in subjects with N antibody to the infecting serotype and strain Regulatory issues: use of immune correlates for vaccines where field efficacy cannot be shown (WN, YF)

Session 1 - Infection with flaviviruses: disease bu rden, epidemiology Epidemiology and Disease Burden of Dengue Duane J Gubler, Sc.D. Professor and Director, Signature Research Program in Emerging Infectious Diseases Duke-NUS Graduate Medical School, Singapore, and Director, Asia-Pacific Institute for Tropical Medicine and Infectious Diseases, John A. Burns School of Medicine, University of Hawai‘i, Honolulu, Hawaii The global spread of Dengue viruses has resulted in a dramatic increase in the frequency and magnitude of epidemic Dengue/Dengue hemorrhagic fever. With an estimated 50 to 100 million infections annually, Dengue is the most important vector-borne viral disease of humans and is one of the most important emerging tropical diseases globally. Efforts to control the disease have been largely unsuccessful. The changing epidemiology and disease burden of Dengue, and challenges for reversing the trend of increasing epidemic disease will be discussed.

The epidemiology of Dengue is really driven by a number of trends that have became more intensified through the years specially in the last 15 years. First the unprecedented population growth and the economic growth that has occurred in developing countries such in Asia as well as in Latin America; this whole leading to a greater urban growth and globalization favoring the seeding and spreading of the disease. Increased movement of people, animals, and commodities and within the vector and pathogens of several diseases including Dengue. Furthermore, in the case of Dengue it has been introduce in areas of the world where public health infrastructures do not really exist. The 20th Century Re-emergence of Dengue

Average annual number of DF/DHF cases reported to WHO, 1955-2007

908 15,497

122,174

295,554

479,848

968,564

0

200000

400000

600000

800000

1000000

1200000

1955-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2007

Num

ber

of c

ases

0

10

20

30

40

50

60

70

Num

ber

of c

ount

ries

• Expanding geographic Distribution • Increased epidemic activity • Hyperendemicity • Emergence of DHF and increasing • Disease severity

Urbanization is the key factor for the propagation of diseases such as Dengue by creating a perfect environment for the nesting of the vectors; it plays a role whether for limiting the control of the diseases as well as for the introduction of vaccines or preventive measures. The impact of these trends in the different regions, this following graph just shows the impact of degue re-emergence in the Americas.

When looking individual countries within a region there is a lot of variations from country to country, this also portraits that the surveillance from country to country is not the same and in many cases is totally inadequate, often Dengue cases are underreported. Singapore is one of the countries were reported data is reliable.

Re-Emergence of Dengue in the Americas

0 8,822

77,480104,412

266,818

656,240

0

100000

200000

300000

400000

500000

600000

700000

1955-1959 1960-1969 1970-1979 1980-1989 1990-1999 2000-2007

Num

ber

of c

ases

0

5

10

15

20

25

30

35

40

45

Num

ber

of c

ount

ries

When looking at country common data, one can observe the discrepancies of reported data, according to the bottom graph must Dengue cases recorded come out of the American region yet if you look at the data with deaths (graph following this one) the scenario is exactly the opposite. There is a gross underreporting of Dengue.

Burden of Dengue Total population at risk •3.6 billion people at risk for symptomatic Dengue •55% of the world population living in countries at risk for locally acquired Dengue infection Total Dengue infections •270 million Dengue infections annually

- 230 million asymptomatic infections •Social impact of epidemic Dengue Total symptomatic Dengue cases •34 million cases of Dengue fever annually •2 million DHF cases annually Total Dengue deaths •21,000 deaths annually New Dengue Estimates

Comparison of Dengue burden of disease with other mosquito borne diseases

Disease Cases Deaths Estimated Countries Affected

Dengue 36 million 21,000 124

Yellow Fever* 200,000 30,000 >42

Japanese Encephalitis* 50,000 >10,000 >10

Malaria* 500 million >1 million >105

Comprehensive costs of Dengue: Thailand, Panama, and Puerto Rico In terms of travelers Potential impact of outbreak on tourist revenues

In conclusion Dengue produces Public Health Impact, Social Impact, Economic Impact.

Item Thailand PanamaPuerto

RicoThailand vs

PanamaPopulation (million) 62 3,3 3,9

Cost of dengue illness (US $ million) $158 $11,8 $21,1Per capita cost of illness $2,55 $3,58 $5,41 -29%

Cost of dengue vector control (US $ million) $62 $5,0 $7,7Per capita cost of vector control $1,00 $1,52 $1,97 -34%

Total cost of dengue (US $ million) $220 $16,9 $28,8Per capita cost of dengue $3,55 $5,22 $7,38 -32%

Per capita GDP $2 750 $4 630 $17 100 -31%Dengue / GDP 0,13% 0,11% 0,04% 15%

Vector control share of dengue costs 28% 30% 27% -7%

Thailand based on officially reported cases only, $48 millionArmien B et al.: Am J Trop Med Hyg 2008, 79(3):364–371.Kongsin S et al.; Dengue Bulletin 2010, in press.Perez C et al.: Dengue Bulletin 2010, in press.Halasa Y et al.: Unpublished dataIndex mundi, CIA factbook - http://www.indexmundi.com/puerto_rico/gdp_per_capita_(ppp).html

Epidemiology and Disease Burden of Infections with Encephalitic Flaviviruses J. T. Roehrig, Arboviral Diseases Branch, Division of Vector-Borne Diseases, U.S. Centers for Disease Control and Prevention, Fort Collins, CO. Arthropod-borne flaviviruses cause a wide variety of human illness ranging from hemorrhagic fever to encephalitis. The flaviviruses that cause human encephalitis or neuroinvasive disease (ND) belong to two antigenically-related serocomplexes: the tick-borne encephalitis virus (TBEV) complex and the Japanese encephalitis virus (JEV) complex. Medically important flaviviruses that are members of these serocomplexes cause disease throughout the world: TBEV and TBEV-related viruses, e.g., Powassan virus (Europe, Asia, and North America); JEV (Asia, northern Australia); West Nile virus, WNV (Africa, Asia, Europe, North and South America, and Australia as Kunjin virus); St. Louis encephalitis virus, SLEV (North and South America), and Murray Valley encephalitis virus, MVEV (Australia). Of these viruses only TBEV, JEV, and WNV currently cause significant numbers of human infections, with JEV and WNV being responsible for human epidemics. The last major epidemic of SLEV occurred in the late 1970’s in North America. For all encephalitic flaviviruses the human is a dead-end host. Identification of laboratory-confirmed human cases of JEV and WNV encephalitis is confounded by the serological cross-reactivity of human infection-immune sera with other flaviviruses that might co-circulate with these viruses (e.g., SLEV and WNV in North America, and JEV and Dengue viruses in Asia). Identification of either virus or viral RNA in clinical specimens is the surest way to confirm infections with these viruses. When diagnosis requires serological testing, identification of virus-specific IgM or virus-neutralizing antibody is needed for confirmation of infection. Human TBE Cases in Europe and Russia, 1990-2006

1990 1 2 3 4 5 6 7 8 9 2000 1 2 3 4 5 6

Cas

e N

umb

er

YearLehrer and Holbrook, Tick-Borne Encephalitis, in Vaccinefor Biodefense and Emerging Infectious Diseases, p. 713, 2009.

EuropeRussia

Because most human infections with TBEVs occur through the bite of an infected Ixodid tick, these viruses primarily cause endemic disease. TBEV infection of humans is controlled through vaccination with inactivated TBEV vaccines that have been in use for many years. TBEV Vaccines Both JEV and WNV are transmitted through the bite of an infected Culex sp. mosquito. JEV is still a major cause of encephalitis in children throughout Asia. Multiple JEV vaccines are used to interdict and prevent human outbreaks, however their use is not yet universal. Both inactivated and attenuated JEV vaccines are used. No vaccine is currently available for WNV, even though this virus has been responsible for a recent major human epidemic in the Western hemisphere. JEV Vaccines Kurane, Japanese Encephalitis, in Vaccine for Biodefense and Emerging Infectious Diseases, p. 527, 2009.

Type of Vaccine Organ or cells Virus Strain Countries

Inactivated Chick embryofibroblast (CEF) cells

Neudorfl (W) Aus, Ger, Swi, Hun, CRep, UK, Canada, Baltics

K-23 (W) Aus, Ger, Swi, CRep, Russia, Baltics

205 (E) Russia

Sofjin (E) Russia

Type of Vaccine Organ or cells Virus Strain Countries

Inactivated Chick embryofibroblast (CEF) cells

Neudorfl (W) Aus, Ger, Swi, Hun, CRep, UK, Canada, Baltics

K-23 (W) Aus, Ger, Swi, CRep, Russia, Baltics

205 (E) Russia

Sofjin (E) Russia

Type of Vaccine Organ or cells Virus Strain Countries

Inactivated Mouse brain Beijing-1 Japan, Thailand

Nakayama Japan, Taiwan, Vietnam, India

Primary hamster kidney (PHK) cells

P3 China

Vero cells P3 China

SA-14-14-2 Austria, USA-(Ixiaro)

Beijing 1 Japan

Live attenuated PHK cells SA14-14-2 China

Vero cells SA14-14-2* USA

Vero cells YF/SA14-14-2* France (IMOJEV)

Type of Vaccine Organ or cells Virus Strain Countries

Inactivated Mouse brain Beijing-1 Japan, Thailand

Nakayama Japan, Taiwan, Vietnam, India

Primary hamster kidney (PHK) cells

P3 China

Vero cells P3 China

SA-14-14-2 Austria, USA-(Ixiaro)

Beijing 1 Japan

Live attenuated PHK cells SA14-14-2 China

Vero cells SA14-14-2* USA

Vero cells YF/SA14-14-2* France (IMOJEV)

In 1999 WNV was discovered in New York City, and since then the virus has moved throughout the United States (U.S.) and Canada, and southward into Mexico, Central and South America. WNV is currently the leading cause of viral encephalitis in the U.S., with 20% of human infections resulting in clinical disease, and less than 1% of human infections resulting in WNND. WNV Vaccines

• Horses (approved) – TC killed vaccine – Avipoxvirus – DNA - Birds

• Humans (in development) – WNV/YF ChimeriVax – DNA – Subunit (E protein)

The overall case-fatality rate for individuals with WNND is 9-10%. Prior to the approval of an equine WNV vaccine, the equine case-fatality rate was 30% for horses in the U.S. There are no approved prophylactic or therapeutic agents that have been shown to be effective for flaviviral infections. Both human and humanized monoclonal antibodies (MAbs) are currently undergoing intensive study in this regard. A humanized MAb that has been shown to be both protective and therapeutic in animal models for WNV infection is currently in human clinical trials in the U.S. SLEV: Background and Epidemiology First isolated in the United States in 1933 and then was found in North, Central, and South America. Transmited by mosquito (Culex)-bird-mosquito-human transmission cycle. CFR = 5-30%; greatest in elderly and approximately 80 cases/yr; 4,232 cases in U.S. (1964-2008) Conclusions • Five medically important encephalitic flaviviruses: TBEV, JEV, WNV, SLEV, and MVEV • World-wide distribution • Zoonoses: mammal and bird vertebrate reservoirs • Transmitted to humans by the bite of infected ticks or mosquitoes • Disease ranges from inapparent infection to frank encephalitis and death • Approved human vaccines only for TBEV and JEV • No known treatment

Epidemiology and disease burden of infection with y ellow fever virus Alan D. T. Barrett, Ph.D. Sealy Center for Vaccine Development and Department of Pathology, University of Texas Medical Branch at Galveston The disease yellow fever (YF) has been known for over 400 years, today YF virus is found predominantly in tropical and subtropical regions of Africa and South America with over 40 countries and a population of >900 million at risk. Case:fatality rate is up to 50%. Yellow fever (YF) is caused by yellow fever virus (YFV) and is found in sub-Saharan Africa and tropical regions of South America where it is endemic and intermittently epidemic. The majority of cases are reported in sub-Saharan Africa. Only a small proportion of yellow fever cases are officially reported because of the occurrence of the disease in remote areas and lack of specific diagnostic facilities. The virus is transmitted between primates by mosquitoes (Aedes species in Africa, and Hemagogus and Sabethes species in South America). The normal vertebrate hosts are non-human primates but humans can get the disease if bitten by a virus-infected mosquito. When humans come into contact with the jungle vector, they are at risk for infection (jungle transmission cycle), an urban cycle (in which transmission is carried from human to human by domestic mosquitoes) can occur. The Paradigms of YF transmission are that has two cycles: sylvatic and urban, was introduced to South America from Africa (sometime of the slave trade along with vector Aedes aegypti), circulates in wandering epizootics, reservoir hosts are non-human primates, and there are three zones of transmission:

–endemic zone –zone of emergence –zone at risk for epidemics

In the last 25 years there has been an increase in the number of cases of YF and accordingly it has been classified as a reemerging disease. The virus exists as at least seven genotypes; two in West Africa, three in Central and East Africa, and two in South America. Evolutionary studies indicate the virus originated in Central or East Africa, moved to West Africa, and then spread to South America during the slave trade era. Geographic distribution of genetic variants of YFV

• West African YFV: largest population, least diverse, slowest evolving, constant growth = endemic

• YFVs circulating in South America are comprised of two major genotypes; • Genotype I viruses circulate in wandering epizootics; • Genotype II viruses persist in discrete enzootic foci.

Percent Nucleotide and Amino Acid Variation Among Genotypes

Yellow Fever Clinical Manifestations - courtesy of Pedro Vasconcelos, IEC

YF epidemiology

YF virtually disappeared in francophone West African countries as a result of YF mass vaccination campaigns carried out between 1940 and 1953. However, because of the failure to continue mass vaccination campaigns, a resurgence of YF in many African countries began in the early 1980s. There has been a dramatic increase in the number of cases of YF since the late 1980s due to a combination of factors:

• declining vaccine-induced immunity in the population, • deforestation, • urbanization, • population movements, and climate change

Therefore, YF is considered a re-emerging disease There is no antiviral therapy to treat the disease and the disease is prevented by the use of a live attenuated vaccine, strain 17D, which is considered to be one of the most effective and safe vaccines. The vaccine was developed in the 1930s and has been administered to over 550 million people. It is effective against all known genotypes of the virus despite genetic and antigenic variation. Despite the availability of an effective vaccine there are still estimated to be 200,000 cases of wild-type YF each year, including 30,000 deaths. Without the availability of a vaccine, YF would be considered a biosafety level-4 pathogen. Recently, reports of serious adverse events, including fatal outcomes, in temporal association with 17D yellow fever vaccination have highlighted the need for good surveillance and assessment of the benefit/risk of receiving the vaccine. YF vaccine: manufacturing and production issues

• Only 6 producers in the World. • Can produce 200 million doses annually. • Vaccine still produced in embryonated chicken eggs using technology that has not

changed since the 1940s. Issues facing today Epidemiology: Risk Yellow Fever in Urban Areas Epidemiology: Risk Yellow Fever Unvaccinated Travellers Summary •17D vaccine has proved to be a very successful vaccine and has been given to over 500 million people. •Despite the availability of a very successful vaccine there are still 200,000 cases of YF each year. •YF is an re-emerging disease with increasing number of cases in the past 20 years. •The risk of urban YF is continuing to increase with time. •Rare severe adverse events (YEL-AVD and YEL-AND) following immunization are a cause for concern in a society with expects a vaccine to be 100% effective and 102% safe.

Session 2 - Molecular structure of flaviviruses; cellular receptors of flaviviruses; flaviviruses cell entry mechanisms Molecular and antigenic structure of flaviviruses Franz X. Heinz, Department of Virology, Medical University of Vienna Flaviviruses are isometric lipid-enveloped particles with a diameter of about 50 nm.

Virus assembly occurs in the endoplasmic reticulum and first leads to the formation of non-infectious immature virions, composed of a spherical core (containing the genomic RNA and multiple copies of the capsid protein) and a lipid bilayer with two membrane-associated proteins, prM and E. These proteins form an icosahedral scaffold of 60 spikes of heterodimers.

The generation of infectious virions requires the proteolytic cleavage of prM by furin or similar proteases which occurs in the trans-Golgi-network during exocytosis of virus particles, shortly before their release from infected cells. Fully mature virions (in which all of the prM proteins have been cleaved) have a smooth surface and the E proteins are re-organized to form a closed shell of densely packed E-dimers. Parallel sets of three such homodimers form rafts that are arranged in a ‘herringbone’ pattern at the virion surface. The acquisition of infectivity does not require the cleavage of all prM-molecules in a virion and the release of partially mature but infectious particles is a hallmark of cells infected with certain flaviviruses. The E protein has a double membrane-spanning anchor, an alpha-helical ‘stem’ region and an ectodomain - composed of three distinct domains (DI, DII, DIII) - that is oriented parallel to the membrane in mature virions. It mediates interactions with the cellular surface as well as membrane fusion in endosomes after uptake by receptor-mediated endocytosis and is the major target of virus-neutralizing antibodies. Epitopes involved in virus neutralization have been mapped to each of the three domains. The avidity of antibodies to these epitopes and their accessibility determine the degree of occupancy at any given concentration and are the most important determinants of neutralization potency. So far, the highest neutralizing activity was described for monoclonal antibodies directed to DIII. The extent of cross-reactivity between flaviviruses and cross-neutralization between and within the flavivirus serocomplexes is a reflection of amino acid sequence divergence in E. Broadly flavivirus cross-reactive antibodies are primarily directed to the highly conserved fusion peptide loop at the tip of domain II and their contribution to virus neutralization varies among different flaviviruses. Antigenic Relationships of Flaviviruses

Non-neutralizing antibodies or antibodies at sub-neutralizing concentrations can lead to the phenomenon of antibody-dependent enhancement of infection of Fc-gamma receptor-bearing cells, implicated in the exacerbation of disease during secondary Dengue infections or in infants born to seropositive mothers. Such enhancement phenomena can also be observed with barely neutralizing antibodies to prM in combination with partially immature viruses, as is characteristic especially for Dengue viruses. Even completely immature virions can be rendered infectious by prM-specific antibodies that mediate virus uptake into Fc-receptor positive cells followed by prM cleavage at a post-entry stage. The possible biological implications of the induction of antibodies to prM are an important issue in the context of Dengue vaccines. Immature Dengue virus: A veiled pathogen? Izabela A. Rodenhuis-Zybert, Julia Da-Silva Voorham, Bastiaan Moesker, Vanesa Ayala-Nuñez, Heidi van der Ende-Metselaar, Jan Wilschut, and Jolanda M. Smit Department of Medical Microbiology, Molecular Virology Section, University Medical Center Groningen, University of Groningen, Groningen, The Netherlands Dengue virus (DENV) is a major emerging pathogen which causes disease symptoms ranging from febrile illness to devastating hemorhagic manifestations. Increased disease severity is associated with pre-existing DENV antibodies and high circulating virus titers, which suggests that antibodies directly influence the infectious properties of the virus. The molecular mechanism by which antibodies enhance DENV infection however remains elusive. Cells infected with DENV release a high proportion of prM-containing immature virions. Flavivirus-infected cells release a high proportion of particles containing unprocessed prM

It is generally believed that immature particles are irrelevant by-products of infected cells since numerous functional studies have demonstrated that fully immature particles lack the ability to infect cells. On the other hand, Dengue-positive patients secrete substantial levels of prM antibodies, which suggest that immature particles are involved in disease pathogenesis. In attempt to unravel these contrasting paradigms, we investigated the infectious properties of anti-prM antibody-opsonized immature DENV in various FcR- expressing cell lines and human primary PBMCs (peripheral blood mononuclear cells). We found that immature DENV particles become virtually as infectious as wild type virus particles in presence of antibodies. We observed that prM antibodies facilitate efficient binding and cell entry of immature particles into Fc-receptor expressing cells. In addition, enzymatic activity of furin is critical to render the internalized immature virus infectious. Furthermore, we found that multiple E antibodies interact with immature particles and some of these antibodies stimulate the

infectious properties of immature virions in a furin-dependent manner. Infectious properties of immature DENV in presence of antibodies

Immature virions are highly infectious in presence of anti-prM antibodies, at least 10,000 fold increase in virus particle production

Anti-prM antibody mediates efficient binding of immature DENV to cells, blocking Fc ã receptors impedes enhancement of infection

however, increased binding does not fully explain observed enhancement , prM protein present in immature particle obstructs the infectious potential of immature virions, Is furin, the host protease, which circulates between the Trans Golgi Network and early endosomes, involved in maturation of internalized prM-containing virions? Furin activity is crucial to render immature virions infectious, prM antibodies enhance infectivity of wtDENV in a furin-dependent manner. Summary 1

Are E antibodies also able to stimulate the infectious properties of immature DENV particles? A vast majority of E antibodies interact with immature DENV

-12 anti-E DI/II, 14 DIII and 3 anti-E antibodies werevrandomly selected –generated in alpha/beta interferon receptor-deficient C57BL/6v mice -Diamond lab

-86% of the antibodies tested bind to prMDENV as measured by direct ELISA Limited number of anti-E Abs stimulate infectivity of immature DENV

-8 out of 25 antibodies (1/3) enhance immature DENV infectivity –3 anti-E DI/II; 3 anti-E DIII; 2 anti-E

-Distinct enhancement pattern for E vs prM antibodies One can determine that enhancement is strictly dependent on enzymatic furin activity Also, we investigated the cell entry characteristics of prM-opsonized immature DENV particles in macrophages by single particle tracking analysis and revealed that the dynamics and route of viral entry is different compared to that of wild-type DENV particles. Altogether, our data suggests that in presence of antibodies, immature DENV particles have the potential to be highly infectious and hence may contribute to DENV pathogenesis. During the conference the critical determinants in immature DENV cell entry will be discussed.

Summary-2 •Enhancement of infection by E antibodies is epitope specific •Antibody-mediated infection of immature DENV is dependent on enzymatic activity of endosomal furin •Dissociation of the pr peptide following furin cleavage is most likely mediated by the low pH environment in endosomes •Antibodies that do not stimulate infectivity presumably interfere with virus maturation upon entry “Structural basis and mechanisms of neutralization” Kimberly Dowd, NIH, USA Neutralizing antibodies are a significant component of the host’s protective response against flavivirus infection, and are directed primarily against the envelope (E) protein. Reporter virus particles (RVPs) are a quantitative tool for the study of flavivirus entry and its inhibition.

Neutralization of flaviviruses occurs when individual virions are engaged by antibodies with a stoichiometry that exceeds a required threshold. WNV neutralization is a multiple-hit mechanism governed by a stoichiometric threshold, WNV neutralization requires binding of ~30 mAbs per virion. From this "multiple-hit" perspective, the neutralizing activity of antibodies is governed by the affinity with which it binds its epitope and the number of times this determinant is displayed on the surface of the virion. These two governing factors determine whether or not neutralization will occur. (Pierson et al., 2007)

If the number of times an epitope is displayed on the virion surface does not exceed the stoichiometric threshold, neutralization will not occur, regardless of antibody affinity. Antibody affinity determines the fraction of epitopes bound by antibody, High affinity antibodies are better.

(Pierson et al., 2007) At least two factors modulate epitope accessibility. Several classes of epitopes are bound by antibodies whose neutralization potency is sensitive to the maturation state of the virion. In these cases, the process of maturation reduces the accessibility of epitopes on the mature virion below the threshold required for neutralization. Epitope accessibility controls the number of epitopes bound per virion, Occupancy requirements can vary significantly based on the location of a given epitope on the E protein. Thus Epitope accessibility is also modulated by the dynamic motion of the E proteins on the virion that occur at equilibrium.

Poorly accessible/ cryptic epitopes may never reach the threshold for neutralization, regardless of antibody affinity.

Epitope accessibility is a critical factor that governs flavivirus neutralization What modulates epitope accessibility?

Lok et al., 2008

Experiments utilizing monoclonal antibodies (MAbs) known to recognize relatively inaccessible epitopes on mature virions identified a role of virus "breathing" in regulating neutralization activity. Neutralization potency of certain classes of mAbs are sensitive to the maturation state of the virus. The maturation state of WNV affects epitope accessibility.

Nelson et al., 2008

Both time- and temperature-dependent increases in neutralization were observed that could be explained by exposure of inaccessible epitopes through dynamic movement of E proteins on the virion surface.

This phenomenon was observed in studies with a large panel of MAbs specific for epitopes in each domain of the WNV envelope protein, with sera from recipients of a live attenuated WNV vaccine, and in experiments with Dengue virus. Neutralization by some polyclonal sera is also maturation-sensitive. Functionally important viral epitopes targeted during infection may vary between individuals.

Virion maturation modulates antibody-mediated neutralization by changing epitope accessibility.

Viral dynamics modulates antibody-mediated neutralization by changing epitope accessibility

The dynamic movement of flaviviruses exposes epitopes that may then be stabilized by antibody binding. Under these conditions, all antibodies tested become neutralizing. Dengue appears more dynamic than WNV. Our findings suggest that the structural dynamics of flaviviruses has a widespread impact on antibody-mediated neutralization via changes in epitope accessibility; given enough time, even poorly neutralizing antibodies can dock on the virion with a stoichiometry sufficient for neutralization. Taken together, epitope accessibility is a critical yet dynamic factor that governs the functional properties of anti-flavivirus antibodies. Conclusions Epitope accessibility controls the potency of antibody-mediated neutralization and is modulated by:

• Maturation state of the virus- antibodies become less potent • Dynamic movement of virions- antibodies become more potent

Vaccine implications: Focusing the immune response on highly accessible epitopes will allow neutralization at lower concentrations of antibody AND will be able to neutralize a larger fraction of heterogeneous virion populations.

Session 3 - Pathogenesis and immune response; mecha nisms underlying in vivo protection;neutralization and en hancement of infections Clinical aspects of Dengue (the potential impact of vaccines) Dr Laurent Thomas, Emergency Department, University Hospital, Fort-de-France, Martinique (FWI) Potential impacts of Dengue vaccines Medical ° Prevention of symptomatic Dengue disease, i.e. co vering the spectrum from Dengue fever to severe Dengue cases due to serotypes 1, 2, 3 and 4

°Improve quality of life,decrease CFR,etc … Epidemiology ° Decrease the frequency, the magnitude and the sev erity of epidemics Main clinical data gap: under-reporting Issues concerning reporting

Case notification systems - the example of Martinique Martinique sentinel network - Distribution of 52 GPs

• GPs weekly report on : Flu-like, Dengue-like, Bronchiolitis, Gastroenteritis, Conjunctivitis, Asthma.

• Emergency network • Surveillance and reporting of severe cases • The surveillance system is linked to the vector control unit

Dengue epidemics in Martinique (2000- 2010) (data from CIRE-AG www.invs.sante.fr/surveillance/Dengue/ )

Lesson # 1 •A good case notification system will facilitate the measurement of the impact of Dengue vaccines on the frequency, the magnitude and the severity of epidemics •Captive population: good candidate for long term phase IV trial Case classification systems The 2 nd edition (1997) The 3rd edition (2009)

Main topics of the 1997 WHO classification •Full description of DHF and DSS •Recognition of plasma leakage as the main pathophysiological process of DHF •Clear recommandations about monitoring of patients around days 3-5 of illness, particularly around time of defervescence •Clear recommandations about the risks of fluid overload Limitations of the 1997 WHO classification •Difficulties in bedside use •Inappropriate for triage •Plasma leakage develops only in a little proportion of patients (17% of hospitalized patients in Martinique) •Up to 50% of patients with plasma leakage do not meet other DHF/DSS criteria •Under-reporting of other severe clinical forms (hepatitis, encephalitis, myocarditis) Lesson # 2 •WHO 1997 and DENCO 1999 are not a way to divide the world! They are not mutually exclusive and are complementary. •DENCO 1999 is useful for triage and monitoring of patients during the acute febrile phase; WHO 1997 should be used for the management of patients with plasma leakage •There is a considerable overlapping between clinical forms. Clinicians should be aware of the potential occurrence of severe clinical forms as a function of the time elapsed since the onset of fever. •Possible differences in clinical presentation due to genetic backgrounds Lesson # 3 & conclusion •Global Dengue threat: urbanization, vector expansion, resistance to insecticides •Hyperendemicity: more epidemics, more children and adults with severe forms •Emergence of co-epidemics: Dengue / bronchiolitis and/or gastroenteritis; Dengue and YF;Dengue and Chikungunia;etc •Urgent need for a vaccine!

Novel mechanisms of flavivirus innate immune evasio n and control Michael S. Diamond, Stephane Daffis, Kristy J. Szretter, Jianqing Li, Jill Schriewer, Roland Zust, Hongping Dong, Volker Thiel, R. Mark Buller, Michael Gale, Jr., and Pei-Yong Shi

Washington University School of Medicine, St Louis, MO 63110. Saint Louis University School of Medicine, St. Louis, MO 63104. Kantonal Hospital St. Gallen, St. Gallen, Switzerland. New York State Department of Health, Albany NY 12208. University of Washington School of Medicine, Seattle, WA 98195. (Lecture restricted) Type I interferon (IFN) cell-intrinsic antiviral defenses protect against many virus infections by signaling host blockade of viral translation, transcription, and replication, thus limiting spread and pathogenesis. Cellular mRNA of higher eukaryotes and many viral RNA are methylated at the N-7 and 2’O positions of the 5’ guanosine cap by specific nuclear and cytoplasmic methyltransferases (MTases), respectively. West Nile Virus - Immune control

Whereas N-7 methylation is essential for RNA translation and stability, the function of 2’O methylation and its role in virus infection has remained uncertain since its discovery 35 years ago. We show that a West Nile virus (WNV) mutant that lacks 2’O MTase activity was attenuated in wild type primary cells and mice but was pathogenic in the absence of IFN signaling. 2’O methylation of viral RNA did not affect IFN induction in WNV-infected cells but instead modulated the antiviral effects of IFN-induced proteins with tetratricopeptide repeats (IFIT), which are interferon-stimulated genes (ISG) implicated in regulation of protein translation. The first part of this talk will focus on recent results demonstrated that the 2’O methylation of the 5’ cap of viral RNA functions to subvert innate host antiviral responses through escape of IFIT-mediated suppression. Thus, differential methylation of cytoplasmic RNA serves as a paradigm for pattern recognition and restriction of propagation of foreign viral RNA in host cells. The second part of the talk discussed recent progress on using high-throughput shRNA screens in mouse and human cells to identify novel interferon stimulated genes that restrict flavivirus infection.

Improved Dengue-specific adaptive immune responses in humanized BLT mice Anuja MATHEW and Smita Jaiswal,Alan L.Rothman,Leonard D.Shultz,Dale L.Greiner, Michael Brehm ; Center for Infectious Disease and Vaccine Research -USA Challenges with human clinical studies: •Most patients who present to the hospital live in endemic areas and are experiencing a secondary infection.

- the serotype of the previous DENV infection is difficult to determine. •Dengue illness is more commonly seen in pediatric patients living in Asia or South America, and obtaining large volumes of blood from children is difficult. •Cryopreservation or prompt shipping of infectious samples (known to impact the recovery of activated plasmablasts) from remote areas makes cell-based studies difficult. •Controlled virus challenge studies in humans are not feasible. Human Tissue Experimental systems available to study the pathogenesis of DENV disease.

Animal Models for Dengue: Primates

•Viremic after subcutaneous infection •No disease (even after secondary infection) •Expensive (Guirakhoo et al., 2004; Schiavetta et al., 2000; Angsubhakorn et al., 1987)

Mice •Immunodeficient –SCID mice engrafted with human cell lines (Wu et al, 1995; Lin et al, 1998; An et al, 1999;)

•Pathological effect of Dengue in relation to human tissues –AG129 (IFN a/b/g Receptor -/- ) and STAT1 -/- (Johnson and Roehrig, 1999; Shresta et al, 2004 (a), 2004 (b), 2005, 2006;Zellweger 2010;Balsitis 2010)

•Type I interferons are essential for the control of DENV and utilizes both a STAT1-dependent and independent mechanism.

•Immunocompetent (Rothman et al, 1996; Spaulding et al, 1999; Beaumier 2009; Beaumier 2010; Huang et al, 2000; Atrasheuskaya et al, 2003; Chen et al, 2004; Paes et al, 2005; Barth et al, 2006; Barretto et al, 2007) The lack of a suitable animal model to study viral and immunological mechanisms of human Dengue disease has been a constraint to Dengue research. Why Do We Need Humanized Mouse Models? –Outcomes predicted by murine studies are not always representative of actual outcomes in humans –Permits study of human-specific T and B cell immunity to Dengue virus infections Goal –Enable clinically relevant in vivo studies of human cells, tissues, and immune systems without putting patients at risk. Humanized mice are defined as immunodeficient mice engrafted with haematopoietic cells or tissues from humans or mice that transgenically express human genes.

We recently established an animal model for Dengue virus (DDENV) infection and immunity using non-obese diabetic/severe combined immunodeficiency interleukin-2 receptor ã-chain knockout (NOD-scid IL2rã null ) mice engrafted with human hematopoietic stem cells. We sought to further improve this model by using humanized BLT mice which involves the cotransplantation of human fetal thymus and liver tissues and CD34+ fetal liver cells into NOD- scid IL2rã null mice. The education and maturation of human T cells on autologous human thymic tissue in the BLT model was a key factor which led us to

hypothesize that infection with DENV would lead to improved virus-specific cellular and humoral immune responses.

Are humanized NOD-scid IL-2r g null mice a useful animal model to study Dengue immunity and pathogenesis ? •Do we get productive DENV infection in humanized NOD-scid IL-2r g null mice? -YES •Is virus cleared? -YES •Can we generate human DENV- specific antibody responses? -YES (IgM responses) Dengue virus infection and immune response in humanized RAG2(-/-)gamma(c)(-/-) (RAG-hu) mice. (Kuruvilla JG, Troyer RM, Devi S, Akkina R. Virology. 2007 Dec 5;369(1):143-52.) •Can we generate antigen-specific T cell responses? -YES Dengue virus infection and virus-specific HLA-A2 restricted immune responses in humanized NOD-scid IL2r g null mice. (Jaiswal S, Pearson T, Friberg H, Shultz LD, Greiner DL, Rothman AL, Mathew A. PLoS One. 2009 Oct 5;4(10):e7251.) •Do we detect any signs and symptoms of Dengue disease? - inconsistent symptoms Humanized mice show clinical signs of Dengue fever according to infecting virus genotype. (Mota J, Rico-Hesse R. J Virol. 2009 Sep;83(17):8638-45.) Our data show consistent DENV replication in the bone marrow of engrafted mice. We detected significantly higher anti-Dengue specific antibody titers in the sera of BLT mice compared to the previously tested standard and HLA-A2-ttransgenic NOD- scid IL2rã null engrafted mice. Antibody responses were sustained for up to 6 weeks in the sera of infected BLT mice. Human T cells that developed following engraftment of BLT mice with HLA-A2+ human cord blood hematopoietic stem cells secreted IFN-ã in response to stimulation with overlapping peptides that spanned the entire DENV genome. Are immune responses in humanized BLT-NSG superior to responses in cord blood engrafted NSG mice? •Human DENV- specific antibody responses? -YES (IgM responses) •Human antigen-specific T cell responses? -YES

•Do we detect any signs and symptoms of Dengue disease? - inconsistent symptoms Furthermore, four A2-restricted Dengue peptides-NS4b 2353 (111-119),NS4b 2423 (181-189),NS4a 2148 (56-64)and a novel epitope on NS5 specifically stimulated IFN-ã secretion in human T cells. Overall, BLT mice will provide a much needed platform to assess human immune responses to DENV vaccines and the effects of prior immunity n subsequent ENV infections. Innate and adaptive immunity to Dengue vaccines in development Bruno Guy, Claire Balas, Anke Harenberg, Audrey Kennel, Florence Deauvieau, Regis Sodoyer, Nadege Arnaud-Barbe, Melanie Saville, Jean Lang. sanofi pasteur (Lecture restricted) Dengue infection is a major and growing public health issue worldwide. Different vaccine candidates are being developed, including YFV 17D vaccine-based chimeric Dengue virus vaccines (CYDs). These vaccine candidates have been characterized from the early stages of research through to clinical development, to assess in particular their safety and immunogenicity. Firstly, innate responses were investigated in vitro in human monocyte-derived dendritic cells (mDCs), upon infection by each of the 4 CYDs and their tetravalent combination. This was first done by flow cytometry and ELISA on a limited set of parameters, and then in a second step by using Agilent DNA microarrays. In this latter study, CYDs were also compared to wt serotype 3 virus, or to a classically attenuated serotype 3 virus (VDV3) shown to be reactogenic in a clinical trial. The results of the second study confirmed and expanded upon the first: we observed a very reproducible signature for each of the 4 CYDs, involving stimulation of Type I IFN genes and associated ISGs, together with genes encoding chemokines and other mediators involved in the initiation of adaptive responses. In contrast, the wt virus induced a predominantly inflammatory profile, while VDV3 appeared to induce a blunted response, which may have been insufficient to trigger early immune responses and prevent initial viral replication. This could have contributed to VDV3 symptomatic outcome in clinical trials. These studies contributed to documenting the safety and immunogenicity of the 4 CYD candidate vaccine viruses, which are currently in evaluation in large scale efficacy trials. Secondly, adaptive immune responses induced by the vaccine candidates were monitored in Phase 1 and Phase 2 clinical trials. Th1 and CD8 responses were induced with an IFN-γ/TNF-α ratio favouring IFN-γ in both cases, regardless of whether the vaccine recipients were flavivirus naive or not. There was an absence of Th2 response in all cases. The Th1 response was dominated by serotype 4 in flavivirus naive individuals after initial vaccination but broadened to include all serotypes after second vaccination. This broadened response was also observed after primary Dengue TV vaccination in subjects previously administered monovalent Dengue 1 and Dengue 2 vaccines. Extending these analyses to children will be essential in the Dengue TV development program, and the techniques we used so far are being modified in an ongoing trial, in order to monitor cellular responses from only 3–5 ml of blood. In conclusion, pre-clinical and clinical results support the favorable immunogenicity and short-term safety of the Dengue TV. Ongoing and future studies will establish the longevity of the vaccine-induced immunity and requirements for boosters.

The yellow fever vaccine immunity in HIV infected p atients: development of new assays for virological and immunological monitoring in HIV infected patient. ( Essai EP46 ANRS NOVAA) N Colin de Verdière1, S Delarue1, V Melfreidy4, B Labrosse2, B Autran3, JP Aboulker4, JM Molina1, 2, F Simon1, 2, 1CHU Saint LOUIS, 2INSERM U 941, 3CERVI-Pitié Salpetrière, 4INSERM SC10, Paris, France In order to compare the virological and immune responses in HIV-positive and HIV-negative individuals YF naive-vaccine populations, we initiated a clinical trial phase III multicentric protocol in Paris, France. Specific tools for the evaluation of the cell-mediated and humoral immunity after YFV and virological responses were developed: - an ELISPOT technology for YF - a pseudotype -based assay for quantifying the anti-YF virus neutralizing antibodies. Thirty HIV-negative subjects and 40 HIV-positive voluntary subjects (CD4 > 350/mm3 under HAART for at least one year, with a viral load < 50 copies/mL since at least 6 months) will be included. HIV positive subjects will be matched according to age (18-40 years and 40-55 years) and sex, with the HIV negative subjects. Subjects will be vaccinated at J0 with STAMARIL and followed over one year. . The responses will be analysed within 7, 14, 28, 90 and 365 days of administration of YFV in terms of : (1) seroconversion by EIA, (2) cytotoxic response in ELISPOT, (3) neutralizing antibody levels in PRNT (reference method) and a new pseudotype based method, (4) post-vaccination viremia and (5) diversity of viral quasi-species. The titles and kinetics of viremia, neutralizing antibodies and ELISPOT will be considered as surrogate markers of response in terms of groups. Impact of YFV on the T-lymphocyte response against HIV by ELISPOT and HIV-1 RNA viral load will be assessed. Subjects with previous vaccination against YF or YF anti-IgG positive, recent immunoglobulins therapy, HCV infection, history of thymic dysfunction (including thymoma and thymectomy) or others immunosuppression, whether congenital, idiopathic will not be included in the study. Clinical and biological tolerance: At all follows up will be measured the incidence of CDC classification events (for HIV+) and general and local reactions of degree ≥ 2 in the setting of the injection of STAMARIL® Shedule : Date of first enrolment: December 2010 Inclusion period: 18 months

Session 4 - Vaccination against flaviviruses: vacci nes in development

IMOJEV TM a new single dose recombinant vaccine for Japanese encephalitis Farshad Guirakhoo, Senior Director External R&D North America, Sanofi Pasteur Flaviviruses are a global public health concern affecting nearly half of the world’s population. Japanese encephalitis (JE) is a mosquito-borne infection with high epidemic potential, ~30% case-fatality rate, and high severe sequelae among survivors long after their acute illness. About 3 billion people live in endemic regionsAlthough several types of JE vaccines are available, disadvantages including reactogenicity, production shortages, and need for multiple doses hinder their effectiveness to combat JE. IMOJEV™ is a live attenuated single dose vaccine developed to provide convenient immunization against JE. It was constructed by inserting coding sequences of the premembrane and envelope structural proteins of the JE vaccine strain SA14-14-2 virus into the core and non-structural genes of yellow fever (YF) vaccine strain 17D virus. IMOJEV™ Vaccine • Live attenuated • Constructed from two widely used LAVs

(Live Atenuated Viruses): o YFV strain 17D o JEV strain SA14-14-2

• Grown in Vero serum-free • Freeze-dried • Saline as diluent • No preservative or adjuvant • Single dose for primary immunization

o >4 log PFU in 0.5 mL per injection • Subcutaneous administration F 17D JE SA14-14-2

Recovery of Chimeric Vaccine Virus

At the end of the above process you have a virus which has 17 D C and a non structure gene from Yellow Fever and covered by the JE envelope proteins which are the immunogens of interest. Generally, the single does LAV produce longer duration of immunity and T cell response which is superior to recombinant or inactivated vaccine if you compare one dose vs. one dose. Since it is a recombinant virus we have to go through a series of Nonclinical safety studies using a series of animal models. Guirakhooflavivirusmerieux10 5 A single dose of a LAV induces

IMOJEV™: Nonclinical Safety Studies Neurovirulence (IC inoculation)

Less neurovirulent than YF 17D vaccine virus (mice, monkeys) Neuroinvasiveness (IP inoculation)

Not neuroinvasive (mice, hamsters, monkeys) Viremia (IC or SC)

Low, transient viremia (monkeys) Toxicology (SC)

No adverse clinical signs and no toxicological findings (monkeys) No organ dysfunction and no histopathological lesions (monkeys)

Biodistribution (SC) No shedding, and no virus detected in any organs (monkeys)

Environmental risks Genomic stability in vitro (Vero) and in vivo (monkeys) No natural recombination (artificial recombination with WT virus reduced virulence of WT virus) Does not infect mosquitoes by oral route

The studies showed that IMOJEV™ was significantly less neurovirulent than an YF 17D vaccine (YF-VAX®) in both mouse and monkey models inoculated by the intracerebral (IC) routes. A single subcutaneous low dose of IMOJEV™ conferred 100% protection in monkeys against a lethal IC challenge with a highly virulent JE virus. Environmental safety testing included vector transmission, biodistribution and recombination studies showed that the vaccine virus is genetically stable in vitro and in vivo, does not shed to the environment from vaccinees, is unable to infect mosquitoes, and when artificially recombined with a wild type virus, its neurovirulence phenotype is lower than that of the wild type parental virus. Once the nonclinical studies where showed good results, we passed to the Clinical phases, to demonstrate efficacy based on a serological correlate of protection

• PRNT50 ¡Ý1:10 dilution • Non-inferiority to a licensed JE vaccine • Widely accepted by WHO experts* • Accepted by health authorities (FDA, EMEA, TGA)

First trials were completed in adults in studies run in the USA and Australia ; ongoing studies in pediatric populationsin Thailand, the Philippines and India. IMOJEV™: Phase III Study Designs in Adults

IMOJEV™: Immunogenicity vs. JE-VAX ® in Efficacy Population

IMOJEV™: Conclusion on Immunogenicity and Safety in Adults Safety*

Systemic reactions • IMOJEV™= placebo (very similar reactions between both)

Reactions at injection site • Significantly lower with IMOJEV™(67.6%) than with JE-VAX ® (82.2%)

(p<0.001) Immunogenicity*

IMOJEV™ elicits a high level of neutralizing antibodies after a single dose • GMT: 1,392 on Day 28 • >93% seroconversion after 14 days • 99% seroconversion after 28 days

Non-inferiority* Seroconversion after a single IMOJEV™ vaccination (99.1%) was statistically non-inferior to that after three-doses of JE-VAX ® (95.1%) vaccination

Duration of Immunity** 87% predicted to remain protected 5 years after a single IMOJEV™ vaccination (Kaplan Meier analysis)

**: Nasveld et al. Human Vaccines 2010, in press *: Torresi et al., Vaccine 2010 Nov 23;28(50):7993-8000. Epub 2010 Oct 8

IMOJEV™: Immunogenicity and Safety Study in Toddlers 12 to 24 months

Viremia and SAEs in Toddlers 12- 24M, viremia was assessed on D4 and only 5.1% presented low level viremia just over the level of quantitation (20.0 PFU/mL). In terms of SAEs, no SAEs related to vaccination nor death was reported. Immune Response, Toddlers 12-24M*

Antibody persistence 1 year after Vaccination in JE-naïve toddlers (Phase III JEC05)

Summary of Clinical Studies Adults • Completed • 9 clinical studies in Australia and USA

-Phase III studies in comparison to JE-VAX® and placebo -Long-term follow-up up to 5 years after single dose of IMOJEV™

• More than 2,500 subjects received IMOJEV™ Pediatric populations • Long-term follow-up up to 5 years after single dose of IMOJEV™ • Concomitant administration with pediatric vaccines • Ongoing Phase II and Phase III clinical studies in India, Thailand, The Philippines,

Taiwan • 9 months to 17 years of age • More than 1,500 infants/toddlers/children received IMOJEV™ As part of global clinical studies more than 4,000 subjects received IMOJEV™. Overall Conclusions: IMOJEV™ Single dose : Increased compliance, increased coverage Rapid protection : very positive for ‘last minute’ travelers High immunogenicity : High level of protection, Long term protection expected Good safety profile :

• Similar to placebo • Better than JE-VAX ® • No preservative, no adjuvant

Modern, high quality manufacturing • Vero cell substrate • Serum-free • Lyophilised, stable vaccine (36 months shelf life)

In total the nine clinical studies (Phase I/II to Phase III) in healthy adult populations in the USA and Australia and three clinical studies (Phase II and Phase III) in healthy children, toddlers and infants conducted with IMOJEV™ demonstrated a profile of a highly safe and effective vaccine. Further trials are underway in pediatric populations. IMOJEV™ is currently licensed in Australia as a single dose vaccine. A new cell culture inactivated vaccine for yellow f ever Dennis Trent, University Texas Medical Branch –USA Despite the existence of a yellow fever vaccine - YF, there are issues that need further consideration. Live, attenuated yellow fever (YF)17D vaccine is highly efficacious but causes rare, serious adverse events resulting from active replication of the virus in the host resulting in direct injury to vital organs. Following some of the issues found concerning the current 17D vaccine: Live, attenuated 17D, Vaccine •Precautions : Age >60 years •Contraindications in :

–Age <6 months –Immune deficiency, immune suppression –Thymic disease/thymectomy –Egg allergy –Pregnancy

•Possible future contraindications –Autoimmune disorders –Breast-feeding

Live, attenuated YF 17D –Serious Adverse Events •Yellow fever vaccine associated neurotropic adverse events (YEL-AND)

–Previously infants, now seen in adults –Reporting rate as high as 0.8 per 100,000 –CFR 1-2%

•Yellow fever vaccine associated viscerotropic adverse events (YEL-AVD) –Similar to wild-type YF disease –Reporting rate 0.4 per 100,000 (Peru 2007, 7.9/100,000) –CFR 64%

•Anaphylaxis (egg, gelatin allergy) –Incidence 1.8 per 100,000

Based on the above mentioned the development of a potentially safer â-propiolactone-inactivated whole virion YF vaccine (XRX-001) that will be targeted for the population that 17D vaccine cannot cover safely, this vaccine has shown to be highly immunogenic in mice, hamsters, and monkeys (Vaccine 2010;28:3827-33840). The indication for a new Safer Vaccine •Prevention of yellow fever in travelers, military, (endemic zone populations) •Safer vaccine than current live 17D

–No serious AEs related to replicating virus (neurotropic, viscerotropic adverse events) –No egg or other animal derived allergens –No age restriction (indicated for infants, elderly)

–Not contraindicated for immune suppressed, thymic disease, pregnant, lactating patients

The Formulation of the new vaccine is :

•Whole virion vaccine, BPL-inactivated •Liquid, adsorbed (0.2% aluminum hydroxide) •Proprietary excipient formulation

–GRAS substances •Single dose vial (0.65mL) •Shipment/storage 2-8 o C

•Starting material: YF-VAX®, Lot # NDC49281-915-05, Sanofi Pasteur, Swiftwater PA. •The virus was adapted for increased replication in Vero cells by 10 serial virus passages at terminal dilution. •At Virus Passage 10, a single plaque was picked and passed in fluid culture to produce a mini-seed stock at Virus Passage 11. •This virus showed a 3-7 fold increased replication capacity in Vero cells compared to the YF 17D at Virus Passage 1. •The Virus Passage 11 virus stock was used for RNA extraction and the RNA used to produce cGMP grade virus seeds •Full genome sequencing showed that P11 contained a single mutation in the E protein associated with increased replication .

Immunization of hamsters with two doses of the XRX-001 stimulated YF virus neutralizing antibody titers that were significantly higher than a single dose of YF-VAX ®vaccine.

Hamsters given a single dose or two doses of XRX-001 or a single dose of YF-VAX ®vaccine were fully protected against hepatitis, viremia, weight loss and death when challenged with the virulent Jimenez strain of YF virus. To determinate protective efficacy of neutralizing antibodies stimulated by the inactivated XRX-001 YF virus vaccine, graded doses of serum from hamsters immunized with the inactivated vaccine or live attenuated YF-VAX ®were given to hamsters 24 hours prior to virulent YF virus challenge. Passive antibody neutralizing antibody (PPRNT 50) titers in sera of treated hamsters were determined 4 hours before virus challenge and at 4 and 21 days after virus challenge.

Neutralizing antibodies mediated protection from YF virus infection and related disease. Animals that had PRNT 50 antibody titers of ¡Ý 20 four hours before virus challenge were protected from disease as evidenced by lower viremia, liver enzyme levels and infection associated illness (weight change). Passive PRNT 50 antibody at a titer ¡Ý40 produced sterilizing immunity in most of the hamsters, shown byt he absence of an immune response to the challenge YF virus. Immunization with the XRX-001 vaccine stimulated YF neutralizing antibodies equally effective (based on dose response) as antibodies stimulated by the live attenuated YF-VAX ® vaccine. Conclusions •XRX-001 YF vaccine is highly immunogenic, eliciting neutralizing antibodies (nAb) in 100% of animals immunized. •Two inoculations of ~4.4 mcg of vaccine are required for effective immunization. •Hamsters immunized with XRX-001 developed nAb that protected them from disease after virulent YF virus challenge. •Passive transfer of XRX-001 hyperimmune and YF-VAX® serum to hamsters protected them against YF disease following virulent YF virus challenge. •Minimum serum nAb titers of ¡Ý40 before challenge protected hamsters from YF disease when infected with virulent YF virus. •Some animals treated with low doses of nAb were protected from YF disease w/o detectable YF virus nAb in the serum. •Antibodies to XRX-001 and YF-VAX® were equally effective in protecting against YF virus disease in the hamster model. Recombinant live attenuated Dengue vaccine based on YF 17D virus Jean Lang, MD, Ph.D, Associate Vice-President, R&D, Dengue Vaccine Program Head Sanofi Pasteur, Marcy l’Etoile, France (Lecture restricted) The development of the Dengue tetravalent (TV) recombinant YF17D-based chimeric vaccine has been continuously driven by a benefit/ risk evaluation and mitigation strategy. Earlier Pre-clinical studies have demonstrated that the sanofi pasteur Dengue vaccine is genetically and phenotypically stable, non-hepatotropic, less neurovirulent than YF17D and does not infect mosquitoes by the oral route. In vitro and in vivo preclinical studies also showed that the TV Dengue vaccine induced controlled stimulation in human dendritic cells, and a significant immunogenicity in monkeys. Results of Phase II trials in the USA, the Philippines and Mexico showed that the majority of adverse events were mild to moderate and transient in nature. Viraemia was transient and low, and was not increased after initial Dengue TV administration, even in the case of incomplete responses. PRNT50 seroconversion ranged between 80 and 100% for all four serotypes in subjects injected with 2 to 3 doses of TV Dengue vaccine.Responses were also monitored at the cellular level in humans: Th1 and CD8 responses were induced with an IFN-γ/TNF-α ratio favoring IFN-γ. A worldwide clinical development program for Dengue TV is underway including completion of the enrollment of 4000 4-11 years old children in an efficacy trial in Thailand and the planned evaluation of industrial scale clinical lots in Phase III by year end. Assuming continued successful outcomes, initial submissions to national regulatory authorities could now been envisaged within a 5-year period.

A recombinant live attenuated Dengue vaccine based on the delta-30 mutation Stephen Whitehead, NIAID, NIH The goal of the National Institutes of Allergy and Infectious Diseases (NIAID) intramural DENV vaccine program is to produce a minimally reactogenic, highly immunogenic, genetically stable, live attenuated DEN vaccine that is cost-effective and safe for the community. Taking into account that for instance adults on the streets of Vietnam are Dengue immune, the question that rises is : How did they get that way? And Can we safely induce this immunity in children? In endemic areas, Dengue immunity is most likely acquired by sequential infections, the majority of which are asymptomatic. This leads to neutralizing antibody with broad specificity to all 4 DENV serotypes. However, studies show that Sequential administration of monovalent vaccines would not be safe in the endemic setting. Alternatives: -- Administer a single vaccine that is cross-protective against all four serotypes -- Simultaneously administer four vaccines against each Dengue serotype Taking the above into account, the LID / NIAID Dengue Vaccine Goals are : Tetravalent:

a) DEN1 b) DEN2 c) DEN3 d) DEN4

Safe: a) For vaccinee b) For community c) Genetically stable

Immunogenic: Significant response after 1 dose Economical: Available to those most in need homotypic antibody to each serotype Over the past ten years, the NIH has developed numerous live attenuated candidate vaccines against the four individual DENV serotypes using a variety of techniques based on the delta-30 mutation in the 3’ untranslated region of the genome. We have tested 8 monovalent vaccines in 16 separate Phase I clinical trials to identify DEN1, DEN2, DEN3, and DEN4 candidate vaccine viruses that are safe and maintain optimal infectivity and immunogenicity profiles for inclusion in a tetravalent formulation. Each monovalent candidate was well tolerated by volunteers with no volunteer experiencing a Dengue-like illness.

Monovalent Clinical Summary (10 1 - 10 3 pfu doses)

a. No volunteer experienced systemic illness defined as > 2 of the following symptoms lasting > 2 days: headache, malaise, anorexia, myalgia/arthralgia, nausea, eye pain. b. Mean peak titer (log 10 pfu/ml) calculated only for those volunteers with detectable viremia. c. Rash was macular / maculopapular and asymptomatic. d Neutropenia is defined as an ANC of < 1500 /mm 3 . e. ALT < 72 IU/mm 3 is defined as normal for males, < 52 IU/mm 3 is normal for females. Max. ALT elevation 1.7X ULN. f. Unrelated to vaccination. g. Max. temperature was 100.5 o F on single reading, unconfirmed in clinic. Following a single subcutaneous injection of 1,000 PFU, each of the selected candidate vaccines induced seroconversion rates of 80 – 100% to wildtype DENV. Factors such as infectivity, immunogenicity, and reactogenicity were used to determine which vaccine candidates were appropriate to include in a tetravalent formulation. Conclusions for the monovalent vaccine candidates Goal Findings Safe a) For vaccinee Minimal reactogenicity b) For community Poorly transmissible c) Genetically stable .30 mutation maintained Immunogenic 80 –100% seroconversion Economical Produced at >10 7 pfu/mL

Administered at 10 3 pfu ( 1 L = 10 million doses)

A Phase I Clinical trial evaluating 3 different tetravalent admixtures given as a single dose of 1000 PFU/serotype was initiated in 2010 in flavivirus-naïve subjects. Clinical summary of DEN tetravalent vaccine candidates

Conclusions from tetravalent study 1. The tetravalent mixtures are safe. 2. Viremia remained very low. 3. 70 - 90% of subjects had at least a trivalent antibody response. 4. It appears that there is very little cross reactivity between strains. 5. Booster immunization?

• May be needed to achieve a tetravalent response • Will likely increase cross reactivity • Should increase antibody durability

Immediate clinical plans for the development of the tetravalent Dengue virus vaccine 1. Complete initial evaluation of the tetravalent vaccine mixtures, with second vaccination at 6 months. 2. Provide each of the four components to our licensees in Brazil, India, and Vietnam. 3. Determine the number and the timing of doses needed for an optimal antibody response. 4. Evaluate tetravalent vaccine in endemic region : Safety, Age de-escalation

Phase 1 Clinical Trials of DENVax - A Live Attenuat ed, Tetravalent Dengue Vaccine Jorge Osorio1, Aurelia A. Haller1, Joseph Brewoo1, Charalambos Partidos1, Shawn Silengo1, John Arguello1, Tim Powell1, Jill Livengood1, Yuping Ambuel1, Cynthia Thomson2, Joseph Santangelo2, Claire Huang3, Sarah George4, Ivan Velez5, and Dan Stinchcomb 1

1Inviragen, Inc., Fort Collins, CO and Madison, WI, USA; 2Inviragen Singapore, Singapore; 3Centers for Disease Control, Fort Collins, CO, USA; 4St. Louis University, St. Louis, MO, USA; 5Universidad de Antioquia, Medellin, Colombia. (Restricted Lecture) Inviragen’s DENVax tetravalent Dengue vaccine consists of a mixture of the live, attenuated DEN-2PDK-53 virus and three chimeric recombinant viruses. Each chimera contains the attenuating mutations in the DEN-2 PDK-53 non-structural gene backbone while expressing DEN-1, DEN-3 or DEN-4 structural genes. The four DENVax strains have been shown to be attenuated, immunogenic and protective in mice and in nonhuman primate Dengue models. Tetravalent formulations generate neutralizing antibodies to all four Dengue serotypes and preclinical data suggest that intradermal administration of DENVax improves vaccine immunogenicity. Currently, two Phase 1 studies are ongoing to evaluate the safety, tolerability and immunogenicity of DENVax in healthy adults. One study is sponsored by the National Institute of Allergy and Infectious Diseases in the United States (St. Louis, MO) and the second study is sponsored by Inviragen in Rionegro, Colombia, a high altitude area with no Aedes aegypti and no Dengue exposure. The clinical trial design is similar for both studies: two different formulations of DENVax are administered by intradermal or subcutaneous injection in two doses, three months apart. Initially, volunteers receive a low dose of DENVax and safety data collected for 21 days is reviewed prior to administrating the booster or immunizing the next cohort with the high dose formulation of DENVax. These trials will allow direct comparison of subcutaneous and intradermal delivery in placebo controlled trials in two different continents. A third Phase 1 trial is planned to directly compare needle and syringe to needle free delivery of DENVax. Results obtained to date indicate that the low dose of DENVax is well-tolerated in both U.S. and Colombian healthy adult populations; additional data will be summarized as they become available. Development of a Recombinant Subunit Vaccine for De ngue Beth-Ann Coller, Ph.D., Hawaii Biotech, Inc., Merck Research Laboratories

Dengue viruses are a major cause of morbidity and mortality throughout the tropics and subtropics with an estimated 50-100 million cases of Dengue annually. To date no specific vaccine or therapy has been licensed to combat this important disease. To address this unmet medical need, and to avoid issues with viral interference seen with live virus vaccines, Merck is evaluating a recombinant subunit vaccine originally developed by Hawaii Biotech. The tetravalent subunit vaccine is based on the envelope glycoprotein of the Dengue viruses and designed to protect individuals against Dengue virus induced disease. Hawaii Biotech originally developed subunit vaccine based on carboxy truncated envelope protein –80E – recently acquired by Merck. The Drosophila S2 insect cell culture expression system is used to produce Dengue subunits and is a no live virus, potential to achieve balanced tetravalent immunity without interference.

Rationale for Envelope Protein Based Subunit Vaccine Candidate: •Functional regions of Envelope are associated with binding to cell receptors and fusion •Envelope important in protective immune response – key target for neutralizing Ab •Envelope also contributes as cell mediated immune target (CD4 and CD8) •Balance of multivalent immunity may be manipulated by varying antigen doses –no viral interference •No prM protein in purified product –minimize risk of enhancement linked to anti-prM antibodies Formulation development: •HBI has tested Dengue and WN vaccines with variety of adjuvants in preclinical models. •Saponin-based adjuvants and MPL/Alum in combination with DEN antigens induce potent neutralizing responses and protective efficacy in primates •Access to potent adjuvants and regulatory challenges led to focus in short term on alum-based adjuvants •Alum based formulations immunogenic and efficacious if properly formulated and administered. •Focus on alum-based vaccines for initial clinical trial. Other adjuvants may offer significant advantages in long term (durability, dose sparing, magnitude and other).

Preclinical studies conducted in mice, rats, or rabbits have demonstrated the immunogenicity of both monovalent and tetravalent formulations adjuvanted with alum or ISCOMATRIX™ adjuvant and formal toxicology studies have demonstrated acceptable safety of all formulations. Non-Human Primate Studies Monovalent DEN-80E studies : 1 - Rhesus macaques –DEN2-80E + ISCOMATRIX adjuvant 2 - Rhesus macaques –DEN2-80E + GSK adjuvants – Putnak et al., 2005 3 - Rhesus macaques –DEN1-80E + alum adjuvant – stand alone and prime boost with LAV Tetravalent DEN-80E studies 1 - Rhesus macaques –ISCOMATRIX Adjuvant – + DEN2 NS1 pilot study 2 - Rhesus macaques –ISCOMATRIX adjuvant –+/- DEN2 NS1 study –large safety, immunogenicity

Following some of the results concerning neutralizing Ab:

An alum-based monovalent DEN1 formulation and a tetravalent ISCOMATRIX™ adjuvant-based formulation were shown to be immunogenic and protective in non-human primates. A Phase 1 clinical study of monovalent DEN1 recombinant envelope protein adjuvanted with alum has been conducted in healthy volunteers demonstrating the immunogenic potential of the recombinant antigens in human subjects. Conclusions: •Based on preliminary, unaudited, blinded data –monovalent DEN1-80E vaccine adjuvanted with aluminum hydroxide is immunogenic in healthy, adult, flavivirus-naïve volunteers –consistent with preclinical data. •Study included a dose escalation with monitoring by a Safety Committee - no safety signals

•Demonstrates potential of the recombinant proteins. •Plans for testing of tetravalent product under development Implementation of Dengue Vaccination: Role of Publi c-Private Partnerships Scott B Halstead, MD, Director, Supportive Research and Development Program, Pediatric Dengue Vaccine Initiative (PDVI) An enormous challenge confronting the world will be to deploy vaccines to reduce the burden of Dengue disease among an at-risk population of between 2.5 and 3 billion persons. Fortunately, each of the four Dengue viruses is highly immunogenic; natural infection conveys life-long homologous protective immunity and some degree of heterotypic immunity. There is epidemiological evidence that two different Dengue infections protect individuals from severe Dengue disease during a third or fourth infection. Primary Dengue Infections •Approximately 50 million/year. •One-half are in adults. •In adults predominantly overt, causing DF and in older adults with pre-existing conditions, more severe disease. •In children, predominantly mild or silent. The good news is that severe Dengue is rare, only 2 - 4% of secondary Dengue infections are hospitalized. First goal is to protect those persons who have been infected by one Dengue virus, second goal is to protect against primary infection disease and against primary infection “sensitization.” Applying this strategy, sanofipasteur seeks to protect individuals based by administering two or three doses of a chimeric live-attenuated tetravalent Dengue vaccine. Multiple doses of tetravalent Dengue vaccines raise hetorotypic neutralizing antibiodies, it is important to study infection outcome in models that discriminate protection vs. enhancement. New research models - the AG 129 mouse and intrinsic ADE in primary human monocytes/macrophages – now available may permit more accurate assessment of protection afforded by these heterotypic Dengue antibodies than has been possible in the past. i ADE •New method to measure “protective ” role of Dengue antibodies. •i ADE responses differ between primary human monocytes and macrophages. (Boonnak et al J Virol (in press)) (Recent reviews: Halstead SB et al Lancet Infect Dis 10:710, 2010, Ubol et al Clin Vacc Immunol Dec 2010) INTRINSIC ADE –10-20 X increase in virus production per infected cell compared to EXTRINSIC ADE with 3 X increase in number of infected monocytes/macrophages. •Observed with wide taxa of organisms capable of replicating in monocytes / macrophages. •Requires attachment of non-protective “enhancing ”IgG antibodies. •Enhancing immune complex ligates Fc Receptor which sends signal via tail. •May be a central role for IL-10.

Dengue vaccines should be deployed strategically to protect the unvaccinated, especially young infants who may be at risk to severe disease accompanying a first Dengue infection. This will require use of herd immunity. To prepare for this option, herd immunity to the four Dengue viruses should be measured in several at-risk populations. Implementing Dengue Vaccination: Strategy • Use protective synergy from broadened antibody responses to vaccination of partial

Dengue-immunes. • Vaccinate to stop transmission • Vaccinate to protect the unvaccinated or those who cannot easily be reached by vaccines. • Use protective synergy from broadened antibody responses to vaccination of partial

Dengue-immunes. • Vaccinate to stop transmission • Vaccinate to protect the unvaccinated or those who cannot easily be reached by vaccines. This same strategy was used to eradicate smallpox and it worked, this strategy requires herd immunity. What is known about herd immunity in Dengue? •Dengue 1 stopped in Cuba in 1979 at around 50% antibody prevalence (house index ~ 70%). •Dengue 1 stopped in Iquitos, Peru in 1995 at around 80% antibody prevalence (house index > 95%) Mathematical models may then be applied to design vaccine programs that achieve herd immunity efficiently. The Dengue Vaccine Initiative (DVI), successor to PDVI, a new Public-Private Partnership is ready to contribute to these and other components of the strategic implementation of new Dengue vaccines. PDVI is a Product Development Public-Private Partnership founded in 2001 with support from the Rockefeller and Gates Foundations. Program goals for PDVI : 1) to accelerate the development, evaluation and introduction of Dengue vaccines; 2) to make safe and effective Dengue vaccines available to reduce disease incidence in populations at risk in Dengue endemic countries.

Annex I – Meeting’s Program