Safety and efficacy of an rAd26 and rAd5 vector-based heterologous prime-boost COVID-19 vaccine: an interim analysis of a randomised controlled phase 3 trial in Russia

Introduction

The COVID-19 pandemic has led to more than 98 million confirmed cases and more than 2 million deaths (at the time of publication). There are a few provisionally licensed vaccines against COVID-19, and global efforts are focusing on developing safe and efficacious vaccines for COVID-19 prevention. According to the WHO draft landscape of COVID-19 candidate vaccines,¹ 64 candidates are in clinical assessment (including 13 at phase 3) and 173 are in preclinical analyses. The phase 3 vaccine candidates include a variety of vaccine platforms: vector vaccines (Gamaleya National Research Centre for Epidemiology and Microbiology [NRCEM; this study], University of Oxford/AstraZeneca,² CanSino Biological Inc/Beijing Institute of Biotechnology, and Janssen Pharmaceutical Companies), mRNA-based vaccines (Moderna/National Institute of Allergy and Infectious Diseases³ and BioNTech/Fosun Pharma/Pfizer⁴), inactivated vaccines (SinoVac, Wuhan Institute of Biological Products/Sinopharm, Beijing Institute of Biological Products/Sinopharm, and Bharat Biotech), and adjuvanted recombinant protein nanoparticles (Novavax).

The safety of adenoviral vector vaccines has been extensively studied, and adenoviral vector-based therapeutic drugs are used in clinical practice.5, 6, 7 Adenoviral vector-delivered antigens are known to induce both cellular and humoral immunity after a single immunisation, allowing their use as an emergency prophylaxis tool in a pandemic. Furthermore, the use of two immunisations gives a durable and long-lasting immune response.8, 9 These characteristics make recombinant replication-deficient adenovirus (rAd)-based vaccines suitable candidates for the WHO target product profiles for long-term protection of people at high risk of COVID-19 in outbreak settings because they stimulate rapid onset of protective immunity. Although adenoviral vectors might induce immune responses against vector components and attenuate antigen-induced responses, prime-boost heterologous vaccination with two different vectors allows minimisation of this effect.9, 10, 11 Thus, the most effective approach for generating a powerful and long-lasting immune response that does not depend on the presence of a pre-existing immune response to the vector is the heterologous prime-boost vaccination approach. We used this approach when developing a vaccine for the prevention of COVID-19.

Gam-COVID-Vac is a combined vector vaccine, based on rAd type 26 (rAd26) and rAd type 5 (rAd5)—both of which carry the gene for SARS-CoV-2 full-length glycoprotein S (rAd26-S and rAd5-S). rAd26-S and rAd5-S are administered intramuscularly separately with a 21-day interval. The phase 1/2 clinical trials of the vaccine were completed in August, 2020.¹² The results showed that the vaccine was well tolerated and highly immunogenic in healthy participants. As a result, the vaccine candidate was provisionally approved in Russia according to national legislation. Such registration allows the vaccine to be used in high-risk groups, with enhanced pharmacovigilance, while a post-marketing efficacy study is conducted. Here, we present preliminary efficacy and safety results of a phase 3 multicentre study using Gam-COVID-Vac in adults, with subanalysis of adults older than 60 years.

Methods

Results

Between Sept 7 and Nov 24, 2020, 21 977 adults were eligible and randomly assigned to receive placebo (n=5476) or vaccine (n=16 501; figure 1 ).

The first database lock was on Nov 18, 2020, when 20 cases of COVID-19 had been reported. The interim safety analysis (analysis of rare adverse events) was done with data up to the first database lock. Since there was an increase in COVID-19 incidence in Moscow during November, the second database lock was done on Nov 24, 2020, when 78 COVID-19 cases had been reported. Data for the interim efficacy analysis and serious adverse events analysis are presented up to the second database lock.

74 participants from the vaccine group and 41 from the placebo group were excluded from analyses (figure 1). This preliminary analysis included 16 427 participants in the vaccine group and 5435 in the placebo group, who received at least one dose and continued participation in the trial. 14 964 in the vaccine group and 4902 in the placebo group had received two doses at the time of database lock (Nov 24, 2020) and were included in the primary outcome analysis (table 1 ). Median time from participants receiving the first dose to the date of database lock was 48 days (IQR 39–58). Among the participants who received two doses, the mean age was 45·3 years (SD 12·0) in the vaccine group and 45·3 years (SD 11·9) in the placebo group; the distribution by sex (p=0·619), incidence of concomitant diseases (p=0·420), and infection risk (p=0·851) were similar between the two groups (table 1).

From 21 days after the first dose of vaccine (the day of dose 2), 16 COVID-19 cases were confirmed in the vaccine group (of 14 964 participants; 0·1%) and 62 cases were confirmed in the placebo group (of 4902 participants; 1·3%); vaccine efficacy was 91·6% (95% CI 85·6–95·2; table 2 ). The observed vaccine efficacy was greater than 87% in all age and sex subgroups. Notably, vaccine efficacy was 91·8% (67·1–98·3) in participants older than 60 years. There were no cases (vaccine group) and 20 cases (placebo group) of moderate or severe COVID-19 confirmed at least 21 days after dose 1; thus, vaccine efficacy against moderate or severe COVID-19 was 100% (94·4–100·0). From 15 to 21 days after the first dose, efficacy was 73·6% (p=0·048), then from day 21, efficacy was 100% (p<0·0001; appendix p 11).

97 confirmed cases of COVID-19 (63 in the vaccine group and 34 in the placebo group) are not reflected in the analysis of the primary endpoint because they occurred fewer than 21 days after dose 1 (ie, before dose 2; table 2, figure 2 ). Estimated vaccine efficacy against confirmed COVID-19 occurring at any time after dose 1 was 73·1% (95% CI 63·7–80·1). Notably, in the vaccine group, most cases of COVID-19 occurred before dose 2. Rates of disease onset were similar for the vaccine and placebo groups until about 16–18 days after dose 1, after which, early onset of protection led to the number of cases in the vaccine group increasing much more slowly than in the placebo group (figure 2).

The interim immunogenicity analysis included samples transferred from the central laboratory that were collected before Nov 30, 2020, and showed that the vaccine induces an immune response in participants. Before first vaccination, no RBD-specific antibodies (of 456 participants tested) or virus-neutralising antibodies (of 100 participants tested) were detected in the blood serum of participants. In the analysis of humoral immune response, serum samples of 456 participants (342 from the vaccine group and 114 from the placebo group) were analysed for the presence of antibodies specific to the receptor-binding domain of SARS-CoV-2 glycoprotein S 42 days from the start of vaccination (figure 3 A). In the vaccine group, RBD-specific IgG was detected in 336 (98%) of 342 samples, with a geometric mean titre (GMT) of 8996 (95% CI 7610–10 635), and a seroconversion rate of 98·25%. In the placebo group, RBD-specific IgG was detected in 17 (15%) of 114 samples, with a GMT of 30·55 (20·18–46·26), and a seroconversion rate of 14·91% (p<0·0001 vs the vaccine group). When comparing the level of RBD-specific antibodies between age strata, we noted that the age 18–30 years group (combined male and female) had a significantly higher GMT than the other age groups (p=0·0065). There were no differences between the other age groups (p=0·343). Antibody levels did not differ significantly between men (n=179) and women (n=159; p=0·258). Descriptive statistics by age stratum and sex are in the appendix (p 3).

To assess the induction of a humoral immune response, serum samples from 100 participants were analysed for the presence of neutralising antibodies on day 42 after first vaccination (figure 3B); the GMT of neutralising antibodies was 44·5 (95% CI 31·8–62·2) and the seroconversion level was 95·83% in the vaccine group. GMT in the placebo group was 1·6 (1·12–2·19) and the seroconversion rate was 7·14%, which was significantly lower than that in the vaccine group (p<0·0001). Levels of neutralising antibodies were similar between age strata (p=0·222) and between men and women (p=0·639). Descriptive statistics by age stratum and sex are in the appendix (p 3).

Cellular immune response in participants was characterised by secretion of IFN-γ of peripheral blood mononuclear cells upon SARS-CoV-2 glycoprotein S restimulation in culture. To assess cellular immune response, serum samples from 58 participants (44 from the vaccine group and 14 from the placebo group) were analysed. By day 28 after first vaccination, all participants in the vaccine group had significantly higher levels of IFN-γ secretion upon antigen restimulation (median 32·77 pg/mL [IQR 13·94–50·76]) compared with the day of administration of the first dose (figure 4 ). Descriptive statistics for IFN-γ immune responses are in the appendix (p 4).

The general safety and rare adverse event analyses included 12 296 participants who received both doses up to the database lock on Nov 18, 2020. The most common adverse events were flu-like illness, injection site reactions, headache, and asthenia. Most of the reported adverse events (7485 [94·0%] of 7966) were grade 1; 451 were grade 2 (5·66%) and 30 were grade 3 (0·38%). 122 rare adverse events were reported in the study (91 in the vaccine group and 31 in the placebo group; appendix pp 8–9).

The analysis of serious adverse events included 21 862 participants who received at least one dose (of whom 19 866 received two doses) up to database lock on Nov 24, 2020. 70 episodes of serious adverse events, considered not related to COVID-19, were recorded in 68 participants: in 45 (0·3%) of 16 427 participants from the vaccine group and 23 (0·4%) of 5435 participants from the placebo group (appendix pp 5–7). None of the serious adverse events were considered associated with vaccination, as confirmed by the independent data monitoring committee (IDMC).

Because there were few serious adverse events, it was possible to process and verify the serious adverse events data up to the second database lock; however, full adverse events data, which has not yet been processed, will be provided in a later publication to avoid discrepancies with the final report after full data processing is complete.

During the study, four deaths were recorded: three (<0·1%) of 16 427 participants in the vaccine group and one (<0·1%) of 5435 participants in the placebo group. No vaccine-related deaths were reported. In the vaccine group, one death was associated with fracture of the thoracic vertebra and the other two were associated with COVID-19 (one patient with a severe cardiovascular background who developed symptoms on day 4 after first dose and one patient with a background of endocrinological comorbidities who developed symptoms on day 5 after first dose; appendix p 12). Based on the incubation period of the disease, both participants were deemed to be already infected before being included in the study, despite a negative PCR test. In the placebo group, the death was associated with haemorrhagic stroke.

The study included 2144 participants older than 60 years (1611 in the vaccine group and 533 in the placebo group). The mean age in this subgroup was 65·7 years (SD 4·5) in the vaccine group and 65·3 years (4·3) in the placebo group (appendix p 10). The maximum ages of the participants were 87 years in the vaccine group and 84 years in the placebo group. Proportions of participants by sex (p=0·378), incidence of concomitant diseases (p=0·774), and risk of infection (p=0·090) were similar between the vaccine and placebo groups. The vaccine was well tolerated in these participants. 1369 participants older than 60 years (who received two doses and for whom data in the case report form was verified at the time of database lock [Nov 18, 2020]) were included in the safety analysis. The most common adverse events were flu-like illness in 156 (15·2%) and local reaction in 56 (5·4%) of 1029 participants in the vaccine group and 30 (8·8%) and four (1·2%) of 340 participants in the placebo group (appendix p 11). There were three episodes of adverse events of grade 3 or worse, considered not associated with vaccination: an exacerbation of urolithiasis and acute sinusitis in the vaccine group and a flu-like illness in the placebo group. All these adverse events were resolved. In the participants older than 60 years, there were three serious adverse events reported in the vaccine group: renal colic and deep vein thrombosis (both associated with pre-existing comorbidities) and extremity abscess (due to physical injury and subsequent infection of the wound surface of the soft tissues of the finger). No association was found between serious adverse events and vaccine administration.

Discussion

Our interim results of the phase 3 Gam-COVID-Vac trial show that the vaccine is 91·6% (95% CI 85·6–95·2) efficacious against COVID-19 (from day 21 after first dose, the day of receiving second dose). Our results also showed that the vaccine was 100% (95% CI 94·4–100) efficacious against severe COVID-19, although this was a secondary outcome so the results are preliminary. The vaccine was well tolerated, with 45 (0·3%) of 16 427 participants in the vaccine group reporting serious adverse events, all of which were considered not related to the vaccine. According to the study design, the starting point for counting COVID-19 cases for estimation of vaccine efficacy was 21 days after dose 1 (day of dose 2 administration). Although the study was not designed to assess the efficacy of a single-dose regimen, our early starting point allows us to observe a possible partial protective effect of a single dose. The cumulative COVID-19 incidence curves of COVID-19 cases among the placebo and vaccine groups begin to diverge 16–18 days after the first immunisation, showing early onset of a partially protective effect after a single-dose immunisation; however, the study design does not allow us to draw conclusions from these observations.

The vaccine induced robust humoral (n=342) and cellular (n=44) immune responses in all age strata. Notably, there were a few non-responders in the vaccine group (six of 342), possibly due to immunosenescence in older people, individual characteristics of the formation of an immune response, or concomitant immunological disorders.

17 (15%) of 114 participants in the placebo group had RBD-specific antibodies on day 42, probably associated with asymptomatic COVID-19; however, none of these participants were SARS-CoV-2 PCR positive, nor did they report the onset of respiratory symptoms in the electronic diary or when interviewed as part of the telemedicine follow-up.

Given the importance of protecting populations at risk because of older age, we assessed the ability of the vaccine to induce an immune response and protect against COVID-19 in individuals older than 60 years. Our results show that the two-component vaccine Gam-COVID-Vac was able to induce a virus-neutralising humoral response in participants older than 60 years. Furthermore, vaccine efficacy in this group of participants did not differ significantly from the efficacy of the age 18–60 years group.

The limitations of the interim analysis of efficacy include the small sample sizes within age strata. Further data collection will allow for clarification of efficacy data within age groups. Furthermore, COVID-19 cases were detected through self report of symptoms by participants, followed by a PCR test, so only symptomatic cases of COVID-19 are included in the efficacy analyses.

Initially, we developed a vaccine in two forms: liquid (which is stored at –18°C) and freeze dried (which is stored at 2–8°C). In this study, we studied the liquid form of the vaccine that requires storage at –18°C. Storage at 2–8°C, a favourable temperature profile for global distribution, has been approved by the Ministry of Health of the Russian Federation.

We previously reported on the local and systemic post-vaccine adverse reactions of the Gam-COVID-Vac vaccine in a small sample of participants.¹² In this interim analysis, we report serious adverse events in more than 21 000 participants (of whom more than 16 000 received the vaccine).

70 episodes of serious adverse events were recorded in 68 participants across the two groups; none of these events were considered related to the vaccine. During the study, four deaths were recorded: three in the vaccine group and one in the placebo group. None were considered related to the vaccine, with confirmation by the IDMC. No post-vaccination adverse events were reported in any of these participants after vaccination.

The two COVID-19-related deaths were due to pre-existing cardiovascular and endocrinological conditions exacerbated by COVID-19. Taking into account the length of the incubation period described by WHO and the Centers for Disease Control and Prevention (CDC),14, 15 these two participants were probably already infected with SARS-CoV-2 at the time of randomisation and vaccination. On the basis of WHO and CDC guidelines and review of the underlying clinical data, the IDMC confirmed that participants were infected and disease had progressed before any immunity from the vaccine developed. A detailed description of the condition of participants with COVID-19 is in the appendix (p 12). Among the seven participants assigned to the placebo group who were confirmed to have COVID-19 within the first 7 days after the first dose, there were no comorbid conditions, in contrast to the vaccine group, in which there were three participants with a comorbidity among 25 who were confirmed to have COVID-19 within the first 7 days.

In summary, both COVID-19-related deaths have several principal points to be considered. First, despite the negative PCR test at screening and absence of increased temperature at the time of first vaccine dose administration, the onset of the first COVID-19 symptoms (4–5 days after first dose, similar to the average COVID-19 incubation period) testifies that participants had been infected with SARS-CoV-2 before or near the vaccination day, which was additionally confirmed by the IDMC, on the basis of WHO and CDC guidelines and review of underlying clinical data. Second, both participants self-administered non-steroidal anti-inflammatory drugs without informing clinicians, which interfered with diagnosis and receipt of medical help upon hospital admission. Third, because of limited diagnostics at screening (limitations of medical examination and testing and patient unaware of comorbidities) each participant had comorbidities that were only known after hospital admission. Participants who have not developed protective immunity to SARS-CoV-2 (ie, were infected before vaccination or early after vaccination) showed the natural clinical course of COVID-19. We did another analysis of the severity of the course of COVID-19 in the two groups, which showed that in the first 2 weeks after the first dose, there was no significant difference in the severity of the course of COVID-19 between the vaccine and placebo groups. From 15 to 21 days after the first dose, efficacy was 73·6% (p=0·048), then from day 21, efficacy was 100% (p<0·0001; appendix p 11). Therefore, in this study, the efficacy analysis was done 21 days after the first dose, because by that time, the immune response is formed.

Currently, scrupulous monitoring continues, in particular for cases of COVID-19. All safety data will be provided to the regulator for analysis.

In this interim analysis, we have not been able to assess duration of protection; median follow-up time was 48 days after first dose. Although the study enrolled participants with comorbidities, not all risk groups are represented. There is a need to further investigate the vaccine in adolescents and children under Pediatric Investigational Plans, as well as pregnant and lactating women. Most participants in our trial were white, so we welcome further investigation in a more diverse cohort.

Interim results on efficacy have been announced for several vaccine candidates against SARS-CoV-2. The safety and efficacy study of the ChAdOx1 nCoV-19 vaccine (AZD1222) provides an analysis of data from four randomised controlled trials in Brazil, South Africa, and the UK. 11 636 participants were included in the primary efficacy analysis. Among participants who received two standard doses, the reported efficacy of the vaccine was 62·1%, and in participants who received a low dose followed by a standard dose, efficacy was 90·0%, resulting in overall efficacy of 70·4%.² BNT162b2, an mRNA-based vaccine developed by Pfizer/BioNTech has reported 95% efficacy against COVID-19 in a multinational, placebo-controlled, observer-blinded, pivotal efficacy trial.⁴ 36 523 participants without baseline infection were included in the primary efficacy analysis. Eight cases of COVID-19 with onset at least 7 days after the second dose were observed in the study group and 162 cases were observed in the placebo group. There was one case of severe COVID-19 in a study group with onset at least 7 days after the second dose of BNT162b2.⁴ A phase 3, randomised, stratified, observer-blind, placebo-controlled study of an mRNA-1273 vaccine has enrolled 30 000 participants, 25% of whom are age 65 years or older. Interim results of the trial suggest efficacy of 94·1% based on 95 cases of asymptomatic COVID-19 among participants: 90 in the placebo group and five in the study group.³ The results of this Gam-COVID-Vac trial are not dissimilar to those reported for the other vaccines.

Vaccination strategies should account for a number of concerns regarding the priority of access to COVID-19 vaccines in various population groups, reliable risk assessment of adverse effects of vaccination in population groups with increased risk of severe COVID-19 (older adults and individuals with comorbidities), vaccine logistics (cold chain supply), sufficient coverage of immunisation, and duration of protective immune response. According to WHO target product profiles for COVID-19 vaccines,¹⁶ the characteristics required for emergency use during an outbreak include efficacy of at least 50%, suitability for use in older adults, maximum of two-dose regimen, and protection for at least 6 months. Further studies of candidate vaccines are needed to obtain information on duration of post-vaccine protective immune response. Yet, the results on efficacy and safety of COVID-19 vaccine candidates thus far are promising.

Our interim analysis of this phase 3 trial of Gam-COVID-Vac has shown promising results. In parallel with implementation of multiple clinical trials (in Russia, Belarus, United Arab Emirates, and India), the vaccine has already been released in Russia for use by the public, largely in at-risk populations, medical workers, and teachers, and as of Jan 23, 2021, more than 2 million doses of Gam-COVID-Vac have already been administered to the public (pharmacovigilance and monitoring of the incidence of rare adverse events is controlled by the Federal Service for Surveillance in Healthcare).

We are conducting research to investigate a single-dose regimen of the vaccine (the clinical trial was approved by the Regulator and Ethics committee on Jan 8, 2021, number 1). Our interim analysis of the randomised, controlled, phase 3 trial of Gam-COVID-Vac in Russia has shown high efficacy, immunogenicity, and a good tolerability profile in participants aged 18 years or older.

Data sharing

Anonymous participant data will be available upon completion of clinical trials and publication of the results of the completed study upon request to the corresponding author. Proposals will be reviewed and approved by the sponsor, security department, researcher, and staff on the basis of scientific merit and absence of competing interests. Once the proposal has been approved, data can be transferred through a secure online platform after the signing of a data access agreement and a confidentiality agreement.

Supplementary Material