Which SARS-CoV-2 variants reduce the effectiveness of vaccines?

Several new variants of the SARS-CoV-2 virus have been detected around the world and the impact of new variants on the performance of vaccines is a matter of concern. The more SARS-CoV-2 circulates in the population, the more likely it is to accumulate mutations that could potentially affect the immune response produced by current vaccines, potentially requiring new vaccines to be developed. However, there is increasing evidence that the same mutations are emerging in different SARS-CoV-2 strains, potentially posing a biological limit on how many different variants there could be and allowing for the better design of future vaccines.

How many new SARS-CoV-2 variants are there in the UK?

Public Health England regularly tracks the presence of SARS-CoV-2 variants in the UK. When first detected, new variants are initially classified as ‘variant under investigation’ (VUI). The New and Emerging Respiratory Virus Threats Advisory Group (NERVTAG) is responsible for classifying them as a ‘variant of concern’ (VOC) following a risk assessment based on indicators such as transmissibility, infection severity, and effectiveness of vaccines, drugs and therapeutics.

According to data reported up to 14 April 2021, there are 4 VOCs in the UK:

• 1.1.7 (also known as ‘Kent variant’ or VOC 202012/01), the main VOC in the UK, accounting for a total of 209,492 cases;
• 1.351 (also known as ‘South African variant’ or VOC 202012/02), the second most common VOC with 600 cases reported;
• 1 (also known as ‘Brazilian variant’ or VOC-202101/02), with 40 cases reported so far
• ‘B.1.1.7 with E484K’ (also known as VOC-202102/02), a VOC that evolved from B.1.1.7 that also had the E484K mutation (one of the mutations also present in the ‘South African variant’, which is key in evading the antibody response). Only 43 cases have been reported.

There are another seven variants currently under investigation (VUI), that account for around 700 cases altogether. This Rapid Response summarises the evidence available on the effectiveness of the vaccines that are currently licensed in the UK on the most common UK VOCs (the ‘Kent’ variant and the ‘South African’ variant).

An up-to-date review on how single mutations are able to avoid the immune response is available. A short review of the effectiveness of other vaccines on B.1.1.7, B.1.351 and P.1 variants is also available.

What advice has the UK Government received on vaccine effectiveness against the new variants?

On 27 January 2021, NERVTAG produced a brief note on the SARS-CoV-2 variant B.1.1.7 and a note on the SARS-CoV-2 variant B.1.351. They summarised the data available on whether antibodies triggered by a natural infection with the ‘original’ version of SARS-CoV-2 (known as ‘wild-type’) or by immunisation with the Pfizer/BioNTech or the Moderna vaccine were able to recognise and neutralise these two variants. According to the data available up to that point, NERVTAG concluded that antibodies (either induced by natural infection or by immunisation) were able to neutralise the B.1.1.7 (Kent) variant, but their neutralising activity was reduced against the B.1.351 (South Africa) variant. Reductions in neutralising activity (ability to bind and stop infectious virus) were higher for antibodies triggered by natural infection than from the two mRNA vaccines tested. NERVTAG also highlighted that the actual impact on immunity in the population was still unclear, as these results are based on laboratory experiments and not on clinical trials.

The effectiveness of the current COVID-19 vaccines on variants and how to adapt them to new variants were also discussed at the SAGE meeting on 11 March 2021. The SAGE Vaccine Update Group reviewed all the evidence available on the ability of antibodies from people who have recovered from COVID-19 or from vaccinated individuals (either with the Pfizer/BioNTech or with the Moderna RNA vaccines) to recognise and neutralise SARS-CoV-2 variants. The paper shows that small reductions could be observed for the B.1.1.7 (Kent) variant, while an up to 10-fold decrease in neutralising activity against the B.1.351 (South Africa) variant was seen in vaccinated people and those who have recovered from natural infection.

What about T-cells?

An important consideration of many of the studies discussed at SAGE is that they are based on antibodies. These  are just a single component of the immune response and there is increasing evidence that T-cells have a more complex and comprehensive role in the immune response against SARS-CoV-2. A recent study (not peer reviewed) demonstrated that T cell responses from recovered patients or recipients of the Pfizer or Moderna vaccines are not affected by several variants tested, including B.1.1.7 and B.1.351. Another study showed that some rare mutations in specific parts of SARS-CoV-2 (including the spike protein) can affect how well T-cells recognise the virus, leading to evasion of the T cell-mediated immune response. See the latest British Society for Immunology report for more details about immunity and COVID-19.

Another important caveat in the majority of the studies discussed is that they are based on in vitro assays (performed outside a living organism). Studies analysing infections (either in human or in animal models) could provide a more biologically accurate understanding of how variants interact with the vaccine-triggered immune response.

What else do we know about COVID-19 vaccine effectiveness against the new variants?

The following paragraphs analyse the latest research developments on this subject for all the vaccines currently approved in the UK (synthesised in the table below). It is correct as of 21 April 2021.

Neutralising antibody activity compared with wild-type SARS-CoV-2

The Pfizer/BioNTech vaccine

There is increasing evidence that antibodies triggered by the Pfizer/BioNTech vaccine are able to recognise the (Kent) B.1.1.7 variant, with only minor changes in their neutralising activity, while their ability to recognise the B.1.351 (South Africa) variant is impaired by up to 7.6-fold. There is also some evidence that neutralising activity decreases profoundly against ‘B.1.1.7 with E484K’. These data would suggest that the E484K mutation, a mutation present in the B.1.351 (South Africa), the P.1 (Brazilian) and the ‘B.1.1.7 with E484K’ variants, could impact the effectiveness of the antibody response triggered by the Pfizer/BioNTech vaccine. However, one study showed that the ability of antibodies triggered by the Pfizer/BioNTech vaccine to recognise the P.1 variant is only reduced by 2.6-fold.

Recent preliminary research sponsored by Pfizer analysed antibodies from 15 participants and identified roughly equivalent neutralisation for B.1.1.7 and P.1, and a significant reduction in neutralisation for B1.of at least 2-fold. The authors note that real-world data are required to confirm the impact of this on vaccine-mediated protection.

According to these data, B.1.351 and ‘B.1.1.7 with E484K’ could require an update of the Pfizer/BioNTech vaccine. However, a new study (not yet peer reviewed) suggests that the response of vaccine-induced T-cells was similar between the B 1.1.7 and B.1.351 variants and wild-type SARS-CoV-2. This suggests that vaccination could provide protection against these new variants through a T-cell response.

The Moderna vaccine

Use of the Moderna vaccine started in Wales on 7 April 2021. There is increasing evidence that vaccine-induced neutralising antibodies are able to recognise the B.1.1.7 variant and therefore vaccine-induced immunity may be produced. Recent data by Moderna showed similar neutralising activity against the B.1.1.7 (Kent) variant, a major (6.4 fold) reduction against the B.1.351 (South Africa) variant and smaller reductions against the P.1 (Brazil) and the ‘B.1.1.7 with E484K’ variants (3.5 and 3.1 fold, respectively). These reductions were attributed (at least partially) to the E484K mutation.

A recent study (not yet peer reviewed) hypothesised that although neutralising antibodies induced by the Moderna vaccine were lower against the B.1.351 (South Africa) variant, they should still be able to confer protection against it.

The Oxford / AstraZeneca vaccine

A recent study based on UK data found that, although there is an 8.9-fold reduction in the level of vaccine-induced neutralising antibodies, the Oxford/AstraZeneca vaccine is 70.4% effective against symptomatic COVID-19 disease caused by the B.1.1.7 variant and vaccination reduces the duration of shedding and viral load, with potential impact on the transmission of COVID-19. Data on neutralising activity or vaccine efficacy against ‘B.1.1.7 with E484K’ is not available yet.

A study published in February based on blood serum from people vaccinated with the Oxford/AstraZeneca vaccine found a 9-fold reduction in neutralising antibodies against the (South Africa) B1.351 variant. A more recent, South Africa-based study on 2,026 participants aged between 24 and 41 years found that the Oxford/AstraZeneca vaccine was only 10%  effective against mild disease from the B.1.351 variant. Neutralising antibody activity against the B.1.351 variant was evaluated in 12 participants: seven had no detectable activity and five showed between a 4.1 to 31.5-fold reduction. Analysis of vaccine-triggered T-cells in 17 participants found that these should still be able to recognise several components of the spike protein of the B.1.351 variant. The impact of this on severe disease is still unknown. Experts warned about overinterpreting these results because of the very limited number of participants in the study and the fact that it only focused on mild COVID-19 cases.

Current evidence on the effectiveness of the Oxford/AstraZeneca vaccine against the P.1 (Brazilian) variant is limited. However, one study found a minor (2.9-fold) reduction of vaccine-triggered neutralising antibodies against this variant.

The World Health Organization (WHO) published a statement highlighting how the Oxford/AstraZeneca vaccine has been proven effective in preventing severe COVID-19 and its effectiveness in preventing severe illness caused by the B.1.351 variant is still unknown. The WHO recommends using the Oxford/AstraZeneca vaccine, even in the presence of variants, but highlights the need for increased surveillance on vaccine effectiveness in the presence of new variants.

How quickly can the current vaccines be adapted to new variants?

Updating vaccines to respond to new versions of a virus is a well-established process in the case of seasonal influenza vaccines. The WHO Global Influenza Surveillance and Response System monitors the global influenza virus variants and recommends vaccine updates when needed (such as when antibodies triggered by the current vaccines cannot neutralise a new variant). According to the SAGE Vaccine Update Group analysis, a reliable ‘correlate of protection’ for SARS-CoV-2 (a quantifiable sign, such as a certain level of neutralising antibodies, that could confirm that a person is protected from the virus) is needed in order to know when adaption of current vaccines for new SARS-CoV-2 variants is needed. At the moment the most reliable correlate is the level of neutralising antibodies that are able to recognise live SARS-CoV-2 virus. However, the absence of detectable neutralising antibodies does not necessarily mean someone is not immune to SARS-CoV-2 and other parts of the immune system may be just as important.

The minutes of the 83^rd meeting of SAGE provisionally concluded that a drop in vaccine efficacy is expected following the reductions in neutralising antibody levels seen against variants. This reduction will depend on vaccine type, neutralising antibody levels and how different the variant is from the ‘wild type’ SARS-CoV-2. Developing a process to update current vaccines will be necessary and a decision will depend on whether the vaccines are used to protect from COVID-19 disease or from SARS-CoV-2 transmission. For further information see the POST Rapid Response analysing the impact of vaccines on transmission.

Usually new vaccines are recommended when decreases in neutralising activity similar to the one observed with B.1.351 are seen. The current vaccine platforms based on RNA (such as Pfizer/BioNTech and Moderna) and on viral vectors (such as the Oxford/AstraZeneca) can rapidly being re-designed to target a new version of the spike protein:

• On 25 February, Pfizer/BioNTech announced that they were discussing with regulators how to begin a clinical study using a new version of their vaccine based on the B.1.351 (South African) variant.
• Similarly on 24 February, Moderna announced the creation of a variant-specific vaccine candidate, mRNA-1273.351, against the SARS-CoV-2 variant known as B.1.351 to use as a booster or as an initial vaccine to immunise people. On 10 March, Moderna announced two different new versions of their vaccine, a variant-specific one based on B.1.351 and a ‘multivalent one’, combining their original mRNA sequence with the new B.1.351-based mRNA. They are currently evaluating them as new vaccines or as boosters for those who have received the ‘original’ vaccine first dose (see below).On 15 March, Moderna also announced the beginning of clinical trials for a new version of its vaccine, encoding for other parts of the spike protein, called the Receptor Binding Domain (RBD) and N-terminal Domain (NTD).
• On 23 March, AstraZeneca announced that they will soon start trials on a new version of their vaccine that will work against all current SARS-CoV-2 variants. These are thought to be available for use in late 2021.

Changing the genetic information embedded in the vaccines is only the first requirement. Adjustments to the manufacturing pipeline are complex and are likely to require more time. Read more in the previous POST rapid response about Manufacturing COVID-19 vaccines.

How quickly can modified vaccines be approved?

Another key aspect of adapting current vaccines to new variants is the extent to which new clinical trials are required for regulatory approval.

On 4 March 2021, the ACCESS Consortium (a coalition of regulatory authorities from the UK, Australia, Canada, Singapore and Switzerland) published a guideline for adapting authorised COVID-19 vaccines for SARS-CoV-2 variants. The guidance is based on a well-established system used for influenza virus vaccines, which need annual updates as influenza viruses mutate so quickly. To be granted approval, manufacturers will need to show that the new version of the vaccine is able to stimulate an immune response (both antibody and T-cell response), is safe, and meets high quality standards. Only short-term clinical trials results (up to 2 months) will be required. Clinical trial data from the original vaccine version and real-world data on ongoing immunisation programmes will be considered, as well as post-approval effectiveness studies.

Similar policies to allow the rapid approval of adapted COVID-19 vaccines have been published by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA). The International Coalition of Medicines Regulatory Authorities held a series of meetings to discuss global regulators’ work towards alignment on policy approaches and regulatory flexibility during COVID-19.

What other strategies can be used in the future?

There is increasing consensus among experts that SARS-CoV-2 is very likely to become an endemic virus, such as the chickenpox virus in the UK. Modifying existing vaccines is only one of several alternatives that could improve current immunisation strategies and offer higher protection against SARS-CoV-2 variants. These include:

1. Combining different vaccine types. This could potentially lead to a stronger immune response and is currently being tested in the UK with the Pfizer and AstraZeneca vaccines, however the publication timing of any results is currently unknown.
2. An additional booster dose. Vaccines approved so far in the UK require two doses; a prime dose to trigger an immune response and a second dose to boost it. A third dose could boost the immune system’s response even more, leading to enough antibodies to neutralise any new variants. Relying on multiple booster doses is a strategy already established for some vaccinations, such as tetanus. This strategy is currently under investigation by Pfizer/BioNTech, who are planning to administer a third dose to 144 Phase 1 participants in two age cohorts (18–55 and 65–85 years) 6 to 12 months after their second dose.
3. Developing a multivalent vaccine, able to target several variants at once. An example of a multivalent vaccine is the quadrivalent influenza vaccine that protects against four different strains of influenza virus. Moderna is currently testing a two-dose regime for its multivalent COVID-19 vaccine. Novel computational tools are also being used to identify the best way to develop such vaccines.
4. An alternative booster dose. Moderna is currently testing the possibility of using their original vaccine against the ‘wild-type’ SARS-CoV-2 as a prime dose and either a variant-specific one or a multivalent vaccine as a booster.
5. Using vaccines based on whole inactivated virus. These could potentially develop a stronger immune response that is capable of recognising different parts of the SARS-CoV-2 virus and not just the spike protein. The UK Government has already secured 100m doses of a whole inactivated virus vaccine, the Valneva vaccine, which is currently in Phase 1/2 trials. Phase 3 trials are planned for the end of April 2021.
6. Developing a universal coronavirus vaccine, for all variants that could potentially emerge. This is technically possible thanks to advances in vaccine technology, but would require a world-wide coordination network. There are around two dozen research projects worldwide focusing on the development of a universal coronavirus vaccine.

What is the UK Government doing to facilitate the development of vaccines against new SARS-CoV-2 variants?

On 22 February, the UK Government published its COVID-19 Response – Spring 2021, the roadmap out of lockdown for England. The document describes preparations for a revaccination campaign, likely to happen in autumn or winter 2021, with a variant-specific booster, and the establishment of a partnership with the German vaccine manufacturer CureVac to quickly develop new mRNA vaccines.

On 3 March, Budget 2021 announced £5 million to support the creation of a ‘library’ of mRNA vaccines for future variants and £22 million to further fund vaccine trials, including those combining different vaccine types and one assessing the effectiveness of a third dose.

Photo by Mufid Majnun on Unsplash