第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.

3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

doi: 10.7499/j.issn.1008-8830.2101133

CLINICAL RESEARCH

Efficacy and safety of COVID-19 vaccines:

a systematic review
XING Kai, TU Xiao-Yan, LIU Miao, LIANG Zhang-Wu, CHEN Jiang-Nan,
LI Jiao-Jiao, JIANG Li-Guo, XING Fu-Qiang, JIANG Yi.

Department of Pediatrics, Renmin Hospital of Wuhan University, Wuhan 430060, China

Abstract: Objective To evaluate systematically the efficacy and safety of COVID-19 vaccines. Methods
PubMed, Embase, Cochrane Library, Clinicaltrial.gov, CNKI, Wanfang Data, China Biomedical Literature Service
System, and China Clinical Trial Registry were searched for randomized controlled trials of COVID-19 vaccines
published up to December 31, 2020. The Cochrane bias risk assessment tool was used to assess the quality of studies. A
qualitative analysis was performed on the results of clinical trials. Results Thirteen randomized, blinded, controlled
trials, which involved the safety and efficacy of 11 COVID-19 vaccines, were included. In 10 studies, the 28-day
seroconversion rate of subjects exceeded 80%. In two 10,000-scale clinical trials, the vaccines were effective in 95% and
70.4% of the subjects, respectively. The seroconversion rate was lower than 60% in only one study. In six studies, the
proportion of subjects who had an adverse reaction within 28 days after vaccination was lower than 30%. This proportion
was 30%-50% in two studies and >50% in the other two studies. Most of the adverse reactions were mild to moderate
and resolved within 24 hours after vaccination. The most common local adverse reaction was pain or tenderness at the
injection site, and the most common systemic adverse reaction was fatigue, fever, or bodily pain. The immune response
and incidence of adverse reactions to the vaccines were positively correlated with the dose given to the subjects. The
immune response to the vaccines was worse in the elderly than in the younger population. In 6 studies that compared
single-dose and double-dose vaccination, 4 studies showed that double-dose vaccination produced a stronger immune
response than single-dose vaccination. Conclusions Most of the COVID-19 vaccines appear to be effective and safe.
Double-dose vaccination is recommended. However, more research is needed to investigate the long-term efficacy and
safety of the vaccines and the influence of dose, age, and production process on the protective efficacy.
[Chin J Contemp Pediatr, 2021, 23(3): 221-228]
Key words: COVID-19; SARS-CoV-2; Vaccine; Systematic review; Efficacy; Safety; Clinical trial

It has been more than a year since the outbreak primary health threat for all mankind, and impacted
of the novel coronavirus pneumonia (COVID-19). the world’s political, economic and cultural greatly[2-3].
Although the spread of severe acute respiratory SARS-CoV-2 is a β-coronavirus with RNA as genetic
syndrome coronavirus 2 (SARS-CoV-2) that caused material, which enters cell through a spike protein
COVID-19 in China was effectively controlled, the combined with angiotensin converting enzyme 2[4-5].
global epidemic has not stopped. According to data COVID-19 generally manifests as fever and dry
from the World Health Organization, as of 16:05 cough, and injuries multiple organ, especially the
on February 15,2021, Central European Time, the lungs [2,5-6]. Wearing mask and maintaining social
cumulative number of confirmed COVID-19 cases distancing have been confirmed as the most effective
worldwide reached 108,579,352, and the cumulative measures to stop the spread of the virus form China’s
[1]
deaths reached 2,396,408 . The COVID-19 epidemic experience of fighting the epidemic[3,7-9], and isolation
as a major global public health event has become the and symptomatic supportive treatment still dominate

［Received］January 25, 2021; ［Accepted］February 19, 2021

［Fund project］Fundamental Research Funds for the Central Universities (2042020kf1011).
［Corresponding author］Prof. JIANG Yi. Email: [email protected].

· 221·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

for COVID-19 patients[5]. However, the efficacy of 1.2　Literature exclusion criteria

antiviral drugs and traditional Chinese medicines Documents that meet one of the following
[10-11]
needs more evidence . Due to the low penetration conditions were excluded: (1) Medical news, popular
rate of masks and the limitations of treatment options science articles, non-medical papers, reviews, letters,
[12-13]
abroad , more and more hopes are pinned on the comments, basic research, case reports, conference
development of a COVID-19 vaccine. According to abstracts; (2) No full text or literature published in
different targets and technologies, vaccines can be a third language other than Chinese and English;
divided into the following categories: inactivated (3) One of overlapping two studies were excluded;
vaccines, recombinant spike protein vaccines, viral (4) If the data of the literature included in the later
vector vaccines, RNA vaccines, live attenuated published literature, The former was excluded.
[14-16]
vaccines and virus-like particle vaccines, etc . 1.3　Literature search
Currently, hundreds of COVID-19 candidate vaccine The English databases PubMed, Embase,
projects have been registered in the US clinical trial Cochrane Library and clinicaltrials.gov databases
[15,17]
database (clinicaltrials.gov) . Results of phase 3 were searched. The Chinese databases searched
[18-22]
clinical trials of several vaccines are published . included CNKI, Wanfang Database, China Biomedical
As of January 1, 2021, China, Russia, the United Literature Service System and China Clinical
States, Britain and other countries have approved Trial Registration Center. In order to ensure the
their own mass vaccination plans for the population. comprehensiveness of the search results, this system
This study evaluated the safety and effectiveness of evaluation used Boolean logic to search by “subject
the COVID-19 vaccine through systematic literature words + free words”. The main search terms include:
review and qualitative analysis for the published COVID-19, 2019-nCoV, SARS-CoV-2, 2019 novel
COVID-19 vaccine clinical trial results. coronavirus, vaccines, vaccination, COVID-19
vaccines, mRNA-1273 vaccine, Ad5-nCoV vaccine,
1　Information and methods ChAdOx1 COVID-19 vaccine, BNT162 vaccine,
controlled clinical trial, randomized controlled trials,
This systematic review was completed in controlled clinical trial, random, blind, placebo, trial,
accordance with the guidelines in the "Preferred Meta, and etc. Chinese search terms include: 新型冠
Reporting Project for Systematic Evaluation and 状病毒、新冠肺炎、新型冠状病毒肺炎、疫苗、试验、
[23-24]
Meta-Analysis (PRISMA)" . 随机对照试验、随机对照研究、随机对照、随机、
1.1　Literature inclusion criteria 元分析、Meta、荟萃 , etc.
The literature inclusion criteria: (1) The healthy 1.4　Literature screening and data extraction
men or non-pregnant women aged 18 and above; The literature screening and data extraction were
(2) COVID-19 vaccination as the intervention measure; done independently by two researchers. Differences in
(3) The randomized, controlled, and blinded trials; the summary of the results will be discussed and dealt
(4) The clinical trial results indicators include at least with by the two researchers or the third researcher. All
one or more as following: local adverse reactions (pain, results obtained in the database were imported into
itching, redness, swelling and induration, etc.), systemic Note Express (Wuhan University Library Edition)
adverse reactions (fever, diarrhea, fatigue, nausea/ software, and duplicate documents were removed
vomiting, etc.), the last vaccine neutralizing antibody mechanically using the software's duplicate check
geometric mean titer (GMT), seroconversion rate and function. The initial screening by reading the title and
other laboratory test indicators measured by live virus abstract, and the secondary screening by reading the
neutralization test 14 days or 28 days after inoculation. full text were completed. The extracted data included:

· 222·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

the first author, vaccine type, inoculation dose, interval 2.3　The characteristics of the included studies
between inoculations, number of subjects and baseline The 13 included studies were randomized,
characteristics (race, sex ratio, age range or average blinded, and controlled trials, involving 5 inactivated
age), research design, local and systemic adverse vaccines [21-22,27-29,34] , 2 recombinant spike protein
reactions, laboratory indicators, as well as funds, vaccines[30,32], 2 RNA vaccines[20,31,33] and 2 adenovirus
sponsors and registration number. vector vaccines[19,35]. Table 2 for details of vaccine
1.5　Methodological quality evaluation characteristics and developer information). There were
Assess the risk of bias according to the Cochrane 6 studies comparing the effects of single-dose and
[25-26]
Systematic Review Manual . double-dose vaccination[19,27,30-31,33,35]. Most of the 13
1.6　Statistical analysis studies compared the difference of two doses of vaccine
The main results of this systematic review at intervals of 2, 3 or 4 weeks. Most studies also
included the safety and effectiveness of the vaccine. compared the difference between low, medium and high
Indicators for evaluating safety included local adverse injection doses. Participants in all trials were adults, and
reactions (pain, itching, redness, induration, etc.) 5 articles reported the results of vaccines in the elderly
and systemic adverse reactions (cough, diarrhea, population[19-20,32-33,35]. The baseline characteristics of the
fatigue, fever, headache, nausea/vomiting, itching, participants were shown in Table 3.
muscle pain, joint pain/discomfort, anorexia, etc.).
The immunogenicity indicators included GMT, Records identified (n=753) :
PubMed (n=294), Embase (n=130), Cochrane Library (n=69), clinicaltrials.
seroconversion rate, and the response of IgG or other gov (n=33), China Biomedical Literature Service System (n=46), CNKI (n=84),

specific antibodies to the receptor binding domain. Wanfang Data (n=97)

Records excluded for duplication (n=174)

2　Results
Records screened (n=579)

2.1　Literature search results Medical news, popular science articles,

non-medical papers, reviews, letters,
There were 753 relevant articles published before comments, basic research, case reports,

December 31, 2020. After screening, 13 papers were and conference abstracts were excluded
by reading titles and abstracts (n=547)
included in the system evaluation[19-22,27-35]. The process
Full-text articles assessed for
of document screening was shown in Figure 1.
eligibility (n=32)
2.2　Methodological quality evaluation
Excluded by reading the full text (n=19):
The 13 included studies all adopted a randomized non-RCT (n=9)
control method[19-22,27-35], with a double-blind method preprints (n=7)
duplication (n=3)
in 10 studies[21-22,27-32,34-35], and a single-blind method
in 2 studies[20,33], and bothsingle-blind method and Studies included in the analysis (n=13)

double-blind methodin one study[19]. All trials hid the

Figure 1 A flow diagram of literature screening
allocation plan. Nine trials had incomplete data or
selective reports[19,22,27,29-31,33-35], of which 2 had more
Random sequence generation (selection bias)
missing data in the preprint[22,29], and the remaining 7 Allocation concealment (selection bias)
[19,27,30-31,33-35] Blinding of participants and personnel (performance bias)
missed individual data ; 9 trials had other Blinding of outcome assessment (detection bias)
[19-20,22,29-32,34-35]
types of bias , for example, Keech et
Incomplete outcome data (attrition bias)
Selective reporting (reporting bias)

al.[30] did not perform virus neutralization test in the Other bias

0% 25% 50% 75% 100%

experimental design. In general, the included literature Low risk of bias Unclear risk of bias High risk of bias

had a low risk of bias (Figure 2 & Table 1).

Figure 2 Risk assessment of literature bias
· 223·
第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

Table 1 Methodological quality evaluation of included studies

Random Blinding of Blinding

Allocation Incomplete Selective
Studies sequence participants and of outcome Other bias
concealment outcome data reporting
generation personnel assessment
Voysey 2021[19] low risk low risk high risk low risk high risk low risk high risk
Polack 2020[20] low risk low risk high risk low risk low risk low risk high risk
Xia 2020[21] low risk low risk low risk low risk low risk low risk low risk
[22]
Pu 2020 low risk low risk low risk low risk high risk low risk high risk
Xia 2021[27] low risk low risk low risk low risk high risk low risk low risk
[28]
Che 2020 low risk low risk low risk low risk low risk low risk low risk
Ella 2020[29] low risk low risk low risk low risk high risk low risk high risk
Keech 2020[30] low risk low risk low risk unclear high risk low risk high risk
[31]
Mulligan 2020 low risk low risk low risk low risk high risk low risk high risk
Richmond 2020[32] low risk low risk low risk low risk low risk low risk high risk
[33]
Walsh 2020 low risk low risk high risk low risk high risk low risk low risk
Zhang 2021[34] low risk low risk low risk low risk high risk high risk high risk
Zhu 2020[35] low risk low risk low risk low risk low risk high risk high risk

Table 2 Experimental design and developers of the included studies

Studies Vaccines Adjuvant Research type Phase Developers Registration ID

[19]
Voysey 2021 Adenovirus recombinant No Randomized double I/II/III AstraZeneca NCT04324606,
vector vaccine (ChAdOx1 /single blind control NCT04400838,
nCoV-19/AZD1222) NCT04444674

Polack 2020[20] RNA vaccine (BNT162b2) Lipid Randomized single- II/III BioNTech / Pfizer NCT04368729
nanoparticle blind control
Xia 2020[21] Inactivated vaccine Aluminum Randomized double- I/II Wuhan Institute of ChiCTR2000031809
hydroxide blind control Biological Products
Co. Ltd
Pu 2020[22] Inactivated vaccine Aluminum Randomized double- I Institute of Medical NCT04412538
hydroxide blind control Biology, Chinese
Academy of Medical
Sciences
Xia 2021[27] Inactivated vaccine Aluminum Randomized double- I/II Beijing Institute of ChiCTR2000032459
(BBIBP-CorV) hydroxide blind control Biological Products
Che 2020[28] Inactivated vaccine Aluminum Randomized double- II Institute of Medical NCT04412538
hydroxide blind control Biology, Chinese
Academy of Medical
Sciences
Ella 2020[29] Inactivated vaccine Algel-IMDGor Randomized double- I Bharat Biotech NCT04471519
(BBV152) Algel blind control
Keech 2020[30] Recombinant spiroprotein Mareix-m1 Randomized double- II Novavax NCT04368988
nanoparticle vaccine blind control
(NVX-CoV2373)

Mulligan 2020[31] RNA vaccine (BNT162b1) Lipid Randomized double- I/II BioNTech/Pfizer NCT04368728
nanoparticle blind control
Richmond 2020[32] Recombinant spiroprotein ASO3 or CpG/ Randomized double- I Clover NCT04405908
vaccine (SCB-2019) Alum blind control Biopharmaceuticals

Walsh 2020[33] RNA vaccine (BNT162b1/ Lipid Randomized single- I BioNTech/ Pfizer NCT04368728
BNT162b2) nanoparticle blind control

Zhang 2021[34] Inactivated vaccine Aluminum Randomized double- I/II SINOVAC NCT04352608
hydroxide blind control BIOTECH CO.LTD.

Zhu 2020[35] Adenovirus type-5- No Randomized double- II Beijing Institute of NCT04341389

vectored vaccine blind control Biotechnology and
Citic Biological

· 224·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

Table 3 Baseline characteristics of the participants

Experimental
Age Male/Female Injection
Studies group/control Injected dose Country/ethnic group
(years) (n) procedure***
group (n)
Voysey 2021[19] ≥18* 55,447/54,360 55,048/54,759 2.2×1010, (3.5-6.5) ×1010 or Single injection Brazilian, South African
(5-7.5)×1010 virus particles or (0, 28) and British/White
Polack 2020[20] 52#/≥16 19,129/18,394 19,198/18,325** 30 μg (0, 21) American, Argentinian,
Brazilian, South African,
German, Turkish/White
Xia 2020[21] 41.2/18-59 120/200 240/80 2.5 μg, 5 μg or 10 μg (0, 14), (0, 28) Chinese/Asian
or (0, 56)
Pu 2020[22] 18-59* unclear 144/48 50 EU, 100 EU or 150 EU (0, 14) or (0, 28) Chinese/Asian
[27]
Xia 2021 53.7/≥18 301/339 470/170 2 μg, 4 μg or 8 μg (0, 28) Chinese/Asian
[28]
Che 2020 41.4/18-59 258/486 595/149 100 EU or 150 EU (0, 14) or (0, 28) Chinese/Asian
Ella 2020[29] 18-55* unclear 297/73 3 μg or 6 μg (0, 14) Indian/unclear
[30]
Keech 2020 30.8/18-59 63/62 102/23 5 μg or 25 μg (0, 21) Australian /White
Mulligan 2020[31] 35.4/18-55 23/22 36/9 10 μg, 30 μg or 100 μg Single injection German/ White
or (0, 21)
Richmond 2020[32] 35.7/18-75 70/78 118/30 3 μg, 9 μg or 30 μg (0, 21) Australian/White
Walsh 2020[33] 35.9/18-85 94/101 156/39 10 μg, 20 μg, 30 μg Single injection American/White
or 100 μg or (0, 21)
Zhang 2021[34] 42.6/18-59 345/389 568/166 3 μg or 6 μg (0, 14) or (0, 28) Chinese/Asian
[35] 10 11
Zhu 2020 39.7/≥18 445/445 382/508 5×10 or 1×10 v i r u s Single injection Chinese/Asian
particles
*
The study did not report a mean age; # The median age; ** Only the data of subjects without any evidence of SARS-CoV-2 infection before
vaccination were selected; *** The numbers in parentheses indicate when vaccine was injected, for example (0, 28) means that the vaccine is injected
again on the 28th day after the first injection.

2.4　Qualitative analysis mostly mild to moderate, and could be relieved within

2.4.1 The effectiveness and safety of vaccines In 24 hours after vaccination. The most common local
10 studies, the 28-day seroconversion rate of testee adverse reaction included pain or tenderness at the
exceeded 80%[21-22,27-34]. The RNA vaccine (BNT162b2) injection site[19-22,27-35]. Fatigue was reported as the
reported by Polack achieved 95% efficiency[20], the most common systemic adverse reaction in 9 stud
recombinant adenovirus vector vaccine (ChAdOx1 ies[19-20,22,28-29,31,33-35]. In addition, fever was reported
nCoV-19) reported by Voysey achieved an effective as the most common systemic adverse reaction in 2
rate of 70.4%[19], but Zhu reported that the 28-day studies[21,27], and 2 studies reported somatic pain as the
seroconversion rate of the adenovirus recombinant most common systemic adverse reaction[30,32] (Table 4).
vector vaccine in testee was less than 60%[35]. 2.4.2 Dose difference The difference in
In 6 studies, the incidence of adverse reactions injection dose is an important factor affecting the
in volunteers within 28 days for vaccination was immunogenicity and safety of the vaccine. A total of
less than 30%[20-22,27-28,34]. The adverse reaction rates 9 studies [21-22,27-29,32-35] found significant differences
of the recombinant spike protein vaccine (SCB- in GMT and seroconversion rates obtained from
2019) reported by Richmond[32] and the RNA vaccine testee with different doses of vaccination, 8 of
reported by Walsh [33] were 34.7% and 39.1%, which [20-22,28-29,31,34-35] found that GMT increased,
respectively, and the adverse reaction rates of the RNA and 4[22,28-29,32] found that the seroconversion rate of
vaccine (BNT162b1) reported by Mulligan[31] and the testee increased with the increase of vaccine dose,
adenovirus recombinant vector vaccine reported by but the incidence of adverse reactions also increases
Zhu[35] were 52.8% and 73.0%, respectively. Three relatively [22,28-29,32]. Therefore, when the clinical trial
studies could not obtain the adverse reaction rate[19,29-30]. entered Phase III, the researchers set the medium dose
The adverse reactions of all vaccinated testee were as the standard dose of the vaccine [19-20].

· 225·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

Table 4 Effectiveness and safety of vaccines

Total incidence Incidence of The most common

of adverse serious adverse adverse reactions
Studies Key effectiveness indicators
reactions reactions Local Systemic
[%(n/N)) [%(n/N)) reactions reactions
0.15
Voysey 2021[19] Efficacy 70.4%* Unclear Pressing pain fatigue
(84/55,048)
Polack 2020[20] Efficacy 95%** 27.0%# Unclear Pain Fatigue
[21]
Xia 2020 Day 14 seroconversion rates: 97.6% in the middle dose 15.0(36/240) 0(0/240) Pain Fever
group; Day 14 GMT: 121 in the standard dose group
Pu 2020[22] Day 28 seroconversion rates: 80%, 96% and 92% in the 25.7(37/144) 0(0/144) Pain Fatigue
low dose, middle dose and high dose groups respectively;
Day 28 GMT: 10.6, 15.4 and 19.6 in the low dose, middle
dose and high dose groups respectively
Xia 2021[27] Day 28 seroconversion rates: 100% each in the low dose, 29.2(42/144) 0(0/144) Pain Fever
middle dose and high dose groups; Day 28 GMT: 13.4, 18.9
and 23.7 in the low dose, middle dose and high dose groups
respectively
Che 2020[28] Day 28 seroconversion rates: 92% in the middle dose group 24.5(146/595) 0(0/595) Pain Fatigue
and 96% in the high dose group; Day 28 GMT: 19 in the
middle dose group and 21 in the high dose group
Ella 2020[29] Day 28 seroconversion rates: 87.9% in the low dose group Unclear Unclear Pain Fatigue
and 91.9% in the high dose group; Day 28 GMT: 61.7 in
the low dose group and 66.4 in the high dose group
Keech 2020[30] Day 35 GMT: 4-6 times higher than that of serum in Unclear 1.96(2/102) Pressing pain Arthralgia
convalescent period
Mulligan 2020[31] Day 28 GMT: 168 in the low dose group and 267 in the 52.8(19/36) 5.6(2/36) Pain Fatigue
middle dose group
Richmond 2020[32] Day 36 seroconversion rates: 95%, 100% and 100% in the 34.7(41/118) 1.69(2/118) Pain Headache
low dose, middle dose and high dose groups respectively
Walsh 2020[33] Day 28 GMT (BNT162b1 vaccine): 168, 167 and 267 in the 39.1(61/156) 4.49(7/156) Pain Fatigue
low dose, middle dose and high dose groups respectively;
Day 28 GMT (BNT162b2 vaccine): 157, 263 and 361 in the
low dose, middle dose and high dose groups respectively
Zhang 2021[34] Day 28 seroconversion rates: 25% in the low dose group 26.6(151/568) 1.04(1/96) Pain Fatigue
and 83% in the high dose group; Day 28 GMT: 5.4 in the
low dose group and 15.2 in the high dose group
Zhu 2020[35] Day 28 seroconversion rates: 59% in the low dose group 73.0(279/382) 6.5(25/382) Pain Fatigue
and 47% in the standard dose group; Day 28 GMT: 18.3 in
the low dose group and 19.5 in the standard dose group
GMT: geometric mean titers; * The efficacy is calculated from the corrected relative risk; ** Efficacy =100×(1-IRR), IRR is the ratio of the
number of confirmed COVID-19 cases per 1000 person-years of follow-up in the vaccine group to the corresponding cases in the placebo group;
#
The original literature only gave the incidence of adverse reactions, but did not give the specific number of people.

2.4.3 Difference of age Four studies specifically seroconversion time was later than that of the 18-59
recruited the elderly 60 years and older, and conducted years-old group. The incidence of systemic adverse
[32]
a special subgroup analysis in the results. Richmond reactions in the elderly within 7 days after vaccination
reported that the GMT range measured by the micro- was 28.6%, which was lower than 41.7% of the 18-59
neutralization test in the elderly group was 1567-3625, years-old group. Polack[20] and Walsh[33] also reported
which was lower than 2510-4452 in the 18-59-year-old similar results. In short, compared with healthy people
group. The incidence of systemic adverse reactions in aged 18 to 59, the GMT detected in the serum was
the elderly after the first injection was 17%, which was significantly lower in elderly population vaccinated
[27]
lower than 38% in the 18-59 years-old group. Xia with the same vaccine according to the same
also reported that the GMT of the elderly group was procedure, but the incidence of adverse reactions in the
lower than that of the 18-59 years-old group, and the elderly population was also significantly lower [20,27,32-33].

· 226·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

2.4.4 Differences in vaccination procedures had poor immunogenicity to elderly people over 60,
Although a number of studies designed a comparison but the adverse reaction rate was also low. One of the
of different vaccination procedures, the results of possible reasons was low immunity of the older. A lot
the experiment were complicated. Zhang 's research of studies on the tolerance of the elderly population
showed that testee who vaccinated at 2-week intervals to the vaccine still are needed. In addition, there
got a faster immune response, but a stronger immune are currently no published results of clinical trials
[34]
response at 4-week intervals . Che detected a targeting juveniles. (4) Most studies recommend
stronger immune response in testee who were double-dose vaccination, but the interval needs further
[28]
vaccinated at 2-week intervals . Xia also found that study.
the incidence of adverse reactions in testee vaccinated However, this systematic review has some
at 2-week intervals was lower than that at 4-week limitations: (1) No evidence of the long-term
[21]
intervals . In 6 studies that compared single-dose effectiveness and safety of the vaccine. Due to the
and double-dose vaccination, 4 studies showed that urgency of vaccine development, most trials only
double-dose vaccination produced a stronger immune followed up to 28 days after vaccination. Whether
[19,31,33,35]
response than single-dose vaccination . neutralizing antibodies can be maintained for a long
2.4.5 Differences of vaccine type The RNA time and whether there are delayed adverse reactions
[20]
vaccine (BNT162b2) reported by Polack and the after vaccination still require a longer period. (2) In
recombinant adenovirus vector vaccine (ChAdOx1 order to get more up-to-date evidence, this systematic
[19]
nCoV-19) reported by Voysey involved more than review also includes preprinted documents, which
10,000 people, and two both used relative risk to have not been peer reviewed and some of the data
calculate the effective rate, showing that effective rate are not available. (3) Only randomized, double-
[20] [19]
of the former was 95% , and the latter was 70.4% . blind, and controlled trials were included, while
Owing to differences in the design, the small sample observational studies, retrospective case analysis,
size, and different outcome indicators of other clinical and early animal experiments were all excluded. For
trials, their effective rates were not yet comparable. example, an open label trial conducted by Anderson[36]
found that mRNA-1273 vaccine had a good safety
3　Discussion in the elderly population. Logonov[37] reported two
adenovirus recombinant vector vaccine preparations
The system evaluation draws the following (rAd26) in a non-random clinical trial (rAd26-S
conclusions: (1) All candidate vaccines have a good and rAd5-S) had a good safety and immunogenicity
immunogenicity and safety except the vaccine in healthy people aged 18 to 60. (4) There were
[35]
reported by Zhu . Within 28 days after vaccination, differences in the design of various clinical trials,
the testee' serum GMT increased significantly, and the which made it impossible to compare the advantages
seroconversion rate was mostly greater than 80%. The and disadvantages of different types of vaccines. For
adverse reaction rate of most vaccines was less than example, Voysey[19] and Polack[20] used relative risk to
30%, degree was mild to moderate, and symptoms calculate the effective rate. Although the remaining
were alleviated within 24 hours. (2) The potency and 10 studies have completed the virus neutralization
adverse reaction rate after vaccination were positively test, the experimental design schemes were quite
[21-22,27-29,31-35]
related to the dose. Most clinical trials chose the different . (5) Only Chinese and English
middle dose when the phase III. This might be the documents were searched in this systematic review,
result of comprehensive consideration of effectiveness and documents published in other languages such as
and safety. (3) Under the same conditions, the vaccine Japanese and French were excluded.

·227·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

In conclusion, this systematic review [19] Voysey M, Clemens SAC, Madhi SA, et al. Safety and efficacy of the ChAdOx1
nCoV-19 vaccine (AZD1222) against SARS-CoV-2: an interim analysis of four
summarized the results of clinical trials related to the randomised controlled trials in Brazil, South Africa, and the UK[J]. Lancet, 2021,

COVID-19 vaccine, showing that most vaccines had 397(10269): 99-111.

[20] Polack FP, Thomas SJ, Kitchin N, et al. Safety and efficacy of the BNT162b2 mRNA
a good safety and effectiveness. It is believed that COVID-19 vaccine[J]. N Engl J Med, 2020, 383(27): 2603-2615.

with the widespread vaccination of COVID-19, it is [21] Xia SL, Duan K, Zhang YT, et al. Effect of an inactivated vaccine against SARS-
CoV-2 on safety and immunogenicity outcomes: interim analysis of 2 randomized
possible to control and end the global pandemic of clinical trials[J]. JAMA, 2020, 324(10): 951-960.

COVID-19. [22] Pu J, Yu Q, Yin ZF, et al. An in-depth investigation of the safety and
immunogenicity of an inactivated SARS-CoV-2 vaccine[J]. medRxiv, 2020. DOI:
10.1101/2020.09.27.20189548. Epub ahead of print.

Conflict of interest: The authors have no conflicts [23] Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews
and meta-analyses: the PRISMA statement[J]. PLoS Med, 2009, 6(7): e1000097.
of interest to disclose. [24] Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for reporting
systematic reviews and meta-analyses of studies that evaluate health care
interventions: explanation and elaboration[J]. Ann Intern Med, 2009, 151(4):
[References] W65-W94.
[25] Higgins JPT, Thomas J, Chandler J, et al. Cochrane handbook for systematic reviews

[1] World Health Organization. WHO coronavirus disease (COVID-19) dashboard[EB/ of interventions[M]. 2nd edition. Chichester (UK): John Wiley & Sons, 2019.

OL]. (2021-02-15)[2021-02-16]. https://covid19.who.int/. [26] Cumpston M, Li TJ, Page MJ, et al. Updated guidance for trusted systematic reviews:

[2] Sun JM, He WT, Wang LF, et al. COVID-19: epidemiology, evolution, and cross- a new edition of the Cochrane Handbook for Systematic Reviews of Interventions[J].
disciplinary perspectives[J]. Trends Mol Med, 2020, 26(5): 483-495. Cochrane Database Syst Rev, 2019, 10: ED000142.

[3] Xi JP. Remarks at the National Commendation Conference on COVID-19 (8th [27] Xia SL, Zhang YT, Wang YX, et al. Safety and immunogenicity of an inactivated

September 2020) [J]. Qiushi, 2020(20): 4-15. SARS-CoV-2 vaccine, BBIBP-CorV: a randomised, double-blind, placebo-controlled,

[4] Zhu N, Zhang DY, Wang WL, et al. A novel coronavirus from patients with phase 1/2 trial[J]. Lancet Infect Dis, 2021, 21(1): 39-51.

pneumonia in China, 2019[J]. N Engl J Med, 2020, 382(8): 727-733. [28] Che YC, Liu XQ, Pu Y, et al. Randomized, double-blinded, placebo-controlled phase

[5] National Health Commission of the People's Republic of China. The clinical 2 trial of an inactivated severe acute respiratory syndrome coronavirus 2 vaccine in

significance of COVID-19 Diagnosis and Treatment Guidelines ( Interim version 8) [J]. healthy adults[J]. Clin Infect Dis, 2020. DOI: 10.1093/cid/ciaa1703. Epub ahead of

Journal of Tropical Diseases and Parasitology, 2020, 13(5): 321-328. print.

[6] Oliveira BA, Oliveira LC, Sabino EC, et al. SARS-CoV-2 and the COVID-19 disease: [29] Ella R, Vadrevu KM, Jogdand H, et al. A phase 1: safety and immunogenicity

a mini review on diagnostic methods[J]. Rev Inst Med Trop Sao Paulo, 2020, 62: e44. trial of an inactivated SARS-CoV-2 vaccine-BBV152[J]. medRxiv, 2020, DOI:

[7] Wang J, Pan LJ, Tang S, et al. Mask use during COVID-19: a risk adjusted strategy[J]. 10.1101/2020.12.11.20210419. Epub ahead of print.

Environ Pollut, 2020, 266(Pt 1): 115099. [30] Keech C, Albert G, Cho I, et al. Phase 1-2 trial of a SARS-CoV-2 recombinant spike

[8] Li T, Liu Y, Li M, et al. Mask or no mask for COVID-19: a public health and market protein nanoparticle vaccine[J]. N Engl J Med, 2020, 383(24): 2320-2332.

study[J]. PLoS One, 2020, 15(8): e0237691. [31] Mulligan MJ, Lyke KE, Kitchin N, et al. Phase I/II study of COVID-19 RNA vaccine

[9] General Office of the National Health Commission, PRC. Technical guidelines on BNT162b1 in adults[J]. Nature, 2020, 586(7830): 589-593.

prevention and control of novel coronavirus infection in medical institutions (First [32] Richmond P, Hatchuel L, Dong M, et al. A first-in-human evaluation of the safety

edition) [J]. Chinese Journal of Infection Control, 2020, 19(2): 189-191. and immunogenicity of SCB-2019, an adjuvanted, recombinant SARS-CoV-2

[10] Cao YC, Deng QX, Dai SX. Remdesivir for severe acute respiratory syndrome trimeric S-protein subunit vaccine for COVID-19 in healthy adults; a phase 1,

coronavirus 2 causing COVID-19: an evaluation of the evidence[J]. Travel Med Infect randomised, double-blind, placebo-controlled trial[J]. medRxiv, 2020. DOI:

Dis, 2020, 35: 101647. 10.1101/2020.12.03.20243709. Epub ahead of print.

[11] Pardo J, Shukla AM, Chamarthi G, et al. The journey of Remdesivir: from Ebola to [33] Walsh EE, Frenck RW Jr, Falsey AR, et al. Safety and immunogenicity of two RNA-

COVID-19[J]. Drugs Context, 2020, 9: 2020-2024. based COVID-19 vaccine candidates[J]. N Engl J Med, 2020, 383(25): 2439-2450.

[12] Tirupathi R, Bharathidasan K, Palabindala V, et al. Comprehensive review of mask [34] Zhang YJ, Zeng G, Pan HX, et al. Safety, tolerability, and immunogenicity of an

utility and challenges during the COVID-19 pandemic[J]. Infez Med, 2020, 28(Suppl inactivated SARS-CoV-2 vaccine in healthy adults aged 18-59 years: a randomised,

1): 57-63. double-blind, placebo-controlled, phase 1/2 clinical trial[J]. Lancet Infect Dis, 2021,

[13] Provenzani A, Polidori P. COVID-19 and drug therapy, what we learned[J]. Int J Clin 21(2): 181-192.

Pharm, 2020, 42(3): 833-836. [35] Zhu FC, Guan XH, Li YH, et al. Immunogenicity and safety of a recombinant

[14] Romero JR, Bernstein HH. COVID-19 vaccines: a primer for clinicians[J]. Pediatr adenovirus type-5-vectored COVID-19 vaccine in healthy adults aged 18 years or

Ann, 2020, 49(12): e532-e536. older: a randomised, double-blind, placebo-controlled, phase 2 trial[J]. Lancet, 2020,

[15] Sharma O, Sultan AA, Ding H, et al. A review of the progress and challenges of 396(10249): 479-488.

developing a vaccine for COVID-19[J]. Front Immunol, 2020, 11: 585354. [36] Anderson EJ, Rouphael NG, Widge AT, et al. Safety and immunogenicity of SARS-

[16] Korang SK, Juul S, Nielsen EE, et al. Vaccines to prevent COVID-19: a protocol for a CoV-2 mRNA-1273 vaccine in older adults[J]. N Engl J Med, 2020, 383(25): 2427-

living systematic review with network meta-analysis including individual patient data 2438.

(The LIVING VACCINE Project)[J]. Syst Rev, 2020, 9(1): 262. [37] Logunov DY, Dolzhikova IV, Zubkova OV, et al. Safety and immunogenicity of an

[17] Wang JL, Peng Y, Xu HY, et al. The COVID-19 vaccine race: challenges and rAd26 and rAd5 vector-based heterologous prime-boost COVID-19 vaccine in two

opportunities in vaccine formulation[J]. AAPS PharmSciTech, 2020, 21(6): 225. formulations: two open, non-randomised phase 1/2 studies from Russia[J]. Lancet,

[18] Ramasamy MN, Minassian AM, Ewer KJ, et al. Safety and immunogenicity of 2020, 396(10255): 887-897.

ChAdOx1 nCoV-19 vaccine administered in a prime-boost regimen in young and old

adults (COV002): a single-blind, randomised, controlled, phase 2/3 trial[J]. Lancet, （本文编辑：邓芳明）
2021, 396(10267): 1979-1993.

· 228·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

doi: 10.7499/j.issn.1008-8830.2101133

论著·临床研究

COVID-19 疫苗的有效性和安全性的系统评价
邢凯　涂晓燕　刘苗　梁章武　陈江南　李姣姣　江利国　邢富强　姜毅
（武汉大学人民医院儿科，湖北武汉　430060）

［摘要］　目的　系统评价新型冠状病毒肺炎（COVID-19）疫苗的有效性和安全性。方法　通过计算机检
索有关 COVID-19 疫苗的临床随机对照试验文献，对临床试验结果进行定性分析。检索时间为各数据库建库至
2020 年 12 月 31 日。所检索的数据库包括 PubMed、Embase、Cochrane 图书馆、Clinicaltrial.gov、中国知网、万
方数据、中国生物医学文献服务系统和中国临床试验注册中心。使用 Cochrane 偏倚风险评估工具评估文献质量。
结果　纳入了 13 项随机、盲法、对照试验，涉及 11 种 COVID-19 疫苗接种的安全性和有效性。在其中 10 项研
究中，受试者的 28 d 血清转化率超过 80%；2 项万人规模的临床试验中，分别取得了 95% 和 70.4% 的有效率；
1 项研究的血清转化率低于 60%。在对接种后 28 d 内不良反应发生率的分析显示，6 项研究不良反应发生率低
于 30%，2 项研究为 30%~50%，2 项研究高于 50%。在 13 项研究中，疫苗接种不良反应事件绝大部分为轻度
到中度，在接种后 24 h 内缓解；最常见的局部不良反应为注射部位疼痛或压痛，最常见的系统性不良反应为疲劳、
发热或躯体痛。受试者对疫苗的免疫反应和不良反应发生率与接种剂量呈正相关。老年人对疫苗的免疫反应较
年轻人差。6 项研究比较了疫苗单剂量与双剂量接种的效应，其中 4 项研究显示双剂量接种比单剂量接种产生
更强的免疫反应。结论　大部分 COVID-19 疫苗具有较好的有效性和安全性；推荐双剂量接种。然而 COVID-19
疫苗的长期有效性、安全性及剂量、年龄和工艺差异对保护效力的影响需要更多的研究证实。
［中国当代儿科杂志，2021，23（3）：221-228］
［关键词］　新型冠状病毒肺炎；严重急性呼吸综合征冠状病毒 2；疫苗；系统评价；有效性；安全性；
临床试验

Efficacy and safety of COVID-19 vaccines: a systematic review

XING Kai, TU Xiao-Yan, LIU Miao, LIANG Zhang-Wu, CHEN Jiang-Nan, LI Jiao-Jiao, JIANG Li-Guo, XING Fu-
Qiang, JIANG Yi. Department of Pediatrics, Renmin Hospital of Wuhan University, Wuhan 430060, China (Jiang Y,
Email: [email protected])

Abstract: Objective To evaluate systematically the efficacy and safety of COVID-19 vaccines. Methods
PubMed, Embase, Cochrane Library, Clinicaltrial.gov, CNKI, Wanfang Data, China Biomedical Literature Service
System, and China Clinical Trial Registry were searched for randomized controlled trials of COVID-19 vaccines
published up to December 31, 2020. The Cochrane bias risk assessment tool was used to assess the quality of studies. A
qualitative analysis was performed on the results of clinical trials. Results Thirteen randomized, blinded, controlled
trials, which involved the safety and efficacy of 11 COVID-19 vaccines, were included. In 10 studies, the 28-day
seroconversion rate of subjects exceeded 80%. In two 10 000-scale clinical trials, the vaccines were effective in 95% and
70.4% of the subjects, respectively. The seroconversion rate was lower than 60% in only one study. In six studies, the
proportion of subjects who had an adverse reaction within 28 days after vaccination was lower than 30%. This proportion
was 30%-50% in two studies and >50% in the other two studies. Most of the adverse reactions were mild to moderate
and resolved within 24 hours after vaccination. The most common local adverse reaction was pain or tenderness at the
injection site, and the most common systemic adverse reaction was fatigue, fever, or bodily pain. The immune response
and incidence of adverse reactions to the vaccines were positively correlated with the dose given to the subjects. The
immune response to the vaccines was worse in the elderly than in the younger population. In 6 studies that compared

［收稿日期］2021-01-25；［接受日期］2021-02-19
［基金项目］中央高校基本科研业务费专项资金资助项目（2042020kf1011）。
［作者简介］邢凯，男，本科生。
［通信作者］姜毅，男，主任医师。Email：[email protected]。

· 221·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

single-dose and double-dose vaccination, 4 studies showed that double-dose vaccination produced a stronger immune
response than single-dose vaccination. Conclusions Most of the COVID-19 vaccines appear to be effective and safe.
Double-dose vaccination is recommended. However, more research is needed to investigate the long-term efficacy and
safety of the vaccines and the influence of dose, age, and production process on the protective efficacy.
[Chin J Contemp Pediatr, 2021, 23(3): 221-228]
Key words: COVID-19; SARS-CoV-2; Vaccine; Systematic review; Efficacy; Safety; Clinical trial

新 型 冠 状 病 毒 肺 炎（COVID-19） 疫 情 暴 发 1.1　文献纳入标准
至今已 1 年余。虽然 COVID-19 疫情在我国已经 文献纳入标准包括：（1）试验对象为 18 岁
得 到 了 有 效 控 制， 但 全 球 整 体 疫 情 形 势 依 然 严 及以上的健康男性或未孕女性；（2）干预措施为
峻。根据世界卫生组织的数据，截至欧洲中部时 接种 COVID-19 疫苗；（3）试验类型为随机、对
间 2021 年 2 月 15 日 16 : 05， 全 球 累 计 COVID-19 照、盲法试验；（4）临床试验结果指标至少包括
确诊病例达到 108 579 352 例，累计死亡人数达到 以下一项或几项：局部不良反应（疼痛、瘙痒、
[1]
2 396 408 人 。作为全球的重大公共卫生事件， 发红、肿胀和硬结等）、全身不良反应（发热、
COVID-19 疫情成为全人类首要的健康威胁，世界 腹泻、疲劳、恶心 / 呕吐等）、末次疫苗接种 14 d
[2-3]
政治经济文化也受到巨大冲击 。导致 COVID-19 或 28 d 后以活病毒中和试验测得的中和抗体几何
的严重急性呼吸综合征冠状病毒 2（SARS-CoV-2） 平均滴度（GMT）、血清转化率及其他实验室检
是以 RNA 为遗传 物 质 的 β 属 冠 状 病 毒， 通过 刺 测指标。
[4-5]
突蛋白结合血管紧张素转化酶２进入细胞 。 1.2　文献排除标准
COVID-19 患者的首发症状以发热和干咳多见，在 具备以下条件之一的文献被排除：（1）文献
[2,5-6]
多脏器损伤中，肺脏受损最为严重 。在疫情控 类型为医学新闻、科普文章、非医学类论文、综
制上，佩戴口罩和保持社交距离已经在中国抗击 述、信件、评论、基础研究、病例报告、会议摘要；
疫情的实践中被确认为阻断病毒传播最为有效的 （2）无法获取全文或以中文、英文外的第三种语
[3,7-9]
措施 。在对 COVID-19 患者的治疗上，隔离和 言发表的文献；
（3）若两项研究的受试者存在重叠，
[5]
对症支持治疗仍占主要地位 ，而关于抗病毒药物 则其中之一被排除；（4）若文献的数据被之后发
[10-11]
和中药等的疗效还需更多证据的支持 。由于口 表的文献包含在内，前者予以排除。
[12-13]
罩在国外普及率的低下和治疗方案的局限性 ， 1.3　文献检索
越来越多的希望被寄托在 COVID-19 疫苗的开发 对英文数据库 PubMed、Embase、Cochrane 图
上。根据靶点和技术的不同，疫苗可以被分为以 书馆和 clinicaltrials.gov 数据库进行了检索。检索
下几类：灭活疫苗、重组刺突蛋白疫苗、病毒载 的中文数据库包括中国知网、万方数据库、中国
体 疫 苗、RNA 疫 苗、 减 毒 活 疫 苗 和 病 毒 样 颗 粒 生物医学文献服务系统和中国临床试验注册中心。
[14-16]
疫苗等 。 目 前， 已 有 数 百 项 COVID-19 候 选 为了保证检索结果的全面性，本系统评价运用布尔
疫苗的项目在美国临床试验数据库（clinicaltrials. 运算逻辑，采取“主题词 + 自由词”方式进行了
[15,17]
gov）注册 ，数种疫苗的 3 期临床试验结果予 检索。主要检索词包括：COVID-19、2019-nCoV、
[18-22]
以发表 。 截 至 2021 年 1 月 1 日， 中、 俄、 SARS-CoV-2、2019 novel coronavirus、vaccines、
美、英等国家先后批准了本国疫苗在人群中的大 vaccination、COVID-19 vaccines、mRNA-1273
规模接种计划。本研究通过系统文献复习及定性 vaccine、Ad5-nCoV vaccine、ChAdOx1 COVID-19
分析已发表的 COVID-19 疫苗临床试验结果，评估 vaccine、BNT162 vaccine、controlled clinical trial、
COVID-19 疫苗的安全性与有效性。 randomized controlled trials、controlled clinical trial、
random、blind、placebo、trial、Meta 等。中文检索
1　资料与方法 词包括新型冠状病毒、新冠肺炎、新型冠状病毒
肺炎、疫苗、试验、随机对照试验、随机对照研究、
本系统评价遵循《系统评价和 Meta 分析的 随机对照、随机、元分析、Meta、荟萃等。
[23-24]
首选报告项目（PRISMA）》中的准则完成 。
· 222·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

1.4　文献筛选和资料提取 倚风险较低。见图 2 和表 1。
文献筛选和资料提取工作由两位研究者独立 2.3　纳入研究的基本特征
完成。若结果汇总时出现分歧，由两位研究者讨 所纳入的 13 项研究均为随机、盲法、对照试
论处理或交由第 3 位研究者决定。在数据库中获 验，共涉及灭活疫苗 5 种 [21-22, 27-29, 34]、重组刺突蛋
得的所有检索结果导入 NoteExpress（武汉大学图 白 疫 苗 2 种 [30,32]、RNA 疫 苗 2 种 [20,31,33] 和 腺 病 毒
书馆版）软件中，使用软件的查重功能机械地去 载体疫苗 2 种 [19,35]，疫苗特性、开发者等信息见
除重复文献。然后通过阅读标题和摘要完成初次 表 2。有 6 项研究比较了疫苗单剂量与双剂量接种
筛选，通过阅读全文完成二次筛选。在第二次筛 的效应 [19,27,30-31,33,35]。大部分研究比较了以 2 周、
选中，每一篇文献被剔除的原因均被记录。所提 3 周或 4 周为间隔注射两剂疫苗的差别。大部分
取数据包括：第一作者、疫苗类型、接种剂量、 研究也比较了低、中、高不同注射剂量的差别。
接种间隔时间、受试者人数及基线特征（种族、 所有试验的参与者均为成年人，有 5 篇文献报道
性别比例、年龄范围或平均年龄）、研究设计方案、 了疫苗在老年人群体中的结果 [19-20,32-33,35]。所纳入
局部和全身不良反应、实验室检查指标，以及基金、 研究参与者的基线特征见表 3。
赞助商和注册号等。
1.5　方法学质量评价 检索获得文献（n=753）：

依据 Cochrane 系统评价手册评估偏倚风险 [25-26]。 PubMed（n=294）、Embase（n=130）、Cochrane Library（n=69）、

clinicaltrials.gov（n=33）、中国生物医学文献服务系统（n=46）、中国知
1.6　统计学分析 网（n=84）、万方数据库（n=97）

本系统评价的主要结果包括疫苗的安全性和
有效性。评估安全性的指标包括局部不良反应（疼 重复文献（n=174）

痛、瘙痒、红肿、硬结等）及全身不良反应（咳嗽、
腹泻、疲倦、发烧、头痛、恶心 / 呕吐、瘙痒、肌 剔重后获得文献（n=579）

肉疼痛、关节痛 / 不适、厌食等）。评估免疫原性
阅读标题和摘要排除文献（n=547）：
的指标包括 GMT、血清转化率、IgG 或其他特异 医学新闻、科普文章、非医学类论
文、综述、信件、评论、基础研究、
性抗体对受体结合域的反应。
病例报告、会议摘要

2　结果 初筛获得文献（n=32）

阅读全文排除文献（n=19）：
2.1　文献检索结果 非 RCT（n=9）
已发表的预印本（n=7）
检索了截至 2020 年 12 月 31 日之前发表的所
数据重叠（n=3）
有相关文献，共得到 753 篇。经过筛选后纳入 13
篇 [19-22,27-35] 进入本系统评价。文献筛选的具体流程 纳入文献（n=13）

见图 1。
图 1　文献筛选流程图
2.2　纳入研究的方法学质量评价
纳入的 13 项研究 [19-22,27-35] 均采用了随机对照
随机分配方法
的 方 法， 其 中 10 项 [21-22,27-32,34-35] 实 施 了 双 盲 法， 分配方案隐藏
对研究对象、治疗方案实施者采用盲法
2 项 [20,33] 实施了单盲法，1 项 [19] 在不同试验地点 对研究结果测量者采用盲法
结果数据的完整性
分别使用了单盲法和双盲法；所有试验均隐藏了 选择性报告研究结果
其他偏倚风险
分配方案；9 项 [19,22,27,29-31,33-35] 数据不完整或选择性
0% 25% 50% 75% 100%

报告，其中 2 项 [22,29] 预印本缺失数据较多，其余 低风险 不清楚 高风险

7 项 [19,27,30-31,33-35] 缺失个别数据；9 项 [19-20, 22, 29-32, 34-35]

图 2　文献偏倚风险评估
存在其他类型偏倚，如 Keech 等 [30] 在试验设计中
未做病毒中和试验。总的来讲，所纳入文献的偏

· 223·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

表 1　纳入研究的方法学质量评价

对研究对象、
对研究结果测量 结果数据的 选择性报告
研究 随机分配方法 分配方案隐藏 治疗方案实施 其他偏倚风险
者采用盲法 完整性 研究结果
者采用盲法
Voysey 2021[19] 低风险 低风险 高风险 低风险 高风险 低风险 高风险
[20]
Polack 2020 低风险 低风险 高风险 低风险 低风险 低风险 高风险
[21]
Xia 2020 低风险 低风险 低风险 低风险 低风险 低风险 低风险
Pu 2020[22] 低风险 低风险 低风险 低风险 高风险 低风险 高风险
[27]
Xia 2021 低风险 低风险 低风险 低风险 高风险 低风险 低风险
Che 2020[28] 低风险 低风险 低风险 低风险 低风险 低风险 低风险
[29]
Ella 2020 低风险 低风险 低风险 低风险 高风险 低风险 高风险
Keech 2020[30] 低风险 低风险 低风险 不清楚 高风险 低风险 高风险
[31]
Mulligan 2020 低风险 低风险 低风险 低风险 高风险 低风险 高风险
Richmond 2020[32] 低风险 低风险 低风险 低风险 低风险 低风险 高风险
[33]
Walsh 2020 低风险 低风险 高风险 低风险 高风险 低风险 低风险
Zhang 2021[34] 低风险 低风险 低风险 低风险 高风险 高风险 高风险
[35]
Zhu 2020 低风险 低风险 低风险 低风险 低风险 高风险 高风险

表 2　纳入研究的试验设计和开发者

研究 疫苗 佐剂 研究类型 分期 开发者 注册号

腺病毒重组载体疫苗 NCT04324606,
随机双盲 /
Voysey 2021[19] (ChAdOx1 nCoV-19/ 无 Ⅰ/ Ⅱ/ Ⅲ AstraZeneca NCT04400838,
单盲对照
AZD1222) NCT04444674
Polack 2020[20] RNA 疫苗 (BNT162b2) 脂质纳米粒 随机单盲对照 Ⅱ/ Ⅲ BioNTech 公司和辉瑞公司 NCT04368729
Xia 2020[21] 灭活疫苗 氢氧化铝 随机双盲对照 Ⅰ/ Ⅱ 武汉生物制品研究所有限公司 ChiCTR2000031809
Pu 2020 [22]
灭活疫苗 氢氧化铝 随机双盲对照 Ⅰ 中国医学科学院医学生物研究所 NCT04412538
Xia 2021[27] 灭活疫苗 (BBIBP-CorV) 氢氧化铝 随机双盲对照 Ⅰ/ Ⅱ 北京生物制品研究所 ChiCTR2000032459
Che 2020 [28]
灭活疫苗 氢氧化铝 随机双盲对照 Ⅱ 中国医学科学院医学生物研究所 NCT04412538
Algel-IMDG
Ella 2020[29] 灭活疫苗 (BBV152) 随机双盲对照 Ⅰ Bharat 生物技术公司 NCT04471519
或 Algel
重组刺突蛋白纳米颗
Keech 2020[30] mareix-m1 随机双盲对照 Ⅱ Novavax 公司 NCT04368988
粒疫苗 (NVX-CoV2373)
Mulligan 2020[31] RNA 疫苗 (BNT162b1) 脂质纳米粒 随机双盲对照 Ⅰ/ Ⅱ BioNTech 公司和辉瑞公司 NCT04368728
重组刺突蛋白疫苗 ASO3 或
Richmond 2020[32] 随机双盲对照 Ⅰ 三叶草生物制药 NCT04405908
(SCB-2019) CpG/Alum
RNA 疫苗 (BNT162b1/
Walsh 2020[33] 脂质纳米粒 随机单盲对照 Ⅰ BioNTech 公司和辉瑞公司 NCT04368728
BNT162b2)
Zhang 2021[34] 灭活疫苗 氢氧化铝 随机双盲对照 Ⅰ/ Ⅱ 北京科兴生物制品 NCT04352608
Zhu 2020 [35]
腺病毒 5 载体疫苗 无 随机双盲对照 Ⅱ 北京生物技术研究所和中信生物 NCT04341389

· 224·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

表 3　纳入研究的基线特征

年龄 试验组 /
男/女
研究 ( 平均 / 范围 , 对照组 注射剂量 注射程序 *** 国家 / 主要人种
(例)
岁) (例)
2.2×1010、(3.5~6.5)×1010 巴西、南非、英国 /
Voysey 2021[19] ≥ 18* 55 447/54 360 55 048/54 759 单剂量或 (0, 28)
或 (5~7.5)×1010 病毒颗粒 白人
美国、阿根廷、巴西、
Polack 2020[20] 52#/ ≥ 16 19 129/18 394 19 198/18 325** 30 μg (0, 21) 南非、德国及土耳其 /
白人
(0, 14)、(0, 28)
Xia 2020[21] 41.2/18~59 120/200 240/80 2.5 μg、5 μg 或 10 μg 中国 / 亚洲人
或 (0, 56)
Pu 2020[22] 18~59* 不详 144/48 50 EU、100 EU 或 150 EU (0, 14) 或 (0, 28) 中国 / 亚洲人
[27]
Xia 2021 53.7/ ≥ 18 301/339 470/170 2 μg、4 μg 或 8 μg (0, 28) 中国 / 亚洲人
[28]
Che 2020 41.4/18~59 258/486 595/149 100 EU 或 150 EU (0, 14) 或 (0, 28) 中国 / 亚洲人
Ella 2020[29] 18~55* 不详 297/73 3 μg 或 6 μg (0, 14) 印度 / 不详
[30]
Keech 2020 30.8/18~59 63/62 102/23 5 μg 或 25 μg (0, 21) 澳大利亚 / 白人
[31]
Mulligan 2020 35.4/18~55 23/22 36/9 10 μg、30 μg 或 100 μg 单剂量或 (0, 21) 德国 / 白人
Richmond 2020[32] 35.7/18~75 70/78 118/30 3 μg、9 μg 或 30 μg (0, 21) 澳大利亚 / 白人
10 μg、20 μg、30 μg 或
Walsh 2020[33] 35.9/18~85 94/101 156/39 单剂量或 (0, 21) 美国 / 白人
100 μg
Zhang 2021[34] 42.6/18~59 345/389 568/166 3 μg 或 6 μg (0, 14) 或 (0, 28) 中国 / 亚洲人
[35] 10 11
Zhu 2020 39.7/ ≥ 18 445/445 382/508 5×10 或 1×10 病毒颗粒 单剂量 中国 / 亚洲人

注： 该研究未报道平均年龄； 中位年龄； 仅选取了接种前无任何 SARS-CoV-2 感染迹象的受试者的数据；*** 括号中的数字表示注

* # **

射疫苗的时间，如（0，28）表示在第 1 次注射后，第 28 天再次注射。

2.4　定性分析结果 反应 [19-20,22,28-29,31,33-35]。此外，发热在 2 项研究中被

2.4.1　 疫 苗 的 有 效 性 和 安 全 性　　 在 10 项 研 究 报道为最常见的系统性不良反应 [21,27]，也有 2 项研
中，受试者的 28 d 血清转化率超过 80%[21-22, 27-34]； 究报道躯体痛为最常见的系统性不良反应 [30,32]。见
在两项万人规模的临床试验中，Polack 等 [20] 报道 表 4。
的 RNA 疫苗（BNT162b2）取得了 95% 的有效率， 2.4.2　剂量差异的影响　　注射剂量的不同是影
[19]
Voysey 等 报道的腺病毒重组载体疫苗（ChAdOx1 响疫苗免疫原性和安全性的重要因素。共有 9 项
[35]
nCoV-19）取得了 70.4% 的有效率；Zhu 等 报道 研究 [21-22,27-29,32-35] 发现接受不同剂量疫苗接种的受
的腺病毒重组载体疫苗在受试者中的 28 d 血清转 试者获得的 GMT 和血清转化率存在显著性差异，
化率低于 60%。见表 4。 其中 8 项 [20-22,28-29,31,34-35] 发现 GMT 随着疫苗剂量的
在 6 项研究中，志愿者在接种疫苗后的 28 d 增加而增加，4 项 [22,28-29,32] 发现受试者血清转化率
内不良反应发生率低于 30%[20-22,27-28,34]；Richmond 随疫苗剂量的增加而增加。但随着接种剂量的加
[32]
等 报 道 的 重 组 刺 突 蛋 白 疫 苗（SCB-2019） 和 大，不良反应的发生率也相对增加 [22,28-29,32]。因此，
Walsh 等 [33] 报道的 RNA 疫苗的不良反应率分别为 当临床试验进入 Ⅲ 期阶段，研究者将中等剂量设
[31]
34.7% 和 39.1%；Mulligan 等 报 道 的 RNA 疫 苗 定为疫苗的标准剂量 [19-20]。
（BNT162b1）和 Zhu 等 [35] 报道的腺病毒重组载体 2.4.3　年龄差异的影响　　有 4 项研究专门招募了
疫苗的不良反应率分别为 52.8% 和 73.0%；3 项研 60 岁及以上的老年人群，并在结果中进行了专门
[19,29-30]
究无法获取不良反应率 。所有疫苗接种的受 的亚组分析。Richmond 等 [32] 报道使用微量中和试
试者发生不良反应事件绝大部分都是轻度到中度， 验在老年人组测得的 GMT 范围为 1 567~3 625，低
且在接种后 24 h 内可缓解；所有疫苗接种最常见 于 18~59 岁组的 2 510~4 452；而老年人在第 1 次
[19-22,27-35]
的局部不良反应均为注射部位疼痛或压痛 ； 注射后的全身不良反应发生率为 17%，低于 18~59
疲劳在 9 项研究中被报道为最常见的系统性不良 岁组的 38%。Xia 等 [27] 也报道老年人组 GMT 低于

·225·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

18~59 岁组，且达到血清转化时间晚于 18~59 岁组； 了更强的免疫反应，Xia 等 [21] 也发现以 2 周为间隔

而老年人在接种后 7 d 内的全身不良反应发生率为 接种疫苗的受试者不良反应发生率低于以 4 周为间
[20]
28.6%，低于 18~59 岁组的 41.7%。Polack 等 和 隔接种疫苗的受试者。在 6 项比较了疫苗的单剂量
[33]
Walsh 等 两项研究也报道了相似结果。总之， 与双剂量接种的研究中，4 项研究显示疫苗双剂量
相比于 18~59 岁的健康人群，老年人群按照相同 接种比单剂量接种产生更强的免疫反应 [19,31,33,35]。
的程序接种同种疫苗后，血清中所检测到的 GMT 2.4.5　疫苗类型差异的影响　　Polack 等 [20] 报道
显著偏低，但相应地老年人群中不良反应发生率 的 RNA 疫苗（BNT162b2）和 Voysey 等 [19] 报道的
也显著偏低 [20,27,32-33]。 腺 病 毒 重 组 载 体 疫 苗（ChAdOx1 nCoV-19） 受 试
2.4.4　接种程序差异的影响　　虽然多项研究设计 者人数超过 10 000 人，都采用相对危险度计算有
了不同接种程序的对比，但试验结果是复杂的。 效率，显示前者有效率为 95%[20]，后者有效率为
Zhang 等 [34] 的研究表明，以 2 周为间隔接种疫苗 70.4%[19]。其他临床试验的设计存在差异，受试者
的受试者获得了更快的免疫反应，但以 4 周为间隔 规模较小，结局指标也有所不同，其有效率尚无
接种疫苗的受试者获得了更强的免疫反应。但 Che 法比较。
[28]
等 在以 2 周为间隔接种疫苗的受试者中检测到

表 4　疫苗的有效性和安全性

总不良反应 严重不良反应 最常见的不良反应

研究 主要有效性指标 发生率 发生率
[%(n/N)] [%(n/N)] 局部 系统性
Voysey 2021[19] 有效率 70.4%* 不详 0.15(84/55 048) 压痛 疲劳
[20]
Polack 2020 有效率 95%** 27.0# 不详 疼痛 疲劳
14 d 血清转化率：标准剂量组 97.6%；
Xia 2020[21] 15.0(36/240) 0(0/240) 疼痛 发热
14 d GMT：标准剂量组 121
28 d 血清转化率：低、中、高剂量组分别为 80%、96%、92%；
Pu 2020[22] 25.7(37/144) 0(0/144) 疼痛 疲劳
28 d GMT：低、中、高剂量组分别为 10.6、15.4、19.6
28 d 血清转化率：低、中、高剂量组均为 100%；
Xia 2021[27] 29.2(42/144) 0(0/144) 疼痛 发热
28 d GMT：低、中、高剂量组分别为 13.4、18.9、23.7
28 d 血清转化率：中剂量组 92%，高剂量组 96%；
Che 2020[28] 24.5(146/595) 0(0/595) 疼痛 疲劳
28 d GMT：中剂量组 19，高剂量组 21
28 d 血清转化率：低剂量组 87.9%，高剂量组 91.9%；
Ella 2020[29] 不详 不详 疼痛 疲劳
28 d GMT：低剂量组 61.7，高剂量组 66.4
Keech 2020[30] 35 d GMT：比恢复期血清高 4~6 倍 不详 1.96(2/102) 压痛 关节痛
Mulligan 2020[31] 28 d GMT：低剂量组 168，中剂量组 267 52.8(19/36) 5.6(2/36) 疼痛 疲劳
[32]
Richmond 2020 36 d 血清转化率：低、中、高剂量组分别为 95%、100%、100% 34.7(41/118) 1.69(2/118) 疼痛 头痛
BNT162b1 疫苗 28 d GMT：低、中、高剂量组分别为 168、
Walsh 2020[33] 167、267；BNT162b2 疫苗 28 d GMT：低、中、高剂量组分 39.1(61/156) 4.49(7/156) 疼痛 疲劳
别为 157、263、361
28 d 血清转化率：低剂量组 25%，高剂量组 83%；
Zhang 2021[34] 26.6(151/568) 1.04(1/96) 疼痛 疲劳
28 d GMT：低剂量组 5.4，高剂量组 15.2
28 d 血清转化率：低剂量和标准剂量组分别为 59%、47%；
Zhu 2020[35] 73.0(279/382) 6.5(25/382) 疼痛 疲劳
28 d GMT：低剂量组 18.3，标准剂量组 19.5

注：GMT 为中和抗体几何平均滴度； * 有效率由校正后的相对危险度计算；** 有效率 =100×（1-IRR），IRR 为疫苗组每 1 000 人年

随访中确诊的 COVID-19 病例数与安慰剂组相应病例的比率；# 原文献仅给出不良反应率，未给出具体人数。

· 226·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

3　讨论 苗相关的临床试验结果，表明大部分疫苗都具有
较好的安全性和有效性。这让我们有理由相信，
本 系 统 评 价 得 出 以 下 结 论：（1） 除 了 Zhu 随着 COVID-19 疫苗的广泛接种，有望控制、终结
[35]
等 报道的疫苗外，所有候选疫苗都具有良好 COVID-19 的全球大流行。
的免疫原性和安全性。接种后 28 d 内，受试者血
清 GMT 显著增加，血清转化率大多大于 80%，大 利益冲突声明：所有作者均声明不存在利益
部分疫苗的不良反应率低于 30%，且以轻到中度 冲突。
为主，24 h 内缓解。（2）接种后产生的效价和不
良反应率与剂量呈正相关，因此，大部分临床试 ［参　考　文　献］
验进入 Ⅲ 期阶段后，选择了中等剂量作为标准剂
[1] World Health Organization. WHO coronavirus disease
量，这可能是对有效性和安全性综合考虑的结果。
(COVID-19) dashboard[EB/OL]. (2021-02-15)[2021-02-16].
（3）相同条件下，疫苗对 60 岁以上的老年人的 https://covid19.who.int/.
免疫原性较差，但不良反应率也偏低，一种可能 [2] Sun JM, He WT, Wang LF, et al. COVID-19: epidemiology,
evolution, and cross-disciplinary perspectives[J]. Trends Mol
的解释是这与人体的免疫衰老有关。老年人群对
Med, 2020, 26(5): 483-495.
疫苗的耐受性需要继续研究。此外，目前尚没有 [3] 习近平 . 在全国抗击新冠肺炎疫情表彰大会上的讲话 (2020
针对未成年人的临床试验结果发表。（4）大部分 年 9 月 8 日 )[J]. 求是 , 2020(20): 4-15.
[4] Zhu N, Zhang DY, Wang WL, et al. A novel coronavirus from
疫苗研究都推荐双剂量接种，但接种间隔时间需
patients with pneumonia in China, 2019[J]. N Engl J Med, 2020,
进一步研究。 382(8): 727-733.
然而，本系统评价有一定的局限性：（1）缺 [5] 中华人民共和国国家卫生健康委员会 . 新型冠状病毒肺炎诊
疗方案 ( 试行第八版 )[J]. 中华临床感染病杂志 , 2020, 13(5):
乏疫苗的长期有效性和安全性的证据。由于疫苗
321-328.
研 发 的 急 迫 性， 大 部 分 试 验 只 随 访 到 了 接 种 后 [6] Oliveira BA, Oliveira LC, Sabino EC, et al. SARS-CoV-2 and
28 d，中和性抗体能否长期维持，接种疫苗后是 the COVID-19 disease: a mini review on diagnostic methods[J].
Rev Inst Med Trop Sao Paulo, 2020, 62: e44.
否有迟发的不良反应，仍需要更长时间的随访。
[7] Wang J, Pan LJ, Tang S, et al. Mask use during COVID-19:
（2）为了纳入更多最新证据，本系统评价也将预 a risk adjusted strategy[J]. Environ Pollut, 2020, 266(Pt 1):
印本文献包含在内，这些文献没有经过同行评议， 115099.
[8] Li T, Liu Y, Li M, et al. Mask or no mask for COVID-19: a
且其中一些数据无法获取。（3）本系统评价只纳
public health and market study[J]. PLoS One, 2020, 15(8):
入了随机、双盲、对照试验，而观察性研究、回 e0237691.
顾性病例分析及早期的动物试验均被排除在外。 [9] 中华人民共和国国家卫生健康委员会办公厅 . 医疗机构内新

如 Anderson 等 [36]
实施的一项开放标签试验发现 型冠状病毒感染预防与控制技术指南 ( 第一版 )[J]. 中国感染
控制杂志 , 2020, 19(2): 189-191.
mRNA-1273 疫苗在老年人群体具有较好的安全性， [10] Cao YC, Deng QX, Dai SX. Remdesivir for severe acute
Logunov 等 [37] 在非随机临床试验中报道了两种腺 respiratory syndrome coronavirus 2 causing COVID-19: an
病 毒 重 组 载 体 疫 苗 制 剂（rAd26-S 和 rAd5-S） 在 evaluation of the evidence[J]. Travel Med Infect Dis, 2020, 35:
101647.
18~60 岁健康人群具有较好的安全性和免疫原性。 [11] Pardo J, Shukla AM, Chamarthi G, et al. The journey of
（4）各项临床试验的设计存在差异，导致无法对 Remdesivir: from Ebola to COVID-19[J]. Drugs Context, 2020,
不同类型疫苗的优劣进行比较，如 Voysey 等 [19]
和 9: 2020-2024.
[20] [12] Ti r u p a t h i R , B h a r a t h i d a s a n K , P a l a b i n d a l a V, e t a l .
Polack 等 采用相对危险度计算有效率，Keech
Comprehensive review of mask utility and challenges during the
[30]
等 未做病毒中和试验，其余 10 项研究虽然均 COVID-19 pandemic[J]. Infez Med, 2020, 28(Suppl 1): 57-63.
完 成 了 病 毒 中 和 试 验， 但 试 验 设 计 方 案 差 异 较 [13] Provenzani A, Polidori P. COVID-19 and drug therapy, what we
learned[J]. Int J Clin Pharm, 2020, 42(3): 833-836.
大 [21-22,27-29,31-35]。（5）本系统评价只检索了中英文
[14] Romero JR, Bernstein HH. COVID-19 vaccines: a primer for
文献，以日文、法文等其他语言发表的文献被排 clinicians[J]. Pediatr Ann, 2020, 49(12): e532-e536.
除在外。 [15] Sharma O, Sultan AA, Ding H, et al. A review of the progress
and challenges of developing a vaccine for COVID-19[J]. Front
综上所述，本系统评价总结了 COVID-19 疫
Immunol, 2020, 11: 585354.

· 227·
 第 23 卷 第 3 期 中国当代儿科杂志 Vol.23 No.3
2021 年 3 月 Chin J Contemp Pediatr Mar. 2021

[16] Korang SK, Juul S, Nielsen EE, et al. Vaccines to prevent Lancet Infect Dis, 2021, 21(1): 39-51.
COVID-19: a protocol for a living systematic review with [28] Che YC, Liu XQ, Pu Y, et al. Randomized, double-blinded,
network meta-analysis including individual patient data (The placebo-controlled phase 2 trial of an inactivated severe acute
LIVING VACCINE Project)[J]. Syst Rev, 2020, 9(1): 262. respiratory syndrome coronavirus 2 vaccine in healthy adults[J].
[17] Wang JL, Peng Y, Xu HY, et al. The COVID-19 vaccine race: Clin Infect Dis, 2020. DOI: 10.1093/cid/ciaa1703. Epub ahead
challenges and opportunities in vaccine formulation[J]. AAPS of print.
PharmSciTech, 2020, 21(6): 225. [29] Ella R, Vadrevu KM, Jogdand H, et al. A phase 1: safety
[18] Ramasamy MN, Minassian AM, Ewer KJ, et al. Safety and and immunogenicity trial of an inactivated SARS-
immunogenicity of ChAdOx1 nCoV-19 vaccine administered C o V- 2 v a c c i n e - B B V 1 5 2 [ J ] . m e d R x i v, 2 0 2 0 . D O I :
in a prime-boost regimen in young and old adults (COV002): a 10.1101/2020.12.11.20210419. Epub ahead of print.
single-blind, randomised, controlled, phase 2/3 trial[J]. Lancet, [30] Keech C, Albert G, Cho I, et al. Phase 1-2 trial of a SARS-
2021, 396(10267): 1979-1993. CoV-2 recombinant spike protein nanoparticle vaccine[J]. N
[19] Voysey M, Clemens SAC, Madhi SA, et al. Safety and efficacy Engl J Med, 2020, 383(24): 2320-2332.
of the ChAdOx1 nCoV-19 vaccine (AZD1222) against SARS- [31] Mulligan MJ, Lyke KE, Kitchin N, et al. Phase I/II study of
CoV-2: an interim analysis of four randomised controlled COVID-19 RNA vaccine BNT162b1 in adults[J]. Nature, 2020,
trials in Brazil, South Africa, and the UK[J]. Lancet, 2021, 586(7830): 589-593.
397(10269): 99-111. [32] Richmond P, Hatchuel L, Dong M, et al. A first-in-human
[20] Polack FP, Thomas SJ, Kitchin N, et al. Safety and efficacy of evaluation of the safety and immunogenicity of SCB-2019,
the BNT162b2 mRNA COVID-19 vaccine[J]. N Engl J Med, an adjuvanted, recombinant SARS-CoV-2 trimeric S-protein
2020, 383(27): 2603-2615. subunit vaccine for COVID-19 in healthy adults; a phase 1,
[21] Xia SL, Duan K, Zhang YT, et al. Effect of an inactivated randomised, double-blind, placebo-controlled trial[J]. medRxiv,
vaccine against SARS-CoV-2 on safety and immunogenicity 2020. DOI: 10.1101/2020.12.03.20243709. Epub ahead of print.
outcomes: interim analysis of 2 randomized clinical trials[J]. [33] Walsh EE, Frenck RW Jr, Falsey AR, et al. Safety and
JAMA, 2020, 324(10): 951-960. immunogenicity of two RNA-based COVID-19 vaccine
[22] Pu J, Yu Q, Yin ZF, et al. An in-depth investigation of the safety candidates[J]. N Engl J Med, 2020, 383(25): 2439-2450.
and immunogenicity of an inactivated SARS-CoV-2 vaccine[J]. [34] Zhang YJ, Zeng G, Pan HX, et al. Safety, tolerability, and
medRxiv, 2020. DOI: 10.1101/2020.09.27.20189548. Epub immunogenicity of an inactivated SARS-CoV-2 vaccine in
ahead of print. healthy adults aged 18-59 years: a randomised, double-blind,
[23] Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting placebo-controlled, phase 1/2 clinical trial[J]. Lancet Infect Dis,
items for systematic reviews and meta-analyses: the PRISMA 2021, 21(2): 181-192.
statement[J]. PLoS Med, 2009, 6(7): e1000097. [35] Zhu FC, Guan XH, Li YH, et al. Immunogenicity and safety of
[24] Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement a recombinant adenovirus type-5-vectored COVID-19 vaccine
for reporting systematic reviews and meta-analyses of studies in healthy adults aged 18 years or older: a randomised, double-
that evaluate health care interventions: explanation and blind, placebo-controlled, phase 2 trial[J]. Lancet, 2020,
elaboration[J]. Ann Intern Med, 2009, 151(4): W65-W94. 396(10249): 479-488.
[25] Higgins JPT, Thomas J, Chandler J, et al. Cochrane Handbook [36] Anderson EJ, Rouphael NG, Widge AT, et al. Safety and
for Systematic Reviews of Interventions[M]. 2nd ed. Chichester, immunogenicity of SARS-CoV-2 mRNA-1273 vaccine in older
UK: John Wiley & Sons, 2019. adults[J]. N Engl J Med, 2020, 383(25): 2427-2438.
[26] Cumpston M, Li TJ, Page MJ, et al. Updated guidance for [37] Logunov DY, Dolzhikova IV, Zubkova OV, et al. Safety
trusted systematic reviews: a new edition of the Cochrane and immunogenicity of an rAd26 and rAd5 vector-based
Handbook for Systematic Reviews of Interventions[J]. Cochrane heterologous prime-boost COVID-19 vaccine in two
Database Syst Rev, 2019, 10: ED000142. formulations: two open, non-randomised phase 1/2 studies from
[27] Xia SL, Zhang YT, Wang YX, et al. Safety and immunogenicity Russia[J]. Lancet, 2020, 396(10255): 887-897.
of an inactivated SARS-CoV-2 vaccine, BBIBP-CorV: a
randomised, double-blind, placebo-controlled, phase 1/2 trial[J]. （本文编辑：邓芳明）

· 228·