Safety and reactogenicity of the BNT162b2 COVID-19 vaccine: Development, post-marketing surveillance, and real-world data

ABSTRACT

The pandemic caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) led to urgent actions by innovators, vaccine developers, regulators, and other stakeholders to ensure public access to protective vaccines while maintaining regulatory agency standards. Although development timelines for vaccines against SARS-CoV-2 were much quicker than standard vaccine development timelines, regulatory requirements for efficacy and safety evaluations, including the volume and quality of data collected, were upheld. Rolling review processes supported by sponsors and regulatory authorities enabled rapid assessment of clinical data as well as emergency use authorization. Post-authorization and pharmacovigilance activities enabled the quantity and breadth of post-marketing safety information to quickly exceed that generated from clinical trials. This paper reviews safety and reactogenicity data for the BNT162 vaccine candidates, including BNT162b2 (Comirnaty, Pfizer/BioNTech COVID-19 vaccine) and bivalent variant-adapted BNT162b2 vaccines, from preclinical studies, clinical trials, post-marketing surveillance, and real-world studies, including an unprecedentedly large body of independent evidence.

KEYWORDS:

• SARS-CoV-2
• safety
• reactogenicity
• vaccine development
• post-marketing surveillance
• real-world studies

Introduction

Following the emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), an urgent need to develop vaccines to counter the rapidly spreading pandemic was quickly recognized. While maintaining rigorous clinical trial standards, multiple stakeholders collaborated to expedite vaccine development, authorization, and roll-out.¹

Vaccine clinical development timelines were compressed through leverage of existing research on previous coronavirus outbreaks, the utilization of novel vaccine technologies with shorter design-to-production times, increased funding and collaboration, at-risk investment in commercial production from manufacturers prior to approval, and the use of accelerated regulatory procedures and expedited review timelines.^1–3 In addition, high infection rates and participant willingness to take part in clinical trials led to more rapid enrollment and study completion, when compared with standard vaccine development timelines. Pre-existing manufacturing processes for messenger RNA (mRNA) vaccines enabled rapid production and scale-up.¹ Subsequently, development and commercialization timelines were much shorter compared with other vaccines, such as influenza, while every step of the pathway required for regulatory approval was fulfilled (Figure 1).²

Figure 1. Accelerated timelines for the development and approval of COVID-19 vaccines.^1,2,4 COVID-19, coronavirus disease 2019.

The clinical development programs for coronavirus disease 2019 (COVID-19) vaccines, such as BNT162b2, the focus of this review, generated a large amount of safety data. The initial approvals were based on compelling efficacy and short-term safety data (up to 2 months follow-up post-primary schedule for BNT162b2, in line with regulatory guidance).⁵ Variations of these approvals to include a booster dose were based on 2.6 months follow-up post-BNT162b2 booster.⁶ After authorization, vaccine manufacturers continued clinical trials and collaborated with regulatory authorities and other organizations for post-marketing pharmacovigilance activities to monitor longer-term safety. The quantity and breadth of this post-authorization safety data quickly surpassed that generated from clinical trials. This was supported by real-world data from countries initiating mass vaccination programs,¹ as well as independent clinical and real-world trials in special populations and with different vaccination regimens, which confirmed the safety profile observed in clinical trials, allowed analysis of real-world practice patterns, and created an unprecedented level of transparency.⁷

The ongoing emergence of new variants of SARS-CoV-2 has led to further generation of safety data for mRNA vaccines, as booster doses and variant-adapted bivalent vaccines have been developed and brought to the market. Here, we review safety and reactogenicity data for the BNT162 mRNA vaccine candidates, including BNT162b2 (Comirnaty, Pfizer/BioNTech), from preclinical studies, clinical trials, post-marketing surveillance, and real-world studies, as well as from studies of booster doses and variant-adapted BNT162b2 vaccines.

Methods

A literature search was performed in Medline (PubMed) and pre-print servers medRxiv and bioRxiv for English-language articles. No additional inclusion or exclusion criteria were applied. Additional references were obtained by searching citation lists of retrieved articles, and the authors identified further appropriate references for inclusion based on expert knowledge. Other sources of information included pharmaceutical company press releases and public health websites. Systematic search methodology was not used.

The initial formal literature search, comprising terms such as BNT162b2, SARS-CoV-2, adverse event (AE), AE following immunization, and AE of special interest (AESI; including itemized terms such as ‘myocarditis’, ‘Bell’s Palsy’, ‘syncope’, etc.), was performed on October 05, 2023. After the date of the initial search, owing to the fast-moving nature of the field, additional targeted follow-up searches were performed throughout manuscript development, from first draft to finalization.

Results

Clinical development of BNT162b2 vaccines

Clinical development of BNT162 vaccines was carried out in line with regulatory agency standards and applicable guidelines.^5,8,9 Although timelines were compressed because of the public health emergency, the overall clinical development process used was the same as that used for the development and approval of other vaccines in terms of trials performed and data collected (Figure 1).¹⁰ Expectations for safety evaluations, including the volume and quality of data collected, were the same for vaccines developed in accelerated and non-emergency pre-approval environments.⁷

Both the European Medicines Agency (EMA) and the United States Food and Drug Administration (FDA) devoted extra resources to support the rapid development and authorization of vaccines.⁷ The FDA EUA pathway enabled the public to access COVID-19 vaccines in the emergency pandemic setting quickly while requiring manufacturers to supply additional vaccine data from clinical settings for rigorous review.⁷ Similarly, the EMA expedited its evidence appraisal process through rolling review stages,^3,11 which allowed regulators to receive and review the data as soon as they became available. Regular dialogue between manufacturers, regulatory authorities, and other stakeholders was essential in expediting timelines while adhering to the development process. Increased familiarity with the data, and preplanned opportunities for sponsor/regulator discussions, enabled regulators to provide expedited assessments of marketing authorization applications.³ Conditional marketing authorization for BNT162b2 was first granted by the European Commission on December 21, 2020.¹²

Vaccine development was also able to be expedited by building upon experience gained from development of severe acute respiratory syndrome coronavirus (SARS-CoV) and Middle East respiratory syndrome coronavirus (MERS-CoV) vaccines,¹³ and because developers had clinical, regulatory, and manufacturing experience with mRNA formats and lipid nanoparticle (LNP)-based formulations.^14,15 In accordance with guidance issued in early 2020, led by the International Coalition of Medicines Regulatory Authorities (ICMRA), well-characterized toxicology data could be leveraged across closely related vaccines using the same platform technology (e.g., same RNA formats and LNPs).¹⁶ Therefore, a platform-based toxicology program could be conducted to explore diverse combinations of mRNA chemistries and formats with different receptor-binding domain (RBD) or full-length SARS-CoV-2 spike antigen designs in conjunction with the selected LNP formulation.

Highly reproducible preclinical data were obtained from BNT162 vaccine candidates across independent toxicological evaluations in animal models.¹⁷ These efforts established an early platform toxicology profile that enabled rapid initiation of a first-in-human clinical trial investigating several of these vaccine candidates (Figure 1).¹⁷ Early stage clinical trials used sentinel cohorts and dose escalation to optimize enrollment rate and monitor safety, with stopping rules in place in the event of safety issues.¹⁸ An independent data monitoring committee reviewed safety data from clinical trials to determine whether any changes to planned doses should be made, and any treatment group should be terminated early.¹⁸ Double-blind study designs allowed for true assessment of safety events, and the large numbers of subjects enabled robust assessment of AEs. For example, clinical evaluation of BNT162b2 vaccines in approximately 20,000 vaccinated participants enabled the detection of events with an incidence of > 1 event in 1,000 vaccinated individuals.¹⁹ Key safety and reactogenicity data from BNT162 preclinical and clinical trials are described below.

Preclinical studies of original BNT162 candidate vaccines

Preclinical assessment of BNT162 candidate vaccines was performed in rats, mice, and monkeys. BNT162 vaccine candidates coding for the RBD of the spike glycoprotein (S protein) have vaccine identifier names that end in “−1, −3,” whereas those coding for the full-length S protein have vaccine identifier names that end in “−2” (Table 1).¹⁵ The preceding letter refers to distinct RNA platforms differing in chemistry or format used to construct the vaccine (a = unmodified uridine-containing RNA [uRNA]; b = pseudouridine-modified RNA [modRNA]; c = self-amplifying RNA [saRNA]).^15,20 All variants were formulated with the same lipid formulation.

Table 1. BNT162 vaccine candidates evaluated in (A) preclinical trials and (B) early clinical trials.

BNT162 vaccine candidate (product code)^†	Platform	SARS-CoV-2 variant^†	Encoded antigen	Evaluation
(A) Preclinical trials^Citation17,Citation23
BNT162a1	uRNA	Wild type	SARS-CoV-2 RBD, a secreted variant	Rats^‡
BNT162b1	modRNA	Wild type	SARS-CoV-2 RBD, a secreted variant	Rats, mice, rhesus macaques
BNT162b2	modRNA	Wild type	Full-length SARS-CoV-2 S protein bearing mutations preserving neutralization-sensitive sites^,§ a secreted variant	Rats, mice, rhesus macaques
BNT162b3	modRNA	Wild type	SARS-CoV-2 RBD, a membrane-bound variant	Rats^‡
BNT162c1	saRNA	Wild type	SARS-CoV-2 RBD, a secreted variant	Rats^‡
(B) Early clinical trials^Citation26–30,Citation32,Citation105
BNT162a1	uRNA	Wild type	SARS-CoV-2 RBD, a secreted variant	Germany, NCT04380701
BNT162b1	modRNA	Wild type	SARS-CoV-2 RBD, a secreted variant	Germany, USA, China NCT04380701 NCT04368728 NCT04523571
BNT162b2	modRNA	Wild type	Full-length SARS-CoV-2 S protein bearing mutations preserving neutralization-sensitive sites, a secreted variant	Germany, USA, Brazil, Argentina, Turkey, Germany NCT04380701 NCT04368728
BNT162c2	saRNA	Wild type	Full-length SARS-CoV-2 S protein, a secreted variant	Germany NCT04380701

^†Wild type refers to the USA-WA1/2020 strain. ^‡Data on file. ^§Two sequence variants, V8 and V9, were evaluated in preclinical studies; V9 was included in the final approved vaccine. modRNA, modified RNA; RBD, Receptor-Binding Domain; S protein, spike protein; saRNA, self-amplifying RNA; SARS-CoV-2, Severe Acute Respiratory Syndrome Coronavirus 2; uRNA, uridine RNA.

Five vaccine candidates were evaluated in preclinical repeat-dose toxicity studies in Wistar Han rats: four encoded different sequence variants on the RBD of the S protein (BNT162a1, BNT162b1, BNT162b3 and BNT162c1), and one encoded the full-length S protein in its pre-fusion conformation (BNT162b2, with sequence variants V8 and V9; Table 1A).¹⁷ Each candidate vaccine was tolerated without evidence of systemic toxicity at all doses evaluated, and vaccine-related findings for all candidates were similar. Clinical signs, including swelling at the injection site and increases in white blood cells, were as expected and consistent with vaccine administration and immune response; all effects were reversible.¹⁷ There were no effects of BNT162b2 on fertility parameters or fetal development.^21,22 The BNT162b1 and BNT162b2 vaccine candidates were also evaluated in mice and rhesus macaques (Table 1A).²³

Early clinical trials of original BNT162 vaccines

Clinical dose-evaluation studies were performed to select the vaccine candidate with an optimal safety, reactogenicity, and immunogenicity profile.^24,25

In an initial Phase I/II dose-escalation study conducted in Germany in adults 18–85 years of age (NCT04380701), four BNT162 vaccine candidates were evaluated: BNT162a1, BNT162b1, BNT162b2, and BNT162c2 (Table 1).^26,27 Injection-site reactions within 7 days of administration were mainly injection-site pain and tenderness, and the severity of reported reactions was mostly mild or moderate, with an occasional severe (Grade 3) event. The incidence of reactogenicity events was dose-dependent. All AEs resolved spontaneously. No serious AEs were reported, and no withdrawals due to AEs were observed for any dose.^27,28

A Phase I, randomized, placebo-controlled, dose-escalation trial in the United States in adults 18–55 years and 65–85 years of age evaluated BNT162b1 and BNT162b2 (NCT04368728).²⁹ The most common local reaction within 7 days after receipt of BNT162b1 and BNT162b2 was injection-site pain; this was more frequent after the second dose.^29,30 Most local reactions and systemic events resolved by day 7.^29,30 No participant who received either candidate reported a Grade 4 reactogenicity event, and there were no serious AEs.²⁹ Systemic reactogenicity events were milder with BNT162b2 than with BNT162b1, and fewer recipients of BNT162b2 reported using antipyretic or pain medication, while the neutralizing antibody profiles of the vaccines were similar. BNT162b2, which encodes the full-length SARS-CoV-2 S protein, was selected for further development. The 30 μg dose was selected as the preferred dose level,²⁹ driven by neutralizing antibody response and tolerability data across all subgroups by age and sex.

Similar safety and reactogenicity profiles were also observed for BNT162b1 in Chinese adults and for BNT162b2 in Japanese adults in Phase I and Phase I/II clinical trials (NCT04523571 and NCT04588480, respectively).^31,32

Phase II/III clinical trials of original BNT162b2

The pivotal Phase II/III clinical trial of BNT162b2 (NCT04368728) was a continuation of the Phase I trial in the United States, expanded to a global scope. This trial randomized 43,548 participants ≥16 years of age to receive two doses of 30 μg BNT162b2 or placebo, and included participants from 130 sites in the United States, Argentina, Brazil, South Africa, Germany, and Turkey.¹⁹ The overall safety population included 18,860 participants who received BNT162b2. In total, 58% were 16–55 years of age and 42% were >55 years of age.¹⁹

The prescribing information for BNT162b2 presents early reactivity data from the trial among 5,807 participants 16–55 years of age (2,682 randomized to BNT162b2 and 2,684 randomized to placebo; data cutoff March 13, 2021) (Table 2).³³ In this dataset, the majority of local reactions after dose 2 were mild, with a mean duration of <3 days.³³ In a larger reactogenicity subset (N = 8,183), the most commonly reported local reaction was injection-site pain within 7 days of dose 1 administration, occurring in 83% and 71% of BNT162b2-vaccinated participants 16–55 years of age and >55 years of age, respectively. The proportion of participants vaccinated with BNT162b2 reporting local reactions within 7 days of administration was lower after the second dose (78% and 66%, respectively).¹⁹ No participant reported a Grade 4 local reaction.

Table 2. Frequency and percentages of participants with solicited local reactions, by maximum severity, within 7 days after each dose – participants 16–55 years of age – reactogenicity subset.^†,33.

	BNT162b2 Dose 1 N^‡  = 2,899 n^§ (%)	Placebo Dose 1 N^‡  = 2,908 n^§ (%)	BNT162b2 Dose 2 N^‡  = 2,682 n^§ (%)	Placebo Dose 2 N^‡  = 2,684 n^§ (%)
Redness^‖
Any (>2.0 cm)	156 (5.4)	28 (1.0)	151 (5.6)	18 (0.7)
Mild	113 (3.9)	19 (0.7)	90 (3.4)	12 (0.4)
Moderate	36 (1.2)	6 (0.2)	50 (1.9)	6 (0.2)
Severe	7 (0.2)	3 (0.1)	11 (0.4)	0
Swelling^‖
Any (>2.0 cm)	184 (6.3)	16 (0.6)	183 (6.8)	5 (0.2)
Mild	124 (4.3)	6 (0.2)	110 (4.1)	3 (0.1)
Moderate	54 (1.9)	8 (0.3)	66 (2.5)	2 (0.1)
Severe	6 (0.2)	2 (0.1)	7 (0.3)	0
Pain at the injection site^¶
Any	2,426 (83.7)	414 (14.2)	2,101 (78.3)	312 (11.6)
Mild	1,464 (50.5)	391 (13.4)	1,274 (47.5)	284 (10.6)
Moderate	923 (31.8)	20 (0.7)	788 (29.4)	28 (1.0)
Severe	39 (1.3)	3 (0.1)	39 (1.5)	0

Reactions were collected in the electronic diary (e-diary) from day 1 to day 7 after vaccination. No Grade 4 solicited local reactions were reported in participants 16–55 years of age. ^†Randomized participants in the safety analysis population who received at least one dose of the study intervention. Participants with chronic, stable human immunodeficiency virus (HIV) infection were excluded. ^‡N = number of participants reporting at least one yes or no response for the specified reaction after the specified dose. The N for each reaction was the same; therefore, this information was included in the column header. ^§N = number of participants with the specified reaction. ^‖Mild: > 2.0 to ≤ 5.0 cm; moderate: > 5.0 to ≤ 10.0 cm; severe: > 10.0 cm. ^¶Mild: does not interfere with activity; moderate: interferes with activity; severe: prevents daily activity.

Systemic events were more frequent after the second dose and in younger (16–55 years of age) versus older (≥55 years of age) participants.¹⁹ In total, ≤2% of participants vaccinated with BNT162b2 reported solicited systemic events graded as severe within 7 days of administration, with the exception of fatigue, which occurred in 3.8% of participants after dose 2.¹⁹ Two participants vaccinated with BNT162b2 reported a temperature of ≥40°C within 7 days of administration, as did two participants in the placebo group.¹⁹ Overall, reactogenicity events were transient and resolved within 1–2 days after onset.¹⁹ The reactogenicity profile of a larger subgroup of participants, including the initial 8,183 participants plus another 1,656 enrolled after the initial data cutoff, remained consistent with the earlier datasets.³⁴

Among 43,252 participants with variable follow-up time for AEs, lymphadenopathy was reported by 64 (0.3%) BNT162b2 recipients and six (<0.1%) placebo recipients. This generally resolved within 10 days and was likely a result of a robust vaccine-elicited immune response in the BNT162b2 recipients.¹⁹ Participants were followed for serious AEs up to 6 months after dose 2. Four serious AEs considered to be related to vaccination were reported by recipients of BNT162b2; these were shoulder injury, right axillary lymphadenopathy, paroxysmal ventricular arrhythmia, and right leg paresthesia. No deaths were considered to be related to the vaccine, and few patients had AEs that led to trial withdrawal.^19,34

After authorization of BNT162b2, participants were given the option to learn their trial assignment, and those receiving placebo were offered BNT162b2 in the context of the pandemic.³⁴ To ensure long-term follow-up of both BNT162b2 and placebo recipients, participants continued to be followed after unblinding. No safety signals were observed during the extended follow-up period.³⁴

BNT162b2 has also been evaluated in a Phase II trial in 960 Chinese adults 18–85 years of age, in which 720 individuals received BNT162b2 (NCT04649021). The reactogenicity and safety profile of BNT162b2 in this population was consistent with that of the global Phase II/III trial.³⁵

Pediatric development

Following authorization of the BNT162b2 vaccine for adult populations, a clinical trial was performed in participants 6 months–11 years of age to assess pediatric age-related doses of BNT162b2. In the Phase I portion, the cohort of children 5–11 years of age received two doses of BNT162b2 10 μg, 20 μg, or 30 μg (NCT04816643). Owing to a higher frequency of fever with the higher doses, and because the neutralizing antibody profiles of the 10 μg and 20 μg doses were similar, the 10 μg dose was selected for further assessment in the Phase II/III portion.³⁶ In the cohort of children 6 months–4 years of age, participants received either 10 μg or 3 μg BNT162b2. Owing to a higher frequency and greater severity of reactogenicity to the 10 μg dose versus the 3 μg dose, and similar neutralizing antibody profiles across dose levels to that observed in older age groups, the 3 μg dose was selected for further assessment.³⁷

The pivotal Phase II/III clinical trial of BNT162b2 (NCT04368728) also included a cohort of 2,260 participants 12–15 years of age, of whom 1,131 received BNT162b2. Solicited local and systemic events were generally mild or moderate in severity and were reported at a similar frequency to participants 16–25 years of age. One participant discontinued the study due to a vaccine-related event of temperature ≥ 40°C after dose 1. Lymphadenopathy was reported by nine (0.8%) recipients of BNT162b2. Up to 1 month after dose 2, no vaccine-related serious AEs and no deaths had been reported.³⁸ Of the 1,131 participants who received BNT162b2, 786 were followed for ≥4 months after the second dose, with the overall safety profile remaining similar to that seen in participants ≥16 years of age.²¹

In the Phase II/III portion of the clinical trial in children <12 years of age (NCT04816643), 1,517 children 5–11 years of age were randomized to receive 10 μg BNT162b2. Local reactions were generally mild to moderate in intensity and lasted 1–2 days. Consistent with the trial in adults, injection-site pain was the most common local reaction, and fatigue and headache were the most common solicited systemic reactions. One recipient of BNT162b2 reported a temperature of ≥ 40°C after the second dose, which resolved with antipyretics.³⁶ After a median follow-up time of 2.3 months after the second dose (95% of BNT162b2 recipients had a follow-up of ≥2 months), no vaccine-related serious AEs, AEs leading to withdrawal, or deaths had been reported.³⁶ Safety evaluation in this study is ongoing.²¹

In a cohort of 3,013 children 6 months–4 years of age who received BNT162b2, most local and systemic reactions were mild to moderate in intensity and no Grade 4 local reactions were reported. The frequency of AEs was similar in BNT162b2 and placebo recipients, few participants were withdrawn due to AEs, and no deaths occurred.³⁷

Booster doses

The safety and reactogenicity of third and fourth doses of BNT162b2 were assessed in participants from the pivotal Phase II/III clinical trials. In 5,081 participants who received a third dose of BNT162b2, reactogenicity was similar to that observed after the second dose. With a median of 2.5 months of follow-up from dose 3, the safety profile of the vaccine was consistent with earlier trials, and no new safety signals were identified. Three participants who received BNT162b2 experienced serious AEs that were considered to be related to the vaccine; these were tachycardia in one participant and increased hepatic enzyme levels in two participants.³⁹ A subset of 306 participants 18–55 years of age were followed for a median of 8.3 months post-booster dose, with 301 followed for ≥6 months. The overall safety profile of the booster dose remained consistent with that seen after two doses.²¹

In the trial in children <12 years of age, 401 participants 5–11 years of age received a third dose at least 5 months (range 5–9 months) after completing the primary series. With a median follow-up time of 1.3 months, the overall safety profile was similar to that seen after the primary course.²¹ In 570 infants 6–23 months of age with a median of 1.3 months of follow-up after dose 3, and 886 children 2–4 years of age with 1.4 months of follow-up after dose 3, reactogenicity after dose 3 was consistent with the known safety profile of BNT162b2.²¹

In adults 18–55 years of age (N = 325) and >55 years of age (N = 305) who received a fourth dose of BNT162b2, the safety profile was similar to that observed after dose 3, after a median of 1.4 and 1.7 months follow-up post-dose 4, respectively.²¹ Because these studies enrolled participants from the previous trials who consented to further study, results may be subject to selection bias.

Overall, the clinical development program for BNT162b2 included the largest pivotal registrational COVID-19 vaccine trial conducted to date and evaluated more than 44,000 participants >12 years of age from around the world (Table 3).^19,38 These findings were supported by further study in different world regions^31,35 and Phase II/III trials in children >6 months of age.^21,36,37

Table 3. Safety data for approved BNT162b2 doses from clinical trials. Data for approved formulation and doses included only.

Trial	Phase	Location	Recipients (N,^† age)	Doses	Reactogenicity/safety	Publication
Original vaccine primary series
NCT04380701	I/II	Germany	N = 12 18–55 years 66.7% male, 100% Caucasian	Two doses 30 μg	Local reactions mostly mild to moderate in intensity, occasionally severe (Grade 3) All AEs resolved spontaneously and were managed with simple measures No serious AEs or withdrawals due to related AEs	Sahin et al.^Citation28
NCT04368728	I	United States	N = 12 18–55 years 50% male, 83% White, 17% Asian N = 12 65–85 years 33% male, 83% White, 17% Asian	Two doses 30 μg	Most common local reaction was injection-site pain, more frequent after dose 2 No Grade 4 reactogenicity events Related AEs reported by 25% of participants 18–55 years of age and 0% participants 65–85 years of age No serious AEs	Walsh et al.^Citation29
	II/III	Argentina, Brazil, Germany, South Africa, Turkey, United States	N = 21720 ≥16 years 51.1% male, 82.9% White, 9.2% Black or African American, 4.2% Asian	Two doses 30 μg	No Grade 4 local reactions; most local events resolved within 1–2 days Severe systemic events in ≤2% except fatigue (3.8%) within 7 days after dose 2 Four related serious AEs and no vaccine-related deaths with up to 6 months follow-up	Polack et al.^Citation19 Thomas et al.^Citation34
			N = 1,131 12–15 years 50.1% male, 85.9% White, 4.6% Black or African American, 6.4% Asian	Two doses 30 μg	Local and systemic reactions generally mild to moderate, resolved within 1–2 days Severe injection-site pain in 1.5%, one discontinuation due to temperature ≥40°C Lymphadenopathy in 0.8% No serious AEs related to vaccine In 786 participants with ≥4 months follow-up after dose 2, the safety profile remained similar to that in ≥16 years of age after	Frenck et al.^Citation38 EMA SmPC^Citation21
NCT04816643	I/II	United States	N = 16 5–11 years 31.3% male, 68.8% White, 18.8% Black or African American, 12.5% Asian	Two doses 10 μg	Most local reactions mild to moderate, all transient	Walter et al.^Citation36
			N = 16 6 months to <2 years 62.5% male, 87.5% White, 6.3% Asian N = 16 2–4 years 56.3% male, 75.0% White, 6.3% Asian	Two doses 3 μg	Local and systemic reactions generally mild to moderate	Munoz et al.^Citation37
	II/III	Finland, Poland, Spain, United States	N = 1,518 5–11 years 52.6% male, 79.3% White, 5.9% Black, 5.9% Asian	Two doses 10 μg	Local and systemic reactions generally mild to moderate Severe reactions: injection-site pain 0.6%, fatigue 0.9%, headache 0.3% No vaccine-related serious AEs, AEs leading to withdrawal, or deaths with median follow-up of 2.3 months after dose 2 (95% with ≥2 months follow-up)	Walter et al.^Citation36
	II/III	Brazil, Finland, Poland, Spain, United States	N = 1,178 6 months to <2 years 50.0% male, 78.3% White, 3.6% Black, 7.7% Asian N = 1,835 2–4 years 49.1% male, 80.1% White, 5.1% Black, 6.9% Asian	Two doses 3 μg	Most reactogenicity events were mild to moderate, no Grade 4 local reactions Few participants withdrawn due to AEs No deaths occurred	Munoz et al.^Citation37
NCT04588480	I/II	Japan	N = 97 20–64 years 51.5% male, 100% Asian N = 22 65–85 years 40.9% male, 100% Asian	Two doses 30 μg	Most common local reaction: injection-site pain Reactogenicity events mild to moderate and generally transient No serious/life-threatening AEs and no deaths	Haranaka et al.^Citation31
NCT04649021	II	China	N = 720 18–85 years 51.5% male, 100% Chinese (Han ethnicity)	Two doses 30 μg	Most reactogenicity events mild to moderate and transient No vaccine-related serious AEs or deaths	Hui et al.^Citation35
Original vaccine booster doses
NCT04955626	II/III	Brazil, South Africa, United States	N = 5,081 ≥16 years 48.4% male, 78.7% White, 9.3% Black, 5.7% Asian	Third dose 30 μg	Reactogenicity after dose 3 similar to that after dose 2 With median follow-up of 2.5 months after dose 3, the safety profile remained consistent with primary series Three vaccine-related serious AEs Subset of 306 participants followed for median 8.3 months, no new safety signals identified	Moreira et al.^Citation39 EMA SmPC^Citation21
	III	Not stated	N = 305 >55 years	Fourth dose 30 μg	With median follow-up of 1.7 months after dose 4, the safety profile was similar to that seen after dose 3	EMA SmPC^Citation21
	III	Not stated	N = 325 18–55 years	Fourth dose 30 μg	With median follow-up of 1.4 months after dose 4, reactogenicity was consistent with known safety profile of BNT162b2	EMA SmPC^Citation21
NCT04816643	II/III	Not stated	N = 401 5–11 years	Third dose 10 μg	With median follow-up time of 1.3 months after dose 3, overall safety profile was similar to that seen after the primary course	EMA SmPC^Citation21
		Not stated	N = 1,456 6 months to 4 years	Third dose 3 μg (any primary course dose)	In 570 infants 6–23 months of age with median 1.3 months follow-up, and 886 children 2–4 years of age with 1.4 months follow-up after dose 3, reactogenicity after dose 3 was consistent with the known safety profile of BNT162b2	EMA SmPC^Citation21
Original/Omicron BA.1 variant-adapted vaccine
NCT04955626	III	United States	N = 305 >55 years 53.1% male, 89.8% White, 4.3% Black, 5.2% Asian	Fourth dose 30 μg	Most reactogenicity events mild to moderate No Grade 4 reactogenicity events No new adverse reactions identified No vaccine-related serious AEs, AEs leading to withdrawal, or deaths	Winokur et al.^Citation40
Original/Omicron BA.4–5 variant-adapted vaccine
NCT05472038	II/III	United States	N = 74 ≥18 years	Fourth dose 30 μg	Safety profile similar to that of the original vaccine	BioNTech^Citation41
XBB.1.5 variant-adapted vaccine
NCT05997290	II/III	United States	N = 412 ≥12 years 41.3% male, 79.1% White, 12.6% Black, 5.3% Asian	Fourth to seventh dose 30 μg	After 1 month follow-up, most reactogenicity events mild to moderate AEs infrequent (7.5% in the total population); none led to study withdrawal No myocarditis or pericarditis reports No new safety signals	Gayed et al.^Citation42

^†Total N randomized to BNT162b2 at dose shown.

AE, adverse event; EMA, European Medicines Agency; SmPC, Summary of Product Characteristics.

Variant-adapted vaccines

The emergence of SARS-CoV-2 variants became apparent shortly after the first COVID-19 vaccines were deployed, prompting developers to evaluate updated vaccine candidates. When the Omicron BA.1 variant of SARS-CoV-2 and its subsequent lineages emerged, BioNTech/Pfizer pursued the development of several variant-adapted vaccines, including a monovalent BA.1 vaccine candidate, a bivalent vaccine candidate targeting both the original wild-type virus and BA.1, and a bivalent vaccine targeting the original wild-type virus and BA.4/BA.5.⁴⁰

In view of the public health urgency for protection against Omicron subvariants, the bivalent vaccines encoding the Original/Omicron BA.1 and Original/Omicron BA.4/5 spike proteins were approved by regulators in different regions. These approvals were based on clinical trials conducted with BA.1 vaccine candidates (monovalent and bivalent, at different dose levels) demonstrating significant increases in neutralizing antibody titers,⁴⁰ as well as preclinical immunogenicity data, and were supported by the extensive and robust safety database for the original BNT162b2 vaccine and the clinical data generated across multiple variant-adapted vaccine candidates encoding earlier SARS-CoV-2 variants.⁴³ In parallel, clinical trials were initiated to further evaluate the safety and reactogenicity of these vaccines.

In a Phase III clinical trial, adults >55 years of age who had previously received three doses of the original BNT162b2 vaccine were randomized to receive either bivalent Original/Omicron BA.1 BNT162b2 vaccine or original BNT162b2 (NCT04955626) (Table 3). In total, 306 participants were randomized to receive a 30 μg dose and 316 received a 60 μg dose of bivalent vaccine.⁴⁰ The bivalent Original/Omicron BA.1 BNT162b2 vaccine had a similar local reaction and systemic event profile to the original BNT162b2 vaccine.^21,40 Most reactogenicity events were mild to moderate in intensity and no Grade 4 reactogenicity events were reported. Injection-site pain and fatigue were the most common local and systemic reactogenicity events, respectively. In participants >55 years of age, mild-to-moderate injection-site pain, fatigue, and muscle pain were more common with the 60 μg dose than with the 30 μg dose of bivalent vaccine.⁴⁴ No serious AEs were considered related to the bivalent vaccine, and there were no AEs leading to withdrawal or deaths.⁴⁰ No new safety signals were detected.²¹

The 30 μg bivalent Original/Omicron BA.4–5 vaccine was assessed in a Phase II/III trial in adults 18–55 years of age and >55 years of age who had previously received three doses of the original BNT162b2 vaccine (NCT05472038).⁴⁵ Early clinical safety data from 7 and 30 days post-vaccination indicate that the bivalent Original/Omicron BA.4–5 BNT162b2 vaccine is well tolerated, with a safety profile similar to that of the original vaccine.^41,45

Continued evolution of SARS-CoV-2 led to recommendations from the WHO Technical Advisory Group on COVID-19 Vaccine Composition (TAG-CO-VAC) to include a component of an XBB.1 descendant sub-lineage of SARS-CoV-2 in COVID-19 vaccines for fall 2023.⁴⁶ Following consistent guidance issued by the EMA/European Centre for Disease Control and Prevention,⁴⁷ and by the FDA,⁴⁸ BioNTech and Pfizer developed a monovalent XBB.1.5 vaccine that has received regulatory approvals in various countries.^49,50 1-month safety follow-up from an ongoing Phase II/III study of this vaccine in participants ≥12 years of age (NCT05997290) did not detect any new safety signals. Local reactions and systemic events were mostly mild to moderate in severity, AEs were infrequent (7.5% of the total population), and no AEs led to study withdrawal.⁴²

Post-marketing surveillance and real-world data

In both the United States and European Union, post-authorization safety monitoring and risk minimization procedures were put in place following BNT162b2 commercialization. International Risk Management Plans (RMPs), or comprehensive documents summarizing the vaccine safety profile and measures taken to further investigate and mitigate risks, were submitted alongside dossiers for marketing authorization.⁵¹ Post-authorization commitments included in the European Union RMP included studies to assess the risk of vaccine-associated enhanced disease, impacts on special populations (including pregnant/breastfeeding women and those with compromised immune systems or comorbidities), and the potential for interactions between vaccines, as well as studies to obtain longer-term safety data.⁵² After FDA approval following a period of EUA, additional post-marketing studies to assess the risk of myocarditis and pericarditis were required. A registry to evaluate pregnancy and infant outcomes was also established.⁵³ In the European Union, the EMA Pharmacovigilance Risk Assessment Committee published regular safety update reports based on data from the marketing authorization holder and data reported by either patients or healthcare providers to EudraVigilance.^12,54

Post-marketing safety reporting for COVID-19 vaccines has been extremely robust compared with other vaccines. COVID-19 vaccine safety surveillance in the United States has been described as the most intensive in the country’s history.⁵⁵ During the peak of the pandemic, manufacturers of COVID-19 vaccines were required to submit monthly safety reports to the EMA.⁵⁶ These safety updates subsequently reduced in frequency and were eventually discontinued when the EMA concluded that the safety profile of the vaccines had been well established.^56,57 Monthly safety reports continue to be submitted to the FDA.⁵⁸ Marketing authorization holders are legally obliged to set up and maintain database systems for authorized products to receive, record, and analyze safety data and signals from spontaneous reporting.⁵⁹

Post-authorization, the safety of BNT162b2 and other COVID-19 vaccines has been monitored through both passive and active surveillance systems.⁶⁰ Together, these surveillance systems (Table 4) capture safety data from an extensive and heterogenous global population. In European Union/European Economic Area countries alone, as of June 16, 2023, almost 535 million doses of original BNT162b2 and 32 million doses of bivalent BNT162b2 have been administered.⁶⁴ In the United States, as of May 10, 2023, more than 360 million and 36 million doses, respectively, have been administered.⁶⁵

Table 4. Surveillance systems.

	Passive surveillance systems	Active surveillance systems	Other
Purpose	Collect spontaneous AE reports	Collect all reports of pre-specified AEs from a representative population, such as a sentinel site	Track post-marketing surveillance AEs Collaborate with academic and non-government healthcare systems
Example	CDC VAERS in the United States and the EMA EudraVigilance system in the European Union^Citation57,Citation61,Citation62	United States VSD and BEST^Citation61,Citation62	CISA and the CBER^Citation61,Citation62 WHO VigiBase captures reports from national centers from over 130 countries^Citation63

AE, Adverse Event; BEST, Biologics Effectiveness and Safety; CBER, Center for Biologics Evaluation and Research; CDC, United States Centers for Disease Control and Prevention; CISA, CDC Clinical Immunization Safety Assessment; EMA, European Medicines Agency; VAERS, Vaccine Adverse Event Reporting System; VSD, Vaccine Safety Datalink; WHO, World Health Organization.

When rare events are detected by a safety monitoring system, manufacturers, as well as regulatory bodies such as the FDA and EMA, review the signals. This is done, for example, by comparing rates in vaccinated and unvaccinated people, comparing rates in an early reporting period with a later interval when more data have been collected, and comparing with data from other surveillance systems, databases, and published literature. If an event is determined as likely to be associated with vaccination, regulatory agencies may then request changes to vaccine recommendations or product information. Monitoring and analysis of the event will continue in order to detect any changes to a signal, until the event is closed.^55,57

Implementation of mass vaccination programs also enables real-world studies of vaccine safety in vaccinated populations. These studies can complement post-marketing surveillance by providing additional data on safety signals. They may also provide estimates on more common events, such as reactogenicity, in different populations from those enrolled in the clinical trials. For COVID-19 vaccines, real-world studies had an important impact on post-authorization regulatory agency and government decision-making due to their unprecedented number, early conduct, and rapid publication.

Selected AESIs reported with BNT162b2 vaccination

AESIs are pre-identified and pre-defined serious or non-serious AEs of scientific and medical interest for which ongoing monitoring, further investigation, and/or rapid communication by the sponsor to the regulator or other stakeholders may be appropriate.^66,67 AESIs are usually identified through active vaccine safety surveillance systems.⁶⁷ Prior to vaccine use in the community, AESIs are identified based on known occurrence patterns in the population and their potential to be associated with one or more vaccine platforms.⁶⁷ AE case definitions are developed by the Brighton Collaboration, which also maintains a list of vaccine platform-related AESIs.^11,67,68 AESIs that have been reported following BNT162b2 vaccination include Bell’s palsy, and myocarditis and pericarditis.

Facial paralysis and swelling, bell’s palsy

Acute peripheral facial paralysis, or Bell’s palsy, is a rare event associated with BNT162b2 vaccination, occurring in ≥1 in 10,000 to <1 in 1,000 recipients (Table 5).²¹ This rare event was initially detected during clinical trials,²¹ and cases have since been observed post-BNT162b2 administration during mass vaccination campaigns,^69,70 although a causal association between these events and BNT162b2 vaccination has not been clearly demonstrated.^71,72 In case – control studies in Hong Kong and Israel, rates of facial nerve palsy/Bell’s palsy reported after COVID-19 vaccination were similar to background rates recorded during years prior to the emergence of COVID-19.^71,73

Table 5. Frequency of selected adverse events and AESIs following BNT162b2 vaccination in individuals ≥12 years of age based on clinical trials and post-authorization experience²¹.

Adverse event/AESI	Frequency
Acute peripheral facial paralysis (Bell’s palsy)	Rare (≥ 1/10,000 to < 1/1,000)
Myocarditis/pericarditis	Very rare (< 1/10,000)
Hypersensitivity reactions	Uncommon (≥ 1/1,000 to < 1/100)
Heavy menstrual bleeding	Not known (cannot be estimated from available data)
Anaphylaxis^†	Not known (cannot be estimated from available data)

^†Appropriate medical treatment and supervision should always be readily available in case of an anaphylactic reaction following administration of the vaccine. Close observation for at least 15 minutes is recommended following vaccination. No further dose of the vaccine should be given to those who have experienced anaphylaxis after a prior dose.²¹ AESI, adverse event of special interest.

Facial nerve palsy has been described as an AE following administration of other vaccines, including influenza, varicella, human papillomavirus, diphtheria – tetanus – pertussis, and meningococcal vaccines.⁷⁴ Causality is not confirmed, with the exception of an intranasal influenza vaccine containing Escherichia coli heat-labile toxin as an adjuvant, leading to the hypothesis that this toxin or the route of administration may be a cause.^72,75 BNT162b2 does not include this toxin and uses a different mechanism to stimulate an immune response.⁷¹

Myocarditis and pericarditis

Myocarditis/pericarditis is a very rare AE that has been observed after administration of COVID-19 mRNA vaccines (Table 5).⁷⁶ Owing to its rarity, this event was not detected in clinical trials, but was reported post-authorization.²¹ Myocarditis and pericarditis are considered important identified risks for COVID-19 mRNA vaccines, including original BNT162b2 vaccine and variant-adapted vaccines, and are included and described accordingly in the European Union RMP and aggregate safety reports. The frequency varies depending on the mRNA vaccine used.⁷⁷ After BNT162b2 vaccination, myocarditis has occurred in <1 in 10,000 recipients.²¹

Myocarditis has been reported within a few days of vaccination, usually within 14 days. It is more frequently observed after the second dose than the first dose of the vaccine, and its incidence is highest in young men.²¹ A large European pharmacoepidemiologic study estimated that the excess risk of myocarditis over the 7 days post-dose 2 was 0.265 (95% confidence interval [CI]: 0.255–0.275) extra cases per 10,000 for men 12–29 years of age, compared with unexposed persons. Another similar study estimated an excess of 0.56 extra cases per 10,000 during the 28 days post-dose 2 in men 16–24 years of age.²¹ In the United States, the estimated incidence of myocarditis during the 7 days post-dose 2 of any COVID-19 mRNA vaccine based on cases reported to the United States Centers for Disease Control and Prevention (CDC) Vaccine Adverse Event Reporting System (VAERS) was 40.6 cases per million doses in men 12–29 years of age and 2.4 per million in those ≥30 years of age; corresponding rates for women were 4.2 and 1.0 per million, respectively.⁷⁶ Estimates from a systematic literature review reported a range of 50–139 and 28–147 cases per million in young men 12–17 years of age and 18–29 years of age, respectively, after COVID-19 mRNA vaccination.⁷⁷

The clinical course and prognosis of post-vaccination myocarditis is comparable with myocarditis of other causes.⁵⁴ In fact, the available literature supports that COVID-19 vaccine-associated myocarditis has a milder presentation, higher recovery rate, and lower mortality compared with myocarditis of other causes (Table 6).^78–82 In the few cases observed, the severity of myocarditis and pericarditis following BNT162b2 vaccination varied; most patients responded well to medications and rest, with prompt improvement of symptoms. Preliminary United States CDC surveys conducted ≥90 days post-diagnosis showed that most patients fully recovered.⁸³ Data from the United States CDC, VAERS, and a European observational study suggest that SARS-CoV-2 infection-related myocarditis occurs in 15–400 people per 10,000 infected.^84–87 Based on a study evaluating magnetic resonance imaging from the University of Toronto and a systematic literature review, the patterns of myocardial injury with vaccine‐associated myocarditis are similar to those seen with idiopathic or viral myocarditis, such as that induced by COVID-19.^81,82 Immunopathology data from the United States suggest that myocarditis post-vaccination might be driven by elevation of circulating inflammatory cytokines and corresponding lymphocytes, with no evidence of cardiac-targeted autoantibodies, hypersensitivity, or hyperimmune humoral mechanisms.⁸⁸

Table 6. Key studies comparing post-vaccination myocarditis with myocarditis from other causes including COVID-19.

Publication	Location	Design	Population	Outcomes
Le Vu et al. 2022^Citation78	France	Matched case – control study	1,612 cases of myocarditis, within a population of 32 million 12–50 years of age (21.2 million received first and 19.3 million received second doses of BNT162b2; 2.86 million received first and 2.58 million received second doses of mRNA-1273)	A lower frequency of intensive care unit admission, mechanical ventilation, and death was observed in patients with post-mRNA vaccination myocarditis than in unvaccinated patients with myocarditis from any cause
Patone et al. 2022^Citation79	England	Self-controlled case series	2,861 cases of myocarditis, within a population of 42,842,345 people receiving at least one dose of COVID-19 vaccine	Risk of myocarditis was increased in the 1 to 28 days after a first, second, and booster dose of BNT162b2 (1.52 [95% CI 1.24–1.85], 1.57 [95% CI 1.28–1.92], and 1.72 [95% CI 1.33–2.22]) but was lower than the risks after a positive SARS-CoV-2 test before or after vaccination (11.14 [95% CI 8.64–14.36] and 5.97 [95% CI 4.54–7.87])
Husby et al. 2023^Citation80	Denmark, Finland, Norway, and Sweden	Nationwide register data	7,292 patients with new onset myocarditis (530 associated with mRNA vaccination, 109 associated with COVID-19, 6,653 with conventional myocarditis), within a population of 23 million individuals	At 90 days follow-up, relative risk of heart failure was higher for myocarditis associated with COVID-19 than for myocarditis associated with vaccination (1.48 [95% CI 0.86–2.54] and 0.56 [95% CI 0.37–0.85] versus conventional myocarditis) At 90 days follow-up, relative risk of death was higher for myocarditis associated with COVID-19 than for myocarditis associated with vaccination (2.35 [95% CI 1.06–5.19] and 0.48 [95% CI 0.21–1.09] versus conventional myocarditis) Among patients 12–39 years of age with no predisposing comorbidities, the relative risk of heart failure or death was markedly higher for myocarditis associated with COVID-19 disease than for myocarditis associated with vaccination (relative risk 5.78 [95% CI 1.84–18.20])

CI, confidence interval; COVID-19, coronavirus disease 2019; mRNA, messenger RNA; SARS-CoV-2, severe acute respiratory syndrome coronavirus 2.

Other select AEs following BNT162b2 immunization

Hypersensitivity

Hypersensitivity to BNT162b2 has been reported, with symptoms such as rash and pruritus occurring in ≥1 in 1,000 to <1 in 100 recipients ≥12 years of age (Table 5).²¹ Allergic reactions may be due to active ingredients or excipients used in the vaccine formulation.⁸⁹ The incidence of anaphylaxis (severe, life-threatening allergic reaction) is low and, thus, difficult to estimate from available data.²¹

After administration of 1,893,360 first doses of BNT162b2, 21 cases meeting Brighton Collaboration criteria for anaphylaxis were identified in the CDC’s VAERS, corresponding to a rate of 11.1 cases per million doses.⁹⁰ A later report after 9,943,247 doses of BNT162b2 estimated a rate of 4.7 cases of anaphylaxis per million doses administered.⁹¹ In total, 77% of individuals had a past history of allergic reaction, and 34% had a past history of anaphylaxis.⁹¹

Although anaphylaxis was one of the first safety signals to be detected with BNT162b2 during post-marketing surveillance, with progressive exposure, it became apparent that the cumulative number of cases was very small.¹² Late hypersensitivity reactions (appearing or lasting >24 h after vaccination) linked with prior exposure to hyaluronic acid dermatological fillers have been reported in a study in Israel, but these did not prevent individuals from receiving subsequent doses of the vaccine.⁹²

Anxiety-related reactions

Anxiety-related reactions, such as syncope, dizziness, palpitations, increases in heart rate, alterations in blood pressure, paresthesia, hypoesthesia, and sweating, may occur after BNT162b2 vaccination. These are not AEs caused by the vaccine formulation, but are rather due to stress and anxiety associated with the vaccination process itself.²¹ These stress-related reactions are temporary and will resolve on their own. Individuals suffering from anxiety should bring symptoms to the attention of the vaccination provider for evaluation. Precautions should be taken to avoid injury from fainting.²¹

Heavy menstrual bleeding

Heavy menstrual bleeding (defined as increased volume or duration that interferes with quality of life) has been reported after BNT162b2 administration.^21,93,94 The incidence of this event is unknown (Table 5), and based on evidence from nearly 9,000 cases from clinical trials, observational studies, post-marketing surveillance activities, and spontaneous reports, most cases have appeared to be non-serious and temporary in nature.⁹⁴

A small number of cases involving positive re-challenge suggested a possible causal relationship.⁹⁴ COVID-19 vaccination is a powerful immune stimulant; thus, it could be hypothesized that the sensitive immune system of the endometrium is briefly modified by vaccination, potentially leading to menstrual disorders.⁹⁵ However, other studies have failed to detect an association between change in menses cycle length and COVID-19 vaccination.⁹⁶ Menstrual disorders can occur for a wide range of reasons, including some underlying medical conditions.⁹⁴ There is no evidence to suggest that menstrual changes occurring after BNT162b2 vaccination have any impact on fertility.⁹⁴

Comparison with other prophylactic vaccines

An analysis of the WHO international database, VigiBase, comparing the incidence of AEs following COVID-19 mRNA vaccination with that of approved influenza vaccines, showed that serious cardiovascular events were more prevalent with the COVID-19 vaccines, while influenza vaccines were more frequently associated with neurological AEs.⁹⁷ Overall, the risk of serious events with the COVID-19 mRNA vaccines was lower than that of the influenza vaccines. Similarly, in a study evaluating AEs in children <5 years of age in Germany, symptoms following BNT162b2 vaccination were generally comparable to those following non-SARS-CoV-2 vaccines including influenza, meningococcal, measles – mumps – rubella, tetanus – diphtheria – pertussis, hepatitis A/B, and human papillomavirus vaccines.⁹⁸

Discussion

The overall experience accrued over time on the safety and reactogenicity of BNT162b2 is extensive, robust, and comprehensive, and it has confirmed the initial safety profile observed in clinical trials, demonstrating that the vaccine is well tolerated. Most adverse reactions are transient, can be managed with medications, or can be reduced through preventive measures. The BNT162 original and variant-adapted vaccines have been well tolerated, and their safety and reactogenicity profiles have remained consistent throughout the clinical development process, which included the largest pivotal registrational COVID-19 vaccine trial conducted to date.

Occurrences of suspected AEs following vaccination should be interpreted in the context of overall numbers of doses administered and background rates of events. Serious AEs potentially related to vaccination are very rare^21,99; therefore, vaccination should be encouraged, as it provides protection against a potentially deadly illness. Although the benefit – risk profile of BNT162b2 varies by age and sex, the prevention of hospitalization, severe disease, and death outweighs the risk of AEs after vaccination in the populations for which vaccination is recommended. Vaccine-associated myocarditis/pericarditis is a rare adverse reaction identified for COVID-19 vaccines. As the risk of myocarditis/pericarditis resulting from COVID-19 disease is higher than the risk of myocarditis/pericarditis associated with vaccination,¹⁰⁰ and as the benefits of prevented COVID-19 cases and related severe outcomes outweigh the risks of myocarditis and pericarditis after receipt of mRNA COVID-19 vaccines, the overall benefit – risk profile of BNT162b2 is positive.^83,101 It has been estimated that COVID-19 vaccines saved 19.8 million lives during the first year of rollout alone.¹⁰²

Vaccine developers continue to collaborate with regulatory authorities for ongoing safety surveillance of COVID-19 vaccines as a key priority, while variant-adapted vaccines are rolled out.^{49,50,103,104} Ongoing collection of safety data for the XBB.1.5 monovalent vaccine, as well as any future variant-adapted vaccines, will continue.¹⁰⁵

In summary, the BNT162b2 vaccines have undergone thorough and extensive safety testing and monitoring, consistently demonstrating a favorable safety profile. Although no longer considered a public health emergency, COVID-19 continues to pose a threat to public health. Vaccines are critical for disease prevention and have been an essential tool in the management of the COVID-19 pandemic. Annual immunization against SARS-CoV-2 may be an option for the future. Education for healthcare providers and the general public regarding the safety of BNT162b2 and other COVID-19 vaccines are essential to ensure the success of vaccination programs.

Author contributions

All authors contributed to the manuscript conception, writing, and review process, and approved the final version for submission.

Acknowledgments

Medical writing support, including assisting authors with the development of the outline and initial draft and incorporation of comments was provided by Rachel Wright, PhD, and Helene Wellington, MS, and editorial support was provided by Ian Norton, PhD, all of Scion, London, UK, supported by BioNTech SE according to Good Publication Practice guidelines (Link).

Disclosure statement

ÖT is a management board member and employee at BioNTech SE (Mainz, Germany) and co-founder of the company. FvO, NC, FJM, CL, SP, and RR are employees at BioNTech SE. ÖT is an inventor on patents and patent applications related to RNA technology and COVID-19 vaccines. SP, RR, CL and ÖT have securities from BioNTech SE.

Additional information

Funding

The work was supported by the BioNTech .