Interview: Why a safe and effective dengue vaccine is proving elusive

The United States Food and Drug Administration last week approved restricted use of Dengvaxia. But the dengue vaccine, manufactured by Sanofi, has a controversial past.

In 2016, the World Health Organisation recommended the vaccine for all children aged nine to 16, based on early results of several large trials that Sanofi conducted globally. The studies, though, showed some children who had not been exposed to dengue went on to develop severe forms of the disease after receiving the vaccine.

On WHO’s recommendation, the Philippines used the vaccine in a mass immunisation programme covering over 1,00,000 children in 2016. A year later, Sanofi released new data showing the vaccine heightened the risk of hospitalisation and plasma leakage syndrome, in which blood vessels start to leak plasma. This brought the Philippines’ immunisation drive to an abrupt halt. Its government started looking into the deaths of children who had been given the vaccine and launched an investigation into the procurement of the vaccine, leading to charges against several vaccine researchers.

Yet, a dengue vaccine is one of the most sought-after medical interventions in the world today. The International Research Consortium on Dengue Risk Assessment, Management and Surveillance estimates, based on data for 2010, that 390 million people globally get dengue every year. Around 96 million are symptomatic cases, with Indian alone accounting for nearly 34 million of them.

In India, however, dengue is vastly underreported despite being a notifiable disease. The National Vector Borne Disease Control Programme’s data shows that only about 1,00,000 cases are reported every year.

Most people who contract a mild dengue infection may not even know it since the symptoms are mild – fever, headaches and joint aches. But 2% to 4% of people who have suffered a dengue infection previously can develop a severe, potentially fatal, infection when they encounter another kind of dengue virus, says Sathyamangalam Swaminathan, a virologist at the International Centre for Genetic Engineering and Biotechnology in Delhi. Swaminathan has been working on dengue for the past 20 years and on vaccine development for nearly a decade.

In an interview to Scroll.in, he explained the complex mechanism by which dengue infections occur, what the problems with Dengvaxia are and why a safe vaccine for dengue has proven elusive. Edited excerpts:

The US Food and Drug Administration just approved Dengvaxia for use in very restricted settings, that is, only for children aged 9 to 16 who have been previously exposed to the virus. What does this mean for India and other countries that are looking for a dengue vaccine?
We have a serious dengue problem in our country. We have about 33-34 million infections. We need a vaccine. A vaccine by definition needs to protect people from a disease they were not exposed to. Here we have this very enigmatic conundrum that you need to have been exposed to the disease to benefit from the vaccine. So that is a very strange situation.

Is this because of the nature of the disease or the nature of the vaccine?
It is inter-related. The disease is caused by the dengue virus of which there are four types that are closely related but different enough to be recognised as independent entities. All four can cause the disease. The peculiarity of dengue is that if someone were to get infected by one of these viruses they get protected for life from reinfection with the same virus. Suppose this is dengue type 1, then the antibodies produced will not protect the person from dengue types 2, 3 and 4 beyond a certain period of time.

Why does this happen?
If a person gets a dengue 1 infection, then he produces antibodies that protect against further infection by dengue 1. It is called homotypic immunity to dengue 1. But homotypic immunity to dengue 1 may give you protection from infection from dengue 2, 3 and 4 for a short period of time after which such cross protection goes down. In fact, the antibodies produced stop protecting the body but start spreading the infection, making it more vigorous when a new dengue virus comes along. This is called antibody dependent enhancement or ADE.

If a person who has had a dengue 1 infection encounters a dengue 2 virus after this cross protection period has elapsed, then he is likely to get a severe dengue disease. The antibodies from the first dengue 1 infection actually help the second infection by dengue 2 to get into the cells, replicate and produce more of dengue 2.

Is this a problem that is also seen with Dengvaxia?
Dengvaxia is a mixture of four live attenuated dengue-yellow fever chimeric viruses. The ideal scenario for a tetravalent vaccine – one with four antigens – is that the immune system recognises all four viruses and produces antibodies to all of them. But what we see is that when the four viruses are mixed, a phenomenon called interference occurs where one virus tends to replicate more at the expense of the others. The immune system responds to that virus and the vaccinated person ends up getting more immunity to that type of dengue.

The vaccine is, what I call, physically tetravalent but ends up being functionally monovalent and it gives only partial immunity. For a short period of time, though, this partial immunity can protect from all four types of dengue.

This was seen in trials where they administered the vaccine and checked the efficacy one year later. That is still in the early stages of interaction of the four types of virus and they saw all four types of antibodies were being made. When they did a long-term follow up they discovered problems. But the vaccine got pushed forward based on the early results. The world was still learning about dengue and no one anticipated these problems.

How long is a dengue vaccine expected to provide immunity?
The hope is that it would give lifetime immunity. There is a belief that the dengue vaccine is needed in endemic countries where the dengue virus circulates. If somebody has been vaccinated and he truly develops tetravalent immunity and he is in an environment where he encounters the virus so there will be a natural boosting of the immune response, then this person will not need further vaccination. That is the ideal scenario but it is not happening.

Is it possible to make a proper tetravalent vaccine?
I believe we cannot make viral vaccines truly tetravalent. I have seen this problem of interference for many years, even before Sanofi developed its vaccine. It is there in the data but people have not really talked much about it.

The vaccine for yellow fever is a robust vaccine and yellow fever virus is related to the dengue virus – they are both flaviviruses. To overcome the problem of ADE, Sanofi created a chimera using the yellow fever virus structure as a backbone for inserting the antigens of dengue. Since all four dengue types have the same basic backbone, the expectation was that all four would replicate comparably and elicit immune responses to each. But that did not happen – there is something else going on that indicates interference is an inherent property of this tetravalent mixture. This is happening with other viruses that are being developed for dengue.

The other problem is when the dengue virus infects the human body it creates two kinds of antibodies. One type is the good antibodies, or neutralising antibodies, that stop the disease. The other type, the enhancing antibodies, are involved in ADE and help spread the disease. The dengue virus on infection makes more of the second type. The live attenuated viral vaccines also created these two types of antibodies. The disease-spreading antibodies do not distinguish between serotypes whereas the disease-protecting ones work only on their particular serotype. So a viral vaccine is going to be a very tough nut to crack.

Is there an alternative approach to develop a dengue vaccine?
One has to probably find ways of getting only the good antibodies to be made for all four viruses. That can probably be achieved only through recombinant technology to make designer vaccines.

Viral vaccines are considered to be more durable and this is why the whole world is focused on them. Recombinant protein-based vaccines are not as robust because they are inert – they stimulate the immune system but a person is probably going to need boosters periodically. But we have such recombinant vaccines for hepatitis B that give durable immunity. Similar vaccines are approved for human papilloma virus as well as malaria.

In your opinion, since we don’t have an alternative yet, is there merit in trying Dengvaxia in India in a restricted way, like the US is doing?
Yes, but there are limitations and concerns that we must be aware of.

This vaccine is for people who have had a prior dengue infection. We can identify prior dengue exposure in a lab test which takes about five hours. But for a vaccination campaign we need a certified rapid test that can be done in the field. We don’t have one for dengue right now.

The test also has to have 100% specificity to identify people who have had only one dengue infection before. This is very difficult because dengue, yellow fever and Zika are all flaviviruses and you cannot eliminate cross-reactivity between them. So your specificity is going to be less than 100%.

Another complication is that antibodies induced by a virus-based vaccine such as dengvaxia to dengue can exacerbate Zika infections. Dengvaxia being a virus-based vaccine to dengue can sensitise people to future Zika infections.

Safety and efficacy are two sides of the same coin in vaccine development. We need to have a vaccine that is efficacious in terms of its ability to stop the virus, all four of them, as well as its ability to not enhance any of the dengue viruses as well as other flaviviruses like Zika, yellow fever or Japanese encephalitis.

Also read: Here’s why we don’t have a vaccine for Zika (and other mosquito-borne viruses)

Lab notes: How bacteria cause premature births, dangers of the dengue vaccine and more