Document text
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 1of 144NON-INTERVENTIONAL ( NI) STUDY CONCEPT PR OTOCOL
Title Post-Emergency Use Authorization Active
Safety Surveillance Study among Individuals
in the Veteran’s Affairs Health Sy stem
Receiving Pfizer -BioNTech Coronavirus
Disease 2019 (COVID -19) Vaccine
Protocol number C4591012
Protocol version identifier Final Version 1.0
Date of last version of protocol 27January 2021
EUPost Authori zation Study (PAS)
register numberTo be registered before the start of data
collection
Active substance COVID-19 mRNA Vaccine is single -stranded,
5’-capped messenger RNA (mRNA) produced
using a cell -free in vitro transcription from the
corresponding DNA templates, encoding the
viral spike (S) protein of SARS -CoV-2.
Medicinal product Pfizer-BioNTech COVID- 19 Vaccine
Research question and objectives Research question: what are the incidence rates
of safety events of interest (based on adverse
events of special interest[AESI]) among
individuals vaccinated with the Pfizer-
BioNTech COVID-19 vaccine within the US
Veterans Health Administration (VHA) s ystem
overall and in sub- cohorts of interest, as
compared to expected rates of those events?
Primary study objectives:
To assess whether in dividuals in the
VHA system experience increased risk
of safety events of interest following
receipt of the Pfizer- BioNTech
COVID-19 vaccine;
To assess whether sub- cohorts of
interest (i.e., immunocompromised,
elderly, individuals with specific
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073210
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 2of 144comorbiditi es, individuals receiving
only one dose of the Pfizer -BioNTech
COVID-19 vaccine, and individuals
with prior SARS -CoV-2 infection) in
the VHA s ystem experience increased
risk of safet y events of interest
following receipt of the Pfizer-
BioNTech COVID -19 vaccine.
Secondary study objective :
To characterize utilization patterns of
the Pfizer -BioNTech COVID-19
vaccine among individuals within the
VHA, including estimating the
proportion of individuals receiving
vaccine, 2 -dose vaccine completion
rate, and distri bution of time gaps
between the first and second dose,
demographics and health histories of
recipients, overall and among the sub-
cohorts of interest.
Authors Yinong Young- Xu, ScD, MA, MS
Director, Clinical Epidemiology Program
Veterans Affairs Medical Center
White River Junction, VT
Cynthia de Luise, PhD, MPH
Senior Epidemiologist/ Safety Surveillance
Research Scientist ; Risk Management and
Safety Surveillance Research
Pfizer, Inc.
New York, NY
Mei Sheng Duh, ScD, MPH
Managing Principal and Chief Epidemiologist
Analysis Group, Inc.
Boston, MA
This document contains confidential information belonging to Pfizer. Except as otherw ise agreed to in w riting,
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 suspected breach of this obligation, Pfizer must be promptly notified.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073211
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 3of 1441.TABLE OF CONTENTS
1.TABLE OF CONTENTS.......................................................................................................3
2. LIST OF ABBREVIAT IONS................................ ................................ ................................ 5
3. RESPONS IBLE PARTI ES................................ ................................ ................................ ....7
4.ABSTRACT ................................ ................................ ................................ ........................... 8
5.AMENDMENTS AND UP DATES................................ ................................ ..................... 21
6.MILESTONES ................................ ................................ ................................ ..................... 22
7. RATIONALE AND BAC KGROUND ................................ ................................ ................ 23
8. RESEARCH QUESTION AND O BJECTIVES................................ ................................ .24
9. RESEARCH METHODS................................ ................................ ................................ ....25
9.1. Study Design................................ ................................ ................................ ...........25
9.1.1. Self -Controlled Risk I nterval (SCRI) Design ................................ .............25
9.1.2. Active Comparator Design ................................ ................................ .........27
9.1.3. Study Period................................ ................................ ................................ 28
9.2. Setting ................................ ................................ ................................ ...................... 28
9.2.1. Inclusion Criteria ................................ ................................ ........................ 28
9.2.2. Exclusion criteria ................................ ................................ ........................ 28
9.2.3. Subgroups ................................ ................................ ................................ ...28
9.3. Variables ................................ ................................ ................................ .................. 29
9.3.1. Exposure of I nterest................................ ................................ .................... 29
9.3.1.1. Pfizer -BioNTech COVID- 19 Vaccine Groups of Interest ........30
9.3.2. Baseline Characteristics ................................ ................................ ..............31
9.3.3. Outcomes ................................ ................................ ................................ ....32
9.4. Data Source................................ ................................ ................................ .............39
9.5. Study Size................................ ................................ ................................ ................ 40
9.5.1. Power ................................ ................................ ................................ ..........40
9.6. Data Management ................................ ................................ ................................ ...42
9.6.1. Case report forms (CRFs)/Electronic data record ................................ ......42
9.6.2. Record retention ................................ ................................ .......................... 42
9.7. Data Anal ysis................................ ................................ ................................ ..........43
9.7.1. Baselin e Characteristics ................................ ................................ ..............43
9.7.2. Vaccine Utilization Patterns ................................ ................................ .......44
9.7.3. Safet y Signal Analyses ................................ ................................ ...............44
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073212
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 4of 1449.7.3.1. Signal Detection ................................ ................................ ........46
9.7.3.2.Signal Evaluation ................................ ................................ ......49
9.7.3.3. Signal Verification ................................ ................................ ....51
9.7.4. Seasonality -Adjusted Cases -Centered Method ................................ ...........51
9.7.5. End-of- Season and End -of-Surveillance Analy ses................................ .....52
9.7.6. Subgroup Analy sis................................ ................................ ...................... 53
9.7.7. Incidence Rates and Time to Safety Event of Interest Anal ysis................. 53
9.8. Quality Control................................ ................................ ................................ ........53
9.9. Strengths and Limitations of the Research Methods ................................ ...............54
9.10. Other Aspects ................................ ................................ ................................ ........55
10. PROTECTI ON OF HU MAN SUBJECTS ................................ ................................ ........55
10.1. Patient I nformation ................................ ................................ ................................ 55
10.2. Patient Consent ................................ ................................ ................................ ......56
10.3. Institutional Review board (I RB)/Independent Ethics Committee (I EC).............56
10.4. Ethical Conduct of the Study ................................ ................................ ................ 56
11. MANAGEMENT AND R EPORTING OF ADVERSE EVENTS/ADVERSE
REACTIONS................................ ................................ ................................ ...................... 56
12. PLANS FOR DI SSEMINATING AND COMMUNI CATING STUDY RESUL TS........58
13. REFERENCES ................................ ................................ ................................ .................. 59
14. LIST OF TABLES ................................ ................................ ................................ .............63
15. LIST OF FIGURES ................................ ................................ ................................ ...........63
16. ANNEX 1. LIST OF STAND ALONE DOCUMEN TS................................ ................... 64
17. ANNEX 2. ENCEPP CHECKLIST FOR STUDY PROTOCOL S................................ ...64
18. ANNEX 3. ADDITIO NAL INFORMATION ................................ ................................ ...64
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073213
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 5of 1442. LIST OF ABBREVIATIONS
Abbreviation Definition
ACIP Advisory Committee on Immunization Practices
ADEM Acute disseminated encephalomy elitis
AE Adverse event
AEM Adverse event monitoring
AESI Adverse event of special interest
AIDS Acquired immunodeficiency syndrome
AMI Acute myocardial infarction
BMI Body mass index
CAD Coronary artery disease
CI Confidence Interval
CCI Charlson comorbidity index
CDC Centers for Disease Control and Prevention
CDW Corporate Data Warehouse
CEP Clinical Epidemiology Program
CIDP Chronic inflammatory demyelinating pol yneuropathy
CMA Conditional Marketing Authorization
COPD Chronic obstructive pulmonary disease
COVID-19 Coronavirus Disease 2019
CPT Current Procedural Terminology
CRADA Cooperative Research and Data Agreement
CRFs Case report forms
DIC Disseminated intravascular coagulation
DVT Deep vein thrombosis
TDap Diphtheria, tetanusand (acellular) pertussis
Td Diphtheria and tetanus
ED Emergency department
EMA European Medicines Agency
EMR Electronic medical records
EU European Union
EUA Emergency Use Authorization
EU PAS European Union Post -Authorization Safety
FDA Food and Drug Administration
GBS Guillain-Barré syndrome
GEP Good Epidemiological Practice
GPP Good Pharmacoepidemiology Practices
H0 Null hypothesis
Ha Alternative h ypothesis
HBV Hepatitis B virus
HCPCS Healthcare Common Procedure Coding S ystem
HCV Hepatitis C virus
HIV Human immunodeficiency virus
HPV Human papillomavirus
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073214
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 6of 144Abbreviation Definition
ICD-10-CM International Classification of Diseases, Tenth Revision, Clinical
Modification
ICD-10-PCS International Classification of Diseases, Tenth Revision, Procedure
Coding Sy stem
IEA International Epidemiological Association
IEC Independent Ethics Committee
IQR Interquartile range
IRB Institutional Review Board
KD Kawasaki disease
LLR Log-likelihood ratio
MaxSPRT Maximized sequential probability ratio test
MenACWY Meningococcal conjugate
MenB Serogroup B meningococcal
MIS-A Multisystem inflammatory syndromein adults
mRNA Messenger RiboNucleic Acid
MS Multiple sclerosis
NIS Non-interventional study
ON Optic neuritis
PASS Post-Authorization Safety Study
PRISM Post-Licensure Rapid Immunization Safety Monitoring
RCA Rapid cycle analysis
RR Relative risk
SAP Statistical analy sis plan
SARS-CoV-2 Severe acute respiratory syndrome coronavirus 2
SAS SAS Institute
SCRI Self-controlled risk interval
SD Standard deviation
SPEAC Safety Platform for Emergency vACcines
TM Transverse m yelitis
UK United Kingdom
US United States
VA Department of Veterans Affairs
VAERS Vaccine Adverse Event Reporting S ystem
VHA Veterans Health Administration
VINCI VA Informatics and Computing Infrastructure
VISN Veterans Integrated Service Networks
VSD Vaccine Safet y Datalink
VTE Venous thromboembolism
WHO World Health Organization
WOC Without compensation
YRR Your Reporting Responsibilities
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073215
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 7of 1443.RESPONSIBLE PARTIES
Name, degree(s) Job Title Affiliation Address
Principal Investigator:
Yinong Young -Xu,
ScD, MA, MSDirector, Clinical Epidemiology
ProgramVeterans Affairs
(VA) Medical
Center163 Veterans Drive,
White River Junction,
VT 05009
Cynthia de Luise,
PhD, MPHSenior Epidemiologist /Safety
Surveillance Research Scientist ;
Risk Management and Safety
Surveillance ResearchPfizer, Inc. 235 East 42nd Street,
New York, NY 10017
Mei Sheng Duh,
ScD, MPHManaging Principal and Chief
Epidemiologist
Visiting Scientist, Department of
BiostatisticsAnalysis Group,
Inc.
Harvard T. H.
Chan School of
Public Health111 Huntington Ave
14thFloor
Boston, MA 02199
677 Huntington Ave
Boston, MA 02115
Maral DerSarkissian,
PhD Vice President and Senior
Epidemiologist
Adjunct Assistant ProfessorAnalysis Group,
Inc.
Fielding School of
Public Health,
University of
California, Los
Angeles333 South Hope Street
27th Floor
Los Angeles, CA
90071
650 Charles E Young
DriveSouth
Los Angeles, CA
90095
Rachel Bhak, MS Manager and Senior Biostatistician Analysis Group,
Inc. 111 Huntington Ave
14thFloor
Boston, MA 02199
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073216
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 8of 1444.ABSTRACT
Title:Post-Emergency Use Authorization Active Safety Surveillance Study among
Individuals in the Veteran’s Affairs Health S ystem Receiving Pfizer- BioNTech Coronavirus
Disease 2019 (COVID -19) Vaccine
Protocol Version: 1.0; Dateof Protocol : 27January 2021
Authors: Yinong Young Xu, ScD, MA, MS , Veterans Affai rs Medical Center; Cy nthia de
Luise, PhD, MPH, Pfizer, I nc.; Mei Sheng Duh, ScD, MPH, Anal ysis Group, I nc.
Rationale and b ackground :
In March 2020, the World Health Organization (WHO) declared a global pandemic for the
coronavirus disease 2019 (COVID -19) due to the severe acute respiratory syndrome
coronavirus 2 (SARS -CoV-2), which was first identified by public health officials in China
in December 2019.1The COVID -19 pandemic presents an unprecedented public health
crisis.As of January 7, 2021, over 21.4million COVID -19 cases and 364,000 deaths have
been reported in the United States (US) alone.2
Pfizer and BioNTech have partnered to develop a novel messenger RiboNucleic Acid
(mRNA)vaccine against SARS -CoV-2 for the prevention of COVID -19 (Candidate
BNT162b2). Pfizer is conducting a Phase 1/2/3, randomized, placebo -controlled, observer -
blind, dose-finding, vaccine candidate- selection, and efficacy study among healthy
individuals (NCT04368728). The Food and Drug Administration (FDA) reviewed the
available safet y data from 37,586 participants 16 years of age and older and did not identify
anyspecific safet y concerns. In addition, the anal ysis of available efficacy data from 36,523
participants 12 years of age and older without evidence of prior SARS- CoV-2 infection at
least 7 day s after receiving the second dose demonstrated 95% efficacy of the vaccine in the
prevention of COVID -19 (as confirmed by 8 vs. 162 COVID -19 cases in the vaccine and
placebo groups, respectively ).3,4Based on these safety and efficacy data, as well as a revi ew
of manufacturing information regarding product quality and consistency , the FDA
determined that the known and potential benefits of the vaccine outweighed the known and
potential risks for the prevention of COVID -19 in individuals 16 y ears of age and ol der.4
Therefore on December 11, 2020, the Pfizer -BioNTech COVID-19 vaccine was granted an
Emergency Use Authorization (EUA) by the FDA to prevent COVID -19 in individuals 16
years of age and older.5
With respect to geographi c regions other than the US, on December 2, 2020, the United
Kingdom (UK) was the first country in the world to grant temporary authorization for
emergency use of the Pfizer -BioNTech COVID -19 vaccine.6On December 21, 2020, the
European Medicines Agency (EMA) granted the Pfizer -BioNTech COVID- 19 vaccine a
conditional marketing authorization (CMA) for use among individuals 16 years of age and
older throughout all of the European Union’s (EU) 27 member states.7
As required b y the EUA, post-authorization observational studies using real -world data are
needed in order to assess the association between Pfizer -BioNTech COVID- 19 vaccine and
pre-determined safet y events of intere st (including deaths , hospitalizations, and severe
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073217
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 9of 144COVID-19) among individuals administered the vaccine in both the population at large and
in populations of interest ( e.g., immunocompromised individuals, elderl y, and those with
specific comorbidities) .4Pfizer in collaboration with the US Veterans Health Administration
(VHA) and Anal ysis Groupherein propose post -EUA active safet y surveillance of safety
events of interest based on the Priority List of Adverse Events of Special Interest from the
Brighton Collaboration’s Safety Platform for Emergency vACcines (SPEAC) Project , the
FDA and the Centers for Disease Control and Prevention’s (CDC) Advisory Committee on
Immunization Practices (ACI P)enhanced safet y monitoring recommendation. This safety
surveillance study willidentify and evaluate rapid,near real-time potential safet y signals
associated with the Pfizer -BioNTech COVID-19 vaccine in the large -scale VHA electronic
medical record (EMR) database. The observed safety event of interest rates will be compa red
to expected rates derived from self-controls and active comparators receiving seasonal
influenza vaccination. Part of the methodologies used in this study are constructed based on
approaches previousl y used by the Post -Licensure Rapid I mmunization Safe ty Monitoring
(PRISM) program for the H1N1 vaccine.8This non- interventional study is designated as a
Post-Authorization Safety Study (PASS) commitment to the US FDA and is a Category 3
commitment in the EU Risk Management Plan.
Research question and objectives :
Research question: what arethe incidence rates of safet y events of interest (based on adverse
events of special interest [AESI ])among individuals vaccinated with the Pfizer-BioNTech
COVID-19 vaccine within the US VHA system overall and in sub -cohorts of interest, as
compared to expected rates of those events?
Primary study objectives:
To assess whether individuals in the VHA s ystem experience increased risk of safet y
events of interest following receipt of the Pfiz er-BioNTech COVID -19 vaccine;
To assess whether sub- cohorts of interest (i.e., immunocompromised, elderly ,
individuals with specific comorbidities, individuals receiving onl y one dose of the
Pfizer-BioNTech COVID
-19 vaccine, and individuals with prior SARS -CoV-2
infection) in the VHA system experience increased risk of safet y events of interest
following receipt of the Pfizer-BioNTech COVID- 19 vaccine.
Secondary study objective:
To characterize utilization patterns of thePfizer -BioNTech COVID -19 vaccine
among individuals within the VHA , including estimating the proportion of
individuals receiving vaccine, 2-dose vaccine completion rate,anddistribution of
time gaps between the first and second dose , demographics and health histories of
recipients, overall and among the sub- cohorts of interest .
Study design: This post-EUAactive safet y surveillance program will employ a rapid-cycle,
longitudinal, observational cohort study designto provide early real-world safet y
information .
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073218
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 10of 144The self-controlled risk interval (SCRI) design will be used to sequentially monitor
occurrence of safety events of interest while controlling for time- invariant
confounders. The SCRI design uses data from cases ( i.e., individuals who experience
safety events of interest following vaccination) to compare the risk interval following
vaccination to pre -or post-vaccination non -risk intervals (“pre -vaccination control
interval” and “post -vaccination control interval”) in the same individual.
An active comparator design will be used t o sequentially monitor occurrence of safety
eventsof interest with Pfizer -BioNTech COVID -19 vaccinations as compared to
recipients of influenza vaccinein the VHA during 2014/2015 through 2018/2019 flu
seasons. Data in peri -COVID time periods from January 2020 to present are excluded
because of pandemic- associated under -utilization of health resources and under -
reporting of medical events.
Population : The exposed population will be kept as broad as possible in order to capture
safety events of interest that occur among all individuals receiving the Pfizer -BioNTech
COVID-19 vaccine in the period from December 11, 2020 to present . Individuals will be
included if they have a record of at least one dose of Pfizer -BioNTech COVID-19 vaccine.
Individuals who rec eive at least one dose of COVID -19 vaccine from a manufacturer other
than Pfizer -BioNTech will be identified and reported, but they will be excluded from further
analysis. All individuals will be required to be enrolled in and not disenrolled from VHA
benefits during the 1 y ear prior to vaccination date ( i.e., baseline period). Depending on the
attrition rate, the length of the baseline period may be modified to 6 months.
The influenza vaccine comparator cohort will be identified based on a record of at lea st one
dose of seasonal influenza vaccine during prior flu seasons, from 2014/2015 through
2018/2019.
Variables :
Exposure s: Administration of Pfizer-BioNTech COVID- 19 vaccine post -EUA
approval will be identified based on the following:
oCurrent Procedural Terminology (CPT) code 91300 (Severe acute respiratory
syndrome coronavirus 2 (SARS -CoV-2) (coronavirus disease [COVID- 19])
vaccine, mRNALNP, spike protein, preservative free, 30 mcg/0.3mL dosage,
diluent reconstituted, for intramuscular use) and associate d vaccine
administration HCPCS codes corresponding to the first dose: 0001A (ADM
SARS-CoV-2 30 mcg/0.3mL 1st), and the second dose: 0002A (ADM SARS -
CoV-2 30 mcg/0.3mL 2nd);9,10OR
o10 and 11-dig it National Drug Codes (NDCs) 59267-1000 -1 (corresponds to
first dose), 59267 -1000-01 (corresponds to second dose);9OR
oImmunization records that contain data on vaccine code descriptor, vaccine
manufacturer ( i.e., Pfizer), lot number, injection site, and date(s) of
immunization;9
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073219
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 11of 144Relevant codes will be continuously reviewed and amended if new codes are added.
Administration of the seasonal influenza vaccine during 2014/2015 through
2018/2019 flu seasons will be identified based on the following:
oCPT codes
90654 (Influenza virus vaccine, trivalent (IIV3), split virus,
preservative -free, for intradermal use); OR
90656 (Influenza virus vaccine, trivalent (IIV3), split virus,
preservative free, 0.5 mL dosage, for intramuscular use); OR
90658 (Influenza virus vaccine, trivalent (IIV3), split virus, 0.5 mL
dosage, for intramuscular use); OR
o10 and 11- digit NDCs; OR
oImmunization records that contain data on vaccine code descriptor, vaccine
manufacturer, lot number, injection site, and date(s) of immunization.
Outcomes : Safety events of interest for active surveillance (see Table 1and Appendix
Table 2) are based on the Priority List of Adverse Events of Special Interest from the
Brighton Collaboration’s Safety Platform for Emergency vACcines (SPEAC) Project ,
the FDA and the Centers for Dise ase Control and Prevention’s (CDC) Advisory
Committee on I mmunization Practices (ACIP) enhanced safet y monitoring
recommendations.
The list of safety events of interest may be revised over the course of the study, and if
unanticipated po tential safety events of interest are identified during the course of
surveillance, they will be added to the list and included in the anal yses. The risk and
control intervals for each safety event of interest are based on biological plausibility
and precedents in the litera ture (see Table 1). Outpatient (including emergency
department [ED]) and/or inpatient settings will be used to identify safety events of
interest depending on the ty pe of event. The specific encounter setting to be
considered for each safety event of interest is summarized in Table 1
and can be
assigned to 1) the risk interval following vaccination Pfizer-BioNTech COVID-19
vaccination, 2) the pre-v accination self-c ontrol interval, 3) the post-vaccination self-
control interval, or 4) risk interval for the active comparators of receiving seasonal
influenza vaccine. Events outside the intervals will not b
e counted.
Only the individual’s first instance of asafety eventof interest following a specified
clean window ( i.e., the occurrence -free baseline period used to define incide nt
outcomes during which individuals enter the study cohort only if the safety event of
interest did not occur during this period) will be captured; this means that if a safety
event of interest is identified but diagnosis codes corresponding to the safety event of
interest are also observed during the clean window, it will not be counted . The
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073220
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 12of 144duration of the pre-specified clean window will differ by safety event of interest (see
Appendix Table 2) in order to rule out pre -existing events.
Key Covariates: Baseline demographic ( i.e., age, sex, race/ethnicit y, state) and
clinical characteristics ( i.e., smoking, body mass index [ BMI],history of
anaphylaxis/allergic reactions, previous anaphy laxis to vaccine component, history of
hospitalizations, Charlson Comorbidity Index [CCI ], selected comorbidities, and
concurrent immunizations)11will be assessed based on available data ( i.e., during 1-
year baseline ) prior to the date of vaccination with Pfizer -BioNTech COVID -19
vaccine and date of seasonal influenza vaccination for active comparators .
Subgroups : Immunocompromised individuals, elderly , individuals with specific
comorbidities, those receiving only one dose of Pfizer -BioNTech COVID -19 vaccine,
thosewith prior SARS -CoV-2 infection, those with regular use of VHA medical care,
and VA priorit y group 1 veterans will be identified .
Data source :The VHA is the largest integrated health care s ystem in the US, providing both
inpatient and outpatient clinical care to over 9 million Veterans enrolled at more than 170
medical centers and 1,074 community -based outpatient clinics.12This studywill use data
from VHA’s Corporate Data Warehouse (CDW), which is an integrated electronic medical
record (EMR) s ystem with a centralized data warehouse that is updated on a daily basis.The
CDWdoes not include information on an y care received outside of a VHA facility .The VA
offers eligible Veterans long -term care services ranging from nursing homes and assisted -
living centers to caregiver support in the Veterans’ own homes.13
Study size: The sample size achieved will depend on the number of recipients of Pfizer -
BioNTech COVID -19 vaccine within the VHA database, which will increase over time with
subsequent anal yses. As of January 21, 2021, 112,201 doses of Pfi zer-BioNTech COVID -19
vaccine have been administered within the VHA (based on CPT code 91300) to a total of
107,458 patients.
Data analy sis: A stepwise approach, illustrated in the diagram , will be performed for signal
detection, evaluation, and verification.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073221
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 13of 144
Notes:
[1] List of safety events of interest and corresponding definitions may be refined as the study progresses based on additiona l
available information.
[2] The risk and cont rol intervals selected for the SCRI analysis for each safety event of interest are based on biological
plausibility and precedents in the literature. Only the individual’s first instance during the specified clean window ( i.e., the
interval used to define incident outcomes) will be included. Note that only the first inpatient or outpatient occurrence of a
safety event of interest following the clean window will be used to identify incident events ( e.g., if an inpatient safety event
of interest occurs in the clean window, a repeat occurrence will not be counted in the risk interval). However, event
worsening will be counted as a safety event of interest. For example, if an outpatient safety event of inte rest occurs in the
clean window and an inpatient occurrence for the same type of safety event of interest occurs in the risk interval, the
inpatient occurrence will be counted as a safety event of interest.
1) Signal detection : The goal is to provide rapid- cycle, near real -time safety surveillance. In
the signal detection phase, the SCRI analysis will only include pre -vaccination control
intervals as the post -vaccination control intervals will require a longer time to accumulate
andwill be used in the sig nal evaluation phase. To account for multiple testing and bi-weekl y
review of the data, the maximized sequential probability ratio test (MaxSPRT )using a
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073222
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 14of 144binomial probability model will be applied. For comparison with individuals who received
seasonal infl uenza vaccination, the Poisson -based MaxSPRT will be applied.
Sequential anal yses for each safet y event of interest will commence once at least 3 events
occur. This approach is consistent with the FDA’s COVID -19 Vaccine Safety Surveillance
Projectto avoid spurious signals from a few early events.14Signals will be detected if the
critical values are reached via the SCRI or active comparator anal ysis.Critical values will be
determined for each safety event of interest base d on historical incidence rate, expected upper
limit of the number of events under the null hy pothesis, and pre -specified significance level
and power. Incidence rates will also be calculated and Kaplan -Meier methods wi llbe used to
analyze time to safety event of interest .
2) Signal evaluation: If signals are detected for safety events of interest based on the anal ysis
described above, further evaluation will be conducted to refine and confirm such detections.
This will include comprehensive quality assurance (for example, check for possible
duplications of claims or medical records, checking for unusual clustering in claim or
medical record accrual by service date for potential coding issues, check for geographical
distribution of cases that may be related to lot numbers or diagnostic practice ) and
multivariate adjustment using Poisson regression to account for baseline differences between
Pfizer-BioNTech COVID- 19 vaccinated and active comparator cohorts. SCRI analyses using
the post-vaccination control intervals will be conducted as an additional inferential analy sis
once enough post -vaccination time has accumulated. Lastly, the assessment of temporal
clustering will also be conducted.
3) Signal verification: diagnostic validation of the detected safety eventsof interest via
adjudication of medical records by VHA clinicians for outcome verification will be
conducted in a representative sample of cases. For rare events, potentiall y all cases may be
adjudicated.
End-of-season anal yses (over the course of the 30 -month period) and an end-of- surveillance
analysis (i.e., at 30 months) will be conducted. Various subgroup anal yses will also be
conducted, examining different age groups, immunocompromised individuals, individuals
with specific comorbidities, those who only received one dose of the Pfizer -BioNTech
COVID-19 vaccine, those with prior SARS -CoV-2 infection based on medical history or pre-
vaccination serology , those receiving care regularly at VA facilities, and lastly those with VA
Priority group 1 statu s, which determines these individuals are of highest priority for VHA
care and likel y receive all of their care within the VHA s ystem.
Milestones:
Registration in the EU PAS register: To be registered before the start of data
collection;
VHA Cooperative R esearch and Data Agreement ( CRADA) and Institutional Review
Board (IRB) approvals (estimated) : March-April 2021 ;
Start of data collection( estimated planned date for starting data extraction for
analysis): May2021;
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073223
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 15of 144Interim reports: 30 June 2021; 31 December 2021; 30June 2022, 31 December 2022;
End of data collection ( estimated planned date for final data cut): 10 June2023;
Final study report: 31 December 2023
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073224
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 16of 144SUMMARY
Objective Primary 1 Primary 2 Secondary
Aim To assess whether individuals in the
Veterans Health Administration
(VHA) system experience increased
risk of safety events of interest
following receipt of the Pfizer -
BioNTech COVID -19 vaccine. To assess whether sub -cohorts of
interest (i.e.immunocompromised,
elderly, with specific comorbidities,
individuals receiving only one dose
of the Pfizer -BioNTech COVID -19
vaccine, and individuals with prior
SARS-CoV-2 infection ) in the VHA
system experience increased risk of
safety events of interest following
receipt of the Pfizer -BioNTech
COVID-19 vaccine. To characterize utilization patte rns of
thePfizer-BioNTech COVID -19
vaccine among individuals within the
VHA including estimating the
proportion of individuals receiving
vaccine, 2 -dose vaccine completion
rate, and distribution of time gaps
between the first and second dose,
demographics and health histories of
recipients, overall and among the
sub-cohorts of interest.
Study design This post-EUA active safety surveillance program will employ a rapid -cycle, longitudinal, observational cohort study
design to provide early real -world safety information.
The self-controlled risk interval (SCRI) design to sequentially monitor occurrence of safety events of interest
while controlling for time -invariant confounders. This design allows inclusion of either a pre -vaccination
control interval or a p ost-vaccination control interval, depending on the safety event of interest (e.g., post-
vaccination control intervals are used for outcomes where there is concern for bias due to indication or
contraindication);
An active comparator design will be used to sequentially monitor occurrence of safety events of interest with
Pfizer-BioNTech COVID -19 vaccinations as compared to recipients of influenza vaccine in the VHA during
2014/2015 through 2018/2019 flu seasons. Data in peri- COVID time periods from January 2020 to present are
excluded because of pandemic -associated under -utilization of health resources and under -reporting of medical
events.
Study population The study will be kept as broad as possible in order to capture safety events of interest that occur among vaccinated
individuals.
Inclusion criteria:
Record of at least one dose of Pfizer -BioNTech COVID -19 vaccine in the period of December 11, 2020 to
present, or
Record of at least one dose of seasonal influenza vaccine during prior flu seasons, from 2 014/2015 through
2018/2019 ( applies to active comparator s only); and
At least 1 year of enrollment in and no disenrollment from VHA benefits ( i.e., the baseline period) prior to
Pfizer-BioNTech COVID -19 or seasonal influenza vaccination date.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073225
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 17of 144Exclusion criteria:
Individuals who receive at least one dose of Pfizer -BioNTech COVID -19 vaccine in addition to a COVID -19
vaccine from a manufacturer other than Pfizer- BioNTech will be identified and reported, but they will be
excluded from further analysis.
Study Period The study will be conducted for a period of 30 months, starting on December 11, 2020 onward, with data collection
concluding on June 10, 2023.
Exposure Administration of Pfizer -BioNTech COVID -19 vaccine post -EUAapproval will be identified based on records of the
following:
Current Procedural Terminology (CPT) code 91300 (Severe acute respiratory syndrome coronavirus 2
(SARS-CoV-2) (coronavirus disease [COVID- 19]) vaccine, mRNALNP, spike protein, preservative fre e, 30
mcg/0.3mL dosage, diluent reconstituted, for intramuscular use) and associated vaccine administration
HCPCS codes corresponding to the first dose: 0001A (ADM SARS -CoV-2 30 mcg/0.3mL 1st), and the second
dose: 0002A (ADM SARS -CoV-2 30 mcg/0.3mL 2nd); OR
10 and 11 -digit National Drug Codes (NDCs) 59267 -1000-1 (corresponds to first dose), 59267 -1000-01
(corresponds to second dose); OR
Immunization records that contain data on vaccine code descriptor, vaccine manufacturer ( i.e., Pfizer), lot
number, injec tion site, and date(s) of immunization;
Administration of the seasonal influenza vaccine during 2014/2015 through 2018/2019 flu seasons will be identified
based on records of the following:
CPT codes
o90654 (Influenza virus vaccine, trivalent (IIV3), spli t virus, preservative -free, for intradermal use);
OR
o90656 (Influenza virus vaccine, trivalent (IIV3), split virus, preservative free, 0.5 mL dosage, for
intramuscular use); OR
o90658 (Influenza virus vaccine, trivalent (IIV3), split virus, 0.5 mL dosage, f or intramuscular use);
OR
10 and 11 -digit NDCs; OR
Immunization records that contain data on vaccine code descriptor, vaccine manufacturer, lot number,
injection site, and date(s) of immunization.
Safety Events of Interest Safety events of interest for active surveillance were identified based on the Priority List of Adverse Events of Special
Interest from the Brighton Collaboration’s Safety Platform for Emergency vACcines (SPEAC) Project, the FDA and
the Centers for Disease Control and Prevention’s (CDC) Advisory Committee on Immunization Practices (ACIP)
enhanced safety monitoring recommendations. The list of safety events may be revised over the course of the study,
and if unanticipated potential safety events of interest are identified during the course of surveillance, they will be
added to the list and included in the analyses. The risk and control intervals for each safety event of interest are based
on biological plausibility and precedents in the literature. Outpatient (including emergency dep artment) and/or
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073226
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 18of 144inpatient settings will be used to identify safety events of interest depending on the type of event. Safety events of
interest can be assigned to 1) the risk interval following vaccination Pfizer -BioNTech COVID -19 vaccination, 2) the
pre-vaccination self -control interval, 3) the post -vaccination self -control interval, or 4) risk interval for the active
comparators of receiving seasonal influenza vaccine. Events outside the intervals will not be counted. Only the
individual’s first instance of a safety event of interest following a specified clean window (i.e. , the occurrence -free
baseline period used to define incident outcomes during which individuals enter the study cohort only if the safety
event of interest did not occur during this peri od) will be included; this means that if a safety event is identified but
diagnosis codes corresponding to the safety event are also observed during the clean window, it will not be counted .
The duration of the pre -specified clean w indow will differ by typ e of safety event of interest in order to rule out pre -
existing events.
Neurologic
Generalized convulsions/seizures;
Guillain-Barré syndrome (GBS);
Aseptic meningitis;
Encephalitis/encephalomyelitis;
Other acute demyelinating diseases;
Transverse myelitis (TM);
Multiple sclerosis (MS);
Optic neuritis (ON);
Bell’s palsy
Immunologic
Anaphylaxis;
Vasculitides;
Arthritis and arthralgia/joint pain;
Multisystem inflammatory syndrome in adults (MIS -A);
Kawasaki disease (KD);
Fibromyalgia;
Autoimmune thyroiditis
Cardiac
Myocarditis;
Pericarditis;
Acute myocardial infarction (AMI)
Hematologic
Thrombocytopenia;
Disseminated intravascular coagulation (DIC)
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073227
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 19of 144COVID-19 (for all COVID- 19-related safety events of interest listed below, an inpatient diagnosis of COVID -
19 will be required in combination with the codes or laboratory values specified in Appendix Table 2; in
addition, COVID- 19 related safety events of interest will only be evaluated using data from 2020 onw ard
using the SCRI design ):
Severe COVID- 19 disease;
Microangiopathy;
Heart failure and cardiogenic shock;
Stress cardiomyopathy;
Coronary artery disease (CAD);
Arrhythmia;
Deep vein thrombosis (DVT);
Pulmonary embolus;
Cerebrovascular hemorrhagic stroke;
Cerebrovascular non -hemorrhagic stroke;
Limb ischemia;
Hemorrhagic disease;
Acute kidney injury;
Liver injury;
Chilblain-like lesions;
Single organ cutaneous vasculitis;
Erythema multiforme
Other
Death;
Narcolepsy/cataplexy;
Non-anaphylactic allergic reactions;
Appendicitis
Data source The VHA Corporate Data Warehouse (CDW) database w ill be used.
Data analysis A stepwise approach w ill be performed for signal detection, evaluation, and verification.
1) Signal detection: The goal is to provide rapid -cycle, near real -time safety surveillance. In the signal detection phase,
the SCRI analysis will only include pre -vaccination control intervals as the post -vaccination control intervals will
require a longer time to accumulate and will be used in the signal evaluation phase. To account for multiple testing and
bi-weekly review of the data, the maximized sequential probability ratio test (MaxSPRT) using a binomial probability
model will be applied. For compar ison with individuals who received seasonal influenza vaccination, the Poisson -
based MaxSPRT w ill be applied.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073228
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 20of 144Sequential analyses for each safety event of interest will commence once at least 3 events occur. This approach is
consistent with the FDA’s COVI D-19 Vaccine Safety Surveillance Project to avoid spurious signals from a few early
events. Signals will be detected if the critical values are reached via the SCRI or active comparator analysis. Critical
values will be determined for each safety event of interest based on historical incidence rate, expected upper limit of
the number of events under the null hypothesis, and pre -specified significance level and pow er. Incidence rates will
also be calculated and Kaplan -Meier methods will be used to analyze t ime to safety event of interest.
2) Signal evaluation: If signals are detected for safety events of interest based on the analysis described above, further
evaluation will be conducted to refine and confirm such detections. This will include comprehensive quality assurance
(for example, check for possible duplications of claims or medical records, checking for unusual clustering in claim or
medical record accrual by service date for potential coding issues, check for geographical distribution of cases that may
be related to lot numbers or diagnostic practice) and multivariate adjustment using Poisson regression to account for
baseline differences between Pfizer -BioNTech COVID -19 vaccinated and active comparator cohorts. SCRI analyses
using the post -vaccination control intervals will be conducted as an additional inferential analysis once enough post -
vaccination time has accumulated. Lastly, the assessment of temporal clustering will also be conducted.
3) Signal verification: diagnostic validation of the detected safety events of interest via adjudication of medical records
by VHA clinicians for outcome validation will be conducted in a representative sample of cases. For rare events,
potentially all cas es may be adjudicated.
End-of-season analyses (over the course of the 30 -month period) and an end -of-surveillance analysis ( i.e., at 30
months) will be conducted. Various subgroup analyses will also be conducted, examining different age groups,
immunocompr omised individuals, individuals with specific comorbidities, those who only received one dose of the
Pfizer-BioNTech COVID -19 vaccine, those with prior SARS -CoV-2 infection based on medical history or pre -
vaccination serology, those receiving care regularl y at VA facilities, and lastly those with VA Priority group 1 status,
which determines these individuals are of highest priority for VHA care and likely receive all of their care within the
VHA system.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073229
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 21of 1445.AMENDMENTS AND UPDAT ES
None.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073230
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 22of 1446.MILESTONES
Milestone Planned date
Registration in the EU PAS register To be registered before the start of
data collection
VHA CRADA and IRB approvals (estimated) March -April 2021
Start of data collection (estimated) May2021[1]
Interim reports 30June 2021
31December 2021
30June 2022
31 December 2022
End of data collection (estimated) 10June 2023[2]
Final study report 31December 2023
Abbreviations : CRADA , Cooperative Research and Data Agreement; IRB, Institutional Review Board;
VHA, Veterans Health Administration.
Notes:
[1] Start of data collection is the planned date for starting data extraction for the purposes of the study
analysis. The initial data analysis will include the Pfizer -BioNTech COVID -19 vaccine exposure since
December 11, 2020, the EUA approval date by the US FDA.
[2] End of data collection is the planned date on which the Pfizer -BioNTech COVID -19 vaccine exposure
reached 30 months post -EUA approval.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073231
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 23of 1447.RATIONALE AND BACKGR OUND
In March 2020, the World Health Organization (WHO) declared a global pandemic for the
coronavirus disease 2019 (COVID -19) due to the severe acute respiratory syndrome
coronavirus 2 (SARS -CoV-2), which was first ide ntified by public health officials in China
in December 2019.1The COVID -19 pandemic presents an unprecedented public health
crisis. As of January 7, 2021, over 21.4million COVID -19 cases and 364,000 deaths have
been reported in the United States (US) alone.2To date, the incidence of COVID -19 has
continued to rise, largel y affecting the elderl y and middle -aged individuals, with worsening
clinical sequelae linked to increasing age and comorbid conditions ( e.g., cardiovascular
disease, active cancer, obesity , diabetes and chronic lung disease).15,16SARS-CoV-2is a
well-adapted highl y infectious human pathogen with a case fatality rate that ranges between
0.5% and 20%, based on the individual’s age, gender, race, and comorbidites.17
Pfizer and BioNTech have partnered to developa novel messenger RiboNucleic Acid
(mRNA)vaccine against SARS -CoV-2for the prevention of COVID -19 (Candidate
BNT162b2). To this end, Pfizer is conducting a Phase 1/2/3, randomized, placebo -controlled,
observer-blind, dose -finding, vaccine candidate- selection, and efficacy study among healthy
individuals (NCT04368728). In their Phase 1 trial evaluating safety and immunogenicit y of
two mRNA vaccine candidates ( i.e., BNT162b1, BNT162b2) at various dose levels,
candidate BNT162b2 was selected for advancement to a pivotal Phase 2/3 safet y and efficacy
evaluation due to its milder systemic reactogenicity profile, especially in older adults.18The
study was initiated in July 2020 with a target enrollment of 43,998 individuals.19
The US Food and Drug Administration (FDA) announced that regulatory emergency use
authorization (EUA) as well as full approval of any COVID-19 vaccine will require
demonstrating prevention of the disease or decrease in its severity in at least 50% of the
individuals who receive it. I n addition, data from Phase 3 studies are required to include a
median follow -up duration of at least 2 months after completion of the full vaccination
regimen to assess the vaccine’s benefit -risk profile, especiall y adverse events and cases of
severe COVID -19 in vaccinated study subjects.20,21The FDA reviewed the available safet y
data of the Phase 1/2/3 trial from 37,586 partici pants 16 y ears of age and older and did not
identify any specific safety concerns. In addition, the anal ysis of available efficacy data from
36,523 participants 12 years of age and older without evidence of prior SARS- CoV-2
infection at least 7 day s after receiving the second dose demonstrated 95% efficacy of the
vaccine in the prevention of COVID -19 (as confirmed by 8 vs. 162 COVID -19 cases in the
vaccine and placebo groups, respectivel y).3,4Based on these safety and efficacy data, as well
as a review of manufacturing information regarding product quality and consistency , the
FDA determined that the known and potential benefits of the vaccine outweighed the known
and potential risks for the pre vention of COVID -19 in individuals 16 y ears of age and older.4
Therefore on December 11, 2020, the Pfizer -BioNTech COVID-19 vaccine was granted an
Emergency Use Authorization (EUA) by the FDA to prevent COVID -19 in individuals 16
years of age and older.5
With respect to geographic regions other than the US, on December 2, 2020, the United
Kingdom (UK) was the first country in the world to grant temporary authorization for
emergency use of the Pfizer -BioNTech COVID -19 vaccine.6On December 21, 2020, the
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073232
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 24of 144European Medicines Agency (EMA) granted the Pfizer -BioNTech COVID- 19 vaccine a
conditional marketing authorization (CMA) for use among individuals 16 years of age and
older throughout all of the European Union’s (EU) 27 member states.7
As required b y the EUA, post-authorization observational studies using real -world data are
needed in order to assess the association between Pfizer -BioNTech COVID- 19 vaccine a nd
pre-determined safet y events of interest (including deaths , hospitalizations, and severe
COVID-19) among individuals administered the vaccine in both the population at large and
in populations of interest ( e.g., immunocompromised individuals, elderl y, and those with
specific comorbidities) .4Post-authorization safety evaluations are important for identify ing
rare, serious safet y events of interest in larger populations that may not have been detected
during clinical trials (either due to sample size or s elected stud y populations), and ensure a
favorable benefit -risk ratio post -trial.Pfizer in collaboration with the US Veterans Health
Administration ( VHA)and Analy sis Group herein propose post -EUA active safet y
surveillance of safet y events of interest ba sed on the Priority List of Adverse Events of
Special Interest from the Brighton Collaboration’s Safety Platform for Emergency vACcines
(SPEAC) Project , the FDA andthe Centers for Disease Control and Prevention’s (CDC)
Advisory Committee on Immunization P ractices (ACI P)enhanced safet y monitoring
recommendation . This safety surveillance study willidentify and evaluate rapid, near real-
time potential safety signals associated with the Pfizer -BioNTech COVID -19 vaccine in the
large-scale VHA electronic medical record ( EMR)database. The observed rates of safet y
event of interest will be compared to expected rates derived from self -controls and active
comparators. Part of the methodologies used in this study are constructed based on
approachespreviouslyused by the Post-Licensure Rapid I mmunization Safety Monitoring
(PRISM) program for the H1N1 vaccine.8This non- interventional study is designated as a
Post-Authorization Safety Study (PASS) commitment to the US FDA andisa Category 3
commitment in the EU Risk Management Plan .
8.RESEARCH QUESTION AND OBJECTIVES
Research question: what are the incidence rates of safety events of interest (based on adverse
events of special interest [AESI ]) among individuals vaccinated with the Pfizer-BioNTech
COVID-19 vaccine within the US VHA system overall and in sub -cohorts of interest a s
compared to expected rates of those events?
Primary study objectives:
To assess whether individuals in the VHA system experience increased risk of safet y
events of interest following receipt of the Pfizer- BioNTech COVID -19 vaccine;
To assess whether sub-cohorts of interest ( i.e., immunocompromised, elderly , with
specific comorbidities, individuals receiving onl y one dose of the Pfizer -BioNTech
COVID-19 vaccine, and individuals with prior SARS -CoV-2 infection) in the VHA
system experience increased risk of safety
events of interest following receipt of the
Pfizer-BioNTech COVID- 19 vaccine.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073233
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 25of 144Secondary study objectiv es:
To characterize utilization patterns of the Pfizer -BioNTech COVID -19 vaccine
among individuals within the VHA including estimating the proportion of individuals
receiving vaccine, 2-dose vaccine completion rate,and distribution of time gaps
between the first and second dose , demographics and health histories of recipients,
overall and among the sub-cohorts of interest.
9.RESEARCH METHODS
9.1.Study Design
This post-EUAactive safety surveillance program will employ a rapid-cycle, longitudinal,
observational cohort study designto provide earl y real-world safet y information . The self-
controlled risk interval (SCRI) design will be used to sequentially monitor occurrence of
safety events of interest while controlling for time- invariant confounders (such as sex, race,
chronic illness, and state). I n addition, safety events of interest associated with Pfizer-
BioNTech COVID -19 vaccinations will be sequentially monitored and compared to
recipients of influenza vaccine in the VHA between 2014/2015 to 2018/2019 .8,22
9.1.1.Self-Controlled Risk Interval (SCRI) Design
The SCRI design uses data from cases ( i.e., individuals who experience safety events of
interest following vaccination) to compare the risk interval following vaccination to pre -or
post-vaccination non -risk intervals (“pre -vaccination control interval” and “post- vaccination
control interval”) in the same individual.23Whether a pre -or post-vaccination control
interval is used will depend on the clinical nature, seasonality , and frequency of the safety
event of interest , as described in greater detail below. A length of 42 day s has been used to
define the risk interval in SCRI design studies for signal detection to ascertain the safet y
profile of the H1N1 vaccine.8,22The same length of risk interval is proposed here, subject to
further modification based on clinical input, clinical trial data, biologic plausibility , and
published literature. The day of vaccination will only be included in the risk period for those
safety eventsof interest for which a same -day occurrence is biologically plausible
(e.g.,anaphylaxis).
As some individuals may choose to decline or delay Pfizer- BioNTech COVID -19
vaccination soon after an illness (known as the “healthy vaccinee effect”),24the
pre-vaccination control interval will exclude the 14- day period before vaccination.25While
using a pre -vaccination control period allows for timely analysis, especially pertinent for
rarer safety events of interest , a post-vaccination control interval would be more appropri ate
and will be used for certain safet y events of interest for the following reasons: (1) a recent
prior safet y event of interest might preclude vaccination ( i.e.,anaphylaxis), (2) individuals
might have an underl ying condition that is also a contraindication for vaccination ( i.e.,
seizure disorder), or (3) safet y event of interest and vaccination may be seasonal in nature.26
The time between the risk and control intervals will be determined based on the biological
mechanism of action for each safet y eventsof interest assessed , and may be subject to change
based on further clinical input. Examples of the SCRI design with a pre -vaccination control
interval and a post -vaccination control interval (in an individual who only receives the first
dose of vaccine) is presented in Figure 1below.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073234
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 26of 144Figure 1. Example of SCRI Design for Assessment of a Safety Event of Interest with a
42-day Risk Interval in an Individual who Receives Only One Vaccine Dose, Showing
Both Pre -and Post-vaccination Control Intervals
*The risk interval may include day 0, date of Pfizer -BioNTech COVID -19 vaccination, for some of the safety
eventsof interest assessed ( e.g., anaphylaxis). The length of the risk interval will vary across each safety event
of interest and may be subject to change based on clinical input. Note that some individuals may not receive the
complete course of vaccination, and thus may only receive the first dose of vaccine. This is represented in
Figure 1while Figure 2represents an example where the complete course with 2 doses are received.
Two doses of the Pfizer -BioNTech COVID -19 vaccine are recommended 3 weeks apart.
This studyprogram will monitor safety eventsof interest that occur after dose 1 before dose
2 (i.e., during risk interval 1), after dose 2 ( i.e., during risk interval 2), and aggregate for
doses 1 and 2 ( i.e., risk interval 1 + risk interval 2), respectively , for individuals receiving
both doses.
For individuals who receive two doses of the vaccine, two separate control intervals will be
defined to correspond to the risk interval associated with each dose (regardless of whether
pre
- or post-vaccination control intervals are used). See Figure 2 below fo r an example in an
individual who receives tw o doses of Pfizer-B ioNTech COVID-19 vaccine, with the second
dose received 21 days after the
first. Safety events of in terest that occur during the
overlapping period of risk interval 1 and risk interval 2 (shown in gray shading in Figure
2)
may be flagged for separate analy ses to discern th
e additive effect of Pfizer-B ioNTech
COVID-19 vaccine dose 1 and dose 2.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073235
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 27of 144Figure 2. Example of SCRI Design with Overlapping Risk Intervals when Two Doses of
Pfizer-BioNTech COVID -19 Vaccine are Administered, Showing a Pre- and
Post-vaccination Control Interval
9.1.2.Active Comparator Design
In the active comparato rdesign, the frequency of safety events of interest among individuals
who received Pfizer -BioNTech COVID -19 vaccine from December 11, 2020 onwardwill be
compared with the event frequency among recipients of the seasonal influenza vaccination in
five prior seasons, between 2014/2015 t hrough 2018/2019. Data in peri -COVID time periods
from January 2020 to present are excluded because of pandemic- associated under -utilization
of health resources and under- reporting of medical events. The same risk interval len gth
(e.g.,42 days)will be used to evaluate safety eventsof interest following vaccination with
Pfizer-BioNTech COVID- 19 vaccine and to assess safet y eventsof interest occurring after
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073236
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 28of 144vaccination for seasonal influenza in prior seasons. The observed num ber of safety eventsof
interest for Pfizer -BioNTech COVID -19 vaccine will be compared to the expected number
calculated for the influenza vaccine in past seasons.8
9.1.3.Study Period
The studywill be conducted for a period of 30 months , starting on December 11, 2020
onward, with data collection concluding on June 10, 2023.
9.2. Setting
The study population will be kept as broad as possible in order to capture safety eventsof
interest that occur among all vaccinated individuals.
9.2.1. Inclusion Criteria
Record of at l east one dose of Pfizer -BioNTech COVID- 19 vaccine in the period of
December 11, 2020 to present, or
Record of at least one dose of seasonal influenza vaccine during prior flu seasons,
from 2014/2015 to 2018/2019 (applies to active comparator s only); and
Atleast 1 year of enrollment in and no disenrollment from VHA benefits ( i.e., the
baseline period) prior to Pfizer -BioNTech COVID- 19 or seasonal influenza
vaccination date.
9.2.2.Exclusion criteria
Individuals who receive at least one dose of Pfizer- BioNTech COV ID-19 vaccine in
addition to a COVID -19 vaccine from a manufacturer other than Pfizer -BioNTech
will be identified and reported, but they will be excluded from further analy sis.
9.2.3.Subgroups
Safety surveillance may be conducted for subgroups of interest, incl uding, but not limited to:
Immunocompr omised individuals;
Different age groups, with a focus on the elderl y(e.g., <35, 35 - <45, 45 - <55, 55 -
<65, 65 -<75, >75);
Individuals with specific comorbidities;
Individuals receiving only one dose of Pfizer-BioNTech COVID-19 vaccine;
Individuals with prior SARS -CoV-2 infection based on medical history or
pre-vaccination serology ;
Individuals with regular use of VHA medical care, defined as at least two outpatient
(excluding ED, as ED visits may not be considered regular) or inpatient encounters in
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073237
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 29of 144the one year prior to vaccination .The encounters must be separated b y >30days (for
inpatient, by admission date), and at least one must be within six months prior to the
date of va ccination.This will ensure that individuals have ongoing health care
encounters, particularl y near the vaccination date, and regularly receive their
healthcare from VHA facilities, rather than outside facilities that would not be
captured in the CDW;
Individualswho are in the VApriority group 1Veteran. These individuals have either
the highest levels of service connected disability (>50% disabling), are considered
unemploy able, or have received the medal of honor.27Individuals categorized as
priority group 1 are the highest priorit y for VHA care. This will ensure that the
individual is more likely to receive all of their care from a VA facility .
Additional subgroups of interest will be assessed as additional information becomes
available from ongoing clinical trials, Vaccine Adverse Event Reporting Sy stem (VAERS),
and other sources that will inform the Pfizer -BioNTech COVID -19 vaccine safet y profile.
Giventhat VA population has a median age of over 46 y ears for females and is comprised of
approximatel y 90% males , the evaluation of the Pfizer-BioNTech COVID -19 vaccine safet y
during pregnancy , including fetal death and infant outcomes, may have poor feasibility and
will therefore not be conducted.
9.3.Variables
9.3.1.Exposure of Interest
Administration of Pfizer -BioNTech COVID -19 vaccine post-EUA approval will be identified
based on the following:
Current Procedural Terminology (CPT) code 91300 ( Severe acute respiratory
syndrome coronavirus 2 (SARS -CoV-2) (coronavirus disease [COVID-19]) vaccine,
mRNALNP, spike protein, preservative free, 30 mcg/0.3mL dosage, diluent
reconstituted, for intramuscular use ) and associated vaccine administration HCPCS
codes corresponding to the first dose: 0001A (ADM SARS -CoV-2 30 mcg/0.3mL
1st), and the second dose: 0002A (ADM SARS- CoV-2 30 mcg/0.3mL 2nd);9,10OR
10 and 11-digit National Drug Codes (NDCs) 59267-1000- 1 (corresponds to first
dose), 59267-1000- 01 (corresponds to second dose); OR
Immunization records that contain data on vaccine code descriptor, vaccine
manufacturer ( i.e., Pfizer), lot number, injection site, and date(s) of immunization.9
Relevant codes will be continuously reviewed and amended if new codes are added.
Person-time at-risk exposure to the first dose onl y, overlapping first and second doses, and
second dose onl y will be anal yzed separatel y.
Administration of the seasonal influenza vaccine during 2014/2015 through 2018/2019 flu
seasons will be identified based on the following:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073238
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 30of 144CPT codes
o90654 (Influenza virus vaccine, trivalent (IIV3), split virus, preservative -free,
for intradermal use) ; OR
o90656 (Influenza virus vaccine, trivalent (IIV3), split virus, preservative free,
0.5 mL dosage, for intramuscular use) ;OR
o90658 (Influenza virus vaccine, trivalent (IIV3), split virus, 0.5 mL dosage,
for intramuscular use );OR
10 and 11- digit NDCs; OR
Immunization records that contain data on vaccine code descriptor, vaccine
manufacturer, lot number, injection site, and date(s) of immunization.
9.3.1.1.Pfizer-BioNTech COVID -19 Vaccine Groups of Interest
While the primary vaccination group of interest is all in dividuals receiving Pfizer- BioNTech
COVID-19 vaccine (irrespective of receipt of seasonal influenza vaccination), additional
subsets of the study population will be studied, similar to the PRI SM safety surveillance
program of H1N1 vaccine safet y:8
Cohort A: Individuals vaccinated with Pfizer -BioNTech COVID- 19 vaccine who did not
receive the influenza vaccine during the flu season in which COVID -19 vaccination
occurred;
Cohort B: Individuals vaccinated with Pfizer -BioNTech CO VID-19 vaccine who received
the seasonal influenza vaccine at least 42 day s prior to COVID -19 vaccination during the
same flu season in which COVID -19 vaccination occurred;
Cohort C: Individuals vaccinated with Pfizer -BioNTech COVID- 19 vaccine who received
the seasonal influenza vaccine within 42 day s before or any time after COVID -19
vaccination during the same flu season in which COVID- 19 vaccination occurred;
Cohort D: Individuals vaccinated with both Pfizer -BioNTech COVID -19 vaccine and the
seasonal inf luenza vaccine on the same day .
The following sub -cohorts will be assessed for each of the Cohorts A -D:
Individuals vaccinated with only 1 dose (i.e., incomplete course) of Pfizer -BioNTech
COVID-19 vaccine;
Individuals vaccinated with 2 doses ( i.e., comple te course) of Pfizer -BioNTech
COVID-19 vaccine.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073239
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 31of 1449.3.2.Baseline Characteristics
The following data elements regarding baseline demographic and clinical characteristics will
be assessed based on a 1-year baseline period prior to the date of vaccination with Pfizer -
BioNTech COVID -19 vaccine and date of seasonal influenza vaccination for active
comparator s.Depending on the attrition rate, the length of the baseline period may be
modified to 6 months. All diagnoses, procedures, and medications will be identified by the
International Classification of Diseases, Tenth Revision, Clinical Modification ( ICD-10-CM)
diagnosis codes, ICD -10-PCS (procedure coding system) codes, ICD-10-CM Current
Procedural Terminology (CPT), orHealthcare Common Procedure Coding Sy stem (HCPCS )
procedure codes, and generic drug names, as appropriate( Appendix Table 1). The following
demographic and clinical characteristics will be assessed:
Demographic s:
Age
Sex
Race/ethnicit y
State
Clinical characteristics:
Smoking status
Body mass index (BMI)
History of anaphylaxis/
allergic reactions
Previous anaph ylaxisof vaccine component
History of hospitalizations
Charlson c omorbidity index (CCI)
Selected c omorbidities
oAutoimmune disease
oAsthma
oBleeding diathesis or condition associated with prolonged bleeding
oCancer
oCardiovascular conditions
oChronic kidney disease/dialy sis
oChronic obstructive pulmonary disease (COPD)/interstitial lung disease
oDiabetes mellitus
oDown syndrome
oSickle cell disease
oHepatitis B virus ( HBV)
oHepatitis C virus ( HCV)
oHuman immunodeficiency virus (HIV)
oHyperlipidemia
oHypertension
oLiver disease
oNeurological disease
oOther immune deficiencies
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073240
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 32of 144oSolid organ transplant
oVenous thromboembolism (VTE)
Concurrent immunizations
oSeasonal influenza vaccine
oTetanus diphtheria and pertussis (Tdap or Td)
oChickenpox (varicella)
oShingles (herpes zoster recombinant and/or live)
oHuman papillomavirus (HPV)
oPneumococcal conjugate
oPneumococcal polysaccharide
oHepatitis A
oHepatitis B
oMeningococcal conjugate (MenACWY) and serogroup B meningococcal
(MenB)
oHaemophilus influenza type b
9.3.3.Outcomes
The safety events of interest for active surveillance wereidentified based on the Priority List
of Adverse Events of Special Interest from the Brighton Collaboration’s Safety Platform for
Emergency vACcines (SPEAC) Project, the FDA and Centers for Disease Control and
Prevention (CDC) enhanced safet y monitoring recommendations.28,29Endpoints of special
interest in signal detection, as noted by the FDA and CDC’s Advisory Committee on
Immunization Practices (ACIP) are denoted in italics .29If unanticipated pote ntial safety
eventsof interest are identified during the course of surveillance, they will be added to the
list and included in the analy ses.See Appendix Table 2for the operational definitions of the
outcome variables based on I CD-10-CM diagnosis codes, which may be refined as the study
progresses based on additional available information and the publ ished literature (e.g.,
frequency of ICD-10-CMcodes).Outpatient (including ED) and/orinpatient setting s will be
used to identify safety events of interest, depending on the t ype of event. The specific
encounter setting considered for each safet y event of interest is summarized in Table 1.Any
record of death will be captured, regardless of whether the individual died in a healthcare or
non-healthcare setting. The following safet y events of interest will be assessed:
Neurologic :
Generalized convulsions/seizures
Guillain-Barré syndrome (GBS)
Aseptic meningitis
Encephalitis/encephalomyelitis
Other acute dem yelinating diseases
Transverse m yelitis (TM)
Multiple sclerosis (MS)
Optic neuritis (ON)
Bell’s pals y
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073241
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 33of 144Immunologic :
Anaphylaxis
Vasculitides
Arthritis and arthralgia/joint pain
Multisystem inflammatory syndrome in adults (MIS-A)
Kawasaki disease (KD)
Fibromyalgia
Autoimmune thy roiditis
Cardiac:
Myocarditis
Pericarditis
Acute myocardial infarction (AMI)
Hematologic :
Thrombocytopenia
Disseminated intravascular coagulation (DIC)
COVID-19(for all COVID -19-related safet y eventsof interest listed below, an inpatient
diagnosis of COVID -19 will be required in combination with the codes or laboratory values
specified in Appendix Table 2; in addition, COVID- 19 related safety events of interest will
only be evaluated using data from 2020 onward using the SCRI design):
Severe COVID -19 disease
Microangiopath y
Heart failure and cardiogenic shock
Stress cardiom yopathy
Coronary artery disease (CAD)
Arrhythmia
Deep vein thrombosis (DVT)
Pulmonary embolus
Cerebrovascular hemorrhagic stroke
Cerebrovascular non -hemorrhagic stroke
Limb ischemia
Hemorrhagic disease
Acute kidney injury
Liver injury
Chilblain-like lesions
Single organ cutaneous vasculitis
Erythema multiforme
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073242
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 34of 144Other:
Death
Narcolepsy/cataplexy
Non-anaphylactic allergic reactions
Appendicitis
The risk and control intervals selected for the SCRI anal ysis for each safety event of interest
arebased on biological plausibility and precedents in the publ ished literature ( Table 1). A
safety event of interest will only be counted if it can be assigned to 1) the risk interval
(following Pfizer -BioNTech COV ID-19 vaccination, 2) the pre -vaccination control interval,
3) the post -vaccination control interval, or 4) the risk interval for the active comparators
receiving seasonal influenza vaccine. Events outside the intervals will not be counted. Onl y
the individual’s first instance of a safet y event of interest following a specified clean window
(i.e., the occurrence -free baseline period used to define incident outcomes during which
individuals enter the study cohort only if the safety event of interest did not occur during this
period) will be included ; this means that if a safet y event of interest is identified but
diagnosis codes (or laboratory values in the case of se lect safety events of interest)
corresponding to the safety event of interest are also observed during the clean window, it
will not be counted. The duration of the pre -specified window will differ by safet y events of
interest in order to rule out pre -existing events. This approach is consistent with the FDA’s
COVID-19 Vaccine Safety Surveillance Project.14By way of example, a safety events of
interest for the SCRI design can be considered in the following way s:
If a safetyevent of interest occurs in the individual’s pre- vaccination control
interval and there are no other diagnosis codes for the same safety event of
interest in the clean window ( e.g., 1-year prior to that date), the safet y event of
interest should be assign ed to the pre -vaccination control interval.
oIf a safety event of interest occurs in the pre -vaccination control interval
but another diagnosis code for the same safety event of interest is
identified during the risk interval, then the safet y event of inter est will not
be assigned to the risk interval and will onl y be assigned to the pre -
vaccination control interval as it will have occurred in the required clean
window preceding the risk interval. However, if an outpatient safety event
of interest occurs in the clean window and an inpatient occurrence for the
same type of safety eventof interest occurs in the risk interval, the
inpatient occurrence will be counted in order to capture event
exacerbation.
If a safety event of interest occurs in the risk inter val and there are no other
diagnoses for the same safety event of interest in the clean window ( e.g., one-year
prior to this date), which also includes the pre -vaccination control interval, then
the safety event of interest will be assigned to the risk int erval.
The same approach will be applied for the post- vaccination control intervals.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073243
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 35of 144The risk intervals for outcome evaluation for the active comparators who received
seasonal influenza vaccination will be the same as for the individuals who received Pfizer-
BioNTech COVID -19 vaccine.
However, it is possible that some safety eventsof interest do not have a precise time
interval from which to evaluate risk, for example if biological plausibility is unknown or the
diagnostic time window is more delayed than anticipated. In these cases, misspecification of
the risk (and control) intervals could result in misclassification and introduce bias, often
toward the null. For instance, the assumption of a longer risk interval than is true may result
in “washing out” the signal, and an erroneousl y short risk interval may similarly result in
underestimation of effect when using post -vaccination time intervals for self -control. To
address this, sensitivity analyses may be conducted with vary ing risk intervals (longer as well
as shorter) in order to increase the likelihood that the safet y risk is detected accuratel y.
Additionally , if further refinement and evaluation is necessary , temporal scan statistics may
be used to empiricall y identify the at-risk time int erval by evaluating clusters of safety events
of interest. This will be further described in the SAP.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073244
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 36of 144Table 1.Outcome algorithms for SCRI analysis , with risk and control intervals
Safety Event of Interest * Setting
(Inpatient [IP],
Outpatient [OP])Clean
windowPre-vaccination
control interval
(days)Risk
interval
(days)Post-
vaccination
control interval
(days)
Neurologic
Generalized convulsion/seizures8IP or OP86 months N/A 0-14 15-29
GBS8,22IP, primary position
only141 year N/A 1-42 43 -84
Aseptic meningitis30IP only141 year N/A 1-42 43-84
Encephalitis/ encephalomyelitis8IP only141 year -56 through -15 1-42 N/A
Other acute dem yelinating diseases8IP or OP81 year -98 through -15 1-42 N/A
TMaIP only141 year -98 through -15 1-42 N/A
MS8,22IP or OP81 year -98 through -15 1-42 N/A
ON8,22IP or OP81 year -98 through -15 1-42 N/A
Bell’s pals y8,22IP or OP141 year -56 through -15 1-42 N/A
Immunologic
Anaphylaxis8,22IP or OP146 months N/A 0-2 7-9
VasculitideseIP only 1 year N/A 1-28 29-56
Arthritisand arthralgia /joint paincIP or OP 1 year N/A 1-42 43-84
MIS-AbIP only141 year N/A 1-42 43-84
KD31IP only311 year N/A 1-28 29-56
FibromyalgiacIP or OP 1 year N/A 1-42 43-84
Autoimmune thy roiditiscIP or OP 1 year N/A 1-42 43-84
Cardiac
Myocarditis8,22IP or OP141 year -56 through -15 1-42 N/A
Pericarditis8,22IP or OP141 year -56 through -15 1-42 N/A
AMIdIPonly141 year -56 through -15 1-42 N/A
Hematologic
Thrombocy topenia30IP or OP141 year N/A 1-42 43-84
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073245
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 37of 144Table 1.Outcome algorithms for SCRI analysis , with risk and control intervals
Safety Event of Interest * Setting
(Inpatient [IP],
Outpatient [OP])Clean
windowPre-vaccination
control interval
(days)Risk
interval
(days)Post-
vaccination
control interval
(days)
DICeIPonly141 year N/A 1-42 43-84
COVID-19(for all COVID -19-related safet y events of interest listed below, an inpatient diagnosis of COVID -19 will be required in
combination with the codes or laboratory values specified in Appendix Table 2; in addition, COVI D-19 related safet y events of
interest will only be evaluated using data from 2020 onward using the SCRI design):
Severe COVID -19 diseasebIP only 1 year N/A 1-42 43-84
Microangiopath yeIP only 1 year N/A 1-42 43-84
Heart failure and cardiogenic shockdIP only 1 year -56 through -15 1-42 N/A
Stress cardiom yopathydIP only 1 year -56 through -15 1-42 N/A
CADdIP only 1 year -56 through -15 1-42 N/A
ArrhythmiadIP only 1 year -56 through -15 1-42 N/A
DVTeIP or OP141 year N/A 1-42 43-84
Pulmonary emboluseIP or OP141 year N/A 1-42 43-84
Cerebrovascular hemorrhagic stroke8IP only141 year N/A 1-42 43-84
Cerebrovascular non -hemorrhagic
stroke8IP only141 year N/A 1-42 43-84
Limb ischemiaeIP only 1 year N/A 1-42 43-84
Hemorrhagic diseaseeIP only 1 year N/A 1-42 43-84
Acute kidney injurygIP only 6 months N/A 1-42 43-84
Liver injurygIP or OP 1 year N/A 1-42 43-84
Chillblain -like lesionseIP or OP 1 year N/A 1-42 43-84
Single organ cutaneous vasculitiseIP only 1 year N/A 1-42 43-84
Erythema multiformefIP only 6 months N/A 1-2 8-9
Other
Narcoleps y and cataplexyaIP or OP141 year -98 through -15 1-42 N/A
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073246
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 38of 144Table 1.Outcome algorithms for SCRI analysis , with risk and control intervals
Safety Event of Interest * Setting
(Inpatient [IP],
Outpatient [OP])Clean
windowPre-vaccination
control interval
(days)Risk
interval
(days)Post-
vaccination
control interval
(days)
Non-anaphylactic allergic
reactions8,22IP or OP86 months N/A 1-2 8-9
Appendicitis32IPonly146 months N/A 0-42 43-84
*Safety events of interestare based on the Priority List of Adverse Events of Special Interest from the Brighton Collaboration’s Safety Platform for
Emergency vACcines (SPEAC) Project, the FDA and the Centers for Disease Control and Prevention’s (CDC) Advisory Committee on Immunization
Practices (ACIP) enhanced safety monitoring recommendations.
Notes:
aPublished risk and control intervals for demyelinating diseases and cranial disorders w ere applied to TM and narcolepsy/catap lexy.
bAs severe COVID- 19 ranges from severe pneumoni a, 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.
c Published risk and control intervals for autoimmune disorders w ere applied to similar autoimmune rheumatic conditions (i.e., fibromyalgia and au toimmune
thyroiditis).
dPublished 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).
eSimilar 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.
fPublished risk and control intervals for non -anaphylactic allergic reactions were applied to hypersensitivity disorders (i.e., erythema multiform e).
gRisk 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 .
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073247
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 39of 1449.4.Data Source
The VHA is the largest integrated health care s ystem in the US, providing both inpatient and
outpatient clinical care to over 9 million Veterans enrolled at more than 170 medical centers
and 1,074 community -based outpatient clinics.12VHA’s health care delivery system is
organized regionall y around 18 Veterans Integrated Service Networks (VISNs) across the
US. Each VISN is responsible for health care planning and resource allocation in a particular
geographical region. For example, the VA New England Healthcare S ystem (VISN 1) covers
VHA facilities in Massachusetts, Connecticut, New Hampshire, Maine, and Rhode I sland,
while the VA Heart of Texas Health Care Network (VI SN 17) oversees the facilities in
Texas.
The VHA also maintains its own mortality data where 99% of enrollees’ deaths are reported
within one month of occurrence .As of January 7, 2021, the VHA has ha d over 174,000
confirmed COVID -19 cases.33Among active and convalescent cases , approximately 145,000
are Veterans and approximately 15,000are employees (with an estimated 630 as Veteran
employees).33While African America n Veterans make up approximately 12% of the VHA,34
the burden of COVID -19 cases are skewed, with African American Veterans comprising
approximately 20% of all COVID -19 cases.33Approximately 7,099COVID-19-infected VA
patientshave died, an estimated 2,738in VHA hospitals.33
The objectives of this study will be addressed using data from VHA’s Corporate Data
Warehouse (CDW), which is an integrated EMR system with a centralized data warehouse
that is updated on a dail y basis. The CDW stores data in separate databases, one for each t ype
of clinical information ( e.g., inpatient medication, inpatient admission, outpatient medication,
outpatient visit). Individualdemographic information such as date of birth and gender are
also available. Immunization records include information on manufacturer, lot number,
injection site, and concurrent immunizations. The CDW does not include information on an y
care receiv ed outside of a VHA facility .
Each individual is assigned a unique identification number to allow for longitudinal
follow-up as well as to cross -reference to the various separate databases. For example, in
each inpatient admission record, there is informa tion on the primary discharge diagnosis (and
as many as 15secondary diagnoses), date of admission, date of discharge, and length of stay .
This record can then be linked to other information of that inpatient stay located in other
files, including procedur es that the patient underwent during the hospitalization, medical
specialty of the provider, and prescriptions dispensed. Other files are similarly structured,
and therefore may be linked together to provide comprehensive information about the patient
and his/her medical encounters.
The VHA database is an appropriate data source to evaluate the safet y of the Pfizer -
BioNTech COVID-19 vaccine for the following reasons. First, as the vaccine will be
distributed through government facilities (including VHA) as part of initial distribution,
analysis of VHA data will provide earl y data on the safet y of the vaccine. Veterans living in
long-term care facilities and Veteran s who are healthcare workers will be prioritized in the
first wave of Pfizer -BioNTech COVID -19vaccinations.35The VA offers eligible Veterans
long-term care services ranging from nursing homes and assisted- living centers to caregiver
support in the Veterans’ own homes.13Secondly, and relatedl y, VHA data are refreshed dail y
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073248
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 40of 144and would thus enable early and rapid data anal ysis. Third, the VHA population is on
average older than the general US population.36Of these, about 30% (roughly
1,000,000 individuals) use VHA health services almost exclusively (i.e., those with a priority
group of 1 or 4; Veterans assigned to Priorit y group 4 are either accepting VA assistance or
housebound benefits, or have been determined to be “catastrophicall y disabled” b y the
VA.27), which lends itself to having complete, longitudinal healthcare data for such
individuals who may be at higher risk of COVID -19 due to older age.37,38These priorit y
groups include Veterans with the highest levels of service- connected disability and are
therefore, the highest priority for VHA care.27Finally, the VHA population ha s, on average,
more comorbid conditions t han the general population, which also indicates that these
individuals may be at higher risk of COVID -19.39While the VHA pop ulation is
predominantly male (approximately 90%), and thus lacks generalizability to females, it will
still provide a useful setting to examine real -world vaccine safet y.
9.5.Study Size
The sample size achieved will depend on the number of recipients of Pfizer -BioNTech
COVID-19 vaccine within the VHA database during the study period, which will increase
over time with subsequent analy ses. The population size will increase with each bi-weekl y
analysis as the Pfizer -BioNTech COVID-19 vaccine becomes more readily available and a
greater number of individuals are vaccinated. Specificall y, the data wi ll be refreshed on a
biweekly basis and a continuous sequential test procedure will be used to reevaluate data
according to this schedule. As of January 21, 2021, 112,201 doses of Pfizer -BioNTech
COVID-19 vaccine have been administered within the VHA (base d on CPT code 91300) to a
total of 107,458 patients.
As a result of the ability to perform near -real-time analysis, the risk interval (and post -
vaccination control interval, for applicable safety events of interest ) may have only partially
elapsed in some cases. To account for this, we will use methods adopted in previous
studies,8,25,40whereby risk intervals will be scaled (or truncated) in order to ensure an
equivalent length (or a fixed ratio) of time is assessed between the control and risk intervals.
9.5.1.Power
Power calculations for the rapid cy cle analysis (RCA) approaches proposed for safety event
of interest signal detection will be conducted according to t he methods of Kulldorff et al.41,42
Table 2 illustrates the estimated po wer for the RCA approach using the Poisson -based
maximized sequential probability ratio test (MaxSPRT), and provides an overview of the
power required to detect vary ing relative risk (RR ) estimates with an alpha level of 0.01. T
denotes the expected number of safety events of interest to occur du ring the risk interval of
interest (Table 2 and Table 3). Power of ≥ 80% is typically desirable in drug safety research.
Usually
the FDA views a RR of > 3 as meaningful, so this has been used for power
calculations here.43 As an example, as shown in Table 2, the surveillance system would have
sufficient power (80.0%) to detect an increased risk of
safety events of interest associated
with the Pfizer -BioNTech COVID- 19 vaccine b y 3 fold when the expected number of safety
events of interest
reaches 6 events.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073249
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 41of 144Table2. Estimated Statistical Power for the Poisson -based MaxSPRT41
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073250
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 42of 1449.6.Data Management
Data for this stud y will be stored and extracted from the VHA database (p reviously described
in Section 9.4) that contain information about patient demographics, vaccinations,
procedures, diagnoses, and death.
9.6.1.Case report forms (CRFs)/Electronic data record
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 should be completed for each included patient in the signal
verification phase that requires EMR and chart review ( see Section 9.7.3.3). The completed
original CRFs should not be made available in any form to third parties, except for
authorized representatives of Pfizer or appropriate regulatory authorities, without written
permission from Pfizer. The CRF will consist of two parts: (1) a database CRF that will be
populated based on a direct extraction of data from the VA CDW for review by the
adjudicators; (2) an adjudication page that will be completed by an adjudicator after
reviewing data in the completed CRFs. Analysis Group shall ensure that the CRFs are
securely stored on VHA servers in an encry pted electronic and/or paper] form and will be
password protected or secured in a locked room to prevent access b y unauthorized third
parties.
Analysis Group has ultimate responsibility for the collection and reporting of all clinical,
safety, and laboratory data entered on the database CRFs and an y other data collection forms
(source documents) and ensuring that they are accurate, authentic/original, attributable,
complete, consistent, legible, timely (contemporaneous), enduring, and available when
required. The adjudication page must be signed by the adjudication committee members to
attest that the data contained on the formsare true and accurate based on their review of the
EMR data . Any corrections to entries made in the CRFs or source documents must be dated,
initialed, and explained (if necessary) and should not obscure the original entry .
The source documents are the hospital or the ph ysician's chart. In these cases, data collected
on the CRFs must match those char ts.
9.6.2.Record retention
To enable evaluations and/or inspections/audits from regulatory authorities or Pfizer,
Analysis Group agrees to keep all study -related records, which includes study documents and
deliverables such as the protocol, SAP, aggregated results tables, SAS programming files,
and study report. The records should be retained by Analysis Group according to local
regulations or as specified in the vendor contract, whichever is longer. Analy sis Group must
ensure that the records continue to be sto red securely for so long as they are retained.
If Analysis Group becomes unable for any reason to continue to retain study records for the
required period, Pfizer should be prospectivel y notified. The study records must be
transferred to a designee accepta ble to Pfizer.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073251
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 43of 144Study records must be kept for a minimum of 15 years after completion or discontinuation of
the study, unless Analy sis Group and Pfizer have expressly agreed to a different period of
retention via a separate written agreement. Record must be retained for longer than 15 years
if required b y applicable local regulations.
Analysis Group must obtain Pfizer's written permission before disposing of any records, even
if retention requirements have been met.
9.7.Data Analysis
Detailed methodology for summary and statistical anal yses of data analyzedin this study will
be documented in a statistical anal ysis plan (SAP), which will be dated, filed ,and maintained
by the sponsor. The SAP may modify the plans outlined in the protocol; any major
modification s of primary endpoint definitions or their anal yses would be ref lected in a
protocol amendment. The SAP will also provide additional detail regarding the evaluation of
a threshold of excess risk for each of the safet y events of interest. Consistent with th e
approach of Kulldorff et al. , this will be determined based on background incidences for each
event (e.g., based on historical influenza vaccinated active comparator cohort data to be
evaluated during the study), in addition to pre -specified significance level (e.g., alpha=0.01
or 0.05) and power.41This information, in conjunction with a clinically meaningful RR ( e.g.,
2 or 3) and the expected upper limit of events under the null hy pothesis will allow for the
calculation of critical values of each safet y event of interest using the MaxSPRT method.
Greater power ( e.g., 80%) is also a natural criterion to use when selecting the upper limit on
the length of surveillance, and in turn, the expected number of events to occur, although there
is ultimately a tradeoff between that power and the time allowed to identify the expected
number of events to occur.
Data analy ses will be conducted using SAS Enterprise Guide version 7.1 (SAS I nstitute Inc.,
Cary, NC) or R Version 3.5.3 or its latest version (R Core Team, Vienna, Austria). In
addition, SaTScan will also be used to conduct specific temporal anal yses.
9.7.1.Baseline Characteristi cs
Baseline demographics and clinical characteristics for individuals receiving Pfizer -BioNTech
COVID-19 vaccine and individuals who received seasonal influenza vaccination will be
summarized using descriptive statistics, consisting of the mean and standard deviation (SD)
and median (interquartile range [IQR]) values for continuous variables and frequency
distributions for categorical variables. Incidence rates ( i.e., per-patient per -month) for prior
hospitalizations may be calculated as the number o f events divided b y person-time of
observation since the length of the baseline period may vary between individuals.
Standardized differences will be calculated between Pfizer BioNTech COVID -19 vaccine
recipients and active comparators who received seasona l influenza vaccination to evaluate
whether there are an y major differences in individuals’ baseline characteristics. Standardized
differences <10% will indicate that matching has appropriatel y balanced the charac teristics
between recipients of the Pfizer-BioNTech COVID- 19 vaccine and seasonal influenza
vaccine.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073252
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 44of 1449.7.2.Vaccine Utilization Patterns
Descriptive statistics will also be used to summarize vaccine utilization patterns, including
proportion of individuals receiving vaccine, 2 -dose completion rate ,distribution of time gaps
between the first and second dose , and care setting where immunization was received ( e.g.,
outpatient clinic, pharmacy , inpatient ward). Counts of individuals who received a COVID -
19 vaccine from a different manufacturer in addition to the Pfizer -BioNTech COVID-19
vaccine will be reported.
9.7.3.Safety Signal Analyses
Several analy ses corresponding to the designs discussed previousl y will be conducted to
detect safet y signals associated with Pfizer -BioNTech COVID -19 vaccine. Analy ses will be
conducted among all individuals receiving the vaccine, individuals who received
Pfizer-BioNTech COVID-19 vaccine without seasonal flu vaccine (Cohort A will be used for
SCRI; Cohort B +Cwill be used for active comparator analyses), and individuals receivin g
Pfizer-BioNTech COVID- 19 vaccine and seasonal flu vaccine on the same day (Cohort D),
along with sub- cohorts receiving onl y one dose vs. two doses.
A stepwise process, illustrated below, will be performed for signal detection, evaluation, and
verification (Figure 3).This approach has been adapted from the Active Monitoring Protocol
of the FDA’s COVID -19 Vaccine Safet y Surveillance Project.14The statistical approach
described below may be modified further based on data availability , additional clinical input,
and for consistency or to complement similar studies of Pfizer -BioNTech COVID-19
vaccine.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073253
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 45of 144Figure 3. Steps in Signal Detection, Evaluation , and Verification
Notes:
[1] List of safety events of interest and corresponding definitions may be refined as the study progresses based
on additional available information.
[2] The risk and control intervals selected for the SCRI analysis for each safet y event of interest are based on
biological plausibility and precedents in the literature. Only the individual’s first instance during the specified
clean window (i.e., the interval used to define incident outcomes) will be included. Note that only the first
inpatient or outpatient occurrence of a safety event of interest following the clean window will be used to
identify incident events ( e.g., if an inpatient safety event of interest occurs in the clean window , a repeat
occurrence will not be counted i n the risk interval). However, event worsening will be counted as a safety event
of interest. For example, if an outpatient safety event of interest occurs in the clean window and an inpatient
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073254
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 46of 144occurrence for the same type of safety event of interest occurs in the risk interval, the inpatient occurrence will be counted as
a safety event of interest.
9.7.3.1.Signal Detection
9.7.3.1.1. Sequential Testing -SCRI Design using the Binomial -based MaxSPRT for
Comparison to Pre -vaccination Control Intervals
The goal is to provide ra pid-cycle, near real -time safet y surveillance. In the signal detection
phase, the SCRI analysis will only include pre -vaccination control intervals as the post -
vaccination control intervals will require a longer time to accumulate and thus will not allow
for timely analysis. The post -vaccination control period will be assessed during the signal
evaluation phase (see Section 9.7.3.2), to allow for additional observation time to accrue as
well as to more deeply investigate potential signals. This will allow for timely RCA without
the need to wait for data to accumulate for safety events of interest with post-vaccination
control intervals.
To account for multiple testing and bi -weekly review of the data, the MaxSPRT using a
binomial probability model will be applied. The null hy pothesis (H 0) assumes that the risk of
a safety event of interest during the risk interval is equivalent to the risk of the same safety
event of interest developing during the control interval, accounting for differences in interval
duration as needed ( e.g., for safety events of interest such as demyelinating disease), meaning
a RR of 1 is specified under H 0.22The one-sided composite alternative h ypothesis (H a)
assumes that the risk of a safety event of interest during the risk interval is greater than the
risk of the same safety eventof interest developing during the control interval, accounting for
differences in interval duration ( i.e., RR>1, H ais applicable across a range of RRs).41
Specifically, for the Pfizer -BioNTech COVID- 19 vaccine, let xrepresent the total count of
safety events of interest in the control interval ( Figure 4), let y represent the total count of
safety events of interest in the risk i nterval, and let rrepresent the ratio of yto x under the
null hypothesis. Thus, when the total control interval duration and total risk interval duration
are equal, r will be 1. The RR is estimated by ..
.25The RR and corresponding 9 9%
confidence intervals (CIs) will be calculated.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073255
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 47of 144Figure 4. Example of SCRI Design for a Safety Event of Interest with a 42- day Risk
Interval and a Pre- vaccination Control Interval
For the binomial mode l, the log-likelihood ratio (LLR) is calculated as the log probability of
observing this distribution of y under H a, divided by the probability of this occurring under
H0.41This ratio is calculated whenever new data arereceived to account for the continuous
data stream until the full 42 -day risk period is complete.
=ln ( | )
( | 0)
Once the LLR test statistic reaches a pre -specified critical value , a signal is detected.
Specificall y, the null hypothesis will be rejected if the LLR exceeds the critical value. The
null hypothesis will not be rejected if the LLR does not reach or exceed the critical value, if
the total number of safety events of inte restreaches a pre -specified upper limit, or if
surveillance ends without reaching this upper limit.25
For each safety event of interest (and specific to each age group, if age- stratified anal yses are
conducted), the criti cal value of the LLR will be determined based on the safet y event of
interest specific upper limit of expected safety events of interest and alpha level.25Upper
limits will be determined based on the expected number of safety events of interest under the
null hypothesis, assuming the risk after Pfizer -BioNTech COVID- 19 vaccination is no
greater than the risk of safety events of interest after seasonal influenza vaccination.
Therefore, upper limits will be chosen such that t hey would not usually be reached.
9.7.3.1.2. S equential Testing -Poisson-based MaxSPRT for Comparison to Active
Comparators who Received Seasonal Influenza Vaccination
For comparison with active comparators who received seasonal influenza vaccination, the
Poisson-based MaxSPRT will be applied, following the same statistical approach as
described above, but using a Poisson probabilit y distribution. In the Poisson MaxSPRT
approach, the event frequency of safety events of interest in the risk interval after Pfizer-
BioNTech COVID -19 vaccination will be compared to a background rate ofsafety events of
interest in the risk interval after seasonal influenza vaccination in five prior seasons, ranging
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073256
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 48of 144from 2014/15 through2018/19. This approach is particularl y important for extremely rare
safety events of interest (i.e., less than 50 anticipated based on historical influenza vaccine
ratesof safety events of interest ).22Poisson MaxSPRT is used to mo nitor very rare safet y
events of interest as binomial MaxSPRT may not detect a signal, despite a clinically
meaningful RR.25This will also allow for more timely anal ysis using historical data, as well
as improved power and sample size.
GBS is of particular interest relative to the safety profile of Pfizer -BioNTech
COVID-19vaccine. As GBS is an extremely rare safety event of interest , the primary RCA
proposed will focus on Poisson MaxSPRT and apply an alpha of 0.05. The Poi sson
MaxSPRT has increased power to detect a signal with fewer occurrences of the safet y event
of interest. However, this method cannot fully control for confounding by indication.
9.7.3.1.3. Critical Values and Alpha Spending
Critical values for the LLR test statistic are shown below in Table 3based on calculations
conducted b y Kulldorff et al 2011.41For example, assuming T =6 (number of expected
events under the null) and RR =3, which corresponds to a power of 80.0% (See
Section9.5.1), the critical value would be 5.14 using alpha of 0.01 for the Poisson -based
MaxSPRT. As noted previously , each safety event of interest will be evaluated separatel y to
determine a critical va lue based on background incidence, alpha, power, and clinically
meaningful RR. These details will be addressed in the SAP.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073257
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 49of 144Table 3. Critical Values for Poisson -based MaxSPRT
Multiple ty pes of alpha spending functions can be employ ed to calculate the cumulative rate
at which Ty pe 1 error (alpha) probabilit y is spent during sequential testing.44To achieve
optimal expected time -to-signal, especiall y when historical Poisson data are used with
surveillance data, a power-typeconvex alpha spending shape will be used based on published
literature.44Additionally , ρ =1.5 is referenced as a “rule of thumb” as it is suggested to be
appropriate in most applications.
9.7.3.2.Signal Evaluation
Signals are detected when the event frequency of asafety eventof interest during the risk
interval following vaccination with Pfizer -BioNTech COVID -19 vaccine is significantl y
increased compared to the event frequency of the same safety events of interest in the control
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073258
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 50of 144comparator (i.e., the critical value is achieved and surpassed). If signals are indeed detected
for safety events of interest based on the anal ysis described above, further evaluation is
warranted to refine and conf irm such detections. This will include the following additional
analyses to assess the robustness of the findings.
9.7.3.2.1. Post-Signal Quality Assurance
Quality assurance will first be conducted in order to assess the quality of the data and
analysis that produce d the signal. While quality control measures will be conducted during
the signal detection phase (seeSection 9.8), post-signal qualit y assurance will also be
performed during the signal evaluation phase. This will include a comprehensive quality
assurance (for example, check for possible duplications of claims or medical records,
checking for unusual clustering in claim or medical record accrual b y service date for
potential coding issues, check for geographical distribution of cases that may be related to lot
numbers or diagnostic practice) . In addition, for signals detected via active comparison,
additional anal yses comparing to pre- vaccination contro l intervals may be formed to check
for consistency . Signals will also be confirmed across all of the safety studies planned to be
performed (i.e., C4591008, C4591011, C4591012) to confirm that specific data sources are
not biased.
9.7.3.2.2. Multivariate Adjustment using Poisson Regression
If signals are detected and persist after conducting quality assurance, further evaluation via
statistical measures are warranted. Specifically , to investigate whether potential signals
identified via Poisson MaxSPRT for the compar ison to active comparators with seasonal
influenza vaccination are not confounded ( i.e., to take into account baseline differences
between the Pfizer BioNTech COVID -19 vaccinated and active comparator populations), a
multivariate Poisson regression anal ysis will be conducted to compare the incidence rates of
the safety events of interest occurring within the risk intervals. T he predictor would be
whether the individual had received the Pfizer -BioNTech COVID -19 vaccine or had received
the influenza vaccine d uring historical seasons. Analy ses will be adjusted for relevant
baseline and/or clinical characteristics ( e.g., age, sex, race, CCI and/or specific comorbidities
of interest, state, etc.).8
If the signal remains, based on an IRR > 3 with a p- value <0.01 from the adjusted Poisson
regression, further evaluation may be considered via signal verification.
9.7.3.2.3. Assessment of Temporal Clusters
Vaccine safety surveillance must allow for sufficient ty pe I error probability for rapid
detectionofsafety events of interest , and statistically significant signals must be studied
further to ensure that a true association is present.45Therefore, the presence of temporal
clusters will be assessed using the so ftware SaTScan to calculate temporal scan statistic in
order to further refine safety signals detected from the signal detection analy ses.22A temporal
scan statistic accounts for multiple testing present during overlapping risk intervals. The null
hypothesis assumes that there is no association between the safet y eventsof interest and
immunization, and safety events of interest are assumed to be distributed independentl y and
uniformly during a period of time subsequent to Pfizer -BioNTech COVID- 19 vaccination.22
A temporal scan statistic will be generated b y moving a time interval of fixed length across
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073259
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 51of 144the risk interval, comparing the number of observed versus expected safety events of inter est
within the time interval under the null hy pothesis.46
9.7.3.2.4. Sequential Testing -SCRI Design using the Binomial MaxSPRT for
Comparison with Post -Vaccination Control Intervals
Similar to the SCRI design using the binomial -based MaxSPRT method for pre -vaccination
control intervals, sequential testing anal yses will be conducted using the post- vaccination
control intervals as appropriate for specific safety events of interest . This will be conducted
during the signal evaluation p hase in order to allow time to accumulate during the post -
vaccination control period. The same statistical methodology as described for the pre -
vaccination control intervals will be applied.
9.7.3.3.Signal Verification
If a signal persists after conducting signal evaluation , signal verification through medical
records review may be conducted .
9.7.3.3.1. Medical Records Review
As part of the signal evaluation process, diagnostic validation of the detected safet y events of
interest (i.e.,cases) via a djudication of patient medical records by VHA clinicians for
outcome verification in a representative sample of cases will be conducted. The total number
of charts to be reviewed will depend on the number of safety events of interest detected, such
that all cases may be reviewed for safety events of interest where a small number of events
result in signal detection and a representative sub -sample may be reviewed for safety events
of interest where a larger number of events results in signal detection .47For rare events,
potentially all cases may be adjudicated. An adjudication charter will be developed to govern
signal evaluation and medical records review. Specificall y, validation of detected safet y
events of interest will be performed through patient medical chart review in collaboration
withan adjudication committee comprised ofthe treating or trained healthcare
professionals.47
9.7.4.Seasonality- Adjusted Cases -Centered Method
A case-centered anal ysis for specific safety events of interest for which signals were detected
may also be conducted in order to account for bias caused b y seasonality of safety events of
interest and vaccination.23This method will use data on all safety event of interest cases that
occur after vaccination with Pfizer -BioNTech COVID -19 vaccine. Logistic regression will
be used to compare the number of safety event of interest cases that were vaccinated inside
versus outsid e a pre-specified risk interval , as of the date of the safet y events,where the total
number of vaccinations given inside versus outside the risk interval (in the population of all
vaccinees) is used as the offset term.25Specifically, the association of vaccination with risk
of safety events of interest will be estimated from a logistic regression model that includes
summarized data with one record per risk set. The key independent variable will be the
proportion of the risk set who were in the risk interval on the date of the safety eventof
interestoccurrence. In this way , risk sets are anchored to calendar dates, and confounding b y
seasonality of the safet y eventsof interest and vaccination is addressed .48Note that other
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073260
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 52of 144confounders may also be adjusted for b y restricting risk sets to vaccinees similar with respect
to select characteristics ( i.e., through stratification).
9.7.5.End-of-Season and End -of-Surveillance Analyses
For any safety event of interest with signals detected, end-of- season anal yses (over the course
of the 30- month period) and an end -of-surveillance anal ysis (i.e., at 30 months, after the end
of surveillance) will be conducted. Similar methodology will be applied for the end -of-
surveillance anal ysis and end-of- season anal ysis conducted for seasonal influenza vaccine in
order to adjust for the seasonality of both disease and vaccine administration.8This approach
will be able to define the true risk intervals after each dose and estimate the risk for potential
safety events of interest after both dose 1 and 2 of the Pfizer -BioNTech COVID- 19 vaccine,
as well as the abilit y to discern whether or not on e or two doses of seasonal influenza vaccine
were administered during the same period.
The number of events in the sum of three distinct risk intervals will be compared to the
control interval, adjusting for potential differences in interval length, to e stimate the RR of
Pfizer-BioNTech COVID-19 vaccine compared to the influenza vaccine . In order to monitor
the safety after the first and full course of the vaccine, the number of potential safety events
of interest occurring in three separate risk intervals (P 1, P2, P3) will be estimated ( Figure 5).
P1represents the risk interval after the first dose only , excluding any overlap in risk intervals
with the second dose. P 2represents the overlapping risk intervals for first and second dose of
the vaccine. P 3represents the risk interval of the second dose of the vaccine, excluding the
overlapping risk interval alread y captured in P 2. This design will allow fo r the assessment of
risk during the appropriate periods, regardless of the time interval between vaccine doses. As
multiple endpoints will be assessed, 99% CIswill be calculated around the RR in order to
ascertain whether the Pfizer- BioNTech COVID -19 vaccine is associated with safety events
of interest.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073261
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 53of 144Figure 5. Example of Risk (P1, P2, P3) and Pre -vaccination Control Intervals for the
SCRI End -of-surveillance Analyses of 1 or 2 Doses of Pfizer-BioNTech COVID -19
Vaccine
In Figure 5A, P1+ P2+ P3represent the risk intervals where a safet y event of interest may
occur. In Figure 5B,there is no overlapping risk interval so that P 1+ P3represent the risk
intervals where a safety event of interest may occur.The timing of the risk and control
intervals maybe adjusted for in order to control for the effect of seasonalit y across the
intervals assessed.
9.7.6.Subgroup Analysis
Separate anal yses of baseline characteristics, vaccine utilization patterns, signal detection,
signal evaluation, and signal verification insubgroups of interest may be conducted based on
feasibility , sample size, and data available.
9.7.7.Incidence Rates and Time to Safety Event of Interest Analysis
Incidence rates (and corresponding CIs) will be calculated from safety event of interest signal
detection anal yses. Kaplan -Meier methods will be used to anal yze time-to-event (i.e., time to
safety event of interest ). If individuals do not experience thesafet y events of interest , they
will be censored at the end of the risk interval. Median time to safety event of interest and
corresponding CI swill be reported.
9.8.Quality Control
Data for the study will be extracted from electronic databases in the CDW of the VHA. Each
data content area in the CDW is subjected to similar checks, from high level variable
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073262
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 54of 144name/type checks, to detailed trending comparisons. As an example, the diagnostic data is
subject to the following checks:
Referenced table exists
Diagnosis type is correctly assigned b y codes defining the diagnosis
Percentages, rates, are as expected (check ranges and for missing)
Both inpatient and outpatient diagnosis codes are captured. Referenced variables exist
and are of appropriate length and t ype
Data retrieval will be coordinated b y an experienced programmer/anal yst. The anal yst will
write programm ing for retrieval of each data element from the electronic databases. Double
programming will be performed for the first iteration of the anal yses; results/datasets will be
compared, and if an y discrepancies are identified, both programmers will determine a
resolution, bringing in a third programmer if needed. Subsequent iterations of anal yses (i.e.,
re-runs of the anal yses) will be audited by a senior programmer. All tables will be reviewed
by the project manager and the principal investigator to evaluate for internal consistency of
counts and totals. All calculated variables will be checked against the component variables
(cross tabs) to ensure accuracy . For example, categorical age would be compared with
continuous age to confirm that each category of agecontained onl y individuals of the
expected age ranges within that category .
9.9.Strengths and Limitations of the Research Methods
To identify individuals who experienced safety events of interest associated with Pfizer-
BioNTech COVID -19 vaccine, the SCRI method of signal detection offers some key
advantages. The SCRI approach inherentl y adjusts for within-individual confounders, such as
age, sex, and confounding b y indication. Additionally , the inclusion of a post -vaccination
control period will account for in creased detection bias from stimulated reporting of safet y
events of interest due to heightened vigilance on COVID -19 vaccines.49Specificall y, safety
events of interest may be more likely to be reported or sought care for after vaccination with
Pfizer-BioNTech COVID-19 vaccine than before ( i.e., during the pre -vaccination control
interval)which may result in bias against the Pfizer -BioNTech COVID -19 vaccine . Lastly,
SCRI allows for near real- time monitoring of safety risksassociated with the Pfizer -
BioNTech COVID -19 vaccine.
The VHA CDW provides a range of benefits, including its comprehensive structure, large
number of variables, and electronic accessibility. The VHA CDW also includes EMR data
that include structured fie lds (which will be used for signal detection) and open fields (such
as physician notes, which will be used for signal evaluation and case validation, as needed).
Importantly , the VHA CDW retains electronic immunization records that include
manufacturer nam e and lot numbers, facilitating the identification of brand- specific vaccines,
such as the Pfizer -BioNTech COVID -19 vaccine. Moreover, the VHA CDW data are
updated on a dail y basis, enabling near real -time rapid monitoring of potential safet y signals.
However, there are several limitations when rely ing on VHA that should be noted. First,
there could be gaps in the data since individuals may receive healthcare services outside of
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073263
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 55of 144VHA facilities. As such, if individuals receive the Pfizer-BioNTech COVID-1 9 vaccine
outside of a VHA facility, this information will not be captured in the VHA EMR sy stem.
Similarly, individuals may have also received past seasonal influenza vaccinations outside of
the VHA s ystem, and thus would be misclassified as not having re ceived vaccine in the
current anal ysis. For example, veterans with secondary insurance or veterans who are
65years of age or older who have Medicare may receive health care services outside of VHA
facilities. One stud y on VHA enrollees in seven different states found that of all individuals
admitted to VHA hospitals in 2007, one fifth also had a non VHA hospitalization during that
year.50Another study reported that about 53% of Veterans 65 years of age and older who
were dually eligible for VHA and Medicare services in 2003 2004 used both.51Hence, it is
important to note that data on vaccination status ma y be incomplete. However, this limitation
will be addressed b y examining subgroups of individuals who receive care regularl y at VHA
facilities, as well as those with Priority group 1 status, to ensure that their healthcare data are
complete to the extent p ossible in the CDW. Second, to the extent that the individuals in the
VHA database are different from individuals outside of the VHA, the results may not be
generalizable to the broader US population. For example, since the VHA includes
predominantly male Veterans (approximately 90% male), findings from this study may not
be generalizable to women in the US.
9.10.Other Aspects
Not applicable.
10. PROTECTION OF HUMAN SUBJECTS
10.1.Patient Information
All parties will comply with all applicable laws, including laws regarding the implementation
of organizational and technical measures to ensure protection of patient personal data. Such
measures will include omitting patient names or other directl y identifiable data in an y
reports, publications, or other disclosures, except where required b y applicable laws.
To protect the rights and freedoms of natural individuals with regard to the processing of
personal data, when study data are compiled for transfer to Pfizer and other authorized
parties, an y patient names will be removed and will be replaced by a single, specific,
numerical code. All other identifiable data transferred to Pfizer or other authorized parties
will be identified by this single, patient -specific code. In case of data transfer, Pfizer will
maintain high standar ds of confidentialit y and protection of individuals’ personal data
consistent with the vendor contract, and applicable privacy laws.
No personal data is planned to be transferred off the VA servers. Specifically, the Clinical
Epidemiology Program (CEP) at White River Junction VA Medical Center will conduct this
safety surveillance stud y with sponsorship from Pfizer and assistance from Analy sis Group,
Inc. The project will be led by the VA, with Dr. Yinong Young -Xu, Director of CEP, serving
as the Principal Investigator. Data access will be granted through VA Informatics and
Computing I nfrastructure (VINCI). VHA data will not be provided to Pfizer or Anal ysis
Group.Rather, onl y VA employees, including those with research service without
compensation (WOC) e mployee status, who have completed necessary VA training and have
proper clearance will access and anal yze data on secure VA servers and behind necessary
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073264
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 56of 144firewalls, under the direction and supervision of Dr. Young -Xu. Given the sensitive nature of
healthcare data, comprehensive securit y measures will be implemented to ensure the
confidentiality , integrity, and protection of Veterans’ privacy and healthcare data.
10.2.Patient Consent
As this study does not involve data subject to privacy laws according to applica ble legal
requirements, obtaining informed consent from individuals by Pfizer is not required.
10.3. Institutional Review board (IRB)/Independent Ethics Committee (IEC)
There must be prospective approval of the study protocol, protocol amendments, and their
relevant documents from the relevant IRBs/IECs. All correspondence with the I RB/IEC must
be retained. Copies of IRB/IEC approvals must be forwarded to Pfizer. The study protocol
will be reviewed b y the IRB of the VA Medical Center, White River Junction, VT.
10.4.Ethical Conduct of the Study
The study will be conducted in accordance with legal and regulatory requirements, as well as
with scientific purpose, value and rigor and follow generall y accepted research practices
described in Guidelines for Good Pharmacoepi demiology Practices (GPP) issued by the
International Societ y for Pharmacoepidemiology ,53the FDA Guidance for Industry and FDA
Staff: Best Practices for Conducting and Reporting, Pharmacoepidemiologic Safet y Studies
Using Electronic Healthcare Data52and Good Epidemiological Practice (GEP) guidelines
issued by the International Epidemiological Association (IEA).54
11.MANAGEMENT AND REPORTING OF ADVERSE EVENTS/A DVERSE
REACTIONS
Signal Detection and Signal Evaluation
This study involves data that exist as structured data by the time of study start or 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 language processing. In these data sources,
individual patient data are not retrieved or validated, and it is not possible to link (i.e.,
identifya potential association between) a particular product and medical event for any
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.
Signal Verification
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, imag es of physician notes, neurological
scans, X-rays, or narrative fields in a database. The reviewer is obligated to report adverse
events (AEs) with explicit attribution to any Pfizer drug that appear in the reviewed
information (defined per the patient po pulation and study period specified in the protocol).
Explicit attribution is not inferred b y a temporal relationship between drug administration
and an AE, but must be based on a definite statement of causality by a healthcare provider
linking drug admin istration to the AE.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073265
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 57of 144The requirements for reporting safet y events of interest on the non -interventional study
(NIS) adverse event monitoring (AEM) Report Form to Pfizer Safety 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) and reported, within 24 hours of awareness, to Pfizer Safet y
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
efficacy, and occupational exposure associated with the use of a Pfizer product must
be reported, within 24 hours of awareness, to Pfizer Safet y using the NI S AEM
Report Form.
For these AEs with an explicit attribution or scenarios involving exposure to a Pfizer
product, the safety information identified in the unstructured data reviewed is captured in the
Event Narrative section of the report form, and constitutes all 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, since not all elements will be available due to privacy concerns with
the use of secondary data sources. While not all demographic fields will be completed, 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 a valid one in accordance with pharmacovigilance 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/year
(DD/MMM/YYYY ) format.
All research staff members must complete the following Pfizer training requirements:
Your Reporting Responsibilities ( YRR)Training for Vendors Working on Pfizer
Studies
These trainings must be completed by research staff members prior to the start of data
collection. All trainings include a “Confirmation of Training Certificate” (for signature b y
the trainee) as a record o f completion of the training, which must be kept in a retrievable
format. Copies of all signed training certificates must be provided to Pfizer.
Re-training must be completed on an annual basis using the most current Your Reporting
Responsibilities training materials.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073266
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 58of 14412.PLANS FOR DISSEMINAT ING AND COMMUNICATING STUDY RESULTS
This protocol will be posted on publicly available registers following its finalization. The
final study results will be made publicly available via the European Union Post Authorisation
Safety (EU PAS) Register and may be submitted for publication in a peer reviewed medical
journal.
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 inves tigator is aware of an y new
information which might influence the evaluation of the benefits and risks of a Pfizer
product, Pfizer should be informed immediately.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073267
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 59of 14413.REFERENCES
1. World Health Organization (WHO). WHO Director-General’s opening remarks at th e
media briefing on COVID -19 -11 March 2020.
https://www.who.int/dg/speeches/detail/who- director-general-s-opening-remarks-at-
the-media-briefing-on-covid-19---11-march-2020. Accessed November 11, 2020.
2.Johns Hopkins University. Coronavirus Resource Cen ter.
https://coronavirus.jhu.edu/. Accessed November 10, 2020.
3. U.S. Food and Drug Administration (FDA). Vaccines and Related Biological
Products Advisory Committee Meeting: FDA Briefing Document Pfizer -BioNTech
COVID-19 Vaccine. Accessed December 10, 2020.
4. U.S. Food & Drug Administration (FDA). Pfizer COVID-19 Vaccine EUA Letter of
Authorization reissued 12-23 -20. In:December 23, 2020.
5.U.S. Food and Drug Administration (FDA). FDA Takes Key Action in Fight Against
COVID-19 By Issuing Emergency Use Authorization for First COVID- 19 Vaccine.
https://www.fda.gov/news -events/press -announcements/fda -takes-key-action-fight-
against-covid-19- issuing-emergency -use-authorization -first-covid-19.
6.Pfizer Inc. Pfizer and BioNTech Achieve First Authorization in the World for a
Vaccine to Combat COVID -19. https://www.pfizer.com/news/press -release/press-
release-detail/pfizer -and-biontech-achieve-first-authorization -world. Accessed
January 4, 2021.
7.Pfizer Inc. Pfizer and BioNTech Receive Authorization in the European Union for
COVID-19 Vaccine. https://www.pfizer.com/news/press -release/press- release-
detail/pfizer -and-biontech-receive-authorization- european- union. Accessed December
28, 2020.
8.Yih WK, L ee GM, Lieu TA, et al. Surveillance for adverse events fol lowing receipt
of pandemic 2009 H1N1 vaccine in the Post- Licensure Rapid Immunization Safety
Monitoring (PRI SM) System, 2009 -2010. Am J Epidemiol. 2012;175(11):1120 -1128.
9.American Medical Association (AMA). Appendix Q: Severe Acute Respiratory
Syndrome Coronavirus 2 (SARS -CoV-2) (coronavirus disease [COVID -19])
Vaccines. https://www.ama -assn.org/s ystem/files/2020-11/covid-19- immunizations -
appendix-q-table.pdf. Accessed January 12, 2021.
10.Centers for Medicare & Medicaid Services. COVID -19 Vaccines and Monoclonal
Antibodies. 2020; https://www.cms.gov/medicare/medicare- part-b-drug-average-
sales-price/covid -19-vaccines- and-monoclonal -antibodies. Accessed January 14,
2021.
11. Center for Disease Control (CDC). People with Certain Medical Conditions.
Decembe r 29, 2020; https://www.cdc.gov/coronavirus/2019 -ncov/need -extra-
precautions/people -with-medical-conditions.html. Accessed January 4, 2021.
12.U.S. Department of Veterans Affairs. Veterans Health Administration. 2020;
https://www.va.gov/health/. Accessed November 10, 2020.
13. U.S. Department of Veterans Affairs. VA nursing homes, assisted living, and home
health care. 2020; https://www.va.gov/health -care/about -va-health-benefits/long -
term-care/. Accessed January 4, 2020.
14.U.S. Food and Drug Administrat ion (FDA). COVID -19 Vaccine Safet y Surveillance
Project: Active Monitoring Protocol. December 16, 2020.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073268
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 60of 14415.Center for Disease Control (CDC) COVID -19 Response Team. Preliminary Estimates
of the Prevalence of Selected Underl ying Health Conditions Among Patients with
Coronavirus Disease 2019 —United States, February 12–March 28, 2020. MMWR
Morbidity and Mortality Weekly Report 2020;69(13):382 -386.
16.Gupta S, Hay ek SS, Wang W, et al. Factors Associated With Death in Critically Ill
Patients With Coronavirus Disease 2019 in the US. JAMA Intern Med.
2020;180(11):1 -12.
17.Williamson EJ, Walker AJ, Bhaskaran K, et al. Factors associated with COVID -19-
related death using OpenSAFELY. Nature. 2020;584(7821):430-436.
18.Walsh EE, Frenck RW, Jr., Falsey AR, et al. Safety and Immunogenicity of Two
RNA-Based Covid-19 Vaccine Candidates. N Engl J Med. 2020.
19.U.S. National Library of Medicine. Study to Describe the Safet y, Tolerability,
Immunogenicit y, and Efficacy of RNA Vaccine Candidates Against COVID -19 in
Healthy Individuals. NCT04368728.
20. U.S. Food and Drug Administration (FDA) Center for Biologics Evaluation and
Research (CBER). Emergency Use Authorization for Vaccines to Prevent CO VID-
19: Guidance for Industry . https://www.fda.gov/media/142749/download Accessed
November 10, 2020.
21. U.S. Food and Drug Administration (FDA) Center for Biologics Evaluation and
Research (CBER). Development and L icensure of Vaccines to Prevent COVID -19;
Guidance for Industry, June 2020. https://www.fda.gov/media/142749/download.
Accessed November 10, 2020.
22.Lee GM, Greene SK, Weintraub ES, et al. H1N1 and seasonal influenza vaccine
safety in the vaccine safety datalink project. Am J Prev Med. 2011;41(2):121 -128.
23.Fireman B, Lee J, L ewis N, Bembom O, van der Laan M, Baxter R. I nfluenza
vaccination and mortality: differentiating vaccine effects from bias. Am J Epidemiol.
2009;170(5) :650-656.
24.Virtanen M, Peltola H, Paunio M. Day -to-day reactogenicity and the healthy vaccinee
effect of measles -mumps-rubella vaccination. Pediatrics. 2000;106(5):E62.
25.Greene SK, Kulldorff M, Yin R, et al. Near real -time vaccine safety surveillance with
partially accrued data. Pharmacoepidemiol Drug Saf. 2011;20(6):583 -590.
26.Keren R, Zaoutis TE, Bridges CB, et al. Neurological and Neuromuscular Disease as
a Risk Factor for Respiratory Failure in Children Hospitalized With I nfluenza
Infection. JAM A. 2005;294(17):2188-2194.
27.U.S. Department of Veterans Affairs. VA priority groups. 2020;
https://www.va.gov/health- care/eligibility /priority-groups/.
28.Law B, Sturkenboom M. D2.3 Priority List of Adverse Events of Special Interest:
COVID-19. 2020.
29.Shimabukuro T. CDC post -authorization/post -licensure safety monitoring of COVID -
19 vaccines. National Center for Immunization & Respiratory Diseases. 2020.
30.Liu CH, Yeh YC, Huang WT, Chie WC, Chan KA. Assessment of pre -specified
adverse events follow ing varicella vaccine: A population- based self -controlled risk
interval study . Vaccine. 2020;38(11):2495-2502.
31.Baker MA, Baer B, Kulldorff M, et al. Kawasaki disease and 13 -valent pneumococcal
conjugate vaccination among young children: A self -controlled risk interval and
cohort study with null results. PLoS Med. 2019;16(7):e1002844.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073269
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 61of 14432.Gee J, Naleway A, Shui I, et al. Monitoring the safety of quadrivalent human
papillomavirus vaccine: findings from the Vaccine Safety Datalink. Vaccine.
2011;29(46):827 9-8284.
33.U.S. Department of Veterans Affairs. COVID -19 National Summary . 2020;
https://www.accesstocare.va.gov/Healthcare/COVID19NationalSummary . Accessed
November 10, 2020.
34.U.S. Department of Veterans Affairs. Racial and Ethnic Minority Veterans. 2020;
https://www.va.gov/HEALTHEQUITY/Race_Ethnicity .asp. Accessed January 18,
2021.
35.U.S. Department of Veterans Affairs. COVID -19 vaccines at VA. 2020;
https://www.va.gov/health- care/covid -19-vaccine/.
36.U.S. Department of Veterans Affairs. Profile o f Veterans: 2014. 2016;
https://www.va.gov/vetdata/docs/SpecialReports/Profile_of_Veterans_2014.pdf.
Accessed November 10, 2020.
37.U.S. Department of Veterans Affairs. Number of Veteran Patients b y Healthcare
Priority Group: FY2000 to FY2017. 2018;
https://www.va.gov/vetdata/docs/Utilization/Number_of_Veteran_Patients_by _HC_P
riority_Groups_2000_2017.pdf. Accessed November 10, 2020.
38.U.S. Department of Veterans Affairs. Average Expenditures Per Patient by
Healthcare Priorit y Group: FY2000 to FY2013. 201 3;
https://www.va.gov/vetdata/docs/Utilization/AvgCost_FINAL2.xlsx. Accessed
November 10, 2020.
39.Agha Z, Lofgren RP, VanRuiswy k JV, Layde PM. Are patients at Veterans Affairs
medical centers sicker? A comparative anal ysis of health status and medical re source
use. Arch Intern Med. 2000;160(21):3252-3257.
40.Li R, Stewart B, McNeil MM, et al. Post licensure surveillance of influenza vaccines
in the Vaccine Safet y Datalink in the 2013 -2014 and 2014 -2015 seasons.
Pharmacoepidemiol Drug Saf. 2016;25(8):928 -934.
41.Kulldorff M, Davis RL, Kolczak† M, Lewis E, L ieu T, Platt R. A Maximized
Sequential Probability Ratio Test for Drug and Vaccine Safet y Surveillance.
Sequential Analysis. 2011;30(1):58 -78.
42.Kulldorff M, Silva I R. Continuous Post -Market Sequentia l Safety Surveillance with
Minimum Events to Signal. arXiv preprint arXiv. 2015(1503.01978).
43.When Are Risks Real? I n. Clinical Nutrition Insight. Vol 332007:6.
44.Silva IR, Lopes WM, Dias P, Yih WK. Alpha spending for historical versus
surveillance Poisson data with CMaxSPRT. Stat Med. 2019;38(12):2126 -2138.
45.Silva IR. Type I error probability spending for post -market drug and vaccine safet y
surveillance with bin omial data. Stat Med. 2018;37(1):107-118.
46. Kulldorff M. A spatial scan statistic. Communications in Statistics - Theory and
Methods. 2007;26(6):1481 -1496.
47. U.S. Food and Drug Administration (FDA) Center for Biologics Evaluation and
Research (CBER). G uidance for Industry: Good Pharmacovigilance Practices and
Pharmacoepidemiologic Assessment. 2005. Accessed November 12, 2020.
48.Baker MA, Lieu TA, L i L, et al. A vaccine study design selection framework for the
postlicensure rapid immunization safety monitoring program. Am J Epidemiol.
2015;181(8):608 -618.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073270
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 62of 14449. Hoffman KB, Demakas AR, Dimbil M, Tatonetti NP, Erdman CB. Stimulated
reporting: the impact of US food and drug administration -issued alerts on the adverse
event reporting s ystem (FAERS). Drug Saf. 2014;37(11):971 -980.
50.West AN, Charlton ME, Vaughan -Sarrazin M. Dual use of VA and non -VA hospitals
by Veterans with multiple hospitalizations. BMC Health Serv Res. 2015;15:431.
51.Petersen LA, By rne MM, Daw CN, Hasche J, Reis B, Pietz K. Relationship between
clinical conditions and use of Veterans Affairs health care among Medicare -enrolled
veterans. Health Serv Res. 2010;45(3):762 -791.
52.U.S. Food and Drug Administration (FDA) Center for Drug Evaluation and Research
(CDER). Best Practices for Condu cting and Reporting Pharmacoepidemiologic
Safety Studies Using Electronic Healthcare Data. 2013;
https://www.fda.gov/media/79922/download . Accessed January 19, 2021.
53.International Societ y for Pharmacoepidemiology (ISPE). Guidelines for good
pharmacoepi demiology practices (GPP). Pharmacoepidemiol Drug Saf.
2008;17(2):200 -208.
54. International Epidemiological Association (IEA). Good Epidemiological Practice
(GEP), IEA Guidelines for Proper Conduct of Epidemiological Research. 2007;
https://ieaweb.org/IEA Web/Content/I EA_Publications.aspx . Accessed January 25,
2021.
55.Centers for Medicare & Medicaid Services. 2021 ICD -10-CM Official Guidelines for
Coding and Reporting. https://www.cms.gov/files/document/2021 -coding-
guidelines -updated-12162020.pdf . Accessed January 17, 2021.
56.Baxter R, Eaton A, Hansen J, Aukes L , Caspard H, Ambrose CS. Safet y of
quadrivalent live attenuated influenza vaccine in subjects aged 2 -49years. Vaccine.
2017;35(9):1254 -1258.
57.Johns Hopkins Vasculitis Center. Ty pes of Va sculitis.
https://www.hopkinsvasculitis.org/ty pes-
vasculitis/#:~:text=%E2%80%9CAngiitis%E2%80%9D%20and%20%E2%80%9CAr
teritis%E2%80%9D,lit'%20i%20deez%E2%80%9D . Accessed January 19, 2021.
58.OptumInsight Inc. Guide to Clinical Validation, Documentation and Coding: Acute
Kidney Injury.
https://www.optum360coding.com/upload/pdf/ECDCG14/CDCG14_v2.pdf .
Accessed January 17, 2021.
59. U.S. Department of Health and Human Services. Common Terminology Criteria for
Adverse Events (CTCAE). 2017;
https://ctep.cancer.g ov/protocoldevelopment/electronic_applications/docs/CTCAE_v
5_Quick_Reference_5x7.pdf . Accessed January 17, 2021.
60.Forns J, Cainzos -Achirica M, Hellfritzsch M, et al. Validity of ICD-9 and ICD -10
codes used to identify acute liver injury : A study in three European data sources.
Pharmacoepidemiol Drug Saf. 2019;28(7):965 -975.
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073271
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 63of 14414.LIST OF TABLES
Table 1. Outcome algorithms for SCRI analysis, with risk and control intervals ................... 36
Table2. Estimated Statistical Power for the Poisson -based MaxSPRT41............................... 41
Table 3. Critical Values for Poisson -based MaxSPRT................................ ............................ 49
Appendix Table 1. Demographic and Clinical Characteristics Definitions ..Error! Bookmark
not defined.
Appendix Table 2. Operational Definitions of Safety Events of Interest.....Error! Bookmark
not defined.
15.LIST OF FIGURES
Figure 1. Example of SCRI Design for Assessment of a Safet y Event of Interest with a
42-day Risk Interval in an I ndividual who Receives Only One
Vaccine Dose, Showing Both Pre- and Post-vaccination Control
Intervals................................ ................................ ................................ ....26
Figure 2. Example of SCRI Design with Overlapping Risk Intervals when Two Doses
of Pfizer-BioNTech COVID- 19 Vaccine are Administered,
Showing a Pre-and Post-vaccination Control Interval ............................. 27
Figure 3. Steps in Signal Detection, Evaluation, and Verification ................................ ..........45
Figure 4. Example of SCRI Design for a Safet y Event of Interest with a 42 -day Risk
Interval and a Pre -vaccination Control I nterval................................ ........47
Figure 5. Example of Risk (P1, P2, P3) and Pre- vaccination Control I ntervals for the
SCRI End-of-surveillance Anal yses of 1 or 2 D oses of Pfizer -
BioNTech COVID -19 Vaccine ................................ ................................ .53
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073272
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 64of 14416.ANNEX 1. LIST OF STAND ALONE DOCUMENTS
None.
17.ANNEX 2. ENCEPP CHEC KLIST FOR STUDY PROT OCOLS
N/A
18.ANNEX 3. ADDITIONAL INFORMATI ON
Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Demographic Characteristics
Age Continuous variable;
Dichotomous variable:
18-64
>65;
Categorical variable:
<35
35 -<45
45 -<55
55 -<65
65 -<75
≥75Age as of the date prior toPfizer-
BioNTech COVID -19 vaccination
(and/or date prior to seasonal
influenza vaccination for active
comparators)
Sex Categorical variable:
Male
Female
Unknown
Race/ethnicit yCategorical variable:
White
Asian or Pacific
Islander
Black
American Indian or
Alaskan native
Other
Unknown
State Geographic regions in the US State of residence
Clinical Characteristics
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073273
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 65of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Smoking Dichotomous variable Defined b ythe “tobacco” variable.
‘Y’ indicates the person is a tobacco
user
ICD-9-CM codes:
305.1, Tobacco use disorder
V15.82, History of tobacco
use
ICD-10-CM codes:
F17.200, Nicotine
dependence, unspecified,
uncomplicated
Z7.20, Tobacco use
Z87.891, Personal history of
nicotine dependence
Body mass index
(BMI)Continuous variable;
Categorical variable:
Underweight (<18.5)
Normal weight (18.5 -
24.9)
Overweight (25 -29.9)
Obese (≥30 -<40)
Severe obesit y (>40)Calculated from height and weight
data (kg/m2)
ICD-9-CM codes:
V85.0, Bod y Mass Index less
than 19, adult
V85.1, Bod y Mass Index
between 19- 24, adult
V85.2, Bod y mass index
between 25- 29, adult
V85.3, Bod y mass index
between 30- 39, adult
V85.4, Bod y mass index 40
and over, adult
ICD-10-CM codes:
Z68.1,Body Mass Index
19.9 or less, adult
Z68.2, Body mass index 20-
29, adult
Z68.3, Bod y mass index
between 30- 39, adult
Z68.4, Bod y mass index 40
and over, adult
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073274
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 66of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
History of
anaphylaxis/allergic
reactionsDichotomous variable ICD-9-CM code:
V13.81, Personal history of
anaphylaxis
V14.0 -V14.6, V14.8,
V14.9, Personal history of
allergy to drugs, medications
and biological substances,
excluding serum and vaccine
V15.0x, Other allergy
525.66, Allergy to existing
dental restorative material
995.0, Other anaph ylactic
shock, not elsewhere
classified
995.1, Angioneurotic edema,
not elsewhere classified
995.21, Arthus phenomenon
999.27, Other drug allergy
995.3, Allergy , unspecified,
not elsewhere classified
995.6x, Anaphy lactic shock
due to food
999.41, Anaph ylactic
reaction due to
administration of blood and
blood products
999.49, Anaph ylactic
reaction due to other serum
ICD-10-CM code:
Z87.892 Personal history of
anaphylaxis
Z88.0 -Z88.6, Z88.8, Z88.9,
Allergy status to drugs,
medications and biological
substances, excluding serum
and vaccine
T78.00xx -T78.09xx,
Anaphylactic reaction due to
food, initial encounter,
subsequent encounter and
sequela
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073275
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 67of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
T78.2xxx, Anaphy lactic
shock, initial encounter,
subsequent encounter and
sequela
T78.3xxx, Angioneurotic
edema, initial encounter,
subsequent encounter and
sequela
T78.41xx, Arthus
phenomenon
T80.51xx, Anaphy lactic
reaction due to
administration of blood and
blood products, initial
encounter, subsequent
encounter and sequela
T80.59xx, Anaphy lactic
reaction due to other s erum,
initial encounter, subsequent
encounter and sequela
T88.6xxx, Anaphy lactic
reaction due to adverse effect
of correct drug or
medicament properl y
administered, initial
encounter, subsequent
encounter and sequela
Previous anaph ylaxis
of vaccine
componentDichotomous variable ICD-9-CM code:
999.42, Anaph ylactic
reaction due to vaccination
V14.7, Personal history of
allergy to serum or vaccine
ICD-10-CM codes:
T80.52xx, Anaphy lactic
reaction due to vaccination,
initial encounter, subsequent
encounter and sequela
Z28.04, Immunization not
carried out because of patient
allergy to vaccine or
component
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073276
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 68of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Z88.7, Allergy status to
serum and vaccine
History of
hospitalizations Dichotomous variable;
Continuous variableDefined b y having any
hospitalizations (dichotomous) and
number of hospitalizations
(continuous)
Charlson
Comorbidity Index
(CCI)Continuous variable ICD-9-CM codes:
410.x, 412.x, My ocardial
infarction
398.91, 402.01, 402.11,
402.91, 404.01, 404.03,
404.11, 404.13, 404.91,
404.93, 425.4 -425.9, 428.x,
Congestive heart failure
093.0, 437.3, 440.x, 441.x,
443.1 -443.9, 447.1, 557.1,
557.9, V43.4, Peripheral
vascular disease
362.34, 430.x -438.x,
Cerebrovascular disease
290.x, 294.1, 331.2,
Dementia
416.8, 416.9, 490.x -505.x,
506.4, 508.1, 508.8, Chronic
pulmonary disease
446.5, 710.0 -710.4, 714.0 -
714.2, 714.8, 725.x,
Rheumatic disease
531.x -534.x, Peptic ulcer
disease
070.22, 070.23, 070.32,
070.33, 070.44, 070.54,
070.6, 070.9, 570.x, 571.x,
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073277
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 69of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
573.3, 573.4, 573.8, 573.9,
V42.7, Mild liver disease
250.0 -250.3, 250.8, 250.9,
Diabetes without chronic
complication
250.4 -250.7, Diabetes with
chronic complication
334.1, 342.x, 343.x, 344.0 -
344.6,344.9, Hemiplegia or
paraplegia
403.01, 403.11, 403.91,
404.02, 404.03, 404.12,
404.13, 404.92, 404.93,
582.x, 583.0 -583.7, 585.x,
586.x, 588.0, V42.0, V45.1,
V56.x, Renal disease
140.x -172.x, 174.x -195.8,
200.x -208.x, 238.6, Any
malignancy , including
lymphoma and leukemia,
except malignant neoplasm
of skin
456.0 -456.2, 572.2 -572.8,
Moderate or severe liver
disease
196.x -199.x, Metastatic
solid tumor
042.x -044.x, Acquired
immunodeficiency syndrome
(AIDS)/Human
immunodeficiency virus
(HIV)
ICD-10-CM codes:
I21.x, I21.xx, I 22.x, I25.2,
Myocardial infarction
I09.9, I11.0, I13.0, I13.2,
I25.5, I42.0,
I42.5 -I42.9, I43, I43.x,
I50.x, I50.xx, Congestive
heart failure
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073278
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 70of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
I70.x, I71.x, I 73.1, I73.8,
I73.9, I77.1, I79.0, I79.2,
K55.1, K55.8, K55.9, Z95.8,
Z95.9, Peripheral vascular
disease
G45, G45.x, G46.x, H34.0,
I60.x -I63.x, I60.xx -I63.xx,
I60.xxx - I63.xxx, I65.x -
I69.x, I65.xx - I69.xx,
I65.xxx - I69.xxx,
Cerebrovascular disease
F00.x -F03.x, F00.xx -
F03.xx, F05, F05.1, G30.x,
G31.1, Deme ntia
I27.8, I27.9, J40.x - J47.x,
J40.xx -J47.xx, J40.xxx -
J47.xxx, J60.x -J67.x, J68.4,
J70.1, J70.3, Chronic
pulmonary disease
M05, M05.x, M05.xx,
M05.xxx, M06, M06.x,
M06.xx, M06.xxx, M31.5,
M32.x -M34.x, M32.xx -
M34.xx, M35.1, M35.3,
M36.0, Rheum atic disease
K25.x -K28.x, Peptic ulcer
disease
B18.x, K70.0 - K70.3,
K70.9, K71.3 - K71.5,
K71.7, K73.x, K74.x,
K74.xx, K76.0, K76.2 -
K76.4, K76.8, K76.9, Z94.4,
Mild liver disease
E10.0, E10.1x, E10.6x,
E10.6xx, E10.8, E10.9,
E11.0x, E11.1x, E11.6x,
E11.6xx, E11.8, E11.9,
E12.0, E12.1, E12.6, E12.8,
E12.9, E13.0x, E13.1x,
E13.6x, E13.6xx, E13.8,
E13.9, E14.0, E14.1, E14.6,
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073279
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 71of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
E14.8, E14.9, Diabetes
without chronic complication
E10.2x -E10.5x, E10.2xx -
E10.5xx, E10.7, E11.2x -
E11.5x, E11.2xx - E11.5xx,
E11.7, E12.2 - E12.5, E12.7,
E13.2 -E13.5x, E13.7, E14.2
-E14.5, E14.7, Diabetes
with chronic complication
G04.1, G11.4, G80.1, G80.2,
G81.x, G81.xx, G82.x,
G82.xx, G83.0, G83.1 -
G83.3, G83.1x -G83.3x,
G83.4, G83.9, Hemiplegia or
paraplegia
I12.0, I13.1x, N03.2 - N03.7,
N05.2 -N05.7, N18.x, N19,
N25.0, Z49.0x -Z49.3x,
Z94.0, Z99.2, Renal disease
C00-C75, C00.x -C75.x,
C00.xx-C75.xx (excluding
C44, C44.x and C44.xx),
C7A., C7A.x, C7A.xx, C7B.,
C7B.x, C7B.xx, C76- C80,
C76.x-C80.x, C76.xx -
C80.xx, C81- C96, C81.x-
C96.x, C81.xx -C96.xx, Any
malignancy , including
lymphoma and leukemia,
except malignant neoplasm
of skin
I85.0, I85.9, I86.4, I98.2,
K70.4x, K71.1x, K72.1x,
K72.9x, K76.5, K76.6,
K76.7, Moderate or severe
liver disease
C77.x -C80.x, C77.xx -
C80.xx, Metastatic solid
tumor
B20, B97.35, AIDS/HIV
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073280
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 72of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Comorbidities Categorical variable:
Autoimmune
disease
Asthma
Bleeding diathesis
or condition
associated with
prolonged bleeding
Cancer
Cardiovascular
conditions ( e.g.,
heart failure, CAD,
cardiomyopathies)
Chronic kidney
disease/dial ysis
COPD/interstitial
lung disease
Diabetes mellitus
(ie, Type 2 diabetes)
Down syndrome
Sickle cell disease
HBV
HCV
HIV
Hyperlipidemia
Hypertension
Liver disease
Neurological
disease
Other immune
deficiencies
Solid organ
transplant
VTEAutoimmune disease
(immunocompromised state
[weakened immune s ystem] from
solid organ transplant):
ICD-9-CM codes:
245.2, Chronic
lymphocytic thyroiditis
340, Multiple sclerosis
357, Acute infective
polyneuritis
357.4, Poly neuropathy in
other diseases classified
elsewhere
696.1, Other psoriasis
694.3, Impetigo
herpetiformis
696.1, Other psoriasis
696, Psoriatic
arthropath y
695.4, Lupus
erythematosus
714, 714.x, 714.xx,
Rheumatoid arthritis and
other inflammatory
polyarthropathies
359.6, Symptomatic
inflammatory myopathy
in diseases classified
elsewhere
357.1, Poly neuropathy in
collagen vascular disease
714.89, Other specified
inflammatory
polyarthropathies
714.9, Unspecified
inflammatory
polyarthropathy
446.5, Giant cell arteritis
710.2, Sicca syndrome
ICD-10-CM codes:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073281
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 73of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
D69.3, Immune
thrombocytopenic
purpura
E06.3, Autoimmune
thyroiditis
G35, MS
G61.0 and G65.0, GBS
and sequelae of GBS
L40.x, L40.5x, Psoriasis
L93.x, Lupus
erythematosus
M05.x, M05.xx,
M05.xxx, Rheumatoid
arthritis with rheu matoid
factor
M06.x, M06.xx,
M06.xxx, Other
rheumatoid arthritis
M31.5, M31.6, Giant cell
arteritis
M35.0x, Sicca
(Sjogren’s) s yndrome
E10, E10.x, E10.xx,
Type 1 diabetes mellitus
N05.9,
Glomerulonephritis
D84.9,
Immunodeficiency ,
unspecified
Asthma:
ICD-9-CM codes:
o493.xx, Asthma
ICD-10-CM codes:
oJ45.2x -J45.3x,
Mild intermittent
asthma
oJ45.4x, Moderate
persistent asthma
oJ45.5x, Severe
persistent asthma
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073282
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 74of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oJ45.9x, Other and
unspecified
asthma
Bleeding diathesis or condition
associated with prolonged bleeding:
ICD-9-CM codes:
o286.x,
Coagulation
defects
o289.8x, Other
specified diseases
of blood and
blood-forming
organs
o287, 287.x,
287.xx, Purpura
and other
hemorrhagic
conditions
ICD-10-CM codes:
oD65,
Disseminated
intravascular
coagulation
oD66, Heredita ry
factor VIII
deficiency
oD67, Hereditary
factor IX
deficiency
oD68, D68.x,
D68.xx, Other
coagulation
defects
oD69, D69.x,
D69.xx, Purpura
and other
hemorrhagic
conditions
Cancer:
ICD-9-CM codes:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073283
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 75of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o140.x -149.x,
Malignant
neoplasm of lip,
oral cavit y, and
pharynx
o150.x -159.x,
Malignant
neoplasm of
digestive organs
and peritoneum
o160.x -165.x,
Malignant
neoplasm of
respiratory and
intrathoracic
organs
o170.x -176.x,
Malignant
neoplasm of
bone, connective
tissue, skin, and
breast
o179.x -189.x,
Malignant
neoplasm of
genitourinary
organs
o190.x -199.x,
Malignant
neoplasm of other
unspecified sites
o200.xx -208.xx,
Malignant
neoplasm of
lymphatic and
hematopoietic
tissue
o209.0x -209.3x,
Malignant
neuroendocrine
tumors
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073284
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 76of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o230.x -234.x,
Carcinoma i n situ
of digestive
organs
ICD-10-CM codes:
oC00-C75, C00.x -
C75.x, C00.xx -
C75.xx, C7A.,
C7A.x, C7A.xx,
C7B., C7B.x,
C7B.xx,
Malignant
neoplasms, stated
or presumed to be
primary (of
specified sites),
and certain
specified
histologies,
except
neuroendocrine,
and of lymphoid,
hematopoietic
and related tissue
oC76-C80, C76.x -
C80.x, C76.xx -
C80.xx,
Malignant
neoplasms of ill-
defined, other
secondary and
unspecified sites
oC81-C96, C81.x -
C96.x, C81.xx -
C96.xx,
Malignant
neoplasms of
lymphoid,
hematopoietic
and related tissue
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073285
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 77of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Cardiovascular conditions ( e.g.,
heart failure, coronary artery disease
[CAD], cardiomy opathies):
ICD-9-CM codes:
o428.xx, Heart
failure
o414.01, 429.2,
411.1, 413.9,
414.11, 414.12,
414.05, 414.02,
414.04, 414.03,
414.06, 414.07,
414.2, 411.81,
411.89, CAD
o425.xx,
Cardiomy opathy
ICD-10-CM codes:
o150.x, 150.xx,
Heart failure
oI24.0, I24.8,
I24.9, I25.10,
I25.110, I25.111,
I25.118, I25.119,
I25.41, I25.42,
I25.700, I25.701,
I25.708, I25.709,
I25.710, I25.711,
I25.718, I25.719,
I25.720, I25.721,
I25.728, I25.729,
I25.730, I25.731,
I25.738, I25.739,
I25.750, I25.751,
I25.758, I25.759,
I25.760, I25.761,
I25.768, I25.769,
I25.790, I25.791,
I25.798, I25.799,
I25.810, I25.811,
I25.812, CAD
oI42.x,
Cardiomy opathy
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073286
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 78of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
Chronic kidney disease/dialy sis:
ICD-9-CMcodes:
o283.11,
Hemolytic-
uremic syndrome
o403, 403.x,
403.xx,
Hypertensive
chronic kidney
disease
o404, 404.x,
404.xx,
Hypertensive
heart and chronic
kidney disease
o440.1,
Atherosclerosis of
renal artery
o442.1, Aneury sm
of renal artery
o572.4,
Hepatorenal
syndrome
o274.1, Gouty
nephropath y,
unspecified
o710, Systemic
lupus
erythematosus
o710.2, Sicca
syndrome
o580, 580.x,
580.xx, Acute
glomerulonephriti
s
o581.x, 581.xx,
Nephrotic
syndrome
o582, 582.x,
582.xx, Chronic
glomerulonephriti
s
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073287
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 79of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o583, 583.x,
583,xx, Nephritis
and nephropath y,
not specified as
acute or chronic
o591,
Hydronephrosis
o593.3, Stricture or
kinking of ureter
o592, Calculus of
kidney
o592.1, Calculus
of ureter
o590.9, Infection
of kidney ,
unspecified
o584.x, Acute
kidney failure
o585.x, Chronic
kidney disease
o588.x, 588.xx,
Disorders
resulting from
impaired renal
function
o587, Renal
sclerosis,
unspecified
o753.1x, Cy stic
kidney disease
o753.2, 753.2x,
Obstructive
defects of renal
pelvis and ureter
ICD-10-CM codes:
oD59.3,
Hemolytic-
uremic syndrome
oI12.x,
Hypertensive
chronic kidney
disease
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073288
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 80of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oI13.x, I13.xx,
Hypertensive
heart and chronic
kidney disease
oI70.1,
Atherosclerosis of
renal artery
oI72.2 Aneury sm
of renal artery
oK76.7,
Hepatorenal
syndrome
oM10.30-M10.39,
M10.30x-
M10.37x, Gout
due to renal
impairment
oM32.14,
Glomerular
disease in
systemic lupus
erythematosus
oM32.15, Tubulo -
interstitial
nephropath y in
systemic lupus
erythematosus
oM3504, Sicca
syndrome with
tubulo-interstitial
nephropath y
oN00.x-N07.x,
N08, Glomerular
diseases
oN13.1, N13.2,
N13.3x,
Obstructive a nd
reflux uropathy
oN14.x,
Nephropath y
oN15.x, Other
renal tubulo -
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073289
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 81of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
interstitial
diseases
oN16, Renal
tubulo-interstitial
disorders in
diseases classified
elsewhere
oN17.x, N18.x,
N19, Acute
kidney failure and
chronic kidney
disease
oN25.x, N26.x,
N25.xx, Other
disorders of
kidney and ureter
oQ61.02, Q61.11x,
Q61.2-Q61.9,
Cystic kidney
disease
oQ62.x, Q62.xx,
Congenital
obstructive
defects of renal
pelvis and
congenital
malformation of
ureter
COPD/interstitial lung disease:
ICD-9-CM codes:
o491.9,
Unspecified
chronicbronchitis
o492.8, Other
emphysema
o491.x, 491.xx,
Chronic
bronchitis
o493.2, Chronic
obstructive
asthma,
unspecified
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073290
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 82of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o496, Chronic
airway
obstruction, not
elsewhere
classified
o516, 516.x,
516.xx, Other
alveolar and
parietoalveolar
pneumonopathy
o515,
Postinflammatory
pulmonary
fibrosis
o518.x, 518.xx,
Other diseases of
lung
o714.81,
Rheumatoid lung
ICD-10-CM codes:
oJ41.x Simple and
mucopurulent
chronic bronchitis
oJ42, Unspecified
chronic bronchitis
oJ43.x,
Emphysema
oJ44.x, Other
COPD
oJ80, J81.x,
J82.xx, J84.xx,
J84.xxx, Other
respiratory
diseases
principally
affecting the
interstitium
oM05.10,
Rheumatoid lung
disease with
rheumatoid
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073291
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 83of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
arthritis of
unspecified site
Diabetes mellitus (ie, Type 2
diabetes):
ICD-9-CM codes:
o250.xx, Diabetes
mellitus
ICD-10-CM codes:
oE11.x, E11.xx,
E11.xxx, Ty pe 2
diabetes mellitus
Down syndrome:
ICD-9-CM codes:
o758.x, Down
syndrome
ICD-10-CM codes:
oQ90.x, Down
syndrome
Sickle cell disease:
ICD-9-CM codes:
o282.xx, Sickle -
cell disease
ICD-10-CM codes:
oD57, D57.x,
D57.xx, D57.xxx,
Sickle-cell
disorders
HBV:
ICD-9-CM codes:
o70.33, Chronic
viral hepatitis B
without mention
of hepatic coma
with hepatitis
delta
o70.32, Chronic
viral hepatitis B
without mention
of hepatic coma
without mention
of hepatitis delta
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073292
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 84of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o70.3, Viral
hepatitis B
without mention
of hepatic coma,
acute or
unspecified,
without mention
of hepatitis delta
o70.2, Viral
hepatitis B with
hepatic coma,
acute or
unspecified,
without mention
of hepatitis delta
ICD-10-CM codes:
oB18.0, B18.1,
Chronic viral
hepatitis B
oB19.1, B1 9.1x,
Unspecified viral
hepatitis B
HCV:
ICD-9-CM codes:
o70.7, Unspecified
viral hepatitis C
without hepatic
coma
o70.71,
Unspecified viral
hepatitis C with
hepatic coma
o70.54, Chronic
hepatitis C
without mention
of hepatic coma
ICD-10-CM codes:
oB18.2, Chronic
viral hepatitis C
oB19.2x,
Unspecified viral
hepatitis C
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073293
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 85of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
HIV:
ICD-9-CM codes:
o42, HIV disease
o79.53, HIV t ype 2
ICD-10-CM codes:
oB20, HIV disease
oB97.35, HIV t ype
2 as the cause of
diseases classified
elsewhere
Hyperlipidemia
ICD-9-CM codes:
o272.0x, Pure
hypercholesterole
mia
o272.1x, Pure
hyperglyceridemi
a
o272.2x, Mixed
hyperlipidemia
o272.4x,
Hyperlipidemia,
NOS
ICD-10-CM codes:
oE78.0-E78.5,
E78.0x, E78.4x,
Hyperlipidemia
Hypertension:
ICD-9-CM codes:
o401.1, Benign
essential
hypertension
o401.9, Essential
hypertension,
NOS
o405.1, Benign
secondary
hypertension
o405.9, Secondary
hypertension,
NOS
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073294
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 86of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o997.91,
Hypertension,
NOS
ICD-10-CM codes:
oH35.03x,
Hypertensive
retinopathy
oI10, I11.x -I16.x,
I13.xx,
Hypertensive
diseases
oI67.4,
Hypertensive
encephalopa thy
diseases
Liver disease:
ICD-9-CM codes:
o571, 571.x,
Alcoholic fatt y
liver
o572, 572.x,
Hepatic
encephalopath y
o573.x, Other
disorder of liver
o570, Acute and
subacute necrosis
of liver
ICD-10-CM codes:
oK70.x, K70.xx,
Alcoholic fatt y
liver
oK71.x, K71.xx,
Toxic liver
disease
oK72.xx, Hepatic
failure, not
elsewhere
classified
oK73.x, Chronic
hepatitis, not
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073295
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 87of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
elsewhere
specified
oK74.x, K74.xx,
Fibrosis and
cirrhosis of liver
oK75.x, K75.xx,
Other
inflammatory
liver diseases
oK76.x, K76.xx,
Other diseases of
liver
oK77, Liver
disorders in
diseases classified
elsewhere
Neurological disease:
ICD-9-CM codes:
o780.97, Altered
mental status
o780.93, Memory
loss
o781.8, Neurologic
neglect syndrome
o797, Senility
without mention
of psychosis
oV62.89, Other
psychological or
physical stress,
not elsewhere
classified
o799.5x, Signs and
symptoms
involving
cognition
o780.99, Other
general symptoms
o780.4, Dizziness
and giddiness
o781.1,
Disturbances of
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073296
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 88of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
sensation of smell
and taste
oV41.5, Problems
with smell and
taste
o368.16,
Psychophysical
visual
disturbances
o307.9, Other and
unspecified
special symptoms
or syndromes, not
elsewhere
classified
o300.9,
Unspecified
nonpsychotic
mental disorder
o300.9,
Unspecified
nonpsychotic
mental disorder
o308.9,
Unspecified acute
reaction to stress
o307.9, Oth er and
unspecified
special symptoms
or syndromes, not
elsewhere
classified
oV62.85,
Homicidal
ideation
oV62.84, Suicidal
ideation
o799.24,
Emotional lability
o799.23,
Impulsiveness
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073297
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 89of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o799.29, Other
signs and
symptoms
involving
emotional state
oV40.39, Other
specified
behavioral
problem
ICD-10-CM codes:
oR41, R41.x,
R41.xx, Other
symptoms and
signs involving
cognitive
functions and
awareness
oR42, Dizziness
and giddiness
oR43, R43.x,
Disturbances of
smell and taste
oR44, R44.x,
Other symptoms
and signs
involving general
sensations and
perceptions
oR45, R45.x, R45.
xx, Symptoms
and signs
involving
emotional state
oR46, R46.x, R46.
xx, Symptoms
and signs
involving
appearance and
behavior
Other immune deficiencies:
ICD-9-CM codes:
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073298
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 90of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o279.x, 279.xx,
Deficiency of
humoral
immunity
o135, Sarcoidosis
o273.x, Disorders
of plasma protein
metabolism
ICD-10-CM codes:
oD80, D80.x,
Immunodeficienc
y with
predominantly
antibody defects
oD81, D81.x,
D81.xx,
Combined
immunodeficienci
es
oD82, D82.x,
Immunodeficienc
y associated with
other major
defects
oD83, D83.x,
Common variable
immunodeficienc
y
oD84, D84.x,
D84.xx, Other
immunodeficienci
es
oD86, D86.x,
D86.xx,
Sarcoidosis
oD89, D89.x,
D89.xx, Other
disorders
involving the
immune
mechanism, not
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073299
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 91of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
elsewhere
classified
Solid organ transplant:
CPT codes:
o32850-32856,
Transplantation
of lung
o33930-33945,
Transplantation
of heart
o44132, 44133,
47133, 47135,
47140-47147,
Transplantation
of liver
o44135-44137,
44715, 44720,
44721,
Transplantation
of intestine
o48160, 48550 -
48552, 48554,
48556,
Transplantation
ofpancreas
o50300, 50320,
50323, 50325,
50327, 50328,
50329, 50340,
50340, 50360,
50365, 50370,
50380, Renal
transplantation
ICD-9-PCS codes:
o00.91 -00.93,
Transplant from
donor or cadaver
o37.51, Heart
transplantation
o33.51, Unilateral
lung
transplantation
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073300
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 92of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o33.52, Bilateral
lung
transplantation
o46.97, Transplant
of intestine
o50.59, Other
transplant of
intestine
o52.82,
Homotransplant
of pancreas
o55.69, Other
kidney transplant
ICD-10-PCS codes:
o02YA0Z0,
02YA0Z1,
Transplantation
of heart
o0BYC0Z0,
0BYC0Z1,
0BYD0Z0,
0BYD0Z1,
0BYF0Z0,
0BYF0Z1,
0BYG0Z0,
0BYG0Z1,
0BYH0Z0,
0BYH0Z1,
0BYJ0Z0,
0BYJ0Z1,
0BYK0Z0,
0BYK0Z1,
0BYL0Z0,
0BYL0Z1,
0BYM0Z0,
0BYM0Z1,
Transplantation
of lung
o0DY60Z0,
0DY60Z1,
Transplantation
of stomach
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073301
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 93of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o0DY80Z0,
0DY80Z1,
Transplantation
of small i ntestine
o0DYE0Z0,
0DYE0Z1,
Transplantation
of large intestine
o0FY00Z0,
0FY00Z1,
Transplantation
of liver
o0FYG0Z0,
0FYG0Z1,
Transplantation
of pancreas
o0TY00Z0,
0TY00Z1,
0TY10Z0,
0TY10Z1,
Transplantation
of kidney
VTE:
ICD-9-CM codes:
o415.1x,
Pulmonary
embolism and
infarction
o451.x, 451.xx,
Phlebitis and
thrombophlebitis
o452, Portal vein
thrombosis
o453.x, 453.xx,
Other venous
embolism and
thrombosis
ICD-10-CM codes:
oI26, I26.x, I26.xx,
Pulmonary
embolism
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073302
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 94of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oI80, I80.x, I80.xx,
I80.xxx, Phlebitis
and
thrombophlebitis
oI81, Portal vein
thrombosis
oI82, I82.x, I82.xx,
I82.xxx Other
venous embolism
and thrombosis
Concurrent
immunizationsCategorical variable:
Seasonal influenza
Tetanus diphtheria and
pertussis (Tdap or Td)
Chickenpox (Varicella)
Shingles (Herpes Zoster
recombinant and/or
live)
Human papillomavirus
(HPV)
Pneumococcal
conjugate
Pneumococcal
polysaccharide
Hepatitis A
Hepatitis B
Meningococcal
conjugate (MenACWY)
and serogroup B
meningococcal (MenB)
Haemophilus influenza
type bDescription of immunization,
immunization I D, lot number, and
manufacturer code will be available.
Seasonal influenza:
CPT codes:
o90653, Influenza
vaccine, inactivated
(IIV), subunit,
adjuvanted, for
intramuscular use
o90724, Influenza virus
vaccine
o90662, Influenza virus
vaccine (IIV), split virus,
preservative free,
enhanced
immunogenicit y via
increased antigen
content, for
intramuscular use
o90662, Influenza virus
vaccine (IIV), split virus,
preservative free,
enhanced
immunogenicit y via
increased antigen
content, f or
intramuscular use
o90694, Influenza virus
vaccine, quadrivalent
(aIIV4), inactivated,
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073303
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 95of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
adjuvanted, preservative
free, 0.5 mL dosage, for
intramuscular use
o90756, Influenza virus
vaccine, quadrivalent
(ccIIV4), derived from
cell cultures, subunit,
antibiotic free, 0.5 mL
dosage, for intramuscular
use
o90674, Influenza virus
vaccine, quadrivalent
(ccIIV4), derived from
cell cultures, subunit,
preservative and
antibiotic free, 0.5 mL
dosage, for intramuscular
use
o90688, Influenza virus
vaccine, quadrivalent
(IIV4), split virus, 0.5
mL dosage, for
intramuscular use
o90686, Influenza virus
vaccine, quadrivalent
(IIV4), split virus,
preservative free, 0.5 mL
dosage, for intramuscular
use
o90630, Influenza virus
vaccine, quadrivalent
(IIV4), split virus,
preservative free, for
intradermal use
o90682, Influenza virus
vaccine, quadrivalent
(RIV4), derived from
recombinant DNA,
hemagglutinin (HA)
protein only , preservative
and antibiotic free, for
intramuscular use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073304
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 96of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o90672, Influenza virus
vaccine, quadrivalent,
live (LAIV4), for
intranasal use
o90661, Influenza virus
vaccine, trivalent
(ccIIV3), derived from
cell cultures, subunit,
preservative and
antibiotic free, 0.5 mL
dosage, for intramuscular
use
o90658, Influenza virus
vaccine, trivalent (IIV3),
split virus, 0.5 mL
dosage, for intramuscular
use
o90656, Influenza virus
vaccine, trivalent (IIV3),
split virus, preservative
free, 0.5 mL dosage, for
intramuscular use
o90654, Influenza virus
vaccine, trivalent (IIV3),
split virus, preservative -
free, for intradermal use
o90673, Influenza virus
vaccine, trivalent (RIV3),
derived from
recombinant DNA,
hemagglutinin (HA)
protein only , preservative
and antibiotic free, for
intramuscular use
o90660, Influenza virus
vaccine, trivalent, live
(LAIV3), for intranasal
use
o90659, Influenza vir us
vaccine, whole virus, for
intramuscular or jet
injection use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073305
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 97of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
HCPCs codes:
oG0008,
Administration of
influenza virus
vaccine
oG8482, Influenza
immunization
administered or
previously received
oQ2034, Influenza
virus vaccine, split
virus, for
intramuscular use
(Agriflu)
oQ2035, Influenza
virus vaccine, split
virus, when
administered to
individuals 3 y ears of
age and older, for
intramuscular use
(Afluria)
oQ2036, Influenza
virus vaccine, split
virus, when
administered to
individuals 3 y ears of
age and older, for
intramuscular use
(Flulaval)
oQ2037, Influenza
virus vaccine, split
virus, when
administered to
individuals 3 y ears of
age and older, for
intramuscular use
(Fluvirin)
oQ2038, Influenza
virus vaccine, split
virus, when
administered to
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073306
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 98of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
individuals 3 y earsof
age and older, for
intramuscular use
(Fluzone)
oQ2039, Influenza
virus vaccine, not
otherwise specified
Tetanus diphtheria and pertussis
(Tdap or Td):
CPT codes:
o90714, Tetanus and
diphtheria toxoids
adsorbed (Td), preservative
free, when administered to
individuals 7 y ears or
older, for intramuscular use
o90715, Tdap administered
to individuals 7 y ears or
older, for intramuscular use
o90718, Tetanus and
diphtheria toxoids (Td)
adsorbed when
administered to
individuals 7 y ears or
older, for intramuscular
use
Chickenpox (Varicella)
CPT codes:
o90396, Varicella -zoster
immune globulin, human,
for intramuscular use
o90716, Varicella virus
vaccine, live, for
subcutaneous use
Shingles (Herpes Zoster
recombinant and/or live)
CPT codes:
o90396, Varicella -zoster
immune globulin, human,
for intramuscular use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073307
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 99of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o90736, Zoster (shingles)
vaccine (HZV), live, for
subcutaneous injection
o90750, Zoster (shingles)
vaccine (HZV),
recombinant, subunit,
adjuvanted, for
intramuscular use
Human papillomavirus (HPV)
CPT codes:
o90649,Human
Papillomavirus vaccine,
types 6, 11, 16, 18,
quadrivalent (4vHPV), 3
dose schedule, for
intramuscular use
o90650, Human
Papillomavirus vaccine,
types 16, 18, bivalent
(2vHPV), 3 dose
schedule, for
intramuscular use
o90651, Human
Papillomavirus vaccine
types 6, 11, 16, 18, 31, 33,
45, 52, 58, nonavalent
(9vHPV), 2 or 3 dose
schedule, for
intramuscular use
Pneumococcal conjugate
CPT codes:
o90669, Pneumococcal
conjugate vaccine, 7
valent, for intramuscular
use
o90670, Pneumococcal
conjugate vaccine, 13
valent (PCV13), for
intramuscular use
HCPCS codes (used
pneumococcal conjugate and
polysaccharide):
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073308
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 100of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
oG0009, Administration of
pneumococcal vaccine
oG8864, Code for
Pneumococcal vaccine
administered or
previously received
Pneumococcal pol ysaccharide:
CPT code:
o90732, Pneumococcal
polysaccharide vaccine,
23-valent (PPSV23), adult
or immunosuppressed
patient dosage, when
administered to
individuals 2 y ears or
older, for subcutaneous or
intramuscular use
Hepatitis A
CPT codes
o90632, Hepatitis A
vaccine, adult do sage, for
intramuscular use
o90633, Hepatitis A vaccine
(HepA),
pediatric/adolescent
dosage-2 dose schedule,
for intramuscular use
o90634, Hepatitis A vaccine
(HepA),
pediatric/adolescent
dosage-3 dose schedule,
for intramuscular use
o90730, Hepatitis A vacci ne
o90636, Hepatitis A and
hepatitis B vaccine (HepA -
HepB), adult dosage, for
intramuscular use
Hepatitis B
CPT codes:
o907311, Hepatitis B
vaccine
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073309
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 101of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
o90739, Hepatitis B vaccine
(HepB), adult dosage, 2
dose schedule, for
intramuscular use
o90740, Hepatitis B vaccine
(HepB), dial ysis or
immunosuppressed patient
dosage, 3 dose schedule,
for intramuscular use
o90743, Hepatitis B vaccine
(HepB), adolescent, 2 dose
schedule, for intramuscular
use
o90744, Hepatitis B vaccine
(HepB),
pediatric/adolescent
dosage, 3 dose schedule,
for intramuscular use
o90745, Hepatitis B
vaccine, adolescent/high
risk infant dosage, for
intramuscular use
o90746, Hepatitis B
vaccine (HepB), adult
dosage, 3 dose schedule,
for intramuscular use
o90747, Hepatitis B
vaccine (HepB), dial ysis
or immunosuppressed
patient dosage, 4 dose
schedule, for
intramuscular use
HCPCS codes:
oG0010, Administration of
Hepatitis B vaccine
Meningococcal conjugate
(MenACWY) and serogroup B
meningococcal (MenB)
CPT codes:
o90619, Meningococcal
conjugate vaccine,
serogroups A, C, W, Y,
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073310
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 102of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
quadrivalent, tetanus
toxoid carrier
(MenACWY -TT), for
intramuscular use
o90620, Meningococcal
recombinant protein and
outer membrane vesicle
vaccine, serogroup B
(MenB-4C), 2 dose
schedule, for
intramuscular use
o90621, Meningococcal
recombinant lipoprotein
vaccine, serogroup B
(MenB-FHbp), 2 or 3
dose schedule, for
intramuscular use
o90733, Meningococcal
polysaccharide vaccine,
serogroups A, C, Y, W -
135, quadrivalent
(MPSV4), for
subcutaneous use 90734,
Meningococcal conjugate
vaccine, serog roups A, C,
W, Y, quadrivalent,
diphtheria toxoid carrier
(MenACWY -D) or
CRM197 carrier
(MenACWY -CRM), for
intramuscular use
Haemophilus influenza type b
CPT codes:
o90645, Hemophilus
influenza b vaccine (Hib),
HbOC conjugate (4 dose
schedule), for
intramuscular use
o90646, Hemophilus
influenza b vaccine (Hib),
PRP-D conjugate, for
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073311
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 103of 144Appendix Table 1. Demographic and Clinical Characteristics Definitions
Variable Description Operational definition
booster use onl y,
intramuscular use
o90647, Haemophilus
influenzae ty pe b vaccine
(Hib), PRP -OMP
conjugate, 3 dose
schedule, for
intramuscular use
o90648, Haemophilus
influenzae ty pe b vaccine
(Hib), PRP -T conjugate, 4
dose schedule, for
intramuscular use
o90737, Hemophilus
influenza B
o90748, Hepatitis B and
Haemophilus influenzae
type b vaccine (Hib-
HepB), for intramuscular
use
090177e19624f4aa\Approved\Approved On: 29-Jan-2021 20:01 (GMT)
FDA-CBER-2021-5683-1073312
BNT162b2 (COVID -19 vaccine)
C4591012 NON- INTERVENTIONAL STUDY PROTOCOL
Final Version 1.0, 27 January 2021
PFIZER CONFIDENTIAL
Page 104of 144Appendix Table 2. Operational Definitions of Safety Events of Interest
Variable Operational Definition
Defined b y the presence of an y of the following
ICD-9-CMcodes (inclusive)1:Defined b y the presence of an y of the
following ICD -10-CMcodes
(inclusive)1:
Neurologic
Generalized convulsions/seizures8,22345, Epilepsy and recurrent seizures
780.3, Convulsions
780.31, Febrile convulsions (simple),
unspecified
780.39, Other convulsionsG40.A01, Absence epileptic
syndrome, not intractable, with
status epilepticus
G40.A09, Absence epileptic
syndrome, not intractable, without
status epilepticus
G40.A11, Absence epileptic
syndrome, intractable, with status
epilepticus
G40.A19, Absence epileptic
syndrome, intractable, without
statusepilepticus
G40.101, L ocalization -related
(focal) (partial) sy mptomatic
epilepsy and epileptic s yndromes
with simple partial seizures, not
intractable, with status epilepticus
G40.109, L ocalization -related
(focal) (partial) sy mptomatic
epilepsy and epilep tic syndromes
0
…[truncated]