An Omicron-specific, self-amplifying mRNA booster vaccine for COVID-19: a phase 2/3 randomized trial – Nature.com

Study design

This was a prospective, multicenter, open-label, randomized phase 2 seamlessly followed by a phase 3 study to evaluate the safety, tolerability and immunogenicity of GEMCOVAC-OM as a booster in participants 18years of age and older. In this seamless study, phase 2 safety data till day 7 were analyzed and presented to an independent DSMB. The DSMB evaluated these data and provided their approval to initiate the phase 3 part of the study. The phase 3 study was conducted at 20 hospitals in 13 cities across India in compliance with the principles defined in the Declaration of Helsinki, International Conference for Harmonisation Good Clinical Practice Guideline. The study protocol was approved by the local ethics committee at each study site and Central Drugs Standard Control Organisation, the central licensing authority in India. This clinical trial is registered with the Clinical Trial Registry India, CTRI/2022/10/046475. Details on the sites and Ethics Committees can be found in Supplementary Information.

An interim analysis was planned at day 29 of phase 3 where the immunogenicity and safety of the participants was assessed and presented to the Central Drugs Standard Control Organisation for Emergency Use Authorization.

In phase 2, the safety and immunogenicity of GEMCOVAC-OM as a booster was compared with the prototype vaccine GEMCOVAC-19 designed against the spike protein of the D614G strain of SARS-CoV-2 (n=140). Participants were randomized to receive the vaccines in a 1:1 ratio.

The phase 3 study comprised a safety and an immunogenicity cohort. The safety cohort consisted of 3,140 participants of whom 3,000 were enrolled into the GEMCOVAC-OM arm and 140 were enrolled into the ChAdOx1 nCoV-19 arm. Within the safety cohort, the immunogenicity cohort consisted of 420 participants of whom 280 were enrolled into the GEMCOVAC-OM arm and 140 were enrolled into the ChAdOx1 nCoV-19 arm. Participants were healthy adults (male or female participants reported by self), 18years of age or older, who have received two doses of either BBV152 or ChAdOx1 nCoV-19, 4months before the screening visit. Additionally, the participants should have had no known COVID-19 infection at least 3months before the screening visit. Key exclusion criteria included pregnant or lactating mothers, individuals with illnesses that in the opinion of the investigator may affect safety, and the immunocompromised. Detailed inclusion and exclusion criteria are provided in the protocol (Supplementary Information). Participants were screened on the basis of medical history, vital signs and physical examination before enrollment. Eligible participants provided signed informed consent forms at enrollment. Participants were compensated at every visit for their time and cost of travel.

Participants who met the inclusion criteria and successfully completed all screening procedures were randomized in the study by using the interactive web response system (IWRS). Unique randomization codes were assigned to the participants and remained unchanged until the completion of the trial. The randomization codes were generated through Proc Plan using SAS version 9.4 or higher (SAS Institute) by an independent biostatistician. The final randomization list was filed securely by the independent biostatistician and accessible to authorized persons only. Participants were enrolled by investigators with the help of the IWRS.

In phase 3, consecutive 420 in the immunogenicity cohort were randomized in a 2:1 ratio into GEMCOVAC-OM and ChAdOx1 nCoV-19 by stratified block randomization through the IWRS. A randomization code was assigned to each participant in sequence in the order of enrollment, and then the participants received the investigational products labeled with the same code. This was an open-label study, and no masking was performed.

GEMCOVAC-OM consists of an in vitro transcribed mRNA encoding for the spike protein of the Omicron variant of the SARS-CoV-2 virus and cationic lipid nano-emulsion in a buffer containing 10% sucrose in 10mM sodium citrate, pH 6.5 (ref. 45). The complete antigenic sequence that was used has been published in the DDBJ database (accession no. LC769018). GEMCOVAC-OM, 10g in 0.1ml, was administered intradermally using a Tropis needle-free injection system (PharmaJet). More information on the vaccine, mRNA platform and development can be found in Supplementary Information.

ChAdOx1 nCoV-19 (COVISHIELD), the comparator vaccine in phase 3, consisted of Corona Virus Vaccine (Recombinant) 51010 viral particles. This vaccine is based on recombinant, replication-deficient chimpanzee adenovirus vector encoding the SARS-CoV-2 spike glycoprotein, produced in genetically modified human embryonic kidney 293 cells. ChAdOx1 nCoV-19 was administered intramuscularly.

GEMCOVAC-19 consists of an in vitro transcribed mRNA encoding for the spike protein of the D614G variant of the SARS-CoV-2 virus and cationic lipid nano-emulsion in a buffer containing 10% sucrose in 10mM sodium citrate, pH 6.5. The complete antigenic sequence that was used has been published in the DDBJ database (accession no. LC776732.1). GEMCOVAC-19, 10g in 0.5ml, was administered intramuscularly.

The participants were screened on visit 1 (day 1), which included a validated reverse transcription polymerase chain reaction (RTPCR) for SARS-CoV-2. Regardless of the outcome of the RTPCR, the participants who fit the inclusion criteria were enrolled and the vaccine was administered on the same day. Those found to be RTPCR positive would be excluded from the immunogenicity analysis to avoid confounding. Importantly, during the trial, in India, a third dose of BBV152 or ChAdOx1 nCoV-19 (precautionary dose) was approved for participants who had received primary doses of BBV152 or ChAdOx1 nCoV-19, respectively. However, individuals who had taken two doses of BBV152 as their primary vaccination were not eligible for a third dose of ChAdOx1 nCoV-19. Keeping in line with these vaccination guidelines, participants with ChAdOx1 nCoV-19 as their primary vaccination were randomized to get ChAdOx1 nCoV-19 or GEMCOVAC-OM, whereas participants with BBV152 as their primary vaccination received GEMCOVAC-OM only in the clinical trial.

Participants were provided an e-diary or a paper diary to record the solicited AEs till day 7 and unsolicited AEs as well as concomitant medication taken, if any, till the end of the study. A telephone call was placed to all the participants at day 7 to record any additional AEs, if any. Participants visited the study site for visit 2 (day 29+7), visit 3 (day 90+14) and visit 4 (day 180+14). Blood for assessing immunogenicity was drawn at visit 1 before vaccination (baseline), day 29 and day 90. Safety was assessed throughout the duration of the study.

In phase 2, the primary endpoints were to compare the safety and anti-spike IgG antibodies between the two vaccinated arms at day 29. Secondary endpoints included comparison of seroconversion as assessed by 2-fold rise in anti-spike IgG antibody titers from baseline, percentage neutralization by a surrogate neutralization (cPass) assay and cellular immune responses at day 29. Exploratory endpoints included comparison of anti-spike IgG antibodies, percentage neutralization by cPass assay and cellular immune responses at day 90.

In phase 3, the primary endpoint was the demonstration of noninferiority of neutralizing antibody GMT assessed by a plaque reduction neutralization test (PRNT50) assay in terms of LSGMR at day 29 and difference in seroconversion (2-fold rise in antibody titers at day 29 from baseline) between GEMCOVAC-OM and ChAdOx1 nCoV-19. Secondary endpoints included comparison of safety, LSGMR and seroconversion in terms of anti-spike IgG antibody titers, percentage neutralization by a surrogate virus neutralization assay (cPass assay, GenScript) and cell-mediated immunity assessment by intracellular cytokine expression at day 29. Exploratory endpoints included humoral and cellular immune response assessment at day 90.

Although the trial was open-label, laboratory analysis was conducted in a blinded manner. Measurements were taken from distinct samples. Information on the materials used is provided in detail in Supplementary Information.

Neutralizing antibody titers were assessed by the PRNT50 assay at the Interactive Research School for Health Affairs (IRSHA, Bharati Vidyapeeth, Deemed to be University, Pune) that was previously developed46 and then optimized for the BA.1 Omicron variant of SARS-CoV-2 (SARS-CoV-2-IND/0005/2022; B.1.1.529.1 lineage). In brief, Vero E6 cells were initially seeded at a density of 1105 cellsml1 in 24-well plates using Minimum Essential Medium (MEM) containing 10% fetal bovine serum (FBS) and antibiotics and allowed to incubate overnight at 37C with 5% CO2. Serum samples, initially diluted at 1:5 ratios, were subjected to heat inactivation for 30min at 56C. Subsequently, a fourfold serial dilution was executed, and these serum dilutions were mixed in equal proportions with the SARS-CoV-2 virus with a titer ranging from 600 to 1,000 plaque-forming units (pfu) per milliliter. The serumvirus mix was then incubated for 1h at 37C within a humidified incubator with 5% CO2. Following incubation, 100l of the resultant mixture was introduced into duplicate wells of the seeded 24-well plate and subjected to an additional 3-h incubation at 37C in a humidified incubator with 5% CO2. Then, 1ml of an overlay medium, constituting MEM, Aquacide-II, 2% FBS and antibiotics, was added to the Vero cell monolayer. Plates were then incubated for 6days at 37C within a humidified incubator with 5% CO2. At the end of this incubation period, the overlay medium was removed, and cells were fixed through the application of 3.7% formaldehyde. After washing with phosphate-buffered saline, cells were stained using 1% crystal violet. Plates were washed once more and air-dried. Viral plaques were counted using the C.T.L. ImmunoSpot platform. PRNT50 titers were determined using standard logistic regression model. Neutralization was also assessed using a semi-quantitative surrogate virus neutralization assay (cPass, GenScript)47 for the BA.1 variant.

Anti-IgG responses against the spike glycoprotein of B.1.1.529 Omicron variant of SARS-CoV-2 was assessed by an in-house developed indirect ELISA. In brief, 96-well ELISA plates (Nunc Maxisorp) coated with spike protein (full length from Sino Biologicals, 40589-V08H26) were washed thrice with phosphate-buffered saline-Tween (PBS-T). Plates were blocked with 3% nonfat dried milk. Diluted sera samples were added to the blocked plates and incubated at room temperature for 2h. Plates were then washed thrice with PBS-T and incubated at room temperature with detection antibody (1:5,000), anti-human IgG (Fc region specific from Sigma A0170) for 1h at room temperature. After secondary antibody incubation, plates were washed thrice with PBS-T before addition of TMB substrate. Color development was quenched with 3M HCl after 20min of incubation at room temperature. Plates were read at 450nm using a plate reader. The assay background was calculated from the 10s.d. added to the average of the readouts where there was no sample but diluent in the wells. For all samples, IgG titers were an interpolation of previously calculated assay background in 5 parameter logistic fit of sample dilution versus absorbance.

PBMCs were isolated using BD Vacutainer CPT with sodium citrate tubes following the manufacturers guidelines and subsequently cryopreserved in liquid nitrogen. For immune-phenotyping purposes, frozen PBMCs were thawed and allowed to rest in complete RPMI 1640 culture medium (CRPMI) supplemented with 10% FBS, 100Uml1 penicillin and 0.1mgml1 streptomycin (1 pen-strep) for 1822h. Gating strategies for both T cell and B cell experiments are given in Extended Data Fig. 3. In T cell response analysis, intracellular cytokine staining (ICS) was performed using 0.5 million PBMCs in 100l CRPMI medium per well in a V-bottom plate. These cells were stimulated with a 1gml1 epitope mapping 15-mer peptide pool derived from the Omicron B.1.1.529/BA.1 spike glycoprotein peptides, specifically the PepTivator SARS-CoV-2 Prot S B.1.1.529/BA.1 Mutation Pool. Stimulation was carried out in the presence of 1gml1 BD FastImmune (anti-CD28/49d antibody)48 for 6h, with the addition of 1l of Brefeldin-A during the final 4h of stimulation. After stimulation, PBMCs were washed and subjected to surface and ICS staining using antibodies targeting CD3 PE-Cy7 (BD 557851, clone SK7, 1:20), CD4 BV480 (BD 566104, clone SK3, 1:20), CD8 FITC (BD 555366, clone RPA-T8, 1:5), IFN PE (BD 559327, clone B27, 1:5), TNF APC (BD 551384, clone MAb11, 1:5), IL-2 BV421 (BD 562914, clone 5344.111, 1:20), IL-2 BV786 (BD 564113, clone MP4-25D2, 1:10), IL-13 BV711 (BD 564288, clone JES10-5A2, 1:10) and CD19 PerCP-Cy5.5 (BD 561295, clone HIB19, 1:20) markers. ICS was executed utilizing the BD cytofix/cytoperm kit following the manufacturers instructions. Antibody incubation was carried out for 30min at 4C. Phorbol myristate acetate (PMA)ionomycin was used as positive control. To assess the B cell population specific to the B.1.1.529 spike protein, PBMCs were initially labeled with biotinylated spike protein specific to Omicron B.1.1.529. Subsequently, surface staining was performed with common surface markers CD3 BV605 (BD 563219, clone SK7, 1:20), CD19 PerCP-Cy5.5 (BD 561295, clone HIB19, 1:20), CD20 APC-H7 (BD 560734, clone 2H7, 1:20). Following staining and washing steps, PBMCs were resuspended in fluorescence-activated cell sorting buffer, acquired using the FACSLyric system (BD Biosciences), and analyzed using FlowJo software version 10.8.1 (FlowJo LLC, BD Biosciences).

Solicited events data were captured up to 7days after booster vaccine administration through an electronic or paper diary. Local solicited events included pain, redness, swelling, warmth, pruritus and bruising. Systemic solicited events included fever, headache, myalgia, arthralgia, fatigue, malaise, nausea and chills. Unsolicited events were assessed throughout the duration of the study. AE terms were coded using Medical Dictionary for Regulatory Activities. These AEs were graded on the basis of the Division of AIDS criteria49. Myocarditis was considered as an AE of special interest; site investigators were asked to thoroughly evaluate participants with any symptoms of chest pain, breathlessness or palpitations.

No formal sample size calculations were performed for phase 2. Phase 3 consisted of a safety and an immunogenicity cohort. The safety cohort consisted of 3,140 participants of whom 3,000 were included in the GEMCOVAC-OM arm. The immunogenicity cohort was analyzed for two primary endpoints based on World Health Organization guidelines50, with individuals randomized to GEMCOVAC-OM and ChAdOx1 nCoV-19 in a 2:1 ratio. A sample size of 420 (280 in GEMCOVAC-OM and 140 in ChAdOx1 nCoV-19) was found adequate for assessing noninferiority of neutralizing antibody titers if the lower limit of the two-sided 95% CI of the LSGMR (GMTGEMCOVAC-OM/GMTChAdOx1 nCoV-19) was >0.67 considering a standard deviation of 1.82, alpha error of 5%, power of 90% and dropout rate of 20%. A sample size of 381 (254 in GEMCOVAC-OM and 127 in ChAdOx1 nCoV-19) was found adequate for assessing the noninferiority of seroconversion difference considering a margin of 10%, alpha error of 5%, power of 90% and dropout rate of 20%. The sample size of 420 (280 in GEMCOVAC-OM and 140 in ChAdOx1 nCoV-19 arm) was considered in this study to provide adequate numbers for the statistical analysis of both the primary endpoints.

All immunogenicity analysis was performed in the full analysis set (intention to treat) population. The observed GMT and associated 95% CIs (ClopperPearson method) were calculated on the basis of log-transformed antibody titers. The rise in neutralizing antibody titers from baseline to day 29 was compared using a paired t-test. The LSGMR of neutralizing antibody titers in both the arms at day 29 from the PRNT50 (BA.1 strain of SARS-CoV-2) assay was calculated using ANCOVA with baseline titers as covariates. If the lower limit of the two-sided 95% CI of the LSGMR was >0.67, GEMCOVAC-OM would be considered noninferior to ChAdOx1 nCoV-19. Seroconversion in terms of neutralizing antibody titers from PRNT50 assay was defined as a 2-fold rise in titers at day 29 from baseline. The difference in seroconversion was determined using the MeitinenNurminen method. If the lower bound of the two-sided 95% CI for seroconversion difference was >10%, GEMCOVAC-OM would be considered noninferior to ChAdOx1 nCoV-19.

Similarly, for the secondary endpoint, LSGMR of anti-Spike IgG antibody titers from ELISA at day 29 was assessed using ANCOVA, with baseline titers as covariates. The 95% CI was calculated for percentage by using the ClopperPearson method. The difference in seroconversion at day 29 for anti-spike IgG antibodies from ELISA were calculated using the MeitinenNurminen method. The difference in the change in mean percentage neutralization from baseline to day 29, assessed by the cPass assay, was analyzed using ANCOVA with baseline neutralization considered as a covariate. In cell-mediated immunity, change in expression from baseline to day 29 was assessed by either a two-sided paired t-test or Wilcoxon signed-rank test depending on normality of the data. Expression at day 29 was compared using a two-sided t-test or Wilcoxon rank sum base based on normality. The normality of the data was assessed using a ShapiroWilk test. This was also performed for subgroups based on their primary vaccination of either BBV152 or ChAdOx1 nCoV-19.

Sex-disaggregated analysis was conducted for humoral and safety data from phase 3. For humoral immunogenicity, ANCOVA was used to compare the differences in neutralizing antibody and anti-spike IgG between men and women at days 29 and 90, using baseline titers as covariates. For safety, unadjusted OR was calculated along with the 95% CI.

Data were collected using Clinion (version 3.1). Statistical analysis for humoral immunogenicity was performed using Statistical software SAS version 9.4 (SAS Institute). Figures were generated using GraphPad Prism (Version 9.5.1). The Spearman rank correlation coefficient (denoted as r) was computed for all pairs of parameters utilizing the corrplot package (version 0.92) within RStudio (version 2022.12.0.0). To complement the correlogram, two-tailed P values associated with Spearman rank correlations were calculated through the corr.mtest function and visualized using the corrplot function.

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Originally posted here:

An Omicron-specific, self-amplifying mRNA booster vaccine for COVID-19: a phase 2/3 randomized trial - Nature.com