In general, the desired characteristics of vaccines for travelers can differ substantially from those for endemic populations. For example, the target age and immune status are likely to differ in travelers. In endemic populations, children are often the most vulnerable group while adults have often attained a certain degree of natural protective immunity [34]. In contrast, travelers of all ages are likely to be immunologically naïve and susceptible to infection. This difference in travelers is accentuated by the fact that international travelers are also more likely to be adults than children [35]. Thus, vaccines for travelers need to target both adults and children, especially those who are immunologically naive. In the case of dengue, an additional complicating factor is that secondary DENV infection is more likely to lead to severe disease than primary infection. Although this does not negate the benefit to individual travelers in preventing primary DENV infection, repeat travelers to dengue endemic regions or expatriates who have already been exposed to DENV may benefit more from dengue vaccination than first time travelers.

Many additional factors specific to each traveler's situation need to be considered when deciding whether to administer a travel vaccine. These factors may relate to the level of exposure risk during travel including the destination country, specific travel locations within the country, type of transportation and accommodations, purpose of travel (e.g., tourism, adventure travel, business, and visiting friends or relatives, etc.), travel conditions, and length of stay. These risk factors may be different for different diseases. For example, the risk of DENV exposure is often higher in urban areas than rural areas, which is the converse of the situation with malaria. Furthermore, since DENV transmission can be spatiotemporally heterogeneous [36], assessing risk based on specific travel locations within a dengue endemic country may be extremely difficult.

Vaccines should ideally be highly efficacious whether for travelers or for endemic populations. However, for travelers, who typically have a much shorter period of exposure and in whom decisions about vaccination are based more on individual factors rather than population-level public health considerations, higher vaccine efficacy is typically required to make the vaccine acceptable to travelers. For malaria, the U.S. military has designated a threshold efficacy of >80% for military personnel and travelers [37]. However, lower efficacies may be acceptable for individual travelers depending on the likelihood of exposure [38]. For example, a traveler who plans to enter an area with an ongoing disease outbreak with high probability of exposure may be more likely to accept a moderately effective vaccine. Therefore, no absolute efficacy threshold is necessarily required for a travel vaccine in all situations. In the case of dengue, vaccine efficacy considerations are complicated by the fact that efficacy may differ for each of the four DENV serotypes [39]. Since dengue epidemiology and circulating serotypes can vary dramatically in different countries and regions, decisions about dengue vaccine use for travel can become complex. Travelers also expect vaccines to be efficacious against both mild and severe disease, since both can have a large impact on travel. Thus, travel vaccines are typically expected to have high efficacy and protect against a wide clinical spectrum of disease.

Vaccination in travelers also has the potential to decrease the risk of spreading the virus from endemic to non-endemic areas. How much impact the vaccine might have in this way would depend in part on the type of protection produced by the vaccine. If a vaccine is efficacious only against severe disease, then the risk of virus spread to non-endemic areas by infected travelers may only be minimally reduced. For example, DENV has been shown to be transmissible to mosquitoes from asymptomatic and pre-symptomatic individuals, even more than symptomatic patients [40]. Therefore, dengue vaccines that prevent severe disease but not mild or asymptomatic infection may still allow significant transmission to occur.

Vaccines require time to induce a sufficient immune response to elicit protection. According to the International Trade Administration Office of Travel and Tourism Industries from the United States Department of Commerce, preparation for leisure travel typically begins 60-90 days prior to the onset of the trip. However, business travelers start preparation for travel 30 days before departure [41]. Many travelers request vaccinations just days before departure. Therefore, vaccines with rapid onset of immunity are an important requirement for travelers. For vaccine regimens that require more than one dose, the level of immunogenicity and the time to attain that level may differ depending on the dose number. In addition, prior vaccination may affect the time required to achieve an immune response. Booster doses of a previously administered vaccine may elicit a faster immune response than de novo vaccination. Therefore, vaccines with booster doses are well suited for use in travelers.

Although all vaccines should ideally provide long-term protection against disease, a travel vaccine requires protection primarily during the travel period. Therefore, a shorter duration of protection may be sufficient for most travel situations. For example, the Department of Tourism of Thailand, where dengue is hyperendemic, indicates that the average visit lasts for about 10 days [42]. If necessary, booster doses can be administered prior to each travel event. For repeat travelers or expatriates, the duration of protection may need to be longer.

The number of doses and schedule is an important factor for travel vaccines [43]. Few numbers of doses and accelerated vaccine schedules can accommodate travelers in need of vaccines on short notice, and encourage compliance with pre-travel vaccination [44]. Fewer doses can also reduce costs to the traveler, which in turn can further increase compliance [45]. In addition, as most dengue-endemic areas in the tropics pose a risk for other diseases, suitability for co-administration with other recommended travel vaccines would be relevant for dengue vaccines for travelers. In this regard, non-live vaccines have more flexibility for co-administration than live vaccines.

The individual risk of infection in travelers is weighed against the risk of complications from vaccination in otherwise healthy individuals. Since the risk of infection in travelers is usually low, the risk of vaccine complications should be similarly very low. This is usually true for most vaccines among general travelers. However, unique to dengue vaccines, there are concerns about the theoretical risk of vaccine-induced immune enhancement leading to more severe disease. It is unclear to what degree this risk exists in various dengue vaccine candidates. However, such a safety signal was indeed detected in very young children during clinical trials of Sanofi Pasteur's dengue vaccine [46].

In special populations such as immunocompromised individuals [47] or pregnant women [48], the risks of the vaccine are usually not fully evaluated. Conversely, the benefits of vaccination in these groups may potentially be greater than in the general population. The decision-making about vaccine administration in these groups may differ substantially from other groups.

CYD-TDV (or Dengvaxia) is a tetravalent live recombinant chimeric dengue vaccine based on a yellow fever vaccine strain (YFV17D) backbone with premembrane (prM) and envelope (E) structural protein genes from each of the four DENV serotypes [50]. CYD-TDV underwent two large phase III trials in five countries in Asia (Indonesia, Malaysia, the Philippines, Thailand, and Vietnam) and five countries in Latin America (Brazil, Colombia, Honduras, Mexico, and Puerto Rico) [5152]. Pooled results during the 25-month active surveillance period from the two trials showed that vaccine efficacy varied depending on serotype, age, baseline dengue serostatus, and severity of disease [46]. Efficacy against symptomatic virologically confirmed dengue among individuals aged 9 years and older was 65.6% (95% confidence interval [CI], 60.7 to 69.9), substantially higher than the efficacy of 44.6% (95% CI, 31.6 to 55.0) in younger children. Dengue seronegative individuals aged 9 years and older had efficacy of 52.5% (95% CI, 5.9 to 76.1), while seropositive individuals in the same age group had efficacy of 81.9% (95% CI, 67.2 to 90.0). Among children younger than 9 years of age, efficacy was 14.4% (95% CI, -111 to 63.5) in seronegative individuals and 70.1% (95% CI, 32.3 to 87.3) in seropositive individuals. Serotype-specific efficacy among children aged 9 years and older ranged from 47.1% (95% CI, 31.3 to 59.2) against DENV-2 to 83.2% (95% CI, 76.2 to 88.2) against DENV-4. Efficacy in children under 9 years ranged from 33.6% (95% CI, 1.3 to 55.0) against DENV-2 to 62.1% (95% CI, 28.4 to 80.3) against DENV-3. Efficacy against severe dengue and hospitalization was substantially higher in individuals aged 9 years and older: 93.2% (95% CI, 77.3 to 98.0) and 80.8% (95% CI, 70.1 to 87.7), respectively; whereas, efficacy among younger children was 44.5% (95% CI, -54.4 to 79.7) and 56.1% (95% CI, 26.2 to 74.1), respectively. Long-term follow-up of hospitalized and severe dengue cases in the Asian phase III trial demonstrated increased risk of hospitalization and severe dengue among the youngest vaccinated children age 2-5 years during the third study year (relative risk, 7.45; 95% CI, 1.15 to 313.80).

Based on these phase III results, CYD-TDV, trademarked as Dengvaxia, was licensed in Mexico, the Philippines, Brazil, El Salvador, and Paraguay, and has been submitted for licensure in other dengue-endemic countries [53]. The vaccine was approved for use in individuals 9-45 years of age (or 9-60 years in Paraguay) living in endemic areas, with three doses administered on a 0/6/12 month schedule. In April 2016, the WHO Strategic Advisory Group of Experts (SAGE) on immunization recommended countries consider introduction of Dengvaxia only in national or subnational geographic settings where dengue is highly endemic [54]. The first school-based dengue immunization program was launched in April 2016 in Marikina, a suburb of Manila, the Philippines [55].

As a travel vaccine, Dengvaxia has substantial drawbacks. Low efficacy in seronegative individuals limits its utility in travelers, who are likely to be dengue naïve [39]. The current three-dose schedule over 12 months is impractical for pre-travel administration and would likely not be acceptable to most travelers [56]. While many travelers to dengue endemic regions are adults [35], a large group of child travelers under 9 years of age would not be eligible for vaccination. Dengvaxia may have a role in repeat travelers or expatriates who may have had prior exposure to DENV [39], and be more motivated to comply with a three-dose 12-month schedule.

The National Institute of Allergy and Infectious Diseases (NIAID) of the U.S. NIH developed TV003/TV005, a tetravalent live recombinant dengue vaccine candidate [57]. Vaccine serotype constructs for DENV-1, DENV-3, and DENV-4 are based on full length viruses. The DENV-2 component is a chimera, in which the prM and E protein genes of DENV-4 vaccine strain were replaced by those of DENV-2 [53]. Each of the four serotype constructs was optimized in phase I trials assessing various monovalent candidates [58]. Among several different tetravalent formulations evaluated for safety and immunogenicity in phase I, TV003 and TV005 have been further studied [5960]. TV005 differs from TV003 only in the 10-fold higher dose of the DENV-2 component in TV005. A single dose of TV005 elicited seroconversion rates of >90% for each DENV serotype, with a tetravalent response attained in over 90% of flavivirus-naïve individuals [60].

The U.S. NIH provided a license for the vaccine to in-country vaccine manufacturers in Brazil (Instituto Butantan), Vietnam (Vabiotech), and India (Panacea Biotec and Serum Institute of India), and to Merck [61]. The Brazilian National Health Surveillance Agency (ANVISA) approved a large phase III trial, based on preliminary safety and immunogenicity results from a phase II trial in Brazil [62], as well as the results from clinical studies performed in the United States [59], including a successful challenge study using a DENV-2 challenge strain [63]. Instituto Butantan initiated the phase III trial of TV003 in February 2016, involving 16,944 healthy subjects aged 2-59 years in Brazil [62].

Although the efficacy of TV003/TV005 in different age groups will not be known until results of the phase III trial become available, one potential advantage of this candidate as a travel vaccine is that a single dose appears to induce tetravalent seroconversion in flavivirus-naïve individuals. A single dose schedule in immunologically naïve travelers would be a strong advantage for a travel vaccine.

Takeda developed a tetravalent live recombinant dengue vaccine candidate (TDV) that consists of an attenuated full length DENV-2 component and three chimeras containing the prM and E protein genes of DENV-1, DENV-3, and DENV-4 expressed in a DENV-2 backbone [64]. Several phase I trials evaluated different formulations, doses, and routes of administration [656667]. A phase II study assessed the safety and immunogenicity of TDV in subjects aged 1.5-45 years in Puerto Rico, Colombia, Singapore, and Thailand with a two-dose schedule at 0/90 days [68]. After a single dose, 59%-86% of individuals had tetravalent responses with the greatest increase observed in the youngest age group (1.5-11 years). After the second dose, the seropositivity rate for DENV-1, -2, and -3 in all age groups was over 95%, while DENV-4 seropositivity rate was 72.7%-100%. Neutralizing antibody titers for DENV-1, DENV-3, and DENV-4 increased significantly in seronegative subjects after the second dose. The optimal dosing schedule (0 day, 0/3 months, or 0/12 months) is being further refined in a phase II trial in Asia and Latin America. A multi-country phase III trial is expected to be initiated in the near future [53].

TDV's current two-dose schedule over three months may limit acceptance by travelers. A single dose or compressed two-dose schedule in dengue-naïve adults may need to be evaluated in additional studies to be broadly useful as a travel vaccine.

GlaxoSmithKline (GSK), Fiocruz, and the Walter Reed Army Institute of Research (WRAIR) have been collaborating to develop tetravalent purified inactivated whole virus candidates (TDENV-PIV and DPIV). The DENV strains were grown in Vero (African green monkey kidney epithelial) cells, purified on sucrose gradients and inactivated with formalin [69]. Phase I trials in the continental U.S. and Puerto Rico have been conducted comparing different formulations of TDENV-PIV with various adjuvants (alum, AS01E, and AS03B) [70]. The vaccine is administered in two doses at 0/4 weeks. In dengue-naïve subjects, neutralizing antibody titers were highest at 4 weeks after the second dose, and waned to a plateau by month seven. A phase I/II study has been initiated in the United States to evaluate different formulations and duration of immune responses.

Multiple dosing and the need for boosters are general characteristics of inactivated vaccines [70]. TDENV-PIV and DPIV's two-dose regimen administered over one month may be reasonable as a travel vaccine. Possible limitations in duration of immunity could be overcome with booster doses prior to travel. Co-administration with or around other travel vaccines would likely not be an issue. Inactivated vaccines could also potentially be used for immunization of immunocompromised travelers [71]. A safe and efficacious tetravalent inactivated dengue vaccine administered in a few doses over a short interval could play a role as a travel vaccine.

Merck's V180 is a tetravalent recombinant subunit protein candidate based on wild-type prM and 80% of E protein (DEN-80E) via expression in the Drosophila S2 cell expression system [72]. A phase I trial evaluated monovalent DENV1-80E adjuvanted with alhydrogel at high/low dose in flavivirus-naïve adults in a three-dose regimen at one month intervals. DENV1-80E induced DENV-1 neutralizing antibodies in most individuals, but the titers in the majority of subjects were modest and waned over time [73]. A larger phase I trial of tetravalent V180 with ISCOMATRIX adjuvant was conducted in 98 flavivirus-naïve adults in Australia. Three doses were administered at one month intervals at three different dosages (high/medium/low). One month after the third dose, V180 with ISCOMATRIX resulted in seroconversion rates of 85.7%-100% [74].

Generally, recombinant subunit vaccines require multiple dosing and adjuvant to achieve suitable immunogenicity [70]. They also may have shorter durations of immune response [75], with requirement for booster doses. However, if robust immunogenicity and efficacy can be demonstrated in future studies, a recombinant subunit dengue vaccine may have a role as a travel vaccine, with similar advantages as whole inactivated vaccines.

The U.S. Naval Medical Research Center (NMRC) developed a tetravalent plasmid DNA vaccine candidate using prM and E protein genes expressed in plasmid vector [76]. A phase I clinical trial evaluated safety and immunogenicity of a DENV-1 monovalent candidate in healthy flavivirus-naïve adults using a three-dose schedule at 0/1/5 months, with poor immunogenicity [57]. Although it is possible that TVDV may have a role as a travel vaccine in the future, the available data is currently insufficient to anticipate its potential use as a travel vaccine.