Bharat Biotech claims it can introduce a Zika vaccine in two years, but given the challenges involved in actually developing a vaccine and regulatory hurdles, is the estimate too hopeful?
When Hyderabad-based firm Bharat Biotech announced earlier this month that it had developed two vaccine candidates against the Zika virus, the development was hailed by some media outlets as a breakthrough. At a press conference, the company reportedly said it would be able to complete animal testing for one of the vaccines within five months and bring it to market in two years. In the days that followed, however, the early media optimism was replaced by more conservative estimates of how long it would take Bharat Biotech (or anyone else) to launch such a vaccine. Early this week, while announcing that the bio-pharmaceutical firm did indeed have one of the two most advanced vaccine candidates against Zika, the World Health Organization said human trials were at least 18 months away. This puts the launch of an effective vaccine at least three or four years away, if Bharat Biotech is very lucky.
Why is the road ahead for Zika vaccine developers going to be uphill? The Wire spoke to several vaccine researchers to find out.
Developing a vaccine
First, the good news: the Zika virus is a flavivirus, the family of viruses to which the eponymous yellow fever virus belongs (‘flavus’ is Latin for ‘yellow’). Vaccines have been developed against several viruses of this family, such as chikungunya, Japanese encephalitis and the tick-borne fever. This means these viruses are vulnerable to vaccines unlike, say, the fast-changing human immunodeficiency virus (HIV).
But one flavivirus that has flummoxed vaccine-makers in the past is dengue. That’s because of dengue’s four serotypes, or the four variations of the virus that cause illness. Each serotype typically has a different antigen: the molecule on the viral surface, which the human body recognises and responds to by producing antibodies. So, the ideal dengue vaccine had to work against all four serotypes. Any less, and there’s a chance that the vaccine itself could be deadly. When a person vaccinated against one serotype of dengue becomes exposed to another, the antibodies against the first can actually aggravate infection by the second. This phenomenon, known as antibody-dependent enhancement (ADE), is why a person catching dengue the second time is often worse. It is also why the attempts to create a dengue vaccine have been stymied for over 80 years, with the first vaccine launched only in 2015.
Zika is a comparatively simpler beast to handle. It has only one serotype, putting it in the category of chikungunya and Japanese encephalitis. So, researchers can skip the tedious process of mixing and matching vaccines against all four serotypes to create a tetravalent preparation. “Flavivirus vaccines are very successful wherever you have one serotype. You take the virus, weaken it and you have a candidate that can go through regular clinical trials. Zika falls in that category. So, it shouldn’t be too challenging,” says Navin Khanna, a researcher working on a recombinant dengue vaccine at the International Centre for Genetic Engineering and Biotechnology, New Delhi.
Testing on animals
The bigger problem is going to be the development of an animal model to test the vaccine. Currently, next to nothing is known about the Zika virus because it wasn’t thought to pose a significant threat to humans until the recent epidemic in Brazil. Very few researchers worldwide have looked for animal models for the virus – animals that are infected by the virus so that vaccine efficacy and safety can be tested in them. It is not known if Bharat Biotech has such models, although the firm has indicated that it will start animal trials in late February. Krishna Ella, chairman of Bharat Biotech, declined to comment for this story.
To understand the challenges in finding animal models, one only needs to look at dengue again. The dengue virus neither replicates, nor causes disease in mice. In monkeys, on the other hand, dengue replicates but stops short of triggering illness. So, monkeys provide only a part-model to test dengue vaccines.
Getting around this problem took decades, with vaccine researchers experimenting with several ways to make mice susceptible to the virus. One approach developed in 1999 saw mice being genetically engineered to lack genes for immunity. These genes, known as interferon-receptor genes, enable proteins called interferons to fight viruses. The dengue virus multiplied more readily in the interferon-gene-knockout mice, causing disease symptoms like capillary leakage. Sadly, these mice models too have shortcomings. For one, they can be infected by one dengue serotype, but not others. Plus, the immunity triggered in interferon-gene-knockout mice by a vaccine is probably too aberrant to reflect what happens in humans.
Researchers have tried other hacks too. They have adapted dengue strains to infect mice more readily and have used mice with natural genetic mutations that suppress their immunity. None of these models are ideal, but each helps vaccine-makers test a different aspect of the vaccine. “These are time consuming and expensive experiments to set up, but they are vital for any vaccine to move forward,” says Khanna.
With Zika, a monkey model is possible because African monkeys are a host for the virus, says Ravi Vasanthapuram, a neurovirologist at the National Institute of Mental Health and Neurosciences (NIMHANS), Bengaluru, who has worked on a Japanese encephalitis vaccine. However, regulatory clearance for primate research is difficult in India, he cautions.
When good animal models are hard to come by, researchers just press on without them. Even when an animal doesn’t develop the disease, it develops an immune response when injected with a vaccine, says Kavitha Singh, program director of the Malaria Vaccine Development Program in New Delhi. “We check if the response is specific to the disease, carry out toxicology studies and take it to humans,” she says. No good animal models are available for malaria either and this increases the chances that the vaccine will fail in human trials.
If the vaccine doesn’t work well enough during animal experiments, the developer goes back to the drawing board and tinkers with the formula. There is much scope for tinkering. Like other flaviviruses, Zika has around ten proteins, out of which three are antigens that vaccine makers can work with. Which antigens will work best and whether the vaccine should be a combination of one, two or all three, can be ascertained through experimentation alone. “It’s trial and error,” says Sudhanshu Vrati, who worked on a rotavirus vaccine launched by Bharat Biotech in March last year. “We think we know the science of making vaccines, but that is not true. Basically you make a calculated judgement and you go for human trials, where the majority of vaccines fail.”
In the case of the rotavirus vaccine, Vrati points out, two vaccines from Merck and Glaxo Smithkline were already in the global market when Bharat Biotech began working on its own. So, the team knew how to go about it. “We knew what we needed: we needed the virus, we needed to weaken it. We only had to develop the process and complete the certification.” Despite these advantages, he says, it took his team 15 years from drawing board to launch. On the other hand, “in Zika, nobody knows what is required,” he says.
Testing on humans
If and when the Zika vaccine make it through animal experiments, it may face a fresh set of challenges in human trials. Just as antibodies to one serotype of the dengue virus can enhance infection from another, antibodies to one flavivirus can possibly talk to other flaviviruses, warn researchers.
It’s a double edged sword. On the one hand, there is the possibility of cross-immunity, which means neutralising antibodies to dengue, Japanese encephalitis or west Nile fever in people exposed to these viruses (such as Indians and Brazilians) can protect against the Zika too. On the other, antibody-dependent enhancement, à la the dengue virus, can kick in. “We know large sections of population has neutralising antibodies to (other flaviviruses). We don’t know if this will affect people who receive the vaccine,” says Vasanthapuram. Such cross-talk can throw a spanner in the works for developers.
Bharat Biotech has indicated that it expects its inactivated vaccine to be developed faster than its recombinant vaccine. An inactivated vaccine is a virus that has been killed off using chemicals, heat or radiation so that it cannot cause disease. However, since it still carries its antigens, its structure can trigger immunity. The polio vaccine is the most famous example of this. An inactivated Zika vaccine is believed to have a good chance at winning regulatory approval because it is likely to be safe for pregnant women, who will be the key targets for the vaccine.
Bharat Biotech’s other vaccine is a recombinant one, normally developed by inserting a bit of the DNA from a Zika antigen into a cell culture such as yeast. When the yeast cells reproduce, they create large quantities of the antigen, which is then injected into humans, triggering an immune response. Designing and testing recombinant vaccines can be more complex and time-consuming.
The question of where clinical trials for Bharat Biotech’s Zika vaccine will be conducted remains moot. PN Rangarajan, a researcher at the Indian Institute of Science who has worked on a recombinant hepatitis B vaccine, says it is unlikely that Bharat Biotech will be able to carry out human trials in India, where there is no Zika outbreak. “When Zika is not a problem in this country, why would the country give permission?” he asks. Vasanthapuram and Singh say safety studies, the first stage of human clinical trials can be conducted in India because they involve giving the vaccine to healthy individuals who are then observed for adverse effects.
But efficacy studies are a different ball game. They require a population at-risk of developing Zika so that researchers can compare rates of illness between the vaccinated arm and the placebo arm to calculate how often the vaccine prevents Zika fever. If such efficacy studies are to be conducted anywhere other than an outbreak country, researchers have to find what is known as a surrogate marker for immunity. A surrogate marker, such as the level of antibodies in a vaccinated person, can predict whether the vaccine will prevent disease or not, when the actual infection is not present in a population.
Such surrogate markers are often elusive, says Singh. In dengue, for example, the levels of neutralising antibodies in a vaccinated person may or may not predict the chances of developing dengue. Without a surrogate marker for immunity, Bharat Biotech is going to have to take its human trials to an outbreak country, Singh believes.
A final unknown in Bharat Biotech’s vaccine-development program is regulatory clearances. Some researchers, including Vrati and Rangarajan, have questioned how the firm was able to import the Zika virus into India, given that Zika hasn’t yet affect Indians. When asked by The Wire whether his firm received an import permit from the Drug Controller General of India, Ella refused to comment. Member secretary of the Review Committee for Genetic Modification, SR Rao, also did not respond to an email about whether Bharat Biotech had permission to develop a recombinant vaccine. Before the company continues development, it ought to answer these questions, argue Vrati and Rangarajan.
Developing a Zika vaccine in the normal course of events would have taken around 15 years. With expedited clearances, given the emergency in South America, Bharat Biotech can avoid some regulatory lags. But the sheer R&D challenge of developing any vaccine is more difficult to bypass. Having two candidates definitely gives the firm a headstart in the Zika vaccine race, but this is just a few minutes shaved off the beginning of a marathon.