by Reynald Castaneda in London.

Despite its advantages over other vaccine technologies for Covid-19, adenovirus vector vaccines are likely to be tripped up by pre-existing antibodies to the vectors used and the need for a second injection to boost protection.

CanSino Biologics’ Ad5-nCoV and Johnson & Johnson’s AdVac platform-based vaccine use a human adenovirus vector, but a significant chunk of people may already have neutralising antibodies against the vector, decreasing efficacy prospects. Phase I Ad5-nCoV data is also underwhelming, adding credence to the issue of pre-existing antibodies.

AstraZeneca’s AZD1222 and Rome-based ReiThera’s Covid-19 vaccines are also adenovirus vectored but use nonhuman vectors. However, AZD1222’s recent animal data also leave questions about its utility to prevent virus spread. A possible way to improve efficacy is to add a booster shot down the line, perhaps using a different adenovirus vector or even a different vaccine technology. Perhaps AZD1222 only carrying SARS-CoV-2’s spike protein may not be enough.

There are at least 38 companies or universities with a recombinant adenovirus-derived vaccine asset for Covid-19 from preclinical to Phase II/III stages, according to GlobalData. AstraZeneca partnered with Oxford University on 30 April to further develop AZD1222, and on 3 June, the US Federal Government’s vaccine initiative, Operation Warp Speed, named it as one of five finalists along with Johnson & Johnson’s (J&J) candidate. Phase I/II AZD1222 trial (NCT04324606) data are expected shortly, while J&J’s Phase I/IIa will start in July. ReiThera will also begin a clinical investigation in the summer. CanSino is the only company that has released clinical trial data, with Ad5-nCoV already in a Phase II trial (NCT04341389) that has a primary completion date of January 2021.

AZD1222 is concurrently in Phase II/III trial (NCT04400838), with data timelines dependent on community viral transmission rates, the Oxford University website states. A registrational field trial expected to start in the summer is likely to require around 25,000–30,000 volunteers if the annualised incidence rate is 1.5%, this news service reported on 14 May. While a challenge trial design could be considered to quickly gather protection data, such a trial design also has its own operational and ethical issues.

Covid-19 vaccines: Human adenovirus vectors need to overcome neutralising antibody obstacle

The advantage of using adenovirus vectors is that the vector has surface proteins that help facilitate the delivery of the SARS-CoV-2 nucleic acid into the host cell, which then transcribes it to become viral antigens, explained Dr Hildegund Ertl, professor, Vaccine and Immunotherapy Center, The Wistar Institute, Philadelphia, Pennsylvania. In contrast, mRNA and DNA plasmid vaccines do not have comparable proteins to do so, noted Ertl. Moderna’s mRNA-1276 is lipid-nanoparticle-encapsulated (LNP); LNPs enter the host cell via endocytosis. Inovio’s INO-4800 requires a hand-held device to insert the DNA plasmid into the cell.

Adenovirus vectors also trigger a stronger inflammatory response compared to mRNA, DNA and protein subunit vaccines, Ertl noted. A way to differentiate between Chengdu, China-based Clover Biopharmaceuticals’, Novavax’s and IMV’s protein subunit vaccine is via their choice of adjuvant, this news service reported 26 May.

Adenovirus vectors can be divided into human and nonhuman adenovirus groups, Ertl said. AstraZeneca/Oxford’s AZD1222 uses a chimpanzee adenovirus vector and ReiThera’s is based on a gorilla adenovirus vector. CanSino’s Ad5-nCoV and J&J’s vaccine utilise human adenovirus types 5 (Ad5) and 26 (Ad26), respectively.

The disadvantage of human adenovirus vectors is that such adenoviruses are present in the community, Ertl noted. Hence, a significant part of the population would have pre-existing neutralising antibodies against the vaccine vector, said Don Diamond, PhD, professor, Department of Hematology & Hematopoietic Cell Transplantation, City of Hope Comprehensive Cancer Center, California. The vaccine would thus be neutralised before it has the chance to deliver its payload, Ertl explained.

About 40% of people in the US and Europe may have neutralising antibodies against Ad5, with some 50% in China and 80% in Africa, Ertl noted. In the China-based Phase I Ad5-nCoV trial, 44–56% of participants divided into three dose groups had pre-existing Ad5 neutralising antibodies (Zhu FC et al., Lancet. 2020 Jun 13;395(10240):1845-1854).

The attraction behind the Ad26 vector is that the adenovirus is not as prevalent, noted a virologist, who has investigated human adenoviruses as vectors. Yet, noted Ertl, Ad26 is still very common in Africa.

Because Ad5-nCoV and J&J’s vaccine are administered via intramuscular injection, concerns about pre-existing neutralising antibodies may not be as severe, said a second virologist, who has investigated adenovirus vectors in oncology. The vaccine’s vectors would primarily be contained in the injection site, limiting the negative impact of neutralising antibodies, he explained. However, neutralising antibodies disperse throughout the body, said the first virologist.

Phase I Ad5-nCoV data showed high levels of neutralising antibodies, which is not surprising considering adenovirus-vectored vaccines are potent, explained Dr David Curiel, director, Biologic Therapeutics Center, Washington University School of Medicine, St Louis, Missouri. ELISA antibodies and neutralising antibodies reportedly increased significantly at day 14 and peaked 28 days postvaccination.

But the overall immune system effect is attenuated by pre-existing neutralising antibodies, noted Curiel. Phase I data demonstrated that pre-existing Ad5 immunity could slow down rapid immune responses to SARS-CoV-2 and lower the peak of the responses, particularly for humoral immunity. Future human data should underscore IgA and IgG titres, the first virologist said. IgA is found in the mucosal surface and is relevant for protection, and while IgG has the same function, IgG concentration needs to be higher than IgA for protection, he explained.

The Phase I data’s 14- and 28-day time points postvaccination is logical, as there is a need for an early signal for efficacy, the first virologist said. However, it is still unknown if the response is temporary or sustained, both virologists noted. CanSino’s Phase II is investigating the low (5×10^10 viral particles) and middle doses (1×10^11), which had better safety profiles.

That said, it is understandable that the Ad5 vector is under investigation, as it has gone through many clinical trials through the years, the second virologist said. Yet, Diamond noted, in some studies, like in HIV, the placebo arm outperformed the vaccine group. No Ad5-vectored vaccines have been approved for Covid-19; the only such vaccine is for Ebola, but it is only sanctioned in China for emergency use and stockpiling.

With regards to the human-based adenovirus vector being leveraged by J&J, Ad26 is not as immunogenic as Ad5 based on available data, the second virologist said. Ertl agreed based on unpublished results from her laboratory, which produced its own Ad26 vector and compared it with Ad5. J&J’s vaccine features Janssen Pharmaceuticals’ AdVac platform. A Janssen spokesperson said it is too early to speculate on the immunogenicity of an Ad26-based vaccine versus an Ad5-based vaccine.

The attraction behind a less immunogenic vector is that it overcomes issues of pre-existing neutralising antibodies, but to the detriment of overall vaccine efficacy, Diamond said. However, the first virologist noted, increasing vector dose or adding immunomodulatory molecules could help boost the vaccine’s immunogenicity. The vector could also be engineered to induce higher immunogenicity and reduce pre-existing neutralising antibody impact, Curiel added. The Janssen spokesperson said AdVac’s vectors have been genetically modified so that it can no longer replicate in humans and cause disease. CanSino, AstraZeneca and ReiThera did not respond to a comment request.

AZD1222 data underwhelming; second injection might be needed

Meanwhile, the allure of using nonhuman adenovirus vectors such as those in AstraZeneca’s and ReiThera’s Covid-19 vaccines is that people would not have neutralising antibodies against the vector, experts said.

However, AZD1222 rhesus macaque data are not overwhelming, said Ertl and Diamond. Viral load in the lower respiratory system was decreased, but a reduction in viral shedding from the nose was not observed (van Doremalen N, et al., bioRxiv. Preprint. 2020 May 13). Furthermore, infectious diseases can resolve spontaneously in monkeys, which makes the results hard to interpret, Diamond added.

If the animal data are replicated in people, AZD1222 may only reduce disease deterioration but not reduce disease spread, the second virologist said. This is an issue considering herd immunity is a critical element of vaccine use, Ertl explained. If AZD1222’s animal data are consistent in humans, perhaps the asset would be best repositioned as a therapeutic vaccine, the second virologist said.

Perhaps AZD1222 only featuring SARS-CoV-2’s spike protein may not be enough, said Ertl, who is an investigator on a preclinical chimpanzee adenovirus-vectored vaccine carrying both the spike protein and another protein on the SARS-CoV-2 envelope. While vaccines that code for the spike protein is logical, there is still the chance the viral envelope may be more immunodominant, this news service reported 16 April.

Another potential way to improve AZD1222’s animal data in humans is to have another dose administered after about four months, Ertl said. AZD1222 is under investigation for single administration. Four months allows the immune system to ease because a booster too soon does not provide much additional efficacy benefit, Ertl said.

But the drawback with adenovirus-vectored vaccines, in general, is that one can get a maximum of two doses over a lifetime; any more than that and the immune system could generate neutralising antibodies to attack the vector, said Dr Kathryn Edwards, professor of paediatrics, Division of Infectious Diseases, Vanderbilt University School of Medicine, Nashville, Tennessee. Perhaps a different type of vector could be used for the second injection, the first virologist noted. Even using a different vaccine technology, like a DNA plasmid vaccine, for the booster shot could be considered, Curiel said.

Additionally, adenovirus-based vaccines typically need refrigeration, added Curiel, who is investigating an ovine adenovirus-vectored, thermostable vaccine designed to overcome cold chain issues underscored by the pandemic.

CanSino has an $18.19bn market cap, while J&J’s and AstraZeneca’s are $380.75bn and $140.16bn, respectively.

Reynald Castaneda is as Associate Editor for Clinical Trials Arena parent company GlobalData’s investigative journalism team. A version of this article originally appeared on the Insights module of GlobalData’s Pharmaceutical Intelligence Center. To access more articles like this, visit GlobalData.