Six vaccine candidates in clinical trials for COVID-19 employ viruses to deliver genetic cargo that, once inside our cells, instructs them to make SARS-CoV-2 protein. This stimulates an immune response that ideally would protect recipients from future encounters with the actual virus. Three candidates rely on weakened human adenoviruses to deliver the recipe for the spike protein of the pandemic coronavirus, while two use primate adenoviruses and one uses measles virus.
Most viral vaccines are based on attenuated or inactivated viruses. An upside of using vectored vaccines is that they are easy and relatively cheap to make. The adenovirus vector, for example, can be grown up in cells and used for various vaccines. Once you make a viral vector, it is the same for all vaccines, says Florian Krammer, a vaccinologist at the Icahn...
Once inside a cell, viral vectors hack into the same molecular system as SARS-CoV-2 and faithfully produce the spike protein in its three dimensions. This resembles a natural infection, which provokes a robust innate immune response, triggering inflammation and mustering B and T cells.
But the major downside to the human adenoviruses is that they circulate widely, causing the common cold, and some people harbor antibodies that will target the vaccine, making it ineffective.
Human adenovirus vectors
CanSino reported on its Phase II trial this summer of its COVID-19 vaccine that uses adenovirus serotype 5 (Ad5). The company noted that 266 of the 508 participants given the shot had high pre-existing immunity to the Ad5 vector, and that older participants had a significantly lower immune response to the vaccine, suggesting that the vaccine will not work so well in them.
With vectors you are always trying to find the sweet spot. Too weak, and they don’t work. Too strong, and they are too toxic.—Nikolai Petrovsky, Flinders University
“The problem with adenovirus vectors is that different populations will have different levels of immunity, and different age groups will have different levels of immunity,” says Nikolai Petrovsky, a vaccine researcher at Flinders University in Australia. Also, with age, a person accumulates immunity to more serotypes. “Being older is associated with more chance to acquire Ad5 immunity, so those vaccines will be an issue [with elderly people],” Krammer explains. Moreover, immunity against adenoviruses lasts for many years.
“A lot of people have immunity to Ad5 and that impacts on how well the vaccine works,” says Krammer. In the US, around 40 percent of people have neutralizing antibodies to Ad5. As part of her work on an HIV vaccine, Hildegund Ertl of the Wistar Institute in Philadelphia previously collected serum in Africa to gauge resistance levels to this and other serotypes. She found a high prevalence of Ad5 antibodies in sub-Saharan Africa and some West African countries—80 to 90 percent. A different group in 2012 reported that for children in northeast China, around one-quarter had moderate levels and 9 percent had high levels of Ad5 antibodies. “I don’t think anyone has done an extensive enough study to do a world map [of seroprevalence],” notes Ertl.
J&J’s Janssen is using a rarer adenovirus subtype, Ad26, in its COVID-19 vaccine, reporting in July that it protects macaques against SARS-CoV-2 and in September that it protects against severe clinical disease in hamsters. Ad26 neutralizing antibodies are uncommon in Europe and the US, with perhaps 10–20 percent of people harboring antibodies. They are more common elsewhere. “In sub-Saharan Africa, the rates are ranging from eighty to ninety percent,” says Ertl.
Also critical is the level of antibodies in individuals, notes Dan Barouch, a vaccinologist at Beth Israel Deaconess Medical Center and Harvard Medical School. For instance, there was no neutralizing of Ad26-based HIV and Ebola vaccines in more than 80,000 people in sub-Saharan Africa, he says. “Ad26 vaccine responses do not appear to be suppressed by the baseline Ad26 antibodies found in these populations,” because the titres are low, Barouch writes in an email to The Scientist. Barouch has long experience with Ad26-based vaccines and collaborates with J&J on their COVID-19 vaccine.
The Russian Sputnik V vaccine, approved despite no published data or Phase 3 trial results, starts with a shot of Ad26 vector followed by a booster with Ad5, both of which carry the gene for the spike protein of SARS-CoV-2. This circumvents a downside of viral vector vaccines, specifically, once you give the first shot, subsequent injections will be less efficacious because of antibodies against the vector. Ertl says she has no idea of the proportion of the Russian population with Ad26 or Ad5 antibodies, and there seems to be little or no published data from countries that have expressed interested in this virus, such as Venezuela and the Philippines.
Simian adenovirus vectors
An alternative is look to our nearest relatives. Chimp adenoviruses were the focus of vaccine interest by Ertl for HIV and by Adrian Hill at the University of Oxford for malaria. “About one percent of people have antibodies to the chimp adenovirus, probably because of cross reactivity, which is why we use it,” explains Hill, referring to the COVID-19 vaccine candidate ChAdOx1 nCoV-19, which has shown antibody and T cell responses in an early phase clinical trial. This candidate, which also encodes the instructions for producing SARS-CoV-2 spike protein, is now in Phase 3 trials in the UK, US, South Africa, and Brazil and is to be manufactured by AstraZeneca.
Unfortunately, says Ertl, use of the attenuated chimp virus in a COVID-19 vaccine means it cannot now be used for malaria, because those vaccinated for the coronavirus will have antibodies against the vector. But there are other simian vectors. In Italy, a Phase 1 trial of a COVID-19 vaccine with a gorilla adenovirus vector has begun recruiting healthy volunteers. Ertl says that having multiple adenoviruses from different species is “a good thing, because it broadens the range of diseases we could tackle.” It could also allow animal virus vectors for COVID-19 vaccines to be used in places where human adenovirus immunity is high.
Not everyone is enthusiastic about vector-based vaccines. “Their reactogenicity profile is not great,” says Petrovsky, meaning they stimulate a strong immune response. “Even [President Vladimir] Putin commented that his daughter had a fever [after taking Sputnik V]. Generally, fevers are a no-no for a vaccine.” He says headache and fever have been relatively common in early results from vaccines based on viral vectors. Some people are prone to having convulsions from fevers, so extreme reactions cannot be ruled out, he adds.
Petrovsky says children generally react more strongly to vaccines than adults do, and that could be a huge drawback in countries with young populations such as India. “With vectors you are always trying to find the sweet spot,” says Petrovsky, which is their Achilles’s heel. “Too weak, and they don’t work. Too strong, and they are too toxic.” Petrovsky is involved in the development of Covax-19, a recombinant protein–based vaccine plus adjuvant that is in early clinical trials and was developed by his company Vaxine Pty in Australia.
So far, there is not much experience with vector-based vaccines on the market. The European Medicines Agency granted market authorization in May for a new Ebola vaccine that consists of a prime shot with an Ad26 vector, and a booster with an attenuated poxvirus (MVA). An HIV vaccine trial based on Ertl’s research was to have started this fall, but has been delayed until next year due to COVID-19. “We don’t have post-licensing experience,” says Ertl, in relation to vector-based vaccines, “but these things have been in multiple trials, so we have a reasonably good idea about what doses are tolerated and about safety concerns.”
A measles vector
In August, a trial in France and Belgium began recruiting volunteers to test a COVID-19 vaccine based on a replicating measles vaccine virus. This so-called Schwartz strain was weakened in the 1960s by serial passaging on chicken cells. The virus expresses the full-length spike protein of SARS-CoV-2 and has been tested in mice, say scientists at the Pasteur Institute in France who licensed the vector technology to Themis in Austria. It was previously tested on mice for SARS and for MERS.
It was shown previously that pre-existing immunity to measles acquired by infection in the elderly or vaccination in young people did not dampen responses to a Chikungunya vaccine based on this same vector. The measles vector “goes into cells, then makes more measles vaccine. It will come out again, infect more cells, but after a few cycles it stops,” says vaccine scientist Christiane Gerke of the Pasteur Institute who is leading the COVID-19 vaccine trial. That the measles strain replicates distinguishes it from the adenovirus vectors and could explain why pre-existing antibodies do not matter. “So long as measles antibodies at the start do not eliminate all of the vaccine, then the vaccine replicates itself,” says Gerke.
The live nature of the measles vaccine strain means that it could not be given to immunocompromised individuals. However, the Swartz strain has about 50 mutations and measles vaccine strains have never escaped these attenuation shackles and caused disease in healthy people. “It is a promising candidate,” says Krammer, though a little behind the others. The Pasteur Institute could not confirm whether volunteers had begun receiving the vaccine. In June, Themis was acquired by Merck, a company with a significant vaccine portfolio.
Success with viral vectors has implications for vaccine development overall. “It took a very long time for viral vectors to end up on the market, which they did with the Ebola vaccines,” says Krammer. “The way I see it, this is going to speed up vaccine development in general.” That is, as long as there is a successful outcome with a COVID-19 vaccine. Any misstep by a regulator with one of these vaccines could retard the potential of vector-based vaccines for multiple diseases, says Krammer.