Dengue Vaccines

History of Disease

Early in the 20th century, investigators studying dengue fever outbreaks successfully transmitted infection from person to person through the bite of Aedes aegypti mosquitoes. The epidemiology and clinical descriptions of this disease made it possible to identify outbreaks of dengue fever from historical records. A well-characterized outbreak of “break-bone fever” occurred in late summer of 1780 on the Philadelphia waterfront. Benjamin Rush fully described the classical signs and symptoms of dengue fever. Importantly, he recognized the postillness depression, unique to dengue. Remarkably, in 1779, David Bylon “stads cirurgyn” of Batavia (Jakarta) described another epidemic febrile exanthema. This had an acute onset with joint involvement leading to persisting pain and swelling. ^, The febrile period was shorter than that of break-bone fever. These features identify the 1779 outbreak as chikungunya. For many years the Java outbreak was cited as the classic description of dengue fever.

A dengue zoonosis was recognized in Philippine monkeys in 1931. This was formally identified by Rudnick who recovered dengue viruses (DENVs)-1, -2, and -4 in Malaysia from wild-caught monkeys. ^, It is likely that the four DENVs evolved from an ancestral virus that circulated in nonhuman primates on mainland Asia that was subsequently fragmented into the Indonesian archipelago and the Philippines by the rise of ocean levels during the current interglacial period. ^, This separation permitted the evolution of viruses whose envelope proteins differed sufficiently to escape cross-neutralization—hence the four virus types. One of these, DENV-2, was introduced into Africa where it re-established a diverse zoonosis in nonhuman primates.

Aedes aegypti , a vector mosquito of enormous consequence to human health, may have been introduced into Africa from Indian Ocean islands 85,000 years ago. There it evolved to two subspecies, Ae aegypti formosus and Ae aegypti aegypti , respectively, sylvatic and domestic vectors. The spread during historical times of A. aegypti aegypti from Africa throughout the world created an ecologic niche that supported the urban transmission of yellow fever, Zika, chikungunya, and the DENVs. ^, Genetic studies of Aedes aegypti populations suggest that the species was imported during the colonial era from Africa to the Western Hemisphere and from there to Asia. ^, Analysis of the few existing zoonotic DENV strains suggests that the four sylvatic virus types were independently imported into the urban cycle within the past 1000 years. From ecological evidence this likely occurred in tropical Asia following the introduction of Ae aegypti .

Prior to World War II, localized dengue outbreaks were recognized by their clinical and epidemiological features. Some were dramatic. A classical outbreak of dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS) was reported in northern Queensland, Australia in 1897, and another in Athens in 1928, with one million cases and a thousand deaths. ^, A pan-Pacific dengue fever outbreak of DENV-1 occurred during World War II. With infections supported by postwar population growth, urbanization, and air travel, epidemic dengue hemorrhagic fever/dengue shock syndrome (DHF/DSS) emerged into recognition in the Philippines in 1956 and Thailand in 1958. DENVs inexorably spread globally. DENV-1 of SE Asian origin was introduced into the Western Hemisphere in 1977, and was quickly followed by an outbreak of DHF/DSS accompanying the introduction of DENV 2 into Cuba in 1981. Since then, the other two DENVs have been imported from SE Asia establishing hyper-endemicity in the American tropics.

Why the Disease Is Important

Since World War II, the four DENVs have spread throughout tropical and many subtropical areas to achieve pandemic status with recent annual aggregate estimates of 58–96 million overt disease occurrences annually ( Fig. 19.1 ). Of these, about 10.5 million were hospitalized. The most recent (2019) global burden estimates (95% uncertainty interval) are 56.9 (37.1–101) million cases, 36,199 (9200–44,500) deaths, and 2.38 (0.83–3.27) disability-adjusted life-years (DALYs). Of these, perhaps 1–2 million persons are hospitalized with severe illnesses, including DHF and DSS, and 0.1–5% die. Estimated costs are illustrated by a 2018 study in Indonesia with each nonfatal case in hospitals costing $ 316, each ambulatory case $22, each nonmedically attended case $7.50, with an overall average of $50 per case of dengue. Costs of nonfatal episodes were borne by the patient’s household (37%), social contributors (relatives and friends, 20%), national health insurance (25%), and other sources (government, charity, and private insurance, 18%). Including fatal cases, the average cost per episode was $90.41. Indonesia had an estimated 7.535 million dengue episodes in 2017, with a national aggregate cost of $681.26 million. The study found this cost was 73% higher than previously estimated. Dengue exacts a high socioeconomic toll on inhabitants of 128 tropical countries.

An eight-country survey found costs for both outpatient and inpatient dengue illnesses to be substantial, ranging from 8 to 56 days of gross domestic product (GDP) per capita. The global annual economic burden of dengue is $8.9 (3.7–19.7) billion. This cost exceeds that of other major infectious diseases with comparable data such as cholera, rotavirus, gastroenteritis, Chagas, and canine rabies. While underestimation of passive surveillance system creates substantial uncertainty, the global burden of dengue is substantial. ^, The worldwide DALYs due to dengue grew from 822,800 in 1990 to 2,920,000 in 2017, contradicting the global shift in disease burden from communicable to noncommunicable diseases. ^, Dengue exacts a burden on productivity. A 2016 study in India estimated 53,210,706 cases and 22,527 deaths, resulting in $5.71 billion in economic costs. In a recent review of 20 studies, productivity costs varied between U.S.$6.7–1445.9 and U.S.$3.8–1332 for hospitalized and outpatient nonfatal episodes, respectively. The productivity cost associated with fatal dengue episodes varied between U.S.$12,035 and $1,453,237, variations reflecting the wide range of countries and socioeconomic populations affected. Finally, dengue symptoms often persist resulting in chronic disability and consequently exact a severe impact on quality of life.

In addition, visitors to dengue-endemic countries are infected. In 2007, 141 million persons traveled to dengue-affected areas in Asia. The risk of travel-acquired dengue depends on destination, season, duration of travel and activities during travel. Seroconversion rates reported in returning travelers vary between < 1% and > 20%.

Clinical Description

Infection with DENV-1 (five genotypes), DENV-2 (six genotypes), DENV-3 (five genotypes), or DENV-4 (four genotypes) may be asymptomatic, result in a mild febrile illness, dengue fever (DF), or severe illness including DHF/DSS. ^, Dengue fever is a self-limited febrile illness, with an average incubation period of 5 days, characterized by acute onset of fever and, variously, headache, myalgia/arthralgia, inappetence, prostration, gastrointestinal symptoms, and rash. The continuum from DF to severe dengue is differentiated by the onset of the dengue vascular permeability syndrome (DVPS), which may rapidly appear around the time of defervescence, when virus-infected cells are being immunologically eliminated ( Table 19.1 ). Dengue vascular permeability syndrome, the central component of DHF/DSS, is well described. Hemorrhage, mild or severe, occurs with both DF and DHF/DSS. Vascular permeability is of short duration that spontaneously and rapidly self-heals. Proper management requires careful monitoring of clinical status obtaining multiple measurements of hematocrit and platelets. Evidence from recent in vitro studies and a relevant animal model shows that nonstructural protein 1 (NS1) released from all DENV-infected cells circulates at high concentrations and may activate toll-receptor 4 (TLR4), directly damage endothelial cells, contribute to the damage of liver cells, activate complement, and stimulate the release of cytokines. All of these are the observed features of DVPS.

Complications

The most important but avoidable complication of DVPS is the failure to promptly resuscitate with adequate fluid and, when required, protein solution. The next most important complication is the continued administration of fluid after the vascular leak has healed. This may rapidly result in hypervolemia leading to heart failure and pulmonary edema. Each of these complications contribute to fatal outcomes in DENV infections. For treatment guidelines, clinicians should consult the WHO Technical Guide for the Diagnosis, Treatment and Prevention of Dengue.

Virology

DENV-1–4 are single stranded positive RNA members of the Flavivirus group in the virus family Flaviviridae. The flavivirus genome consists of approximately 11,000 base pairs, which translate into three structural and seven nonstructural proteins (see “Chimeric virus vaccines,” later). The four distinct DENVs all evolved from a common sylvatic ancestor, with separate introductions into the urban cycle of transmission—humans to Aedes mosquitoes to humans which has limited virus evolution compared to other RNA viruses needing to accommodate two hosts. ^, The basic virology of the four DENVs is similar to that of yellow fever (see Chapter 64) and Japanese encephalitis (see Chapter 35). The structural arrangement of DENV surface proteins controls how antibodies neutralize viruses. Dengue virus is an enveloped, positive-strand RNA spherical viral particle with a diameter of approximately 500 A. RNA, when complexed with capsid proteins, forms the inside of the virus particle. It is surrounded by a lipid bilayer membrane. The viral envelope contains two integral membrane proteins designated envelope (E) and premembrane (prM). E protein binds to cellular receptors and mediates fusion of viral and cellular membranes during viral entry into cells and is the main target of neutralizing antibodies. Anchored on the outside of the membrane are the M and E proteins. The M protein is a small protein that is hidden under the E protein but that projects to the surface in its immature state. Each virus particle has 180 monomers of E that are organized into 90 tightly packed dimers that lie flat on the surface of the viral membrane.

Each E protein monomer has three domains: DI, DII, and DIII. The hinge that connects DI to DII is used to flex DII in the low pH environment of the endosome, leading to the exposure of its fusion loop. ^, DIII is thought to be involved in receptor binding. Individual subunits of E protein consist of three β-barrel domains designated domains I (EDI), II (EDII), and III (EDIII), with the native protein forming a head-to-tail homodimer. The hydrophobic viral fusion peptide is located at the tip of domain II and is shielded by domain III of the adjacent subunit. Domain III appears to be responsible for binding to cellular receptors as several mutations that affect receptor binding are located in this domain.

Pathogenesis as It Relates to Prevention

Classical DHF/DSS accompanies first DENV infections in 5–11-month-old infants born to dengue-immune mothers. Altogether 90+% of cases accompany second (occasionally third) heterotypic DENV infections in individuals older than 1 year. ^{, ,} These two epidemiologic settings have in common a single immune factor—IgG ₁ DENV antibodies. Numerous studies show that at subneutralizing concentrations DENV antibodies enhance DENV infections in Fc-receptor-bearing cells, a process referred to as antibody-dependent enhancement (ADE). ^, In humans, peak viremia titers measured early in secondary DENV infections were positively correlated with the subsequent occurrence of DHF/DSS. DHF/DSS is rare, seen in only 2–4% of second DENV infections.

The discovery that DVPS is a toxicosis, mediated by NS1 a protein with endotoxin-like properties, has fundamentally changed the dengue pathogenesis landscape. ^, For 60 years, competing camps promoted many alternative hypotheses of the pathogenesis of severe dengue: intrinsic viral virulence, autoimmunity, direct DENV infection of endothelial cells, “original antigenic sin” and second-infection pathogenic immune responses (“cytokine storm”). These hypotheses centered on explaining DHF/DSS due to second heterotypic DENV infections but, ignored or made no attempt to explain DVPS during a first infection in infants born to dengue-immune mothers. NS1 toxicosis provides a unitary pathogenesis hypothesis. Antibodies, actively or passively acquired form immune-complexes with circulating virus, infect Fc-R-bearing cells leading to increased production of DENV NS1.

Comprehensive pathology studies on 13 fatal DSS cases found splenic and lymph node macrophages were productively infected. Endothelial cells were not infected. There was evidence of nonreplicative DENV infection of hepatocytes and widespread staining of parenchymal cells by DENV NS 1 and complement subunits. ^, While there is a correlation between DVPS and peak NS1 blood levels measured soon after onset of fever, when measured during the period of severe vascular permeability NS1 blood concentrations may be rather low. During the late acute stage, anamnestic antibodies nearly obliterate viremias and RNAemias. Formal attempts to correlate NS1 blood levels with vascular permeability are probably only possible during primary DENV infections. In the literature almost all studies on NS1 blood levels and vascular permeability have been performed during secondary DENV infections. The few studies on infants with severe primary dengue found relatively low viremias compared with peak values in secondary infections. Consecutive daily measurements were not made. Damage to endothelial cells may be a threshold phenomenon and related to duration of exposure to NS1. It is also possible that circulating NS1 antibody complexes, not detectable by immunoassays, may retain endothelial toxicity, in vivo. Progress in understanding dengue pathogenesis is hindered by the absence of an aggressive research agenda.

Specific antiviral medications are not available, but supportive intensive care and careful fluid management is life-saving. The single available preventive measure, comprehensive mosquito control, has not been effective. Increases in ambient temperatures increase vectoral capacity. ^,

Epidemiology

Infection of humans follows the bite of an infected Aedes mosquito (primarily Aedes aegypti , and to a lesser extent Aedes albopictus or Aedes polynesiensis ). Ae aegypti eggs are laid individually by the female mosquito on the walls of both artificial and natural water containers. Eggs resist desiccation for weeks to months and hatch when submerged in water. Larvae and pupae prefer clean water in many different types of artificial containers: water storage containers (tanks, jars, cisterns, pots), ornamental containers (flower holders, ant traps, shrine objects), and discarded items (rubber tires, plastic containers, bottles). The species also occasionally uses natural larval habitats, such a bromeliads and tree holes. Containers located outside dwellings can be filled with rainwater and are productive during the rainy season. Indoor or sheltered containers may produce pupae throughout the year. When extensive use is made of outdoor larval habitats the prevalence of larvae and adults are often subject to marked seasonal variation. The duration of the larval stages is 7–9 days at 25°C and that of the pupal stage is 2–3 days at the same temperature.

The ecology of adult Ae aegypti in a domestic urban environment is characterized by strong anthropophilia and diurnal feeding usually with two peaks, one in mid-morning and another in late afternoon. It seems likely that most females can feed twice or even three times during a single gonotrophic cycle. The preferred resting sites of adults are sheltered dark spaces inside houses. The average life span for females is 8–15 days and that for males about 3–6 days. Geographic dispersal of adults is usually limited, averaging about 30–50 m a day for females, which means that a female rarely visits more than two or three houses during her lifetime. In some locales, such as Puerto Rico, female flight distance may be related to the availability of oviposition sites (breeding sites) and so might be much longer. Passive dispersal of eggs and larvae is common, including trains, boats, and aircraft. Because of the weak spontaneous dispersal of the species and its easy passive dispersal, the International Sanitary Regulations require that the area within 400 m of international ports and airports be kept free of Ae aegypti .

Two main factors regulate Ae aegypti populations: climate and the availability of breeding sites. Population changes may or may not correlate with weather. Daily, seasonal, and interannual variability in temperature, atmospheric moisture, and rainfall all influence mosquitoes in a variety of ways. From an epidemiological perspective, the increased number of interrupted feeding per replete feeds that is the consequence of 2° or 3° warmer temperatures is equivalent to a doubling of the density of Ae aegypti . Under circumstances where a majority of breeding sites are indoors, warm temperature and high moisture contribute to increased adult survival together with the effect of warmer temperatures on shortening the extrinsic incubation period contribute to the occurrence of dengue outbreaks during the hot, rainy season of Southeast Asia. In countries with seasonal variation, DENV transmission is highest during warm or rainy seasons. In the true tropics, such as Singapore, DENV transmission is year-round, but, impacted by seasonal variations in nearby countries whose populations either work in or may visit Singapore.

Immunity in Dengue

Immune responses of humans to DENV infections are both simple and complex. A first DENV infection produces lifelong protection to homotypic virus reinfection. Interactive immunity by the different DENV infections is complex and poorly understood. A first infection with any DENV conveys early but waning cross-protection against a second heterotypic DENV infection. For a period of roughly 2 years this results in the important benefit of protecting against or reducing the severity of secondary infection DVPS. Unfortunately, as the gap between first and second DENV infections widens, the rate and severity of secondary DENV infection DVPS rises. It is this attribute of dengue immunity that endangers the outcomes of imperfect dengue vaccines. Sequential infections with any two DENVs produce strong and durable protection against severe disease accompanying a third DENV infection. ^, We neither have reliable markers of protection nor a comprehensive understanding of the mechanisms of dengue immunity, protective or immunopathogenic. A belief, long held, unsubstantiated by evidence, is that antibodies are the sole or a major mechanism of life-long protective immunity. There is no question that in the right concentrations and composition, antibodies are protective. For example, antibodies passively transferred from women with two or more previous DENV infections protect their infants from DENV infection and disease. But, in a matter of months, these same antibodies degrade to efficiently enhance dengue disease. Insights into mechanisms of protection were improved when it was recognized that antibody interactions with three dimensional virion structures were crucial to neutralization of homotypic DENVs.

All candidate tetravalent DENV vaccines rely on the tetravalent protection demonstrated in monkeys given four live DENVs as a single dose. Protection was evidenced by tetravalent neutralizing antibodies plus solid protection to challenge with each of the four wild-type DENVs. “Solid protection” was defined as the absence of viremia and anamnestic hemagglutination-inhibition (HI) or neutralizing antibody responses.

Diagnosis

Specific etiology is established by detecting the agent or viral protein in blood or by documenting a specific antibody response after infection. Etiology during the acute phase can be determined by isolating virus in any of several host systems or detecting circulating DENV RNA. The standard method is the detection of NS1 by ELISA or immunoassay while it circulates in the blood in the acute and early convalescent phases. ^, In addition, there are many commercial tests on the market that identify antibody responses to DENV infection. The first to be marketed detected IgM and IgG antibodies using the enzyme-linked immunosorbent assay (ELISA). ^, Clinically, the tests of choice now are the rapid immunoassays. These tests identify acute or recent primary DENV infections when DENV IgM is detected in sera collected 5 days or more after onset of fever. A secondary DENV infection is tentatively established when the etiological agent is recovered in an individual with circulating IgG antibodies during the acute or early convalescent stage. Formal diagnosis requires evidence that antibodies measured derive from a prior DENV and not another flavivirus infection. In the United States, the Food and Drug Administration (FDA) regulates the marketing of commercial tests to detect dengue.

Treatment and Prevention With Antimicrobials or Antivirals

There are no antivirals available to treat or prevent DENV infections/disease.

History of Vaccine Development

The complexity of dengue pathogenesis and dengue virology has deferred the achievement of safe and efficacious dengue vaccines for more than four decades. Each DENV type causes disease and death, and in many dengue-endemic areas, there is concomitant cocirculation of multiple types, mandating vaccines that protect against all four DENVs. Following natural infection of humans with DENV-1 a short-duration cross-protection against infection with DENV-2 was observed. It is now well established that heterotypic protective immunity following infection with any first DENV reduces the risk of severe disease with a second heterotypic DENV infection for a period of 2 years. ^, As the interval between first and second DENV infections lengthens, so does the risk of severe disease. Immunologically primed individuals may be at risk of enhanced disease for a lifetime as evidenced by DHF/DSS accompanying secondary DENV-2 or DENV-3 infections 20 years after DENV-1. ^, The mechanism of early heterotypic protection is not established. Studies on DENV-immune humans and experimental studies using mouse models of severe disease suggest a role for T cells. ^, But antibodies also contribute to protection.

Several flaviviral live-attenuated vaccines (LAV) have proved safe and durably efficacious, 17D yellow fever and SA 14-14-2 Japanese encephalitis vaccine. Live-attenuated viral vaccines replicate in recipients, presumably present all viral antigens in a manner analogous to that of wild virus infections, eliciting antibody and T-cell responses. In dengue endemic countries many adults experiencing 3–4 DENV sequential infections are demonstrably protected against severe dengue disease.