Document text
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 1NON- INTERVENTIONAL (NI) STUDY PROTOCOL
PASS information
Title Post Conditional Approval Active
Surveillance Study Among Individuals in
Europe Receiving the Pfizer- BioNTech
Coronavirus Disease 2019 ( COVID -19)
Vaccine
Protocol number C4591021
Protocol version identifier Version 2 (20 May 2021)
Date 20May 2021
EU Post Authoriz ation Study (PAS)
register numberTo be registered before the start of data
collection
Active substance BNT162b2
Medicinal product COVID -19 messenger ribonucleic
acid (mRNA) vaccine is a nucleoside -
modified ribonucleic
acid (modRNA) encoding the viral spike
glycoprotein S of severe acute respiratory
syndrome coronavirus 2 (SARS -CoV -2)
Marketing Authoriz ation Holder (s)
(MAH)BioNTech Manufacturing GmbH
Joint PASS No
Research question and objectives Theresearch question addressed bythis
study is:Is there an increased risk ofselect
adverse events of special interest ( AESI )
after being vaccinated with the Pfizer -
BioNTech COVID -19vaccine?
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950230
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 2Objectives
Primary study objective
To determine whether an increased risk of
prespecified AESI exists following the
administration of a t least one dose the
Pfizer -BioNTech COVID- 19 vaccine using
two approaches: (a) a cohort design
comparing risk in vaccinated and non-
vaccinated individuals and(b)a self -
controlled risk interval (SCRI) design.
Secondary study objectives
To estimate the incidence rates of
prespecified AESI among individuals
who receive at least one dose of the
Pfizer -BioNTech COVID- 19 vaccine
using a cohort study design.
To describe the incidence rates and
determine whether an increased risk of
prespecified AESI exists following the
administration of a t least one dose the
Pfizer -BioNTech COVID- 19 vaccine
compared with a matched comparator
group with no COVID -19 vaccination
within subcohorts of interest
(i.e., individuals who are
immunocompromised, individuals who
are frail and have comorbidities,
individuals diagnosed with previous
COVID -19 infection, and age -specific
groups) in Europe using a cohort study
design and/or a SCRI design.
To determine whether an increased risk
of prespecified AESI exist sfollowing the
administration of at least one dose of the
Pfizer-BioNTech COVID- 19 vaccine
compared with no COVID -19
vaccination, in pregnant people and their
neonates using a cohort study design.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950231
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 3To characterise utilisation patterns of
Pfizer -BioNTech COVID- 19 vaccine
among individuals within Europe ,
including est imating the proportion of
individuals receiving thevaccine; two-
dose vaccine completion rate and
distribution of time gaps between the
first and second doses; and
demographics and clinical characteristics
of recipients, overall and among
subcohorts of int erest , such as
individuals who are
immunocompromised, elderly , or have
specific comorbidities.
Country( -ies) of study The Netherlands (NL) , Italy (IT), Spain
(ES), United Kingdom (UK) of Great
Britain, Norway (NO)
Author Alejandro Arana
Senior Director Epidemiology
RTI Health Solutions, in collaboration with
University Medical Center Utrecht on behalf
of the Vaccine monitoring Collaboration for
Europe (VAC4EU )Consortium research
team
Av. Diagonal, 605, 9 -1,
08028 Barcelona
SPAI N
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950232
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 4Marketing Authoriz ation Holder(s)
Marketing Authoriz ation Holder(s) BioNTech Manufacturing GmbH
An der Goldgrube 12
55131 Mainz
German y
MAH contact person Constanze Blume
This document contains confidential information belonging to Pfizer. Except as otherw ise agreed to in writing,
by accepting or reviewing this document, you agree to hold this information in confidence and not copy or
disclose it to others (except where required by applicable law) or use it for unauthorized purposes. In the event
of any actual or suspec ted breach of this obligation, Pfizer must be promptly notified.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950233
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 51.TABLE OF CONTENTS
1. TABLE OF CONTENTS ................................ ................................ ................................ .......5
2. LIST OF ABBREVIAT IONS ................................ ................................ ................................ 9
3. RESPONSIBLE PARTI ES................................ ................................ ................................ ..12
4. ABSTRACT ................................ ................................ ................................ ......................... 14
5. AMENDMENTS AND UP DATES ................................ ................................ ..................... 17
6. MILESTONES ................................ ................................ ................................ ..................... 18
7. RATIONALE AND BAC KGROUND ................................ ................................ ................ 18
8. RESEARCH QUESTION AND OBJECTI VES ................................ ................................ .19
8.1. Objectives ................................ ................................ ................................ ................ 19
9. RESEARCH METHODS ................................ ................................ ................................ ....20
9.1. Study design ................................ ................................ ................................ ............ 20
9.1.1. Retrospective cohort design ................................ ................................ ........ 20
9.1.1.1. Matching process ................................ ................................ .......21
9.1.2. Self -controlled risk interval design................................ ............................. 22
9.2. Setting ................................ ................................ ................................ ...................... 24
9.2.1. I nclusion criteria ................................ ................................ ......................... 24
9.2.1.1. Cohort design ................................ ................................ ............ 24
9.2.1.2. Self -controlled risk interval design ................................ ........... 25
9.2.2. Exclusion criteria ................................ ................................ ........................ 25
9.2.2.1. Cohort design ................................ ................................ ............ 25
9.2.3. Source population ................................ ................................ ....................... 25
9.2.4. Study period ................................ ................................ ................................ 25
9.3. Variables ................................ ................................ ................................ .................. 26
9.3.1. Exposure definition, by data source ................................ ............................ 26
9.3.1.1. Cohort design ................................ ................................ ............ 27
9.3.1.2. Self -controlled risk interval design ................................ ........... 28
9.3.2. Outcomes definition ................................ ................................ .................... 28
9.3.2. 1. Safet y outcomes ................................ ................................ ........ 28
9.3.3. Covariate definition ................................ ................................ .................... 32
9.4. Data sources ................................ ................................ ................................ ............ 36
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950234
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 69.4.1. PHARMO (NL) (6 million active individuals)................................ ........... 37
9.4.1.1. Vaccine exposure ................................ ................................ ......38
9.4.2. ARS Toscana database (IT) (3.6 million active individuals) ..................... 39
9.4.2.1. Vaccine exposure ................................ ................................ ......39
9.4.3. Pedianet/Health Search Database (IT) (1 million active individuals) ........ 39
9.4.3.1. Vaccine Exposure................................ ................................ ......41
9.4.4. E piChron – Aragon data sources (ES) (1.3 million active individuals) ..... 41
9.4.4.1. Vaccine exposure ................................ ................................ ......42
9.4.5. Clinical Practice Research Datalink and Hospital Episode Statistics
(UK) (16 million active individuals) ................................ ................................ 42
9.4.5.1. Vaccine exposure ................................ ................................ ......44
9.4.6. Norwegian health registers (NO) (5.3million active individuals) ............. 44
9.4.6.1. Norwegian Immunisation Registry ................................ ........... 44
9.4.6.2. The Norwegian Patient Registry ................................ ............... 44
9.4.6.3. Norway Control and Pay ment of Health Reimbursement ......... 45
9.4.6.4. The Norwegian Prescription Database................................ ......45
9.4.6.5. The Medical Birth Registry of Norway ................................ .....45
9.4.6.6. Statistics Norway ................................ ................................ .......45
9.4.6.7. The National Registry ................................ ............................... 45
9.4.6.8. Norwegian Surveillance S ystem for Communicable
Diseases ................................ ................................ ............................. 45
9.4.6.9. Vaccine Exposure................................ ................................ ......46
9.4.7. SIDIAP (ES) (5.7 million active individuals)................................ ............. 46
9.4.8. Cohort design ................................ ................................ .............................. 52
9.4.8.1. Exposure assignment and follow-up ................................ ......... 52
9.4.8.2. Descriptive statistics ................................ ................................ ..53
9.4.8.3. Description of vaccination categories ................................ .......54
9.4.8.4. Crude outcome measures ................................ .......................... 54
9.4.8.5. Adjustment for baseline imbalances ................................ ......... 55
9.4.8.6. Adjustment for adherence to recommended vaccination
schedule ................................ ................................ ............................. 55
9.4.8.7. Meta- analy sis................................ ................................ ............ 55
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950235
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 79.4.8.8. Vaccine exposure ................................ ................................ ......47
9.5. Stu dy size ................................ ................................ ................................ ................ 47
9.6.Data management ................................ ................................ ................................ ....48
9.6.1. Case report f orms (CRFs)/Data collection t ools (DCTs)/Electronic
data r ecord ................................ ................................ ................................ ........ 49
9.6.2. Record retention ................................ ................................ .......................... 50
9.6.3. Dat a extraction ................................ ................................ ............................ 51
9.6.4. Data processing and transformation ................................ ........................... 51
9.6.5. Data access ................................ ................................ ................................ ..51
9.7. Data analy sis................................ ................................ ................................ ........... 52
9.7.1. Self -controlled risk interval ................................ ................................ ........ 55
9.7.1.1. Descriptive statistics ................................ ................................ ..55
9.7.1.2. Measures of association ................................ ............................ 55
9.8. Quality control ................................ ................................ ................................ ......... 56
9.8.1. PHARMO (NL) ................................ ................................ .......................... 56
9.8.2. ARS Toscana (IT) ................................ ................................ ....................... 56
9.8.3. Pedianet/HSD (IT) ................................ ................................ ...................... 57
9.8.4. EpiChron - Aragon data sources (ES) ................................ ........................ 57
9.8.5. CPRD (UK) ................................ ................................ ................................ 58
9.8.6. SI DIAP (ES) ................................ ................................ ............................... 58
9.9. L imitations of the research methods ................................ ................................ .......58
9.10. Other aspects ................................ ................................ ................................ ......... 60
10. PROTECTI ON OF HU MAN SUBJECTS ................................ ................................ ........ 60
10.1. Patient i nformation ................................ ................................ ................................ 60
10.2. Patient c onsent ................................ ................................ ................................ .......61
10.3. Institutional review board (IRB)/Independent ethics c ommittee (IEC) ................ 61
10.4. Ethical conduct of the s tudy ................................ ................................ .................. 61
11. MANAGEMENT AND R EPORTI NG OF ADVERSE EVENTS/ADVERSE
REACTI ONS ................................ ................................ ................................ ...................... 61
12.PLANS FOR DI SSEMI NATING AND COMMUNI CATI NG STUDY RESUL TS........ 63
13. REFERENCES ................................ ................................ ................................ .................. 64
14. LIST OF TABLES ................................ ................................ ................................ ............. 68
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950236
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 815.LIST OF FIGURES ................................ ................................ ................................ ........... 68
ANNEX 1. LIST OF STAND ALONE DOCUMENTS .........................................................68
ANNEX 2. ENCEPP CHEC KLIST FOR STUDY PROT OCOL S ................................ ......... 68
ANNEX 3. ADDITIONAL INFORMATION ................................ ................................ ......... 73
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950237
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 92. LIST OF ABBREVIATIONS
Abbreviation Definition
ACCESS project vACcine Covid -19monitoring readinESS
AESI adverse events of special interest
ARS Toscana Agenzia Regionale di Sanita’ della Toscana (a
research institute of the Tuscan y region of Ital y)
ATC Anatomical Therapeutic Chemical (classification
system)
BDU user database a t EpiChron
BIFAP Base de Datos para la Investigación
Farmacoepidemiológica en Atención Prim ària(a data
resource for pharmacoepidemiology in Spain)
CDM common data model
CHESS COVID -19 Hospitalisation in England Surveillance
System (UK)
CI confidence interval
COVID -19 coronavirus disease 2019
CPRD Clinical Practice Research Datalink
DAP database access provider
DRE Digital Research Environment (NL)
DSRU Drug Safet y Research Unit (UK)
DTP diphtheria, tetanus, and pertussis vaccine
EMA European Medicines Agency
ENCePP European Network of Centres for
Pharmacoepidemiology and Pharmacovigilance
EpiChron EpiChron Research Group on Chronic Diseases at the
Aragon Health Sciences Institute (Spain)
ES Spain
ETL extract ion, transform ation , and loading (a process for
putting data into a common data model)
EU PAS Register European Union electronic register of post-
authorisation studies
EU European Union
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950238
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 10GOLD General Practitioner Online Database (of the CPRD)
GP general practition er
GPP Good Pharmacoepidemiology Practices
GVP Good Pharmacovigilance Practices
HES Hospital Episode Statistics
HSD Health Search Database (Ital y)
ICD International Classification of Diseases
ICD-9-CM International Classification of Diseases, 9th Revision,
Clinical Modification
ICD-10 International Classification of Diseases, 10th Revision
ICPC International Classification of Primary Care
ISPE International Societ y for Pharmacoepidemiology
IT Italy
KUHR Norway Control and Payment of Health
Reimbursement
MAH marketing authorisation holder
MBRN Medical Birth Registry of Norway
mRNA messenger RNA
MSIS Norwegian Surveillance Sy stem for Communicable
Diseases
NHS National Health Service (UK)
NIPH Norwegian Institute of Health
NL Netherlands
NO Norway
NPR National Patient Register (Norway )
ONS Office for National Statistics
PASS post-authorisation safet y study
PHARMO PHARMO I nstitute for Drug Outcomes Research or
PHARMO Database Network (Netherlands)
PHE Public Health England
QC quality control
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950239
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 11RTI-HS RTI Health Solutions
SAP statistical analy sis plan
SARS -CoV -2 severe acute respiratory syndrome coronavirus 2
(cause of COVID -19 disease)
SCRI self-controlled risk interval (study design)
SIDIAP Sistema d’I nformació per el Desenvolupament de la
Investigació en Atenció Primària [I nformation System
for the Improvement of Research in Primary Care]
(Spain)
SQL Structured Query Language
SSB Statistics Norway
SYSVAK national, electronic immunisation register
UK United Kingdom
UMCU University Medical Center Utrecht
USA United States of America
VAC4EU Vaccine monitoring Collaboration for Europe
VAED vaccine -associated enhanced disease
VV Varicella zoster virus
WHO World Health Organization
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950240
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 123.RESPONSIBLE PARTIES
Principal Investigator (s)of the Protocol
Nam e, degree(s) Job Title Affiliation Address
Heather Rubino Director, Epidemiology Pfizer 235 E 42nd St.,
New York, NY 10017
Daniel Weibel Assistant Professor University Medical
Center UtrechtInternal mail no Str
6.131 | PO Box 85500 |
3508 GA UTRECHT
Alejandro Arana Senior Director,
EpidemiologyRTI Health Solutions Av. Diagonal, 605, 9 -1,
08028 Barcelona, Spain
Miriam Sturkenboom Professor University Medical
Center UtrechtInternal mail no Str
6.131 | PO Box 85500 |
3508 GA UTRECHT
Xabier Garcia de
AlbenizDirector, Epidemiology RTI Health Solutions Av. Diagonal, 605, 9 -1,
08028 Barcelona, Spain
Estel Plana Director, Biostatistics RTI Health Solutions Av. Diagonal, 605, 9 -1,
08028 Barcelona, Spain
Alison Kawai Senior Research
EpidemiologistRTI Health Solutions 307 Waverley Oaks
Road, Suite 101,
Waltham, MA
02452 -8413 USA
Bradley Layton Senior Research
EpidemiologistRTI Health Solut ions 3040 East Cornwallis
Rd, PO Box 12194
Research Triangle Park,
NC 27709 -2194 USA
Rachel Weinrib Research Epidemiologist RTI Health Solutions Av. Diagonal, 605, 9 -1,
08028 Barcelona, Spain
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950241
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 13Country Coordinating Investigators
Nam e, degree(s) Job Title Affiliation Address
Rosa Gini Head
Pharm acoepidemiology
UnitAgenzia regionale di
sanità della ToscanaVia Pietro Dazzi 1,
50141 Firenze, Italy
Ron Herings Director PHARMO Institute for
Drug Outcomes
ResearchVan Deventerlaan 30 -40
3528 AE Utrecht
The Netherlands
Carlo Giaquinto President PENTA foundation Corso Stati Uniti 4
35127 Padova
Italy
Saad Shakir Director Drug Safety Research
Unit (DSRU )Drug Safety Research
Unit
Bursledon Hall
Blundell Lane
Southampton,
Ham pshire
SO31 1AA
United Kingdom
Alexandra Prados Torres National Health Service
(NHS )Senior
ResearcherEpiChron Research
Group. Instituto
Aragonés de Ciencias de
la SaludHospital Universitario
Miguel Servet
Paseo Isabel la Católica
1-3
50009, Zaragoza, Spain Antonio Gimeno Miguel Researcher
Beatriz Poblador -Plou Researcher
Boni Bolibar Scientífic DirectorIDIAP -Jordi Gol Gran Via Corts
Catalanes 587
08007, Barcelona, Spain Felipe Villalobos Researcher
Angela Lupattelli Researcher University of Oslo PO 1068
Blindern
0316 Oslo, Norw ay
Note: Country Coordinating investigators have reviewed and contributed to this protocol.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950242
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 144.ABSTRACT
Title :Post Conditional Approval Active Surveillance Study Among Individuals in Europe
Receiving the Pfizer- BioNTech Coronavirus Disease 2019 ( COVID -19) Vaccine; Version 2
(20May 2021); Main author. Alejandro Arana, Senior Director Epidemiology , RTI Health
Solutions, in collaboration with University Medical Center Utrecht on behalf of VAC4EU
Consortium research team
Rationale and background :The novel coronavirus SARS -CoV -2, the cause of COVID -19,
has resulted in a global pandemic. The Pfizer -BioNTech COVID -19vaccine, tozinameran
(Comirnaty®) a novel mRNA -based vaccine, has been authorised for use in the European
Union (EU) , for the prevention of COVID -19. Efficient and timely monitoring of the safet y
of the vaccine is needed in European countries.
Research question and objectives:
Is there an increased risk ofselect adverse events of special interest (AESI ) after being
vaccinated with the Pfizer-BioNTech COVID- 19 vaccine?
Objectives
Primary study objective
To determine whether an increased risk of prespecified AESI exists following the
administration of a t least one dose the Pfizer -BioNTech COVID -19 vaccine using t wo
approaches: (a) a cohort design comparing risk in vaccinated andnon-vaccinated individuals
and(b)a self -controlled risk interval (SCRI ) design.
Secondary study objectives
To estimate the incidence rates of prespecified AESI among individuals who rece ive at
least one dose of the Pfizer- BioNTech COVID -19 vaccine using a cohort study design.
To describe the incidence rates and determine whether an increased risk of prespecified
AESI exists following the administration of a t least one dose the Pfizer -BioN Tech
COVID -19 vaccine compared with a matched comparator group with no COVID -19
vaccination within subcohorts of interest (i.e., individuals who are immunocompromised,
individuals who are frail and have comorbidities, individuals diagnosed with previous
COVID -19 infection, and age -specific groups) in Europe using a cohort study design
and/or a SCRI design.
To determine whether an increased risk of prespecified AESI exists following the
administration of at least one dose of the Pfizer-BioNTech COVID- 19 vaccine compared
with no COVI D-19 vaccination, in pregnant people and their neonates using a cohort
study design.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950243
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 15To characterise utilisation patterns of Pfizer- BioNTech COVID -19 vaccine among
individuals within Europe ,including estimating the proportion of individuals receiving
thevaccine; two -dose vaccine completion rate and distribution of time gaps between the
first and second doses; and demographics and clinical characteristics of recipients, overall
and among subcoho rts of interest , such as individuals who are immunocompromised,
elderl y, or have specific comorbidities.
Study design :A retrospective cohort design will be used to estimate the incidence of AESI
after receiving vaccine doses and compare this incidence wit h that occurring in an
unvaccinated comparator group matched on relevant individual characteristics ( e.g., age,
comorbidities). Where appropriate, the stud y will also use a SCRI design.
Population :The source population will comprise all individuals registered in each of the
health care data sources who are eligible to receive the Pfizer -BioNTech COVID -19vaccine.
The study period will start on the date of launch of the Pfizer -BioNTech COVID -19vaccine
and will end on the date of the latest data availability or 31 Dec 2023 . It is expected that
follow -up will last for 2 years for AESI. People who are pregnant at time of vaccination or
who become pregnant within two years of study start and their live born infants will be
followed for an additional 12 months to collect information about birth outcomes and linked
infant outcomes.
Variables :Exposure will be based on recorded prescription, dispensing, or administration of
the Pfizer -BioNTech COVID -19vaccine. Vaccine administration and date of vaccination
should be obtained from all possible sources that capture COVID -19vaccination. The
outcomes will be based on the AESI proposed by the European Medicines Agency
(EMA)-sponsored ACCESS project (vACcine COV ID-19monitoring readinESS). A ESIwill
be identified based on patient profile review of electronic records b y health care
professionals. In addition, manual review of patient charts conducted b y clinicians blinded to
COVID -19 vaccine exposure will be performed. Confirmation of an event diagnosis will be
classified against existing definitions of the Brighton Collaboration and those currently being
developed. Key covariates include demographics; COVID -19history , as available in each
data source (will be used to define a subgroup of interest); personal lifestyle characteristics;
comorbidities; immunocompromising conditions (will be used to define subgroups for
secondary anal yses); comedication use during the y ear before time zer o (prescriptions or
dispensing, no over -the-counter medication use); health care utilisation descriptors; other
vaccinations; and surrogates of frailt y.
Data sources :The stud y will be performed within the following selected data sources:
PHARMO (PHARMO I nstitute for Drug Outcomes Research) (NL), ARS Toscana (Agenzia
Regionale di Sanita’ della Toscana) (IT), Pedianet/Health Search Database (HSD) (IT),
EpiChron (EpiChron Research Group on Chronic Diseases at the Aragon Health Sciences
Institute) (ES), CPRD ( Clinical Practice Research Datalink) (UK), the Norwegian health
registers (NO), and SIDIAP (Sistema d’I nformació per el Desenvolupament de la
Investigació en Atenció Primària) [I nformation Sy stem for the Improvement of Research in
Primary Care] (ES).
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950244
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 16Data analy sis:The distributions of baseline characteristics at time zero by exposure group
will be calculated to describe the study cohort and illustrate differences between the groups.
For safet y outcome s, the risk over specific time period(s), incidence rate s and the ir
corresponding 95% confidence intervals (CIs) will be computed after the receipt of a first
dose and similarly after the receipt of a second dose. Crude risks, cumulative incidence over
different time periods , and measures of association (risk differences and risk ratios) for each
AESI after vaccination will be estimated in the entire population overall across both doses
and separatel y by dose. Subgroup anal yses will be conducted b y subgroups defined b y
demographic and clinical characteristics aswell as other covariates of interest. Individuals
following each vaccination category under study (vaccination with at least one dose of the
Pfizer -BioNTech vaccine vs. no vaccination) may have different characteristics that may
determine their risk of AESI . To account for such potential confounding, propensity score
methods will be used to estimate the adjusted risk ratios and 95% CIs. Using the main
estimates from each data source, appropriate random- effects meta -analytic methods will be
used to obtain a combined effect estimate. Where appropriate, the study will also use a SCRI
design.
Milestones
Milestone Planned Date
Registration in theEuropean Union electronic
register of post -authorisation studies (EU PAS
Register )Before the start of data collection
Start of da ta collection 30September 2021
End of data collection 31 December 2023
Progress report130 September 2021
Interim report 1 31 March 2022
Interim report 2 30 September 2022
Interim report 3 31 March 2023
Interim report 4 30 September 2023
Interim report 5 31 March 2024
Final study report 30 Sept 2024
1Data will not be provided in the progress report
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950245
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 175.AMENDMENTS AND UPDAT ES
None
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950246
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 186.MILESTONES
Milestone Planned Date
Start of data collection 30 September 2021
End of data collection 31 December 2023
Progress report130 September 2021
Interim report 1 31 March 2022
Interim report 2 30 September 2022
Interim report 3 31 March 2023
Interim report 4 30 September 2023
Interim report 5 31 March 2024
Final study report 30 Sept 2024
1Data will not beprovided in the progress report
7.RATIONALE AND BACKGR OUND
The novel coronavirus SARS -CoV -2, the cause of COVID -19, has resulted in a global
pandemic. The Pfizer -BioNTech COVID -19vaccine, tozinameran (Comirnaty ®) a novel
mRNA -based vaccine, has been authorised for use in several countries, including those in the
EU,for the prevention of COVID-19. Rapid uptake of the vaccine is expected. Because of
the relatively short prelicensure period a nd limited number of participants in clinical studies,
efficient and timely monitoring of the safet y of the vaccine will be needed in European
countries.
The safet y of the Pfizer -BioNTech COVID -19
vaccine has been investigated in clinical
studies conducted in the United States, Europe, Turkey , South Africa, and South America
and included over 43,000 patients aged 16 years and older. The overall safety profile of the
vaccine was found to be favourable in the trial setting. Reported adverse reactions from
unblinded data (i.e., from the overall trial population) on participants aged 16 years and older
who received two doses of Pfizer-BioNTech COVID- 19 vaccine 21 days apart after 2 months
of follow -up included pain at the injection site (84.1%), fatigue (62.9%), headache (55.1%),
muscle pain (38.3%), chills (31.9%), joint pain (23.6%), fever (14.2%), injection site
swelling (10.5%), injection site redness (9.5%), nausea (1.1%), malaise (0.5%), and
lymphadenopathy (0.3%). The safet y database revealed an imbalance of cases of Bell’s pals y
(four in the vaccine group and none in the placebo group) [1].Severe allergic reactions have
been reported following receipt of the Pfizer -BioNTech COVID -19vaccine in mass
vaccination campaigns outside clinical trials in various countries. Additional safet y events
may become evident with more widespread use in the general population.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950247
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 19The mRNA vaccine requires careful storage in ultra -low temperature freezers at temperatures
between −80°C and −60°C and must be protected from light and ultraviolet radiation before
use [1].Before administration, thawing and dilution of the vaccine are required. These
requirements may restrict vaccination events to larger medical centres with appropriate
storage capabilities.
The Pfizer- BioNTech COVID -19vaccine was investigated and is currently recommended as
a two -dose vaccine series, with two doses of 0.3 mL each, administered intramuscularl y
21days apart [1]. Little or no data are currentl y available on the safet y and effectiveness of
incomplete vaccine series, series mixed with other potentially available vaccines, or altered
dosing schedules.
Public health authorities have identified priorit y populations for vaccination based on health
care or essential worker status, comorbidities, and age [2].Early distribution of the vaccine
may be limited to vulnerable groups at higher risk for COVID -19infection andCOVID -19
complications. As recommendations for vaccination are updated over time, the characteristics
of vaccine recipients are expected to vary considerably . Approaches for investigating vaccine
safet y must flexibly account for changing vaccine distribution, which may vary by country or
jurisdiction in Europe.
This non- interventional study is designated as a post -authorisation safet y study (PASS) and is
a commitment to the EMA.
8.RESEARCH QUESTION AND OBJECTIVES
Research question: Is there an increased risk of select adverse events of special interest
(AESI ) after being vaccinated with the Pfizer -BioNTech COVID- 19 vaccine?
8.1. O bjectives
Primary study objective
To determine whether an increased risk of prespecified AESI exists following the
administration of at least one dose the Pfizer -BioNTech COVID -19 vaccine using two
approaches: (a) a cohort design comparing risk in vaccinated andnon-vaccinated
individuals and(b)a SCRI design.
Secondary study objectives
To estimate the incidence rates of prespecified AESI among individuals who receive at
least one dose of the Pfizer- BioNTech COVID -19 vaccine using a cohort study design.
To describe the incidence rates and determine whether an increased risk of prespecified
AESI exists following the administration of a tleast one dose the Pfizer -BioNTech
COVID -19 vaccine compared with a matched comparator group with no COVID -19
vaccination within subcohorts of interest (i.e., individuals who are immunocompromised,
individuals who are frail and have comorbidities, individuals diagnosed with previous
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950248
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 20COVID -19 infection, and age -specific groups) in Europe using a cohort study design
and/or a SCRI design.
To determine whether an increased risk of prespecified AESI exists following the
administration of at least one dose of the Pfizer-BioNTech COVID- 19 vaccine compared
with no COVI D-19 vaccination, in pregnant people and their neonates using a cohort
study design.
To characterise utilisation patterns of Pfizer- BioNTech COVID -19 vaccine among
individuals within Europe ,including estimating the proportion of individuals receiving
thevaccine; two -dose vaccine completion rate and distribution of time gaps between the
first and second doses; and demographics and clinical characteristics of recipients, overall
and among subcoh orts of interest , such as individuals who are immunocompromised,
elderl y, or have specific comorbidities.
9.RESEARCH METHODS
9.1.Study design
This post -authorisation active surveillance stud y of safet y events of interest associated with
the Pfizer -BioNTech COVID- 19 vaccine will usea retrospective cohort design involving
multiple databases.
In addition to the cohort anal ysis, for a subset of the study endpoints (see Table 1), the SCRI
design will also be used to assess risk .The SCRI design will be used to sequentially monitor
the occurrence of AESI while controlling for time- invariant confounders (such as sex, race,
chronic illness, and health state).
9.1.1. Retrospective cohort desi gn
A retrospective cohort design will be used to estimate the incidence of AESI after receipt of
the vaccine . Incidence rates of prespecified AESI among individuals who receive at least one
dose of the Pfizer -BioNTech COVID 19 vaccine will be calculated.
The primary objective will be addressed in a comparative anal ysis of this incidence with that
occurring in an unvaccinated matched comparator group.
In this retrospective cohort design, time zero will be defined as the time at which the
exposure status is assigned, when inclusion and exclusion criteria are applied and when stud y
outcomes start to be counted [1-5]. Time zero in the exposed groups (i.e., recipients of the
vaccine) will be the day the first vaccination dose was received. Time zero in the unexposed
group will be a day when they did not receive a Pfizer -BioNTech COVID -19 vaccine dose.
This day will be chosen by calendar matching to the time zero of the corresponding exposed
group ;at each calendar day when an individual is vaccinated, those individuals who were not
vaccinated that same day (time zero) or before will be assigned to the unexposed group,
matching them to the vaccinated individual by important clinical variables (e .g.,age,
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950249
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2.0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 21indicated and recommended characteristics to be vaccinated at the time, stratification
variables) at time zero.
9.1.1.1. Matching process
As in prior applications of real -world data studies of the Pfizer -BioNTech COVID -19
vaccine[6], we propose to perform a 1:1 matching without replacement using a “rolling
cohort ”design. Starting from th e first day of the study period, for each day , we will attempt
to match newly vaccinated individual swho meet the eligibility criteria that day, even if they
had previousl y been included as an unvaccinated control. Newl y vaccinated individuals will
be matc hed 1:1 with unvaccinated controls meeting the eligibility criteria that day who were
not previously matched. If at a later date an unvaccinated control is vaccinated, they and their
vaccinated match will be censored from this comparative anal ysis. The new ly vaccinated
individual ( Patient 2 in the figure below) will be censored from the unvaccinated group and a
new matched unvaccinated control will be sought so that this newly vaccinated individual
and
the new pair will be included in the anal yses, if eligi bility criteria are still met on the day
of vaccination.
Patients will be match edon the following variables, which have shown good control of
confounding for vaccine effectiveness in a prior study[6]:
Age –of 2-year age groups (consecutive years)
Sex (male, female) –exact matching
Previous COVID- 19 infection (y es/no) –exact matching
Place of residence –exact matching, at the level of clinical practice, neighbourhood, or
small town or prox y as available (specific for data source)
Influenza vaccines in the past 5years (0, 1 -2, 3
-4, 5) –exact matching
Pregnancy (yes, no ) –exact matching
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950250
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 22Immunocompromised (yes/no) –exact matching
Number of pre -existing conditions considered by the Centers for Disease Control ( CDC )
as risk criteria (0, 1, 2, 3, 4+)[7,8]– exact matching
This selection of variables for matching was based on a prior real -world data study[1]. The
selection of variables will be tailored based on the variable availability of each data source .
Conditional e xchangeability will be evaluated b y study ing the difference in the risk of
outcomes that depend on the antibody -mediated immunogenicit y of the COVID -19 disease,
in the first 14 days after the first dose, which should be close to 0. If conditional
exchangeabi lity is not achieved after implementing all statistical adjustments, the variables
for matching may be revisited to improve it. To ensure exchangeability, additional health
history measures, such as comorbidities associated with an increased risk of AESI willbe
explored. Additional details on the matching process, criteria for establishing
exchangeability ,and analy ses to account for the potential autocorrelation introduced by this
situation will be specified in the statistical analy sis plan (SAP).
A single individual may contribute to both exposed and unexposed groups at different time
points (details will be described in the SAP) . The causal contrast of interest will be the
observational analogue of a per -protocol effect, that is, the event rate di fference that would
be observed if all individuals received at least one dose of the Pfizer -BioNTech vaccine vs. if
no individuals received it. Individuals will be classified into exposure groups that are
compatible with their data at time zero. Follow- up under unexposed status and its
corresponding exposed pair is censored if a n individual receives a COVID -19 vaccine .
Unmatched vaccinated individuals will not be included in the retrospective cohort anal ysis.
They will be described in the descriptive anal ysis.
9.1.2. Self-controlled risk interval design
As an additional and complementary approach for a subset of study outcomes that are acute
and meet other necessary assumptions, a SCRI design will be used. These assumptions
include that the outcome must have acute onset and short latency and must have relatively
well-known risk intervals; the design is less suited to study outcomes that affect the
probability of exposure, but this potential bias can be reduced b y the use of a post -
vaccination control interval.
Vaccine exposure is known to be challenging to measure, particularl y in a pandemic setting
where vaccines may be administered outside the usual healthcare s ystem. Often, this re sults
in underascertainment of exposure and the inclusion of exposed persons in the unexposed
cohort. This underascertainment of exposure could result in a bias towards the null if the
vaccine does increase the risk of an event. Asthe SCRI design includes only people with
known vaccine exposure, it is not subject to this bias.
The SCRI design will compare the risk of each outcome during a prespecified period
following each dose during which there is a h ypothesised increased risk of the outcome (“risk
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950251
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2.0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 23interval”) with a self -matched control interval, used to assess the baseline risk of the
outcome.
The SCRI design will be performed in the overall vaccinated population, including among
vaccinated individuals not included in the retrospective cohort analysis because a matching
comparator could not be found. This design will serve as a sensitivity analy sis and will allow
the evaluat ionof the exclusion of
unmatched pairs from the anal ysis.
Table 1 defines the risk windows proposed for each AESI and indicate sfor which AESI a
SCRI anal ysis would be a valid approach .
A prespecified post -vaccination control interval will be used for each outcome. This
approach avoids bias because of outcomes affecting the probability of exposure (e.g., the
outcome is a contraindication for exposure or delayed exposure). For individuals who receive
two doses of the vaccine, the risk interval will extend beyond each vaccine dose.
For outcomes with short risk intervals, for each dose, the control interval will occur close in
time to the risk interval associated with that dose and before the second dose is given. For
outcomes with risk intervals longer than the gap between doses, among individua ls receiving
two doses, the control interval for each dose will occur after the risk interval of the second
dose (see Figure 1).
Figure 1.Self-controlled risk interval design
T =time measured in day s.
Note: Example with a risk period of 42 days and a control period of 42 days.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950252
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 249.2. Setting
For the implementation of this study , we will use electronic health care databases in Europe.
The selected data sources and two- letter country codes are as follows:
PHARMO (PHARMO I nstitute for Drug Outcomes Research) (NL)
ARS Toscana (Agenzia Regionale di Sanita’ della Toscana) [a research institute of the
Tuscan y region of Ital y] (IT)]
Pedianet/Health Search Database (HSD) (IT)
EpiChron (EpiChron Research Group on Chronic Diseases at the Aragon Health Sciences
Institute) (ES)
CPRD (Clinical Practice Research Datalink) (UK)
The Norwegian health registers (NO)
SIDIAP (Sistema d’Informació per el Desenvolupament de la Investigació en Atenció
Primària) [I nformation Sy stem for the Improvement of Research in Primary Care] (ES)
9.2.1. Inclusion criteria
9.2.1.1. Cohort design
Individuals must meet all the following inclusion criteria to be eligible for inclusion in the
cohort study :
Have a minimum of 12 months (or from birth if enrolled in the data source at birth) of
active enrolment and history in one of the selected data sources to ensure adequate
characterisation of medical history ; this criterion may be met after the start of the study
period.
No history of vaccina tionwith a non–Pfizer -BioNTech COVID -19 vaccine before time
zero.
At any point in time, vaccinated individuals may differ from the remaining population in
characteristics that may determine their risk of AESI . Measured baseline differences will be
adjusted for analy tically (Section 9.7).
For the stud y of pregnancy outcomes, the cohort will be restrict ed to pregnant women.
Details of the differences from the main cohort approach will be described in the SAP.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950253
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 259.2.1.2. Self-controlled risk interval design
For analyses of outcomes assessed with the SCRI design, the following criteria must be met.
Note that the stud y population for each outcome -specific anal ysis will thus be different.
Have received at least one dose of the Pfizer -BioNTech COVID -19 vaccine.
Have experienced an event during the risk or control interval.
Have full accrual of data used to define the eve nt in the risk and control intervals
combined, taking into account the data lag and timing of data extraction.
9.2.2. Exclusion criteria
9.2.2.1. Cohort and SCRI design s
There are no exclusion criteria. Individuals having any specified contraindication to
vaccination or b eing part of a group not recommended for vaccination in the jurisdiction of
the study will be analy sed separately
9.2.3. Source population
The source population for both cohort and SCRI designs will be composed of all individuals
registered in each of the health care data sources. The selected European populations are the
populations underly ing the data sources listed in Section 9.2.
9.2.4. Study period
The study period for both cohort and SCRI designs will start on the date of launch of the
Pfizer -BioNTech COVID- 19 vaccine in each country participating in the study and will end
on the date of the latest data availability. Follow -up will last for 2 years for AESI .
Differences in follow -up for acute and non -acute events will be described in the SAP. An
additional y earwill accrue for pregnancy outcomes to occur in pregnancies occurring during
the 2 years of follow -up (see Figure 2).
Figure 2. Study period and follow- up periods
AESI =adverse events of special interest.Follow-up AESI waiting time pregnacy outcome s2021 2022 2023
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950254
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 269.3.Variables
9.3.1. Exposure definition, by data source
Exposure will be based on recorded prescription, dispensing, or administration of the
Pfizer -BioNTech COVID-19 vaccine. Vaccine receipt and date of vaccination should be
obtained from all possible sources that capture COVID -19 vaccination, such as pharmacy
dispensing records, general practice records, immunisation registers, vaccination records,
medical records, or other secondary data sources. Depending on the data source, vaccines
may be identified via nationally used product codes —including batch numbers—where
possible. The main exposure of interest is in the receipt of at least one dose of the
Pfizer -BioNTech COVID- 19 vaccine. Other exposure groups will also be described.
PHARMO (NL): Data on vaccination will be included in PHARMO’s General Practitioner
(GP)database. Information on vaccines includes Anatomical Therapeutic Chemical ( ATC )
code, brand, batch, and date of application.
ARS Toscana (IT) will identify vaccines using the nationally used product code, including
batch number.
Pedianet/HSD (IT) : Information on COVID- 19 vaccine will include date of immunisation,
type of vaccine, vaccine batches, dose. They will be collected b y the paediatrician at each
contact with the patient.
EpiChron – Aragon data sources (ES ): The Aragon Health Sy stem (Aragon, Spain) has
implemented a specific vaccination register embedded in the electronic health record (EHR)
system. The COVID -19vaccine is being s ystematic ally registered in this register b y health
care professionals. This register can collect all the relevant information regarding the
vaccination process, such as patient’s identifier; date of administration and due date for next
dose, if applicable; centre of administration; part of the body where vaccine is administered;
name of the vaccine; brand (laboratory); batch number; dose; and vaccination criteria (risk
group to which the patient belongs). There is also a free -text section in which health
profession als can include their observations (e.g., presence or not of an allergic reaction).
Clinical Practice Research Datalink (UK) : The CPRD contains information recorded by
National Health Service (NHS) primary care GPs; and information on the administration of
COVID -19 vaccines to individuals will be available. This will include, alongside an
encry pted unique patient identifier; the name of the vaccine; manufacturing company ; dose;
stage of the vaccine schedule; administration route; administration location (e. g.;general
practice); batch identifiers/numbers; date of administration; and medical observations,
events, referrals, test results, and prescribed medications recorded b y the GP prior to, on, or
after the vaccination date. Free- text medical notes may also be available if recorded;
however, this is dependent on patient anony mity being maintained. In addition, patient
demographic, practice -level, and staff-level information is also available.
Furthermore, other CPRD -linked COVID -19 data sets that may provide further follow-up
information on AESI include the Public Health England (PHE) Second Generation
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950255
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 27Surveillance Sy stem (SGSS) COVI D-19 positive virology test data, PHE COVID -19
Hospitalisation in England Surveillance Sy stem (CHESS), and the Intensive Care Na tional
Audit and Research Centre (ICNARC) data on COVID -19 intensive care admissions.
Standard CPRD -linked data sets can also be obtained including Hospital Episode Statistics
(HES) data sets covering hospital secondary care (Accident & Emergency , Admitted Patient
Care, Inpatient and Outpatient), Office for National Statistics (ONS) data sets for Death
Registry information, mother -baby link, and an algorithm -based Pregnancy Register.
Norwegian health registers (NO) : The national, electronic immunisation register
(SYSVAK) was established in 1995 and records an individual’s vaccination status and
vaccination coverage in Norway . All vaccinations are subject to notification to SYSVAK and
are registered without obtaining patient consent. This applies to all CO VID-19 vaccines. In
SYSVAK, the following data are registered: individual personal identifier, vaccine name and
ATC code, vaccine batch number, date of vaccination, reason for vaccination as health care
professional versus risk -group patient, and the centr e where the vaccine was administered.
SIDIAP (ES) : For all 5.8 million individuals of the Catalan Institute of Health–Primary Care
teams, SI DIAP will have available information on the administration of COVID -19 vaccines
to individuals linked to a unique an d anon ymous identifier. The information will be
originated from the electronic medical records. For each patient, SI DIAP will have date and
centre of administration, health professional administering the vaccine, dose, brand, reasons
for vaccination (e.g., risk of group), and other information related to vaccination. As the
Pfizer -BioNTech COVID-19 vaccine is indicated as a two- dose vaccine series, multiple
vaccinations per person will be identified.
9.3.1.1. Cohort design
The vaccination categories for the differen t exposure groups will be defined as follows:
1. Receipt of at least one dose of the Pfizer- BioNTech COVID -19 vaccine, followed or
not by a second dose of the Pfizer -BioNTech COVID- 19 vaccine. Individuals will be
censored if and when they receive a non –Pfizer -BioNTech COVID -19 vaccine during
follow -up.
2.The vaccination category for the matched unexposed group will be defined as not
receiving a COVID -19 vaccine of any brand during the study period. Individuals will
be censored when they receive a dose of an y COVID -19–directed vaccine during
follow -up.
As a sensitivity anal ysis, a vaccination category consisting ofthe receiptof the two
vaccination doses per the recommended schedule will be studied (i.e., receipt of a first dose
of the Pfizer-BioNTech COVID- 19 vaccine, followed by a second dose by week 4 after the
first dose in the absence of an adverse event, and never receiv inga non–Pfizer -BioNTech
COVID -19 vaccine). For this specific sensitivity analy sis, an d not for the main anal ysis,
individuals will be censored if they do not receive the second dose of the Pfizer -BioNTech
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950256
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 28COVID -19 vaccine by week 6 after the first dose in the absence of an adverse event and
if/when they receive a non –Pfizer -BioNTech COVID- 19 vaccine during follow -up.
The operationalisation of these exposure strategies is described in 9.7.
9.3.1.2. Self-controlled risk interval design
For the SCRI design, for each dose, person- time in the risk interval will be considered
“exposed”, while person -time in the control interval will be considered “unexposed ”.Risk
intervals will be specific to the outcome of interest and are defined t o reflect the duration of
time post-vaccine exposure that a n incident vaccine -induced event would be expected to
occur. Events known to have a risk window limited to a defined period after vaccination are
are not well known for COVID- 19 vaccines, but have been defined based on prior post-
marketing studies of other vaccines (where applicable), clinical trial data (where applicable),
and passive post -marketing surveillance activities (when they become available). An acute
even t, while time -limited in duration, does not necessarily have a defined risk window if
there is not a known time -limited window post vaccine exposure that the acute event would
be expected to occur post-vaccination.
Outcome -specific control intervals will also be defined. For outcomes with short risk
intervals, the control interval will occur relativel y close in time to the risk interval of each
dose. For outcomes with long risk intervals, among individuals receiving two doses, the
control interval for both the first and second doses will occur after the risk interval of the
second dose.A sensitivity anal ysis will be performed, where the exposed group of vaccinees
is restricted to those who receive vaccine per the recommended schedule, (i.e.,two doses of
the Pfizer -BioNTech COVID-19 vaccine per the Pfizer -BioNTech recommended dosing
schedule ).
9.3.2. Outcomes definition
9.3.2.1. Safety outcomes
Outcomes will be defined homogeneously across the data sources to the fullest extent
possible. Selected A ESIcurrentl y planned for inclusion in the study are listed in Table 1 and
are based on those proposed by the ACCESS project (vACcine COVID -19monitoring
readinESS), which has been funded b y the EMA to ensure that a European infrastructure will
be in place to eff ectively monitor COVI D-19 vaccines in the real world, once the vaccines
are authorised in the EU (http://www.encepp.eu/encepp/viewResource.htm?id=37274.).
Table 1.List of Selected Adverse Events of Special Interest
Body system /
classificationAdverse event of special
interestEstimated risk window
(days)Analytic Approach
Autoimmune
diseasesGuillain -Barré syndromea1-4230Cohort /SCRI
Acute disseminated
encephalomyelitis1-4230Cohort /SCRI
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950257
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 29Table 1.List of Selected Adverse Events of Special Interest
Body system /
classificationAdverse event of special
interestEstimated risk window
(days)Analytic Approach
Narcolepsya1-42b Cohort /SCRI
Acute aseptic arthritis 1-42d Cohort
Diabetes (type 1 and broader) Any Cohort
(Idiopathic)
thrombocytopeniaa1-4231Cohort/SCRI
Heparin -induced
thrombocytopenia (HIT) –like
eventa1530Cohort/SCRI
Cardiovascular
systemAcute cardiovascular injury
including microangiopathy,
heart failure, stress
cardiomyopathy, coronary
artery disease, arrhythmia,
myocarditisAnye Cohort
Circulatory
systemCoagulation disorders:
thromboembolism,
haemorrhage1-2830Cohort/SCRI
Single organ cutaneous
vasculitis1-28f Cohort/SCRI
Hepato -
gastrointestinal
and renal
systemAcute liver injury 1-42hCohort
Acute kidney injury 1-42hCohort
Acute pancreatitis 1-42h Cohort
Rhabdomyolysis Any Cohort
Nerves and
central
nervous
systemGeneralised convulsion 1-4230Cohort/SCRI
Meningoencephalitis 1-4230Cohort/SCRI
Transverse myelitisa1-4230Cohort/SCRI
Bell’s palsy 1-4230Cohort/SCRI
Respiratory
systemAcute respiratory distress
syndromeAny Cohort
Skin and
mucous
membrane,
bone and
joints systemErythema multiforme 1-42gCohort
Chilblain -like lesions 1-42fCohort
Other system Anosmia, ageusia 1-42 Cohort
Anaphylaxisa130Cohort
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950258
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 30Table 1.List of Selected Adverse Events of Special Interest
Body system /
classificationAdverse event of special
interestEstimated risk window
(days)Analytic Approach
Multisystem inflammatory
syndrome1-42c Cohort
Death (any causes) Any Cohort
Subacute thyroiditis 1-42d Cohort
Sudden death Any Cohort
Pregnancy
outcome,
maternalGestational diabetes Any time pregnancy Sub-cohort
Preeclampsia Any time pregnancy Sub-cohort
Maternal death Any time pregnancy Sub-cohort
Pregnancy
outcome,
neonates .
Define design
taking
trimester into
accountFoetal growth restriction Any time pregnancy Sub-cohort
Spontaneous abortions After vaccination Sub-cohort
Stillbirth After vaccination Sub-cohort
Preterm birth At preterm birth Sub-cohort
Major congenital anomaliesa1year after birth Sub-cohort
Microcephaly At birth Sub-cohort
Neonatal death At birth Sub-cohort
Term ination of pregnancy for
foetal anomalyAt termination Sub-cohort
Any COVID -19 Disease Any Cohort
Vaccine- associated enhanced
disease (VAED)aAny Cohort
AESI =adverse events of special interest; VAED =vaccine -associated enhanced disease.
Notes:
a For this AESI clinical validation will occur.
bPublished risk and control intervals for demyelinating diseases and cra nial disorders w ere applied to TM
and narcolepsy/cataplexy.
cAs severe COVID -19 ranges from severe pneumonia, acute respiratory distress syndrome, and multisystem
organ failure/MIS -A, a 1- 42 day risk interval was applied in order to capture the 14 -day incubation period of
the disease and 4 -5 day period from exposure to symptom onset.
dPublished risk and control intervals for autoimmune disorders w ere applied to similar autoimmune
rheumatic conditions (i.e., fibromyalgia and autoimmune thyroiditis).
ePublished risk and control intervals for myocarditis and pericarditis w ere applied to other cardiovascular
conditions (i.e., heart failure and cardiogenic shock, stress cardiomyopathy, CAD, arrhythmia, AMI).
fSimilar risk and control intervals were applied to all cardiovascular and hematological disorders
characterized by damage to the blood vessels and/or arteries and clotting (i.e., microangiopathy, DVT,
pulmonary embolus, limb ischemia, hemorrhagic disease, DIC, chilblain -like lesions). The published risk and
control intervals for KD were applied to vasculitides given that KD is a type of medium and small -vessel
vasculitis.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950259
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 31Table 1.List of Selected Adverse Events of Special Interest
Body system /
classificationAdverse event of special
interestEstimated risk window
(days)Analytic Approach
gPublished risk and control intervals for non -anaphylactic allergic reactions were a pplied to hypersensitivity
disorders (i.e., erythema multiform e).
hRisk intervals of 42 days were applied for acute kidney injury and liver injury to be consistent with other
COVID -19 related safety events of interest .
Outcomes will be identified in E HR databases with algorithms based on codes for diagnoses,
procedures, and treatments. Definitions, codes, and proposed algorithms for all AESI will
incorporate definitions developed b y the ACCESS project
(https://drive.google.com/drive/folders/1Y_3cuGRN1g- jBv2ec1fC0aYcpxEjtrY9) and will
be described in more detail in the SAP.
9.3.2.1.1. Outcome identification and validation, by data source
AESIwill be identified based on patient profile review of electronic records by health care
professionals. In addition, for selected outcomes mentioned in Table 1 and others ( if
considered necessary in a future evaluation of results), manual review of patient charts
conducted b y clinici ans blinded to COVI D-19 vaccine exposure will be performed when
possible and will be based on data source structure. Confirmation of an event diagnosis will
be classified against existing definitions of the Brighton Collaboration and those currentl y
being developed.
Standard algorithms for each outcome definition will be applied to participant data sources,
based on the results of the ACCESS project. Algorithms will be tailored to the data source
and will consider the nature of the records that have identif ied the outcome, e.g., primary
care, access to hospital care, access to emergency care [9]. Multiple algorithms for the same
outcome may be included in the anal ysis, to assess the potential impact of differe ntial
misclassification.
Potential outcomes will be identified based on patient profile review of electronic records b y
health care professionals.
PHARMO (NL): For the validation study , information on selected endpoints from patient
medical records will be abstracted b y local medical professionals or PHARMO employ ees,
provided that medical chart review is approved b y ethics committees and other local and/or
national governing bodies.
Pedianet/HSD (IT) : A validation mechanism including an individual linkage with the
electronic regional immunisation register will be in place. Furthermore, the validation
process includes the review b y clinicians of the individuals ’ electronic medical records,
which contain information from primary care reports.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950260
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 32EpiChron (ES) : In Aragon (EpiChron) data sources, the proposed validation process is
based on the review of the individuals’ electronic medical records b y clinicians from the
research team who are blinded to COVID-19 vaccination status. These records include
information from primary care reports, hospital discharge reports (including hospital
emergency rooms), and results of diagnostic tests and laboratory tests.
Clinical Practice Research Datalink (UK) : In the United Kingdom (UK), validation will be
conducted b y review of electronic medical record information for selected endpoints by an
adjudication committee who will be blinded to COVID -19 vaccination status.
Norwegian health registers (NO) : In Norway , the validation process is based on the manual
review of hospital charts for a subsample of individuals with the adverse event of interest,
compared with registered diagnoses in the Patient Registry of Norway . Validation studies are
alread y available for selected health outcomes (e.g., intracranial haemorrhage, hip fractures,
cancer). Depending on the adverse event of interest, validation is possible by comparing the
registered diagnosis in two separate registers (e.g., the Norweg ian Patient Registry versus the
Norwegian Stroke Register).
SIDIAP (ES) : In SIDIA P, the validation process is part of data qualit y control. Validation
will be based on the review of the electronic medical record information (ECAP) by
members of the SIDIAP research group who will be blinded to COVID -19 vaccination status.
9.3.3. Covariate definition
The following variables will be assessed at time zero (for the cohort design) or the date of
initial vaccine dose (for the SCRI design) to be used to define patient populations of special
interest or priorit y vaccination groups, to define subgroups of interest for secondary anal yses,
or to control for confounding. The AESI may have different sets of risk factors, and
outcome- speci fic anal yses may contain different covariate sets. Potential covariates may
include the following information, as available in each data source:
Demographics
Age at time zero (will be used to define subgroups for secondary anal yses)
Age will be categorised as agecategories in line with published background incidence
rates from ACCESS (0 -19, 20-29, 30-39, 40-49,50-59, 60-69, 70-79,80+)
Sex
Pregnancy status and pregnancy trimester at time zero
Race and/or ethnicity , as appropriate in each country
Geographi c region, as appropriate in each country
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950261
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 33Socioeconomic status, as available in each country (including housing, employ ment,
and income, if available)
Residency in a long -term care facility
Health care worker or essential worker status, if available
Date of vaccination (categorised as appropriate, e.g., by year or month)
Batch of vaccine received
COVID -19 history , as available in each data source (will be used to define a subgroup of
interest)
Previous diagnosis of COVID -19
Positive test result for COVID -19
Personal lifestyle characteristics
Smoking status (if available)
Body mass index (if available)
Comorbidities
History of anaph ylaxis
History of allergies
Diabetes mellitus (ty pes 1 and 2)
Hypertension
Cardiovascular disease
Cerebrovascular disease
Chronic respiratory disease
Chronic kidney disease
Chronic liver disease
Cancer
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950262
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 34Autoimmune disorders
Influenza infection or other respiratory infections
Charlson Comorbidity Index (may be included as the composite scale, or the scale
components may be included as individual terms)
Immunocompromising conditions (will be used to define subgroups for secondary
analyses)
Immunodeficiencies
Immunosuppressant medication use
Human immunodeficiency virus and other immunosuppressing conditions
Comedication use during the y ear before time zero (prescriptions or dispensing, no
over-the-counter medication use)
Analgesics
Antibiotics
Antiviral medications
Corticosteroids
Non-steroidal anti -inflammatory drugs
Psychotropics
Statins
Novel oral anticoagulants
Warfarin
Health care utilisation in the y ear before time zero and in the 2 weeks before time zero
Number of hospitalisations
Number of emergency department visits
Skilled nursing facilit y, nursing home, or extended care facility stay
Primary care utilisation
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950263
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 35Cancer screening
Other preventive health services, as appropriate
COVID -19 tests
Other vaccinations
Influenza
Pneumococcal
DTP (diphtheria, tetanus, and pertussis)
TPV (polio)
TV (MMR) (measles, mumps and rubella)
Hib (Haemophilus influenzae ty pe b)
HB (hepatitis B virus)
VV (varicella zoster virus )
HZ (herpes -zoster virus )
HPV (human papillomavirus)
Mening ococcal
Rotavirus
Surrogates of frailty
Wheelchair use
Home hospital bed
Paraly sis
Parkinson’s disease
Skin ulcer
Weakness
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950264
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 36Stroke/brain injury
Ambulance transport
Dementia
Difficulty walking
Home ox ygen
Rehabilitation care
Psychiatric illness
Sepsis
Heart failure
Podiatric care
Bladder incontinence
Diabetes complic ations
Arthritis
Coagulation deficiencies
Vertigo
Lipid abnormalities
9.4.Data sources
The study will use data from secondary EHR databases that are population based. All data
sources will have the ability to provide high- quality data on COVID -19 vaccines (product
types and dates), outcomes (diagnoses, procedures, and treatments), and important
covariates. It is not currently known the extent to which COVID -19 vaccines, product t ypes,
and batch numbers will be captured in data sources.
At the proposal stage, members of VAC4EU (Vaccine monitoring Collaboration for Europe)
(https://vac4eu.org/) were offered the option to participate in the study . Several data sources
have indicated the ability to participate in the study and are described in the following
subsec tions. As vaccine delivery and registration are not full y determined, we are exploring
additional options in Portugal and Norway (we include here the Norwegian data sources as
they will be able to contribute with one data extraction per y ear).
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950265
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 37Data availab ility for each institution might be affected by third parties or external
circumstances that are independent from the institution involved in the study as described
below in S ection 9.9.
9.4.1. PHARMO (NL) (6million active individuals)
The PHARMO Database Network, which is maintained by the PHARMO Institute for Drug
Outcomes Research, is a population- based network of EHR databases that combines
anony mous data from different primary and secondary health care settings in the
Netherlands. These different data sources —including data from general practices, in- and
outpatient pharmacies, clinical laboratories, hospitals, the cancer regis ter, the pathology
register, and the perinatal register —are linked on a patient level through validated
algorithms. To ensure data privacy in the PHARMO Database Network, the collection,
processing, linkage, and anony misation of the data are performed by STIZON, which is an
independent, ISO/IEC 27001 certified foundation that acts as a trusted third party between
the data sources and the PHARMO Institute. The longitudinal nature of the PHARMO
Database Network enables the follow -up of more than 9 million ind ividuals of a well -defined
population in the Netherlands for an average of 12 years. Currently , the PHARMO Database
Network covers over 6 million active individuals out of 17 million inhabitants of the
Netherlands[10]. Data collection period, catchment area, and overlap between data sources
differ. Therefore, the final cohort size for an y study will depend on the data sources included.
All electronic patient records in the PHARMO Database Network include information on
age, sex, socioec onomic status, and mortality . Other available information depends on the
data source. A detailed description of the different data sources is given in subsequent
sections. The PHARMO Institute is always seeking new opportunities to link with health care
databases. Furthermore, it is possible to link additional data collections, such as data from
chart reviews, patient-reported outcomes, or general practice trials.
The General Practitioner database comprises data from electronic patient records registered
by GPs. The records include information on diagnoses and s ymptoms, laboratory test results,
referrals to specialists, and health care product/drug prescriptions. Th e prescription records
include information on type of product, prescription date, strength, dosage regimen, quantity ,
and route of administration. Drug prescriptions are coded according to the World Health
Organization (WHO) ATC classification sy stem [www.whocc.no]. Diagnoses and s ymptoms
are coded according to the International Classification of Primary Care (ICPC)
[www.nhg.org ], which can be mapped to the International Classification of Diseases (I CD)
codes but can also be entered as free text. General pr actitioner data cover a catchment area
representing 3.2 million residents (~20% of the Dutch population).
The Out -patient Pharmacy Database comprises GP -or specialist -prescribed health care
products dispensed b y the outpatient pharmacy . The dispensing rec ords include information
on ty pe of product, date, strength, dosage regimen, quantity , route of administration,
prescriber specialty , and costs. Drug dispensings are coded according to the WHO ATC
classification s ystem. Outpatient pharmacy data cover a cat chment area representing
4.2 million residents (~25% of the Dutch population). The PHARMO Database Network is
listed under the European Network of Centres for Pharmacoepidemiology and
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950266
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 38Pharmacovigilance (ENCePP) resources database. PHARMO data capture influ enza
vaccination and may be linked to the PRAEVENTIS database maintained by RIVM, based
on specific permissions.
The GP Database contains vaccinations administered by GPs and by the public health
service, as GPs receive an automated notification when a pa tient has a positive corona test or
has been vaccinated via the public health service (provided that individuals have given their
consent).
The Dutch government wants every one from the age of 18 y ears to have had at least one
COVID -19 vaccination by the be ginning of July 2021. This vaccination schedule depends on
many factors (e.g., approval and effectiveness, delivery and distribution of vaccines to
injection sites, such as hospitals and GPs, new developments and advice from, for instance,
the Health Counc il of the Netherlands [i.e., de Gezondheidsraad]).
The Netherlands Perinatal Registry is maintained by Perined and comprises data on
pregnancies, births, and neonatal outcomes of births in the Netherlands, voluntarily collected
by perinatal caregivers mai nly for benchmarking. For research purposes, the data are linked
with the PHARMO Database Network via the TTP, resulting in the PHARMO Perinatal
Research Network (PPRN)[11].Records include information on mothers (e.g., maternal ag e,
obstetric history , parity ), pregnancy (e.g., mode of conception, mode of delivery ), and
children (e.g., birth weight, gestational age, Apgar score). Diagnoses and sy mptoms are
coded according to the Perinatal Registry code lists. For more information: w ww.perined.nl
Permission to obtain these data on a b y-project basis is needed from PHARMO as well as
from Perined.
PHARMO acknowledges that the data source they have access to includes data on vaccine
delivery and registration and undertakes to cooperate on addressing the study objectives by
contributing to providing reports based on such data.
9.4.1.1. Vaccine exposure
Currently , inthe Netherlands, different health care providers administer the COVID -19
vaccines (i.e., GPs, the public health service, and health care institutions) . The vaccination
data are recorded in a central register (if people have given permission beforehand). T his is
the COVID -19 vaccination I nformation and Monitoring S ystem (CIMS). PHARMO is
currentl y exploring the possibilities of linkingwith this register. Until then, the GP Database
will be the basis for the vaccination data and thus vaccination exposure ma y be
underascertained.
As of 3 May 2021, approximately 3.6million doses of the Pfizer -BioNTech COVID-19
vaccine have been administered in the Netherlands. Assuming that PHARMO covers
approximately 20% of the Dutch population, this should be 700,000 doses in PHARMO’s
catchment area.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950267
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 39Therefore i t is difficult to estimate how many individuals will be vaccinated with the Pfizer -
BioNTech COVID -19vaccine during the study period in the Netherlands. Also, the
administration to the Pfizer- BioNTech COVID -19 vacci nedepends on different factors, such
as type of work (care workers), home living (yes or no) , year of birth ,and existing
comorbidities.
9.4.2. ARS Toscana database (IT) (3.6million active individuals)
The Italian National Healthcare S ystem is organised at the regional level: the national
government sets standards of assistance and tax -based funding for each region, which
regional governments are responsible for providing to all their inhabitants. Tuscan y is an
Italian region, with approximately 3.6million inhabitants. The Agenzia Regionale di Sanita’
della Toscana is a research institute of the Tuscany region. The ARS Toscana database
comprises all information collected b y the Tuscany region to account for the health care
delivered to its inhabitants. Moreover, ARS Toscana collects data from regional initiatives.
All data in the ARS Toscana data source can be linked at the individual level through a
pseudo- anon ymous identifier. The ARS Toscana database routinely collects primary care and
secondary care d rug prescriptions for outpatient use and is able to link them at the individual
level with hospital admissions, emergency care admissions, records of exemptions from
copay ment, diagnostic tests and procedures, causes of death, the mental health services
register, the birth register, the spontaneous abortion register, and the induced terminations
register. A pathology register is available, mostly recorded in free text, but with morphology
and topographic SNOMED codes. Mother -child linkage is possible through the birth register.
Vaccination data since 2016 are available for children and since 2019 for adults. However, to
date, 2019 vaccination data for adults may still be incomplete resulting in an
underascertainment of vaccine exposure . The ARS Toscana database was characterised in the
ADVANCE project and considered fit for purpose for vaccine coverage, benefits, and risk
assessment when using the new vaccine register (from 2019) [12].
ARS Toscana acknowledges that the data source they have access to includes data on
vaccine delivery and registration and undertakes to cooperate on addressing the study
objectives by contributing to providing reports based on such d ata.
9.4.2.1. Vaccine exposure
Using data from the European Center s for Disease control[13], as of 08 May 2021 , 17,801,550
doses of the Pfizer -BioNTech COVID -19 vaccine have been administered . If the distribution
of the vaccine were uniform across Italian regions, it is expected that 1 million doses would
have been administered in Toscana .
9.4.3. Pedia net/Health Search Database (IT) (1million active individuals)
Pedianet, a paediatric general practice research database, was set up in 2000. I t contains
reason for accessing health care, health status (according to the Guidelines of Health
Supervision of the American Academ y of Pediatrics), demographic data, diagnosis and
clinical details (free text or coded using the ICD -9-CM [I nternational Classification of
Diseases, Ninth Revision, Clinical Modification]), prescriptions (pharmaceutical
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950268
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 40prescriptions ide ntified by the ATC code), specialist appointments, diagnostic procedures,
hospital admissions, growth parameters, and outcome data of the children habitually seen by
approximately 140 family paediatricians distributed throughout I taly.
Pedianet can link to other databases using unique patient identifiers. In the first database,
information on routine childhood vaccination is captured, including vaccine brand and dose.
In the second database, information on patient hospitalisation date, reason for hospitalis ation,
days of hospitalisations, and discharge diagnosis (up to six diagnoses) is captured. The
family paediatricians’ participation in the database is voluntary , and individuals and their
parents provide consent for use of their data for research purposes. In Italy, each child is
assigned to a famil y paediatrician, who is the referral for an y health visit or any drug
prescription; thus, the database contains a very detailed perso nal medical history . The data,
generated during routine practice care using common software (JuniorBit®), are anony mised
and sent monthly to a centralised database in Padua, I taly, for validation. The Pedianet
database can be linked to regional vaccination data, which was successfully tested in several
large European projects (e.g., ADVANCE, GRI P, EMIF, EU Alliance) where it was
characterised and deemed fit for purpose to evaluate prescriptions including paediatric
routine vaccines [12].
Children aged younger than 12 years will likely start receiving the Pfizer -BioNTech COVID -
19 vaccine soon. This vaccine is expected to be the first COVID -19 vaccine rolled out among
children, and most children in I taly are likely to receive it. We will be able to capture most of
these individuals in Pedianet, as it is expected that approximately 10,000 vaccinated children
aged 12 to 14 years will have data available in Pedianet.
The HSD, an Italian general practice data source in place since 1998, comprises data from
computer -based patient records registered b y a selected group of GPs uniformly distributed
throughout Ital y. The individuals in the database are representative of th e entire I talian
population. I n HSD, patient demographic details are linked through an encry pted patient code
with medical records ( e.g., diagnoses, tests performed, test results, hospital admissions ),drug
prescription information (trade name, dosage form , ATC code, ministerial code, active
substance, date of filled prescription, number of day s’ supply ), risk factors and determinants
of health (blood pressure, body mass index, smoking habits), and date of death. Diseases are
classified according to ICD -9-CM. Ambulatory procedures are encoded in accordance with
the Nomenclatore Tariffario, a list of all outpatient specialist medical services and related
tariffs, instituted by Ministerial Decree in 1996. Currentl y, almost 900 GPs are caring for
approximately 1million individuals (almost 20% of whom are aged younger than 19years).
Pedianet will have an individual patient linkage with the Immunization administrative
database ,which will allow to have all the required information.
Pedianet/HSD acknowledges that the data source they have access to includes data on
vaccine delivery and registration and undertakes to cooperate on addressing the study
objectives by contributing to providing reports based on such data.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950269
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 419.4.3.1. Vaccine Exposure
Using data from the European Ce nter for Disease control, as of 08 May 2021 , 17,801,550
doses of the Pfizer -BioNTech COVID -19 vaccine have been distributed among 60.36 million
Italians[13]. If the distribution of the vaccine were uniform across Italian regions, it is
expected that 295,000 doses would have been administered in HSD .
In Italy, the COVID -19 vaccination campaign started in December 2020. Every region
(n=20) ha sadopted different vaccination strategies involving hubs and/or general practices.
The primary care setting was activel y involved in the vaccination campaign only at the
beginning of April 2021 ,and only certain age categories and/or t ypesof vaccines were
available for direct administration by GPs. Thus, for the period between January and March
2021, I talian GPs have likely recorded vaccine injections according to three main pathway s:
a) some regions automatically informed GPs regarding their patient s’COV ID-19
vaccination; b) GPs refer redpatient s to a specific hub and register edtheirvaccination status
there ; and c) patients autonomously reported their vaccination to their GPs. For the first
semester of 2021, HSD expects to find complete data for certain age categories ,while for the
first trimester and for some other age categories ,we could find only incomplete data for
some regions. In HSD, after preliminary evaluation of data completeness, the study design
(e.g.,self-controlled or cohort design) will be chosen for the specific objectives.
9.4.4. EpiChron – Aragon data sources (ES) (1.3million active individuals)
The Spanish National Health Sy stem is organised at a regional level. Aragon is one of the
regions, with approximately 1.3million inhabitants. The Aragon data sources to be used in
this project, which cover approximately 98% of the reference population, are the following:
The user database (BDU) with s ociodemographic information
Individuals ’ electronic medical records from primary care (OMI -AP) and hospital care
(Minimum Basic Data Set, CMBD, with data on hospital discharges, and PCH database
with data on visits to the emergency room )
Individuals ’ pharm acotherapeutic history with prescriptions and dispensation of drugs in
community pharmacies (Receta Electrónica database) and hospitals (for hospitalised
patients and outpatients )
Aragon COVID -19 Registry
Furthermore, additional databases and registers at the local (i.e., hospital or primary care
health care centre) and national (e.g., Base de Datos para la Investigación
Farmacoepidemiológica en Atención Prim ària [BIFAP ]database and its CIAP dictionary ;
SINASP ) level, as well as new potential databases or registers that could be developed for the
vaccination process during the project, will be explored and used if appropriate. All the
information contained in these data sources is linked at the patient level th rough a
pseudony misation process and then anony mised for research purposes. The group ’s
researchers have broad experience in the use of these databases for research on chronic
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950270
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 42diseases and COVID -19. From the Aragon Health Department, the following key persons
will be directly involved in the project: Antonio Poncel -Falcó, Maria Isabel Cano del Pozo,
Cristina Navarro Pemán, Ana Cristina Bandrés Liso, Mercedes Aza Pascual Salcedo, and
Francisca González Rubio. The group developed the EpiChron Cohort Study [14]for the
analysis of the clinical epidemiology of chro nic diseases, multimorbidity , and poly pharmacy
using real -world data from some of the aforementioned databases during the period
2010 -2020; more than 30 scientific publications have resulted from this study . The group has
also developed the PRECOVID stud y [15], for the demographic and clinical characterisation
of all COVID -19 patients in the Aragon region and for identifying variables associated with
increased mortality risk. Diagnoses are coded initially according to the ICPC or I CD and are
subsequently grouped into diagnostic clusters, if needed, using open software (i.e., Clinical
Classifications Software). Drug prescriptions and dispensations are coded according to the
WHO ATC classification sy stem. Once the aforementioned data sources have been gathered
and linked at the patient level, data undergo continuous quality control checks that ensure
their accuracy and reliability for research purposes.
Information on pregnancy , pregnancy outcomes, and mother -baby linkage from women who
give birth in at least the two most relevant public hospitals in the Aragon region in which
approximately 70% of births in the region occur is expected to be available in the EpiChron
database. The mother -baby linkage is possible using the Neosoft software at the hospital
level, in which all information about the mother and baby is recorded. This information will
be completed using information from the mother’s electronic health records.
EpiChron acknowledges that the data source they have access to includes data on vaccine
delivery and registration and undertakes to cooperate on addressing the stu dy objectives by
contributing to providing reports based on such data.
9.4.4.1. Vaccine exposure
Approximately 0.25million first doses and 0.1 4million second doses of the
Pfizer -BioNTech COVID-19 vaccine have been administered in the Aragon region from
27Decembe r 2020 to 30 April 2021. During May 2021, Aragon will receive a 0.25 million
dose batch, although the number of individuals who may potentiall y receive at least one dose
of the vaccine in Aragon during the stud y period is currentl y difficult to estimate an d will
depend on the availability of the vaccine in Spain in later stages.
9.4.5. Clinical Practice Research Datalink and Hospital Episode Statistics (UK)
(16million active individuals)
The CPRD from the UK collates the computerised medical records of GPs in the UK who act
as the gatekeepers of health care and maintain patients’ life- long EHRs. Accordingl y, GPs
are responsible for primary health care and specialist referrals, and they also store
information about specialist referrals and hospitalisations. General practitioners act as the
first point of contact for any non- emergency health -related issues, which may then be
managed within primary care and/or referred to secondary care, as necessary . Secondary care
teams also provide inf ormation to GPs about their patients, including key diagnoses. The data
recorded in the CPRD include demographic information, prescription details, clinical events,
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950271
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 43preventive care, specialist referrals, hospital admissions, and major outcomes, including
death. Most of the data are coded using Read or SNOMED codes. Data validation with
original records (specialist letters) is also available.
The data set is generalisable to the UK population based on age, sex, socioeconomic class,
and national geographic co verage when CPRD GOLD (General Practitioner Online
Database) and CPRD Aurum versions are used.
There are currently approximately 59million individuals (acceptable for research
purposes) —16million of whom are active (i.e., still alive and registered with the GP
practice) —in over 2,000 primary care practices (https://cprd.com/Data). Data include
demographics, all GP/health care professional consultations (e.g., phone calls, letters, email s,
in surgery , at home), diagnoses and s ymptoms, laboratory test resul ts, treatments (including
all prescriptions), all data referrals to other care providers, hospital discharge summary (date
and Read/SNOMED codes), hospital clinic summary , preventive treatment and
immunisations, and death (date and cause). For a proportion of the CPRD panel practices
(>80%), the GPs have agreed to permit the CPRD to link at the patient level to HES data.
The CPRD is listed under the ENCePP resources database, and access will be provided by
the Drug Safet y Research Unit (DSRU). The CPRD was not y et characterised in the
ADVANCE project, for which the UK THIN and RCGP databases were used, but has been
largel y used in vaccine studies.
The HES database contains details of all admissions to NHS hospitals in England (Accident
& Emergency , Admitted Patient Care, Outpatients); approximately 44.6 million individuals
in the CPRD are linked to the HES database. Not all patients in the CPRD have linked data
(e.g., if they live outside England, if their GP has not agreed that their data should be used in
this way ). As with standard CPRD patients, HES data are limited to patients who are research
standard. CPRD records are linked to the HES using a combination of the patient ’s NHS
number, sex, and date of birth [16]. Additional CPRD -linked data sets include Death
Registration data from the ONS, which includes information on the official date and causes
of death (using ICD codes), m other -baby link, and an algorithm -based Pregnancy Register.
In addition, other CPRD -linked COVID -19 data sets, which may provide further follow -up
information on AESI , include the PHE Second Generation Surveillance Sy stem (SGSS)
COVID -19 positive virology test data, PHE COVID -19 Hospitalisation in England
Surveil lance Sy stem (CHESS), and the I CNARC data on COVID -19 intensive care
admissions.
The mother -baby link (which uses a probabilistic algorithm, based on data in the primary
care medical records) and the Pregnancy Register are linked data sets available with the
CPRD GOL D database. For patients identified in the CPRD Aurum database, the mother -
baby link and Pregnancy Register information is not available. However, information on
pregnancy status and pregnancy outcomes is likely to be available in both CPRD databases as
events reported by the GP in the primary care medical records.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950272
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 44The DSRU acknowledges that the CPRD data sources they have access to include data on
vaccine delivery and registration and undertakes to cooperate on addressing the study
objectives by contributing to providing reports based on such data .
9.4.5.1. Vaccine exposure
Approximately 1.2million patients were identified in the CPRD Aurum database (March
2021 database release) who had received at least one dose of the Pfizer -BioNTech
COVID -19 vaccine. Based on the current UK COVID- 19 vaccine delivery strategy , with
younger age groups more likely to receive the Pfizer -BioNTech COVID -19vaccine, we
would estimate the ability to identify approximately a further 2 to 3million patients in the
CPRD databases who may potentiall y receive at least one dose of the Pfizer -BioNTech
COVID -19 vaccine during the study period. It is expected that, in the near future , the CPRD
GOLD database will be able to contribute to the overall CPRD study , although it is not
possible to estimate the number of patients that will be suitable for participation in the study .
9.4.6. Norwegian health registers (NO) (5.3million active individuals)
The Norwegian data sources included inthis project ,accessed through a partnership with the
University of Oslo are several national health registers, i.e., the Medical Birth Registry of
Norway (MBRN), the National Patient Register (NPR), Norway Control and Pay ment of
Health Reimbursement (KUHR), the Norwegian Immunisation Registry (SYS VAK), the
National Prescription Registry (NoPD), and Statistics Norway (SSB).
The source population will be identified using the Norwegian Institute of Health’s (NIPH)
copy of the Norwegian population data file from the National Registry . The NPR and KUHR
(and the MBRN for the pregnant population) provide data on inpatient and outpatient
diagnostic codes. Information on population background data is derived from SSB
(e.g., education, occupation status, sex, age). Data on vaccination status are derived from
SYSVAK and the Norwegian Prescription Database. The latter register includes data on
filled prescriptions for possible co -medications and other prescription drug use.
9.4.6.1. Norwegian Immunisation Registry
The SYSVAK is the national electronic immunisation regi ster that records an individual’s
vaccination status and vaccination coverage in Norway . It became nationwide in 1995 and
includes information such as personal identit y number, the vaccine code, disease vaccinated
against, and vaccination date.
9.4.6.2. The Norwegi an Patient Registry
The NPR is an administrative database of records reported b y all government -owned
hospitals and outpatient clinics and by all private health clinics that receive governmental
reimbursement. The NPR contains information on admission to h ospitals and specialist
health care on an individual level from 2008. The data include date of admission and
discharge as well as primary and secondary diagnosis. The NPR has included Norwegian
national identification numbers since 2008. Consequently , person-specific data from 2008
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950273
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 45onwards are available. Diagnostic codes in the NPR follow theInternational Classification of
Diseases, 10th Revision (ICD-10).
9.4.6.3. Norway Control and Payment of Health Reimbursement
The KUHR is an administrative database based on el ectronically submitted reimbursement
claims from phy sicians to the Norwegian Health Economics Administration (HELFO). It
contains information from primary health care, GP, and emergency services on morbidity ,
utilisation of health care services, and health care use. Person -specific data are available for
the y ears2010 through 2018. Diagnostic codes in the KUHR follow ICD -10, but the I CPC is
more frequently used by GPs.
9.4.6.4. The Norwegian Prescription Database
Since January 2004, all pharmacies in Norway have be en obliged to send data electronicall y
to the Norwegian Institute of Public Health regarding all prescribed drugs (irrespective of
reimbursement) dispensed to individuals in ambulatory care. Relevant variables for this
project include detailed information on drugs dispensed and date of dispensing.
9.4.6.5. The Medical Birth Registry of Norway
The MBRN is a population- based register containing information on all births in Norway
since 1967 (more than 2.3 million births) . The MBRN is based on mandatory notification of
all births or late abortions occurring at 12 weeks of gestation onwards. The MBRN includes
identification of the mother and father, including national identification numbers, parental
demographic information, the moth er’s health before and during pregnancy , complications
during pregnancy and delivery , and length of pregnancy ,as well as information on the infant,
including congenital malformations and other perinatal outcomes.
9.4.6.6. Statistics Norway
Statistics Norway provid es microdata for research projects and includes information on
population characteristics, housing conditions, education, income, and welfare benefits.
These data are potential important confounders.
9.4.6.7. The National Registry
The National Registry (Folkeregist eret) holds information about all inhabitants in Norway .
The NIPH holds a cop y of the Norwegian population data file from the National Registry that
will be used to identify the source population in Norway .
9.4.6.8. Norwegian Surveillance System for Communicable Di seases
Notification of infectious diseases to the Norwegian Surveillance Sy stem for Communicable
Diseases (MSIS) is an important part in the surveillance of infectious diseases in Norway .
Microbiological laboratories anal ysing specimens from humans, and all doctors in
Norway ,are required by law to notify cases of certain diseases (71 in total ,including
SARS -CoV -2) to the MSIS central unit at the Norwegian Institute of Public Health. The
following variables are available since 1977: notifiable disease, mon th and y ear of diagnosis,
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950274
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 46age groups, count y of residence, and place of infection. Data on positive COVID -19 tests are
updated continuously .
The University of Oslo acknowledges that the data source sthey have access to (described
above) include data on vac cine delivery and registration and undertakes to cooperate on
addressing the s tudy objectives by contributing toproviding reports based on such data.
9.4.6.9. Vaccine Exposure
According to the Vaccination calendar (version 30 April 2021), in Norway, 7,280,000 doses
of the Pfizer -BioNTech COVID-19 vaccine will be distributed by the end of September
2021 .
9.4.7. SIDIAP (ES) (5.7million active individuals)
The Information Sy stem for the Improvement of Research in Primary Care (Sistema
d’Informació per al Desenvolupament de la Investigació en Atenció Primària’ [SI DIAP]) was
created in 2010 by the Catalan Health Institute and the IDIAPJGol I nstitute. I t includes
information collected since 01 January 2006 during routine visits at 278 primary care centres
pertaining to the Catalan Health Institute in Catalonia (North- East Spain) with
3,414 participating GPs. SI DIAP has pseudo -anony mised records for 5.7 million people
(80% of the Catalan population) and is highl y representative of the Catalan population.
The SIDIAP data compr ise the clinical and referral events registered by primary care health
professionals (e.g., GPs, paediatricians, and nurses) and administrative staff in electronic
medical records, comprehensive demographic information, community pharmacy invoicing
data, s pecialist referrals, and primary care laboratory test results. The SIDIAP data can also
be linked to other data sources, such as the hospital discharge database, on a
project -by- project basis. Health professionals gather this information using ICD- 10 code s,
ATC codes, and structured forms designed for the collection of variables relevant for primary
care clinical management, such as country of origin, sex, age, height, weight, body mass
index, tobacco and alcohol use, blood pressure measurements, and blood and urine test
results. Regarding vaccinations, SI DIAP includes all routine childhood and adult
immunisations, including the antigen and the number of administered doses. Encoding
personal and clinic identifiers ensures the confidentiality of the informat ion in the SI DIAP
database. The SIDIAP database is updated annually at the start of each year.
Currently , with the COVID -19 pandemic, there is the possibility to have shorter term updates
in order to monitor the evolution of the pandemic. Recent reports have shown the SIDIAP
data to be useful for epidemiological research. SIDIAP is listed under the ENCePP resources
database (www.encepp.eu/encepp/resourcesDatabase.jsp ).The SIDIAP database was
characte rised in the ADVANCE project and considered fit for purpose for vaccine coverage,
benefits, and risk assessment [12].
Information on pregnancy , pregnancy outcomes, and mother -baby linkage will be available
in the SI DIAP database .
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950275
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 47SIDIAP acknowledges that the data source they have access to includes data on vaccine
delivery and registration and undertakes to cooperate on addressing the study objectives by
contributing to providing reports based on such data.
9.4.7.1. Vaccine exposure
Appro ximately 1.7million doses of the Pfizer -BioNTech COVID -19 vaccine have been
administered in Catalonia from 27 December 2020 through 28 April 2021
(https://dadescovid.cat/) . Based on the National Strategic Vaccination Plan, 600million
doses of the Pfizer -BioNTech COVID -19 vaccine are expected to be administered in Spain .
The number of doses forCatalonia will depend on the availability of the vaccine in later
stages.
9.5.Study size
The study will be conducted in a source population of 38.9 million individuals captured in
the electronic healthcare data sources.
Table 2 shows the sample size calculations for AESI and different risk ratios assumed . As
examples, a ssuming a two- sided alpha =0.95, power of 80%, and a ratio of 1 to 4 exposed to
unexposed, to dete ct a risk ratio of 3 for vaccine -associated enhanced disease (VAED), we
will need to include 1,209 exposed individuals and 4,836 unexposed individuals ; and
assuming a two -sided alpha =0.95, power of 80%, and a ratio of 1 to 4 exposed to
unexposed, to dete ct a risk ratio of 2 for Guillain -Barré syndrome , we will need to include
1,700,582 exposed individuals and 6,802,328 unexposed individuals.
Table 2.Number of individuals needed to detect different risk ratios for select
AESI awith a range of background rates
Sample size
DiseaseBackground
proportion
during risk
windowRisk
ratioExposed Unexposed
Anaphylaxis 1/40,000 1.5 2,147,100 8,588,400
Anaphylaxis 1/40,000 2 680,233 2,720,932
Anaphylaxis 1/40,000 2.5 365,976 1,463,904
Anaphylaxis 1/40,000 3 241,643 966,572
Guillain -Barré syndrome 1/100,000 1.5 5,367,750 21,471,000
Guillain -Barré syndrome 1/100,000 2 1,700,582 6,802,328
Guillain -Barré syndrome 1/100,000 2.5 914,940 3,659,760
Guillain -Barré syndrome 1/100,000 3 604,106 2,416,424
VAED 1/5,000 1.5 10,736 42,944
VAED 1/5,000 2 3,402 13,608
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950276
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 48Table 2.Number of individuals needed to detect different risk ratios for select
AESI awith a range of background rates
Sample size
DiseaseBackground
proportion
during risk
windowRisk
ratioExposed Unexposed
VAED 1/5,000 2.5 1,830 7,320
VAED 1/5,000 3 1,209 4,836
AESI =adverse events of special interest ; VAED =vaccine -associated enhanced disease.
a. As suming a tw o-sided alpha =0.95, pow er of 80%, and a ratio of 1 :4 exposed to unexposed .
9.6.Data management
This study will be conducted in a distributed manner using a common protocol, common data
model (CDM), and common analy tics programmes based on existing health data. The
following steps will be implemented:
1. Extraction, transformation, and loading (ETL) of data to a CDM. To harmonise the
structure of the data sets stored and maintained b y each data partner, a shared syntactic
foundation i s used. The CDM that will be used has been developed during the
IMI-ConcePTION project ( https://www.imi -conception.eu/wp -
content/uploads/2020/10/ConcePTION- D7.5 -Report -on-existing -common -data-models-
and-proposals -for-ConcePTI ON.pdf ). In this CDM, data are represented in a common
structure, but the content of the data remain in their original format. The ETL design for
each stud y is shared in a searchable FAIR catalogue. The VAC4EU FAIR data catalogue
is a meta- data management tool designed to contain search able meta -data describing
organisations that can provide access to specific data sources. FAI R is defined as
findable, accessible, interoperable, and re -usable. Data qualit y checks will be conducted
to measure the integrit y of the ETL as well as internal c onsistency within the context of
the CDM (see Section 9.8).
2. Second, to reconcile differences across terminologies, a shared semantic foundation is
built for the definition of events under stud y by collecting relevant concepts in a
structured fashion using a standardised event definition template. The Codemapper too l
was used to create diagnosis code lists based on completed event definition templates for
each AESI and comorbid risk condition in the ACCESS project. Based on the relevant
diagnostic medical codes and key words, as well as other relevant concepts
(e.g., medications), one or more algorithms are constructed (t ypicall y one sensitive, or
broad, algorithm and one specific, or narrow, algorithm) to operationalise the
identification and measurement of each event. These algorithms may differ by database,
as the c omponents involved in the study variables may differ. Manual review of
electronic records will be conducted for a sample of the events. Specifications for both
ETL and semantic harmonisation will be shared in the catalogue.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950277
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2.0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 493.Third, following conversion to h armonised study variable sets, R and SAS programs for
the calculation of incidence and prevalence will be distributed to data access providers
for local deplo yment. The aggregated results produced b y these scripts will then be
uploaded to the Digital Resea rch Environment (DRE) for pooled analy sis and
visualisation (see Figure 3). The DRE is made available through UMCU (University
Medical Center Utrecht)/VAC4EU ( https://w ww.andrea -consortium.org/ ). The DRE is a
cloud- based, globall y available research environment where data are stored and organised
securel y and where researchers can collaborate ( https://www.andrea -
consortium.org/azure -dre/).
Figure 3.Data management plan
CDM =common data model.
9.6.1. Case report f orms (CRFs)/Data collection t ools (DCTs)/Electronic data r ecord
This study will use secondary data collected in EHR databases. For the purpose of validating
selected stud y endpoints, special forms will be developed and securel y saved in
environments assuring data protection and patient confidentiality according to the
requirements of each country and database access provider
(DAP ).
As used in this protocol, the term CRF should be understood to refer to either a paper form or
an electronic data record or both, depending on the data collection method used in this study .
A CRF is required and
willbe completed for each patient who is subject to an even t/case
verification/validation procedure. The completed original CRFs are the sole propert y of the
DAPs and will not be made available in any form to third parties, except for authorised
representatives of Pfizer or appropriate regulatory authorities . The DAPs will ensure that the
CRFs are securel y stored at the study site in encrypted electronic form and will be password
protected to prevent access by unauthorised third parties.
The DAPs have ultimate responsibility for the collection and reporting of all clinical and
laboratory data entered on the CRFs for the procedure of event/case verification/validation
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950278
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 50and ensuring that they are accurate, authentic/original, attributable, complete, consistent,
legible, timel y (contemporaneous), enduring, and available when required. The CRFs must
be signed b y an authorised staff member of the DAPs to attest that the data contained on the
CRFs are true. An y corrections to entries made in the CRFs or source documents must be
dated, initialled , and explained (if necessary ) and should not obscure the original entry .
The source documents are the hospital’s or physician’s charts. In these cases, data collected
on the CRFs must match those charts.
9.6.2. Record retention
The investigators must obtain Pfizer’s written permission befor e disposing of an y records,
even if retention requirements have been met.
The final study aggregated results sets and statistical programmes will be archived and stored
on the DRE and the VAC4EU SharePoint site. Validation of the quality control (QC) of the
statistical analy sis will be documented. The final study protocol and possible amendments,
the final statistical report, statistical programmes, and output files will be archived on a
specific and secured central drive.
It is the responsibility of the principal investigator to inform the other investigators or
institutions regarding when these documents no longer need to be retained. Study records or
documents may also include the anal yses files, syntaxes (usually stored at the site of the
database), ET L specifications, and output of data quality checks.
To enable evaluations and/or inspections/audits from regulatory authorities or Pfizer, DAPs
agree to keep all study -related records, including the identity of all participating patients
(sufficient infor mation to link records, e.g., CRFs, hospital records), copies of all CRFs,
safet y reporting forms, source documents, detailed records of treatment disposition, and
adequate documentation of relevant correspondence (e.g., letters, meeting minutes, telephone
call reports). The records should be retained b y DAPs according to local regulations or as
specified in the vendor contract, whichever has a longer retention time. DAPs must ensure
that the records continue to be stored securel y for aslong as they are re tained.
If UMCU becomes unable for an y reason to continue retaining study records for the required
period, Pfizer should be prospectivel y notified. In this case, the study records must be
transferred to a designee acceptable to Pfizer.
Study records must be kept for a minimum of 15 years after completion or discontinuation of
the study , unless UMCU and Penta and Pfizer have expressly agreed to a different retention
via a separate written agreement. Records must be retained for longer than 15 years if
requir ed by applicable local regulations.
UMCU must obtain Pfizer’s written permission before disposing of any records, even if
retention requirements have been met.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950279
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 519.6.3. Data extraction
Each database access provider (DAP) will create ETL specifications using the standard
ConcePTION ETL design template (accessible via this link:
https://docs.google.com/document/d/1SWi31tnNJL7u5jJ LbBHmoZa7AvfcVaqX7jiXgL9uA
Wg/edit ). Following completion of this template and review b y stud y statisticians, eac h DAP
will extract the relevant study data locall y using their software (e.g., Stata, SAS, R, Oracle).
These data will be loaded into the CDM structure in csv format. These data remain local (see
Figure 3).
9.6.4. Data processing and transformation
Data processing and transformation will be conducted using R and SAS code against the
syntactically harmonised CDM. The R and SAS scripts will first transform the data in the
syntactically harmonised CDM to semantically harmonised study variables (see Figure 3).
Following creation o f study variables, the data will be characterised. This characterisation
will include calculation of code counts and incidence rates, as well as benchmarking within
the data source (over time), between data sources and externally (against published
estimat es). Subsequently , R and SAS code to conduct anal ysis against semantically
harmonised study variables will be distributed and run locally to produce aggregated results.
The R and SAS scripts for these processing and analysis steps will be developed and tes ted
centrall y and sent to the DAPs.
The R and SAS scripts are structured in modular form to ensure transparency . Functions to
be used in the modules will be either standard R and SAS packages or packages specificall y
designed, developed, and tested for mul tidatabase studies. Scripts will be double coded in
SAS and R and quality checks will be thoroughl y documented.
The DAPs will run the R and SAS code locall y and send aggregated anal ysis results to the
DRE using a secure file transfer protocol. In the DRE, results will be further plotted,
inspected (for quality assessment), and pooled (if needed) for final reporting.
All final statistical computations will be performed on the DRE using R and/or SAS (SAS
Institute; Cary , North Carolina). Data access providers will have access to the workspace for
script verification.
Aggregated results, ETL specifications, and a repository of study scripts will be stored in the
DRE.
9.6.5. Data access
Within the DRE, each project- specific area consists of a separate secure folder cal led a
“workspace ”.Each workspace is completel y secure, and researchers are in full control of
their data. Each workspace has its own list of users, which can be managed by its
administrators.
The DRE architecture allows researchers to use a solution withi n the boundaries of data
management rules and regulations. Although General Data Protection Regulation and Good
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950280
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 52(Clinical) Research Practice still apply to researchers, the DRE offers tools to more easil y
control and monitor which activities take place wit hin projects.
All researchers who need access to the DRE are granted access to study -specific secure
workspaces. Access to this workspace is onl y possible with double authentication using an
identification code and password together with the user’s mobile phone for authentication.
Upload of files is possible for all researchers with access to the workspace within the DRE.
Download of files is onl y possible after requesting and receiving permission from a
workspace member with an “owner” role.
9.7.Data analysis
Detailed methodology for summary and statistical anal yses of data collected in this study will
be documented in a SAP, which will be dated, filed, and maintained b y the sponsor. The SAP
will also provide additional detail regarding the evaluation of a thre shold of excess risk for
each of the safety events of interest. This will be determined based on background incidences
for each event (e.g., based on a historical influenza vaccinated active comparator cohort data
to be determined during the study ), in addition to prespecified significance level
(e.g., alpha =0.01 or 0.05) and power. All anal yses will be conducted using R version
R-4.0.3 or higher (Foundation for Statistical Computing, Vienna, Austria; https://www.R -
project.org) or SAS version 9.3 software or higher (Cary , North Carolina, United States of
America (USA ); SAS I nstitute, I nc.).
The SAP will contain more detail of the anal ysis and data pooling and may modify the plans
outlined in the protocol; any major modifications of primary endpoint definit ions or their
analyses would be reflected in a protocol amendment.
Data extraction ,descriptive analysesof AESI that are not pregnancy outcomes ,incidence
rates ,and comparative analy ses (when the data exist and are available to support those
analyses with sufficient precision) are planned to occur every 6months during the first
2years of the stud yand will be reported in the interim and the final reports. An additional
data extraction and analysis will take place at the end of year 3 of follow -up and will include
all outcomes, including pregnancy outcomes. Pfizer -BioNTech proposes to include in the
interim reports descriptive results .
9.7.1. Cohort design
9.7.1.1. Exposure assignment and follow -up
The main exposure of interest is in the receipt of at least one dose of the Pfizer -BioNTech
COVID -19 vaccine. Other exposure groups will also be described. Individuals will be
assigned to each vaccination category (see Section 9.1.1 ) at time zero ,as outlined below .
A.Vaccination category “Receipt of a first dose of the Pfizer -BioNTech COVID -19 vaccine,
followed or not b y a second dose”: Individuals will be assigned to this exposure categ ory
if they receive a first dose of the Pfizer -BioNTech COVID- 19 vaccine. Individuals will
be censored if and when they receive a non– Pfizer COVID -19 vaccine during follow -up.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950281
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 53For the matched comparative anal ysis individuals will be censored when their mat ched
pair is censored.
B.No vaccination category . Individuals will be assigned to this exposure group if they do
not receive a vaccination at time zero (see Section 9.1.1 ).Individuals will be censored
when they receive a dose of any COVID -19–directed vaccine during follow -up. Time
zero in the unexposed group will be a day when they did not receive a Pfizer -BioNTech
COVID -19 vaccine dose. This day will be chosen by calendar matching to the time zero
of the corresponding exposed group.
Censoring will only apply to the sensitivity anal ysis for a vaccination category consisting of
the receipt of the two vaccination doses per the recommended schedule. For the sensitivity
analysis study ing the r eceipt of a full vaccination regimen of the Pfizer -BioNTech
COVID -19 vaccin e, individuals will be assigned to this exposure categor y if they receive a
first dose of the Pfizer -BioNTech COVID -19 vaccine. Individuals will be censored if they do
not receive a second dose of the Pfizer-BioNTech COVID- 19 vaccine b y week 4 after the
first administration, in the absence of an adverse event that contraindicates the second dose.
Individuals will also be censored if the second dose of Pfizer -BioNTech COVID- 19 vaccine
is received within 2 weeks of the first dose or if they receive a non– Pfizer -BioNTech
COVID -19 vaccine during follow -up.
Individuals will be followed from time zero (see Section 9.1.1 ) until the censoring described
above, death, or the administrative end of follow -up, whichever occurs first. For anal yses of
AESI with known risk windows, follow- up will be truncated at the end of the risk window.
9.7.1.2. Descriptive statistics
The distributions of baseline characteristics at time zero by exposure group will be calculated
to describe the stud y cohort and illustrate differences between the groups. For continuous
variables, means, standa rd deviations, medians, and other quartiles will be estimated. For
categorical variables, counts and proportions will be estimated. The missingness of variables
will also be described. Interim reports will be limited to descriptive analy ses. Comparative
analyses will be included in the final report. Further details will be described in the SAP.
To describe the relative imbalance of characteristics between expos ed and unexposed groups,
absolute standardised differences will be calculated for each baseline ch aracteristic [15, 16].
Multilevel categorical variables will calculate an overall standardised difference ac ross all
levels [18].The larger the absolute standardised difference values, the greater the imbalance
between baseline characteristics. Balance will also be checked after propensity score methods
are applied to control for confounding.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950282
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 549.7.1.3. Description of vaccination categories
Thecounts and proportions of administered doses with the following characteristics will be
reported:
Receipt of a second dose of the Pfizer-BioNTech COVID- 19 vaccine outside the
Pfizer -BioNTech recommended COVID -19 vaccination schedule
Receipt of a dose of the Pfizer -BioNTech COVID-19 vaccine after receipt of a dose of a
different vaccine
Receipt of the Pfizer -BioNTech COVID- 19 vaccine in individuals with contraindications
to the vaccine or in groups not recommended to receive vaccination (i.e.,adults who are
not health care workers in January 2021)
To characterise utilis ation patterns of Pfizer -BioNTech COVID -19 vaccine, the absolute
and relative frequency of individuals receiving at least one dose of the vaccine and the
two-dose vaccine completion rate will be calculated. The distribution of time gaps in
weeks between th e first and second dose will be described by categories (< 2, 2-4, 5-8,
9-12, 13 -18, > 18 weeks) and by median, other quantiles, and minimum and maximum.
9.7.1.4. Crude outcome measures
For safet y outcomes with a known short risk window (e.g., anaphy laxis), the ris k (number of
events/number of vaccinated persons) and the corresponding 95% CIs will be computed
baseline . Effect estimates will be calculated both as risk differences and as risk ratios, along
with their corresponding 95% CIs.
For safet y outcomes with unknown risk windows or those that require long follow -up
(e.g., death), the cumulative incidence will be computed, which will be estimated with a
1 −Kaplan -Meier survival curve, as well as with adjusted parametric incidence curves [19].
Time to outcome will be defined as the time from the baseline date (time zero) until the
occurrence of the outcome or censoring. For individuals without outcomes, the censoring
date is defined as the earlier of date of death , censoring of the match pair and the end of
follow -up (as described in Section 9.7).The variance will be computed using approaches that
account for autocorrelation (e.g .,the robust estimator or via bootstrapping ). Risk differences
and risk ratios (and their corresponding 95% CIs) will be estimated at different time
intervals, which can be ad apted to the specific nature of each outcome.
Crude risks, cumulative incidence, and measures of association for each AESI after
vaccination will be estimated in the entire population. Subgroup anal yses will be conducted
by subgroups defined by demographic and clinical characteristics, as well as other covariates
of interest. In addition, risks or cumulative incidence of AESI in persons vaccinated with
Pfizer -BioNTech COVID- 19 vaccine will be compared with the expected rates in the absence
of vaccination, w hich will be calculated using unexposed and historical data (i.e., from the
ACCESS project).
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950283
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 559.7.1.5. Adjustment for baseline imbalances
Individuals following each vaccination category under study may have different
characteristics that may determine their risk of AESI . To account for such potential
confounding, propensit y score methods will be used to estimate the adjusted risk ratios and
95% CIs. Propensity scores represent the probability of being vaccinated at any calendar time
given a set of baseline covariates . More details will be provided in the SAP.
9.7.1.6. Adjustment for adherence to recommended vaccination schedule
The censoring described in Section 9.7.1.1 may introduce selection bias if individuals
receiving a non–Pfizer -BioNTech COVID -19 vaccine during the follow- up (the main reason
for being censored) are different from those who do not; thes e differences may be present at
baseline (e.g., age, sex) or may become apparent during follow -up (e.g., a reaction after a
first dose). To account for such potential selection bias, we will use weighting b y inverse
probability of censoring. More details w ill be specified in the SAP. This population with
censoring will be described and compared with the population in the main anal ysis.
9.7.1.7. Meta -analysis
Using the main estimates from each data source, appropriate random -effects meta -analytic
methods will be used to obtain a combined effect estimate. The heterogeneity across data
sources will be checked, and a forest plot will be produced with the data sources and the
pooled estimate.
9.7.2. Self-controlled risk interval
9.7.2.1. Descriptive statistics
The number of cases and incidence rates of each AESI will be reported, overall and b y
important covariates.
9.7.2.2. Measures of association
Conditional Poisson regression will be used to estimate incidence rate ratios and 95% CIs,
and risk differences and 95% CIs will be estimated u sing an appropriate method. The
primary anal ysis will be pooled across both doses, and the secondary anal ysis will stratify by
dose number to assess potential effect modification. AESI for which the SCRI design will be
a complementary design and risk windows for such AESI are describe din Table 1. The
control period will
follow the risk window , will be of the same lengths
,andwill be detailed
in the SAP.
The SCRI inherently adjusts for both measured and unmeasured time constant factor s such as
sex and chronic health conditions with onset before the start of follow -up. Time -varying
confounders may be included as covariates in regression models.
Subgroup anal yses will be conducted b y subgroups defined b ydemographic and clinical
characte ristics, dose number, and other covariates of interest.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950284
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 569.8.Quality control
Rigorous QC will be applied to all deliverables. Data transformation into the CDM will be
conducted b y each subcontracted research partner in its associated database, with processes
asdescribed in the following corresponding sections. Standard operating procedures or
internal process guidance at each research centre will be used to guide the conduct of the
study . These procedures include rules for secure and confidential data storage, backup, and
recovery ; methods to maintain and archive project documents; QC procedures for
programming; standards for writing anal ysis plans; and requirements for scientific review by
senior staff.
At UMCU, as the scientific coordinating centre responsible for central data management and
analysis, all documents undergo QC review and senior scientific review. Data management
and statistical anal ysis follow standard operating procedures. All statistical anal ysis
program meswill be double coded.
At RTI Health Solutions (RTI -HS), as the project coordinating centre and scientific coleader
centre , all key study documents will undergo QC review, senior scientific review, and
editorial review. Senior reviewers with expertise in the appropriate subject matter area will
provide advice on the design of research stud y approaches and the conduct of the study and
will review results, reports, and other key study documents.
9.8.1. PHARMO (NL)
PHARMO adheres to high standards throughout the research process based on robust
methodologies, transparency , and scientific independence. PHARMO conducts studies in
accordance with the ENCePP Guide on Methodological Standards in
Pharmacoepidemiology[20]and the ENCePP Code of Conduct[21].PHARMO is I SO
9001:2015 certified. Standard operating procedures, work instructions, and checklists are
used to guide the conduct of a study . These procedures and documents include internal
quality audits, rules for secure and confidential data storage, methods to maintain and archive
project documents, rules and procedures for execution and QC of SAS programming,
standards for writing protocols and reports, and requirements for senior scientific review of
key study documents.
9.8.2. ARS Toscana (IT)
One or two researchers will review study documents. ARS Toscana receives data on a
bimonthly basis from the Tuscany region (where it u ndergoes a first QC); the ARS Toscana
statistical office appends it to an Oracle database and checks it using a dashboard to identify
any inconsistencies with historical data.
The Pharmacoepi Unit has standardised parametric procedures in Structured Query Language
(SQL) and Stata to extract data from the Oracle database. Parametric procedures are also
available to convert the data into various CDMs. Study -specific procedures are developed,
based on the study protocol and/or SAP, as well as by composing stan dard parametric
procedures in Stata. Standard procedures in R are currentl y under development in the context
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950285
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 57of the ConcePTION project. The Unit also regularly generates simulated data sets and double
programming in R programmes that are originally develop ed in SAS or Stata.
9.8.3. Pedianet/HSD (IT)
Pedianet data processing includes, in addition to standardised procedures in SQL and
Microsoft Access to extract data from database, QC steps aimed at verify ing the
correspondence between a diagnostic code and its open-text descriptor that are conducted
through manual validation of clinical histories. Quality control checks of patient general data
are conducted through the detection of outlier values and validation rules; grouping of
diseases; and regular monitoring of aggregate clinical and drug data. All transformations in
the data are logged in R scripts. To ensure code reliability , double programming in R and
Stata or Py thon is in place for all scripts.
HSD data processing includes, in addition to standardi sedproced ures in SQL and Access to
extract data from adatabase, QC steps aimed at verify ingthe correspondence between a
diagnostic code and its open- text descriptor , which are conducted through manual validation
of clinical histories. Quality control of patient general data is conducted through the detection
of outlier values and validation rules ,grouping of diseases , and regular monitoring of
aggregate clinical and drug data. All transformations in the data are logged in SQL scripts
through version control. Fur thermore, to ensure code reliability ,double programming in Stata
isin place for all scripts.
9.8.4. EpiChron - Aragon data sources (ES)
The data QC process in Aragon is conducted in three steps (i.e., data collection, data request
and extraction, and data proce ssing). Common data collection software and procedures
guarantee standardised data input by all health care professionals. In the case of the hospital
CMBD register and the drug dispensation database, their completion is sy stematic, uniform,
and normative according to legal orders. Online specific training and chart documentation on
the use of EHR software is regularl y provided to physicians and nurses in Aragon. The data
contained in each of the registers is routed to a specific service of the Department of Health,
which performs a pseudony misation of the data to encry pt individual -level identification
codes, protecting individuals ’ privacy and compl ying with data protection laws. This new
encry pted code is applied in all registers, enabling the linkage of data at the patient level. The
resulting databases are stored on a central computer server, and access to the files is restricted
to members of the research group b y a double -entry password. The research group is a
multidisciplinary qualified team includin g public health specialists, epidemiologists,
clinicians, pharmacists, statisticians, and data managers; they are all trained in data
management and patient data protection. Given that original databases are in different
formats (e.g., Microsoft Access, Mi crosoft Excel, plain text), the SQL programming
language is emplo yed to extract the data. Stata statistical software (Release 12) is used for
data processing, which includes a number of s ystematic steps aimed at improving the quality ,
accuracy , and reliabi lity of the data for research purposes (e.g., QC of diagnoses to verify the
correspondence between a diagnostic code and its open- text descriptor through manual
validation of clinical histories and use of specific algorithms to search for specific key word s
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950286
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 58or roots of words in open -text fields, QC of patient general data through the detection of
outlier values and validation rules, grouping of diseases, and regular monitoring of aggregate
clinical and drug data). The original databases also have their own QCprocesses. All changes
conducted in the data are logged in Stata scripts, which are continuously revised and updated
given the d ynamic nature of the data processing.
9.8.5. CPRD (UK)
The DSRU has information securit y policies in place to preserve the confident iality , integrit y
and availability of the organisation’s s ystems and data. These include ensuring that the
premises provide suitable phy sical and environmental securit y, all equipment is secure and
protected against malicious software, the network can be a ccessed onl y by authorised staff,
telecommunication lines to the premises are protected from interception b y being routed
overhead or underground, and personnel receive training regarding securit y awareness. The
study will be conducted according to the Guidelines for Good Pharmacoepidemiology
Practices (GPP)[22]and according to the ENCePP Code of Conduct[21].Data quality is a
high priorit y at the DSRU and is assured through a number of methods based on staff
training, validated s ystems, error prevention, data monitoring, data cleaning, and
documentation, including the following:
Staff training on data processing standard operating procedures
Data management plan for every research study outlining the legal basis for data
collection, data flows, data access rights, data retention periods, etc.
Routine data cleaning to screen for errors, missing values, and extreme values and
diagn ose their cause
System process logs to document staff access, etc.
9.8.6. SIDIAP (ES)
Data qualit y processes are implemented at each phase of the data flow cycle. Quality control
checks are performed at the extraction and uploading steps. To assess data completeness the
elements presence are described b y geographical areas, registering ph ysician, time and the
distribution function of values. Correctness is assessed by valid ity checks on outliers, out of
range values, formatting errors and logical dates incompatibilities. Completeness and
correctness measures are used to inform decisions on the required transformations to improve
data quality (e.g., harmonisation, normalisati on, and clean- up) and the data fitness for the
purpose of specific research projects.
9.9.Limitations of the research methods
This study is subject to limitations related to both the study design and use of secondary
health care data.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950287
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 59A data -related limitation of this study is the reliance on the accuracy of codes and algorithms
to identify outcomes. Outcomes and their dates of occurrence will be validated, but the extent
of validation may be limited because of the use of medical records. Exposure identificatio n
may be based on pharmacy dispensing records, general practice records, immunisation
registers, medical records, or other secondary data sources. T he ability to identify specific
COVID -19 vaccine products and dates of vaccination in these data sources is reflected in
Section 9.3.1 . The vaccine roll- out is just underway in some countries and has y et to begin in
others. I t is possible that individuals v accinated outside the health care s ystem will not be
recorded in secondary EHR databases, thereby leading to potential bias because of exposure
misclassification with the cohort design. Furthermore, it is unknown the extent to which
vaccine brands and batc h numbers/lot numbers will be available in the secondary data
sources. It is also possible that some AESI are the result of immunisation errors occurring
during the administration of the Pfizer-BioNTech COVID- 19 vaccine. This information is
not collected r egularl y and will not be able to be taken into account with the current protocol.
In some databases, the mother -baby link may not be available. This is the case for the CPRD
Aurum database, while the reare Pregnancy Register linked data sets available for the CPRD
GOLD database .It is expected that, in the near future , the CPRD GOLD database will be
able to contribute to the overall CPRD study .
A study design- related limitation of both the cohort and SCRI designs is that any uncertainty
regarding risk periods will lead to misclassification and attenuation of risk estimates. A
limitation of the cohort design is the potential for residual or unmeasured confounding, as it
is unlikely that the data sources wil l have information on all potential confounders. To
address potential confounding, the SCRI , which automatically adjusts for time -invariant
confounders, will be used as a secondary approach. However, the SCRI is not well suited to
study outcomes with gradual onset, long latency , or risk periods that are not well known. I t
also may be subject to bias for outcomes that affect the probability of exposure. The SCRI
design will be complementary to the cohort design for prespecified AESI with defined risk
interva ls.In Ital y, the COVID -19 vaccination campaign started in December 2020. Each
region (n =20) has adopted different vaccination strategies involving hubs and/or general
practices.
The primary care setting was activel y involved in the vaccination campaign only at the
beginning of April 2021 ,and only certain age categories and/or t ype of vaccines were
available for direct administration by GPs. Thus, for the period between January and March
2021, I talian GPs have likely recorded vaccine injections accordin g to three main pathway s:
a) some regions automatically informed GPs regarding their patient s’COVID -19 vaccination
status; b) GPs refer redpatients to a specific hub to register the irvaccination status there ; and
c) patients autonomously reported their v accination to their GPs. For the first semester of
2021, HSD expects to find complete data for certain age categories ,while in the first
trimester and for some other age categories ,we could only find incomplete data for some
regions. In HSD, after prelim inary evaluation of data completeness, the study design (e.g.,
self-controlled or cohort design) will be chosen for the specific objectives.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950288
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 60The main anal ysis for both the cohort and SCRI analy sis will pool together the population
used to estimate the effe ct of a first dose of the Pfizer -BioNTech COVID -19 vaccine and the
population used to estimate the effect of a second dose of such vaccine. This pooling is done
to gain statistical precision, under the assumption that the effect of a first or second dose i n
both populations is homogeneous. If this assumption is inaccurate, e.g., because receiving a
first dose sensitises the immune sy stem to react against a second dose, the estimates of the
main anal ysis will be biased.
9.10. Other aspects
Not applicable
10. PROTECTION OF HUMAN SUBJECTS
This is a non- interventional study using secondary data collection and does not pose any risks
for individuals. Each data source research partner will apply for an independent ethics
committee review according to local regulati ons.
Data protection and privacy regulations will be observed in collecting, forwarding,
processing, and storing data from stud y participants.
10.1. Patient i nformation
This study mainly involves data that exist in anony mised structured format and contain no
patient personal information.
All parties will comply with all applicable laws, including laws regarding the implementation
of organi sational and technical measures to ensure protection of patient personal data. Such
measures will include omitting patient nam es or other directl y identifiable data in an y
reports, publications, or other disclosures, except where required b y applicable laws.
Patient personal data will be stored at DAPs in encry pted electronic form and will be
password protected to ensure that on ly authori sedstudy staff have access.
DAPs will implement appropriate technical and organi sational measures to ensure that
personal data can be recovered in the event of disaster. In the event of a potential personal
data breach, DAPs shall be responsible for determining whether a personal data breach has in
fact occurred and, if so, providing breach notifications as required b y law.
To protect the rights and freedoms of natural persons with regard to the processing of
personal data, w hen study data are compiled for transfer to Pfizer and other authori sed
parties, an y patient names will be removed and will be replaced b y a single, specific,
numerical code .All other identifiable data transferred to Pfizer or other authori sedparties
will be identified by this single, patient -specific code. Inthecase of data transfer, Pfizer will
maintain high standards of confidentialit y and protection of individuals’ personal data
consistent with the vendor contract and applicable privacy laws.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950289
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 6110.2. Patien t consent
As this study does not involve data subject to privacy laws according to applicable legal
requirements, obtaining informed consent from individuals by Pfizer is not required.
10.3. Institutional review board (IRB)/Independent ethics c ommittee (IEC)
Each DAP will be following the local country and data custodian requirements to apply for
access to the data. At the coordinating centre, RTI-HS will ask approval for exemption from
review b y the RTI International institutional review board. All correspond ence with the
institutional review board or independent ethics committee and applicable documentation
will be retained as part of the study materials.
10.4. Ethical conduct of the s tudy
This study will adhere to the Guidelines for Good Pharmacoepidemiology Practices
(GPP)[22]and has been designed in line with the ENCePP Guide on Methodological
Standards in Pharmacoepidemiology[20]. The ENCePP Checklist for Study Protocols[23]will
be completed (see ANNEX 2 ).
The study is a post- authorisation study of vaccine safet y and will compl y with the definition
of the non- interventional (observational) stud y referred to in the International Conference on
Harmonisation tripartite guideline Pharmacovigilance Planning E2E[24]and provided in the
EMA Guideline on Good Pharmacovigilance Practices (GVP) Module VIII:
Post- Authorisation Safety Studies[25], and with the 2012 EU pharmacovigilance legislation,
adopted 19 June 2012 [26].
The study will be registered in the EU PAS Register [27]before data collection commences.
The research team and study sponsor should adhere to the general pri nciples of transparency
and independence in the ENCePP Code of Conduct[21]and the ADVANCE Code of
Conduct[25].
The study will be conducted in accordance with legal and regulatory requirements, as well as
with scientific purpose, value ,and rigour, andwillfollow generally accepted research
practices described in the Guidelines for Good Pharmacoepidemiology Practices (GPP)
issued by theInternational Society for Pharmacoepidemiology (ISPE) , and Good
Epidemiological Practice guidelines issued by the International Epidemiological
Association. An independent scientific advisory committee will be installed, comprising
experts in vaccine safet y studies.
11.MANAGEMENT AND REPORTING OF ADVERSE EVENTS/ADVERSE
REACT IONS
This study involves a combination of existing structured data and unstructured data, which
will be converted to structured form during the implementation of the protocol solely by a
computer using automated/algorithmic methods, such as natural languag e processing.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950290
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 62In these data sources, it is not possible to link (i.e. ,identify a potential association between) a
particular product and medical event for an y individual. Thus, the minimum criteria for
reporting an adverse event (AE) (i.e., identifiable patient, identifiable reporter, a suspect
product, and event) cannot be met.
For non -interventional study designs that are based on secondary use of data, such as studies
based on medical chart reviews or EHRs, sy stematic reviews, or meta -analyses, reporting of
adverse events/adverse drug reactions is not required. Reports of AEs/adverse drug reactions
should only be summarised in the study report, where applicable.
According to the EMA Guideline on Good Pharmacovigilance Practices (GVP), Modul e VI
–Management and Reporting of Adverse Reactions to Medicinal Products [29],
“All adverse events/reactions collected as part of [non -interventional post -authorisation
studies with a design based on secondary use of data] , the submission of suspec ted adverse
reactions in the form of [individual case safety reports] is not required. All adverse
events/reactions collected for the study should be recorded and summarised in the interim
safety analysis and in the final study report ”.
Module VIII – Post- Authorisation Safety Studies [25]echoes this approach. Legislation in the
EU further states that for certain stud y designs such as retrospective cohort studies,
particularl y those involving EHRs, it may not be feasible to make a causality assessment at
the individual case level.
This study protocol requires human review of patient- level unstructured data; unstructured
data refer to verbatim medical data, including text- based descriptions and visual depictions of
medical information, such as medical records, images of ph ysician notes, neurological scans,
x-rays, or narrative fields in a database. The reviewer is obligated to report AEs with explicit
attribution to any Pfizer drug that appear s in the reviewed information (defined per the
patient population and study period specified in the protocol). Explicit attribution is not
inferred b y a temporal relationship b etween drug administration and an AE but must be based
on a definite statement of causality by a health care provider linking drug administration to
the AE.
The requirements for reporting safet y events on the non -interventional study (NIS) adverse
event mo nitoring (AEM) Report Form to Pfizer Safet y are as follows:
All serious and non- serious AEs with explicit attribution to any Pfizer drug that appear
in the reviewed information must be recorded on the data collection tool (e.g., chart
abstraction form) andreported, within 24 hours of awareness, to Pfizer Safety using the
NIS AEM Report Form.
Scenarios involving drug exposure, including exposure during pregnancy , exposure
during breast feeding, medication error, overdose, misuse, extravasation, lack of effi cacy,
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950291
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 63and occupational exposure associated with the use of a Pfizer product , must be reported,
within 24 hours of awareness, to Pfizer Safety using the NI S AEM Report Form.
For AEs with an explicit attribution or scenarios involving exposure to a Pfizer pr oduct, the
safet y information identified in the unstructured data reviewed is captured in the Event
Narrative section of the report form, and constitute sall clinical information known regarding
these AEs . No follow -up on related AEs will be conducted.
All the demographic fields on the NI S AEM Report Form may not necessarily be completed,
as the form designates, asnot all elements will be available due to privacy concerns with the
use of secondary data sources. While not all demographic fields will be c ompleted, at the
very least, at least one patient identifier (e.g., gender, age as captured in the narrative field of
the form) will be reported on the NI S AEM Report Form, thus allowing the report to be
considered valid in accordance with pharmacovigilanc e legislation. All identifiers will be
limited to generalities, such as the statement ,“A 35 -year-old female...” or “An elderl y
male...” Other identifiers will have been removed.
Additionally , the onset/start dates and stop dates for “Illness”, “Study Drug ”, and “Drug
Name” may be documented in month/y ear (mmm/yyyy ) format rather than identify ing the
actual date of occurrence within the month /y ear of occurrence in the day /month/y ear
(DD/MMM/YYYY) format.
All research staff members must complete the following Pfizer training requirements:
“YRR Training for Vendors Working on Pfizer Studies (excluding interventional clinical
studies and non -interventional primary data collection studies with sites/investigators)”.
These trainings must be completed by research staff members before the start of data
collection. All trainings include a “Confirmation of Training Certificate” (for signature b y
the trainee) as a record of completion of the training, which must be kept in a retrievable
format. Copies of all signed tr aining certificates must be provided to Pfizer.
Re-training must be completed on an annual basis using the most current Your Reporting
Responsibilities training materials.
12.PLANS FOR DISSEMINATING AND COMMUNICATING STUDY RESULTS
As per EMA GVP Module VIII, t he stud y and its protocol will be registered in the EU PAS
Register prior to the start of data collection .
Results of analy ses and interpretation will be delivered in report form .
The first report will be a progress report describing the status of the study at each study
site,including ethical or related approvals, and results for the population covered b y ARS
Toscana, Italy .
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950292
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 64Each of the following six interim analy sis will be presented as interim reports, to be
delivered every 6months . Pfizer-BioNTech proposes to include in the interim reports
descriptive results, incidence rates ,and comparative anal yses when the data exist and are
accesible to support those anal yses with sufficient precision.
At the end of the third year of follow -up, the final report will be produced, including the
analysis and interpretation of each outcome ,including pregnancy outcomes.
Study results will be published following guidelines, including those for authorship,
established by the I CMJE[30]. When reporting results of this study, the appropriate
Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist
will be followed [31]. Independent publication rights will be granted to the research team in
line with Section VIII.B.5., Publication of study results, of the EMA Guideline on Good
Pharmacovigilance Practices (GVP) Module VIII: Post -Authorisation Safety Studies[25].
Upon study completion and finalisation of the study report, the results of this PASS will be
submitted for publication, preferabl y in a relevant peer -reviewed journal and posted in the
EU PAS Register .
Communication via other appropriate scientific venues will be considered.
In the event of an y prohibition or restriction imposed (e.g., clinical hold) by an applicable
competent authorit y in any area of the world, or if the investigator party responsible for
collecting data from the participant is aware of any new information that might influence the
evaluation of the benefits and risks of a Pfizer product, Pfizer should be informed
immediately .
13.REFERENCES
1.Pfizer -BioNTech. Fact sheet for healthcare providers administering vaccine (vaccination
providers); emergency use authori sation (EUA) of the Pfizer -BioNTech COVID-19
vaccine to prevent coronavirus disease 2019 (COVID -19). 2020. Available at:
https://www.fda.gov/media/144413/download. Accessed 17 December 2020.
2.ECDC. Overview of COVID -19 vaccination strategies and vaccine deployment plans in
the EU/EEA and the UK. Stockholm: European Centre for Disease Prevention and
Control. 2 December 2020. Available at:
https://www.ecdc.europa.eu/sites/default/files/documents/Overview -of-EU_EEA -UK-
vaccination -deploy ment -plans.pdf . Accessed 17 December 2020.
3. Hernán MA, Sauer BC, Hernández- Díaz S, Platt R, Shrier I . Specify ing a target trial
prevents immortal time bias and other self -inflicted injuries in observational analy ses. J
Clin Epide miol. 2016 Nov;79:70-5.
4. Hernán MA, Robins JM. Using big data to emulate a target trial when a randomized trial
is not available. Am J Epidemiol. 2016 Apr 15;183(8):758-64.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950293
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 655.García -Albéniz X, Hsu J, Hernán MA. The value of explicitly emulating a target trial
when using real world evidence: an application to colorectal cancer screening. Eur J
Epidemiol. 2017 Jun;32(6):495-500.
6.Hernán MA, Hernández -Díaz S, Robins JM. A structural approach to selection bias.
Epidemiology . 2004 Sep;15(5):615-25.
7.Suissa S. I mmortal time bias in observational studies of drug effects. Pharmacoepidemiol
Drug Saf. 2007 Mar;16(3):241 -9.
8. Gini R, Dodd CN, Bollaerts K, Bartolini C, Roberto G, Huerta- Alvarez C, et al.
Quantify ing outcome misclassification in multi- database st udies: the case study of
pertussis in the ADVANCE project. Vaccine. 2020 Dec 22;38 Suppl 2:B56 -b64.
9.Kuiper JG, Bakker M, Penning -van Beest FJA, Herings RMC. Existing data sources for
clinical epidemiology: the PHARMO Database Network. Clin Epidemiol.
2020;12:415 -22.
10.Sturkenboom M, Braey e T, van der Aa L, Danieli G, Dodd C, Duarte -Salles T, et al.
ADVANCE database characterisation and fit for purpose assessment for multi -country
studies on the coverage, benefits and risks of pertussis vaccinations. Vaccine. 2020 Dec
22;38:B8- B21.
11.Prados -Torres A, Poblador -Plou B, Gimeno -Miguel A, Calderón- Larrañaga A, Poncel -
Falcó A, Gimeno -Feliú LA, et al. Cohort profile: the epidemiology of chronic diseases
and multimorbidity . The EpiChron Cohort study . Int J E pidemiol. 2018;47(2):382 -4f.
12.Poblador -Plou B, Carmona -Pírez J, I oakeim -Skoufa I, Poncel-Falcó A, Bliek- Bueno K,
Cano- Del Pozo M, et al. Baseline chronic comorbidity and mortality in
laboratory -confirmed COVID -19 cases: results from the PRECOVI D study in Spain. I nt J
Environ Res Public Health. 2020 Jul 17;17(14).
13. Williams T, van Staa T, Puri S, Eaton S. Recent advances in the utility and use of the
General Practice Research Database as an example of a UK primary care data resource.
Ther Adv Drug Saf . 2012 Apr;3(2):89-99.
14. ISPE. Guidelines for good pharmacoepidemiology practices (GPP). Revision 3.
International Societ y for Pharmacoepidemiology . June 2015. Available at:
https://www.pharmacoepi.org/resources/policies/guidelines- 08027/. Accessed 20 Ja nuary
2021.
15.Austin PC. Balance diagnostics for comparing the distribution of baseline covariates
between treatment groups in propensity -score matched samples. Stat Med.
2009;28(25):3083 -107.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950294
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 6616.Yang D, Dalton JE. A unified approach to measuring the ef fect size between two groups
using SAS. Cary , NC: SAS I nstitute Inc. 2012. Available at:
https://support.sas.com/resources/papers/proceedings12/335 -2012.pdf . Accessed 22
October 2020.
17.Hernan MA, Robins JM. Causal survival anal ysis. In: Causal inference . Boca Raton:
Chapman & Hall/CRC; 2020. Available at: https://www.hsph.harvard.edu/miguel -
hernan/causal -inference -book/. Accessed 22 October 2020.
18.ENCePP. Guide on methodological standards in pharmacoepidemiology
(EMA/95098/2010 Rev. 7). European Netwo rk of Centres for Pharmacoepidemiology
and Pharmacovigilance. July 2018. Available at:
http://www.encepp.eu/standards_and_guidances/methodologicalGuide.shtml . Accessed
22 October 2020.
19.ENCePP. The ENCePP code of conduct for scientific independence and transparency in
the conduct of pharmacoepidemiological and pharmacovigilance studies (Revision 4).
15March 2018. Available at: http://www.encepp.eu/code_of_conduct/. Accessed 17
January 2020.
20.ENCePP. ENCePP checklist for study protocols (revision 4). European Network of
Centres for Pharmacoepidemiology and Pharmacovigilance. 15 October 2018. Available
at: http://www.encepp.eu/standards_and_guidances/checkListProtocols.shtml . Accessed
22 October 2020.
21.ICH. Pharmacovigilance planning. E2E. Internatio nal Conference on Harmonisation of
Technical Requirements for Registration of Pharmaceuticals for Human Use. 2004.
Available at: https://database.ich.org/sites/default/files/E2E_Guideline.pdf. Accessed 22
October 2020.
22.EMA. Guideline on good pharmacovi gilance practices (GVP). Module VIII –Post-
authorisation safet y studies (EMA/813938/2011 Rev 3). European Medicines Agency.
13October 2017. Available at: https://www.ema.europa.eu/en/documents/scientific-
guideline/guideline -good -pharmacovigilance -practic es-gvp-module -viii-post-
authorisation -safet y-studies -rev-3_en.pdf . Accessed 17 January 2020.
23. European Commission. Commission implementing Regulation (EU) No 520/2012 of
19 June 2012 on the performance of pharmacovigilance activities provided for in
Regulation (EC) No 726/2004 of the European Parliament and of the Council and
Directive 2001/83/EC of the European Parliament and of the Council. 20 June 2012.
Available at: http://eur-
lex.europa.eu/LexUriServ/L exUriServ.do?uri=OJ:L :2012:159:0005:0025:EN:PDF.
Accessed 22 October 2020.
24.ENCePP. The European Union electronic register of post-authorisation studies (EU PAS
Register). European Network of Centres for Pharmacoepidemiology and
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950295
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 67Pharmacovigilance. 20 December 2018. Available at:
http://www.encepp.eu/encepp_studies/indexRegister.shtml . Accessed 22 October 2020.
25.Kurz X, Bauchau V, Mahy P, Glismann S, van der Aa LM, Simondon F, et al. The
ADVANCE Code of Conduct for collaborative vaccine studies. Vaccine. 2017 Apr
4;35(15):1844 -55.
26.ENCePP. The EN CePP seal. European Network of Centres for Pharmacoepidemiology
and Pharmacovigilance. 08 September 2020. Available at:
http://www.encepp.eu/encepp_studies/index.shtml. Accessed 22 October 2020.
27. EMA. Guideline on good pharmacovigilance practices (GVP). Module VI – Collection,
management and submission of reports of suspected adverse reactions to medicinal
products (Rev 2). European Medicines Agency . 22 November 2017. Available at:
https://www.ema.europa.eu/en/documents/regulatory -procedural -guideline/gu ideline -
good -pharmacovigilance -practices- gvp-module -vi-collection -management -submission-
reports_en.pdf . Accessed 16 June 2020.
28. ICMJE. Recommendations for the conduct, reporting, editing, and publication of
scholarl y work in medical journals. I nternatio nal Committee of Medical Journal Editors.
December 2019. Available at: http://www.icmje.org/recommendations/. Accessed
22October 2020.
29.von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP. The
Strengthening the Reporting of Observational Studies in Epidemiology (STROBE)
statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008
Apr;61(4):344 -9.
30. Yih WK, L ee GM, Lieu TA, et al. Surveillance for adverse events following receipt of
pandemic 2009 H1N1 vaccine in the Post -Licensure Rapid Immunization Safet y
Monitoring (PRI SM) S ystem, 2009 -2010. Am J Epidemiol. 2012;175(11):1120 -1128.
31. Liu CH, Yeh YC, Huang WT, Chie WC, Chan KA. Assessment of pre -specified adverse
events following var icella vaccine: A population- based self -controlled risk interval study .
Vaccine. 2020;38(11):2495-2502.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950296
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 6814.LIST OF TABLES
Table 1. List of Selected Adverse Events of Special Interest................................ .28
Table 2. Number of individuals needed to detect different risk ratios for
select AESI a with a range of background rates ................................ ......... 47
15.LIST OF FIGURES
Figure 1. Self-controlled risk interval design ................................ ........................... 23
Figure 2. Study period and follow -up periods ................................ ......................... 25
Figure 3. Data management plan ................................ ................................ ............. 49
ANNEX 1. LIST OF STAND ALONE DOCUMENTS
None
ANNEX 2. ENCEPP CHECKLIST FOR STUDY PROTOCOLS
ENCePP Checklist for Study Protocols (Revision 4)
Study title: Post Conditional Approval Active Surveillance Study Among Individuals in Europe Receiving
the Pfizer -BioNTech Coronavirus Disease 2019 (COVID -19) Vaccine
EU PAS Register number:
Study reference number (if applicable):
Section 1: Milestones Yes No N/A Section
Number
1.1 Does the protocol specify timelines for
1.1.1 Start of data collection16
1.1.2 End of data collection26
1.1.3 Progress report(s) 6
1.1.4 Interim report(s) 6
1.1.5 Registration in the EU PAS Register 6
1.1.6 Final report of study results 6
Comments:
1Date from which information on the first study is first recorded in the study data set or, in the case of
secondary use of data, the date from which data extraction starts .
2Date from which the analytical data set is completely available.
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950297
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 69Section 2: Research question Yes No N/A Section
Number
2.1 Does the formulation of the research question and
objectives clearly explain: 8
2.1.1 Why the study is conducted? (e.g., to address an important
public health concern, a risk identified in the risk management plan,
an emerging safety issue)7
2.1.2 The objective(s) of the study? 8.1
2.1.3 The target population? (i.e., population or subgroup to whom
the study results are intended to be generalis ed)8.1
2.1.4 Which hypothesis( -es) is (are) to be tested? 8.1
2.1.5 If applicable, that there is no a priori hypothesis?
Comments:
Section 3: Study design Yes No N/A Section
Number
3.1 Is the study design described? (e.g., cohort, case -control, cross-
sectional, other design) 9.1
3.2 Does the protocol specify whether the study is based on
primary, secondary or combined data collection?9.1, 9.6, 9.6.4
3.3 Does the protocol specify measures of occurrence? (e.g., rate,
risk, prevalence)9.1
3.4 Does the protocol specify measure(s) of association?
(e.g., relative risk, odds ratio, excess risk, incidence rate ratio, hazard ratio,
number needed to harm [NNH])9.7.2.2
3.5 Does the protocol describe the approach for the collection
and reporting of adverse events/adverse reacti ons?
(e.g., adverse events that will not be collected in case of primary data
collection)11
Comments:
Section 4: Source and study populations Yes No N/A Section
Number
4.1 Is the source population described? 9.2.1
4.2 Is the planned study population defined in terms of:
4.2.1 Study time period 9.2.2
4.2.2 Age and sex 9.2.3.1
4.2.3 Country of origin 9.2.1
4.2.4 Disease/indication 9.2.3.1
4.2.5 Duration of follow -up 9.2.3.1
4.3 Does the protocol define how the study population will be
sampled from the source population? (e.g., event or
inclusion/exclusion criteria)9.2.3, 9.2.4
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950298
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 70Comments:
Section 5: Exposure definition and measurement Yes No N/A Section
Number
5.1 Does the protocol describe how the study exposure is
defined and measured? (e.g., operational details for defining and
categorising exposure, measurement of dose and duration of drug
exposure)9.3.1
5.2 Does the protocol address the validity of the exposure
measurement? (e.g., precision, accuracy, use of validation sub -study)
5.3 Is exposure categorised according to time windows? 9.3.2.1
5.4 Is intensity of exposure addressed?
(e.g., dose, duration)9.7.1.1,
9.7.1.3
5.5 Is exposure categorised based on biological mechanism of
action and taking into account the pharmacokinetics and
pharmacodynamics of the drug?9.7.1.4
5.6 Is (are) an appropriate comparator(s) identified? 9.3.1.1,
9.3.1.2
Comments:
Section 6: Outcom e definition and measurement Yes No N/A Section
Number
6.1 Does the protocol specify the primary and secondary (if
applicable) outcome(s) to be investigated?9.3.2
6.2 Does the protocol describe how the outcomes are defined
and measured? 9.3.2.1
6.3 Does the protocol address the validity of outcome
measurement? (e.g., precision, accuracy, sensitivity, specifi city,
positive predictive value, use of validation sub -study)9.3.2.1.1
6.4 Does the protocol describe specific outcomes relevant for
Health Technology Assessment? (e.g., HRQOL, QALYs,
DALYS, health care services utilisation, burden of disease or treatment,
compliance, disease management)
Comments:
Section 7: Bias Yes No N/A Section
Number
7.1 Does the protocol address w ays to m easure confounding?
(e.g., confounding by indication)9.7.1. 2
7.2 Does the protocol address selection bias? (e.g., healthy
user/adherer bias)9.7.1.6
7.3 Does the protocol address information bias?
(e.g., misclassification of exposure and outcomes, time -related bias)9.3.2.1.1
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950299
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 71Comments:
Section 8: Effect m easure modification Yes No N/A Section
Number
8.1 Does the protocol address effect modifiers? (e.g., collection of
data on known effect modifiers, subgroup analyses, anticipated direction
of effect) 9.3.3, 9.7.1.4 ,
9.7.2.2
Comments:
Section 9: Data sources Yes No N/A Section
Number
9.1 Does the protocol describe the data source(s) used in the
study for the ascertainment of:
9.1.1 Exposure? (e.g., pharmacy dispensing, general practice
prescribing, claims data, self -report, face -to-face interview)9.3.1
9.1.2 Outcomes? (e.g., clinical records, laboratory markers or values,
claims data, self -report, patient interv iew including scales and
questionnaires, vital statistics)9.3.2
9.1.3 Covariates and other charac teristics? 9.3.3
9.2 Does the protocol describe the information available from
the data source(s) on:
9.2.1 Exposure? (e.g., date of dispensing, drug quantity, dose, number
of days of supply prescription, daily dosage, prescriber)9.4
9.2.2 Outcomes? (e.g., date of occurrence, multiple event, severity
measures related to event )9.4
9.2.3 Covariates and other characteristics? (e.g., age, sex,
clinical and drug use history, comorbidity, co -medications, lifestyle)9.4
9.3 Is a coding system described for:
9.3.1 Exposure? (e.g., WHO Drug Dictionary, Anatomical Therapeutic
Chemical (ATC) Classification System)9.3.1
9.3.2 Outcomes? (e.g., International Classification of Diseases (I CD),
Medical Dictionary for Regulatory Activities (MedDRA))9.3.2
9.3.3 Covariates and other characteristics? 9.4
9.4 Is a linkage method betw een data sources described?
(e.g., based on a unique identifier or other)9.4
Comments:
Section 10: Analysis plan Yes No N/A Section
Number
10.1 Are the statistical methods and the reason for their choice
described? 9.7
10.2 Is study size and/or statistical precision estimated? 9.5
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950300
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 72Section 10: Analysis plan Yes No N/A Section
Number
10.3 Are descriptive analyses included? 9.7.1.2,
9.7.2.1
10.4 Are stratified analyses included? 9.7.2.2
10.5 Does the plan describe methods for analytic control of
confounding?9.7.1.5,
9.7.1.6,
9.7.2.2
10.6 Does the plan describe methods for analytic control of
outcome misclassification?9.3.2.1.1
10.7 Does the plan describe methods for handling missing data? 9.7.1.2
10.8 Are relevant sensitivity analyses described? 9.3.1.1,
9.3.1.2
Comments:
Section 11: Data m anagem ent and quality control Yes No N/A Section
Number
11.1 Does the protocol provide information on data storage?
(e.g., software and IT environment, database maintenance and anti -fraud
protection, archiving)9.6.3, 9.6.5
11.2 Are methods of quality assurance described? 9.8
11.3 Is there a system in place for independent review of study
results? 10.4
Comments:
Section 12: Lim itations Yes No N/A Section
Number
12.1 Does the protocol discuss the impact on the study results
of:
12.1.1 Selection bias? 9.9
12.1.2 Information bias? 9.9
12.1.3 Residual/unmeasured confounding?
(e.g., anticipated direction and magnitude of such biases, validation sub -
study, use of validation and external data, analytical methods)9.9
12.2 Does the protocol discuss stud y feasibility? (e.g., study size,
anticipated exposure uptake, duration of follow -up in a cohort study,
patient recruitment, precision of the estimates)
Comments:
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950301
Pfizer -BioNTech COVID -19 vaccine
C4591021 NON -INTERVENTIONAL STUDY PROTOCOL
Version 2 (20May 2021)
PFIZER CONFIDENTIAL
CT24- WI-GL02 -RF02 2 .0 Non-Interventional Study Protocol Template For Secondary Data Collection Study
01-Jun-2020
Page 73Section 13: Ethical issues Yes No N/A Section
Number
13.1 Have requirements of Ethics Committee/ Institutional
Revie w Board been described?10.3
13.2 Has any outcome of an ethical review procedure been
addressed?
13.3 Have data protection requirements been described? 10.1
Comments:
Section 14: Am endments and deviations Yes No N/A Section
Number
14.1 Does the protocol include a section to document
amendments and deviations? 5
Comments:
Section 15: Plans for communication of study results Yes No N/A Section
Number
15.1 Are plans described for communicating study results (e.g., to
regulatory authorities) ? 12
15.2 Are plans described for disseminating study results
externally, including publication?12
Comments:
Nam e of the main author of the protocol: Alejandro Arana
Date: 24/02/2021
Signature: To be signed upon PRAC endorsement
ANNEX 3. ADDITIONAL INFORMATION
Not applicable
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950302
Document Approval Record
Document Name:
Document Title:
Signed By: Date(GMT) Signing Capacity
!"# $%&
' ($#$)" ($$
* +
, $%&
090177e19714f608\Approved\Approved On: 20-May-2021 15:15 (GMT)
FDA-CBER-2021-5683-0950303