Trimeric S protein COVID-19 vaccine needs to find its place

When the race towards the development of vaccines against COVID-19 began in early 2020, many of the front-line candidates were based on the spike (S) protein of SARS-CoV-2, coupled with an adjuvant. But it was the mRNA vaccines of BioNTech¹ and Moderna² that not only turned out to be highly efficacious but were produced and supplied in large quantities. The second place in the race has been divided between inactivated whole virion vaccines (from China³ and India⁴) and adenovirus-vectored vaccines.⁵ Even now, protein vaccines are not highly prominent. Novavax in December, 2021 received approval for its virus-like particles protein vaccine in the EU. Companies in other countries, such as Cuba, have successfully developed and introduced adjuvanted protein vaccines,⁶ but these are far from mainstream.

Against this background, the development and testing of Clover Biopharmaceuticals' trimeric S-protein recombinant vaccine SCB-2019 has been met with anticipation and success. In The Lancet, Lulu Bravo and colleagues⁷ report the results of their phase 2 and 3 efficacy trial (SPECTRA) of this vaccine in 30 174 participants in four continents, with the greatest contribution (45·4%) from the Philippines. A trial of this size nowadays is challenging because of the widespread roll-out of COVID-19 vaccinations. In this case, the intention was to find and recruit vaccine naive participants but, of the 30 155 with valid baseline serological data, 13 389 (44·4%) participants, including 61·5% (8406 of 13 676) in the Philippines, were seropositive for SARS-CoV-2 at the start. This situation in fact proves useful because information is needed on the performance of COVID-19 vaccines in seropositive individuals.

The SPECTRA trial was conducted at a time when the spectrum of variants was much different from the early studies in 2020, and allowed for establishing the vaccine efficacy against delta (78·7% [95% CI 57·3–90·4]), gamma (91·8% [44·9–99·8]), and mu (58·6% [13·3–81·5]). 30 128 participants (including 14 119 [46·9%] women) received their first dose of the vaccine (n=15 064) or placebo injection (n=15 064), and the per-protocol population consisted of 12 355 participants who were baseline SARS-CoV-2 naive (6251 vaccinees and 6104 placebo). Overall, in initially seronegative participants, the vaccine efficacy against any strain was 67·2% (95·7% CI 54·3–76·8) for any severity COVID-19, 83·7% (97·86% CI 55·9–95·4) for moderate-to-severe COVID-19, and 100% (97·86% CI 25·3–100·0) for severe COVID-19. One limitation of the study is that the participants were mostly young (mean age 32·1 years [18–86]). Furthermore, the follow-up time was short (mean 82 days). Thus, efficacy in older participants and the duration of protection are not known, although information on duration will be collected as the follow-up is being continued. Reactogenicity was low. Local reactions were reported in 36% of the vaccine recipients (compared with 10% in the placebo group). Systemic reactions were similar between the vaccine and placebo groups.

The study vaccine was not a final formulation but consisted of the vaccine (30 μg of recombinant trimeric S protein) and adjuvant (1·50 mg of CpG-1018 and 0·75 mg alum) mixed on site. A single vial combination is needed before the vaccine can be launched successfully. CpG-1018 was chosen for an adjuvant because it is already being used in a licensed vaccine, the HEPLISAV-B hepatitis B vaccine by Dynavax.⁸ A previous phase 1 trial concluded that 30 μg of trimeric S antigen combined with CpG-1018 was similar to 9 μg of antigen combined with adjuvant system 03.⁹ The antigen and adjuvant system combination induced somewhat higher antibody titres but was more reactogenic.⁹ The choice of CpG-1018 might be wise, because there have been reports of a potential association of adjuvant system 03 with narcolepsy after H1N1 2009 influenza vaccinations.¹⁰ Although the antigen that caused it was probably influenza nucleoprotein or neuraminidase, the potent adjuvant probably had a role.¹¹ However, the case of the SARS-CoV-2 S protein vaccine is different.

© 2022 FG Trade/Getty Images

The study was funded by the Coalition for Epidemic Preparedness Innovations, which is anticipated to add the SCB-2019 vaccine into its COVID-19 arsenal for use in low-income and middle-income countries through the COVAX mechanism.¹² To this end, Gavi, the Vaccine Alliance has already placed a tentative order for 400 million doses pending emergency use listing by WHO.¹³ Given that less than 50% of the world's population has received any COVID-19 vaccine, the new protein vaccine will be a welcome addition to the global response to COVID-19.¹⁴

The trimeric S protein recombinant vaccine used in the SPECTRA trial was based on the original SARS-CoV-2 virus and, therefore, the observed protection against other variants, such as the delta variant, might be called cross-protection. Studies have also shown that immune responses to the adjuvanted SCB-2019 vaccine measured against the alpha variant are similar to those induced by licensed mRNA vaccines.¹⁵ Still, cross-protection against the omicron variant might be lower than against other variants and modification of the vaccine might be required for omicron. This requirement adds to the list of challenges for the SCB-2019 vaccine.

I am the lead investigator of studies on hepatitis B vaccine by VBI Vaccines, unrelated to the topic of this Comment.