COVID-19 vaccines: Trials & errors

As foreign-made vaccines are cleared for emergency use and post-approval parallel bridging clinical trials are waiting to be done, it becomes imperative for the adverse effects monitoring and surveillance system to be more effective than has been the case so far.

On April 13, on the basis of the recom mendation made by the National Expert Group on Vaccine Administration for COVID-19, the government fast-tracked Emergency Use Authorisation (EUA) for all the foreign-made COVID-19 vaccines that have already been granted such approval by the United States Food and Drug Administration, the European Medicines Agency (EMA), the United Kingdom’s Medicines and Healthcare products Regulatory Agency and the Japanese Pharmaceuticals and Medical Devices Agency or have been included in the World Health Organisation’s (WHO) Emergency Use Listing.

The above would be besides the vaccines that the Drugs Controller General of India has already approved for EUA: the Serum Institute of India’s (SII) Covishield/ChAdOx1nCoV, Bharat Biotech International Limited’s (BBIL) Covaxin and the Gamaleya National Centre’s Sputnik V. While both Covishield and Sputnik V, which have been developed in the U.K. and Russia respectively, required pre-approval bridging clinical trials to be conducted in India as required under Indian law, the newer foreign-made vaccines will not have to go through such trials. Instead, these vaccines are mandated to carry out a “post-approval parallel bridging clinical trial”.

A bridging study for a drug or a vaccine that is newly introduced in a region provides clinical data on population-specific variations in safety, efficacy and dosage to enable extrapolations from the existing clinical trial data from elsewhere. But, like the earlier controversial EUA in “clinical trial mode” for Covaxin, a parallel bridging trial is a new concept too. This is clearly a consequence of the steep surge in cases, on the one hand, and the ostensible shortage of vaccines different States are facing, on the other. But, actually, even SII was supposed to submit bridging study data before getting approval for Covishield but provided only partial data. What SII would have done to achieve its bridging study targets is tantamount to a parallel trial being carried out even as Covishield had been rolled out.

Global shortage of COVID-19 vaccines

While, in principle, the doors are now open for several foreign-made vaccines to enter the country, whether this will result in the actual flow of vaccines into the country in the immediate future is moot. Given the global shortage of COVID-19 vaccines and considering that many of these manufacturers are already committed to supplying to other countries under advance purchase contracts that they entered into by mid 2020 itself, it is extremely unlikely that any of them will be in a position to supply to India at this juncture even if some Indian companies have pre-existing agreements with some of them to “fill, finish and market”.

Given the current vaccine supply scenario worldwide, besides Covishield and Covaxin, only Sputnik V, which was granted EUA shortly before this announcement, is likely to enter the Indian market in the next two to three months, especially because the Russia Direct Investment Fund has signed agreements with four Indian companies to manufacture 200 million doses a year, each beginning mid 2021. Therefore, this new policy pronouncement may not amount to much immediately. It is also shocking that the government’s so-called “atmanirbhar” efforts have stopped at mere sloganeering. It has made zero investment in the country’s two main vaccine manufacturers for them to scale up their production capacity the way the U.S. and Europe have done. Indeed, Adar Poonawalla, CEO of SII, has been saying for some time now that he requires an additional Rs.3,000 crore to scale up production, but the government has paid no heed to his appeals. The government earmarked Rs.900 crore in November 2020 towards “Mission COVID Suraksha” aimed at strengthening COVID-19 vaccine–related R&D, of which Rs.500 crore has already been committed, but did not seem to think that it was prudent to invest in scaling up domestic vaccine manufacturing capability as well. If the government had had the slightest foresight, the country would not be facing this crisis situation of imminent vaccine shortage.

The total number of doses of Covishield and Covaxin administered from January 16 when the vaccination drive began to April 15 has touched nearly 115 million, about 92 per cent of which are with Covishield. The current average doses administered a day is about 3.5 million. According to experts, this rate needs to be ramped up to at least 10 million a day if the vulnerable population is to be vaccinated in the next two to three months.

The current production capacity of the two approved COVID-19 vaccines in the country is about 85 million doses a month: around 75 million doses of Covishield and a little over 10 million doses of Covaxin. On April 8, Health Minister Dr Harsh Vardhan tweeted that the government had a stock of 24 million doses, which means that, if the same rate of vaccine consumption continues, at the end of the month the country would only be left with enough doses to go on for about 12 days.

Effectively, the country would be consuming about as many doses a month as it would be producing. But that is not enough; what the country needs is about three times that rate, and so the capacity has to be ramped up at least threefold. And if the companies have export commitments to meet, it will need to be increased even more. Clearly, without any investment on that front, that is unlikely to happen. Already, SII has been forced to renege on its contract with AstraZeneca to export 50 per cent of its production. The logic of opening up imports would seem to be from that calculation. The fact that even at this stage the government has not made a statement about investing in manufacturing capacity amounts to utter disregard for public health to say the least.

An important fallout of a large-scale vaccination campaign is what is called an adverse event following immunisation (AEFI). An AEFI is any untoward medical condition that occurs following vaccination and which does not necessarily have any causal relationship with the usage of the vaccine. Since there is no such thing as a perfect vaccine that protects everyone and is safe for everyone, AEFIs are part and parcel of all immunisation drives. Effective vaccines may produce some undesirable side effects that are usually mild, such as headache, pain at the vaccination site and fever, and get resolved quickly. A large number of events thought to be associated with the administration of the vaccine are simply coincidental events and are not actually due to the vaccine itself. Some can even arise due to human or programme error. But real vaccine-related serious adverse events, which may even turn out to be fatal, need to be identified promptly and acted upon. For that, a robust system of monitoring and surveillance of vaccinated individuals and prompt follow-up action on the reported AEFIs, including hospitalisation and subsequent investigation if required, must be in place.

For the total of about 115 million vaccine doses given so far, the total number of AEFIs recorded to date (April 15) is 18,511, giving an overall rate of 0.016 per cent. But, as mentioned above, a large fraction of these would be mild events. What is of significance is the number of serious adverse events (SAEs) and their causality assessment to determine which were actually due to the vaccine itself. According to N.K. Arora, Adviser to the National AEFI Committee and a member of the National COVID-19 Task Force of the Ministry of Health and Family Welfare (MoHFW), despite the limited resources available, India has one of the best AEFI surveillance systems among the developing countries and the system was doing far better than the WHO’s benchmark of reporting 1-in-100,000 (0.001 per cent) AEFIs even during the current pandemic.

But the system has been found severely wanting in monitoring SAEs and conducting their causal assessment apace. This becomes particularly important in the case of the COVID-19 vaccines, many of which use new technologies hitherto untried in the public space and which are being/will be rolled out under EUA across the country.

According to an April 9 report in The Hindu , as on March 31, among the AEFIs reported up to March 29, there had been 617 SAEs (including 180 deaths). Of these, the Immunisation Technical Support Unit of the Ministry had completed classification of only 492, which included 124 deaths (Table 1), 305 serious events that required hospitalisation (Table 2) and 63 severe events that did not require hospitalisation (Table 3). Significantly, however, complete documentation was available for only 236 (38.3 per cent) of the 617 cases of AEFIs.

As of March 17, causal assessment (CA) seems to have been completed only for 13 SAEs, which included 10 deaths; CA for five of these had been made public on March 5 itself. From the above publicly available information, it appears that autopsies had been carried out only in six cases though Arora told The Hindu in the third week of March that in 38 out of 71 post-mortem reports received so far, a causal link to the vaccine had not been established. “All the events so far are coincidental,” he had said. However, this information, along with their CAs, is not yet in the public domain. He also told BusinessLine on April 6 that the Ministry would make public a report with complete information on AEFIs within a week’s time. More recently, on April 12, in an interview to India Today , Arora repeated the same thing. In reality, however, the CA document on a measly eight SAEs of March 17 remains the last publicly available report on the AEFIs.

To describe the SAEs, the CA uses the following phraseology for classification, “consistent causal association to vaccination”, “inconsistent causal association to vaccination (coincidental)” and “unclassifiable”. But, without doing further analysis, particularly with regard to cases classified with the first epithet, the CA documents have concluded: “None have [ sic ] been found to be due to the COVID-19 vaccine”.

Commenting on the lack of transparency on the AEFIs, K.R. Antony, a paediatrician from Kochi, wrote in The Hindu on April 15: “There is no link to vaccination is the quick response by the authorities even before a systematic investigation, including a post-mortem examination, is completed. A line list of less than 15 deaths are [ sic ] visible to the public on the MoHFW website. The time of death or the period of gap between vaccination and death is not mentioned. Some deaths are not even fully documented. Knowing the timeline pattern of deaths is important…. Mere coincidence is not an easy verdict to pronounce unless investigations are thoroughly completed.”

Lack of transparency on adverse events

On March 16, in a letter marked to several of the officials concerned in the Health Ministry and other relevant government officials, including the Prime Minister’s Office, Amar Jesani, editor, Indian Journal of Medical Ethics , Mumbai; Sandhya Srinivasan, consulting editor to the journal; and 27 other public health specialists said referring to the lack of transparency on AEFIs: “Lakhs of people in India are being administered the COVID-19 vaccines every day in the confidence that the vaccine will protect them against severe disease and death. The vaccine programme owes them complete information on the vaccines, a vaccination protocol that minimises the risk of harm, and an assurance of thorough and transparent investigation of injuries and death following immunisation. They are also owed medical care, and compensation for harm suffered post-vaccination. The government has not met these obligations…. The government must immediately undertake complete, time-bound and transparent investigation of all deaths and other SAEs following vaccination with the COVID-19 vaccines.”

“We sent two letters [on January 31 and March 16] to the ministry, the CDSCO [Central Drugs Standard Control Organisation], the national AEFI committee and others in the authority for systematic investigation of the clusters of the deaths in the AEFI (Cardiac, pulmonary, cerebral causes), and demanded full transparency and make available complete info. But there was no response,” Jesani told The Telegraph (Kolkata). “The investigation of each serious/severe AEFI and deaths needs to be done using high level of pathological expertise to detect clottings in the blood vessels of the body and the emboli emerging from them to damage other vital organs. A casual forensic autopsy would not suffice. That means there is an issue of competency of the experts currently involved in doing correct investigation of cause of death. And above all, the individuals involved in the process and in the committees to determine cause ought to be independent of the power,” Jesani added.

With further expansion of the vaccination campaign, which now includes all individuals above 45 (with or without comorbidities), the number of AEFIs will greatly increase. With the possible entry of foreign-made vaccines following the approval of Sputnik V and opening of the doors for the import of other foreign-made vaccines, the numbers being vaccinated will increase manyfold. The AEFIs and SAEs will concomitantly grow. Even with the limited categories of vaccine beneficiaries and with just two vaccines currently in use, the present system for AEFI monitoring and surveillance has several shortcomings and seems far from being optimal and transparent. These need to be addressed immediately so that the system of pharmacovigilance can respond quickly to the AEFI reports. These reports should be investigated through well-defined procedures and the corresponding findings, the CAs in particular, made public at the earliest. Jesani’s reference to blood clots in the remark above is in the context of the such adverse events seen in scores of vaccinees in several European countries following vaccination with the adenovirus-vectored vaccine developed by Oxford University and the Anglo-Swedish pharma company AstraZeneca. These have led to a suspension of vaccination campaigns in some countries and even to its total withdrawal from their campaigns in some others.

The technology that the vaccine uses is to genetically engineer the gene encoding the antigenic spike (S) protein on to a non-replicating chimpanzee adenovirus. The vaccine is administered intramuscularly to an individual. The S-protein is the protrusion visible on the surface of the virus SARS-CoV-2 that binds to human cells and is responsible for the virus’ entry into the body’s cells. Since the adenovirus is non-replicating, it will not cause any infection, but the S-protein that get expressed in the cells will prime the immune system to recognise the protein when an actual SARS-CoV-2 infection occurs and mount an immune response.

But investigations in Germany and Norway into SAEs involving blood clots following immunisation with the AstraZeneca vaccine have established that there is indeed a causal link between the vaccine and the observed thrombotic events. In its final assessment on the issue, while confirming that the benefit of the vaccine still far outweighed the risk of such rare blood clot events, the EMA said in a release on April 7 that the agency had concluded that unusual blood clots (thrombosis) with low blood platelets (thrombocytopenia) should be listed as very rare side effects of the COVID-19 vaccine AstraZeneca. According to the release, in reaching its conclusion, the committee took into consideration all currently available evidence, including advice from an ad hoc expert group. The WHO, on the other hand, put out a more cautionary statement that said: “Based on current information, a causal relationship between the vaccine and the occurrence of blood clots with low platelets is considered plausible but is not confirmed.”

Rare cases of blood clots

The EMA’s Pharmacovigilance Risk Assessment Committee (PRAC) noted that these rare cases of blood clots combined with low levels of blood platelets (thrombosis accompanied by thrombocytopenia) had occurred within two weeks of vaccination and most of the cases reported had occurred in women under 60 years of age. The PRAC had also noted that the blood clots occurred in veins in the brain (cerebral venous sinus thrombosis, or CVST) and the abdomen (splanchnic vein thrombosis, or SVT) and in arteries, together with low levels of blood platelets and sometimes bleeding.

As for the mechanism, it noted that one plausible explanation for the combination of blood clots and low blood platelets could be that it was an immune response, leading to a condition similar to one seen sometimes in patients treated with heparin, called heparin-induced thrombocytopenia. Heparin is a drug used to prevent the formation of blood clots. Certain platelet activating antibodies (to heparin) are known to lead to these thrombotic conditions. However, the statement added that, at this time, it was not possible to identify specific risk factors.

The EMA arrived at the above conclusion on the basis of its in-depth review of 62 cases of CVST and 24 cases of SVT reported in the European Union drug safety database as of March 22, 18 of which were fatal. The cases had come mainly from the spontaneous AEFI-reporting systems of the European Economic Area and the U.K., where around 25 million people had received the vaccine by that time. As of April 4, these numbers had increased to 169 cases of CVST and 53 cases of SVT, and by this date 34 million people had been vaccinated.

These findings—namely the plausible causal link between life-threatening post-vaccination blood clots in people under 60, particularly women, after vaccination with the AstraZeneca vaccine—assume significance in the Indian context because nearly 92 per cent of the vaccine doses administered so far are of Covishield, which is basically the Indian version of that vaccine manufactured by SII under contract with AstraZeneca.

So, do the AEFIs reported so far to the Indian system include any case of blood clots accompanied by low platelet count? Considering the blood clot event rate in 34 million vaccinees in Europe and with about 98 million people having been vaccinated with Covishield in India so far, it stands to reason that one should have seen (if not at the same rate) at least a fair number of blood clot events.

Interestingly, neither Table 2 nor 3 (which include SAEs with or without hospitalisation) has a category for thrombotic events. Perhaps indeed none was found among the 368 SAEs (not including death) whose categorisation was included as of March 29. Since this peculiar post-immunisation SAE is seen only among those below 60 years old, it is conceivable that hardly any such cases would have been seen because vaccination was yet to be thrown open to people in that age group, except for those with comorbidities. Table 1 (which categorises the 124 deaths) has, however, one case in the category called “COVID+” with thrombocytopenia. While the table does not indicate what COVID+ stands for, one presumes that the vaccinee who died of a low platelet count—it is not clear whether the person had thrombosis as well—was also COVID positive. It is also not clear whether the patient got the infection subsequent to vaccination or already had it. It may be noted here that thrombocytopenia and the pro-thrombotic state are conditions seen in some cases of severe COVID-19.

Interestingly, while the Indian AEFI database in the public domain does not reflect any case of blood clot combined with a low platelet count, a search done by the EMA’s PRAC in AstraZeneca’s safety database up to March 8 seems to have revealed eight cases of embolic and thrombotic SAEs from India, while the Indian database that is available in the public domain does not reflect this. The EMA’s assessment report of March 24, which includes data on embolic and thrombotic events with the AstraZeneca vaccine up to March 16, mentions 18 SAEs CVST/cerebral venous thrombosis events in predominantly women with a median age of 42 years, two of which are from India. While the report says that 6 of the 18 were fatal, it is not known whether any of these fatalities was from India. It is once again surprising that the same has not been disseminated to the public in India.

The EMA document also discusses the possibility of the vector adenovirus as the agent responsible for these thrombotic events accompanied by thrombocytopenia. The document notes: “Replication incompetent adenovirus vectors based on different serotypes of human adenoviruses are known to be able to cause thrombosis, secondary (consumptive) thrombocytopenia and disseminated intravascular coagulation, potentially leading to multiorgan failure and death, especially after intravenous injection of virus particles. This has been consistently described in preclinical models, like nonhuman primates and rabbits, and one fatality has also been reported in a phase 1 clinical trial.”

Further, the document observes that the pro-thrombotic effects of adenovirus vectors are considered to be caused by a combination of several pathways/sub-mechanisms with binding of adenovirus to platelets, causing platelet activation and thrombosis, being an important one. “Due to the overall similarity in the build of adenovirus particles,” the EMA report says, “it is considered likely that the above-mentioned pro-thrombotic effects are also shared by chimpanzee adenoviruses such as the ChAdOx1 vector employed in the AZ vaccine.”

Now that similar thrombotic events combined with low platelet counts have been seen in the U.S. with Johnson & Johnson’s single-shot human adenovirus-vectored vaccine called Janssen (Ad26.COV2.S) as well, which has led to pausing of the vaccination campaign with it in that country, the hypothesis of such SAEs being linked to the adenovirus vector gains ground. In August 2020, J&J tied up with the Indian company Biological E. Limited for the manufacture of Janssen.

Sputnik V is also a human adenovirus-vectored vaccine; it uses two different adenoviruses, Ad5 and Ad26, for the two doses. Dr. Reddy’s Laboratories conducted the bridging trials for Sputnik V. The Chinese vaccine by Cansino (Ad5-nCoV) also uses the inactivated human adenovirus vector Ad5.

So, with the possible entry of these other adenovirus-vectored vaccines into India, and becoming part of the Indian vaccination campaign, it becomes imperative that the pharmacovigilance system for AEFIs is made more effective and transparent, in particular for the hitherto unknown blood-clotting events with adenovirus-vectored vaccines, which have now been given the name vaccine induced thrombotic thrombocytopenia. Now that vaccination has been thrown open to everyone above the age of 45, such events may occur in larger numbers. The COVID-19 vaccine AEFI monitoring and surveillance system should be prepared and equipped to track and analyse them more competently than has been evident so far.