The 45 enrolled participants received their first vaccination between March 16 and April 14, 2020 (Fig. S1). Three participants did not receive the second vaccination, including one in the 25-μg group who had urticaria on both legs, with onset 5 days after the first vaccination, and two (one in the 25-μg group and one in the 250-μg group) who missed the second vaccination window owing to isolation for suspected Covid-19 while the test results, ultimately negative, were pending. All continued to attend scheduled trial visits. The demographic characteristics of participants at enrollment are provided in Table 1.
No serious adverse events were noted, and no prespecified trial halting rules were met. As noted above, one participant in the 25-μg group was withdrawn because of an unsolicited adverse event, transient urticaria, judged to be related to the first vaccination.
The severity of solicited adverse events was graded as mild, moderate, or severe (see Table S1).
After the first vaccination, solicited systemic adverse events were reported by 5 participants (33%) in the 25-μg group, 10 (67%) in the 100-μg group, and 8 (53%) in the 250-μg group; all were mild or moderate in severity (Figure 1 and Table S2). Solicited systemic adverse events were more common after the second vaccination and occurred in 7 of 13 participants (54%) in the 25-μg group, all 15 in the 100-μg group, and all 14 in the 250-μg group, with 3 of those participants (21%) reporting one or more severe events.
None of the participants had fever after the first vaccination. After the second vaccination, no participants in the 25-μg group, 6 (40%) in the 100-μg group, and 8 (57%) in the 250-μg group reported fever; one of the events (maximum temperature, 39.6°C) in the 250-μg group was graded severe. (Additional details regarding adverse events for that participant are provided in the Supplementary Appendix.)
Local adverse events, when present, were nearly all mild or moderate, and pain at the injection site was common. Across both vaccinations, solicited systemic and local adverse events that occurred in more than half the participants included fatigue, chills, headache, myalgia, and pain at the injection site. Evaluation of safety clinical laboratory values of grade 2 or higher and unsolicited adverse events revealed no patterns of concern (Supplementary Appendix and Table S3).
SARS-CoV-2 Binding Antibody Responses
Shown are geometric mean reciprocal end-point enzyme-linked immunosorbent assay (ELISA) IgG titers to S-2P (Panel A) and receptor-binding domain (Panel B), PsVNA ID50 responses (Panel C), and live virus PRNT80 responses (Panel D). In Panel A and Panel B, boxes and horizontal bars denote interquartile range (IQR) and median area under the curve (AUC), respectively. Whisker endpoints are equal to the maximum and minimum values below or above the median ±1.5 times the IQR. The convalescent serum panel includes specimens from 41 participants; red dots indicate the 3 specimens that were also tested in the PRNT assay. The other 38 specimens were used to calculate summary statistics for the box plot in the convalescent serum panel. In Panel C, boxes and horizontal bars denote IQR and median ID50, respectively. Whisker end points are equal to the maximum and minimum values below or above the median ±1.5 times the IQR. In the convalescent serum panel, red dots indicate the 3 specimens that were also tested in the PRNT assay. The other 38 specimens were used to calculate summary statistics for the box plot in the convalescent panel. In Panel D, boxes and horizontal bars denote IQR and median PRNT80, respectively. Whisker end points are equal to the maximum and minimum values below or above the median ±1.5 times the IQR. The three convalescent serum specimens were also tested in ELISA and PsVNA assays. Because of the time-intensive nature of the PRNT assay, for this preliminary report, PRNT results were available only for the 25-μg and 100-μg dose groups.
Binding antibody IgG geometric mean titers (GMTs) to S-2P increased rapidly after the first vaccination, with seroconversion in all participants by day 15 (Table 2 and Figure 2A). Dose-dependent responses to the first and second vaccinations were evident. Receptor-binding domain–specific antibody responses were similar in pattern and magnitude (Figure 2B). For both assays, the median magnitude of antibody responses after the first vaccination in the 100-μg and 250-μg dose groups was similar to the median magnitude in convalescent serum specimens, and in all dose groups the median magnitude after the second vaccination was in the upper quartile of values in the convalescent serum specimens. The S-2P ELISA GMTs at day 57 (299,751 [95% confidence interval CI, 206,071 to 436,020] in the 25-μg group, 782,719 [95% CI, 619,310 to 989,244] in the 100-μg group, and 1,192,154 [95% CI, 924,878 to 1,536,669] in the 250-μg group) exceeded that in the convalescent serum specimens (142,140 [95% CI, 81,543 to 247,768]).
SARS-CoV-2 Neutralization Responses
No participant had detectable PsVNA responses before vaccination. After the first vaccination, PsVNA responses were detected in less than half the participants, and a dose effect was seen (50% inhibitory dilution [ID50]: Figure 2C, Fig. S8, and Table 2; 80% inhibitory dilution [ID80]: Fig. S2 and Table S6). However, after the second vaccination, PsVNA responses were identified in serum samples from all participants. The lowest responses were in the 25-μg dose group, with a geometric mean ID50 of 112.3 (95% CI, 71.2 to 177.1) at day 43; the higher responses in the 100-μg and 250-μg groups were similar in magnitude (geometric mean ID50, 343.8 [95% CI, 261.2 to 452.7] and 332.2 [95% CI, 266.3 to 414.5], respectively, at day 43). These responses were similar to values in the upper half of the distribution of values for convalescent serum specimens.
Before vaccination, no participant had detectable 80% live-virus neutralization at the highest serum concentration tested (1:8 dilution) in the PRNT assay. At day 43, wild-type virus–neutralizing activity capable of reducing SARS-CoV-2 infectivity by 80% or more (PRNT80) was detected in all participants, with geometric mean PRNT80 responses of 339.7 (95% CI, 184.0 to 627.1) in the 25-μg group and 654.3 (95% CI, 460.1 to 930.5) in the 100-μg group (Figure 2D). Neutralizing PRNT80 average responses were generally at or above the values of the three convalescent serum specimens tested in this assay. Good agreement was noted within and between the values from binding assays for S-2P and receptor-binding domain and neutralizing activity measured by PsVNA and PRNT (Figs. S3 through S7), which provides orthogonal support for each assay in characterizing the humoral response induced by mRNA-1273.
SARS-CoV-2 T-Cell Responses
The 25-μg and 100-μg doses elicited CD4 T-cell responses (Figs. S9 and S10) that on stimulation by S-specific peptide pools were strongly biased toward expression of Th1 cytokines (tumor necrosis factor α > interleukin 2 > interferon γ), with minimal type 2 helper T-cell (Th2) cytokine expression (interleukin 4 and interleukin 13). CD8 T-cell responses to S-2P were detected at low levels after the second vaccination in the 100-μg dose group (Fig. S11).