To include your compound in the COVID-19 Resource Center, submit it here.

Vaccine prospects for COVID-19: learnings from a 40-year biotech journey that’s still in progress

Guest Commentary: Why the first vaccines for COVID-19 may not turn out to be the best

BioCentury is providing this content for free given the urgent need for information about the coronavirus crisis. Further analysis can be found in our COVID-19 Resource Center. For more, sign up for our daily email.

I have been in biotech research one way or another for over 40 years. This strange time of quarantine and self-isolation during the COVID-19 outbreak has given me time to reflect on how lessons from past vaccine programs bear on the current moment.

While it’s been rewarding to see many of the technologies I grew up with leading to vaccine candidates against the new virus, what’s often missing from the conversation is a discussion of antigen selection and the quality of the immune response required to confer meaningful protection.

Questions such as “will vaccines using just the spike protein be sufficient to induce protective immunity,” whether delivered as DNA, RNA or as a subunit protein-based vaccine; and “will it be necessary to use live attenuated or killed whole virus-based approaches to obtain the best and most protective immune responses” have been exercising my mind with parallels from previous vaccine efforts over the years.

In 1978, at the Animal Virus Research Institute (AVRI) in Pirbright in the U.K., we were cloning foot and mouth disease virus RNA using the then-new recombinant DNA methods.

Previous data had indicated that isolated VP1, one of the virus capsid proteins, could elicit neutralizing antibodies when inoculated into animals. Our approach was to make VP1 in E. Coli and use it as a subunit vaccine.

The project failed. The reason was that the VP1 “vaccine” did not elicit neutralizing antibodies. This could have been predicted given that a naturally occurring nucleic acid-free virus like particle did not elicit neutralizing antibodies either.

We were not alone in our efforts: both Genentech Inc. (now part of Roche (SIX:ROG; OTCQX:RHHBY)) and Amgen Inc. (NASDAQ:AMGN), relatively new biotech companies at that time, had similar FMDV vaccine programs. We all learned a lesson - the use of a single antigen may not be sufficient to induce an effective immune response.

“Just focusing on the spike protein or even more reductionist, the receptor binding part of the spike protein, may not be enough.”

Tim Harris, Repertoire Immune Medicines

Since then, recombinant DNA methods have been used to make many subunit vaccines, demonstrating that the approach can work for some viruses.

One of the first was for hepatitis B. Here the hepatitis B virus surface antigen (HbSag) was cloned and expressed in yeast, purified and appropriately formulated. Three shots of this vaccine protected 90% of the people inoculated against hepatitis B and it is still the vaccine of choice today.

Yeast became the preferred organism for the expression of many subunit vaccines, notably the highly successful and efficacious human papilloma virus (HPV) vaccine. Here, one of the capsid proteins (L1) was expressed and found to self-assemble into a virus-like particle which -- unlike the FMDV equivalent -- elicits highly neutralizing and protective responses to HPV infection. This vaccine has fundamentally changed the picture in terms of the frequency of the occurrence of cervical cancer and oropharyngeal squamous cell carcinoma.

The take-home message is that for some viruses, expressing a subunit protein and formulating it into a vaccine can give protective immunity. But we do not yet know whether this will be true for SARS-CoV-2.

At Pirbright, we also worked on Vesicular Stomatitis virus (VSV-an equine pathogen) and rabies viruses, both members of the Rhabdovirus family. Rabies virus vaccines are important in their own right, but the ability to use genetically engineered VSV to express the coat protein of other viruses has turned out to be very important for the development of several vaccines, including lethal pathogens such as Ebola virus. Although the VSV version of Ebola vaccine is replication competent, that feature is not a requirement for all viral vector vaccines -- others, including that for Ebola, that are based on replication-deficient adenoviruses, are also effective.

Several companies have pioneered the use of nucleic acid-based vaccines and the progress here has been most impressive. One of the major advantages of these modalities, whether they are DNA delivered by adenovirus derivatives or modified mRNA formulated in liquid nanoparticles, is the speed with which new vaccine constructs can be synthesized, formulated and made ready for use.

Several of the vaccine front-runners for COVID-19 are mRNA or DNA-based spike protein constructs.

I never imagined that mRNA would be a good way to make proteins in people -- either for vaccines or for therapy. In fact, I spent most of my time in the lab trying to keep the RNAs I was working with in one piece. The development of antisense oligonucleotides using modified nucleotide linkages that protect both RNA and DNA from rapid degradation have fundamentally changed the picture. Modified RNAs for example, are now stable in cells long enough to be translated into protein.

When the related coronavirus diseases SARS and MERS first arose, a good deal of effort was spent using these various technologies to make vaccines. Though the vaccines never went all the way through development, the efforts are now proving worthwhile for COVID-19.

These different vaccine approaches are being pursued by well over 50 companies and many academic groups, to make potential vaccines to produce neutralizing antibodies against the virus (see BioCentury’s COVID-19 Resource Center).

Unfortunately, there is a lot of hype about the likelihood of success of many of these approaches. Some “vaccines” will elicit neutralizing antibodies - maybe all will - but it is the quality of the immunity that is important, as defined, for example, by how long the immunity lasts and what virus antigens is it directed against.

For COVID-19, it could well be a case of “tortoise and hare.”

Tim Harris, Repertoire Immune Medicines

Just focusing on the spike protein or even more reductionist, the receptor binding part of the spike protein, may not be enough to confer long-term protection. Data from the previous SARS outbreak suggests that immune responses to more than one antigen are required for the induction of long-term immunity.

There is also a relevant debate about what kind of protective immunity a patient attains if infected with the virus, and how viral load correlates with severity of disease, reflecting just how little is known about the pathology of this new virus.

It is not always understood that the quality of the immune response elicited by a vaccine very much depends on the CD4+ and CD8+ T cell responses that the vaccine invokes. Extensive characterization of the immune responses of the leading SARS-Cov-2 vaccines has not been reported, and given the compressed development timelines of these vaccine candidates, such characterization may not be possible before entering the clinic. For COVID-19, it could well be a case of the tortoise and the hare, where the nucleic acid-based vaccines first out the gate do not end up being the ones that provide the level of long-term immunity required.

All the effort is necessary, and to be appreciated and applauded, but at the same time it is important to be realistic about the prospects: the reality is that by necessity, the vaccines are being rushed into trials without a solid grasp on the immune responses that they elicit.

In my present position working Repertoire Immune Medicines, some 46 years since I started at Pirbright (and over 50 years since I did my first RNA extraction), I hope to contribute and learn more, using the company’s platform to decode the T cell responses to infectious agents as well as to cancers. This approach uses yet more new technology, including single cell sequencing platforms, to uncover the peptides responsible for both HLA class I and class II mediated responses. We plan to apply these methods to people infected with the virus, and to compare what we see to the T cell responses in vaccinated people as a measure of the quality of the immune response induced.

Little did I know back in 1974, that the technologies that I have grown up with would turn out to be so important in helping to eliminate COVID-19.

The speed of response of the biopharma industry all over the world to control the disease and those suffering from the infection has been nothing short of amazing. That is one of the reasons that it is a privilege to have been on this biotech journey and still to be traveling on it.

Tim Harris is EVP of corporate development at Repertoire Immune Medicines Inc. and a venture partner at SV Health Investors.
Signed commentaries do not necessarily reflect the views of BioCentury.

Trial Subscription

Get a two-week free trial subscription to BioCentury


Article Purchase

This article may not be distributed to non-subscribers