Immunogenicity and reactogenicity of intradermal mRNA-1273 SARS-CoV-2 vaccination: a non-inferiority, randomized-controlled trial

Fractional dosing can be a cost-effective vaccination strategy to accelerate individual and herd immunity in a pandemic. We assessed the immunogenicity and safety of primary intradermal (ID) vaccination, with a 1/5th dose compared with the standard intramuscular (IM) dose of mRNA-1273 in SARS-CoV-2 naïve persons. We conducted an open-label, non-inferiority, randomized controlled trial in the Netherlands between June and December 2021. One hundred and fifty healthy and SARS-CoV-2 naïve participants, aged 18–30 years, were randomized (1:1:1) to receive either two doses of 20 µg mRNA-1273 ID with a standard needle (SN) or the Bella-mu® needle (BM), or two doses of 100 µg IM, 28 days apart. The primary outcome was non-inferiority in seroconversion rates at day 43 (D43), defined as a neutralizing antibody concentration threshold of 465 IU/mL, the lowest response in the IM group. The non-inferiority margin was set at −15%. Neutralizing antibody concentrations at D43 were 1789 (95% CI: 1488–2150) in the IM and 1263 (951–1676) and 1295 (1020–1645) in the ID-SN and ID-BM groups, respectively. The absolute difference in seroconversion proportion between fractional and standard-dose groups was −13.95% (−24.31 to −3.60) for the ID-SN and −13.04% (−22.78 to −3.31) for the ID-BM group and exceeded the predefined non-inferiority margin. Although ID vaccination with 1/5th dose of mRNA-1273 did not meet the predefined non-inferior criteria, the neutralizing antibody concentrations in these groups are far above the proposed proxy for protection against severe disease (100 IU/mL), justifying this strategy in times of vaccine scarcity to accelerate mass protection against severe disease.


Supplement B: Prespecified stopping rules
The following stopping rules were in place for all participants, based on review of diary reactogenicity and AE data.
Stopping Rule Criteria Safety: -If any participant vaccinated with mRNA-1273 (at any dose level) develops an SAE or SUSAR that is assessed by the investigator as possibly related, or for which there is no alternative, plausible, attributable cause.
-If any participant vaccinated with mRNA-1273 (at any dose level) develops a Grade 4 local reaction within 14 days after vaccination that is assessed as possibly related by the investigator, or for which there is no alternative, plausible, attributable cause.
-If any participant vaccinated with mRNA-1273 (at any dose level) develops a Grade 4 systemic event within 7 days after vaccination that is assessed as possibly related by the investigator, or for which there is no alternative, plausible, attributable cause -If two participants vaccinated with mRNA-1273 (within the same dose level) develop a fever >40.0°C for at least one daily measurement within 7 days after vaccination that is assessed as possibly related by the investigator, or for which there is no alternative, plausible, attributable cause.
-If any two participants vaccinated with mRNA-1273 (at any dose level) report the same or similar severe (Grade 3) AE within 14 days after vaccination, assessed as possibly related by the investigator, or for which there is no alternative, plausible, attributable cause.
-If any participant dies or requires ICU admission due to SARS-CoV-2 infection; if this stopping rule is met, all available clinical and preclinical safety and immunogenicity data should be reviewed to evaluate for enhanced COVID-19 disease.
-If part 2 is unsafe and ≤7 out of 10 participants of part 1 developed virus neutralizing antibody titers (PRNT80) ≥128 at day 43, The anti-Spike IgG antibody titers may be used as a proxy for the neutralizing antibody titers if the results of the neutralizing antibody assay are delayed.

Local reactions
Redness and swelling were measured and recorded in centimeters and categorized as absent, mild, moderate, or severe based on the grading scale (Supplementary Table 1).Pain at the injection site was assessed by the participant as absent, mild, moderate, or severe according the grading scale in Supplementary Table 1.

Systemic events
Solicited systemic events consisted of vomiting, diarrhoea, headache, fatigue, chills, new or worsened muscle pain, and new or worsened joint pain.The symptoms were assessed by the participant as absent, mild, moderate, or severe according to the grading scale in Supplementary Table 2.

Fever
Participants were instructed on how to measure oral temperature at home.Daily temperature measurements were registered in the diary in the morning and at any time when fever is suspected during the diary data collection periods.Fever was defined as an oral temperature of ≥38.0°C.The highest temperature for each day was recorded in the diary.Temperature was measured and recorded to 1 decimal place and then categorized during analysis according to the scale shown in Supplementary Table 3.  Diary for the solicited adverse events and concomitant medication.Participants were instructed to record their temperature and adverse events daily for fourteen days after each vaccination
All centrifugation steps of non-fixed samples were performed at 450g.Cells were transferred to a 96-wells well, washed with PBS and then resuspended in 100 µl PBS containing 1/500 diluted LIVE/DEAD fixable Blue Dead Cell Stain (Invitrogen Cat#L23105) and 1/50 diluted Fc Receptor binding inhibitor (eBioscience cat# 14-9161-73) for 15 minutes at room temperature (RT) in the dark.Then, samples were washed with PBS containing 2 mM EDTA and 0.5% BSA followed by resuspension in 100 µl of extracellular antibody cocktail (see Supplementary Table 4) for 15 minutes at RT in the dark (see below).This cocktail includes streptavidin-coupled BV421 (BD, cat#563259) as a 'decoy probe' to remove B-cells that react with streptavidin.Then stained samples were resuspended into 100 µl of 2 µg/ml of BUV615-labeled and BUV661-labeled SARS-CoV-2-spike-specific tetramers for 30 minutes on ice in the dark.For the intracellular antibody staining, cells were first resuspended in FoxP3 fixation/permeabilization solution (Invitrogen, cat#00-5521-00) and incubated for 30 minutes on ice in the dark, followed by three washes with 1x permeabilization Buffer (Invitrogen, cat#00-8333-56).All centriguation steps of fixed samples were performed at 800g.Then, samples were resuspended in intracellular antibody cocktail, containing Ki67-BV711 (Biolegend, cat#350515), IRF4-APC (Miltenyi, cat#130-100-915) and caspase-3-V450 (BD, cat#560627).Fixed cells were washed with PBS containing 2 mM EDTA and 0.5% BSA and 3-5x10 6 events were acquired on a Cytek Aurora 5L spectral flow cytometer and unmixed using SpectroFlo software.See Supplementary Table 4 for details on the type of single-stain reference controls used for unmixing.FLOWjo software was used to gate CD19+ B-cells and OMIQ data analysis software was used for further analysis (www.omiq.ai).Cytonorm was used for batch corrections followed by Uniform Manifold Approximation and Projection (UMAP) dimensionality reduction to visualize the phenotypes of SARS-CoV-2-spike-specific B-cells.
For gating strategies for SARS-CoV-2-specific B-cells see Supplementary Figure 2.
Generation of SARS-CoV-2 spike labelled tetramers Fluorescently labelled tetramers were generated by labeling biotinylated SARS-CoV-2 spike protein (R&D, cat#10549-050) with streptavidin-coupled BUV615 (BD, cat#613013) or streptavidin-coupled BUV661 (BD, cat#612979).SARS-CoV-2 spike protein was resuspended in 100 ul PBS to get a 500 ug/ml stock mixture (3730 pmol/ml).SARS-CoV-2 spike protein was then diluted two times to get a 1865 pmol/ml user solution.Biotinylated SARS-CoV-2 spike protein was then mixed in an approximate 4:1 molecular ratio with each of the 2 streptavidincoupled fluorochromes (at 100 ug/ml).Incubation was performed on ice in a stepwise approach where 1/10 th fraction of streptavidin-coupled fluorochrome was added to the biotinylated SARS-CoV-2-spike protein every 10 minutes.At 50 and 100 minutes the tetramers were spun down using a short pulse-spin.After the final incubation, free biotin was added to a final concentration of 30 uM and incubated for 30 min on ice to block all unbound streptavidin-coupled fluorochromes.Labeled SARS-CoV-2 spike tetramers (91 ug/ml) were stored in the dark at 4ᴼC and used within 2 weeks.

Analysis of SARS-CoV-2-spike-specific T-cells using flow cytometry
In addition, we obtained PBMCs at day 1, 29, 43 and month 7 after ID and IM administration.PBMCs were isolated from fresh whole blood using Ficoll-Isopaque and cryopreserved until further use.Upon thawing, the PBMCs were slowly diluted in culture medium consisting of Iscove Modified Dulbecco Medium (IMDM; Lonza) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Sigma-Aldrich), 2.7 mM L-glutamine (Lonza), 100 U/mL penicillin (Lonza) and 100µg/mL streptomycin (Lonza) (1% p/s).After thawing and washing, PBMCs were treated with 1.33 mg/ml DNAse to minimize cell clumping, counted and used for T-cell stimulation assay as well as for peptide-HLA tetramer staining.For the T-cell stimulation assay, up to 2x10⁶ PBMCs were seeded in 100 µL culture medium and stimulated with 15-mer peptides with 11 amino acid overlap which cover the whole SARS-CoV-2 spike antigen (SB peptide, France) in 96-well round bottomed plates.The peptides were dissolved in 20% dimethyl sulfoxide (DMSO) in ddH2O and added to the wells in a 1 µg/mL final concentration.As a negative control, DMSO and ddH2O in the same concentration was added.As a positive control, a pool of peptides from CMV, EBV, Flu and extended (CEFX) was dissolved in 20% DMSO in ddH2O and in a final concentration of 0.25 µg/mL (see Supplementary Table 5 for CEFX peptide pool details).After one hour incubation (37°C, 5% CO2), 5 µg/mL Brefeldin A was added to the well and the plate was further incubated for 15 hours.The stimulation reaction was stopped by washing the cells in PBS followed by viability staining using Zombie-Red.After a PBS wash, cells were fixated and permeabilized using the FOXP3 buffer kit (Thermo Fisher).20 µL antibody staining mix was added containing 0.8 mg/mL albumin, Brilliant Stain Buffer Plus and antibodies directed against CD3, CD4, CD8, CD154, CD137, CD69, IFN-γ, TNF-α, IL-2, IL-4, IL-17, PD-1, FOXP3 and CXCR5 (see Supplementary Table 6 for product details).After incubation for 30 minutes at room temperature (RT), the cells were washed in PBS containing 0.8 mg/mL albumin (FACS buffer) and dissolved in 100 µL FACS buffer for measurement on a 3-laser aurora (Cytek Biosciences).
For the tetramer staining, up to 2x10⁶ PBMCs were incubated for 16 hours in 100 µL culture medium.The cells were washed in FACS buffer and stained in two steps.First, 10 µL of an antibody cocktail directed against CD4, CCR7, CD45RA, PD-1 was added together with a tetramer pool in FACS buffer and incubated for 15 minutes at RT. Second, 10 µL of FACS buffer containing CD8 APC-H7 was incubated for an additional 15 minutes at RT (see Supplementary Table 6 for product detail).The tetramer pool contained 23 in-house-made tetramers, consisting of spike peptides and HLAs, conjugated to PE as well as APC (see Supplementary Table 7 for product detail).Cells were washed and dissolved in 100 µL FACS buffer for measurement on a 3-laser aurora (Cytek Biosciences).
For flow cytometry analysis, OMIQ (www.omiq.ai)was used to set gates and retrieve percentages (see Supplementary Figure 5 and S3 for gating strategy).Due to optimal reference controls used for unmixing on the 3-laser Cytek Aurora, no compensation was needed.The same gate was applied to all samples within one donor and as much as possible between donors.Percentage SARS-CoV-2 spike reactive CD4 + T-cells was identified as CD137 + and/or CD154 + cells of total CD4 + T-cells, corrected for background in DMSO.Percentage SARS-CoV-2-spike reactive CD8 + T-cells was identified as CD137 + and/or CD69 + of total CD8 + T-cells.The same CD137 + and/or CD154 + or CD137 + and/or CD69 + was applied to all samples unless the background in DMSO was 0.1% or higher.Then the gate was adapted and applied to all samples of that donor.
For the analysis, percentage SARS-CoV-2-spike CD4 + T-cells was removed if there were less than 10.000 events in CD4 gate and percentage SARS-CoV-2-spike CD8 + T-cells was removed from analysis if there were less than 10.000 events in CD8 gate.A threshold was set on 0.05% for frequency CD154 + and/or CD137 + of total CD4 + Tcells,0.005%for frequency CD137 + and/or CD69 + AND IFN-γ + and/or TNF-α + of total CD8 + T-cells and 0.005% for frequency of spike/HLA tetramer + of total CD8 + T-cells.Further gating to calculate percentage of spikespecific cells that are IFN-γ, TNF-α, IL-2, IL-4, IL-17, Tfh (T follicular helper cells; CXCR5 + PD-1 + ), PD-1 (CXCR5 -PD-1 + ) or Treg (FOXP3 + IFN-γ -TNF-α -) positive was only done if frequency threshold was met and if there were more than 25 events in the CD154 + and/or CD137 + gate for CD4 + T-cells and CD137 + and/or CD69 + for the CD8 + T-cells.Percentage of naïve (CCR7+CD45RA+), central memory (CM; CCR7+CD45RA-), effector memory (EM; CCR7-CD45RA-) or terminal effector memory (TEMRA; CCR7-CD45RA+) was measured of spike/HLA-tetramer + CD8 + T-cells and only calculated if the frequency of spike/HLA-tetramer + CD8 + T-cells was above threshold and contained at least 10 events.The percentages were exported and further analyzed in Graphpad Prism 9.0.1.All timepoints of the same patient were measured simultaneously to minimize technical variance within one patient.Patients were measured and analyzed in random order to minimize technical variance and bias between cohorts.In the last analysis step the data was separated for the different cohorts.

Supplement J: SARS-CoV-2 specific T-cell responses
Since the frequency of spike-specific T-cell responses was similar between the ID-SN and IM group, the function of the SARS-CoV-2-reactive T-cells was further evaluated by measuring the percentages of SARS-CoV-2-spikespecific Th1 (IFN-γ, TNF-α or IL-2), Th2 (IL-4), Th17 (IL-17), Tfh (PD1 + CXCR5 + ), and Treg (IFN-γ -TNF-α - FOXP3 + ).The profile of cytokine production was different between the CD4 + and CD8 + T-cells (Supplementary Figure 10A and 10B).IL-2 responses were not seen in CD8 + T-cells.In both groups, the spike-specific CD4 + Tcells primarily exhibited a Th1 cell response, as they mostly produced IFN-γ, TNF-α and IL-2.In addition, increased expression of PD-1 was observed.Slight reductions were observed in TNF-α and IL-2 expression in the IM group compared to the ID-SN group, with only the latter being significant (Supplementary Figure 10A).The frequency of SARS-CoV-2-reactive CD4 + and CD8 + T-cells that produced IL-2, IFN-γ and TNF-α was similar between the two groups.The SARS-CoV-2-specific CD8 + T-cells induced by ID and IM produced both IFN-γ and TNF-α in most individuals and showed increased PD-1 expression (Supplementary Figure 10B).None of these parameters were significantly different between the two cohorts.
Finally, the differentiation state of vaccine-induced CD8 + T-cells was studied, and showed that in both IM and ID-SN groups, SARS-CoV-2-specific CD8 + T-cells, measured by pHLA tetramer binding, mainly appear to have an effector memory (EM) phenotype (Supplementary Figure 10C).13.Adverse events related to vaccine administration after vaccination one and two.