The n e w e ng l a n d j o u r na l of m e dic i n e

Original Article

Safety and Efficacy of the BNT162b2 mRNA

Covid-19 Vaccine through 6 Months
S.J. Thomas, E.D. Moreira, Jr., N. Kitchin, J. Absalon, A. Gurtman, S. Lockhart,
J.L. Perez, G. Pérez Marc, F.P. Polack, C. Zerbini, R. Bailey, K.A. Swanson, X. Xu,
S. Roychoudhury, K. Koury, S. Bouguermouh, W.V. Kalina, D. Cooper,
R.W. Frenck, Jr., L.L. Hammitt, Ö. Türeci, H. Nell, A. Schaefer, S. Ünal, Q. Yang,
P. Liberator, D.B. Tresnan, S. Mather, P.R. Dormitzer, U. Şahin, W.C. Gruber, and
K.U. Jansen, for the C4591001 Clinical Trial Group*​​

A BS T R AC T

BACKGROUND
BNT162b2 is a lipid nanoparticle–formulated, nucleoside-modified RNA vaccine The authors’ full names, academic de-
encoding a prefusion-stabilized, membrane-anchored severe acute respiratory syn- grees, and affiliations are listed in the Ap-
pendix. Dr. Dormitzer can be contacted
drome coronavirus 2 (SARS-CoV-2) full-length spike protein. BNT162b2 is highly at [email protected] or at
efficacious against coronavirus disease 2019 (Covid-19) and is currently approved, Pfizer, 401 N. Middletown Rd., Pearl River,
conditionally approved, or authorized for emergency use worldwide. At the time of NY 10965.

initial authorization, data beyond 2 months after vaccination were unavailable. *A list of the investigators in the
C4591001 Clinical Trial Group is pro-
METHODS vided in the Supplementary Appendix,
available at NEJM.org.
In an ongoing, placebo-controlled, observer-blinded, multinational, pivotal efficacy
trial, we randomly assigned 44,165 participants 16 years of age or older and 2264 This article was published on September 15,
2021, at NEJM.org.
participants 12 to 15 years of age to receive two 30-μg doses, at 21 days apart, of
BNT162b2 or placebo. The trial end points were vaccine efficacy against laboratory- DOI: 10.1056/NEJMoa2110345
Copyright © 2021 Massachusetts Medical Society.
confirmed Covid-19 and safety, which were both evaluated through 6 months after
vaccination.
RESULTS
BNT162b2 continued to be safe and have an acceptable adverse-event profile. Few
participants had adverse events leading to withdrawal from the trial. Vaccine ef-
ficacy against Covid-19 was 91.3% (95% confidence interval [CI], 89.0 to 93.2)
through 6 months of follow-up among the participants without evidence of previ-
ous SARS-CoV-2 infection who could be evaluated. There was a gradual decline in
vaccine efficacy. Vaccine efficacy of 86 to 100% was seen across countries and in
populations with diverse ages, sexes, race or ethnic groups, and risk factors for
Covid-19 among participants without evidence of previous infection with SARS-
CoV-2. Vaccine efficacy against severe disease was 96.7% (95% CI, 80.3 to 99.9). In
South Africa, where the SARS-CoV-2 variant of concern B.1.351 (or beta) was pre-
dominant, a vaccine efficacy of 100% (95% CI, 53.5 to 100) was observed.
CONCLUSIONS
Through 6 months of follow-up and despite a gradual decline in vaccine efficacy,
BNT162b2 had a favorable safety profile and was highly efficacious in preventing
Covid-19. (Funded by BioNTech and Pfizer; ClinicalTrials.gov number, NCT04368728.)

n engl j med  nejm.org 1

The New England Journal of Medicine
Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

T
he coronavirus disease 2019 (Covid-19) population (all participants ≥12 years of age)
pandemic continues, with recent estimates reported here.
of more than 187 million cases diagnosed Participants who were healthy or had stable
and more than 4 million deaths.1 Vaccines are chronic medical conditions were eligible. An ac-
currently available by means of full approval, tive immunocompromising condition or recent
conditional marketing approval, and emergency immunosuppressive therapy was an exclusion
use authorization pathways.2-5 BNT162b2 is a criterion. Participants with a history of Covid-19
lipid nanoparticle–formulated,6 nucleoside-mod- were excluded, although evidence of current or
ified RNA7 encoding the severe acute respiratory previous SARS-CoV-2 infection on laboratory test-
syndrome coronavirus 2 (SARS-CoV-2) full-length ing of trial-obtained samples was not an exclu-
spike glycoprotein in a prefusion stabilized con- sion criterion. Trial-related responsibilities and
formation.8 To date, more than 1 billion doses ethical conduct are summarized in the Supplemen-
of BNT162b2 have been distributed. tary Appendix, available with the full text of this
We previously reported safety and efficacy article at NEJM.org. The protocol contains addi-
data obtained through a median of 2 months of tional details of the trial and is available at
postimmunization follow-up from a global NEJM.org. The first draft of the manuscript was
phase 1–2–3 trial of BNT162b2 involving persons written by the fourth author. The authors had
16 years of age or older. Vaccine efficacy against the opportunity to review the data included in
Covid-19 was 95%. BNT162b2 had a favorable this article and confirm the accuracy of the data
safety profile in diverse populations.9 These data presented through the specified data cutoff date.
formed the basis for BNT162b2 emergency or The authors vouch for the accuracy and complete-
conditional authorizations globally.10 Safety, ef- ness of the data and for the fidelity of the trial to
ficacy, and immunogenicity data from partici- the protocol.
pants 12 to 15 years of age in this trial have been
reported.11 Here, we report safety and efficacy Procedures
findings from a prespecified analysis of the The participants were randomly assigned in a
phase 2–3 portion of the trial through approxi- 1:1 ratio to receive two 30-μg intramuscular
mately 6 months of follow-up. These additional injections, 21 days apart, of BNT162b2 (0.3 ml
data contributed to the full approval of BNT162b2 volume per dose) or saline placebo. Random-
in the United States. ization was performed with an interactive Web-
based system. Starting in December 2020, after
BNT162b2 became available under emergency or
Me thods
conditional use authorizations, participants 16
Objectives, Participants, and Oversight years of age or older who became eligible for
This randomized, placebo-controlled, observer- Covid-19 vaccination according to national or
blinded, phase 1–2–3 trial assessed the safety, local recommendations were given the option to
efficacy, and immunogenicity of the BNT162b2 learn their trial assignment. Those who had been
vaccine in adolescents and adults. The current randomly assigned to receive placebo were of-
report of the findings from the phase 2–3 portion fered BNT162b2. After unblinding of the group
of the trial focuses on safety assessments among assignments, participants were followed in an
participants 16 years of age or older and prespeci- open-label trial period.
fied assessments of vaccine efficacy among par-
ticipants 12 years of age or older through 6 months Safety
of follow-up after immunization. Because the en- Safety end points included solicited, prespecified
rollment of participants 12 to 15 years of age local reactions, systemic events, and antipyretic
began on October 15, 2020, 6-month postim- or pain medication use during the first 7 days
munization data are currently unavailable for after receipt of each vaccine or placebo dose,
this age cohort. Shorter-duration safety, immu- which were recorded in an electronic diary; unso-
nogenicity, and efficacy data for participants 12 licited adverse events after receipt of the first dose
to 15 years of age are reported separately11; through 1 month after the second dose; and seri-
however, data for this cohort are included in ous adverse events after receipt of the first dose
the analyses of vaccine efficacy in the overall through 1 and 6 months after the second dose

2 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Safety and Efficacy of BNT162b2 through 6 Months

was received. Safety data are presented for the od, with adjustment for surveillance time, which
blinded follow-up and open-label periods. accounts for potential differential follow-up be-
tween the two trial groups. As described in the
Efficacy statistical analysis plan, available with the pro-
BNT162b2 efficacy against laboratory-confirmed tocol, hypothesis-testing analyses were performed
Covid-19 with an onset of 7 days or more after with the use of a Bayesian approach, and the
the second dose was assessed and summarized descriptive analyses presented here were per-
descriptively in participants without serologic or formed with a frequentist approach for clarity of
virologic evidence of SARS-CoV-2 infection within communication. Because the percentage of par-
7 days after the second dose and in participants ticipants who reported symptoms but were miss-
with or without evidence of previous infection. ing a valid polymerase-chain-reaction test result
Efficacy against severe Covid-19 was also assessed. was small and slightly higher in the placebo group,
Lineages of SARS-CoV-2 detected in midturbinate data for these participants were not imputed in the
specimens are reported here for Covid-19 cases analysis.
that occurred 7 days or more after the second The previously reported primary efficacy ob-
dose in South African participants without evi- jective was achieved on the basis of an analysis of
dence of previous infection. Methods for deter- 170 accrued cases of Covid-19 that could be evalu-
mining SARS-CoV-2 lineages and case definitions ated (data cutoff date, November 14, 2020).9 The
for confirmed and severe cases of Covid-19 are current report provides updated efficacy analyses
summarized in the Supplementary Appendix. that were performed with data from cases that
had accrued up to March 13, 2021.
Statistical Analysis
The analysis populations are summarized in R e sult s
Table S1 in the Supplementary Appendix. Safety
analyses included participants 16 years of age or Participants
older without known human immunodeficiency Between July 27, 2020, and October 29, 2020, a
virus (HIV) infection who provided informed total of 45,441 participants 16 years of age or
consent and received at least one BNT162b2 or older underwent screening, and 44,165 underwent
placebo dose. The results of the safety analyses, randomization at 152 sites (130 sites in the
which are descriptive and not based on formal United States, 1 site in Argentina, 2 sites in Brazil,
hypothesis testing, are presented as counts, per- 4 sites in South Africa, 6 sites in Germany, and
centages, and associated Clopper–Pearson 95% 9 sites in Turkey) in the phase 2–3 portion of the
confidence intervals for adverse events, according trial. Of these participants, 44,060 received at
to terms in the Medical Dictionary for Regulatory Ac- least one dose of BNT162b2 (22,030 participants)
tivities, version 23.1, and reactogenicity events for or placebo (22,030), and 98% (21,759 in the
each trial group. Safety data that were reported BNT162b2 group and 21,650 in the placebo group)
up to March 13, 2021, are summarized here. The received the second dose (Fig. 1). During the
95% confidence intervals in this report were not blinded period of the trial, 51% of the partici-
adjusted for multiplicity. pants in each group had 4 to less than 6 months
The analysis of vaccine efficacy during the of follow-up after the second dose; 8% of the
blinded period of the trial included all partici- participants in the BNT162b2 group and 6% of
pants 12 years of age or older without known those in the placebo group had 6 months of
HIV infection who received at least one BNT162b2 follow-up or more after the second dose. During
or placebo dose. Vaccine efficacy was calculated the combined blinded and open-label periods,
as 100 × (1 – IRR), where IRR (incidence rate ra- 55% of the participants in the BNT162b2 group
tio) is the ratio of the rate (number per 1000 had 6 months of follow-up or more after the
person-years of follow-up) of confirmed cases of second dose. A total of 49% of the participants
Covid-19 in the BNT162b2 group to the corre- were female, 82% were White, 10% were Black,
sponding rate in the placebo group. Descriptive and 26% were Hispanic or Latinx; the median age
analyses of vaccine efficacy were performed and was 51 years. A total of 34% of the participants
associated 95% confidence intervals were calcu- had a body-mass index (the weight in kilograms
lated with the use of the Clopper–Pearson meth- divided by the square of the height in meters) of

n engl j med  nejm.org 3

The New England Journal of Medicine
Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

45,441 Participants underwent screening

1276 Were excluded

1173 Did not pass screening
103 Withdrew

44,165 Underwent randomization

22,085 Were assigned to receive BNT162b2 22,080 Were assigned to receive placebo

55 Did not receive BNT162b2 50 Did not receive placebo

26 Withdrew 26 Withdrew
15 Had protocol deviation 12 Had protocol deviation
7 No longer met eligibility criteria 3 No longer met eligibility criteria
3 Had adverse event 2 Had adverse event
4 Had other reason 7 Had other or unknown reason

22,030 Received the first dose 22,030 Received the first dose

380 Discontinued trial after the first

271 Discontinued trial after the first
dose and before the second dose
dose and before the second dose
108 Withdrew
108 Withdrew
90 Were lost to follow-up
89 Were lost to follow-up
119 No longer met eligibility criteria
25 No longer met eligibility criteria
25 Had adverse event
25 Had adverse event
6 Became pregnant
6 Became pregnant
7 Were withdrawn by physician
3 Were withdrawn by physician
2 Died
2 Died
2 Had medication error without
2 Had medication error without
associated adverse event
associated adverse event
1 Had protocol deviation
11 Had other or unknown reason
20 Had other or unknown reason

21,759 Received the second dose 21,650 Received the second dose

167 Discontinued trial after the second

dose 273 Discontinued trial after the second
81 Were lost to follow-up dose
54 Withdrew 125 Withdrew
14 Died 96 Were lost to follow-up
11 Had protocol deviation 24 Had protocol deviation
3 Were withdrawn by physician 13 Died
1 Had medication error without 3 Were withdrawn by physician
associated adverse event 4 No longer met eligibility criteria
1 Was withdrawn by parent 1 Had adverse event
or guardian 1 Became pregnant
1 No longer met eligibility criteria 6 Had other reason
1 Had other reason

20,334 Entered open-label follow-up 20,794 Entered open-label follow-up

4 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Safety and Efficacy of BNT162b2 through 6 Months

Figure 1 (facing page). Screening, Randomization, and evidence, and 49 who lacked the data needed to
Follow-up. determine previous infection status.
The diagram represents all enrolled participants 16 More participants in the BNT162b2 group
years of age or older through the data cutoff date than in the placebo group reported local reac-
(March 13, 2021). The diagram includes two deaths tions, the most common of which was mild-to-
that occurred after the second dose in human
moderate pain at the injection site (Fig. S1A).
immunodeficiency virus (HIV)–infected participants
(one in the BNT162b2 group and one in the placebo Local reactions were reported with similar fre-
group; these deaths were not reported in the Results quency among the participants with or without
section of this article because the analysis of HIV- evidence of previous SARS-CoV-2 infection, and
infected participants is being conducted separately). the reactions were of similar severity. No local
Information on the screening, randomization, and
reactions of grade 4 (according to the guidelines
follow-up of the participants 12 to 15 years of age has
been reported previously.11 of the Center for Biologics Evaluation and Re-
search12) were reported.
More participants in the BNT162b2 group
than in the placebo group reported systemic
30.0 or more, 21% had at least one underlying events, the most common of which was fatigue
medical condition, and 3% had baseline evidence (Fig. S1B). Systemic events were mostly mild to
of a previous or current SARS-CoV-2 infection moderate in severity, but there were occasional
(Table 1 and Table S2). severe events. Systemic reactogenicity was similar
Between October 15, 2020, and January 12, among those with or without evidence of previous
2021, a total of 2306 participants 12 to 15 years SARS-CoV-2 infection, although BNT162b2 re-
of age underwent screening, and 2264 underwent cipients with evidence of previous infection re-
randomization at 29 U.S. sites. Of these partici- ported systemic events more often after receipt
pants, 2260 received at least one dose of BNT162b2 of the first dose, and those without evidence
(1131 participants) or placebo (1129), and 99% reported systemic events more often after receipt
(1124 in the BNT162b2 group and 1117 in the of the second dose. For example, 12% of recipi-
placebo group) received the second dose.11 Among ents with evidence of previous SARS-CoV-2 in-
participants who received at least one dose of fection and 3% of those without evidence report-
BNT162b2 or placebo, 58% had at least 2 months ed fever after receipt of the first dose; 8% of
of follow-up after the second dose, 49% were those with evidence of previous infection and
female, 86% were White, 5% were Black, and 12% 15% of those without evidence reported fever
were Hispanic or Latinx. Full details of the de- after the second dose. The highest temperature
mographic characteristics of the participants reported was a transient fever of higher than
have been reported previously.11 40.0°C on day 2 after the second dose in a
BNT162b2 recipient without evidence of previ-
Safety ous infection.
Reactogenicity
The subgroup that was evaluated for reactogenic- Adverse Events
ity in the current report, in which reactions were Analyses of adverse events during the blinded
reported in an electronic diary, included 9839 period included 43,847 participants 16 years of
participants 16 years of age or older. In this sub- age or older (Table S3). Reactogenicity events
group, 8183 participants had been included in among the participants who were not in the re-
the previous analysis, and 1656 were enrolled actogenicity subgroup were reported as adverse
after the data cutoff for that analysis.9 The reac- events, which resulted in imbalances between
togenicity profile of BNT162b2 in this expanded the BNT162b2 group and the placebo group with
subgroup did not differ substantially from that respect to adverse events (30% vs. 14%), related
described previously.9 This subgroup included adverse events (24% vs. 6%), and severe adverse
364 participants who had evidence of previous events (1.2% vs. 0.7%). New adverse events at-
SARS-CoV-2 infection, 9426 who did not have tributable to BNT162b2 that were not previously

n engl j med  nejm.org 5

The New England Journal of Medicine
Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 1. Demographic Characteristics of the Participants at Baseline.*

BNT162b2 Placebo Total

Characteristic (N = 22,026) (N = 22,021) (N = 44,047)
Sex — no. (%)
Male 11,322 (51.4) 11,098 (50.4) 22,420 (50.9)
Female 10,704 (48.6) 10,923 (49.6) 21,627 (49.1)
Race or ethnic group — no. (%)†
White 18,056 (82.0) 18,064 (82.0) 36,120 (82.0)
Black or African American 2,098 (9.5) 2,118 (9.6) 4,216 (9.6)
Asian 952 (4.3) 942 (4.3) 1,894 (4.3)
American Indian or Alaska Native 221 (1.0) 217 (1.0) 438 (1.0)
Native Hawaiian or other Pacific Islander 58 (0.3) 32 (0.1) 90 (0.2)
Multiracial 550 (2.5) 533 (2.4) 1,083 (2.5)
Not reported 91 (0.4) 115 (0.5) 206 (0.5)
Ethnicity†
Hispanic or Latinx 5,704 (25.9) 5,695 (25.9) 11,399 (25.9)
Not reported 111 (0.5) 114 (0.5) 225 (0.5)
Country — no. (%)
Argentina 2,883 (13.1) 2,881 (13.1) 5,764 (13.1)
Brazil 1,452 (6.6) 1,448 (6.6) 2,900 (6.6)
Germany 249 (1.1) 250 (1.1) 499 (1.1)
South Africa 401 (1.8) 399 (1.8) 800 (1.8)
Turkey 249 (1.1) 249 (1.1) 498 (1.1)
United States 16,792 (76.2) 16,794 (76.3) 33,586 (76.3)
Age group at vaccination — no. (%)
16–55 yr 13,069 (59.3) 13,095 (59.5) 26,164 (59.4)
>55 yr 8,957 (40.7) 8,926 (40.5) 17,883 (40.6)
Age at vaccination — yr
Median 51.0 51.0 51.0
Range 16–89 16–91 16–91
SARS-CoV-2 status — no. (%)‡
Positive 689 (3.1) 716 (3.3) 1,405 (3.2)
Negative 21,185 (96.2) 21,180 (96.2) 42,365 (96.2)
Missing data 152 (0.7) 125 (0.6) 277 (0.6)
Body-mass index — no. (%)§
≥30.0: obese 7,543 (34.2) 7,629 (34.6) 15,172 (34.4)
Missing data 7 (<1) 6 (<1) 13 (<1)

* Data are summarized for participants 16 years of age or older in the safety population. The demographic characteristics
of participants 12 to 15 years of age were reported previously.11 Percentages may not total 100 because of rounding.
SARS-CoV-2 denotes severe acute respiratory syndrome coronavirus 2.
† Race and ethnicity were reported by the participants. The categories shown are those that were used to collect the data.
‡ Positive status was defined as a positive N-binding antibody result or a positive nucleic acid amplification test (NAAT)
result at visit 1 or medical history of coronavirus disease 2019 (Covid-19). Negative status was defined as a negative
N-binding antibody result or a negative NAAT result at visit 1 and no medical history of Covid-19.
§ The body-mass index is the weight in kilograms divided by the square of the height in meters.

6 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Safety and Efficacy of BNT162b2 through 6 Months

Table 2. Vaccine Efficacy against Covid-19 from 7 Days after Receipt of the Second Dose during the Blinded, Placebo-Controlled Follow-up
Period.*

Vaccine Efficacy
Efficacy End Point BNT162b2 Placebo (95% CI)‡

No. of Surveillance No. at No. of Surveillance No. at

Cases Time† Risk Cases Time† Risk

1000 person-yr 1000 person-yr percent

(N = 20,998) (N = 21,096)
First occurrence of Covid-19 77 6.247 20,712 850 6.003 20,713 91.3
from 7 days after receipt (89.0–93.2)
of the second dose among
participants without evidence
of previous infection
(N = 22,166) (N = 22,320)
First occurrence of Covid-19 81 6.509 21,642 873 6.274 21,689 91.1
from 7 days after receipt (88.8–93.0)
of the second dose among
participants with or without
evidence of previous infection

* This analysis included participants who had no serologic or virologic evidence (within 7 days after receipt of the second dose) of previous
SARS-CoV-2 infection (i.e., negative N-binding antibody [serum] test at visit 1 and SARS-CoV-2 not detected by NAAT [nasal swab] at visits
1 and 2) and had a negative NAAT at any unscheduled visit up to 7 days after receipt of the second dose.
† The surveillance time is the total time (in 1000 person-years) at risk for the given end point across all participants within each group. The
time period for the accrual of Covid-19 cases was from 7 days after the second dose to the end of the surveillance period.
‡ Vaccine efficacy was calculated as 100 × (1 – IRR), where IRR (incidence rate ratio) is the ratio of the rate (number per 1000 person-years of
follow-up) of confirmed cases of Covid-19 in the BNT162b2 group to the corresponding rate in the placebo group. The 95% confidence in-
terval for vaccine efficacy was derived with the use of the Clopper–Pearson method, with adjustment for surveillance time.

identified in earlier reports included decreased and 2 in the original placebo group who received
appetite, lethargy, asthenia, malaise, night sweats, BNT162b2 after unblinding died. None of these
and hyperhidrosis. Few participants had serious deaths were considered to be related to BNT162b2
adverse events or adverse events that led to trial by the investigators. Causes of death were bal-
withdrawal. No new serious adverse events were anced between BNT162b2 and placebo groups
considered by the investigators to be related to (Table S4).
BNT162b2 after the data cutoff date of the previ- Safety monitoring will continue according to
ous report.9 the protocol for 2 years after the second dose for
During the combined blinded and open-label participants who originally received BNT162b2 and
periods, cumulative safety data during follow-up for 18 months after the second BNT162b2 dose for
were available through 6 months after the sec- placebo recipients who received BNT162b2 after
ond dose for 12,006 participants who were origi- unblinding.
nally randomly assigned to the BNT162b2 group.
No new safety signals relative to the previous Efficacy
report were observed during the longer follow- Among 42,094 participants 12 years of age or
up period in the current report, which included older who could be evaluated and had no evidence
open-label observation of the original BNT162b2 of previous SARS-CoV-2 infection, Covid-19 with
recipients and placebo recipients who received an onset of 7 days or more after the second dose
BNT162b2 after unblinding.9 was observed in 77 vaccine recipients and in 850
During the blinded, placebo-controlled peri- placebo recipients up to the data cutoff date
od, 15 participants in the BNT162b2 group and (March 13, 2021), corresponding to a vaccine ef-
14 in the placebo group died; during the open- ficacy of 91.3% (95% confidence interval [CI],
label period, 3 participants in the BNT162b2 group 89.0 to 93.2) (Table 2). Among 44,486 participants

n engl j med  nejm.org 7

The New England Journal of Medicine
Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

with or without evidence of previous infection subgroups defined according to age, sex, race,
who could be evaluated, cases of Covid-19 were ethnic group, presence or absence of coexisting
observed in 81 vaccine recipients and in 873 medical conditions, and country was generally
placebo recipients, corresponding to a vaccine consistent with that observed in the overall
efficacy of 91.1% (95% CI, 88.8 to 93.0). population (Table 3 and Table S7).
Among the participants with evidence of pre- Given the concern about the SARS-CoV-2
vious SARS-CoV-2 infection based on a positive B.1.351 (or beta) variant, which appears to be
baseline N-binding antibody test, Covid-19 was neutralized less efficiently by BNT162b2-immune
observed in 2 vaccine recipients after the first sera than many other lineages,14 whole-viral-
dose and in 7 placebo recipients. Among the genome sequencing was performed on midturbi-
participants with evidence of previous SARS- nate samples from Covid-19 cases observed in
CoV-2 infection based on a positive nucleic acid South Africa, where this lineage was prevalent.
amplification test at baseline, cases of Covid-19 Nine cases of Covid-19 were observed in South
were observed in 10 vaccine recipients and in 9 African participants without evidence of previ-
placebo recipients (Table S5). Covid-19 was less ous SARS-CoV-2 infection, all of whom were
common among the placebo recipients with placebo recipients; this finding corresponds with
positive N-binding antibodies at trial entry (7 of a vaccine efficacy of 100% (95% CI, 53.5 to 100)
542 participants, for an incidence of 1.3%) than (Table 3). Midturbinate specimens from 8 of 9
among those without evidence of infection at cases contained sufficient viral RNA for whole-
trial entry (1015 of 21,521, for an incidence of genome sequencing. All viral genomes were the
4.7%); these findings indicate that previous infec- beta variant (Global Initiative on Sharing All
tion conferred approximately 72.6% protection. Influenza Data accession codes are provided in
Among the participants with or without the Supplementary Appendix).
evidence of previous infection, cases of Covid-19
were observed in 46 vaccine recipients and in
Discussion
110 placebo recipients from receipt of the first
dose up to receipt of the second dose, corre- In this update to the preliminary safety and effi-
sponding to a vaccine efficacy of 58.4% (95% CI, cacy report of two 30-μg doses, at 21 days apart,
40.8 to 71.2) (Fig. 2). During the interval from of BNT162b2, 91.1% vaccine efficacy against
the approximate start of observed protection at Covid-19 was observed from 7 days to 6 months
11 days after receipt of the first dose up to re- after the second dose in participants 12 years of
ceipt of the second dose, vaccine efficacy in- age or older. Vaccine efficacy against severe dis-
creased to 91.7% (95% CI, 79.6 to 97.4). From its ease with an onset after receipt of the first dose
peak after the second dose, observed vaccine effi- was approximately 97%. This finding, combined
cacy declined. From 7 days to less than 2 months with the totality of available evidence, including
after the second dose, vaccine efficacy was 96.2% real-world effectiveness data,15-18 alleviates theo-
(95% CI, 93.3 to 98.1); from 2 months to less than retical concerns over potential enhancement of
4 months after the second dose, vaccine effi- vaccine-mediated disease.19
cacy was 90.1% (95% CI, 86.6 to 92.9); and The benefit of BNT162b2 immunization start-
from 4 months after the second dose to the data ed approximately 11 days after receipt of the first
cutoff date, vaccine efficacy was 83.7% (95% CI, dose, with 91.7% vaccine efficacy from 11 days
74.7 to 89.9). after receipt of the first dose up to receipt of the
Severe Covid-19, as defined by the Food and second dose. The trial cannot provide informa-
Drug Administration,13 with an onset after receipt tion on persistence of protection after a single
of the first dose occurred in 31 participants, of dose, because 99% of the participants received
whom 30 were placebo recipients; this finding the second dose as scheduled during the blinded
corresponds with a vaccine efficacy of 96.7% trial period. A recent trial showed that although
(95% CI, 80.3 to 99.9) against severe Covid-19 nonneutralizing viral antigen–binding antibody
(Fig. 2 and Table S6). Although the trial was not levels rise between the first and second BNT162b2
powered to definitively assess efficacy according dose, serum neutralizing titers are low or unde-
to subgroup, supplemental analyses indicated tectable during this interval.20 Early protection
that vaccine efficacy after the second dose in against Covid-19 without strong serum neutral-

8 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Safety and Efficacy of BNT162b2 through 6 Months

100.0
0.5
Placebo
0.4
8.0
0.3
7.0 Placebo
0.2
BNT162b2
Cumulative Incidence (%)

6.0 0.1

0.0
5.0 0 3 6 9 12 15 18 21

4.0

3.0

2.0

1.0 BNT162b2

0.0
0 14 28 42 56 70 84 98 112 126 140 154 168 182 196 210 224 238
Days since Receipt of First Dose

BNT162b2 Placebo
Efficacy End Point (N=23,040) (N=23,037) Vaccine Efficacy
No. of Surveillance No. at No. of Surveillance No. at
cases time risk cases time risk
1000 person-yr 1000 person-yr % (95% CI)
Overall: first occurrence of Covid-19 after receipt of first dose 131 8.412 22,505 1034 8.124 22,434 87.8 (85.3 to 89.9)
After receipt of first dose up to receipt of second dose 46 1.339 22,505 110 1.331 22,434 58.4 (40.8 to 71.2)
<11 Days after receipt of first dose 41 0.677 22,505 50 0.675 22,434 18.2 (−26.1 to 47.3)
≥11 Days after receipt of first dose up to receipt of second dose 5 0.662 22,399 60 0.656 22,369 91.7 (79.6 to 97.4)
After receipt of second dose to <7 days after 3 0.424 22,163 35 0.422 22,057 91.5 (72.9 to 98.3)
≥7 Days after receipt of second dose 82 6.649 22,132 889 6.371 22,001 91.2 (88.9 to 93.0)
≥7 Days after receipt of second dose to <2 mo after 12 2.923 22,132 312 2.884 22,001 96.2 (93.3 to 98.1)
≥2 Mo after receipt of second dose to <4 mo after 46 2.696 20,814 449 2.593 20,344 90.1 (86.6 to 92.9)
≥4 Mo after receipt of second dose 24 1.030 12,670 128 0.895 11,802 83.7 (74.7 to 89.9)

Figure 2. Efficacy of BNT162b2 against Covid-19 after Receipt of the First Dose (Blinded Follow-up Period).
The top of the figure shows the cumulative incidence curves for the first occurrence of coronavirus disease 2019 (Covid-19) after receipt
of the first dose (efficacy analysis population of participants ≥12 years of age who could be evaluated). Each symbol represents Covid-19
cases starting on a given day, and filled symbols represent severe Covid-19 cases. Because of overlapping dates, some symbols repre-
sent more than one case. The inset shows the same data on an enlarged y axis through 21 days. The bottom of the figure shows the
time intervals for the first occurrence of Covid-19 in the efficacy analysis population, as well as the surveillance time, which is given as
the total time (in 1000 person-years) at risk for the given end point across all participants within each group. The time period for the ac-
crual of Covid-19 cases was from after receipt of the first dose to the end of the surveillance period for the overall row and from the start
to the end of the range stated for each time interval. Vaccine efficacy was calculated as 100 × (1 – IRR), where IRR (incidence rate ratio) is
the ratio of the rate (number per 1000 person-years of follow-up) of confirmed cases of Covid-19 in the BNT162b2 group to the corre-
sponding rate in the placebo group. The 95% confidence interval for vaccine efficacy was derived with the use of the Clopper–Pearson
method, with adjustment for surveillance time.

ization indicates that neutralizing titers alone do and antibody-dependent cytotoxicity) may con-
not appear to explain early BNT162b2-mediated tribute to protection.21-26
protection from Covid-19. Other immune mech- Efficacy peaked at 96.2% during the interval
anisms (e.g., innate immune responses, CD4+ or from 7 days to less than 2 months after the sec-
CD8+ T-cell responses, B-cell memory responses, ond dose and declined gradually to 83.7% from

n engl j med  nejm.org 9

The New England Journal of Medicine
Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

Table 3. Vaccine Efficacy against Covid-19 up to 7 Days after Receipt of the Second Dose among Participants without Evidence of Infection.*

First Occurrence of
Covid-19 after Receipt BNT162b2 Placebo Vaccine Efficacy
of the First Dose (N = 20,998) (N = 21,096) (95% CI)‡

No. of Surveillance No. at No. of Surveillance No. at

Cases Time† Risk Cases Time† Risk

1000 person-yr 1000 person-yr percent

Overall population 77 6.247 20,712 850 6.003 20,713 91.3 (89.0 to 93.2)
Age group — yr
16 or 17 0 0.061 342 10 0.057 331 100 (58.2 to 100)
16 to 55 52 3.593 11,517 568 3.439 11,533 91.2 (88.3 to 93.5)
≥55 25 2.499 8,194 266 2.417 8,208 90.9 (86.3 to 94.2)
≥65 7 1.233 4,192 124 1.202 4,226 94.5 (88.3 to 97.8)
≥75 1 0.239 842 26 0.237 847 96.2 (76.9 to 99.9)
Sex
Male 42 3.246 10,637 399 3.047 10,433 90.1 (86.4 to 93.0)
Female 35 3.001 10,075 451 2.956 10,280 92.4 (89.2 to 94.7)
Race or ethnic group§
White 67 5.208 17,186 747 5.026 17,256 91.3 (88.9 to 93.4)
Black or African 4 0.545 1,737 48 0.527 1,737 91.9 (78.0 to 97.9)
American
Asian 3 0.260 946 23 0.248 934 87.6 (58.9 to 97.6)
American Indian or 0 0.041 186 3 0.037 176 100 (–119.0 to 100)
Alaska Native
Native Hawaiian 0 0.015 54 1 0.008 30 100 (–1961.2 to 100)
or other Pacific
Islander
Multiracial 3 0.151 518 22 0.128 476 88.5 (61.6 to 97.8)
Not reported 0 0.026 85 6 0.030 104 100 (2.8 to 100)
Ethnicity§
Hispanic or Latinx 29 1.786 5,161 241 1.711 5,120 88.5 (83.0 to 92.4)
Non-Hispanic and 47 4.429 15,449 609 4.259 15,484 92.6 (90.0 to 94.6)
non-Latinx
Not reported 1 0.032 102 0 0.033 109 NA
Country
Argentina 15 1.012 2,600 108 0.986 2,586 86.5 (76.7 to 92.7)
Brazil 12 0.406 1,311 80 0.374 1,293 86.2 (74.5 to 93.1)
Germany 0 0.047 236 1 0.048 242 100 (–3874.2 to 100)
South Africa 0 0.080 291 9 0.074 276 100 (53.5 to 100)
Turkey 0 0.027 228 5 0.025 222 100 (–0.1 to 100)
United States 50 4.674 16,046 647 4.497 16,046 92.6 (90.1 to 94.5)

* This analysis of vaccine efficacy during the blinded, placebo-controlled follow-up period included all participants who had undergone ran-
domization and were 12 years of age or older without baseline evidence of previous infection who had undergone randomization. NA de-
notes not applicable.
† Surveillance time is the total time (in 1000 person-years) at risk for the given end point across all participants within each group. The time
period for the accrual of Covid-19 cases was from 7 days after the second dose to the end of the surveillance period.
‡ Vaccine efficacy was calculated as 100 × (1 – IRR). The 95% confidence interval for vaccine efficacy was derived with the use of the Clopper–
Pearson method, with adjustment for surveillance time.
§ Race and ethnicity were reported by the participants. The categories shown are those that were used to collect the data.

10 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Safety and Efficacy of BNT162b2 through 6 Months

4 months after the second dose to the data cut- dose. The safety profile observed at a median of
off date — an average decline of approximately 2 months after immunization was confirmed
6% every 2 months. Ongoing follow-up is needed through 6 months after immunization in the cur-
to understand persistence of the vaccine effect rent analysis. No cases of myocarditis were noted.
over time, the need for booster dosing, and tim- Before immunization, 3% of the participants
ing of such a dose. Most participants who ini- 16 years of age or older had evidence of SARS-
tially received placebo have now been immunized CoV-2 infection. Although this group had a slight-
with BNT162b2, ending the placebo-controlled ly higher incidence of systemic reactogenicity
period of the trial. Nevertheless, ongoing obser- events after receipt of the first dose than those
vation of participants through 2 years in this without evidence of previous infection, the group
trial, together with real-world effectiveness data,15- had a slightly lower incidence of reactogenicity
18
will determine whether a booster is likely to be events after the second dose than those without
beneficial after a longer interval. Booster trials to previous infection. Thus, there was minimal ob-
evaluate safety and immunogenicity of BNT162b2 served difference in the overall reactogenicity
are under way to prepare for this possibility. profile on the basis of infection status at baseline.
From 7 days after the second dose, 86 to 100% Nine cases of Covid-19 were observed among
efficacy was observed across diverse demographic participants with previous serologically defined
profiles, including age, sex, race or ethnic group, natural infection: two cases were observed among
and factors that increase the risk of Covid-19, the vaccine recipients and seven among the pla-
such as high body-mass index and other coexist- cebo recipients. These data support the current
ing medical conditions. BNT162b2 was also practice of immunizing without screening for
highly efficacious in various geographic regions evidence of previous infection.
including North America, Europe, South Africa, This report has several limitations. Duration
and Latin America. Although vaccine efficacy of protection and safety data that could be col-
was slightly lower in Latin American countries, lected in a blinded, placebo-controlled manner
BNT162b2 had a high efficacy of approximately were limited by the ethical and practical need to
86% in Argentina and Brazil. Circulation of immunize eligible initial placebo recipients un-
SARS-CoV-2 variants — some of which are as- der emergency use authorization and according
sociated with more rapid transmission and po- to the recommendations of public health authori-
tentially greater pathogenicity27 — has raised ties. The data presented here do not address
concerns that such variants could evade vaccine- whether vaccination prevents asymptomatic in-
mediated protection. Our studies of in vitro fection; however, evaluation of that question is
neutralization of a variety of SARS-CoV-2 vari- ongoing in this trial, and real-world data sug-
ants have, to date, showed that all tested gest that BNT162b2 prevents asymptomatic in-
BNT162b2-immune sera neutralize all tested fection.33,34 Preliminary analyses of breakthrough
variants.14,28-32 The beta variant, which has shown cases have not yet identified a correlate of pro-
the greatest reduction in neutralization and was tection, since vaccine protection rates remain high.
the dominant strain in South Africa during the This report does not address vaccine efficacy and
reported observation period, is still neutralized safety in pregnant women and in children younger
at serum titers higher than those observed at the than 12 years of age. Studies evaluating BNT162b2
onset of protection against Covid-19 after the in these populations are ongoing.
first vaccine dose.9,14,20 We found that BNT162b2 The data in this report show that BNT162b2
had an observed efficacy of 100% (95% CI, 53.5 prevents Covid-19 effectively for up to 6 months
to 100) against Covid-19 in South Africa (9 cases after the second dose across diverse populations,
occurred in the placebo recipients and 0 cases in despite the emergence of SARS-CoV-2 variants,
the BNT162b2 recipients), and 8 of 9 cases for including the beta variant, and the vaccine con-
which sequence information could be obtained tinues to show a favorable safety profile.
involved the beta variant of SARS-CoV-2.
Safety data are now available for approxi- Supported by BioNTech and Pfizer.
mately 44,000 participants 16 years of age or Disclosure forms provided by the authors are available with
the full text of this article at NEJM.org.
older; 12,006 participants have at least 6 months A data sharing statement provided by the authors is available
of safety follow-up data after a second BNT162b2 with the full text of this article at NEJM.org.

n engl j med  nejm.org 11

The New England Journal of Medicine
Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 The n e w e ng l a n d j o u r na l of m e dic i n e

We thank all the participants who volunteered for this trial; roko, Jason McKinley, Tracey Mellelieu, Neda Aghajani Memar,
the investigators in the C4591001 Clinical Trial Group for their Farheen Muzaffar, Brendan O’Neill, Jason Painter, Elizabeth
contributions; the members of our data monitoring committee Paulukonis, Allison Pfeffer, Katie Puig, Kimberly Rarrick, Balaji
(Jonathan Zenilman [chair], Robert Belshe, Kathryn Edwards, Prabu Raja, Christine Rainey, Kellie Lynn Richardson, Elizabeth
Stephen Self, and Lawrence Stanberry) for their review of the Rogers, Melinda Rottas, Charulata Sabharwal, Uzma Sarwar,
data; Tricia Newell, Sheena Hunt, and Philippa Jack of ICON Vilas Satishchandran, Harpreet Seehra, Judy Sewards, Huiqing
(North Wales, PA) for editorial support, which was funded Si, Helen Smith, David Swerdlow, James Trammel, Elisa Har-
by Pfizer; the following Pfizer staff: Greg Adams, Negar Ali- kins Tull, Sarah Tweedy, Erica Weaver, John Wegner, Jenah West,
abadi, Mohanish Anand, Fred Angulo, Ayman Ayoub, Melissa Christopher Webber, David C. Whritenour, Fae Wooding, Em-
Bishop-Murphy, Mark Boaz, Christopher Bowen, Donna Boyce, ily Worobetz, Nita Zalavadia, and Liping Zhang, as well as the
Sarah Burden, Andrea Cawein, Patrick Caubel, Darren Cowen, Vaccines Clinical Assay Team, the Vaccines Assay Development
Kimberly Ann Cristall, Michael Cruz, Daniel Curcio, Gabriela Team, and all the Pfizer colleagues not named here who contrib-
Dávila, Carmel Devlin, Gokhan Duman, Niesha Foster, Maja uted to the success of this trial; the following BioNTech staff:
Gacic, Juleen Gayed, Ahmed Hassan, Luis Jodar, Stephen Kay, Corinna Rosenbaum, Christian Miculka, Andreas Kuhn, Ferdia
William Lam, Esther Ladipo, Joaquina Maria Lazaro, Marie- Bates, Paul Strecker, Ruben Rizzi, Martin Bexon, Eleni Lagkadi-
Pierre Hellio Le Graverand-Gastineau, Kwok Lee, Zhenghui Li, nou, and Alexandra Kemmer-Brück; and Dietmar Katinger and
Jacqueline Lowenberg, Hua Ma, Rod MacKenzie, Robert Ma- Andreas Wagner at Polymun.

Appendix
The authors’ full names and academic degrees are as follows: Stephen J. Thomas, M.D., Edson D. Moreira, Jr., M.D., Nicholas Kitchin,
M.D., Judith Absalon, M.D., Alejandra Gurtman, M.D., Stephen Lockhart, D.M., John L. Perez, M.D., Gonzalo Pérez Marc, M.D., Fer-
nando P. Polack, M.D., Cristiano Zerbini, M.D., Ruth Bailey, B.Sc., Kena A. Swanson, Ph.D., Xia Xu, Ph.D., Satrajit Roychoudhury,
Ph.D., Kenneth Koury, Ph.D., Salim Bouguermouh, M.D., Ph.D., Warren V. Kalina, Ph.D., David Cooper, Ph.D., Robert W. Frenck, Jr.,
M.D., Laura L. Hammitt, M.D., Özlem Türeci, M.D., Haylene Nell, M.D., Axel Schaefer, M.D., Serhat Ünal, M.D., Qi Yang, Ph.D., Paul
Liberator, Ph.D., Dina B. Tresnan, D.V.M., Ph.D., Susan Mather, M.D., Philip R. Dormitzer, M.D., Ph.D., Uğur Şahin, M.D., William C.
Gruber, M.D., and Kathrin U. Jansen, Ph.D.
The authors’ affiliations are as follows: the State University of New York, Upstate Medical University, Syracuse (S.J.T.), and Vaccine
Research and Development, Pfizer, Pearl River (J.A., A.G., K.A.S., K.K., S.B., W.V.K., D.C., Q.Y., P.L., P.R.D., W.C.G., K.U.J.) — both
in New York; Associação Obras Sociais Irmã Dulce and Oswaldo Cruz Foundation, Bahia (E.D.M.), and iTrials-Hospital Militar Central
(G.P.M.) and Fundacion INFANT, Buenos Aires (F.P.P.) — all in Brazil; Centro Paulista de Investigação Clinica, São Paulo (C.Z.); Vac-
cine Research and Development, Pfizer, Hurley, United Kingdom (N.K., S.L., R.B.); Vaccine Research and Development (J.L.P., X.X.)
and Worldwide Safety, Safety Surveillance, and Risk Management (D.B.T., S.M.), Pfizer, Collegeville, PA; Global Product Development,
Pfizer, Peapack, NJ (S.R.); Cincinnati Children’s Hospital, Cincinnati (R.W.F.); Johns Hopkins Bloomberg School of Public Health,
Baltimore (L.L.H.); BioNTech, Mainz (Ö.T., U.Ş.) and Medizentrum Essen Borbeck, Essen (A.S.) — both in Germany; Tiervlei Trial
Centre, Karl Bremer Hospital, Cape Town, South Africa (H.N.); Hacettepe University, Ankara, Turkey (S.Ü.); and Worldwide Safety,
Safety Surveillance, and Risk Management, Pfizer, Groton, CT (D.B.T., S.M.).

References
1. COVID-19 dashboard. 2021 (https:// 7. Karikó K, Muramatsu H, Welsh FA, et clinical trials. Rockville, MD:​Center for
coronavirus​.­jhu​.­edu/​­map​.­html). al. Incorporation of pseudouridine into Biologics Evaluation and Research. Sep-
2. Food and Drug Administration. mRNA yields superior nonimmunogenic tember 2007 (https://www​.­fda​.­gov/​­media/​
COVID-19 vaccines. Silver Spring, MD:​ vector with increased translational capac- ­73679/​­download).
Food and Drug Administration. 2021 ity and biological stability. Mol Ther 2008;​ 13. COVID-19: developing drugs and bi-
(https://www​.­fda​.­gov/​­emergency​ 16:​1833-40. ological products for treatment or pre-
-­preparedness​-­and​-­response/​­coronavirus​ 8. Wrapp D, Wang N, Corbett KS, et al. vention: guidance for industry. Silver
-­disease​-­2019​-­covid​-­19/​­covid​-­19​ Cryo-EM structure of the 2019-nCoV Spring, MD:​Food and Drug Administra-
-­vaccines). spike in the prefusion conformation. Sci- tion. May 2020 (https://www​.­fda​.­gov/​
3. Medicines and Healthcare Products ence 2020;​367:​1260-3. ­regulatory​-­information/​­search​-­fda​
Regulatory Agency. Decision: conditions 9. Polack FP, Thomas SJ, Kitchin N, et al. -­g uidance​-­documents/​­covid​-­19​
of authorisation for COVID-19 vaccine As- Safety and efficacy of the BNT162b2 -­developing​-­drugs​-­and​-­biological​
traZeneca (regulation 174). 2021 (https:// mRNA Covid-19 vaccine. N Engl J Med -­products​-­t reatment​-­or​-­prevention).
www​.­gov​.­uk/​­government/​­publications/​ 2020;​383:​2603-15. 14. Liu Y, Liu J, Xia H, et al. Neutralizing
­regulatory​-­approval​-­of​-­covid​-­19​-­vaccine​ 10. Pfizer, BioNTech. Pfizer-BioNTech activity of BNT162b2-elicited serum. N Engl
-­astrazeneca/​­conditions​-­of​-­authorisation​ COVID-19 vaccine. FDA briefing docu- J Med 2021;​384:​1466-8.
-­for​-­covid​-­19​-­vaccine​-­astrazeneca). ment. Presented at the Vaccines and Re- 15. Haas EJ, Angulo FJ, McLaughlin JM,
4. Baraniuk C. What do we know about lated Biological Products Advisory Com- et al. Impact and effectiveness of mRNA
China’s covid-19 vaccines? BMJ 2021;​373:​ mittee Meeting, virtual, December 10, BNT162b2 vaccine against SARS-CoV-2
n912. 2020 (https://www​.­fda​.­gov/​­media/​­144245/​ infections and COVID-19 cases, hospitali-
5. Baraniuk C. Covid-19: what do we ­download). sations, and deaths following a nation-
know about Sputnik V and other Russian 11. Frenck RW Jr, Klein NP, Kitchin N, et wide vaccination campaign in Israel: an
vaccines? BMJ 2021;​372:​n743. al. Safety, immunogenicity, and efficacy observational study using national sur-
6. Pardi N, Tuyishime S, Muramatsu H, of the BNT162b2 Covid-19 vaccine in ado- veillance data. Lancet 2021;​397:​1819-29.
et al. Expression kinetics of nucleoside- lescents. N Engl J Med 2021;​385:​239-50. 16. Abu-Raddad LJ, Chemaitelly H, Butt AA.
modified mRNA delivered in lipid nanopar- 12. Guidance for industry: toxicity grad- Effectiveness of the BNT162b2 Covid-19
ticles to mice by various routes. J Control ing scale for healthy adult and adolescent vaccine against the B.1.1.7 and B.1.351
Release 2015;​217:​345-51. volunteers enrolled in preventive vaccine variants. N Engl J Med 2021;​385:​187-9.

12 n engl j med  nejm.org

The New England Journal of Medicine

Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.
 Safety and Efficacy of BNT162b2 through 6 Months

17. Lopez Bernal J, Andrews N, Gower C, memory CD8 T cells against a lethal viral vaccine-elicited sera. Nat Med 2021;​ 27:​
et al. Effectiveness of the Pfizer-BioNTech infection. Mol Ther 2021 May 14 (Epub 620-1.
and Oxford-AstraZeneca vaccines on cov- ahead of print). 29. Zou J, Xie X, Fontes-Garfias CR, et al.
id-19 related symptoms, hospital admis- 24. Tauzin A, Nayrac M, Benlarbi M, et al. The effect of SARS-CoV-2 D614G muta-
sions, and mortality in older adults in A single BNT162b2 mRNA dose elicits an- tion on BNT162b2 vaccine-elicited neu-
England: test negative case-control study. tibodies with Fc-mediated effector func- tralization. NPJ Vaccines 2021;​6:​44.
BMJ 2021;​373:​n1088. tions and boost pre-existing humoral and 30. Muik A, Wallisch A-K, Sänger B, et al.
18. Vasileiou E, Simpson CR, Shi T, et al. T cell responses. March 18, 2021 (https:// Neutralization of SARS-CoV-2 lineage
Interim findings from first-dose mass www​.­biorxiv​.­org/​­content/​­10​.­1101/​­2021​.­03​ B.1.1.7 pseudovirus by BNT162b2 vaccine-
COVID-19 vaccination roll-out and CO- .­18​.­435972v1). preprint. elicited human sera. Science 2021;​ 371:​
VID-19 hospital admissions in Scotland: 25. Gallagher KME, Leick MB, Larson 1152-3.
a national prospective cohort study. Lan- RC, et al. SARS-CoV-2 T-cell immunity to 31. Liu Y, Liu J, Xia H, et al. BNT162b2-
cet 2021;​397:​1646-57. variants of concern following vaccination. elicited neutralization against new SARS-
19. Haynes BF, Corey L, Fernandes P, et May 3, 2021 (https://www​.­biorxiv​.­org/​ CoV-2 spike variants. N Engl J Med 2021;​
al. Prospects for a safe COVID-19 vaccine. ­content/​­10​.­1101/​­2021​.­05​.­03​.­442455v1). 385:​472-4.
Sci Transl Med 2020;​12(568):​eabe0948. preprint. 32. Liu J, Liu Y, Xia H, et al. BNT162b2-
20. Walsh EE, Frenck RW Jr, Falsey AR, et 26. Painter MM, Mathew D, Goel RR, et elicited neutralization of B.1.617 and
al. Safety and immunogenicity of two al. Rapid induction of antigen-specific other SARS-CoV-2 variants. Nature 2021
RNA-based Covid-19 vaccine candidates. CD4+ T cells guides coordinated humoral June 10 (Epub ahead of print).
N Engl J Med 2020;​383:​2439-50. and cellular immune responses to SARS- 33. Dagan N, Barda N, Kepten E, et al.
21. Sahin U, Muik A, Vogler I, et al. CoV-2 mRNA vaccination. April 22, 2021 BNT162b2 mRNA Covid-19 vaccine in a na-
BNT162b2 vaccine induces neutralizing (https://www​.­biorxiv​.­org/​­content/​­10​.­1101/​ tionwide mass vaccination setting. N Engl J
antibodies and poly-specific T cells in hu- ­2021​.­04​.­21​.­440862v1). preprint. Med 2021;​384:​1412-23.
mans. Nature 2021;​595:​572-7. 27. Centers for Disease Control and Pre- 34. Tande AJ, Pollock BD, Shah ND, et al.
22. Goel RR, Apostolidis SA, Painter MM, vention. SARS-CoV-2 variant classifica- Impact of the COVID-19 vaccine on asymp­
et al. Distinct antibody and memory B cell tions and definitions. 2021 (https://www​ tomatic infection among patients under-
responses in SARS-CoV-2 naïve and recov- .­cdc​.­gov/​­coronavirus/​­2019​-­ncov/​­cases​ going pre-procedural COVID-19 molecu-
ered individuals following mRNA vacci- -­updates/​­variant​-­surveillance/​­variant​-­info​ lar screening. Clin Infect Dis 2021 March
nation. Sci Immunol2021;​6 (58):​eabi6950. .­html#print). 10 (Epub ahead of print).
23. Knudson CJ, Alves-Peixoto P, Mura- 28. Xie X, Liu Y, Liu J, et al. Neutralization Copyright © 2021 Massachusetts Medical Society.
matsu H, et al. Lipid-nanoparticle-encap- of SARS-CoV-2 spike 69/70 deletion,
sulated mRNA vaccines induce protective E484K and N501Y variants by BNT162b2

n engl j med  nejm.org 13

The New England Journal of Medicine
Downloaded from nejm.org on October 5, 2021. For personal use only. No other uses without permission.
Copyright © 2021 Massachusetts Medical Society. All rights reserved.