Vector-Borne Viral Infections
Dengue Fever
- Introduction
- Disease Burden
- Virology
- Vaccines
- Research Priorities
- WHO Media Fact Sheet
- WHO Dengue homepage
- Useful Links
Introduction
Dengue, an usually non-severe albeit debilitating viral fever ("breakbone disease"), is the most prevalent mosquito-borne viral disease in people. Dengue is endemic in most tropical and subtropical countries, many of which are heavily populated as well as a popular destination for tourists
[15]
. The disease is caused by the dengue virus (DV), of which four serologically different serotypes exist (DV-1, DV-2, DV-3 and DV-4), which actually are four viruses almost as genetically different from each other as JE, West Nile and SLE viruses are from one another
[16]
. Infection is believed to provide life-long immunity against reinfection by the same serotype, but not against the other serotypes. An individual could therefore experience a case of dengue-1 fever on one year, followed by a case of dengue-2 fever on the following year. Third infections however are very rare, and fourth infections have never been reported.
The dengue viruses are the only known arboviruses that have fully adapted to humans and do not need an animal reservoir, although dengue has been observed to also circulate in nonhuman primates. They are transmitted from human-to-human by the urban-dwelling Aedes aegypti mosquito, which has adapted to humans, laying their eggs in artificial containers in and around houses and feeding only on humans, and by Ae albopictus (the Asian tiger mosquito) and Ae polynesiensis. These latter mosquitoes feed on birds, reptiles, rats, cows, and humans. DV may be transmitted vertically from the female mosquito to her offspring. Most cases of transmission from human-to-human are through female mosquitoes which previously became infected when feeding on an infected person.
Dengue is primarily an epidemic disease of urban and peri-urban settings. Thus, in 1987, Thailand reported 175 000 cases and 1000 deaths; in 1996 Brazil, 180 000 cases; in 1998, countries in Latin America, the South-East Asian and the Western Pacific regions 1 300 000 cases and 3500 deaths. In 2001, almost 400 000 cases were notified in Brazil, which has known repeated epidemics and an increase in severe cases of dengue in adults since then
[17]
. Several large outbreaks also occurred in 2007 in Singapore, Cambodia, Malaysia, The Philippines and Vietnam, with more than 133 000 clinical cases reported and 850 deaths (182 in Cambodia alone).
Disease Burden
Some 2500 million (2.5 billion) people are estimated to currently be at risk of dengue in over 100 countries across the globe. It is estimated that between 50 - 100 million cases of Dengue fever, 500 000 cases of Dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS) and more than 20 000 deaths from DHF/DSS occur each year
[18]
. Dengue has become one of the most important emerging disease problems among international travelers [15] and the second most frequent cause of hospitalization after malaria among travelers returning from the tropics
[19]
[20]
.
Dengue is a major cause of morbidity and mortality and a leading cause of hospitalization of children in many countries in tropical and subtropical areas of the world. The cost of illness to society is considerable. The past 25 years have seen the emergence and reemergence of epidemic dengue, with more frequent and larger epidemics associated with more severe diseases
[21]
[22]
, probably related to population growth, massive unplanned urbanization, modern transportations and the lack of effective mosquito control. Most primary infections cause a non fatal form of illness, but some individuals, mostly children, can experience a more severe form of the disease. Severe disease is referred to as dengue haemorrhagic fever (DHF), characterized by a transient increase in vascular permeability resulting in plasma leakage from blood vessels into tissues, oedema, especially in the chest and abdomen, as well as thrombocytopenia and bleeding seen as petechiae easily detected by the tourniquet test
[23]
. The patient can eventually go into hypovolaemic shock syndrome (DSS) and organ failure and die if not properly managed
[24]
. Children with severe dengue are particularly susceptible to DSS, the highest mortality being in infants, who show case fatality rates of up to 13% in hospital-admitted patients [5].
The risk of severe disease, which is estimated to occur in 250 000-500 000 patients a year, is associated with secondary infection by a different serotype, and may be due to antibody-dependent enhancement of macrophage infection by non-neutralizing but cross-reactive antibodies [24]. The increasing endemicity and co-circulation of several DV strains is therefore a leading contributor to the observed increased severity of dengue. Differences in strain virulence probably also are important
[25]
. In fact, the occurrence of DHF/DSS seems to actually depend on a combination of heterotypic antibody-dependent enhancement of macrophage infection [24]
[26]
, viral load, strain virulence and host immune response
[27]
as well as to enhancement of dendritic cell infection
[28]
. DV enhancing antibody activity in plasma, as measured by in vitro assays, does not necessarily predict subsequent disease severity in secondary dengue virus infection
[29]
[30]
, as viral factors also contribute to disease severity. Thus, all DV-2 epidemics causing substantial DHF in the American region were associated with a South-East Asian genotype introduced to Cuba from Vietnam in 1981 and those of DV-3 with a virus from India or Sri Lanka [10].
The burden of severe disease remains proportionately much greater in Asian and Pacific countries than in the Americas. In a prospective study on 2114 school children in northern Thailand from 1998 to 2002, dengue accounted for 11% of all febrile illness cases, a burden of 3200 DALYs per million per year, in the same order as hepatitis B
[31]
. The study showed that dengue illness which does not require hospitalization accounted for half or more of DALYs lost to dengue and its cost was very substantive
[32]
. The infecting virus serotype also seems to be an important albeit variable determinant of DALYs lost, emphasizing the different contribution of the different serotypes to the disease burden.
Virology
DVs are members of the genus Flavivirus in the family Flaviviridae, which includes more than 70 related but distinct viruses, many of which are mosquito-borne, such as the yellow fever (YF) virus. Flaviviruses are enveloped, 40-50 nm-diameter viruses with an icosahedral capsid that protects the single-stranded, positive sense RNA genome. DV envelope surface projections are composed of dimers of the viral envelope (E) glycoprotein and of the membrane (M) protein, itself derived by furine-mediated cleavage from a prM precursor. The only other protein constituent in the virion is the capsid (C) protein. The E glycoprotein mediates virion attachment to receptor and fusion of the virus envelope with the target cell membrane and bears the virus neutralization epitopes. On native virions, the elongated three-domain E molecule lies tangentially to the virus envelope in a head-to-tail homodimeric conformation. Upon penetration of the virion into the target cell endosome, E dimers are converted to stable target cell membrane-inserted homotrimers that reorient themselves vertically to promote virus-cell fusion.
The 10.5 kb-long genomic RNA is a monocistronic mRNA which is translated into a precursor polyprotein from which the individual viral proteins eventually derive by cleavage, starting with the C, prM and E proteins followed by nonstructural proteins NS1 to NS5. NS3 is a protease and a helicase, whereas NS5 is the RNA polymerase in charge of viral RNA replication. In addition to the E glycoprotein, only one other viral protein, NS1, has been associated with a role in protective immunity. This glycoprotein is not present on the virion, but is found on the surface of infected cells. Immunization with NS1 has been shown to elicit protective immunity in animal models.
Vaccines
As there is no cross-protection between the 4 DV serotypes and because of fear of immune enhancement by heterotypic DV antibodies, only a tetravalent vaccine will be acceptable. Progress in DV vaccine development has been relatively slow, mainly because DV grows poorly in cell culture and because there is no reliable animal model for DHF. Also, tetravalent DV vaccines have often generated disappointing immunogenicity results as compared to monovalent vaccines, due to a phenomenon of interference between the 4 strains, the identity of the dominant serotype(s) depending on the nature and composition of the vaccine. The application of infectious clones technology to dengue vaccine development has greatly stimulated the development of candidate vaccines and the current pipeline of dengue vaccines is diverse and overall promising
[33]
[34]
[35]
[36]
.
Live attenuated vaccines.
The initially favored strategy has been to attenuate DV strains by repeated passage of wt DV in cell culture in order to prepare a vaccine based on live attenuated virus strains. This was undertaken at Mahidol University in Bangkok, Thailand, using primary dog kidney cells, African green monkey cells and/or fetal rhesus monkey cell cultures. Live attenuated vaccines face the difficulty to define a correct balance between insufficient attenuation and over-attenuation of the candidate vaccine strains, the lack of correlation between in vitro markers of attenuation such as small plaque phenotype or thermosensitivity and in vivo attenuation, and the phenomenon of immunological interference between the four DV serotypes
[37]
[38]
. The development of a tetravalent live attenuated vaccine was eventually taken over by Sanofi Pasteur
[39]
before being put on hold.
The Walter Reed US Army Institute of Research (WRAIR) succeeded in developing a tetravalent live attenuated vaccine by serial passage of all four DV strains in dog kidney cells, and tested various tetravalent formulations in adult volunteers, then in Thai children.
[40]
[41]
[42]
. The vaccine was then licensed to GSK, which continued clinical trials
[43]
and is presently going through Phase II trials with the live attenuated tetravalent vaccine candidate.
The US National Institutes of Health (NIH) also have developed attenuated DV strains, using reverse genetics to create an attenuating 30-nucleotide deletion (?30) in the 3' untranslated region of the genome of the four DV strains
[44]
[45]
. Two of the resulting virus strains, DV1-?30 and DV4-?30, were found to be attenuated in rhesus monkeys and safe and highly immunogenic at a dose of 103 PFU/vaccinee in Phase I/II clinical trials in human volunteers
[46]
[47]
. However, this strategy did not yield suitable candidates for serotypes 2 and 3, which had to be generated by replacing the sequence coding for the M and E structural proteins in the attenuated DV4-?30 genome by the corresponding sequences from DV2 or DV3, thus yielding intertypic chimeric viruses DV2/4-?30 and DV3/4-?30, respectively
[48]
. The DV2/4-?30 virus strain was tested in humans and appeared safe and strongly immunogenic at the dose of 103 PFU/dose
[49]
. The replication of these chimeric viruses was also attenuated for Ae aegypti, indicating that they would likely manifest decreased transmissibility by mosquitoes
[50]
. The DV3/4-?30 virus was evaluated in clinical trials
[51]
but immunogenicity appeared to be weak. The 4 attenuated virus strains should eventually be combined together and tested as a tetravalent live attenuated candidate vaccine
[52]
.
Live chimeric virus vaccines
A homotypic chimeric virus approach also has been applied by the US Centers for Disease Control and Prevention to engineer DV2 chimeras by inserting the structural protein genes from DV1, DV3 and DV4 into an attenuated PDK53 DV2 genome that had been attenuated by replacing a portion of the DV terminal 3' stem and loop structure with that of West Nile virus
[53]
[54]
. The clinical development of these DV2-based chimeras is being carried yR Inviragen, in collaboration with the US CDC and Shantha Biotechnic.
Another approach, the engineering of heterotypic chimeric viruses, was initiated in the 1990s by Acambis, using the YFV 17D vaccine strain as the genetic background
[55]
and replacing the M and E structural protein-coding sequences in the YFV genome with those from either the four DV serotypes
[56]
, or from JEV
[57]
, or from WNV
[58]
[59]
. The DEN-YF chimeras were developed by Acambis and licensed to Sanofi Pasteur. When injected to monkeys for safety and protective efficacy tests, the monkeys developed a brief viremia followed by a DV neutralizing antibody response and were protected against challenge with DV
[60]
.
The chimeric viruses were tested and found to be safe and immunogenic in humans. A tetravalent combination of the four DEN-YF chimeras, ChimeriVax-Den vaccine [56]
[61]
, was shown to induce a transient and low grade viremia in nonhuman primate and in human volunteers, followed by a solid immune response against the four serotypes with some strains showing dominant immunogenicity. A dose adjustment for the DV2 chimera resulted in a more balanced response. The chimeric viruses also were found to replicate and disseminate poorly in the body of mosquitoes, indicating that the risk of infection and transmission by mosquitoes in nature would be minimal. A Phase II trial is taking place in the USA and Latin America and a Phase IIb pediatric trial has been launched by Sanofi Pasteur in early 2009 in Thailand.
Live recombinant, DNA and subunit vaccines
DV genes were inserted at the Naval Medical Research Center into a new, non-replicative adenovirus vector (Ad5) to engineer double recombinants expressing the prM and E sequences from both DV1 and DV2 and DV3 and DV4, respectively
[62]
. The pair of recombinants was tested in mice and shown to induce neutralizing antibodies to the four DV serotypes. It also was tested in macaques, in which it induced significant protection against challenge with all four DV serotypes
[63]
.
The Naval Medical Research Center is pursuing a DNA-based vaccine approach, using a Biojector device for immunization. Evaluation of a Phase I study is ongoing.
Finally, a subunit dengue vaccine has been developed by Hawaii Biotech Inc using the truncated amino-terminal 80% of the E glycoprotein from each serotype plus the entire NS1 protein from DV2
[64]
formulated in a proprietary adjuvant. The candidate vaccine elicited robust immune responses in nonhuman primates and clinical trials are envisaged in the near future. A novel dengue subunit vaccine candidate was developed using a consensus dengue virus envelope protein domain III (cEDIII). BALB/c mice immunized with the recombinant cEDIII in the presence of aluminium phosphate developed long-lasting neutralizing antibodies against all 4 serotypes of dengue virus
[65]
. Several groups are developing subunit vaccines based on domain III of the DV E protein, a strategy that is aimed at reducing the induction of crossreactive antibodies.
Another approach has used inactivated virus. A purified inactivated DV2 candidate vaccine has been tested and shown to elicit a good level of neutralizing antibody and protection in nonhuman primates
[66]
.
The question of whether any of the candidate dengue fever vaccines will be ready by 2012 still is an open question, although it is the declared objective of both GSK and Sanofi Pasteur, both of which have presented timelines compatible with licensure by 2012.