Virologist Shahid Jameel says there is no experimental evidence so far that Covaxin will neutralize newly emerging mutants any better
Shahid Jameel, Virologist, and Director, Trivedi School of Biosciences, Ashoka University said different concerns existed regarding both approved vaccines in India adding that it was premature to assume they would be useful against the 'UK strain'.
We've seen two Indian vaccines approved. Covishield by Serum Insitute of India has submitted — but not published — any data on its Indian trial. Bharat Biotech has published safety and immunogenicity data (preprint) for Covaxin but hasn't generated the all important efficacy data. Do you think the regulator has made a judicious call in approving both?
As more details trickle out, it appears that the fear of a second wave, propelled by more contagious viruses of the B.1.1.7 lineage (popularly called UK variant), weighed heavily on the Drug Controller General of India (DCGI) and the regulatory committees. Since what was presented by the companies to them is not available in the public domain, it is hard to second guess.
But it is clear that the process, initially laid out by DCGI, was compromised. Vaccines are a product of science, and there is a methodology and process to science. It is based not on beliefs but on data and evidence, which is found to be lacking in this case.
The relative confidence in Covishield rests on efficacy studies in the UK and Brazil. However there were multiple problems pointed in these trials. Some vials of vaccine had less than intended dosage leading to varying efficacy estimates. The U.S. FDA too hasn't cleared the vaccine in their country. Are these matters of concern?
Yes, these should be matters of concern. You cannot combine Oxford/AstraZeneca efficacy data from Brazil and UK together — 62% and 90% as a weighted average of 70.4% — just to cross the threshold. The doses were different. The concern is indeed highlighted by the US Food and Drug Administration (FDA) not clearing this vaccine as yet. In India if two full doses of Covishield are to be used, the efficacy would be 62% not 70.4%.
The reasons cited by the Indian government, for approving vaccines, inspite of incomplete evidence, is the threat by “mutant strains,” and having “a backup”. Is there evidence that an inactivated whole virus might perform better than sub-unit recombinant ones? What does the history of virology suggest?
Firstly, there is no “backup” vaccine. A vaccine is a vaccine. Regulators never approve something as a backup. Every product must be safe and efficacious for everyone to take. And that data is still lacking for Covaxin.
While it is a theoretically possibility, there is thus far no experimental evidence that Covaxin will neutralize the newly emerging mutants any better. As opposed to Covishield that makes only the Spike (S) protein, Covaxin contains all viral particle proteins, including the Nucleocapsid (NC) protein. Antibodies are mainly generated against the S and NC proteins, and neutralizing antibodies only against the 'S.'
The Covaxin Phase 2 data shows a rising antibody titre against both S and NC proteins after the first and second dose. The point to understand is that only the S protein is displayed on the virus surface and therefore, only anti-S antibodies will neutralize the virus; the anti-N antibodies would have no role.
Where the NC protein would help is to generate better cytotoxic T cell responses that would efficiently eliminate virus-infected cells. Being a killed vaccine, Covaxin is unlikely to raise good cytotoxic T cell immunity; live/replicating viral vaccines do that much better. While cell-mediated immunity data is shown for Covaxin, it is not clear how much is being contributed by the NC protein. Thus, while it is a theoretical possibility, the jury is still out on whether the NC protein is providing added benefit to Covaxin over other vaccines based only on the S protein.
Historically, for another respiratory virus – the Respiratory Syncytial Virus (RSV) inactivated virus vaccines were found to produce antibodies to the surface protein that exacerbated disease in children who were naturally infected following vaccination. Cases of antibody-dependent enhancement have been reported after the use of inactivated vaccines against not just RSV, but also measles and in people with prior antibodies to dengue virus. This becomes a possibility when antibodies are developed to improperly folded proteins or proteins whose structure has been altered during inactivation of the virus with chemicals. Even though it is remote, this possibility has not been addressed by the data presented for the Covaxin phase 2 trial. It would have been addressed in Phase 3. Finally, Covaxin uses an adjuvant that has never been used in a licensed vaccine.
Bharat Biotech has expressed difficulties in recruiting people for trials. Does such apprehension exist globally or are there unique challenges in India?
Indian companies, Bharat Biotech included, are very good vaccine companies. These challenges are not on account of perceived quality. I feel this is because phase 3 trials in India came at a time when the outbreak had declined significantly. Had the Covaxin phase 3 trial started in August or September instead of November, there would have been no recruitment problems. It is both a matter of perception (reducing danger) as well as exclusion criteria. The latter require volunteers who have no exposure to SARS-CoV2 (i.e. antibody negative) and who also don’t share living quarters with someone who tested positive for the virus. Increasing exposure of the population makes the trial increasingly difficult.
I would not call inactivated viral vaccines as vestigial. Most effective vaccines in use today belong to that category. The inherent difficulty with this platform is that it requires viruses to be grown to high titres. That is often difficult and sometimes not possible. Take for example the hepatitis E virus on which I worked for three decades. It grows very poorly in culture and thus no inactivated vaccine has been possible.
Nucleic acid vaccines are a new platform and would be used for the first time with Covid vaccines. So far the results for mRNA vaccines look good and DNA vaccines are yet to be tested. But the long term safety for both is yet to be established. These would be platforms for the future simply because it is much easier to manufacture RNA and DNA than proteins. Viral protein antigens are best made by the body to ensure proper conformation and structure, which is often not possible in vitro. DNA, RNA and viral vectors (eg adenoviruses) use the cell’s machinery to make viral protein antigens.
In the pre-pandemic world, efficacy data from a phase-3 trial was non-negotiable for a vaccine. However exceptions have been made in the case of Ebola in Africa where a vaccine was rolled out because of high mortality. Does the state of the pandemic now in India justify an emergency rollout of an unproven vaccine?
Yes, Ebola vaccines were used without Phase 3 testing to vaccinate about 60,000 people in West Africa. But that cannot be cited as a precedence. The 2014-16 Ebola outbreak had 28,600 cases and 11,325 deaths, giving about a 40% fatality rate. The global fatality rate for Covid-19 is 2.15%; for India it is 1.44%. There is a big difference in the two situations. Further, India is on a continuous declining curve since mid-September. From a peak of about 95,000 per day (7-day average), we are now down to about 18,000 cases per day. What was the tearing hurry?
What are the possible harms from rolling out a vaccine, that in limited human trials, appears to be not harmful but hasn't been shown to be efficacious either?
The antibody-dependent enhancement effect needs to be ruled out as described above and that is better done with Phase 3 data. Otherwise, it is a matter of due process, which was subverted for speculative reasons. We don’t know how many doses of Covaxin would be ready by Jan 13 when vaccines roll out in India. For Covishield, we are aware of 50 million doses ready to roll out. I see no rational reason to compromise the approval process.
Given that 30 million healthcare workers will begin to be inoculated in the coming weeks, is it reasonable for an eligible healthcare worker, who now observes a declining trend in daily positive cases for months, knows that the true exposure is far more than PCR positive cases, to decline to be inoculated on the grounds that vaccine's efficacy analysis is incomplete?
That is a personal choice and it would not be proper to make vaccination mandatory. But HCWs and FLWs should also realize that they are at the highest risk and should get vaccinated. Further, considering the nature of their work, they also pose the highest risk to transmit to others. They should get protected. Most likely all of them will get Covishield. By the time we get to the remaining 270 million who are >50 years of age or those with co-morbidities, we will have efficacy data for Covaxin as well.
Given that India has a long history of administering childhood vaccinations, was a dry run for the COVID vaccine really necessary in that would it have genuinely given a taste of the potential challenges of actually administering a real vaccine to this nation of a billion plus?
Yes. I think this is a very good strategy by the government. India has a lot of experience under UIP [Universal Immunisation Programme], but the logistics of childhood vaccination are different from adult vaccination. For one, children are a captive population (parents bring them in) while adults are not. Secondly, we have never delivered vaccines to curb a pandemic. Any roll out lapses would adversely affect vaccine effectiveness and confidence. A dry run helps on both counts. Full credit to the Ministry of Health for thinking of this and doing it well.
Do you think the nature of vaccine and drug trials has now irrevocably changed forever? Will we ever go back to the regime of 5-10 years for a vaccine?
Yes absolutely. This is a real positive outcome of COVID —it has revealed to us the power of science.