Dengue is an important human viral disease transmitted by insects.Although nearly half the world's population is at risk for infectionand as many as 100 million cases occur annually,1 we have noantiviral drugs to treat it and no vaccines to prevent it. Aclosely related but much more lethal mosquito-borne virus, yellowfever, used to be one of the great scourges among humans. Althoughyellow fever is now largely controlled by vaccination, manyregions are susceptible to a reemergence if the disease is introducedby travelers, and substantial recent problems with vaccine safetywill no doubt change vaccination policy.
Both dengue and yellow fever are single-stranded RNA virusesin the family Flaviviridae, which includes West Nile virus andapproximately 50 others. Substantial progress has been madein understanding the mechanism of the entry of these virusesinto cells, the atomic structure of the viral envelope (seefigure), the interactions between the molecular determinantsand the host antibody, and the mechanism underlying the neutralizationof the virus by antibodies.2 The unraveling of virus–celland virus–antibody interactions at the molecular levelmay lead to the development of antiviral drugs, improved vaccines,and tests for protective and pathological antibodies.
Structure of the Dengue Virus on Cryoelectron Microscopy.
Courtesy of Richard Kuhn, Purdue University.
Dengue and yellow fever are endemic to and epidemic in tropicalregions (see map). Both are zoonoses maintained in nature bytransmission to humans from monkeys or mosquitoes that breedin tree holes. Infected humans have high blood levels of virusand can therefore infect vector mosquitos. After an incubationperiod of about 10 days, during which the virus replicates inthe salivary-gland tissues of blood-feeding aedes mosquitoes,they can transmit the virus to another person. Throughout thetropics, the principal vector for endemic and epidemic spread,Aedes aegypti, has adapted to living among humans in domesticenvironments. Increasing human population density, urbanization,poor sanitation (creating breeding sites for larval mosquitoes),reinfestation (in the 1970s) of South America by A. aegyptiafter a successful eradication campaign, and the movement ofinfected persons by airplanes have contributed to a substantialincrease in dengue incidence during the past 50 years.2A. aegyptimosquitoes are prevalent in the southern United States, whichis therefore receptive to the introduction and spread of bothdengue and yellow fever (see map).
Distribution of Dengue, Yellow Fever, and Their Principal Vector, the Aedes aegypti Mosquito.
Areas infested with A. aegypti are receptive to the introduction (by air travelers with viremia) and epidemic transmission of the dengue and yellow fever viruses. Yellow fever has never occurred in Asia — possibly because immunity to dengue provides a barrier to interhuman transmission by mosquitoes and because Asian strains of A. aegypti are less efficient vectors than strains from Africa and Latin America — but spread to Asia is an important future threat.
In its classic form, dengue is an acute illness characterizedby fever, headache, muscle and joint pain, and rash. There arefour serotypes, and neutralizing antibodies are serotype-specific,so in regions where multiple serotypes cocirculate, people mayhave sequential infections. Immunity against a specific serotypeis lifelong, but previous infection with one serotype is a riskfactor for a more severe form of dengue — dengue hemorrhagicfever — upon subsequent infection with another serotype.Dengue hemorrhagic fever is characterized by the capillary-leaksyndrome, thrombocytopenia, hemorrhage, hypotension, and shock.Its incidence has increased dramatically during the past severaldecades, as multiple dengue serotypes introduced into new environmentsby air travelers with viremia have become endemic. Approximately500,000 cases occur annually, with a case fatality rate rangingfrom 1 to 3% to as high as 10 to 20%, depending on the sophisticationof the available fluid management and intensive care.
Since we lack an animal model of dengue hemorrhagic fever, ourknowledge of pathogenic mechanisms relies on evidence from invitro studies and patients. Differences in virulence among dengue-virusstrains as well as host factors — principally antibody-mediatedenhancement of dengue replication3 — contribute to diseaseexpression. Central to the immune-enhancement mechanism in denguehemorrhagic fever is the usurping of Fc receptors on maturedendritic cells and macrophages as a means of entry for complexesof dengue virus and subneutralizing antibody. These complexesform in the presence of cross-reactive antibody induced by aprevious infection with a heterologous dengue serotype or, ininfants with maternally derived IgG, when antibody levels dropbelow neutralizing levels. In such instances, the viral loadis increased by means of the infection of an increased proportionof Fc-receptor–bearing cells and an increased level ofvirus in each cell. The T-cell activation and clearance of infectedcells by killer cells and cross-reactive cytotoxic T cells thenelicits a proinflammatory "cytokine storm" that causes endothelialdamage and capillary leakage.
These immunopathological mechanisms create a conundrum for vaccinedevelopers: since sequential infection and heterotypic antibodiescause dengue hemorrhagic fever, a successful vaccine must simultaneouslygenerate long-lasting protective immunity against all four dengueserotypes.4 This vexing problem is the reason that no denguevaccine has yet been approved for use, despite considerableefforts (see table) and substantial funding from the Bill andMelinda Gates Foundation for the Pediatric Dengue Vaccine Initiative.Two live, attenuated vaccines are in phase 2 clinical development.One, developed by the Walter Reed Army Institute of Researchand GlaxoSmithKline, consists of viruses empirically attenuatedby means of serial passage in cell culture. The second was rationallydesigned by Acambis through the genetic engineering of the envelopegenes of dengue (which contain the epitopes for neutralizingantibodies) into a clone of the licensed yellow fever 17D vaccine.The resulting live, attenuated dengue–yellow fever chimericviruses elicit antibodies only to dengue. This vaccine was licensedto Sanofi Pasteur and will soon enter phase 3 trials.
Both vaccines contain mixtures of four dengue serotypes designedto induce long-lasting neutralizing antibodies. The coadministrationof four live viruses is associated with competition among serotypeswith respect to replication and the ability to stimulate neutralizingantibodies, probably owing to the activation of toll-like receptorsand the induction of innate immunity. These effects have beenassociated with the generation of an incomplete repertoire ofneutralizing antibodies after a single inoculation; therefore,complete immunization may require multiple doses. Moreover,our limited understanding of viral neutralization and immunecorrelates of protection, and the difficulty of distinguishingcross-reactions from the development of type-specific antibodies,create challenges for vaccine development. Ultimately, large,controlled field trials will be needed to demonstrate vaccineeffectiveness, with follow-up lasting for several seasons ofviral transmission to ensure that sensitization to dengue hemorrhagicfever has not occurred.
Yellow fever, for its part, is a fearsome systemic illness characterizedby high levels of virus in the blood, jaundice, midzonal coagulativenecrosis (apoptosis) of the liver, renal failure, myocardialinjury, hemorrhage, and shock — with case fatality ratesas high as 50%. The true incidence of yellow fever is unknownbut is likely to be a few thousand cases per year, with intermittent,large epidemics involving more than 100,000 cases. In contrastto dengue, yellow fever has only one serotype — a factthat simplified the development of a vaccine. A live, attenuated(17D) vaccine was developed in 1936 by means of serial passagein chicken-embryo tissue; it has since been used in more than400 million people and induces long-lasting neutralizing antibodiesin about 99% of those who are vaccinated. Many countries inwhich yellow fever is endemic routinely immunize infants at9 months of age. In the United States, about 250,000 personsper year are vaccinated to prevent infection during travel ormilitary assignments in tropical regions.
S. Burt Wolbach and John L. Todd Crossing Kandong Bolon, a Creek near Somita, Gambia, Where They Were Researching Yellow Fever and Other Tropical Diseases, in February 1911.
Courtesy of Margot Green.
Until 2001, the 17D vaccine was believed to be extremely safe,but that view has been altered by the recognition of a new syndrome:viscerotropic disease associated with the yellow fever vaccine,5an extensive infection of vital organs by a 17D virus that isindistinguishable from wild-type yellow fever disease and hasa 60% case fatality rate. Genetic factors of the host (possiblyin genes involved in interferon responses) and acquired factors(advanced age and thymectomy) appear to underlie susceptibilityto this condition. The overall incidence is about 1 case forevery 200,000 to 400,000 vaccinations, but among persons over60 years of age, the incidence is as high as 1 for every 50,000vaccinations — which makes 17D one of the least safe vaccinesin use. In addition, neurotropic adverse events (mainly encephalitiscaused by invasion of the brain by the 17D virus) have a similarincidence but a much lower case fatality rate (about 6%).
Such risks result in difficult choices, given that the reportedincidence of yellow fever among unvaccinated travelers is lowerthan that of serious vaccine-related adverse events. The problemis that unvaccinated persons traveling to an area of activetransmission of yellow fever are in great danger, but such areasare often epidemiologically silent, since the indigenous populationis immune or surveillance is poor — rendering it impossibleto selectively vaccinate only persons who are at a real riskfor exposure. Until a safer vaccine is developed, travelersand their physicians must exercise judgment on the basis ofgeographic and epidemiologic data. Travel plans should be carefullyassessed to determine whether the traveler will in fact entera region in which yellow fever is endemic (see map). Improvedsurveillance for adverse events must be instituted wherevervaccinations are routinely given. Finally, research is neededon the individual risk factors for vaccine-associated viscerotropicdisease and on the care of patients in whom this potentiallylethal complication develops.
Dr. Monath reports being the former chief scientific officerof Acambis, for which he has served as an expert witness andhas received consulting fees; holding multiple patents for recombinant,chimeric vaccines that use the yellow fever 17D strain as alive vector; and serving on the Board of Counselors of the PediatricDengue Vaccine Initiative and as a nonexecutive director ofXcellerex, which is developing a yellow fever vaccine.
Source Information
Dr. Monath is a partner at the venture-capital firm Kleiner Perkins Caulfield & Byers, Menlo Park, CA, and an adjunct professor at the Harvard School of Public Health, Boston.
An interview with Dr. Monath and a slide show are available at www.nejm.org.
References
Gubler DJ. Dengue and dengue hemorrhagic fever: history and current status. In: Novartis Foundation. New treatment strategies for dengue and other flaviviral diseases. Chichester, United Kingdom: John Wiley, 2006:3-22.
Kuhn RJ, Zhang W, Rossmann MG, et al. Structure of dengue virus: implications for flavivirus organization, maturation, and cell fusion. Cell 2002;108:717-725. [CrossRef][Web of Science][Medline]
Halstead SB. Antibody, macrophages, dengue virus infection, shock, and hemorrhage: a pathogenic cascade. Rev Infect Dis 1989;11:Suppl 4:S830-S839. [Web of Science][Medline]
Wright, P. F., Durbin, A. P., Whitehead, S. S., Ikizler, M. R., Henderson, S., Blaney, J. E., Thumar, B., Ankrah, S., Rock, M. T., McKinney, B. A., Murphy, B. R., Schmidt, A. C.
(2009). Phase 1 Trial of the Dengue Virus Type 4 Vaccine Candidate rDEN4{Delta}30-4995 in Healthy Adult Volunteers. Am J Trop Med Hyg
81: 834-841
[Abstract][Full Text]
Previous
Audio Interview
Slide Presentation
