Submitting Study Datasets for Vaccines to the Office of Vaccines Research and Review: Guidance for Industry; Technical Specifications Document

Administrative guidance, not enacted legislation. This captured publisher edition preserves the listed document and attachments; later developments are not independently consolidated.

Submitting Study Datasets for
Vaccines to the Office of Vaccines
Research and Review
Guidance for Industry
Technical Specifications Document
This guidance is for immediate implementation.
FDA is issuing this guidance for immediate implementation in accordance with 21 CFR
10.115(g)(4)(i). Submit one set of either electronic or written comments on this guidance at any
time. Submit electronic comments to https://www.regulations.gov. Submit written comments to
the Dockets Management Staff (HFA-305), Food and Drug Administration, 5630 Fishers Lane,
Rm. 1061, Rockville, MD 20852. You should identify all comments with docket number FDA-
2018-D-1216.
Additional copies of this guidance are available from the Office of Communication, Outreach
and Development (OCOD), 10903 New Hampshire Ave., Bldg. 71, Rm. 3128, Silver Spring,
MD 20993-0002, or by calling 1-800-835-4709 or 240-402-8010, or email ocod@fda.hhs.gov, or
from the Internet at https://www.fda.gov/vaccines-blood-biologics/guidance-compliance-
regulatory-information-biologics/biologics-guidances.
For questions on the content of this guidance, contact OCOD at the phone numbers or email
address listed above.
U.S. Department of Health and Human Services
Food and Drug Administration
Center for Biologics Evaluation and Research
December 2019

Submitting Study Datasets for
Vaccines to the Office of Vaccines
Research and Review
Guidance for Industry
Technical Specifications Document
Revision History
Date 	Version 	Summary of Revisions
April 2018 	1.0 	Initial Version
October 2019 	2.0 	Revised Language in Section 3.5 and 4.0
Added Language in Section 2.0, 3.2, 3.3, 4.0 and 7.0
Added Section 6.0
December 2019 	2.1 	Correction to Revision History Summary, Changed Section
7.9 to 7.0

i
Contains Nonbinding Recommendations
Table of Contents
1.0 	INTRODUCTION .............................................................................................................. 1
2.0 	SUBMISSION OF DATA FOR VACCINE TRIALS ...................................................... 2
3.0 	SUBMISSION OF SAFETY DATA FOR VACCINE TRIALS ..................................... 2
3.1 	Reactogenicity Data ................................................................................................ 2
3.2 	Unsolicited Adverse Events (AEs) ......................................................................... 5
3.3 	Laboratory Safety Assessments ............................................................................. 5
3.4 	Medically Attended Adverse Events (MAAEs) .................................................... 5
3.5 	Deaths....................................................................................................................... 6
4.0 	SUBMISSION OF CLINICAL DISEASE ENDPOINT EFFICACY DATA FOR
VACCINE TRIALS ............................................................................................................ 6
5.0 	SUBMISSION OF IMMUNOGENICITY DATA FOR VACCINE TRIALS ............... 7
6.0 	SUBMISSION OF COMCOMITANT MEDICATION DATA FOR VACCINE
TRIALS………………………………………………………………………………… ... 7
7.0 	SUBMISSION OF DATA FOR MATERNAL IMMUNIZATION TRIALS ................ 7

Contains Nonbinding Recommendations
1
Submitting Study Datasets for Vaccines to the Office of Vaccines
Research and Review
Guidance for Industry
Technical Specifications Document
1.0 	INTRODUCTION
This technical specifications document provides you, sponsor/applicant detailed information and
specifications for the content of datasets submitted to FDA’s Center for Biologics Evaluation and
Research (CBER) Office of Vaccines Research and Review (OVRR) and is designed to aid
clinical and statistical reviewers in the review of vaccine applications, e.g., biologics license
applications. We, FDA, recommend submission of these datasets as part of the applicant’s data
tabulation datasets. The submission of standardized datasets facilitates review and analyses of
the data and allows for pooling of data, when appropriate. These specifications reflect current
CBER OVRR thinking, are built to be consistent with the FDA Study Data Technical
Conformance Guide1, and are generally consistent with the Therapeutic Area User Guide
(TAUG) for Vaccines2.
If there is a question regarding a specific submission or implementation of a particular data
standard, the sponsor should contact the appropriate review division in OVRR for specific
submission questions or the appropriate contact for data standards issues at cber-
edata@fda.hhs.gov. For more general recommendations on the use and submission of
standardized trial data, the sponsor should refer to the FDA Study Data Technical Conformance
Guide.
FDA’s guidance documents, including this guidance, do not establish legally enforceable
responsibilities. Instead, guidance describes the FDA’s current thinking on a topic and should be
viewed only as recommendations unless specific regulatory or statutory requirements are cited.
The use of the word should in FDA’s guidance means that something is suggested or
recommended but not required.
1 https://www.fda.gov/media/122913/download
2 https://www.cdisc.org/standards/therapeutic-areas/vaccines
This guidance represents the current thinking of the Food and Drug Administration (FDA or
Agency) on this topic. It does not establish any rights for any person and is not binding on FDA
or the public. You can use an alternative approach if it satisfies the requirements of the
applicable statutes and regulations. To discuss an alternative approach, contact the FDA staff
responsible for this guidance as listed on the title page.

Contains Nonbinding Recommendations
2
2.0 	SUBMISSION OF DATA FOR VACCINE TRIALS
The domains to be utilized in all Study Data Tabulation Model (SDTM) datasets submitted to
OVRR include: Trial Summary (TS), Demographics (DM), Subject Visits (SV), Concomitant
Medications (CM), Exposure (EX), Disposition (DS), Protocol Deviations (DV), and Medical
History (MH). If a physical exam is included in the protocol, the information should be included
in the Physical Exam (PE) domain. Similarly, if safety labs are included in the protocol, the
information should be included in the Laboratory Test Results (LB) domain. Usage of the
Adverse Event (AE), Clinical Event (CE), Findings About (FA), Vital Signs (VS), and
Microbiology Specimen (MB) domains will be detailed below.
The DS domain should always include the Standardized Disposition Name (DECOD), Reported
Term (TERM), Category (CAT), and Start Date/Time of Event (STDTC) variables. If relevant
(i.e., multiple doses administered over time), the DS domain should include the use of the Order
of Element within Arm (TAETORD) timing variable. The TAETORD timing variable should
also be included in other domains when the vaccine is multi-dose.
In addition, the domains and variables to be utilized for safety, immunogenicity, and efficacy
(i.e., clinical disease endpoint) data are detailed below.
3.0 	SUBMISSION OF SAFETY DATA FOR VACCINE TRIALS
Safety data submitted for vaccine clinical trials should include reactogenicity data (i.e., a set of
prespecified AEs collected within a prespecified time frame, often referred to as solicited AEs or
reactions), unsolicited AEs, medically attended adverse events (MAAEs), and death. The
sections below will detail the usage of domains and variables expected in submissions to OVRR.
Any usage of domains and variables that differ from these recommendations will need to be
agreed upon prior to submission of datasets.
3.1 	Reactogenicity Data
Reactogenicity data should be represented primarily in the CE domain with the Findings
About Clinical Event (FACE) domain and VS domain to provide the specific information
for each event. A more detailed description of how reactogenicity data should be
collected is described in the Vaccine TAUG2. We prefer that the Vaccine TAUG “flat
model” be utilized, i.e., that data for each day be included, even if a subject never
experienced a particular event. However, the “nested model,” which includes only a
summary record for a particular event if a subject never experienced that event, may be
necessary for large trials with significant amounts of data. Sponsors should discuss which
model is appropriate with their review team prior to beginning their clinical trial.
In the flat model, the CE dataset should include a record for each event for each subject,
with the Clinical Event Reported Term (CETERM) as the event name and the Clinical
Event Occurrence (CEOCCUR) = “Y” or “N.” CEOCCUR is “null” if the reactogenicity
event was not measured on a particular day (i.e., data are missing) and none of the other
assessment days for that subject are “Y.” If an event occurred, the clinical event start
day/date (CESTDY/CESTDTC) and end day/date (CEENDY/CEENDTC) of the
reactogenicity event should be collected and included in the dataset. If it is unknown

Contains Nonbinding Recommendations
3
when an event started and/or ended (e.g., because the diary card was not completed), the
variable(s) CESTDTC and/or CEENDTC should be null, and a comment should be
provided in the Comments (CO) domain. The pattern of the event (i.e., continuous or
intermittent) can be noted by using the Clinical Event Pattern of Events (CEPATT)
variable, if collected. In both cases, the daily records are represented in FACE. All daily
temperature measurements are represented in the VS domain. Standard units should
either be Celsius (o
C) or Fahrenheit (o F), but not both within a dataset. A dataset-level
relationship in Related Records (RELREC) should be used to represent relationships
between CE and FACE and between VS and assessments of fever in CE.
The annotated Case Report Form (CRF) should include a check box to indicate whether
the reactogenicity event did or did not occur, during the prespecified time frame. A check
box in the annotated CRF should also capture whether the reactogenicity event was
collected every day, during the assessment interval for a subject. If the Occurrence
(OCCUR) variable is null, the Completion Status (--STAT) and Reason Not Done (--
REASND) variables in FACE and VS and CE (covering the assessment interval) should
be utilized.
Reactogenicity events should be indicated by using “reactogenicity” in the variable
Clinical Event Category (CECAT). It should be further subcategorized into
“administration site” or “systemic,” using the variable Clinical Event Subcategory
(CESCAT).
If a reactogenicity event should happen to continue beyond the assessment interval, it
should also be represented in the AE domain, with the CE domain record indicating that
the event extended beyond the assessment interval, using the variables Clinical Event End
Reference Time Point (CEENTPT) (e.g., “Day 7”) and Clinical Event End Relative to
Reference Time Point (CEENRTPT) (“Ongoing”). A check box in the CRF, indicating
whether the reactogenicity event is ongoing after the assessment period, is recommended.
The event should be categorized in the Adverse Event Category (AECAT) variable as
“reactogenicity.” The start day/date (--STDY/--STDTC) and the end day/date (--ENDY/-
-ENDTC) of the reactogenicity event should be identical in both the CE and AE domain;
whereas the duration (--DUR) should report the time that the event occurred as part of the
assessment interval and as part of the continuance separately (e.g., an event that lasted 6
days in the assessment interval and 3 days beyond the assessment would be reported as
Clinical Event Duration (CEDUR) = 6 days and Adverse Event Duration (AEDUR) = 3
days). We recommend one or more check boxes in the CRF, indicating the duration of
the reactogenicity event in the assessment period and the duration of the reactogenicity
event beyond the assessment period. If a reactogenicity event rises to the level of a
Serious AE (SAE) during the prespecified assessment interval, it should be reported in the
AE domain, and the Serious Adverse Event (AESER) variable should be utilized, as well
as the CESER variable. The event should be categorized in the AECAT as
“reactogenicity.” The day and date that the reactogenicity event becomes serious should
be reported in supplemental AE (suppae) domain Qualifier Variable Names (QNAM) =
“SAEDY” and “SAEDTC,” Qualifier Variable Labels (QLABEL) = “SAE day” and
“SAE date,” and Supplemental Qualifier Values (QVAL) = day/date that reactogenicity
event became serious. A dataset-level relationship in RELREC should be used for each
case. We prefer that, in both cases, FACE (and/or VS) be utilized to provide the day-to-
day data for the reactogenicity event from start to end.

Contains Nonbinding Recommendations
4
If a reactogenicity event is assessed by the investigator as unrelated to the vaccine, but
instead related to another diagnosis (e.g., Viral Infection), the definitive diagnosis would
be recorded in the AE domain with the day-to-day signs and symptoms reported in the
Findings About Adverse Event (FAAE) domain. However, the reactogenicity event(s))
(i.e., symptoms), would still be reported in the CE (and FACE/VS) if obtained from the
diary because the event(s) are prespecified and occur during the solicited assessment
period. In this case CEREL (causality) can be reported as “N” as there is a plausible
reason for the event that is not due to vaccine administration. AEREL would also be
reported as “N”. The relationship between AE and CE in the subject should be
established using RELREC.
Please note it would be considered recall data, if you are reporting selected AE preferred
terms (PT) from the CRF that are synonymous with solicited local
administration/systemic reactogenicity events which “occurred” during the pre-specified
reporting time period post-vaccination but were actually collected from the investigator in
a follow-up visit. If you intend to report data based on recall (and not obtained from the
diary), please minimize potential bias by including a reconstructed data flag in
supplemental CE (suppce) using QNAM = “RECON,” (QLABEL) = “reconstructed data”
and (QVAL) “Y” in your dataset. Reconstruction should be limited to investigators who
are still blinded to treatment group in a randomized, controlled trial using a scripted query
for parameters (but perhaps not grading severity) if applied during a specified time period
post-vaccination. A “reconstructed data” flag might support a comparison of results
generated based on different types of data collection (e.g. diary card vs eCRF) as well as
overall compiled safety data; and may be used to determine if randomization was
preserved in “reconstructed” data to get a gross sense of whether the investigators were
truly blinded as they queried subjects and whether all treatment groups had the expected
representation in both the “purest” collection of the safety data (i.e., directly from the
diary) and the “reconstructed” as well as whether there were differences between sites.
Any plans to include reconstructed data in your submission should be discussed with the
review team prior to submission.
If an analgesic/antipyretic is used for prevention or treatment of pain and/or fever
associated with vaccine administration, please report this in the CM domain using the
appropriate variables to report the date and time in ISO 8601 format or other character
format with reference time point. The CMCAT (Concomitant Medication Category)
should be used to differentiate “Prevention” versus “Treatment.” Additionally, you
should flag if analgesic/antipyretic administered (Y/N) using the variable CECONTRT
(Clinical Event Concomitant or Additional Treatment Given).
A telephone contact that is used to ensure diary reporting compliance and/or to ascertain if
any unsolicited AEs have occurred is considered a subject visit. Please report the study day
of start of visit (SVSTDY), the visit number (VISITNUM) and the visit name (VISIT) as
“telephone contact 1 (or 2 or 3, etc.)” for each subject.

Contains Nonbinding Recommendations
5
3.2 	Unsolicited Adverse Events (AEs)
Unsolicited AEs should be represented in the AE domain, regardless of whether they
occurred during the prespecified assessment interval or after the prespecified assessment
interval. The clinical details of the AE should be represented in the FAAE domain and
possibly in the VS, CM, Procedures (PR), Healthcare Encounter (HO), and Death Detail
(DD) domains. The following variables should be utilized in the AE domain: --SEV
(Severity) (or –TOXGR (Standard Toxicity Grade)), --LLT (Lowest Level Term), --
LLTCD (Lowest Level Term Code), --HLT (High Level Term), --HLTCD (High Level
Term Code), --HLGT (High Level Group Term), --HLGTCD (High Level Group Term
Code), --SOC (Primary System Organ Class) and --SOCCD (Primary System Organ
Class Code). Additionally, you should report in the AETRTREF (Time Point Reference)
variable the reference time point of “Dose 1, 2, or 3 etc.” and AEELTM (Planned Elapsed
Time from Time Point Reference) variable to indicate the beginning of the event (in days,
hours/minutes if necessary) from the last vaccination administered (referenced as dose 1,
2, or 3 etc.).
In the AE domain, there should be one record per AE per subject for each unique event.
The AE record reported by the investigator is submitted as a summary record, “collapsed”
to the highest level of severity, causality, seriousness, and final outcome. The variables
for study day/date should be utilized (AESTDY/AESTDTC and AEENDY/AEENDTC).
The FAAE domain should be used to report the day-to-day information, concerning the
AE. The CM, PR, HO, and/or DD domains should be utilized if the corresponding
variables in AE are marked as “yes.” A dataset-level relationship in RELREC should be
provided.
3.3 	Laboratory Safety Assessments
Laboratory safety assessment of clinical chemistry, hematology, and urine should be
included in the LB domain. If the laboratory result for a particular assessment for a
subject is outside the normal range, it should additionally be reported in the AE domain,
with the highest level noted. A dataset-level relationship in RELREC should be used.
3.4 	Medically Attended Adverse Events (MAAEs)
MAAEs should be reported in the AE and HO domain, and if needed, additional data
should be reported in the FAAE, VS, and/or DD domains.
In certain circumstances, OVRR may request that a clinical trial protocol include
collection of data relating to specific MAAEs, particularly autoimmune or auto-
inflammatory diseases, among subjects in all treatment groups through 12 months or
longer, following the last study vaccination. Since the request for this information is
based on a theoretical potential, CBER considers the occurrence of a vaccine-related
potential immune-mediated medical conditions (PIMMC) to be unexpected.
When a particular MAAE is noted as a new onset of a chronic disease (NOCD) or a
PIMMC it should be reported in the AE domain and categorized in CAT with “NOCD” or
“PIMMC,” respectively.

Contains Nonbinding Recommendations
6
If a chronic disease is exacerbated, following a vaccination, the MH domain should
contain information on the disease, prior to the first day of vaccination, and the AE
domain should contain information on the disease, following vaccination. The CAT and
Subcategory (-SCAT) variables should be populated in both domains with a
categorization/subcategorization of the chronic disease, e.g., kidney failure would be
categorized as “renal medical history” and subcategorized as “renal failure.” A dataset-
level relationship in RELREC should be used.
3.5 	Deaths
When a death occurs, it should be reported in the AE domain with supplemental death
detail information provided in the DD domain. The DM domain should also indicate that
a death has occurred, using the flag variable (DTHFL) and date/time of death variable
(DTHDTC).
4.0 	SUBMISSION OF CLINICAL DISEASE ENDPOINT EFFICACY DATA FOR
VACCINE TRIALS
For clinical trials in which efficacy against clinical disease is an objective, efficacy data will
primarily be reported in the CE domain with specific information provided in the MB domain,
FACE domain, VS domain, and PE domain (if necessary).
To differentiate an efficacy clinical event (i.e., case of the disease of interest) from other clinical
events, the variable CECAT should be utilized, and “efficacy” should be stated. Since efficacy is
pre-specified, the variables CEOCCUR, Clinical Event Completion Status (CESTAT), and
Clinical Event Reason Not Collected (CEREASND) should be included. The CETERM for the
efficacy clinical event should be indicative of the disease, e.g., for an influenza vaccine,
“influenza-like illness or ILI;” or for a human papillomavirus vaccine, “cervical intraepithelial
neoplasia or CIN. Test results from assays conducted to confirm the presence of the microbe of
interest (e.g., polymerase chain reaction (PCR), enzyme-linked immunosorbent assay (ELISA),
cell culture, etc.) should be reported in the MB domain. If the disease of interest is confirmed
(e.g., by the clinical investigator or by an adjudication committee, using the information from a
clinical assessment and the results of confirmatory tests), it should be reported in CE variable
CEDECOD (e.g., Influenza), and flagged in a supplemental DM domain with QNAM =
“CDECASE,” (QLABEL) = “clinical disease endpoint case flag,” and (QVAL) = “Y.”
If the disease of interest is not confirmed by either the clinical investigator or by an adjudication
committee, the results should be flagged in a supplemental DM domain with QNAM =
“CDECASE,” QLABEL = “clinical endpoint case flag,” and QVAL = “N.” The manner in which
the occurrence of a disease of interest is confirmed (e.g., by an adjudication committee or by the
clinical investigator or by another mechanism) should be clearly delineated in the clinical
protocol. A check box in the CRF, indicating the status of the disease of interest, is
recommended.
If hospitalization occurs during the clinical trial because of occurrence of the “disease of
interest,” the CE domain variable Clinical Event Requires or Prolongs Hospitalization
(CESHOSP) should be marked with “Y.” Additional data regarding the hospitalization and/or
other healthcare encounter should be provided in the HO domain.

Contains Nonbinding Recommendations
7
If concomitant or additional treatment is prescribed to alleviate the severity of the disease of
interest, the CE domain variable Clinical Event Concomitant or Additional Treatment Given
(CECONTRT) should be utilized.
Variables available in AE that are also available for use in CE consist of Serious (SER),
Outcome of Event (OUT), Involves Cancer (SCAN), Congenital Anomaly or Birth Defect
(SCONG), Persistent or Significant Disability/Incapacity (SDISAB), Is Life Threatening
(SLIFE), Results in Death (SDTH), and Occurred with Overdose (SOD).
5.0 	SUBMISSION OF IMMUNOGENICITY DATA FOR VACCINE TRIALS
Data related to the assessment of immunogenicity should be reported in the Immunogenicity
Specimen (IS) domain. The LB domain should not be used for immunogenicity data.
6.0 	SUBMISSION OF COMCOMITANT MEDICATION DATA FOR VACCINE
TRIALS
Previous immunizations should be reported in the CM domain. You should include the variable
CMPRESP along with the accompanying variables -OCCUR, -STAT and -REASND, if the
particular immunization is prespecified.
7.0 	SUBMISSION OF DATA FOR MATERNAL IMMUNIZATION TRIALS
The relationship between the mother and child should be represented in Relationship of Study
Subjects (RELSUB). To allow for the possibility that a mother is treated before the number of
fetuses is known, fetuses are given a Unique Subject Identifier (USUBJID) value that appends an
A or B to the mother’s USUBJID.
In the DM domain, the informed consent and vaccination administration date for the fetus are the
same as for the mother. The age of enrollment will be a negative number in weeks. The birth
date for the infant should be reported, using the variable Date/Time of Birth (BRTHDTC).
Since the amount of vaccine that reaches the fetus is unknown, the Dose Description (DOSTXT)
variable in EX is utilized, and “Fetal Exposure” should be noted instead of populating the
variable Dose (DOSE).
The reproductive findings (RP) domain should be used to report the maternal date of last
menstrual period, and from this, the gestational age at exposure can be calculated and reported in
RP.