Dengue vaccines: the road to failure or to success?

ABSTRACT

A licensed vaccine against dengue was shown to enhance disease under some circumstances, even though it protected most vaccinees. We discuss the implication of these results for the future of dengue vaccines.

KEYWORDS:

• Dengue
• enhancement
• correlate of protection

A proverb says that “the road to hell is paved with good intentions.” The Sanofi vaccine company embarked with good intentions on a large and costly program to develop a vaccine against dengue, but our understanding of the disease increased during that development, which eventuated in a vaccine that protects some persons but renders others more susceptible to serious disease.

Dengue is a widespread and dangerous infection, prevalent throughout most of the world but most important in tropical areas of Asia, Africa and Latin America. The cause is a virus that comes in four serotypes, all carried by mosquitoes which inject them into humans. The peculiarity of dengue is that the most dangerous infections may not be the first in any individual, but the second, which may be followed by a severe disease. The pathologic phenomenon may be due to the fading of neutralizing heterotypic antibody to the first infection with time, leaving antibody that no longer neutralizes the virus but which enhances its entry into cells of the macrophage system.¹ Poorly neutralizing antibodies to the viral PrM protein or to induction of CD8 + T cells that react with non-structural proteins of the virus are suspected to be involved.² Regardless of the exact cause, the ideal dengue vaccine elicits homotypic antibodies to all of the four serotypes.

Multiple projects have attempted to develop a dengue vaccine, including those led by Sanofi, NIH and Takeda. All of the current candidates seek to protect against the four dengue serotypes with quadrivalent vaccines. However, each candidate is different. The Sanofi candidate is based on the attenuated 17D yellow fever vaccine virus in which the genes coding for the pre-membrane and envelope proteins of each of the four dengue serotypes have been inserted. Thus, it is a chimera in which some antigens are from dengue viruses while others are from Yellow Fever virus. The Takeda candidate originated at CDC, where a dengue type 2 virus was attenuated in cell culture and the PrM and envelope genes from types 1, 3 and 4 dengue were inserted into the attenuated type 2. In a phase 3 trial, the reported efficacy in 4- to 16-y-old subjects was 81% against disease and 95% efficacy against hospitalization.³ A third candidate dengue vaccine was developed at the U.S. National Institutes of Health and consists of attenuated versions of each of the four dengue serotypes mixed together. It is being developed jointly by a number of manufacturers including Merck in the United States, and has the advantage in principle of requiring only one dose.⁴ Other attempts to develop a dengue vaccine are still in early stages.

The Sanofi dengue-yellow fever vector vaccine, called Dengvaxia, showed the efficacy of about 75% in the 9–16 y olds and 60% in younger seropositive subjects in studies conducted in Asia and Latin America. However, in originally seronegative subjects efficacy was reduced to about 40% in the older group but about 20% in the younger group.⁵ Moreover, in the 2- to 5-y-old group efficacy faded by the third year, in which there were more dengue cases in the vaccinees compared to placebo recipients, suggesting enhancement of dengue by the vaccine.⁶ This raised the possibility that in young children who had never experienced dengue that the vaccine actually acted like a first infection which made them more susceptible to the hemorrhagic disease when exposed to a second dengue virus, a phenomenon called antibody-dependent enhancement.⁷ As dengue hemorrhagic fever can be fatal this result caused much consternation.⁸

Moreover, Dr. Scott Halstead had warned against the use of Dengvaxia on the grounds that disease might be enhanced in originally seronegative individuals unless the vaccine-induced homotypic antibodies against all four serotypes.⁹ Various analyses of the data were done, notably by WHO, which recommended vaccination only in countries with a high circulation of dengue viruses such that most adults and older children had already been infected with at least one serotype. As a result of that analysis vaccination of those 9 y or older was recommended in the Philippines. However, authorities in that country had already launched vaccination including younger children, which created great concern in parents and controversy in the press that eventuated in stopping vaccination. Moreover, although an estimated 82% of those over 9 y in the Philippines are thought to be seropositive, the fate of seronegative vaccinees in general seemed uncertain.

Indeed, a philosophic evaluation of the situation raised the analogy of the famous ethical dilemma: if a trolley is about to kill five people down a track, is it ethical to switch the trolley to another track where it would only kill one person?¹⁰ Clearly, the answer is not simple and opinions differ. Importantly, a group of experts brought together by WHO recommended the development of serologic tests for prior dengue infection and use of such tests to identify only seropositive individuals for vaccination.¹¹ However, those tests are not yet widely available.

Part of the problem is that a correlate of protection against dengue is uncertain and indeed a research agenda has been formulated to try to find one.^12,13 Another problem is that dengue type 1 viruses in particular are poorly neutralized by antibodies generated in humans.¹⁴

To some extent, two situations in the past have raised similar questions. The first rotavirus vaccine protected against dehydrating diarrhea that could cause dehydration and death, but also increased intussusception.¹⁵ It was withdrawn from the market and eventually replaced by safer vaccines, but in the interim many infants suffered rotavirus diarrhea.¹⁶ An effective vaccine against Lyme disease was withdrawn from the market because of alleged autoimmune reactions that turned out to be false, thus allowing the disease to increase and to spread.¹⁷

If one adopts the credo of “do no harm” it might seem simple to ban the use of Dengvaxia, but it is certain that such an attitude will result in thousands of dengue cases that could have been prevented.¹⁶ Parenthetically, some have recommended reliance on a positive serologic test for prior dengue infection before vaccination, but no current serologic test is foolproof.¹⁸ Thus, one is left with both scientific and moral questions. Other dengue vaccines are under development but in addition to protection, they may well cause both homotypic and heterotypic responses similar to the case of Dengvaxia. Indeed, an inactivated virus quadrivalent dengue vaccine enhanced disease in a rhesus monkey model.^19,20 In addition, the duration of immune responses and immune memory is likely to be an issue, as vaccinees might become susceptibles with time as immunity wanes and boosters might need to be given, with uncertain results.^21,22 It is likely that the situation will be unclear for some time until we have data on persistence of immunity and severity of breakthrough disease. The best course we can follow now is to restrict the use of Dengvaxia in young children in order to minimize potentiation of disease by vaccination. Nevertheless, the fact is that we can now considerably reduce the incidence of dengue by introducing wise and sophisticated vaccine programs with Dengvaxia and ultimately with improved dengue vaccines. A WHO sponsored meeting reached the conclusion that even in the Philippines the overall impact of Dengvaxia would be positive.^18,23

The ideal properties of an improved dengue vaccine will be development of a vaccine that induces long-lasting homotypic immune responses all four serotypes in all age groups. Also, internal antigens of dengue should be used to elicit T cell responses that add to the efficacy of a vaccine.²⁴ The resulting vaccine should be able to prevent mosquito-borne epidemics if used at their beginning, but in any case, could serve as prophylaxis in large populations if vaccination is offered before the mosquito season. Thus, Dengvaxia will have been the first stop on the road to a broadly effective dengue vaccine.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.