10 February 2021
Co-chairs: Marion Gruber (FDA, US) and Marco Cavaleri (EMA, EU)
1. Welcome and objectives of workshop
Almost one year after the declaration of the pandemic on 12 March 2020, vaccine and therapeutic development to protect against COVID-19 has progressed in an unprecedented pace resulting in authorizations in various part of the world and roll-out of these vaccines for mass vaccination campaigns has started. Other therapeutics are progressing although at a slower pace. However, over the last several months variants of the SARS-CoV-2 virus have emerged in various part of the world and have become dominant in their respective areas of origin raising questions regarding the level of protection that can be conferred by currently authorized vaccines against such variants and what could be the forecast for the future.
SARS-CoV-2 variants are new strains with demonstrably different phenotypes due to specific mutations in their spike protein that may be manifested by differences in antigenicity, transmissibility, or virulence.
The ICMRA workshop was organised in light of the possible spread of these new variants globally and the need to consider rapid updates of current vaccines. The objectives were to discuss current surveillance activities to monitor the spread of the variants, and minimal elements of data that would be required to swiftly approve updated versions of available vaccines against emerging variants of SARS-CoV-2.
2. What do we know about variants and impact on vaccines and therapeutics?
The D614G variant has become the most globally prevalent strain since its first detection in spring 2020. Three variants of concern (VOC) have since been identified; VOC 202012/01 (first identified in the UK), 501Y.V2 (first identified in South Africa) and P.1 (first identified in Brazil).
All these variants carry multiple mutations in the spike protein of SARS-CoV-2 and, importantly, some mutations in the receptor-binding domain (RBD) that alter the antigenicity of the coronavirus.
Data based on in vitro neutralization assays have been generated with sera from subjects vaccinated with different vaccines based upon the original SARS-CoV-2 virus, such as mRNA or viral vectored vaccines. Overall, it appears that most of these vaccines are not significantly altered in their neutralization activity against the VOC 202012/01 variant, while there is more pronounced reduction in neutralization against the 501Y.V2 and P.1 variants.
Emerging data from clinical trials in South Africa with the ChAdOx1- and Ad26-vectored vaccines as well as an adjuvanted sub-unit protein vaccine suggest that protection against the 501Y.V2 variant is decreased at different levels depending on the vaccine. This raises concern with respect to the suitability of vaccines that are based on the ancestral strain to protect against SARS-CoV-2 variants encompassing multiple mutations in the spike protein and in the RBD specifically.
It was agreed that aside from the need for a robust virologic surveillance network globally, data from additional clinical trials and from observational studies would be essential to understand the real impact of SARS-CoV-2 variants on the effectiveness of currently authorized vaccines and to guide decisions regarding the need to update available vaccines accordingly.
The value of setting up a coordinated global system, similar that already in place at WHO for influenza surveillance, was recognized, in order to evaluate and decide when and which emerging strains should be considered for any vaccine update. WHO has outlined a set of activities to address this key aspect and organised a dedicated meeting on 11 February 2021.
3. Vaccines: Possible minimal requirements for regulatory approval, or updated versions, of already-approved vaccines
The focus of the discussion was on the generation of monovalent vaccines containing the spike protein from an emerging variant recognized to pose a significant public health threat because it has shown to escape immunity from current vaccines. Multivalent vaccines against SARS-CoV-2 variants were not considered in this workshop.
There was consensus that large safety and efficacy studies, similar to those required for initial approval of COVID-19 vaccines, are not feasible and would not be compatible with the timeframe for a rapid regulatory decision responsive to public health needs.
Authorization under emergency use approval and approval of variant COVID-19 vaccines against a SARS-CoV-2 variant expressing the spike protein and made by the same manufacturer and process as the prototype COVID-19 vaccine shown to be safe and effective in clinical trials, could be considered based on clinical data bridging the immune response of the variant vaccine to the prototype parent vaccine for which efficacy has been demonstrated in clinical disease endpoint efficacy studies.
Assuming that much of the manufacturing process and controls, as well as the facilities for vaccine production, for the variant COVID-19 vaccine would be identical to that of the prototype COVID-19 vaccine and depending on platform specific aspects, data to be generated may be confined to critical aspect of product characterization, potency assay and stability.
In general, it was agreed that additional non-clinical safety studies would not be required. Data derived from animal challenge/protection studies in a relevant model could be considered supportive. However, clinical immunogenicity and safety data would provide primary evidence for determining the suitability of the variant vaccines.
Inference of efficacy would need to be supported by conducting clinical non-inferiority immunogenicity studies comparing the immune responses induced by a variant COVID-19 vaccine against the SARS-CoV-2 variant of concern to the immune responses induced by the prototype vaccine for which clinical studies demonstrated efficacy and when administered according to the authorized dose and dosing regimen.
Studies in seronegative adults showing non-inferiority after a primary series of the new variant vaccine against the variant strains compared to prototype vaccine against ancestral strain were discussed as critical evidence for inferring efficacy of the variant vaccine.
In addition, booster studies would provide evidence to support the use of the variant vaccine when administered as a booster to individuals who were already vaccinated with the prototype vaccine as a primary series according to the authorized dose and dosing regimen. In this case, one option would be to demonstrate non-inferiority of the variant booster response against the variant strain versus the immune response of the primary series of the prototype vaccine against the ancestral strain.
As no correlate of protection for COVID-19 vaccines has been established to date, immune markers best suited for inferring protection were briefly discussed. The majority view was that neutralizing antibodies would constitute the most suitable immune marker for conducting immuno-bridging studies despite the complexity of comparing results from assays against different strains.
T-cell responses are considered an important component of the immune response elicited by different vaccines contributing to protection, but it would not be feasible to measure them due to lack of currently available validated assays.
The size of the studies would be driven by pre-specified statistical success criteria and endpoints. Safety data should be collected covering both solicited and unsolicited adverse events as part of the immunogenicity studies. The safety of the variant vaccine would likely not differ significantly from that of the prototype, however, should unexpected safety signals arise, the safety database may have to be extended.
The importance of post-authorization studies evaluating safety and effectiveness of the variant COVID-19 vaccines was stressed.
4. Virus variants and impact on therapeutics
There is emerging data that suggest some of the approved or candidate monoclonal antibodies are affected by variants, with the possibility that some will lose effectiveness entirely. Additional studies and analyses are needed to understand the potential impact, including in which patients they could be used.
There was however consensus view in support of combination use of monoclonal antibodies rather than in monotherapy, for prophylaxis or treatment for circulating strains.
5. Closing remarks and next steps
Overall, there was good alignment and agreement on general principles between participating authorities. Regulatory convergence is seen as key to ensuring a consistent and timely response to emerging variants. FDA, US and EMA, EU are due to publish draft guidance shortly and WHO will publish the outcome of the 11 February 2021 meeting, and these will be shared with ICMRA partners to facilitate convergence to the extent possible.
A follow-up ICMRA workshop will be organised when more information is available on the behaviour and evolution of the virus variants and participating authorities have been able to review appropriate regulatory responses.