Over the past fifty years, Marek’s disease—an illness of fowl—has become fouler. Marek’s is caused by a highly contagious virus, related to those that cause herpes in humans. It spreads through the dust of contaminated chicken coops, and caused both paralysis and cancer. In the 1970s, new vaccines brought the disease the under control. But Marek’s didn’t go gently into that good night. Within ten years, it started evolving into more virulent strains, which now trigger more severe cancers and afflict chickens at earlier ages.
Andrew Read from Pennsylvania State University thinks that the vaccines were responsible. The Marek’s vaccine is “imperfect” or “leaky.” That is, it protects chickens from developing disease, but doesn’t stop them from becoming infected or from spreading the virus. Inadvertently, this made it easier for the most virulent strains to survive. Such strains would normally kill their hosts so quickly that they’d die out. But in an immunised flock, they can persist because their lethal nature has been neutered. That’s not a problem for vaccinated individuals. But unvaccinated birds are now in serious trouble.
This problem, where vaccination fosters the evolution of more virulent disease, does not apply to most human vaccines. Those against mumps, measles, rubella, and smallpox are “perfect:” They protect against disease and stop people from transmitting the respective viruses. “You don’t get onward evolution,” says Read. “These vaccines are very successful, highly effective, and very safe. They have been a tremendous success story and will continue to be so.”
He is more concerned about the next generation of vaccines that are being developed against diseases like HIV and malaria. People don’t naturally develop life-long immunity to these conditions after being infected, as they would against, say, mumps or measles. This makes vaccine development a tricky business, and it means that the resulting vaccines will probably leak to some extent. “This isn’t an argument against developing those vaccines, but it is an argument for ensuring that we carefully check for transmission,” says Read.
“The candidate Ebola vaccines are also foremost in my mind,” he adds. “Some of the monkey trials suggest that they may be perfect, but we need to be very confident that they don’t leak. If they do, and some vaccinated individuals are capable of passing on Ebola, that might lead to the evolution of very dangerous pathogens.”
He is also concerned about animal vaccines, which are often leaky. These include vaccines against Newcastle disease in poultry, Brucella in livestock, and especially bird flu. When bird flu outbreaks hit American and European farms, the birds are culled. But in Southeast Asia, they’re often vaccinated, “and those vaccines are leaky,” says Read. “It creates an analogous situation to Marek’s.” The birds might survive more lethal forms of the virus, which they could then spread to each other—and potentially to people.
Read first proposed the “imperfect vaccine hypothesis” back in 2001, on purely theoretical grounds. It proved controversial, not least because he had neither experimental evidence nor case studies to support the idea. Then, a colleague told him that the hypothesis might explain the increasing virulence of Marek’s disease. “I wrote the name down, misspelled it, and couldn’t find anything in the literature!” Read says. He only heard about the condition again when he was asked to speak at a Marek’s conference. There, someone put him in touch with Marek’s expert Venugopal Nair from the Pirbright Institute.
The duo infected vaccinated and unvaccinated chicks with five different strains of Marek’s virus, of varying virulence. They found that when unvaccinated birds are infected with mild strains, they shed plenty of viruses into their surroundings. If they contract the most lethal strains, they die before this can happen, and their infections stop with them. In the vaccinated chicks, this pattern flips. The milder strains are suppressed but the lethal ones, which the birds can now withstand, flood into the environment at a thousand times their usual numbers.
Read and Nair also found that the “lethal” strains could spread from one vaccinated individual to another, and that unvaccinated chickens were at greatest risk of disease and death if they were housed with vaccinated ones.
All of this is consistent with the imperfect vaccine hypothesis. It doesn’t prove that imperfect vaccines drove the evolution of today’s extra-virulent strains, “and we may never know for sure why those evolved in the first place,” Read writes. Other factors, like the fact that modern chickens are genetically similar or raised in dense, crowded conditions, may have also played a role. Still, it’s at least clear that vaccines can keep virulent strains in circulation. “For the chicken industry, these results are actually an argument for getting the vaccine,” says Read. “Any chicken that doesn’t get it is at even greater risk than it would be in the 1950s.”
“This work may drive change in the way that vaccines are developed and tested, so that there is much greater emphasis on their ability to prevent infection and transmission, rather than only on their ability to prevent clinical disease,” says Joanne Devlin from the University of Melbourne. “I think that would be a positive step.”
Katherine Atkins from the London School of Hygiene and Tropical Medicine agrees. “While more theoretical work is now being conducted prior to vaccine roll-outs,” she says, researchers need to look beyond how vaccines curb epidemics. They must also consider “the long-term evolutionary consequences of new vaccine introduction.”
But Vincent Racaniello from Columbia University says, “We still do not have any proof that allowing a virus to replicate in a vaccinated individual will select for more virulent viruses.” The new results simply show that leaky vaccines allow virulent viruses to spread—not that they allow those viruses to evolve in the first place. The only way of doing that is to infect vaccinated chickens with mild strains and see if more virulent ones arise after many rounds of transmission.
Racaniello is also unconvinced that the effect would generalise to other vaccines. For example, the Salk polio vaccine—one of two that are used—is a little leaky. “People who are immunized can be infected with poliovirus and the virus can replicate in their guts, be shed, and transmitted to others,” says Racaniello. “This behaviour has been well documented in human populations, yet the virulence of poliovirus has not increased for the 50+ years during which this vaccine has been used.”
That is no reason to rest on our laurels, says Read. It’s important to at least check for the emergence of deadlier viruses if vaccines are imperfect—and perhaps to take preventative measures. For example, a leaky malaria vaccine could be paired with bed nets that would stop mosquitoes from spreading more virulent strains of malarial parasites to unvaccinated people. “If someone developed [such a vaccine] and it worked, we should go ahead and use it, but not think of it as a magic bullet,” says Read. “I’d say that anyone who is vaccinated against malaria should be under a bed net too.”
Reference: Read, Baigent, Powers, Kgosana, Blackwell, Smith, Kennedy, Walkden-Brown & Nair. 2015. Imperfect Vaccination Can Enhance the Transmission of Highly Virulent Pathogens. PLoS Biol http://dx.doi.org/10.1371/journal.pbio.1002198