The Drug Controller General ofIndia’s (DCGI) rushed approval on January 3 for Restricted Emergency Use (REU)—also sometimes referred to as Emergency Use Authorisation (EUA)—of two COVID-19 vaccines— Covishield of the Serum Institute of India (SII), Pune, and Covaxin of Bharat Biotech (BB), Hyderabad—has revealed once again that politics and not science continues to be in the driving seat in the fight against the pandemic in the country. Not surprisingly, the approvals for both have generated much controversy and criticism.
To recall, Covishield (or its original ChAdOx1 nCOV-19 or AZD1222) is a vectored vaccine. It uses a weakened recombinant Chimpanzee adenovirus (which does not infect humans) vector into which the gene for the Spike protein of the virus SARS-CoV-2, the key component of the virus that enables its entry into human cells, has been engineered. Once injected, the human host begins to produce the Spike protein, triggering an immune response against the antigen. The vaccine is produced by the British-Swedish multinational AstraZeneca on the basis of Oxford University’s technology, which has been transferred to the SII.
While adenovirus-vectored vaccines have been developed for other diseases, none of them has been licensed so far for public use. Adenovirus-vectored vaccines for COVID-19 are the first ones to be licensed for public use. The Russian Sputnik-V has already been licensed for public use in Russia; AZD1222 has recently been licensed for public use in the United Kingdom and Argentina on the basis of Phase-3 results in the U.K., Brazil and South Africa. With the January 3 approval, India will be the next country where it is deployed for public use, albeit in a limited way for now.
Covaxin (also called BBV152) is an adjuvanted inactivated whole SARS-COV-2 virion vaccine developed indigenously by Bharat Biotech in collaboration with the National Institute of Virology (NIV), Pune, of the Indian Council of Medical Research (ICMR). This, being essentially the whole virus particle, is expected to generate an immune response to all the viral proteins, not just the Spike proteins, including the envelope- (E), the membrane- (M) and the nucleocapsid- (N) proteins.
As required by the instituted procedure, the DCGI acts on the recommendations of the Subject Expert Committee (SEC) of the Central Drugs Standard Control Organisation (CDSCO) for approving the use of vaccines under development, particularly for REU. On January 1 and 2, the SEC examined the data of vaccine trials submitted by the SII and Bharat Biotech respectively and recommended that the vaccines be approved for REU. Yet, neither of the companies had met the requirements that the SEC had itself stipulated while examining their applications during meetings in December 2020 and earlier.
On January 1, the SEC gave its green signal for Covishield, the Indian version of AZD1222, even though the SII had not completed its bridging Phase-3 trials in 400 enlisted volunteers and submitted data to the SEC. Significantly, the bridging studies in India were only meant to generate safety and immunogenicity data, not efficacy. So, calling it a bridging Phase-2/3 study is a misnomer; it is actually only a Phase-2 study. This approval was immediately followed up the next day with an okay for Bharat Biotech’s vaccine BBV152 (branded Covaxin), although the company was still in the process of recruiting volunteers for its scheduled Phase-3 trials in 25,800 subjects. It had until then recruited only 22,500 volunteers.
Permission to the SII to conduct Phase-2/3 trials in India was conditioned upon the clinical data thus generated to be considered along with the outcome of the clinical trials conducted by Oxford University in the U.K. and elsewhere. The Indian clinical trials were to be conducted in 1,600 individuals. Since the SII’s trials were intended to generate only safety and immunogenicity data, they would ride piggyback on Oxford University’s efficacy data from its Phase-3 trials. Of the 1,600, a cohort of 1,200 was to be vaccinated for generating safety data and the remaining cohort of 400 for safety and immunogenicity data. It is reliably learnt that when the approval was granted, the SII had presented immunogenicity data from only 185 of the 400 people recruited for the trial.
Lacking: efficacy data from within India
Thus, the SII’s Covishield got the approval on the basis of its submission of safety and efficacy data from Phase-2/3 trial results conducted by Oxford and AstraZeneca in the U.K., Brazil and South Africa, but without adequate bridging study data on the impact of the vaccine on the Indian population. And, Bharat Biotech’s Covaxin got the go-ahead on the basis of safety and immunogenicity data from its Phase-1/2 trials but no Phase-3 data. That is, neither of the approved vaccines has generated efficacy data within the country.
While the SII has been permitted to take refuge under efficacy data from elsewhere, Bharat Biotech is yet to complete its Phase-3 trials, which began only in November 2020. According to Bharat Biotech, nearly 6,000 people have already been given the two doses (to be administered four weeks apart). However, any efficacy data cannot be generated right now because Phase-3 trials have been given to an independent external agency. Analysis to generate even interim data from the Phase-3 trials could take another couple of months. Their primary endpoint, according to the Clinical Trial Registry-India (CTRI), is reaching 130 cases of COVID-19 cases/infections among the participants of the trial.
Criticism from scientific community
The approvals have naturally given rise to a lot of criticism from the scientific community and public health activist groups against the government’s violation of its own guidelines for the roll-out of COVID-19 vaccines. As one had suspected , there seems to have been a political push behind this announcement to time it with the Republic Day celebrations so that the Prime Minister can proclaim it as a major achievement of the regime. Of course, if British Prime Minister Boris Johnson had not cancelled his visit as the Chief Guest in the Republic Day function, both the leaders would have hailed the approval of Covishield as a great milestone in India-U.K. relations and a result of their joint fight against the pandemic.
SEC did not adhere to its own targets
The full minutes of the SEC meetings are not in the public domain. But summary statements have been released on meetings in December 2020 and January 2021, which are relevant for the present discussion. These give an idea of how the SEC did not adhere to the targets that it had set for the companies and how it changed its decision in record time.
At the meeting of December 9, the SEC noted that the SII had submitted safety data until November 14 only. It asked the SII to provide (1) updated safety data of Phase-2/3 trials in the country; (2) immunogenicity data from the clinical trials in the U.K. and India; and, (3) the outcome of the (then ongoing) assessment of Medicines and Healthcare Products Regulatory Agency (MHRA) of the U.K. As for Covaxin, following Bharat Biotech’s submission of the safety and immunogenicity data from its Phase-1/2 trials, the SEC after detailed deliberations recommended that the company should present the safety and efficacy data (emphasis added) from its ongoing Phase-3 study for further consideration of its request for REU. At the next meeting on December 30, the day when Britain’s MHRA gave the thumbs up for emergency use of AZD1222 vaccine in the U.K., the SEC told the SII that it should submit the complete details of the conditions and restrictions under which AstraZeneca was granted EUA in the U.K. along with revised information on its use in the Indian context.
The regimen for the vaccine that the MHRA approved was two full doses (which have shown 62 per cent efficacy in trials), with the second dose to be given between four to 12 weeks after the first. This aspect of the approval for a range of dosing gap period has generated quite a bit of controversy within the U.K. and elsewhere. In the original Phase-1/2 trials, the dosing schedule was 28 days apart, and this regimen was found to be effective in preventing symptomatic COVID-19 from occurring more than 14 days after the second dose, in particular severe cases of infection, or hospitalisation. The Food and Drug Administration (FDA) of the United States has, in fact, made it clear that it will not approve any prime-boost dose interval period that has not gone through trials. We will return to this later.
In its interim efficacy analysis, Oxford/AstraZeneca pooled the results of two dosing regimens—one involving two full doses (in 8,895 participants) showing 62 per cent efficacy, the second involving a half dose priming inoculation followed by a full dose (in 2,741 participants) showing 90 per cent efficacy. Its November 23 announcement of the vaccine’s efficacy of 70.4 per cent was the result of the combined analysis of these two different regimens with the stated efficacy figure being the weighted pooled average. It will be noted, however, that the MHRA has not approved this varying prime-booster dosage regimen, although it has approved a questionable wide range of interval period between the prime shot and the booster shot.
Perhaps because of the urgency generated by the MHRA’s approval, the SEC met again on January 1 to consider the requests of the SII and Bharat Biotech for REU/EUA approval. The SII presented all the information that the SEC had sought on December 30. According to the SEC, the safety and immunogenicity data presented by the SII from the Indian study were comparable to that of the overseas data. How the SEC came to this conclusion is unclear as the SII has put out no data from its clinical trials in the public domain. Also, as we shall see, the dosing schedules of the U.K. Phase-3 trials and the Indian bridging Phase-2/3 trials are different. No information about the conditions and restrictions under which the vaccine was approved in the U.K is publicly available. But more importantly, as mentioned earlier, there is no efficacy data generated in India. Also, the claim of 70 per cent efficacy made by both the MHRA and the DCGI while approving the Oxford vaccine with the two full-dose regimen is wrong. With that dosage, the efficacy is only 62 per cent.
Conditions for administering vaccine
The SEC recommended grant of permission for REU of Covishield subject to various conditions. One of them is that vaccination should be restricted to individuals above 18 (as the trials in the U.K. have been only on people aged 18-55 years), and that the vaccine should be administered intramuscularly in two full doses with an interval of four to six weeks. The company was also required to submit safety, efficacy and immunogenicity data from the ongoing clinical trials nationally and internationally for review at the earliest. The basis for the wider than four-week window for prime-boost dosing interval is not clear as the trials in India had been approved only for a dosing schedule separated by 28 days.
The chief argument for delaying the second dose (up to 12 weeks) in the U.K. is that given the limited vaccine availability, a larger number of people can be vaccinated with a single dose. According to the U.K. Joint Committee on Vaccination (JCVI), modelling of the high level of protection afforded by the first dose suggested that initially vaccinating a greater number of people would prevent more deaths and hospitalisation than vaccinating a smaller number with two doses on a fixed four-week schedule. A counter to this argument by many scientists is that it is not known how the vaccine will perform with an increased gap between injections. Nor is it known how protective a single injection is over a longer term.
Professor Gagandeep Kang of Christian Medical College, Vellore, said: “Actually, even though the dosing schedule was supposed to be four weeks apart, the interval[s] went longer [during the trials] because of lack of availability of doses initially, so there are data included [in the published Phase-3 results] with longer intervals. Messy, but it appears that there is greater than 50 per cent efficacy with a single dose...”
Satyajit Rath of the Indian Institute of Science Education and Research (IISER), Pune, said: “It is an odd decision poorly based on rigorous evidence. That said, the Oxford-AstraZeneca Phase-3 trial data do seem to indicate that the first dose might be providing some protection for a few weeks…[So] it would be reasonable to delay the second dose a little so as to make the most of limited vaccine dose supplies.” But, in the Indian context, where there is a rapid decline in the number of cases (unless new variants cause a sudden surge), increasing the interval up to six weeks does not seem to bear any logic.
Efficacy data from U.K. not robust
Many scientists are questioning even the MHRA’s approval for AZD1222 because of the highly confusing manner in which the U.K. trials have been conducted and the data analysed to give efficacy results that are difficult to interpret sensibly. The Indian authorities have banked upon these dubious data to give approval for Covishield. The Oxford/AstraZeneca trials in the U.K. and elsewhere assessed the efficacy of the vaccine on the basis of 131 COVID infections among 11,636 participants, merging data across two trials from the U.K. and Brazil. The All India Drug Action Network (AIDAN), a non-governmental organisation (NGO), pointed out that “in these data there are wide variations in the duration of the interval between dosing and also differences in the strengths of the vaccine doses. There are differences in the trial designs of the U.K. and Brazil trials.”
Paul Offit, a professor at the paediatrics department of the Children’s Hospital of Philadelphia and a member of the FDA’s vaccine advisory panel, who was quoted in a recent media report, has said: “It is sloppy science and the data isn’t sufficient for efficacy assessments…The trial involved two different dosing regimens, two different time intervals between the two doses, and two different placebos…You shouldn’t combine such data, but they did it.” According to him, the FDA would not approve the AstraZeneca-Oxford vaccine on the basis of the existing data. The strategy of using an adenovirus to generate immune responses against the novel coronavirus may be perfectly viable, but the vaccine would need to be shown as effective through trials involving 30,000 to 40,000 participants, Offit added.
In India, there have been echoes of this sentiment. In a statement addressed to the DCGI (and two others) after the SEC meeting of December 30, AIDAN cautioned the health officials that there were important differences between the SII’s Indian trial and Oxford/AstraZeneca’s foreign trials. It also urged that the SII should be asked to submit the interim safety and immunogenicity data for all the participants. This was crucial for the assessment of the vaccine candidate in an Indian population and the purpose of the bridging study, it said.
AIDAN also pointed out that while the SII’s safety and immunogenicity trials involved two full doses of the vaccine given with an interval of four weeks, there was no corresponding efficacy analysis for the same dosing regimen in the published data of U.K. and Brazil Phase-3 trials for the Oxford/AstraZeneca vaccine. “However,” the NGO’s statement said, “an efficacy analysis for two standard doses administered at an interval of less than 6 weeks shows efficacy of 53.4 per cent based on analysis of 28 cases (in 3400 persons aged 18-55 years) in the U. K. and Brazil trials. While this efficacy may not be particularly robust due to the small sample, it would fail to meet the criteria set out by the WHO for evaluation of COVID-19 vaccines for either prequalification or for Emergency Use Listing (EUL) and draft guidelines of the CDSCO [of September 20, 2020]…it certainly raises the need for more data.”
The above observation, that the available efficacy data (from the U.K. and Brazil trials) are statistically not robust for the interval period of four to six weeks, puts a further question mark on the approval for Covishield. Indeed, two members of India’s National Task Force on COVID-19 have been reported to have said that the approval reflects a concession to poor science and incomplete data. One of the members observed that the approval was an “opaque process”.
As regards Bharat Biotech’s Covaxin, following the company’s presentation of additional data including that of serious adverse events (SAEs) during the trials and data from the ongoing Phase-3 trials, the SEC observed that Covaxin, being an inactivated whole Virion Coronavirus Vaccine, had the potential to target mutated coronavirus strains. “The data generated so far,” the SEC noted, “demonstrates a strong immune response (both antibody as well as T-cell) and in vitro viral neutralisation…However, efficacy is yet to be demonstrated.” In the end, however, the committee recommended that the firm should try to expedite the recruitment and may perform interim efficacy analysis for further consideration of restricted emergency use approval (emphasis added). This observation was overturned soon after in the SEC’s meeting of January 2. Having decided one way, at whose behest the SEC thought it necessary to meet again so soon to discuss Bharat Biotech’s proposal and approve it is not clear.
At the meeting with the SEC, Bharat Biotech seems to have laid stress on its vaccine’s potential—but not demonstrated—to target mutated virus strains (which the SEC, too, had noted in the previous meeting) and sought REU approval. The SEC, in its summary statement on the meeting, said, “[T]he firm further presented the [ sic ] updated data, justification and requested for consideration of their proposal in the wake of incidence of new mutated corona virus infection [emphasis added]… The ongoing clinical [Phase-3] trial is a large trial on 25,800 Indian subjects in which already 22,500 subjects have been enrolled including subjects with comorbid conditions…which has demonstrated safety till date. Moreover, firm has presented the safety and efficacy data from non-human primate challenge study where the vaccine has been found to be safe and effective. In view of above, after detailed deliberation, the committee recommended for grant of permission for restricted use in emergency situation in public interest as an abundant precaution, in clinical trial mode, to have more options for vaccinations, especially in case of infection by mutant strains. Further, the firm shall continue the on-going Phase-3 clinical trial and submit data emerging from the trial as and when available.”
Following the SEC’s recommendations of January 1 and 2, the DCGI, V.G. Somani, granted final approval for the release of the vaccines for REU immediately on January 3, adding his bit about the vaccines to justify the decision. In his press statement, he said, “This vaccine [Covaxin] is developed on vero cell platform, which has well established track record of safety and efficacy in the country and globally. The firm has generated safety and immunogenicity data in various animal species…and also conducted challenge studies on non-human primates (Rhesus macaques) and hamsters…Phase-1 and 2 clinical trials were conducted in approximately 800 subjects and the results have demonstrated that the vaccine is safe and provides a robust immune response.” However, according to the published data of the vaccine BBV152, Phase-1 and 2 trials were conducted only in 375 and 380 participants respectively, and not 800 as the DCGI has claimed.
The SEC’s approval for BBV152 of Bharat Biotech, following a turnaround in its stance, and that too without even interim results of the ongoing Phase-3 trials, has naturally generated a huge controversy. Balram Bhargava, Director-General of the ICMR, was quoted in a news report as having said: “We are in a war-like situation; we need innovative strategies.” The NIV, an ICMR lab, has provided the SARS-CoV-2 virus strain and technology for the vaccine BBV152. “Safety, immunogenicity and efficacy are required in a non-emergency situation,” Bhargava said. “The existing pandemic, the mortality, available science, and lack of definitive treatment were considered [by the regulators] for accelerated approval,” he added.
Bharat Biotech’s defence
Bharat Biotech’s arguments in defence against the mounting criticisms are essentially the following:
There are precedents of approving a vaccine based on Phase-2 trial data on safety and immunogenicity alone in emergency situations here and elsewhere, as during the Ebola and H1N1 outbreaks.
The New Drugs and Clinical Trials (NDCT) Rules of March 2019 allow for such an approval for REU. (For criticisms against this argument, see accompanying article by S. Srinivasan.)
Bharat Biotech’s Phase-2 data on high antibody titres against various viral proteins, in particular neutralising antibodies and animal challenge studies, are “surrogate endpoints” for efficacy.
The whole virion vaccine BBV152 has the potential to prevent infections caused by new SARS-CoV-2 strains as well. (The criticism against this is that this argument is at best heuristic; it has not been demonstrated against any variant of SARS-CoV-2, for example, the variant with D614G mutation.)
They have presented safety data of nearly 25,000 people; no serious adverse events (SAEs) to be concerned about.
No data on Covishield in public domain
While Bharat Biotech has put out all its Phase-1 and 2 data in the public domain, the SII is yet to put out any data about Covishield’s trials in India.
One of the main criticisms has been that, by approving a vaccine without any efficacy data, the CDSCO has gone against its own “Draft Regulatory Guidelines for Development of Vaccines with Special Consideration for COVID-19 Vaccine” of September 2020. The document clearly says, “To ensure that a widely deployed COVID-19 vaccine is effective, the primary efficacy endpoint point estimate for a placebo-controlled efficacy trial should be at least 50 per cent…”
Jacob John, formerly of the CMC and one of the foremost virologists of the country, has said in an interview that in an emergency pandemic situation safety is of utmost importance and he would prefer a vaccine that has robust safety data rather than worry about efficacy figures. So, between Covishield and Covaxin, he would prefer Covaxin.
Critics, however, have found the imprecise use of language and choice of phrases in the approval statement baffling. Both the SEC and the DCGI stated that the BBV152 vaccine would be given in the “clinical trial mode”. In an interview to The Times of India on January 4, Dr Kang said: “I can understand what they [SEC/DCGI] are saying about the Covishield vaccine and then they have this very complicated language for Covaxin. They say it will be given in clinical trial mode. What does that mean? I have no clue. I have never seen anything like this before…Either you are doing a clinical trial or you are not. I am confused.” Indeed, given that the government’s vaccine roll-out plan is to first vaccinate priority groups, including health-care workers, totalling about 30 crore, this so-called “clinical trial mode” cannot be a blinded randomised trial. Also, which is the control group that will get the placebo? These questions have not been spelt out at all by the SEC or the DCGI.
Besides the political lobbying within the Bharatiya Janata Party (BJP) regime, underlined by the “Atmanirbhar” slogan for the indigenous vaccine, a corporate war between the two companies competing for a share of the Indian market is evident. In a situation where COVID-19 spread appears to be on the decline, it could take a long time before Bharat Biotech can attain the primary endpoint of 130 cases, and it could end up losing a good share of the market. And, aided by the political push, Bharat Biotech seems to have played the card of protection against emerging and more infectious strains of SARS-CoV-2 as well.
Where does all this controversy leave the general public? Unfortunately, this may only end up increasing the number of vaccine-hesitant people, even among the priority groups identified, such as healthcare workers.