Document text
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 1of 9NON- INTERVENTIONAL STUDY INTERIM REPORT 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
Date: 22June 2021
Name sand affiliation sof the main authors : Yinong Young Xu, ScD, MA, MS , Veterans
Affairs Medical Center , White River Junction, VT ; Cynthia de Luise, PhD, MPH, Pfizer,
Inc., New York, NY ; Mei Sheng Duh, ScD, MPH, Analy sis Group, Inc. , Boston, MA
Keywords :Pfizer -BioNTech COVID -19 vaccine, emergency use a uthorization (EUA) , rapid
cycle anal ysis, safet y signal, Veterans Health Administration (VHA)
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).1On December 11, 2020, the United States (US) Food and
Drug Administration (FDA) issued itsfirst emergency use authorization (EUA) for a vaccine
for the prevention of COVID-19 disease in individuals 16 y ears of age and older.2This
vaccine (Pfizer -BioNTech COVID- 19 Vaccine) is one of the three COVID- 19 vaccines that
received EUA in the US.
On January 27, 2021, Pfizer , in collaboration with the US Veterans Health Administration
(VHA) and Anal ysis Group, submitted a study protocol (protocol C4591012; Appendix 1 )
for post -EUA a ctive safety surveillance of safet y events of interest among individuals 16
years of age and older in the VHA s ystem. Safet y events of interest for the current stud y
include 42safet y outcomes that were selected based primaril yon the Priority List of Adve rse
Events of Special Interest from the Brighton Collaboration’s Safet y Platform for Emergency
vACcines (SPEAC) Project and from thepreliminary list of safet y events of interest
presented at the September 22, 2020, meeting of Centers for Disease Control a nd
Prevention’s (CDC ’s) Advisory Committee on Immunization Practices (ACI P) on the
enhanced safety monitoring of COVID -19 vaccines .3,4
The study uses data from a large-scale electronic medical record (EMR) database from the
VHA to identify and evaluate rapid -cycle, near real -time potential safet y signals associated
with Pfizer -BioNTech COVID -19 vaccine. Data included for anal ysis commences on
December 11, 2020 and will continue through June 10, 2023 –i.e., 30 months following the
issuance of the EUA for Pfizer -BioNTech COVID-19 vaccine. Four interim reports are
planned on June 30, 2021; December 31, 2021; June 30, 2022; and December 31, 2022, with
a final report to be issued on December 31, 2023.
In this first interim report, the data included for anal ysisspans from December 11, 2020 to
March 12, 2021, and the data were locked on April 26, 2021 . This first interim report
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075887
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 2of 9describes the selection of two samples of individuals within the VHA: individuals who
received Pfizer -BioNTech COVID- 19 vaccine betwe en December 11 , 2020 ( EUA approval
date b y the US FDA) and March 12, 2021 (data cutoff date), and recipients of seasonal
influenza vaccine(s) during the five prior influenza seasons (2014/2015 through 2018/2019) .
Thedemographic and clinical characteristic s and vaccine utilization patterns among the two
samples of individuals are described pursuant to the secondary study objective. Due to the
short time between the data lock date and the first interim report issuance date , this report
does not include results from the safet y signal analy ses pursuant to the primary study
objectives.
All references to the study protocol in this first interim report pertain to the study protocol
Version 1.0 dated January 27, 2021 (see Appendix 1 ). Meanwhile, a protocol amendment is
currentl y under development to address the FDA’s comments on the study protocol Version
1.0,communicated electronically to Pfizer on May 12, 2021.
Research question and objectives :
Research question: what are the incidence rates of safet y events of interest (based on adverse
events of special interest [AESI ]) among individuals vaccinated with 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 s ystem experience increased risk of safet y events
of interest following receipt of Pfizer -BioNTech COVID -19 vaccine; and
To assess whether sub- cohorts of interest (i.e., immunocompromised, elderly ,with specific
comorbidities, individuals receiving onl y one dose of 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 Pfizer -BioNTech COVID -19 vaccine.
Secondary study objective :
To characterize utilization patterns of Pfizer -BioNTech COVID -19 vaccine among
individuals within the VHA including estimating the proportion of individuals receiving
vaccine , two-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.
This first interim report pertains to the secondary study objective.
Study design:
This post -EUA active safety surveillance program will employ a rapid -cycle, longitudinal,
observational cohort study design to provide earl y real-world safet y information .
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075888
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 3of 9The self -controlled risk interval (SCRI) design will be used to sequentially monitor
the occurrence of safet y events of interest while controlling for time -invariant
confounders. The SCRI design uses data from cases (i.e., individuals who experience
safet y events of interest following vaccination) to compare the risk interval following
Pfizer-BioNTech COVID-19 vaccination t o pre -or post -vaccination non -risk
intervals (“pre -vaccination control interval” and “post- vaccination control interval” ,
respectivel y) in the same individual.
An active comparator design will be used to sequentially monitor the occurrence of
safet y events of interest among individuals that receive Pfizer -BioNTech COVID-19
vaccinations as compared to recipients of influenza vaccine in the VHA during
2014/2015 through 2018/2019 influe nza 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.
Setting:
The study population consists of VHA enrollees, w hich largel y include veterans, and iskept
asbroad as possible under inclusion and exclusion criteria to be representative of the real -
world population at the VHA receiving Pfizer -BioNTech COVID -19 vaccine.
Subject sand study s ize:
A total of 750,999 individuals received at least one dose of Pfizer -BioNTech COVID -19
vaccine from December 11, 2020 to March 12, 2021 in the VHA database . The active
comparator group includes a fixed cohort of 4,277,220 historical seasonal influenza vacci ne
recipients who received a total of 10,529,071 seasonal influenza vaccines across successive
influenza seasons between 2014/2015 through 2018/2019 .
Data source, variables, and statistical methods:
Data source: The VHA is the largest integrated health car e system in the US, providing
comprehensive healthcare services, including primary , specialt y and inpatient care,
rehabilitation, long- term and home ca re, and other services to over nine million veterans
through 1, 293health care facilities, includ ing 171 medical centers and 1,112 outpatient sites
of vary ing care complexity .5In addition to veterans, VHA enrollees may also include
employ ees and certain categories of family members of the veterans, but these are a minorit y
among all VHA enrollees. The study relies on secondary data from the Corporate Data
Warehouse (CDW) in the National Veterans Affairs Health Care Network. The CDW data,
which are updated dail y, include standard EMR for all medical encounter information in the
VHA s ystem. The CDW consolidates data from the VHA’s EMR s ystem and contains
information on all outpatient visits, hospital stay s, treatments, dispensed prescriptions,
immunizations, and lab results. Information is also available on date of death, COVID-19
infection status, and COVID-19 and seasonal influenza vaccination status.
Variables:
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075889
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 4of 9Outcomes: Outcomes include 42 safet y events of interest (Table 1 4.2in Section 14 of the
interim report ). The list of outcome safety events may be modified over time as new safet y
information about the COVID -19 vaccines emerges. No outcomes were analy zed as part of
this first interim report.
Exposures: Administration of Pfizer -BioNTech COVID- 19 vaccine post -EUA approval and
date of immunization for each dose were identified in the CDW data based on t he following
codes :
Current Procedural Terminology (CPT) code 91300 (Severe acute respiratory
syndrome coronavirus 2 (SARS -CoV -2) (coronavirus disease [COVID-19]) vaccine,
mRNAL NP, spike protein, preservative free, 30 mcg/0.3mL dosage, diluent
reconstitute d, for intramuscular use) and associated vaccine administration Healthcare
Common procedure Coding S ystem (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);6,7or
10 and 11- digit National Drug Codes (NDCs); 10 -digit NDCs: 59267 -1000-1, 59267 -
1000 -2, 59267 -1000 -3), 11 -digit NDCs: 59267 -1000 -01, 59267 -1000 -02, 59267 -
1000 -03;8or
Immunization records that contain data on vaccine code descriptor, vaccine
manufacturer (i.e., Pfizer), lot number, injection site, and date(s) of immunization .6
Administration of the seasonal influenza vaccine during 2014/2015 through 2018/2019
seasonal influenza seasons and date of immunization were identified in the CDW data based
on CPT, HCPCS , and 10 and 11- digit NDCs codes (Table 14.3 in Section 14 of the interim
report )as well as immunization records that contain data on vaccine code descriptor, vaccine
manufacturer, lot number, injection site, and date(s) of immunization.
Key covariates: Base line demographic (i.e., age, sex, race/ethnicit y, and VHA service area )
and clinical characteristics (i.e., smoking status, body mass index [BMI ], history of
anaph ylaxis/allergic re actions, previous anaphy laxis of vaccine component, history of
hospitalizat ions, Charlson Comorbidity Index [CCI ], select edcomorbidities, and
immunization history ) were assessed in the one- year baseline period before the date of
vaccination with Pfizer- BioNTech COVID -19 vaccine for Pfizer -BioNTech COVID-19
vaccine recipients and prior to the date of seasonal influenza vaccination for active
comparators (with the exception of BM I which was assessed based on a two -year baseline
period to account for a reduction in the visits with BMI records available during the COVID -
19 pandemi c).
Statistical methods: Descriptive statistics (i.e., mean s, standard deviations (SD) ,medians,
inter-quartile ranges [ IQR], and proportions) were used to summarize baseline characteristics
and patterns of vaccination with Pfizer -BioNTech COVID -19 vaccin e (e.g., 2- dose
completion rate, clinical setting where the vaccine was administered) among Pfizer -
BioNTech COVID -19 vaccine recipients and baseline characteristics among historical active
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075890
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 5of 9comparators who received seasonal influenza vaccination. Standardiz eddifferences were
used to compare b aseline demographics and clinical characteristics between individuals
vaccinated with Pfizer -BioNTech COVID- 19 vaccine (at the time of the first Pfizer -
BioNTech COVID -19 vaccine dose) and those vaccinated for seasonal i nfluenza (at the time
of the most recent seasonal influenza vaccine). Standardized difference sbelow 10% were
used to identify patient characteristics that were not well balanced between recipients of
Pfizer -BioNTech COVID -19 vaccine and seasonal influenza vaccine.9,10The count and
proportion of individuals who received COVID -19 vaccine (s) from different manufacturer (s)
in addition to Pfizer -BioNTech COVID-19 vaccine were also reported.
Statistical methods for the primary study objectives, which are not covered in this first
interim report , include safety signal anal yses for signal detection, signal evaluation, and
signal verification, as described in Section 9.7.3 of the study protocol (Appendix 1 ).
Results:
Among 4,648,524 individuals with at least one healthcare encounter from December 11,
2020 to March 1 2, 2021, 17.1% (N = 793,264) received at least one Pfizer -BioNTech
COVID -19 vaccine dose . Among the latter, 752,904 (94.9%) satisfied the one y ear of
continuous e nrollment in VHA healthcare benefits eligibility criteria , of whom most (N =
750,999; 99.7%) only received COVID- 19 vaccine(s) from Pfizer-BioNTech and a minority
(N =1,905; 0.3%) received COVID- 19 vaccine(s) from both Pfizer -BioNTech and another
manufact urer (96.8% Moderna; 2.9% Johnson & Johnson; and 0.5% AstraZeneca). These
750,999 individuals form Pfizer -BioNTech vaccine cohort.
The seasonal influenza vaccine sample consisted of a total of 4,277,220 eligible individuals
who received at least one seaso nal influenza vaccine from the October 2014 -May 2015
influenza season to October 2018 - May 2019 influenza season. Individuals could receive one
seasonal influenza vaccine per influenza season, for a maximum of five seasonal influenza
vaccine sfrom October 2014 to May 2019. Among the individuals in the seasonal influenza
sample that satisfied the study eligibility criteria, an average of 2.46 seasonal influenza
vaccines were observed per individual from October 2014 to May 2019 .
The cohort of Pfizer -BioN Tech COVID -19 vaccine recipients were on average 69.8 years of
age (median: 72.0 years) and included 93.2% males and 64.2% White non -Hispanics, with
the largest proportion of the vaccinees residing in the South (42.8%). While b aseline
demographic and clinical characteristics were generall y similar between Pfizer -BioNTech
COVID -19 vaccine recipients and the seasonal influenza vaccine recipients (standardized
differences < 10%, e.g., smoking status, BMI , comorbidities ), one notable exception was the
higher proportion of elderly individuals among Pfizer -BioNTech COVID -19 vaccine
recipients compared to the seasonal influenza vaccine recipients (age 65 and 74 years: 41.2%
vs. 33.8%, standardized difference 15.3% ; age 75 years or older : 31.6% vs. 25.9% ,
standardized difference 12.5%).
Of the 750,999 individuals in Pfizer -BioNTech COVID -19 vaccine cohort, a total of 485,410
(64.6%) received both doses during the December 11 ,2020 to March 12, 2021 study period.
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075891
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 6of 9Among the remaining 265,589 ( 35.4%)with onl y one dose observed, 94.1% had insufficient
follow -up (<21 day s)to receive a second dose administered per EUA recommended
schedule . Among those who did have at least 21 days o f follow -up between the first Pfizer -
BioNTech COVID -19dose and the data cut -off date , the proportion of Pfizer- BioNTech
COVID -19 vaccine recipients who received both doses was high (95.3 %). Of the 485,410
individuals who completed two Pfizer -BioNTech COVID- 19 vaccine doses, 70.2% received
the second dose exac tly 21 day s after the first dose, as recommended per the product label.
The most common care setting where the first and second doses of Pfizer -BioNTech
COVID -19 vaccine were received was in outpatient clinics (85.3% and 84.8%, respectively ).
Discussion:
This interim report describes sample selection, baseline characteristics and vaccine utilization
patterns among individuals who received Pfizer -BioNTech COVID- 19 vaccine within the
VHA s ystemas compared to a historical sample of individuals who received s easonal
influenza vaccine before the COVID -19 pandemic. Most baseline characteristics were well
balanced between the study sample s, suggesting that seasonal influenza vaccine from the five
prior seasons is an appropriate active comparator for the safet y surveillance of Pfizer-
BioNTech COVID -19 vaccine.
Since CDC’s COVID -19 vaccine rollout recommendations gave priority to individuals 65
years and older in phases 1a through 1c, the higher proportion of older individuals with
Pfizer -BioNTech COVID-19 vaccine was expected.11This distribution of age will likely
change in subsequent interim reports as younger individuals become eligible to be vaccinated
in the VHA. The higher proportion of men in this study was also expected, as the VHA
population is predominantl y male (approximately 90%).
The demographic and clinical characteristics of the Pfizer -BioNTech COVID -19 vaccine
sample were consistent with existing data in the US population, which suggests that vaccine
administration within the VHA s ystem aligned with CDC’s vaccine rollout recommendations
to prioritize the elderly population and individuals with underly ing medical conditions.11
Names and affiliations of principal investigators:
Nam e, degree(s) Title Affiliation Address
Yinong Young -Xu,
ScD, MA, MSPrincipal Investigator Veterans Affairs
(VA) Medical
Center163 Veterans Drive,
White River Junction,
VT 05009
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075892
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 7of 9Cynthia de Luise,
PhD, MPHCo-Investigator Pfizer, Inc. 235 East 42nd Street,
New York, NY 10017
Mei Sheng Duh,
ScD, MPHCo-Investigator Analysis 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 Co-Investigator Analysis 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
Drive South
Los Angeles, CA
90095
Rachel Bhak, MS Co-Investigator Analysis Group,
Inc. 111 Huntington Ave
14thFloor
Boston, MA 02199
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075893
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 8of 9References:
1World Health Organization (WHO). WHO Director- General’s opening remarks at the
media briefing on COVID -19 -11 March 2020 [I nternet]. WHO; 2020 [cited 2020 Nov 11].
Available from: https://www.who.int/dg/speeches/detail/who- director -general -s-opening -
remarks -at-the-media -briefing -on-covid -19---11-march -2020
2U.S. Food and Drug Administration (FDA). Pfizer- BioNTech COVID -19 Vaccine
[Internet]. FDA; 2021 [cited 2021 Apr 21]. Availabl e from: https://www.fda.gov/emergency -
preparedness -and-response/coronavirus- disease -2019 -covid- 19/pfizer -biontech -covid-19-
vaccine
3Law B. SO2 -D2.1.2 Priority List of COVID -19 Adverse Events of special interest:
Quarterl y update [I nternet]. Brightoncollaboration.us; 2020 [cited 2021 Apr 23]. Available
from: https://brightoncollaboration.us/wp-
content/uploads/2021/01/SO2_D2.1.2_V1.2_COVID -19_AESI -update -23Dec2020 -
review_final.pdf
4Shimabukuro T. Enhanced safet y monitoring for COVID -19 vaccines in early phase
vaccination. National Center for Immunization & Respiratory Diseases; 2020. Available
from: https://www.cdc.gov/vaccines/acip/meetings/downloads/slides-2020- 09/COVI D-03 -
Shimabukuro.pdf
5U.S. Department of Veterans Affairs. Veterans Health Administra tion. Providing
Health Care for Veterans [I nternet]. U.S. Department of Veterans Affairs; 2020 [cited 2021
Apr 23]. Available from: https://www.va.gov/health/
6American Medical Association (AMA). Appendix Q: Severe Acute Respiratory
Syndrome Coronavirus 2 (SARS -CoV -2) (coronavirus disease [COVID- 19]) Vaccines
[Internet]. AMA; 2021 [cited 2021 Jan 12] .Available from: https://www.ama -
assn.org/sy stem/files/2020 -11/covid -19-immunizations -appendix -q-table.pdf
7Centers for Medicare & Medicaid Services (CMS) . COVID -19 Vaccines and
Monoclonal Antibodies [Internet]. CMS; 2020 [cited 2021 Jan14]. Available from:
https://www.cms.gov/medicare/medicare -part-b-drug-average -sales -price/covid -19-vaccines-
and-monoclonal -antibodies
8Centers for Disease Control and Prevention (CDC). NDC11 to NDC10 Crosswalk
Reference [Internet]. CDC; 2021 [cited April 30, 2021]. Available from:
https://www2.cdc.gov/vaccines/iis/iisstandards/downloads/NDC/NDC_Reference_1110 -
tag508.pdf
9Austin PC. Using the standardized difference to compare the prevalence of a binary
variable between two groups in observational research. Commun Stat-Simul Comput . 2009
Apr;38(6):1228 -34.
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075894
Non-Interventional Study Interim Report Abstract
Protocol C4591012
22June 2021
PFIZER CONFIDENTIAL
Page 9of 910Ahmed A, Perry GJ, Fleg JL , Love TE, Goff Jr DC, Kitzman DW. Outcomes in
ambulatory chronic s ystolic and diastolic heart failure: a propensity score analy sis. American
heart journal. 2006 Nov 1;152(5):956 -66.
11U.S. Department of Veterans Affairs. VA Caribbean Healthcare S ystem [I nternet] .
2021 [cited 2021 January 15]. Available from:
https://www.caribbean.va.gov/pressreleases/COVID -19_Moderna_Vaccine.asp
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
FDA-CBER-2021-5683-1075895
Document Approval Record
Document Name:
Document Title:
Signed By: Date(GMT) Signing Capacity
! " # $%&%
' ('
) *
+ , ! , $%&%
'-'$ ."/001 *
090177e1975fdd51\Approved\Approved On: 23-Jun-2021 09:48 (GMT)
Page 10
FDA-CBER-2021-5683-1075896