If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.




Adenoviral vectors are the new COVID-19 vaccine front-runners. Can they overcome their checkered past?

CanSino Biologics, Johnson & Johnson, and the University of Oxford are all using genetically engineered common cold viruses to make COVID-19 vaccines. The technology is more than 30 years in the making, but it’s yet to yield an effective vaccine for humans

by Ryan Cross
May 12, 2020 | A version of this story appeared in Volume 98, Issue 19


A photo of a CanSino scientist holding a vial of its Ad5-based COVID-19 vaccine.
Credit: CanSino Biologics
CanSino Biologics began a clinical trial of its Ad5-based COVID-19 vaccine on March 16.

Governments around the world are making a big bet that the first vaccines for COVID-19 could be made with genetically engineered viruses. The engineered viruses, called adenoviral vectors, are designed to shuttle a gene from SARS-CoV-2, the novel coronavirus that causes COVID-19, into our bodies where our cells will read it and make coronavirus spike proteins.

As with all vaccines, the idea is to trick our body into thinking it’s been infected. Those self-made spike proteins would train our bodies to detect and terminate any real SARS-CoV-2 infections before the virus wreaks havoc. The technique has been in development for more than 3 decades, but thanks to COVID-19, it is about to be put to the test like never before.

As soon as the genetic sequence of SARS-CoV-2 was posted online in January, three groups began independently working on adenoviral vector vaccines for COVID-19: CanSino Biologics, the University of Oxford, and Johnson & Johnson. All three teams are chock full of vaccine veterans, and their COVID-19 programs have garnered global attention for their scale and speed.

Support nonprofit science journalism
C&EN has made this story and all of its coverage of the coronavirus epidemic freely available during the outbreak to keep the public informed. To support us:
Donate Join Subscribe

Many scientists believe that a COVID-19 vaccine will be needed to stop the spread of the coronavirus and end the pandemic, which has claimed more than 270,000 lives so far. Over the past 4 months, more than 100 groups have joined the race to develop COVID-19 vaccines. Their efforts cover a spectrum of technologies, including conventional, inactivated viruses and new, unproven technologies like messenger RNA (mRNA) vaccines.

Amid that race, adenoviral vector vaccines have the distinction of reaching humans first.

CanSino’s adenoviral vector vaccine made it into human trials in China in March. Later that month, the US government pledged more than $500 million to help J&J make up to 1 billion doses of its vaccine, which isn’t expected to enter clinical testing until September. Oxford University, meanwhile, is taking the most ambitious approach: in late April, it started a 1,100-person trial to prove its vaccine’s safety while looking for signs that it works. Its goal is to complete that study in just a month and then begin a Phase III trial of 5,000 people as soon as June. If successful, Oxford’s program would leapfrog all other COVID-19 vaccines in development.

“They have the most aggressive timeline of any group,” says Thomas G. Evans, chief scientific officer of Vaccitech, a company founded in 2016 to commercialize Oxford’s adenoviral vector technology. In April, Vaccitech and the big pharma firm AstraZeneca announced a partnership to develop and commercialize the Oxford vaccine. “Oxford is likely to have the first efficacy data in the world” for a COVID-19 vaccine, possibly as early as August, meaning distribution of the vaccine could begin in the fall, Evans says.

Some scientists say that adenoviral vector vaccines, and Oxford’s vaccine in particular, may be society’s best chance for a return to normalcy. “From what I’ve seen out there, they are probably the most promising platform,” says Hildegund Ertl, who studies adenoviral vector vaccines at the Wistar Institute in Philadelphia.

Compared with some of the newer, experimental technologies—such as Moderna’s mRNA vaccine, which was the first to enter human trials in the US—adenoviral vectors are touted as a more tried-and-true approach. J&J calls its adenoviral vector platform a “proven” technology. While adenoviral vectors have been tested in far more people than mRNA vaccines, the technology is used in only one commercial vaccine today: a rabies vaccine used to immunize wild animals. So far, no adenoviral vector vaccines have demonstrated they can prevent disease in humans.

“It is not proven until it is licensed, and in postlicensure, continues to succeed,” Ertl says. “To say it is proven without peer-reviewed efficacy data is a stretch.”

There’s another potential problem. Just as human bodies develop immune responses to most real viral infections, our bodies also develop immunity to adenoviral vectors. That makes booster shots of adenoviral vector vaccines problematic. Upon a second injection, our bodies will unleash an antibody attack on the vaccine itself. And since adenoviral vectors are based on natural viruses that some of us might already have been exposed to, the vaccines might not work for everyone.

Viral vectors for the masses

Adenoviral vectors try to prove their worth

Adenoviral  |  Clinical trial status


CanSino Biologics
Ad5 | Phase II ongoing since April

University of Oxford, Vaccitech, and AstraZeneca
Chimpanzee adenovirus | Phase I/II ongoing since April; Phase III could begin in June

Stabilitech BioPharma
Oral Ad5 | Phase I starts in June

Second-generation Ad5 | Phase I starts by July

Gorilla adenovirus | Phase I starts in July

Johnson & Johnson, Harvard Medical School, and Biomedical Advanced Research and Development Authority
Ad26 | Phase I starts in September

Intranasal Ad5 | Phase I starts by October

Oral Ad5 | Phase I starts by end of year

Sources: Companies.

From failed gene therapy to vaccine

Adenovirus vaccines might be grabbing the limelight amid the coronavirus pandemic, but they have a checkered past.

When scientists began creating adenoviral vectors in the 1980s, most worked with a particular kind of adenovirus called Ad5, which ubiquitously infects humans and causes the common cold. Researchers stripped Ad5 of the genes it needed to replicate and inserted those genes into genetically engineered cell lines. That ensured that the modified viruses could be grown only in these special cells in the lab. It also opened up space in the Ad5 genome for scientists to stitch in new genes of their choosing.

Many scientists hoped to use Ad5 to deliver a human gene that could correct rare genetic mutations—an approach called gene therapy. Those efforts came to a grinding halt in 1999 when a teenage boy with a rare genetic liver disease died after receiving an injection of an Ad5-based gene therapy, which had been designed in James Wilson’s lab at the University of Pennsylvania.

The large dose of 38 trillion viruses the patient was given sparked massive body-wide inflammation and sent his immune system into overdrive. After that, scientists mostly stopped using adenoviral vectors for gene therapy, in which the dose needs to be high to reach many cells of the body.

But vaccine developers viewed adenovirus-induced inflammation as an asset.

“There is an expression out there that a failed gene therapy makes a good vaccine,” says Luk Vandenberghe, a viral vector expert at Harvard Medical School.

One attractive feature is that adenoviruses’ inflammatory effects mean developers don’t have to use adjuvants, molecules added to conventional vaccines to direct the immune system’s attention to the viral protein. The adenoviruses themselves drive the inflammation, which is kept under control by giving the vaccines at low doses.

And all genetic vaccines—DNA vaccines, mRNA vaccines, and adenoviral vector vaccines—mimic a natural viral infection by forcing our bodies to produce viral proteins inside our cells. That spurs the T cells of our immune system to attack these vaccinated cells, and in the process, they learn to seek and destroy cells infected with the real virus in the future.

Traditional vaccines, made from weakened viruses or viral proteins, stimulate B cells to make antibodies against the virus. Those antibodies latch onto invading viruses and prevent them from entering our cells.

The problem is that once the virus infiltrates our cells, the antibodies from a traditional vaccine are useless. It’s at that stage that T cells need to swoop in. Adenovirus vectors “are the best of all vaccines at inducing a T-cell response,” Wistar’s Ertl says.

That’s why some vaccine developers turned to adenoviral vectors in the early 2000s to tackle diseases, such as AIDS, malaria, and tuberculosis, caused by pathogens that hide out in cells. The largest, and most infamous, effort was led by Merck & Co., which had developed an Ad5-based vaccine for HIV. Two large clinical trials were halted early in 2007 when it became clear that the vaccine was not working—and, alarmingly, may have even increased the risk of HIV infections in a subset of people with preexisting immunity to Ad5.

“That put a big kibosh on adenoviruses for the next 5 years,” Vaccitech’s Evans says.

The National Institutes of Health, which partly funded the trials, called a meeting to decide whether to proceed with trials of Ad5-based vaccines. In 2009, it decided to push forward with a modified version of a planned HIV vaccine trial so long as the participants didn’t have preexisting immunity to Ad5. Results from the 2,500-person study showed that the vaccine was safe, but it still didn’t work.

That study curbed enthusiasm for Ad5, but didn’t eliminate it altogether. CanSino, a Chinese company founded by former Sanofi vaccine developers, developed an Ad5-based vaccine for Ebola during the 2014 outbreak, and a Phase II study showed that the vaccine induced an antibody response 4 weeks after injection.

In 2017, China approved the vaccine, but only for emergency use and national stockpiling. That made it the first, and still only, adenoviral vector vaccine approved for humans—with the big caveat that the Phase II study did not prove the vaccine prevented Ebola infections. Furthermore, the antibody levels dropped sharply within 6 months of vaccination. Most participants had preexisting immunity to Ad5, which some scientists believe may have curtailed the vaccine’s ability to induce a longer-lasting immune response.

The firm’s Ebola experience enabled it to move quickly on a COVID-19 vaccine using Ad5. On March 16, CanSino became the first company to begin a clinical trial of a COVID-19 vaccine. The 108-person Phase I safety study is completed, although the results have not yet been released. A 500-person Phase II study is underway.

Some scientists have cast doubt about CanSino’s chances of success, but industry veterans say that preexisting immunity to Ad5 can be overcome with a higher dose of the vaccine—which will require more stringent monitoring of side effects.

“They have a better shot at this than anyone because they have a huge manufacturing facility, great expertise, and the financial and manpower backing of the Chinese government,” Evans says. “If you are discounting CanSino, you are making a huge mistake.”

There is an expression out there that a failed gene therapy makes a good vaccine.
Luk Vandenberghe, Harvard Medical School

Several smaller firms are also developing COVID-19 vaccines based on Ad5. One of them is ImmunityBio, which uses Ad5 vectors with additional gene deletions. CEO Patrick Soon-Shiong says the modification drastically curtails the body’s toxic immune responses to the virus and even allows the vector to be dosed multiple times. The firm has tested the vector in about 200 people in several small clinical trials, mostly for cancer.

Other companies, including Altimmune, Stabilitech BioPharma, and Vaxart, believe they can circumvent preexisting immunity to Ad5 in the bloodstream by administering their vaccines as nasal sprays or pills rather than injections. The experimental formulations could also be easier to manufacture, store, distribute, and use.

“In a pandemic like this, it is not only convenient but almost essential to have something you can easily administer,” says Altimmune chief scientific officer Scot Roberts. “You can even imagine mailing the vaccine to someone.”

Alternatives to Ad5

Even before the failed HIV trials, some scientists believed preexisting immunity to Ad5 would present a problem, so they looked to nature for less common adenoviruses that fewer people would have been exposed to. The vaccine company Crucell Holland and Dan Barouch at Beth Israel Deaconess Medical Center and Harvard Medical School used one of the most promising natural viruses, called Ad26, to make a new adenoviral vector. J&J, which acquired Crucell in 2011, went on to develop multiple Ad26-based vaccines for viruses like HIV, respiratory syncytial virus (RSV), Zika virus, and Ebola virus.

J&J has since administered thousands of doses of its experimental Ebola vaccine to people in the Democratic Republic of the Congo and Rwanda. The vaccine is under review by drug regulators in Europe, meaning it could become the first commercial adenoviral vector vaccine that is proven to prevent a disease in humans.


In January, Barouch began working with J&J on an Ad26-based COVID-19 vaccine. Although J&J won’t start human studies of its vaccine until the fall, it has a leg up on manufacturing capacity. In addition to its own production facilities, it has recruited Emergent BioSolutions and Catalent to help make up to 1 billion doses of the vaccine.

But J&J’s vaccine has potential drawbacks. The firm’s Ebola, HIV, and RSV vaccine regimens all use a shot of an Ad26-based vaccine plus a booster shot of a different vaccine. That combinationmakes it hard to draw comparisons to its COVID-19 vaccine, which uses only Ad26. And Barouch has found that about half the adults in some countries in sub-Saharan Africa and Southeast Asia have preexisting immunity to Ad26, meaning that the vaccine might not work well for these people.

Some labs have sought to avoid the problem of preexisting immunity altogether by using adenoviruses that don’t normally infect humans but do infect our closest relatives. In the early 2000s, Wilson’s lab at Penn began hunting for chimpanzee adenoviruses, which the researchers isolated from the animal’s feces. Soon after, Ertl’s lab at Wistar began collaborating with Wilson to use the chimpanzee adenoviruses as a novel vaccine vector.

Other groups adopted the idea too. “Chimpanzees are very protected, but stools can be easily collected,” says Stefano Colloca, who worked on adenoviral vectors at Merck Research Laboratories’ center in Rome in the early 2000s. He later helped form a company, Okairos, that was spun out of that work when Merck discontinued its Ad5 programs in 2007.

A scientist at MilliporeSigma.
Credit: MilliporeSigma
MilliporeSigma worked with scientists at the University of Oxford to improve the process of manufacturing its chimpanzee adenoviral vector vaccine for COVID-19.

Okairos focused on developing chimpanzee adenoviral vectors that closely resembled human Ad5, and it soon formed a collaboration with a newly founded vaccine center at the University of Oxford called the Jenner Institute. The Oxford team used one of the Okairos chimpanzee-derived vectors to develop a malaria vaccine, which became the first chimpanzee-derived vector to be tested in humans.

In 2012, the Oxford group developed its own chimpanzee-derived vector, dubbed ChAdOx1, based on an adenovirus discovered in chimpanzee feces. The Oxford team went on to create the spin-off company Vaccitech in 2016 and has developed experimental vaccines for a number of diseases, including AIDS, malaria, tuberculosis, and Middle East respiratory syndrome, which is caused by the MERS coronavirus.

A small safety study of that MERS vaccine was conducted in 2018. The results, published this April, showed that most of the 24 people in the trial still had T cells that targeted the MERS virus 12 months after a single injection of the vaccine. They also still had elevated levels of antibodies a year later. But only about half the people who got the highest dose of the vaccine had antibodies that neutralized the MERS virus in lab experiments.

That MERS work allowed the Oxford team to move fast on a COVID-19 vaccine, which essentially swaps in the genetic instructions for the SARS-CoV-2 spike protein. To improve the manufacturing process for its vaccine, Oxford enlisted the help of MilliporeSigma, which will supply equipment to multiple contract manufacturers that could collectively develop tens of millions of doses of the vaccine.

In July, Okairos, which has since morphed into a company now called ReiThera, plans to start a clinical trial of its own COVID-19 vaccine, which is based on an adenovirus discovered in gorilla feces. The biggest drawback to the great-ape adenoviral vector vaccines may be their lack of prior testing in humans. Before the coronavirus pandemic, Oxford’s ChAdOx1 vector had been given to only about 320 people, and ReiThera’s new gorilla-derived vector has never been tested in humans. Although preexisting immunity could limit the effectiveness of the Ad5- and Ad26-based vaccines, at least their developers have a better idea of their vectors’ safety.

And while most vaccine scientists agree that adenoviral vector vaccines are great at spurring T-cell immunity, they disagree on how important that will be for preventing COVID-19. Most research has focused on the immune system’s antibody response to the virus. Adenoviral vector vaccines can induce antibody responses, but they’re usually not as strong as those elicited by more traditional vaccines.

Scientists will be watching the adenoviral vector vaccine trials closely in the coming months to see how they compare with mRNA and DNA vaccines for COVID-19. If the preliminary results are promising and everything goes perfectly, limited numbers of the vaccines could be available for select groups—such as health-care workers—as early as the fall. Many firms hope that a larger number of the vaccines will be available throughout 2021.

“It is possible that many of these vaccine modalities will work or that none of them will work,” Harvard’s Vandenberghe says. “Almost any modality is not going to check all the boxes, and we don’t even know what boxes we need to check. All bets may be needed.”


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.