Promising interim Phase 3 trial data a shot in the arm for acceptance of Covaxin

Following Covaxin’s promising interim Phase 3 results and the WHO’s recommendation on the use of Covishield for the elderly, the pace of vaccinations in India is likely to improve. So far, the total number of people who have been vaccinated is far fewer than half the number of doses the government has procured.

Sanitation workers receiving the first dose of Covaxin at the Hindu Rao Hospital in New Delhi on February 16. Health care workers and front-line workers, who constituted the first priority group to get vaccines, were not given a choice between Covishield and Covaxin.

The big story on the COVID-19 vaccine front in the country is the release on March 3 of the interim Phase 3 trial results of Bharat Biotech’s Covaxin (also called BBV152). The vaccine is based on an inactivated whole SARS-CoV-2 virus platform and was developed in collaboration with the National Institute of Virology (NIV), Pune, of the Indian Council of Medical Research (ICMR). According to these results, following the second dose, the vaccine has about 81 per cent efficacy in preventing COVID-19 in those without any prior infection. The trials are based on a two-dose regimen given 28 days apart. This is a good figure for efficacy, significantly higher than the threshold figure of 50 per cent that the Drugs Controller General of India (DCGI) requires for Restricted Emergency Use (REU) approval. But the DCGI gave Covaxin a conditional REU approval on January 3 when the vaccine did not have any efficacy data to show from its ongoing Phase 3 trials. The approval was strangely described by a new and unexplained phrase: “in the clinical trial mode”. This naturally triggered a lot of criticism from medical researchers and public health specialists.

‘Clinical trial mode’

In “clinical trial mode”, as one has come to understand it, those who get the vaccine have to sign an informed consent form as in a regular clinical trial. But it is not a clinical trial because there is neither randomisation of the participants nor is there a group that is administered a placebo, which serves as the control arm. However, in this mode, the vaccinees are tracked and monitored after vaccination, and in case of serious adverse events, the vaccine manufacturer is liable to pay compensation if the event is proven to be causally related to the vaccine. Also, in case of any adverse event, the company provides the standard care in designated health centres, as in the case of regular clinical trials.

As the ongoing vaccination drive has picked up pace and coverage, this could in significant part have played a role in creating apprehensions in the minds of the general public about accepting Covaxin. This may well be the reason for the vaccine hesitancy observed even among health care workers, the first priority cohort that was vaccinated when the vaccination drive began on January 16. Health care workers and front-line workers were not given a choice between the two vaccines approved for REU—Covishield of the Serum Institute of India and Covaxin—and were faced with a “take it or leave it” situation.

Stated to be the largest clinical trial in India, Bharat Biotech’s Phase 3 trials have recruited 25,800 participants 18-98 years of age who were serologically negative (to SAS-CoV-2), including 2,433 over the age of 60 and 4,500 with comorbidities. The trials are being done in association with the ICMR. As per the information on the Clinical Trials Registry-India, the primary end point of the trials was defined by the first occurrence of 130 PCR (polymerase chain reaction)-confirmed cases of symptomatic (mild, moderate or severe) COVID-19, with the onset of the disease occurring at least 14 days after inoculation with the second dose. Bharat Biotech has outsourced the trial to the external agency IQVIA, which specialises in conducting clinical trials globally.

According to Bharat Biotech’s press release of March 3, the first interim analysis was done on reaching 43 COVID-19 cases. Of the 43 cases, 36 were seen among the placebo group, which yields—on the basis of the number distribution between the vaccinated and the placebo groups—a point efficacy estimate of 80.6 per cent. The second interim analysis will be conducted on reaching 87 cases, and a final analysis will be done on reaching 130 cases. “The interim analysis,” the release stated, “included a preliminary review of the safety database, which showed that severe, serious, and medically attended adverse events occurred at low levels and were balanced between vaccine and placebo groups.”

The Central Drugs Standard Control Organisation (CDSCO) and the DCGI had clearly gambled in giving Covaxin an REU approval solely on the basis of Phase 1 (safety) and Phase 2 (immunogenicity, or the ability to elicit an immune response) data and had to trot out a convoluted rationale to defend their controversial decision. Even though the approval violated their own guidelines on minimum vaccine efficacy of 50 per cent, their statement said that Covaxin was being approved “for restricted use in emergency situation in the public interest as an abundant precaution, in clinical trial mode, to have more options for vaccinations, especially in case of infection by mutant strains”.

Political gamble

The political gamble, which it clearly was, now seems to have paid off. This probably will end the earlier public apprehensions about Covaxin and quell the surrounding controversy. But this should not in any way be taken to mean that the CDSCO and the DCGI should gamble with people’s lives in every such emergency pandemic situation. A proper REU approval would have meant only an additional six-week-long wait, which would have been nothing, especially after more than a year of a raging COVID-19 pandemic that has still not fully subsided.

More significantly, research by scientists of the NIV and Bharat Biotech has shown that Covaxin’s immunogenicity against the new variant of the virus that originated in the United Kingdom (called VOC/VUI 202012/01, or B.1.1.7), which is more transmissible and more infectious, is on a par with what was seen with the more widely prevalent original virus in its Phase 2 trial. The work, which was hosted on the web preprint server biorXiv on January 26, found that vaccine-induced antibodies could neutralise the U.K. variant and other heterologous strains of the virus (strains that differ from the one on which the vaccine is based).

According to the NIV-BB paper, the U.K. variant has 17 mutations (23, according to some reports), eight of which are in the receptor-binding domain (RBD) of the spike (S) protein on the viral surface, which mediates the binding of the virus to the ACE2 receptor on the human cell surface and enables its entry into human cells. The apprehension that vaccine candidates may not be efficacious against new variants stems from the fact that most recombinant vaccines, including the SII/Oxford-AstraZeneca’s Covishield/AZD1222, are designed to specifically target the S-protein.

Given that some of the new variants have been found to be more infectious and perhaps more virulent, it is suspected that some of these new mutations could escape the virus-neutralising ability of the antibodies generated by vaccines that target only the S-protein’s RBD. The hypothetical argument being made in favour of Covaxin’s efficacy against the new variants was that since it is a killed whole virus vaccine, it presents all the epitopes (antigenic sites on the pathogen that antibodies recognise) on the virus surface, not just the single epitope on the S-protein, for the immune system to mount a response. But this had to be demonstrated in some variants at least.

The NIV-BB team performed what is called a plaque reduction neutralisation test (PRNT) to measure the amount of (virus) neutralising antibodies (nAbs) the immune system produced in response to the virus. Unlike nAbs, not all antibodies can neutralise the virus; many just bind to it without actually being able to destroy it. Sera were collected from 26 recipients of Covaxin/BBV152 (made from the original homologous strain designated hCoV-19/India/2020770), which had been earlier found to have 99.6 per cent seroconversion rate with nAbs in the Phase 2 trials. The study performed a PRNT and measured the nAb titres against the U.K. variant isolated from infected Indians returning from the U.K. (strain designated hCoV-19/India/20203522) and a domestic variant strain (designated hCoV27-19/India/2020Q111). The sera of the 26 vaccinated individuals showed comparable neutralisation activity against the U.K. variant as well as the heterologous domestic strain, thus “dispelling the uncertainty of possible neutralisation escape” by the mutations, the paper said.

Until this demonstration, it was only a theoretical argument. This will certainly strengthen Covaxin’s place in the ongoing vaccination campaign because Covishield, or even the original Oxford-AstraZeneca’s (chimpanzee adenovirus) vectored vaccine (called ChAdOx1-nCOV, or AZD1222) on which it is based, is yet to demonstrate its efficacy against any of the newly emergent variants. Oxford-AstraZeneca’s trials in Brazil and South Africa seem to suggest that its vaccine is less efficacious against the new variants found in South Africa (called B.1.351) and Brazil (called P.1), all of which have a defining mutation called N501Y. The South African and the Brazilian variants have in addition another defining mutation called E484K. While on the basis of the above work, the NIV-BB paper stated that it was unlikely that the N501Y mutation would be able to dampen the potential benefits of its vaccine BBV152, a similar comparative analysis of nAb titres has not yet been done with variants having the E484K mutation, which has been reported to be a neutralisation escape mutation and would be of equal concern. However, given that Covaxin has been found to have comparable neutralising activity against the U.K. variant and two indigenous variants, it may well be equally efficacious against E484K and other potential escape mutations.

Covishield trials

The SII is yet to make public any data from its bridging Phase 2/3 clinical trials in India on 1,600 subjects that were aimed at studying only safety and immunogenicity—safety in 1,200 of them and safety and immunogenicity in the remaining 400—and not efficacy. The SII obtained its REU approval on the basis of data Oxford-AstraZeneca generated in its Phase 3 trials in the U.K., Brazil and South Africa and partial bridging (safety/immunogenicity) study data in India, which it submitted to the CDSCO. One of the limitations of the earlier Phase 3 trials of Oxford-AstraZeneca in the U.K., South Africa and Brazil was that the participants did not include those above the age of 55, and so its efficacy in older adults is yet to be demonstrated. However, a later blinded randomised and controlled Phase 2/3 (safety and immunogenicity) trial focussed precisely on this issue, and this study was published in The Lancet on December 19, 2020. The trials included 560 participants divided into three escalating age groups: 18-55 years (100+60); 56-69 years (120+40) and greater than/equal to 70 years (200+40), the two numbers in each age tier referring to the vaccinated and the placebo groups.

The Oxford-AstraZeneca research team found that while after the prime (first) dose vaccination the antibody responses measured on day 28 decreased with age, following the booster second dose on day 28, the antibody titres and T-cell responses were comparable between all the three age groups. Interestingly, the study found that those of age 56 and above were less reactogenic to the Oxford-AstraZeneca vaccine. The adverse reactions included local (injection-site) effects and mild systemic effects such as fatigue, headache and fever. No serious adverse event was seen among the elderly.

While the safety and immunogenicity profiles of the vaccine in the elderly are encouraging and indicative of possible good protective efficacy, this is yet to be measured in a Phase 3 trial. Also, the sample size in the Phase 2/3 trials targeting the elderly was small; the number vaccinated in the above 60 age group must have been only around 250.

Even in the ongoing Phase 3 trials, recruitment data show that only 1 in 10 were 65 years old or older, and interim efficacy data (up to December 7) show an efficacy of only 52 per cent. For this reason, many European countries decided not to use the AZD1222 vaccine for older adults. Germany, in fact, is reportedly yet to pick up the entire quantity of the vaccine committed to it under a bilateral agreement. But these countries, where the demography is skewed towards the elderly, have chosen to use the Pfizer/BioNTech or Moderna mRNA vaccines, despite the logistical problems of storing and transporting at sub-zero temperatures, because these vaccines have demonstrated efficacy in people over the age of 65.

Notwithstanding the above, the World Health Organisation (WHO) recently recommended the use of AZD1222 even for those aged 65 and above. In its interim recommendations on the use of AZD1222, the WHO stated: “Because a relatively small number of participants aged 65 years or over were recruited into the clinical trials, there were few cases of COVID-19 in either the vaccine or the control group in this age category, and thus the confidence interval on the efficacy estimate is very wide. More precise efficacy estimates for this age group are expected soon…. Immune responses induced by the vaccine in older persons are well documented and similar to those in other age groups. This suggests it is likely that the vaccine will be found to be efficacious in older persons. The trial data indicate that the vaccine is safe for this age group. The risk of severe disease and death due to COVID-19 increases steeply with age.... Taking the totality of available evidence into account, WHO recommends the vaccine for use in persons aged 65 years and older.”

The Phase 3 trials of Covaxin, on the other hand, have included a significant number (2,433) of people over the age of 60, though the interim efficacy data has not been presented in age-stratified form yet. According to reports, Bharat Biotech now plans to submit this interim Phase 3 data to the CDSCO to get regular unconditional REU approval, which presumably will include such stratified efficacy data. On the basis of the WHO’s recommendation on Covishield/AZD1222 and the interim efficacy data of Covaxin/BBV152, the elderly of the country can get a shot of either vaccine without much apprehension.

Inter-dose gap controversy

It would be recalled that while approving the Oxford-AstraZeneca vaccine for emergency use in the U.K. in January, the U.K.’s Medicines and Healthcare Products Regulatory Agency (MHRA) permitted a wide window of four to twelve weeks as the interval between the two vaccine doses, though the Phase 3 clinical trials in the U.K. and elsewhere were based on a fixed four-week gap. This was primarily to ease vaccine supply bottlenecks and get the available doses to as many of the vulnerable population as possible. The scientific basis for this decision, not a very strong one though, was the observation (although not in a large sample) that the first dose itself generated a significant immune response, and the assumption was that this response would take longer than four weeks to decline to low efficacy levels. But the move generated considerable controversy within the scientific community.

However, more recently, this scientific basis has been placed on a firmer footing with a recent “exploratory” analysis of Phase 3 trial data of 17,178 participants, which found that the overall efficacy actually increases with increasing the inter-dose gap up to 12 weeks. The paper’s non-peer reviewed preprint version was posted on the web on February 1. It has now been peer reviewed and was published in The Lancet on February 19. The work found that while the efficacy was only 55 per cent for a gap of up to six weeks, it was 81 per cent for a three-month interval. The overall pooled efficacy against symptomatic COVID-19 irrespective of the inter-dose gap is about 63 per cent.

Also, a single dose of the vaccine was highly efficacious (76 per cent efficacy) in the first three months 22 days after the first dose. The authors also note that it is not clear whether the protection with only a single dose will last longer than three months as the trial protocol limited the data-gathering period to three months maximum. And so, a second dose of the vaccine is still recommended. Efficacy results were also backed by binding antibody titres in the 18-55 age group, which were more than twofold higher in the group with a longer delay before the second dose.

On the basis of the above data, in its February 10 document on interim recommendations on the Oxford-AstraZeneca/AZD1222 vaccine, the WHO has recommended an interval of eight to twelve weeks between the doses even though the manufacturer AstraZeneca’s product label says that the vaccine can be administered with an interval of four to twelve weeks. But the WHO document has further stated that if the second dose is inadvertently administered earlier than four weeks after the first, the dose does not need to be repeated. And, if administration of the second dose is inadvertently delayed beyond 12 weeks, it should be given at the earliest possible opportunity. It is recommended that all vaccinated individuals receive two doses.

However, the Ministry of Health and Family Welfare has not changed the dosage schedule that it had recommended earlier, which is 28 days. It is on the basis of the Phase 3 trial data (from the U.K., South Africa and Brazil) with a four-week inter-dose gap that the DCGI had granted REU approval for Covishield. Commenting on these recent developments in The Hindu , the eminent virologists Jacob John and Gagandeep Kang independently suggested that the government should revisit the dosage schedule. Unwittingly, these new results would perhaps be comforting to those who may have missed the second dose after the gap of 28 days.

As on March 4, about 1.7 crore people have been administered the vaccine in the ongoing phase of the campaign that began on January 16. Although the initial progress was tardy, it picked up pace in February, and at present about 10 lakh doses are being administered every day. Administration of the second dose for those who got the first dose on January 16 began on February 13, as per the dosage schedule. However, the cumulative number of doses being administered (see graph), as on date, is fewer than half of the total number of doses that the government has procured. On February 12, the Health Ministry stated in its answer to a question in Parliament that a total of 410 lakh doses of the two vaccines had been procured. Of this, 100 lakh doses were through GAVI, the global vaccine alliance set up to ensure equitable access to vaccines across countries, with support from COVAX, the WHO initiative for COVID-19 vaccine access to poor countries.

Clearly, between February 12 and to date, the figure of 410 lakh must have gone up to say 500 lakh, but the consumption has been only about a little over one third. According to the U.K.’s MHRA, Covishield has a shelf life of about six months if stored at 2-8 °Celsius. The same would perhaps hold good for Covaxin as well. With the footfall of people at the vaccination sites not being commensurate with the accumulated vaccine stock, the government has now allowed vaccination to be carried out round the clock at all the vaccination centres perhaps apprehensive of loss of efficacy of the vaccines if stored for longer periods.

Skewed distribution

Soon after the launch of the vaccination programme, the Ministry stated that only 12 States would receive Covaxin in addition to Covishield. An earlier article pointed out the Centre’s skewed distribution of vaccine doses of Covishield to States. Whether this was done because of hesitancy on the part of the government as Covaxin’s Phase 3 trial data were not in yet or because vaccine production was not geared to meet the demand was not clear. While the number of States receiving Covaxin has now increased from the original 12 to 19, it is still just half of the total number of States/Union Territories. Why this is so is still not clear. In a recent statement to the media, Bharat Biotech CEO Krishna Ella stated that of the total of about one crore doses administered, only 11 per cent were of Covaxin.

Perhaps following its promising interim Phase 3 results and the DCGI’s consequent unconditional REU approval, Covaxin’s distribution to, and its offtake by, States will increase and the distribution of the two vaccines across the country will become more even.