Download Therapeutic Area Data Standards User Guide for QT

Therapeutic Area Data Standards
User Guide for QT
Version 1.0 (Draft)
Prepared by the
CFAST QT Team
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Notes to Readers
•
•
•
This is the draft version 1.0 of the Therapeutic Area Data Standards User Guide for QT.
This document corresponds to the SDTM v1.4 and SDTMIG v3.2 and to the ADaM v2.1 and ADaMIG
v1.0.
The TAUG-QT v1.0 package contains a user guide, a draft domain model, and a draft specification for a
new relationships dataset.
Revision History
Date
TBD
2014-07-31
Version
1.0
1.0 Draft
Summary of Changes
Provisional
Draft for Public Review
See Appendix F for Representations and Warranties, Limitations of Liability, and Disclaimers.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 2
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
CONTENTS
1
INTRODUCTION .............................................................................................................. 5
1.1
1.2
1.3
1.4
1.5
1.6
1.7
PURPOSE.............................................................................................................................................................5
CLINICAL GUIDELINES FOR QT STUDIES ............................................................................................................6
ORGANIZATION OF THIS DOCUMENT...................................................................................................................7
CONCEPT MAPS ..................................................................................................................................................7
CONTROLLED TERMINOLOGY .............................................................................................................................8
RELATIONSHIPS TO OTHER STANDARDS .............................................................................................................8
KNOWN ISSUES...................................................................................................................................................9
2
ECG OVERVIEW .............................................................................................................. 9
2.1
QT/QTC ASSESSMENTS IN CLINICAL STUDIES ...................................................................................................9
2.2
EXPLANATION OF COMMON TERMS ..................................................................................................................10
2.3
THE FUNDAMENTALS OF ECGS ........................................................................................................................12
2.3.1
What is an ECG and What Information Does it Provide? ........................................................................12
2.3.2
ECG Machinery .......................................................................................................................................16
2.3.3
ECG Device Types ...................................................................................................................................20
2.3.4
How is the QT Interval Adjusted (Corrected) for Heart Rate? .................................................................21
3
THE THOROUGH QT (TQT) STUDY .......................................................................... 23
4
TRIAL DESIGN ................................................................................................................ 26
4.1
TQT STUDY DESIGN.........................................................................................................................................26
4.1.1
Parallel Studies ........................................................................................................................................26
4.1.2
Crossover Studies.....................................................................................................................................27
4.1.3
Trial Elements ..........................................................................................................................................29
4.1.4
Trial Summary Parameters .......................................................................................................................30
4.2
TIME POINT PLANNING .....................................................................................................................................32
5
SUBJECT CHARACTERISTICS AND ELIGIBILITY............................................... 35
5.1
5.2
INCLUSION/EXCLUSION CRITERIA ....................................................................................................................35
PHARMACOGENETICS .......................................................................................................................................37
6
STUDY ASSESSMENTS.................................................................................................. 38
6.1
ECG ASSESSMENTS..........................................................................................................................................38
6.1.1
Specification for ECG Test Results ..........................................................................................................40
6.1.2
Examples for ECG Test Results ...............................................................................................................44
6.2
QT CORRECTION ........................................................................................ERROR! BOOKMARK NOT DEFINED.
6.2.1
Specification for ECG QT Correction ......................................................................................................51
6.2.2
Examples for ECG QT Corrections .........................................................................................................53
6.3
PK ASSESSMENTS ......................................................................................ERROR! BOOKMARK NOT DEFINED.
6.4
HEMODYNAMICS/VITAL SIGNS .........................................................................................................................55
6.4.1
Example for Orthostatic Challenge ..........................................................................................................58
7
DATA ANALYSIS ............................................................................................................. 59
7.1
ADEG EXAMPLE ..............................................................................................................................................60
7.1.1
Analysis Data Metadata ...........................................................................................................................60
7.1.2
Analysis Dataset.......................................................................................................................................64
7.2
ADQT EXAMPLE ..............................................................................................................................................67
7.2.1
Analysis Data Metadata ...........................................................................................................................67
7.2.2
Analysis Dataset.......................................................................................................................................68
7.3
BASELINE ALTERNATIVES AND DEFINITION OF TREATMENT DIFFERENCES ......................................................69
Page 3
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
7.3.1
7.3.2
7.3.3
7.3.4
Time-Matched Baseline; Double-Delta Treatment Difference.................................................................69
Time-Averaged Baseline; Double-Delta Treatment Difference ...............................................................71
Predose Averaged Baseline; Double-Delta Treatment Difference ...........................................................72
No Baseline; Single-Delta Treatment Difference .....................................................................................74
APPENDICES ............................................................................................................................. 75
APPENDIX A: PROJECT PROPOSAL ................................................................................................................................75
APPENDIX B: CFAST ORGANIZATIONS ........................................................................................................................75
APPENDIX C: WORKGROUP ..........................................................................................................................................76
APPENDIX D: GLOSSARY AND ABBREVIATIONS ............................................................................................................77
APPENDIX E: REFERENCES ...........................................................................................................................................78
Appendix E1: Figures ..............................................................................................................................................79
Appendix E2: Further Reading ................................................................................................................................79
APPENDIX F: REPRESENTATIONS AND WARRANTIES, LIMITATIONS OF LIABILITY, AND DISCLAIMERS..........................80
LIST OF FIGURES
Figure 1: CDISC Industry Wide Data Standards ...........................................................................................................5
Figure 2: Concept Classification Coding Key for Concept Maps..................................................................................8
Figure 3: Diagram of the Heart’s Conduction System................................................................................................. 13
Figure 4: Sequence of Heart Excitation and the Associated ECG Waveforms............................................................ 14
Figure 5: Electrocardiogram Waveform Illustration.................................................................................................... 15
Figure 6: Standard Electrode/Lead Configuration for a 12-Lead/Waveform ECG ..................................................... 17
Figure 7: 12-Lead Electrocardiogram Waveform ........................................................................................................ 18
Figure 8: High-Level View of ECG Hardware-Software: From Subject’s Chest to Data for Analysis....................... 19
Figure 9: Relationship between the Bazett- and Fridericia-Corrected QT Interval and RR Interval ........................... 22
Figure 10: Parallel Study Design Schema for Example TQT Study 1 ......................................................................... 26
Figure 11: Crossover Study Design for Example TQT Study 2 with Washout Period Combined with each Treatment
Element ...................................................................................................................................................... 28
Figure 12: Crossover Study Design for Example TQT Study 2 with Separate Elements for each Washout Period.... 28
Figure 13: Individual Beats and Relevant Identifying Information Within a 24-hour Continuous Recording ............ 33
Figure 14: ECG Replicates and Nominal Time Points ................................................................................................ 34
LIST OF CONCEPT MAPS
Concept Map 1: Recording of ECG Digital Waveforms ............................................................................................. 19
Concept Map 2: Measurement of RR and Intervals for Use in Analysis ..................................................................... 20
Concept Map 3: Potential Parameters and Their Order of Collection ......................................................................... 35
Concept Map 4: ECG Quantitative Results and Morphological (Qualitative) Findings Determination ...................... 39
Concept Map 5: Population QT Correction for a Pre-specified Model ....................................................................... 49
Concept Map 6: Individual QT Correction for a Pre-specified Model ........................................................................ 49
Concept Map 7: Individual QT Correction (QTc) ‘Best Fit’ Model ............................................................................ 50
Concept Map 8: PK/PD Analysis Steps and Timing ................................................................................................... 55
Concept Map 9: Assessment of Orthostasis ................................................................................................................ 56
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 4
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
1 Introduction
This Therapeutic Area Data Standards User Guide for QT (TAUG-QT) was developed under the Coalition for
Accelerating Standards and Therapies (CFAST) initiative (Appendix A).
CFAST, a joint initiative of Clinical Data Interchange Standards Consortium (CDISC) and the Critical Path Institute
(C-Path), was launched to accelerate clinical research and medical product development by facilitating the
establishment and maintenance of data standards, tools, and methods for conducting research in therapeutic areas
important to public health. CFAST partners include TransCelerate BioPharma Inc. (TCB), the U.S. Food and Drug
Administration (FDA), and the National Cancer Institute Enterprise Vocabulary Services (NCI EVS), with
participation and input from many other organizations. See Appendix B for a description of CFAST participating
organizations.
CDISC has developed industry-wide data standards enabling the harmonization of clinical data and streamlining
research processes from protocol (study plan) through analysis and reporting, including the use of electronic health
records to facilitate study recruitment, study conduct and the collection of high quality research data. CDISC
standards, implementations and innovations can improve the time/cost/quality ratio of medical research, to speed the
development of safer and more effective medical products and enable a learning healthcare system.
19
20
21
22
23
24
The goal of the CFAST initiative is to identify a core set of clinical therapeutic area concepts and endpoints for
targeted therapeutic areas and translate them into CDISC standards to improve semantic understanding, support data
sharing and facilitate global regulatory submission.
25
1.1 Purpose
26
27
28
29
30
31
32
33
34
35
36
37
Figure 1: CDISC Industry Wide Data Standards
This draft version 1.0 (v1.0) of the TAUG-QT highlights the data endpoints for clinical studies characterizing the
QT effects of drugs in healthy volunteers or in patients. The TAUG-QT endeavours to cover all studies specifically
directed at the QT invertal, with extra attention to a specific type of study, the “thorough QT (TQT) study,” which is
operationally defined by an ICH E14 guidance document 1 and the associated Question & Answer document 2 (see
Section 3 for additional details). See Appendix A for the project proposal that was approved by the CFAST Steering
Committee.
Because this TAUG is directed at the evaluation of the QT in clinical studies and even more specifically at the
conduct and analysis of the TQT study, certain aspects of the processing of QT data (e.g., correction of the QT
interval for heart rate, potential complexities of evaluating and interpreting a change in QT) will differ from the
processing of QT data used in routine clinical practice. While routine clinical practices focus on the individual
patient and substantial changes in QT and/or absolute QT values, TQT studies can find even relatively small
Page 5
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
differences between groups of patients treated with an experimental drug compared to a control treatment to be of
interest. This difference in interests leads to the need for different processing. Also, TQT studies are generally
limited to adult healthy volunteers of a non-geriatric age and having normal ECGs at screening. Data collection is
also performed under highly controlled, inpatient conditions. Therefore, a number of considerations in other clinical
trials are not emphasized in this TAUG. For example, while potential differences in incidence of adverse events
based on various demographic characteristics might be included for some large Phase 3 studies, these would be
unusual analyses to include in a TQT study. It would also be highly unexpected to observe adverse events related to
cardiac ischemia, congestive heart failure, or other non-electrophysiological aspects of cardiac function, and
therefore such events do not receive special attention in this TAUG. An analysis of all adverse events during the
course of any study would be expected as a routine component of any study report.
This TAUG-QT v1.0 describes the most common data needed for QT interval or heart rate corrected QT interval
(QTc) studies, especially TQT studies, so that those handling the data (e.g., data managers, statisticians,
programmers) understand the data and can apply standards appropriately. Descriptions addressed in this TAUG-QT
v1.0 include the clinical situations from which the data arise, and the reasons these data are relevant for QT/QTc
studies.
The TAUG-QT v1.0 also strives to define research concepts unambiguously, so that consistent terminology can be
used in QT/QTc studies to enable aggregation and comparison of data across studies and drug programs.
And finally, the TAUG-QT v1.0 describes how to use CDISC standards to represent the data:
• For the Study Data Tabulation Model (SDTM) and the SDTM Implementation Guide for Human Clinical
Trials (SDTMIG), these instructions include guidance on which domains data should be stored in, how
variables should be used, and example datasets and controlled terminology for ECG replicate (generally
conventional 10-second, 12-lead ECGs recorded as such or extracted from continuous recording of longer
duration) and single beat measurements.
• For the Analysis Data Model (ADaM) and ADaM Implementation Guide (ADaMIG), an example of
metadata and dataset for ADEG are provided, and baseline alternatives for treatment differences are
described and modeled with ADaM variables.
These CDISC standards are freely available at http://www.cdisc.org. It is recommended that implementers consult
the SDTM prior to implementing these QT clinical data standards.
It is recommended that the reader be familiar with common electrocardiogram (ECG) terminology, which can be
found in Section 2.2, as these terms have very specific connotations within the context of this document. A
comprehensive ECG overview is also provided in Section 2.3 to ensure that the reader has a strong foundation to
understand the intricacies of these QT data standards.
It is important to note that the inclusion of concepts in this user guide should not be construed as a requirement to
collect data on these concepts in any particular study of QT/QTc. The examples included are intended to show how
data of particular kinds can be represented using CDISC standards, if such data are collected. They are not intended
to imply that such data must be collected, or even that such data should be collected.
This document does not contain guidance from a regulatory authority and should not be construed as substituting for
or replacing any documents published by such an authority.
83
1.2 Clinical Guidelines for QT Studies
84
85
86
87
88
89
90
91
A research concept is a unit of knowledge created by a unique combination of characteristics that define
observations of real world clinical research phenomena, including the question and assessment that result in a
measurable outcome. Research concepts specify what should be observed for a specific subject assessment in a
clinical study, but not how to capture the data or how to group observations together. Research concepts include data
that represent a set of roles such as a topic or qualifiers. Research concepts represent clinical research knowledge
that borrows from medical knowledge, statistical knowledge, BRIDG, and the CDISC standards. The authors of this
user guide considered the International Conference on Harmonisation of Technical Requirements for Registration of
Pharmaceuticals for Human Use (ICH) Efficacy (E) 14 Guidance for Industry: The Clinical Evaluation of QT/QTc
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 6
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
Interval Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic Drugs (May 2005) 1 and associated
Question and Answers (R1 and R2) documents (May 2012 and March 2014, respectively) 2, 3 when drafting this user
guide.
The guidance provides recommendations to sponsors concerning the design, conduct, analysis, and interpretation of
clinical studies to assess the potential of a drug to delay cardiac repolarization. Tests for the effects of new agents on
the QT interval and QTc, as well as the collection of cardiovascular adverse events, were noted as recommended
assessments. The guidance also recommends that the investigational approach used for a particular drug should be
individualized, depending on the pharmacodynamic (PD), pharmacokinetic (PK), and safety characteristics of the
product, as well as on its proposed clinical use. The assessment of the effects of drugs on cardiac repolarization
remains the subject of active investigation. Further details regarding ICH E14 guidance can be found in Section 3.
This user guide focuses primarily on how to represent study design elements, time point planning, recommended
exclusion criteria, pharmacogenetic considerations, study assessments and key data analysis components for QT
studies. This document also covers elements of routinely collected data (e.g., medical history, adverse events) that
might be of particular interest in QT studies.
1.3 Organization of this Document
The TAUG-QT v1.0 differs from that of other CDISC therapeutic area user guides, which categorize the concepts
covered into subject/disease characteristics, disease assessments, and routinely collected data, because QT studies
are associated with drug safety rather than with a particular disease or condition to be treated. This is reflected in the
organization of this document, as follows:
•
•
•
•
•
•
•
•
Section 1, Introduction, provides an overall introduction to the purpose and goals of the QT project.
Section 2, ECG Overview, provides a general overview of ECG terminology and concepts.
Section 3, The Thorough QT (TQT) Study, provides background information regarding TQT studies.
Section 4, Trial Design, covers trial design for studies evaluating QT, particularly TQT studies.
Section 5, Subject Characteristics and Eligibility, highlights considerations for inclusion/exclusion criteria
and pharmacogenetics in QT studies.
Section 6, Study Assessments, provides insight into ECG assessments and machinery, PK assessments and
hemodynamics/vital signs.
Section 7, Data Analysis, includes key data analysis elements for a QT study.
Appendices provide additional background material and describe other supplemental material relevant to
the QT/QTc studies.
A list of domains used in the examples in this document, and the sections in which these examples appear, is given
below:
129
Domains from SDTMIG
Findings
EG – ECG Test Results
QT – ECG QT Correction Model Data*
VS – Vital Signs
Trial Design
TA – Trial Arms
TE – Trial Elements
TS – Trial Summary
* Domain is not final.
130
1.4 Concept Maps
131
132
This document uses concept maps to explain clinical processes and research concepts. Concept maps, also
sometimes called mind maps, are diagrams which include “bubbles” representing concepts/ideas/things and labeled
Page 7
Draft
Section
6.1.2
6.1.2
6.4.1
Domains from SDTMIG-MD Section
DI – Device Identifier
2.3.3.1
DO – Device Properties
2.3.3.1
4.1.1, 4.1.2
4.1.3
4.1.4
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
arrows that represent the relationships between the concepts/ideas/things. They are generally easier to draw and
more accessible than more formal modeling diagrams, such as Unified Modeling Language (UML) diagrams.
The diagrams in this document use the following coding for classification of concepts. This classification is based
on classes in the Biomedical Research Integrated Domain Group (BRIDG) model. These color-symbol pairs have
been used to highlight kinds of things that occur commonly in clinical data and therefore give rise to common
patterns of data. Some concepts are not coded; they have a thinner, black outline, and no accompanying symbol.
These may include the subject of an observation, as well as characteristics, or attributes, of the coded concepts.
Figure 2: Concept Classification Coding Key for Concept Maps
1.5 Controlled Terminology
CDISC Controlled Terminology is a set of standard value lists that are used throughout the clinical research process,
from data collection through analysis and submission. Terminology applicable to SDTM and/or ADaM is either in
production or under development by the CDISC Terminology Team at the time of publication of this document.
Production terminology is published by the National Cancer Institute’s Enterprise Vocabulary Services (NCI EVS)
and is available at: http://www.cancer.gov/cancertopics/cancerlibrary/terminologyresources/cdisc.
CDISC Controlled Terminology is updated quarterly. Because this document is a static publication, it refers readers
to the NCI EVS page for CDISC terminology (at the link given above). For the same reason, this document cannot
claim to use controlled terminology in either the lists of laboratory tests, assessment procedures, or in the examples
provided; users should not refer to these as the ultimate authority on what terms to use or to not use.
1.6 Relationships to Other Standards
This section describes the relationship of this document to other standards, whether CDISC or external. This
document does not replace the foundational CDISC standards or their implementation guides. The user should read
those standards and implementation guides before applying the advice in this user guide.
There are multiple types of data that have existing CDISC-based standards that can be used in QT studies without
additional development or customization (e.g. demographics, clinical laboratory tests, and pharmacokinetics), and so
are not covered in special detail in this document.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 8
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
164
165
166
167
168
169
170
171
172
173
This document uses a domain model, a dataset specification, and two variables which are not final at the time of
publication and therefore are subject to change or deletion without formal notice. Please check the most recent
version of the SDTM, SDTMIG, and SDTMIG-MD to ascertain their current status.
• The ECG QT Correction Data Model (QT) domain is still a work in progress (i.e. a “draft” domain), which
has not appeared prior to this TAUG-QT.
• The Device-Device Relationships (RELDEV) dataset is a draft relationships dataset, designed for eventual
inclusion in the SDTMIG-MD, which has also not appeared prior to this TAUG-QT.
• The --REPNUM (Repetition Number) variable is a variable which has been approved by SDTM
Governance but has not yet been included in a formally published version of the SDTM.
• The EGBEATNO (ECG Beat Number) variable is also an approved but unpublished variable.
174
1.7 Known Issues
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
Location of QT correction coefficients in submission data
For studies in which QTc is a major or the primary endpoint, advanced statistical methods may be used for the
correction the QT interval. Unlike the straightforward population correction formula such as Fridericia's or Bazette's,
these other methods are based on regression modeling techniques and may utilize external data or within-subject
data to derive the corrected values. The correction formula produced by these methods, along with the model
coefficients, is considered important data to be included in a submission. Such statistical analyses are not always
performed by the sponsor and may be done by the same vendor that creates the database of uncorrected, collected
measurements.
There are different perspectives on whether the output from these statistical methods should be submitted in the
SDTM or ADaM folder. There is precedence to consider any data originating from a central lab or other vendor as
raw data, regardless of the number or complexity of calculations performed to obtain them, and therefore should be
submitted in SDTM. However, clearly QT correction coefficients that are determined via statistical modeling
represent derived, analysis-related data, and would be considered data submitted in the ADaM folder.
This TAUG-QT v1.0 was developed using the rationale that data that is vendor-supplied should be submitted in
SDTM (Section 6.2) and sponsor-derived data should be submitted in ADaM (Section 7.2). It is recognized that this
rationale results in variability with respect to the location of these data in submissions. Therefore, this approach may
change.
SDTM baseline flag
It is recognized that the SDTM baseline flag is inadequate for analyses in certain situations, among them crossover
studies where each treatment period could have its own baseline. There are ongoing discussions about whether, in
these situations, the SDTM baseline flag should be populated since it could be misleading. In the vital signs example
in this document (Section 6.4.1), the baseline flag VSBLFL was not populated, and it was noted that this omission
would be explained in the Study Data Reviewers Guide. Other approaches are possible, and future implementation
advice on a preferred approach may be different from that shown in the example
202
2 ECG Overview
203
2.1 QT/QTc Assessments in Clinical Studies
204
205
206
207
208
209
210
When ventricular repolarization (i.e. the relaxing of the bottom chambers of the heart, which perform the majority of
its pumping action) is delayed, it can lead to cardiac arrhythmias which may be fatal. Historically, a number of drugs
have been found to cause such a delay, thus, determining whether a drug delays ventricular repolarization is
important in assessing its safety. One measure of the time required for ventricular repolarization is the QT interval
on an ECG, corrected for heart rate; or QTc. QTc is clearly recognized as an imperfect biomarker for increased risk
of fatal arrhythmia because it can be increased by a number of drugs that are not associated with a significant
incidence of such arrhythmias. On an individual basis, the increase in QTc generally needs to be substantial to place
Page 9
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
the patient at risk. However, for a potential new drug, even a slight mean increase * in QTc can be clinically
meaningful, in that some degree of risk cannot be excluded in a small number of individuals in a large population
that will receive the drug during its use in clinical medicine.
QTc can be assessed in a variety of study types with a variety of analyses:
• QTc is one of many safety parameters assessed in early Phase 1 Single Ascending Dose (SAD) and
Multiple Ascending Dose (MAD) safety studies. These studies are typically small, with participants that are
adult, non-geriatric, healthy, and with normal ECGs at screening. These studies collect and analyze ECG,
PK, and PD data from multiple time points both before and after study treatment.
• QTc is also one of the multiple safety parameters that might be assessed in Phase 2–4 clinical trials. Such
trials collect baseline ECGs and then collect additional ECGs at one or more time points subsequent to
initiation of study treatments. Analyses generally consist of comparing change from baseline for active
drug with the comparable change for placebo and/or active control and also comparing the incidences of
values exceeding absolute upper reference limits (e.g., values above 500 ms regardless of magnitude of
change) and positive change from baseline reference limits (e.g., increases from baseline ≥60 ms regardless
of absolute value).
• The TQT study has QTc as its primary outcome of interest.
Data collected in these types of studies are discussed in this document.
Sometimes QTc is an incorrect or insufficient measure:
• When the drug under study causes substantial changes in autonomic tone, this can cause significant
changes in heart rate, complicate the correction of QT for heart rate, and even alter the relationship between
QT and heart rate. This circumstance is the primary one of those under which a conventional TQT might
not be appropriate and alternative/complementary methods and/or analyses are required. A White Paper has
recognized five such alternative/complementary methods and/or analyses 4.
• Alternatives to the TQT study of QTc as assessment of the potential for a drug to cause fatal arrhythmias
have been published 5. These alternatives may predict the potential for a drug to induce a fatal cardiac
arrhythmia or, conversely, be able to demonstrate absence of such a liability. Furthermore, some of these
proposals have suggested using morphological aspects of the T wave that can be quantitated (e.g.,
ascending and descending slopes height (voltage)) as primary outcomes of interest 6.
• Several groups of authors have described a set of properties of a drug (i.e., triangulation, reverse use
dependence, instability, and dispersion; or TRIaD) that relate to the depolarization and repolarization, or
action potential (AP), of the ventricle, which they believe are more relevant to prediction of potential for
induction of a fatal arrhythmia than is prolongation of the QTc obtained from surface ECGs in human
subjects 7.
Detailed discussion of these topics is out of scope for this document.
2.2 Explanation of Common Terms
The terms discussed below have very specific connotations within the context of this document. Familiarity with this
terminology should improve the understanding of the content.
Common Term
Cardiologist over-read ECG
Central ECG laboratory
*
Explanation
An ECG for which a cardiologist, using ECG processing hardware and software,
makes a final determination of ECG quantitative data (interval measurements) and an
assessment of ECG qualitative findings (abnormalities of axis, rhythm, conduction,
evidence of ischemia, etc.). See also Over-read (ECG).
A commercial facility (some pharmaceutical companies maintain internal central ECG
laboratories) that provides a broad range of services, globally, supporting the
acquisition of ECGs during clinical studies, including project management, ECG
management, data management, ECG over-reading and analyses, and reporting of
>5 milliseconds (msec) would be considered to exceed random variability 19 and a mean increase of ≥10 msec could be clinically significant.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 10
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Common Term
Clinical study
ECG, EKG
ECG machine
Continuous, high-fidelity,
12-lead ECG recording
HR
Machine-read ECG
Over-read (ECG)
QT
QTc
RR Interval
Page 11
Draft
Explanation
ECG results. This term is synonymous with ECG core laboratory, vendor, central
vendor, and core lab.
A study (also known as a trial) conducted in human subjects according to procedures
and methods described in a written protocol.
Electrocardiogram. A graphical recording (also termed a tracing) of the electrical
activity of the heart, conventionally recorded from electrodes placed on the body
surface (surface electrodes). The standard ECG records and displays 12 wave forms
referred to as leads.
Equipment designed to record and graphically display signals from the electrical
activity of the heart as detected from electrodes placed on the body surface (ECGs).
This equipment is also known as an ECG cart or an ECG acquisition device. ECG
machines used for most clinical studies must record high-fidelity (1000 Hz) digital
data for 10-second recording intervals and provide automated machine-read
measurements of quantitative ECG data.
Equipment that continuously records the electrocardiogram for up to 24 hours (or
longer, depending on memory capacity of the equipment being used) for all 12
conventional ECG leads with the same digital quality as electrocardiograms recorded
by a standard, 10-sec, 12-lead ECG machine described immediately above.
Conventional 10-sec ECGs can be extracted from continuous recordings.
Sometimes referred to as a Holter recording; however, the original Holter device
recorded at a lower fidelity and with fewer leads.
Heart rate. The measurement of the number of electrical depolarizations of cardiac
muscle, usually leading to contraction of the heart muscle, and under normal
circumstances, resulting in pumping of blood into the arterial system. The rates of
these electrical depolarizations are often referred to as cycles per minute (cpm) but can
be referred to as heart beats per minute (bpm). HR is assessed on an ECG tracing. HR
is not equivalent to pulse rate because pulse rate is obtained by palpation of blood
being pumped through the arteries by effective contraction of the heart muscle. Under
normal circumstances, pulse rate will be the same numerically as HR. However, with
certain abnormalities, these measurements will be numerically different. Pulse rate is
always referred to as bpm.
Automated measurements of ECGs generated by ECG machines (ECG cart, ECG
acquisition device). The corrected QT (QTc) is generally reported with Bazett’s
correction (QTcB) for HR. This information is useful for patient management but may
not be recommended for data analyses that will be used for definitive internal decision
making or for regulatory submission. Some machines can report Fridericia’s
correction (QTcF) or a proprietary correction. ECG machines also provide automated
measurements of heart rate, PR interval, QRS width and axis (QRS vector measured
in degrees), and other quantitative parameters as well as reporting abnormal
qualitative findings
The process of measuring and interpreting the waveforms recorded on each ECG as
related to the electrical conductivity and functioning of the heart. Methods for overreading may be: 1) fully manual by cardiologist or medical professional (ECG
technician or non-cardiologist physician); 2) computer assisted by cardiologist or
medical professional; or 3) fully automated (computerized interpretation of ECGs).
QT interval. The portion of an ECG between the onset of the QRS complex and the
end of the T-wave, representing the total time for ventricular depolarization and
repolarization.
HR-corrected QT interval. The QT interval is inversely related to HR. When the QT
interval is corrected for HR by use of various formulas, it is expressed as QTc and
allows an assessment of the QT interval that is intended to be independent of HR.
The time between 2 consecutive heart beats/cycles (P-QRS-T complexes), measured
as the time between the peaks of 2 consecutive R waves. The RR interval is the time
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Common Term
Subject
TQT study
Explanation
between individual heart beats and is related to HR in that HR is essentially the
number of R waves in 1 minute. In most QT correction formulas the RR interval is
used for correction.
An individual under study. Subjects may be normal healthy individuals or patients
with a particular disease or condition.
A clinical study focused on the investigation of possible study drug effects on
ventricular repolarization as evidenced by changes in the QTc of the ECG following
the ICH E14 guidelines 1. The primary statistical analysis is a noninferiority
comparison of study drug to placebo. That is to say, the goal of the study is to prove
the absence of a drug-prolonging effect on QTc within a predefined upper bound on
the confidence interval about the difference between drug and placebo that will be
accepted as demonstrating no prolongation.
253
2.3 The Fundamentals of ECGs
254
2.3.1 What is an ECG and What Information Does it Provide?
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
An electrocardiogram (ECG, or EKG) is a graphical recording, or tracing, of the electrical activity of the heart
recorded at the body surface and depicted as a series of waveforms that illustrate voltage over time. A 12-lead ECG
is useful as an independent marker of cardiac disease and reflects abnormalities of the heart that can arise from
anatomic, hemodynamic, molecular, ionic, and/or drug-induced causes.
The heart is a specialized muscle consisting of four chambers: the right and left atria (or receiving chambers) and the
right and left ventricles (or discharging chambers). The right atrium receives blood from the body, the right ventricle
discharges blood to the lungs, the left atrium receives blood from the lungs, and the left ventricle discharges blood to
the body. The cardiac cycle consists of a period of relaxation called diastole, during which the heart fills with blood,
followed by a period of contraction called systole. Alternating contraction and relaxation of the muscle produces the
mechanical pumping action of the heart that carries blood around the circulatory system. Atrial contraction precedes
ventricular contraction by a short time, which is important for the pumping’s effectiveness.
The contraction of the heart is initiated by small amounts of electrical activity in the heart muscle cells. Certain
muscle cells are organized into a specialized conduction system, much as a set of wires would carry electrical
impulses in a computer (Figure 3 and Figure 4). For the heart to function properly as a pump, this electrical activity
must be accurately timed and coordinated in the various parts of the heart muscle in order for contraction and
relaxation to occur properly.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 12
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
274
275
276
277
278
279
280
281
Sinoatrial node
Atrioventricular node
Bundle of His
Left bundle branch
Left posterior fascicle
Left-anterior fascicle
Left ventricle
Ventricular septum
Right ventricle
Right bundle branch
Figure 3: Diagram of the Heart’s Conduction System †
As illustrated in more detail in Figure 4 below, electrical activity (depolarization and repolarization) in differing
components of the specialized conduction system are responsible for different components of the surface ECG
waveforms. In Figure 4, the action potentials are the electrical activity recorded from specific local areas in the
conduction system. These multiple action potentials give rise to the surface ECG waveform at the bottom of Figure.
The specific contribution of the action potentials to the surface wave form are color coded.
†
Heuser J, based on Lynch PJ, Jaffe CC. CC-BY 2.5. Available at: http://commons.wikimedia.org/wiki/File:RLS_12blauLeg.png.
Page 13
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Abbreviations:
SA - sinoatrial
AV - atrioventricular
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
Figure 4: Sequence of Heart Excitation and the Associated ECG Waveforms ‡
ECG waveforms include single electrophysiological events and combinations of electrophysiological events (Figure
4 and Figure 5; see also Figure 7). Figure 5 illustrates an ECG waveform complex and the measurement points for
its various components and intervals. The first part of an ECG waveform is called the P-wave and represents atrial
depolarization (also called activation; electrical activity in heart muscle cells that causes muscular contraction).
There is then a slight time delay before the next ECG waveform component, the QRS complex, which represents
ventricle depolarization. Another slight time delay occurs before the T-wave, which represents ventricular
repolarization (or recovery). The U-wave may not always be observed. There is no ECG component representing
atrial repolarization because this occurs at the same time as ventricular depolarization.
An ECG waveform is captured by an ECG machine and recorded on 1-millimeter (mm) grid paper, with the Y-axis
representing changes in electrical voltage and the X-axis representing change over time. When the ECG machine is
set to a paper speed of 25 mm/second, each mm on the X-axis represents 40 milliseconds (msec). Each mm on the
Y-axis customarily represents 0.1 microvolt (mV).
‡
Public domain.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 14
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
A
b
b
r
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
Figure 5: Electrocardiogram Waveform Illustration
Electrocardiogram recorded on grid paper with lines 1mm apart. X-axis is the time axis. Y-axis is the voltage
axis. Figure assumes a paper speed of 25mm/sec and a calibration of 10mm/mV.
Waveform components:
• PR interval − time from onset of atrial activation to onset of ventricular activation; includes P-wave and PR
segment. The PR interval actually ends at the beginning of the QRS complex and not the peak of the Rwave.
o P-wave − atrial depolarization
o PR segment − atrial ventricular (AV) nodal depolarization
• QT interval − duration of ventricular activation and recovery; includes the QRS, ST segment, and T-wave
o QRS complex − ventricular depolarization; obscures atrial repolarization
 Q-wave − the first negative wave in QRS
 R-wave − the first positive wave in QRS
 S-wave − the first negative wave following the R-wave
o ST segment − delay between ventricular depolarization and repolarization; all ventricular cells are
depolarized
o T-wave − ventricular repolarization
• Corrected QT interval (QTc) − QT interval corrected for heart rate (HR)
• U-wave − Purkinje fiber repolarization or possibly represents “after depolarizations” in the ventricles
caused by delayed repolarization of the ventricle
• RR interval − ventricular rate; time between cycles of depolarization/repolarization of the entire heart
HR represents the number of times the heart beats per minute. On an ECG, HR usually implies ventricular rate,
which means the number of QRS complexes in a minute. Normally, each QRS is preceded by a P-wave; hence, HR
also refers to atrial rate (the number of P-waves in a minute). There are different methods to calculate HR on an
ECG. The most widely used ECG machine algorithm derives the median elapsed time from the peaks of the
R-waves on the dominant beat type within the 10-second recording and, as HR = 60,000/RR when RR is expressed
in msec, derives the HR from the RR.
Page 15
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
The PR interval and QRS complex are of interest because they reflect muscle depolarization. Some drugs cause a
delay in muscle depolarization that can lead to a lack of coordination of cardiac electrical activity and impairment of
cardiac pump function.
The QT interval is primarily driven by the ventricular repolarization time, including the delay after depolarization
(the ST segment). Ventricular repolarization time is particularly important because if it is too long, ventricular
depolarization can begin to occur before repolarization is complete, causing the electrical activity of the heart to
become highly rapid and chaotic, resulting in loss of effective cardiac pump function and death. This abnormal
electrical activity is referred to as ventricular tachydysrhythmia (also termed ventricular tachyarrhythmia, ventricular
tachycardia). Specific types of ventricular tachydysrhythmia can occur, such as Torsades de Pointes, and ventricular
fibrillation.
The regulation of HR changes is based on the body’s need for energy sources. With respect to the ECG waveform,
an increase in HR is equivalent to an increase in the frequency of depolarization and to a decreased time between
individual depolarizations as denoted by a shorter RR interval. Changes in the frequency of cardiac depolarization
are detected by heart muscle cells and by the brain centers that influence cardiac electrical activity, which then alter
the rate of repolarization appropriately to maintain effective cardiac function.
As noted above, the time required for ventricular repolarization to occur (using the QT interval as a biomarker)
shortens as heart rate increases and lengthens as heart rate decreases to maintain normal physiological function.
Therefore, it is necessary to correct QT interval for the HR, by means of the RR interval measurement, in order to
meaningfully compare QT intervals that were measured at different HRs. The QT interval corrected for HR is
referred to as the HR-corrected QT (QTc) interval. QTc intervals are useful when comparing the QT intervals within
an individual from ECGs collected at different times when HRs were different or for comparisons between
individuals or groups. Approaches to QT interval correction are covered in Section 2.3.4.
Unfortunately, a number of factors can delay ventricular repolarization and thereby put individuals at risk of
ventricular tachyarrhythmia and sudden cardiac death. Use of drugs is a common cause of delayed ventricular
repolarization, so the effect of a drug on cardiac function must be clearly determined. Towards this end, the ECG
QTc is used as the standard method of assessing the potential for a drug to influence cardiac repolarization.
ECG assessments are covered in Section 6.1.
2.3.2 ECG Machinery
An ECG acquisition device (i.e. ECG machine, also called an ECG cart or cardiograph) records and graphically
displays the electrical impulses that stimulate the heart muscle to contract (electrical activity → muscle contraction
→ heart pumps blood). ECG acquisition devices used in clinical studies capture the heart muscle’s electrical activity
from electrodes on the torso and left leg (right and left arm leads are placed on the respective shoulders) that are
configured to form various types of leads (Figure 6).
Data collected from these electrodes are stored electronically in the ECG machine (Concept Map 1) and are printed
graphically as waveforms, which represent the electrical potential difference over time recorded from the ECG
leads. Bipolar limb leads (I, II, and III) record the electrical potential differences between two limb electrodes. The
unipolar precordial leads record the electrical potential difference between each of six designated sites on the torso
and a theoretical zero reference potential site, known as the Wilson central terminal that is composed of the mean of
3 limb leads. Three augmented limb leads (aVR, aVL, and aVF) are modified or augmented unipolar leads. These
augmented leads record the electrical potential difference between the limb leads and a different central terminal site
composed of the mean of the two other limb leads.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 16
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
Figure 6: Standard Electrode/Lead Configuration for a 12-Lead/Waveform ECG §
Precordial Electrodes locations:
•
•
•
•
•
•
V1: 4th intercostal space to right of sternum
V2: 4th intercostal space to left of sternum
V3: Half way between V2 and V4
V4: Left mid-clavicular line in the 5th intercostal space
V5: Left anterior axillary line in the 5th intercostal space
V6: Left mid-axillary line in the 5th intercostal space
Leads – potential differences between
Abbreviations:
3 bipolar limb leads:
aVR/aVL/aVF = augmented vector right/left/foot;
I: R and L arms (L is+)
R = right; L = left; V = vector
II: R arm and L leg (leg is +)
III: L arm and L leg (leg is +)
6 precordial leads (central terminal is average of the 3 limb leads not recorded)
V1: V1 (+) and central terminal (R and L arms and L leg)
V2: V2 (+) and central terminal (R and L arms and L leg)
V3: V3 (+) and central terminal (R and L arms and L leg)
V4: V4 (+) and central terminal (R and L arms and L leg)
V5: V5 (+) and central terminal (R and L arms and L leg)
V6: V6 (+) and central terminal (R and L arms and L leg)
3 augmented unipolar limb leads (central terminal is average of the 2 limb leads not recorded)
aVR: R arm (+) and central terminal (L arm and L leg)
aVL: L arm (+) and central terminal (R arm and L leg)
aVF: L leg (+) and central terminal (R and L arms)
On an ECG printout, the name for each lead is associated with its waveform. Each ECG tracing has 12 individual
waveforms (resulting from leads) on the standard ECG printout (Figure 7). ECG machines also construct a median
beat for each lead from all the beats for that lead, and then superimpose them by aligning the peak of the R-wave for
each median beat. This creates a superimposed median beat, which is used for measurement purposes of PR, QRS,
and QT. One lead is used to measure multiple RR intervals to obtain HR.
§
Left image: Public domain. Right image: CC0 1.0.
Page 17
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
Figure 7: 12-Lead Electrocardiogram Waveform
In addition to waveforms, ECG machines provide automated machine-read interval measurements and qualitative
assessments (Concept Map 1). The ECG machine stores the ECG waveforms, interval measurements (annotation
marks are placed on the appropriate lead for RR and on the superimposed median beat for other intervals for
automated measurement purposes, but these are not displayed on the ECG print out) and qualitative assessments for
electronic transmission to a server that is housed at a centralized ECG vendor. The server at the vendor (Concept
Map 2) can be used for human over-read of the ECG that consists of human adjustment of the annotation lines for
interval measurements and correction of any qualitative assessment. The human over-reader can adjust the
measurements made on the superimposed median beat and make measurements in any individual lead. The server
and its software allow high precision visualization of the waveforms by the human over-reader. If human over-read
is not performed, the automated measurements are transmitted to the sponsor for analysis.
In collecting and processing ECG data, moving from the subject to a dataset for analysis, there are generally two
distinct hardware-software devices that are employed. These devices are generally linked as a pair. Figure 8 is a
high-level representation of these devices, with more detailed representation in Concept Map 1 and
Concept Map 2. These concept maps also describe some of the functions of the two respective devices in going from
subject to data output for analysis.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 18
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
432
433
434
435
436
437
438
439
440
441
442
443
444
445
Figure 8: High-Level View of ECG Hardware-Software: From Subject’s Chest to Data for Analysis
Some vendors have their own proprietary hardware-software (a third device) for further processing of the digital
waveform information that is ultimately reported to the sponsor for analysis. In this case, the server for the ECG
machine is simply used as a transmission device for the data received from the ECG machine to the vendor’s
proprietary device.
Concept Map 1: Recording of ECG Digital Waveforms
The above concept map shows the process of capturing analog electrical potential differences, converting to digital
waveforms, and the automated measurement of intervals as well as morphological interpretation.
Page 19
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
446
447
448
449
450
451
452
453
454
This concept map displays the process for automated measurement of intervals, based on a single lead or a
superimposed median beat and RR, with potential human over-reading/adjustments. Relationships involved in
optional human over-read are shown using blue arrows. Furthermore, this concept map represents the server at the
central vendor (second machine). If the central vendor uses a proprietary hardware-software device to generate final
measurements to supply to the sponsor, this machine would simply receive the digital information and transmit it to
the proprietary third device.
455
2.3.3 ECG Device Types
Concept Map 2: Measurement of RR and Intervals for Use in Analysis
456
457
458
459
460
461
462
463
In a multisite study, there would be ECG equipment at each site. Even at a single site, there may be multiple ECG
machines used. However, in a QT study, the specifications of the ECG equipment would be stated in the protocol,
and all individual devices used would have the same characteristics. The SDTM Implementation Guide for Medical
Devices (SDTMIG-MD) allows a single sponsor identifier to be assigned to a group of devices that share the same
characteristics, and this is how ECG devices would be handled in QT studies.
464
2.3.3.1 Examples for Device Types
465
466
467
468
469
470
471
472
473
474
475
476
477
478
Example 1:
This example shows the types of devices used to obtain ECG data for a particular study. Although multiple ECG
machines were used, they were all of the same type, so identifiers were assigned to the type of machine, rather than
to individual devices. There were two separate devices used to obtain the ECG data: an ECG machine that recorded
data from the subject, and a second device that performed further processing of the ECG data from the first machine,
and also allowed manual over-reading of the ECG data. Because these two machines were used together to produce
the results in the ECG domain, the combination of the two devices is treated as a composite device, and the
combination is given its own unique sponsor ID. Data about these devices is represented in the Device Identifier
(DI) and Device Properties (DO) domains and in the proposed Device-Device Relationships dataset.
For the ECG machine, the device that collects data from the subject, and the device that does further processing and
allows manual over-reading, the characteristics of interest are manufacturer, model, and software/firmware version.
Row 1:
Rows 2-4:
Shows the sponsor device identifier for the composite ECG Device. This is the device whose
identifier appears in the EG dataset where subject data are represented. Further information about
what this device consisted of is provided in the Device-Device Relationships dataset, below.
Show the characteristics of the ECG machine, which has been given an SPDEVID value of "2.”
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 20
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
479
480
481
482
Rows 5-7:
Show the characteristics of the machine which performs further processing of the ECG data from the
first machine. This device has been given an SPDEVID value of “3.”
di.xpt
Row STUDYID DOMAIN SPDEVID DISEQ DIPARMCD DIPARM
DIVAL
ABC-123
DI
1
1
TYPE
Device Type Composite ECG Device
1
ABC-123
DI
2
1
TYPE
Device Type
ECG Machine
2
ABC-123
DI
2
2
MANUF
Manufacturer
Acme
3
ABC-123
DI
2
3
MODEL
Model
XYZ 2000
4
ABC-123
DI
3
1
TYPE
Device Type
ECG Analyzer
5
ABC-123
DI
3
2
MANUF
Manufacturer
Acme
6
ABC-123
DI
3
3
MODEL
Model
ECG Wizard
7
483
484
485
486
487
488
489
Row 1:
Row 2:
Shows the software/firmware version of the device with sponsor device identifier “2”. The
software/firmware version is not an intrinsic part of the device, so it is represented in the DO domain
rather than the DI domain.
Shows the software/firmware version of the device with sponsor device identifier “3”.
do.xpt
Row STUDYID DOMAIN SPDEVID DOSEQ DOTESTCD
DOTEST
DOORES
ABC-123
DO
2
4
SFTWRVER Software/Firmware Version
3.6
1
ABC-123
DO
3
5
SFTWRVER Software/Firmware Version
12B
2
490
491
492
493
494
495
496
497
498
499
500
501
The relationships between the two ECG machines and their combination is a simple one, which requires no
additional parameters to describe the relationship, so the permissible variables PARMCD, PARM, and VAL are not
included in the Device-Device Relationship dataset.
Row 1:
Row 2:
Row 3:
Shows that the device with SPDEVID = 1 has no parent and thus is at Level 1.
Shows that the device with SPDEVID = 2 is a part of the composite device with SPDEVID = 1.
Since the device with SPDEVID = 1 is at Level 1, this device is at Level 2.
Shows that the device with SPDEVID = 3 is also a part of the composite device with SPDEVID = 1
and thus is also at Level 2.
reldev.xpt
Row STUDYID SPDEVID PARENT LEVEL
ABC
1
1
1
ABC
2
1
2
2
ABC
3
1
2
3
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
2.3.4 How is the QT Interval Adjusted (Corrected) for Heart Rate?
There are three basic methods to adjust or correct the QT interval for HR. Essentially, all the methods attempt to
adjust the individual subject’s QT interval to a value that would be expected if the subject’s HR were constant.
The methods are:
1. Population-based formula derived from a historical population
2. Population-based formula derived from the population under study
3. Individual-based formula derived for each individual in the population under study
The first two methods are based on finding the mathematical formula (linear or nonlinear) that best describes the
relationship between the absolute QT interval and the RR interval from populations of multiple individuals and
applying the same formula to all ECGs for which a QTc is being computed.
1. Population-Based Formula from a Historical Population
In a historical population, this would be a group of normal, healthy volunteers, generally with 1 ECG from each
subject. Due to the normal variance between different populations, multiple researchers using different groups of
subjects have derived different formulas.
Page 21
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
520
521
The most well-known and clinically used is Bazett’s formula (QTcB), which was derived/developed in 1920 from
ECGs recorded in a small group of healthy subjects:
QTc =
522
523
√RR
Fridericia’s formula (QTcF) was also developed in 1920 from ECGs recorded in a small group of healthy subjects:
QTc =
524
525
526
527
528
529
530
531
532
533
QT
QT
3
√RR
It is reasonably well known that the Bazett formula under-corrects at faster heart rates (over 60 bpm) and conversely
over-corrects at slower heart rates. That is, at faster HRs (smaller RR intervals), the computed QTc is ‘larger than it
should be’ and at slower HRs (larger RR intervals), the computed QTc is ‘smaller than it should be’. When Bazett
corrected QTc is plotted against RR interval and a regression line is plotted, the slope is negative (Figure 9; with a
perfect correction, the slope of the regression line would be 0). In spite of this, Bazett’s formula is still the most
widely used for clinical correction of QT intervals. However, it is becoming more acceptable in regulatory
documents to use the Fridericia formula correction, without use of the Bazett formula, 2, 8 along with additional
correction results as described below.
Fridericia correction
Corrected QT interval (msec)
Corrected QT interval (msec)
Bazett correction
500
450
400
350
450
400
350
0.6
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
500
0.8
1.0
1.2
0.6
RR interval (sec)
0.8
1.0
1.2
RR interval (sec)
Figure 9: Relationship between the Bazett- and Fridericia-Corrected QT Interval and RR Interval
2. Population-Based Formula from the Population under Study
A population formula derived from the population under study uses off-treatment, baseline ECGs to construct a
population formula in a manner similar to construction of a formula from a historic control population.
Because the formula is based on the behavior of the individuals actually under study, such a study populationderived formula presumably accounts for variables (e.g., disease factors, age, and gender distribution) which might
influence the QT-RR relationship. Therefore, such a formula should be more accurate for the individuals under
study than one based on a historical population.
3. Individual-Based Formula (QTcI)
An individual-based QTc requires that a number of ECGs be obtained across a sufficient range of HRs (35 to 50
ECGs covering a range of heart rates of 50 to 80 beats per minute difference 9) for each individual under baseline
(nontreatment) conditions. One author has published data to support the position that at least 400 ECGs (QT-RR
pairs for each individual subject) are needed to compute an adequate individual correction and that there must
definitely be a range of heart rates corresponding to the heart rates that will be observed on the experimental drug 10.
These data are then used to compute a specific correction formula for each individual subject in a manner similar to
that used to compute a population correction. The rationale for such a method is based on the experimental findings
that the QT-RR relationship is different across subjects but relatively stable within subjects. In using QTcI, one subapproach is to use a single, predetermined mathematical model for all subjects. An alternative sub-approach is to fit
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 22
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
the individual subject’s data to several preselected mathematical models and use the best mathematical model for
each individual subject (model that results in flattest regression line after correction (QTcI vs. RR)) As such, this
QTc method is probably the best available, but is also very labor intensive and costly to use. For example, a study
with only 60 subjects would require, at a minimum, approximately 2100 baseline ECGs (60 subjects ×
35 ECGs/subject) be collected and measured to compute the 60 individual correction formulae. If 400 ECGs are
required per subject then 24,000 ECGs are required and this would require acquiring the ECGs from continuous
ECG recording. The 60 individual correction formulae would need to be derived in addition to collecting the actual
experimental ECGs. ECGs acquired for baseline values could be used for this purpose.
Some researchers have developed methods of assessing changes in ventricular repolarization based on the QT
interval that do not rely on an explicit correction of the QT interval for heart rate (the RR interval). One such method
is known as the beat-to-beat method 4. These methods are particularly important when the experimental drug results
in marked changes in autonomic nervous system tone and heart rate. These changes can be so large that it will be
difficult to obtain ECG data at heart rates that will be observed during treatment with the experimental drug, which
would raise concerns about the validity of any correction factor. Discussion of these alternatives beyond the
introduction of the concept is outside the scope of this document. These methods would generally rely on continuous
recording data.
Section 6.2 provides further details on QT correction from the dataset perspective.
3 The Thorough QT (TQT) Study
This section is based primarily upon the May 2005 ICH-E14 document 1, which describes the basic conduct,
purpose, and expected analyses of the TQT study.
The purpose of a TQT study is to evaluate the potential for an experimental drug to delay cardiac ventricular
repolarization, which it does through evaluation of changes in the heart rate corrected QT interval (QTc) during drug
treatment; and also to demonstrate that the study is capable of detecting differences of clinical significance
(approximately 5ms), so as to confirm that any lack of detected change is due to actual lack of change rather than
lack of assay sensitivity. These TQT studies are generally conducted in healthy volunteers highly screened for
normal cardiac electrical activity, for ease of precise measurement of the QT interval and to avoid additional
confounding factors.
The TQT study can be a crossover design when sufficient experimental drug exposure can be achieved with a single
dose or a few doses, or it can be a parallel design when a longer treatment period (several weeks or more) is required
to achieve sufficient experimental drug and/or metabolite exposure (i.e. experimental drug and/or metabolites with
long half-lives). If the study is a crossover design, the order of the treatments should be balanced across subjects,
such as with a Williams design 11. In general, the treatments are:
1. A dose of the experimental drug that is many times higher, if possible, than the intended maximum dose, in
order to account for drug-drug interactions and/or genetic metabolic enzyme deficiencies that might lead to
greater exposure to the experimental drug than otherwise intended with a given dose during routine clinical
use
2. Placebo
3. A positive control for purpose of demonstration of assay sensitivity (most often moxifloxacin, usually oral
but sometimes intravenous)
4. Optionally, a dose of the experimental drug that is within the intended therapeutic range (generally the
maximum intended therapeutic dose)
The administration of the active control has been allowed by regulators to be open-label, but the administrations of
the experimental drug dose(s) and placebo are double-blind, and ECG measurements and readings are performed by
persons completely blinded to associated treatments, subject details, and date/time of the ECG.
ECGs are usually collected at one or more baseline time points: immediately before experimental drug
administration, and/or on the day or days preceding day of first treatment administration at time points
corresponding to those at which ECGs will be collected following treatment. In crossover studies, these baseline
Page 23
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
ECGs are usually collected before each treatment period as opposed to using a single baseline for all treatment
periods. Post-treatment time points are determined by the pharmacokinetic properties of the treatment. The specific
time points at which ECGs would be acquired correspond to and surround the expected time of maximum
concentration of parent experimental drug and relevant metabolites.
ECGs are collected as a set of replicates (in close temporal proximity, e.g., 3 ECGs collected at 1-minute intervals)
of approximately 10 seconds in duration and utilizing all 12 leads. ECGs can be collected as conventional ECGs or
they can be extracted from a continuous high-fidelity ECG recording. Experimental drug and metabolite
concentrations are often assayed immediately after the time of ECG collection for PK/PD analysis, which can be a
useful secondary analysis pertinent to the potential influence of the experimental drug on ventricular repolarization.
An experimental drug is declared to have no influence on QTc if:
1. The comparison of the supratherapeutic dose of the experimental drug versus placebo is a noninferiority
comparison and the upper one-sided 95% confidence interval for the maximum difference (across the
several time-matched time points of ECG acquisition following treatment administration) between change
from baseline (there are multiple potential definitions of baselines discussed in a separate document) for
experimental drug compared to placebo is less than 10 ms (with no adjustment for the multiple
comparisons across the multiple time points) 1
AND
2. The positive control (e.g., moxifloxacin) versus placebo results in at least one difference from baseline
where the lower one-sided 95% confidence interval for the difference is greater than 5 ms (must adjust for
the multiple comparisons across time points) 8. Note that this explicit statistical definition requires not only
the ability to detect a difference between the positive control and placebo but also a certain minimum
difference, based on historical experience with moxifloxacin.
Without passing the test on assay sensitivity, the experimental drug is unlikely to be declared to have no effect,
regardless of the results of the first statistical co-primary test. Unfortunately, across multiple TQT studies, oral
moxifloxacin, while producing a detectable increase in QTc, sometimes fails to result in an increase of sufficient
magnitude such that the one-sided 95% confidence interval for the difference between moxifloxacin and placebo is
greater than 5 ms after adjusting for multiple comparisons. This lack of sufficient difference might occur for a
variety of reasons. As discussed below, PK/PD modeling of moxifloxacin concentration – change in QTc can serve
to support assay sensitivity when exposure to moxifloxacin (concentrations) are less than expected due to some
identified reason.
The ICH-E14 document has specified some secondary, conventional difference (not noninferiority) analyses of the
QTc. These are categorical (i.e., outlier, extreme observation) analyses (on the average for the replicate sets) based
on both absolute values and changes from baseline:
• Incidences of QTc >450 ms, experimental drug vs. placebo
• Incidences of QTc >480 ms, experimental drug vs. placebo
• Incidences of QTc >500 ms, experimental drug vs. placebo
• Incidences of increase from baseline >30 ms, experimental drug vs. placebo
• Incidences of increase from baseline >60 ms, experimental drug vs. placebo
Although not explicitly specified in detail in ICH-E14, both central tendency (i.e., mean change) as well as
categorical (i.e., outlier, extreme observation) analyses for experimental drug vs. placebo of the following are often
expected by regulators:
• Heart rate / RR interval
• PR interval
• QRS
• Incidence of treatment-emergent abnormal morphological (i.e., qualitative) findings: rhythm, axis,
conduction, hypertrophy, ischemia-injury-infarction, ST segment abnormalities, T wave abnormalities, U
wave abnormalities, etc.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 24
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
Some treatment-emergent adverse events which suggest the occurrence of delays in ventricular repolarization might
not be well accounted for by the standard, objective study analyses. Additional analyses of such adverse events are
also important. Such events are likely to be of low frequency in a healthy population.
In addition, as noted above, the time course of change in QTc displayed by the positive control is of interest. It is
expected to rise to a maximum associated with time of maximum concentration and then decline to baseline
consistent with kinetics of elimination. If the observed time course of change in QTc does not parallel change in
concentration, assay sensitivity can be questioned.
Sponsors will often model changes in QTc (difference between drug and placebo in changes in QTc) versus changes
in concentration for the experimental drug, its metabolites, and separately moxifloxacin with linear regression or
other appropriate model. If the experimental drug is associated with increases in QTc, this modeling can be helpful
in defining the expected magnitude of such an increase at maximum expected exposure. If the positive control
results in less increase in QTc than expected, this modeling can be helpful in explaining the results and potentially
demonstrating that the positive control behaved similarly to historical expectations when exposure (maximum
concentrations) was less than expected based on historical experience. Detailed advice on the specifics of these
PK/PD analyses are beyond the scope of this introductory description of TQT study basics.
When an experimental drug causes substantial changes in heart rate, the TQT study paradigm described above may
not be appropriate because it may be impossible with conventional methods described above to adequately correct
the QT interval for heart rate. In such situations, alternative study and/or analytical methodologies may be necessary
4
. Similarly, when the experimental drug is potentially toxic and/or has a narrow therapeutic range, special study
adaptations may be necessary.
Page 25
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
685
4 Trial Design
686
4.1 TQT Study Design
687
688
689
690
691
692
693
694
695
696
As mentioned in Section 2.1, QT interval and QTc might be an outcome of interest evaluated in a number of other types of studies such as Phase 1 single
ascending dose (SAD) and multiple ascending dose (MAD) studies, Phase 3 and Phase 4 studies, and TQT studies with modifications dictated by toxicity
considerations relevant to the test drug. In general, these other study types would be of the parallel or crossover design but may have differences in study
elements such as study population, inclusion/exclusion criteria, and specific treatments included.
697
4.1.1 Parallel Studies
698
699
700
701
702
703
704
705
706
707
708
The example study designs presented below illustrate specific TQT study designs. In a TQT study, QT interval and QTc are the primary outcomes of interest (see
Section 3). A typical TQT study is designed as double-blind (partial double-blind as in some cases the investigator might not be blind to administration of the
active control), placebo- and positive-controlled to determine whether the test treatment fails to prolong QTc and QT intervals (primary statistical test is
noninferiority) and to demonstrate the assay sensitivity of the positive control treatment in the study population. Traditional TQT studies employ parallel or
crossover designs and are often designed with equal study duration and sample size for the different treatment arms or periods 12.
Under certain circumstances (related to the PK characteristics of the test drug), a parallel design may be preferred for a TQT study. Such circumstances include 1:
• Drugs with long elimination half-lives for which lengthy time intervals would be required to achieve steady-state and complete washout.
• If carryover effects are prominent for other reasons, such as irreversible receptor binding or long-lived active metabolites.
• If multiple doses of the investigational drug are required to evaluate the effect on QT/QTc.
Example TQT Study 1
Below is the study schema diagram for Example TQT Study 1, a parallel trial. This trial has 4 treatment arms (placebo, positive control, therapeutic study drug
dose, and supratherapeutic study drug dose), which correspond to the 4 possible left-to-right "paths" through the trial. Moxifloxacin has become the standard
positive control with a well characterized (peak effect and time course), expected influence on QTc in healthy subjects with a mean increase in QTc in the range
of 10 – 15 msec. Other positive control compounds are possible (e.g., low dose ibutalide).
Note: Moxifloxacin is one
example of a positive control.
Note: This is an optional arm.
709
710
711
Randomization
Figure 10: Parallel Study Design Schema for Example TQT Study 1
T = Therapeutic Dose (DRUG A 1 MG), ST = Supratherapeutic Dose (DRUG A 100 MG)
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 26
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
712
713
714
715
716
717
718
719
Below is an example of a trial design matrix for Example TQT Study 1. A trial design matrix is a table with a row for each treatment arm in the trial and a
column for each epoch in the trial. The cells in the matrix represent the study cells, which are populated with trial elements. In this trial, each study cell contains
exactly one element. Note that randomization is not represented in the trial design matrix.
Trial Design Matrix for Example TQT Study 1
Assume the study’s therapeutic and supratherapeutic doses for Drug A are 1 mg and 100 mg, respectively.
Placebo
Moxifloxacin
Drug A 1 MG
Drug A 100 MG
720
721
722
723
724
725
726
727
728
729
730
731
Screen
Screen
Screen
Screen
Screen
Treatment
PLACEBO
MOXIFLOXACIN
DRUG A 1 MG
DRUG A 100 MG
Below is an example of a trial arms dataset for Example TQT Study 1. For this example, the conversion of the trial design matrix into the trial arms dataset is
straightforward. For each cell of the matrix, there is a record in the trial arms dataset. ARM, EPOCH, and ELEMENT can be populated directly from the matrix.
TAETORD acts as a sequence number for the elements within a treatment arm, so it can be populated by counting across the cells in the matrix. The
randomization information, which is not represented in the trial design matrix, is held in TABRANCH in the trial arms dataset. TABRANCH is populated only if
there is a branch at the end of an element for the treatment arm. When TABRANCH is populated, it describes how the decision at the branch point would result
in a subject being in this treatment arm.
Trial Arms Dataset for Example TQT Study 1
Assume the study’s therapeutic and supratherapeutic doses for Drug A are 1 mg and 100 mg, respectively.
ta.xpt
Row STUDYID DOMAIN ARMCD
ARM
TAETORD ETCD
ELEMENT
TABRANCH
TATRANS EPOCH
EX1
TA
PBO
Placebo
1
SCRN
Screen
Randomized to Placebo
Screen
1
EX1
TA
PBO
Placebo
2
PBO
Placebo
Treatment
2
EX1
TA
MOXI
Moxifloxacin
1
SCRN
Screen
Randomized to Moxifloxacin
Screen
3
EX1
TA
MOXI
Moxifloxacin
2
MOXI
Moxifloxacin
Treatment
4
EX1
TA
D1MG
Drug A 1mg
1
SCRN
Screen
Randomized to Drug A 1 mg
Screen
5
EX1
TA
D1MG
Drug A 1mg
2
DT
Drug Therapeutic Dose
Treatment
6
EX1
TA
D100MG Drug A 100mg
1
SCRN
Screen
Randomized to Drug A 100 mg
Screen
7
EX1
TA
D100MG Drug A 100 mg
2
DST Drug Supratherapeutic Dose
Treatment
8
732
733
4.1.2 Crossover Studies
734
735
736
In comparison to parallel group studies, crossover studies have at least two potential advantages 1:
• A smaller number of subjects are typically required. Subjects serve as their own controls, resulting in reduced variability of differences related to intersubject variability.
Page 27
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
•
Heart rate correction approaches based on individual subject data may be more feasible. Since baseline ECGs are collected before each treatment
period, more ECGs are available for each subject for computation.
Example TQT Study 2
This is an example of a Williams Design used for a typical crossover study.
The two most common ways this design could be represented are shown in Figure 11 and Figure 12. How the design is described in the protocol and how the
sponsor intends to analyze the data will determine which representation a sponsor chooses. A washout period may be “built into” a treatment Element (Figure 11)
if observations occurring during the washout portion will be analyzed as part of the treatment. This representation assumes there is a sufficient period at the end
of each Element for the drug to be eliminated from the subjects before the next drug is administered. If data from the time periods where dosing does not occur is
not intended to be associated with the previous treatment, then separate washout Elements would be created (Figure 12).
Randomization
Figure 11: Crossover Study Design for Example TQT Study 2 with Washout Period Combined with each Treatment Element
T = Therapeutic Dose (DRUG A 1 mg), ST = Supratherapeutic Dose (DRUG A 100 mg)
Randomization
Figure 12: Crossover Study Design for Example TQT Study 2 with Separate Elements for each Washout Period
T = Therapeutic Dose (DRUG A 1 mg), ST = Supratherapeutic Dose (DRUG A 100 mg), WO = Washout
The trial design matrix for the crossover example trial is shown below.
Trial Design Matrix for Example TQT Study 2
Assume the study’s therapeutic and supratherapeutic doses for Drug A are 1 mg and 100 mg, respectively.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 28
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Treatment Sequence Screen First Treatment Washout Second Treatment Washout Third Treatment Washout Fourth Treatment
PBO-D100-MOXI-D1
MOXI-PBO-D1-D100
D1-MOXI-D100-PBO
D100-D1-PBO-MOXI
762
763
764
765
766
767
Screen
Screen
Screen
Screen
Placebo
Moxifloxacin
Drug A 1 mg
Drug A 100 mg
Washout
Washout
Washout
Washout
Drug A 100mg
Placebo
Moxifloxacin
Drug A 1mg
Washout
Washout
Washout
Washout
Moxifloxacin
Drug A 1mg
Drug A 100 mg
Placebo
Drug A 1mg
Drug A 100mg
Placebo
Moxifloxacin
Below is an example of a trial arms dataset for Example TQT Study 2. For this example, it is straightforward to produce the trial arms dataset for this crossover
trial from the diagram showing treatment arms and epochs or from the trial design matrix.
Trial Arms Dataset for Example TQT Study 2
ta.xpt
Row STUDYID DOMAIN
1
EX2
TA
2
EX2
TA
3
EX2
TA
4
EX2
TA
5
EX2
TA
6
EX2
TA
7
EX2
TA
8
EX2
TA
9
EX2
TA
10
EX2
TA
…
…
…
ARMCD
PBO-D100MGMOXI-D1MG
PBO-D100MGMOXI-D1MG
PBO-D100MGMOXI-D1MG
PBO-D100MGMOXI-D1MG
PBO-D100MGMOXI-D1MG
MOXI-PBO-D1MG
–D100MG
MOXI-PBO-D1MG–
D100MG
MOXI-PBO-D1MG–
D100MG
MOXI-PBO-D1MG–
D100MG
MOXI-PBO-D1MG–
D100MG
…
ARM
Placebo-Drug A 100MGMoxifloxacin-Drug A 1MG
Placebo-Drug A 100MGMoxifloxacin-Drug A 1MG
Placebo-Drug A 100MGMoxifloxacin-Drug A 1MG
Placebo-Drug A 100MGMoxifloxacin-Drug A 1MG
Placebo-Drug A 100MGMoxifloxacin-Drug A 1MG
Moxifloxacin-Placebo-Drug A
1MG-Drug A 100MG
Moxifloxacin-Placebo-Drug A
1MG-Drug A 100MG
Moxifloxacin-Placebo-Drug A
1MG-Drug A 100MG
Moxifloxacin-Placebo-Drug A
1MG-Drug A 100MG
Moxifloxacin-Placebo-Drug A
1MG-Drug A 100MG
…
TAETORD
ETCD
ELEMENT
1
SCRN
Screen
2
PBO
Placebo
3
DA100MG
Drug A
100mg
4
MOXI
Moxifloxacin
5
DA1MG
Drug A 1mg
1
SCRN
Screen
2
MOXI
Moxifloxacin
3
PBO
Placebo
4
DA1MG
Drug A 1 mg
5
DA100MG
…
…
Drug A
100mg
…
768
Note: The above could be modeled using separate elements for each washout period and follow-up
769
4.1.3 Trial Elements
770
771
772
773
774
775
776
Washout
Washout
Washout
Washout
TABRANCH
TATRANS
Randomized to Placebo-Drug A 100mgMoxifloxacin-Drug A 1mg
EPOCH
Screen
First
Treatment
Second
Treatment
Third
Treatment
Fourth
Treatment
Randomized to Moxifloxacin-PlaceboDrug A 1mg-Drug A 100mg
…
Screen
…
First
Treatment
Second
Treatment
Third
Treatment
Fourth
Treatment
…
The Trial Elements (TE) dataset contains the definitions of the elements that appear in the Trial Arms (TA) dataset. An element may appear multiple times in the
Trial Arms table because it appears either 1) in multiple arms, 2) multiple times within an arm, or 3) both. However, an element will appear only once in the Trial
Elements table. Each row in the TE dataset may be thought of as representing a "unique element" in the sense of "unique" used when a case report form template
page for a collecting certain type of data is often referred to as "unique page."
An element is a building block for creating study cells and an arm is composed of study cells. Or, from another point of view, an arm is composed of elements
(i.e. the trial design assigns subjects to arms, which are comprised of a sequence of steps called elements). Trial elements represent an interval of time that serves
Page 29
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
777
778
779
780
781
782
783
784
785
786
787
788
a purpose in the trial and are associated with certain activities affecting the subject. A valid element has a name that describes the purpose of the element and
includes a description of the activity or event that marks the subject's transition into the element as well as the conditions for leaving the element.
Below are the TE datasets for Example TQT Study 1 and Example TQT Study 2, respectively. These trials are assumed to have fixed-duration Elements. The
wording in TESTRL is intended to separate the description of the event that starts the Element into the part that would be visible to a blinded participant in the
trial (e.g., "First dose of a treatment Epoch") from the part that is revealed when the study is unblinded (e.g., "where dose is 5 mg"). Care must be taken in
choosing these descriptions to be sure that they are "Arm and Epoch neutral." For instance, in a crossover trial such as Example TQT Study 2, where an Element
may appear in one of multiple Epochs, the wording must be appropriate for all the possible Epochs. The wording for Example TQT Study 2 uses the wording "a
treatment Epoch."
Trial Elements Dataset for Example TQT Study 1
te.xpt
Row STUDYID DOMAIN ETCD ELEMENT
TESTRL
EX1
TE
SCRN
Screen
Informed consent
1
EX1
TE
PBO
Placebo
First dose of study drug, where drug is placebo
2
EX1
TE
DST Drug A 100mg First dose of study drug, where drug is Drug A 100mg
3
EX1
TE
MOXI Moxifloxacin First dose of study drug, where drug is Moxifloxacin
4
EX1
TE
DT
Drug A 1 mg
First dose of study drug, where drug is Drug A 1mg
5
789
790
791
TEENRL
TEDUR
3 weeks after start of element P21D
2 weeks after start of element P14D
2 weeks after start of element P14D
2 weeks after start of element P14D
2 weeks after start of element P14D
Trial Elements Dataset for Example TQT Study 2
te.xpt
Row STUDYID DOMAIN
ETCD
ELEMENT
TESTRL
EX2
TE
SCRN
Screen
Informed consent
1
EX2
TE
PBO
Placebo
First dose of treatment epoch, where drug is placebo
2
EX2
TE
DA100MG Drug A 100mg First dose treatment epoch, where drug is Drug A 100mg
3
EX2
TE
MOXI
Moxifloxacin First dose of treatment epoch, where drug is Moxifloxacin
4
EX2
TE
DA1MG
Drug A 1mg First dose of treatment epoch, where drug is Drug A 1 mg
5
792
793
794
795
796
TEENRL
TEDUR
3 weeks after start of element P21D
2 weeks after start of element P14D
2 weeks after start of element P14D
2 weeks after start of element P14D
2 weeks after start of element P14D
4.1.4 Trial Summary Parameters
This section shows a partial trial summary dataset for Example TQT Study 1 described in Section 4.1.3, with Drug T (1 mg) and Drug ST (100 mg) that may be
applicable to TQT studies. Note that the example includes data for proposed terminology for new trial summary parameters in rows 44-50.
ts.xpt
Row STUDYID DOMAIN TSSEQ TSGRPID TSPARMCD
TSPARM
1
EX1
TS
1
TITLE
Trial Title
2
3
4
5
6
7
8
EX1
EX1
EX1
EX1
EX1
EX1
EX1
TS
TS
TS
TS
TS
TS
TS
1
1
1
1
1
2
1
SPONSOR
TPHASE
RANDOM
TBLIND
TCNTRL
TCNTRL
DESIGN
Sponsoring Organization
Trial Phase Classification
Trial is Randomized
Trial Blinding Schema
Control Type
Control Type
Description of Trial Design
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
TSVAL
A Randomized, Double-Blind, Placebo- and Positive-Controlled, 4-arm Parallel-Group Study
to Evaluate the Effect of Drug A on Cardiac Repolarization in Healthy Adult Male Subjects
ACME Pharmaceutical Company
Phase I Trial
Y
DOUBLE BLIND
ACTIVE
PLACEBO
PARALLEL
...
...
...
...
...
...
...
...
...
Page 30
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row STUDYID DOMAIN TSSEQ TSGRPID TSPARMCD
9
EX1
TS
1
OBJPRIM
10
EX1
TS
1
OBJSEC
11
EX1
TS
2
OBJSEC
12
EX1
TS
3
OBJSEC
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
EX1
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
TS
1
2
3
1
1
1
1
1
1
1
2
1
2
1
2
3
4
1
2
3
4
1
2
3
4
1
2
3
4
1
1
1
1
1
1
1
1
1
TTYPE
TTYPE
TTYPE
TDIGRP
AGEMIN
AGEMAX
AGEU
SEXPOP
ADDON
TRT
TRT
COMPTRT
COMPTRT
DOSE
DOSE
DOSE
DOSE
DOSU
DOSU
DOSU
DOSU
DOSFRQ
DOSFRQ
DOSFRQ
DOSFRQ
ROUTE
ROUTE
ROUTE
ROUTE
LENGTH
PLANSUB
EGCTMON
EGRDMETH
EGBLIND
EGTWVALG
EGLEADPR
EGREPLTR
EGREPLBL
Page 31
Draft
A-1
A-100
MOXI
PBO
A-1
A-100
MOXI
PBO
A-1
A-100
MOXI
PBO
A-1
A-100
MOXI
PBO
A-1
A-100
MOXI
PBO
TSPARM
TSVAL
The primary objective of this study is to evaluate the effects of therapeutic and
Trial Primary Objective
supratherapeutic doses of Drug relative to placebo on cardiac repolarization following
multiple oral doses in healthy adult male subjects
To characterize the pharmacokinetic profiles of 1 mg and 100 mg doses of Drug A and its
Trial Secondary Objective
metabolites in healthy adult male subjects
To evaluate the safety and tolerability of 1 mg and 100 mg doses of Drug A in healthy adult
Trial Secondary Objective
male subjects
To establish assay sensitivity to the response in QT/QTc following a single dose of 400 mg
Trial Secondary Objective
moxifloxacin in healthy adult male subjects
Trial Type
SAFETY
Trial Type
PHARMACOKINETIC
Trial Type
THOROUGH QT
Diagnosis Group
HEALTHY SUBJECTS
Planned Minimum Age of Subjects
18
Planned Maximum Age of Subjects
45
Age Unit
YEARS
Sex of Participants
M
Added on to Existing Treatments
N
Reported Name of Test Product
Drug A - 1 mg
Reported Name of Test Product
Drug A - 100 mg
Comparative Treatment Name
moxifloxacin
Comparative Treatment Name
placebo
Dose per Administration
1
Dose per Administration
100
Dose per Administration
400
Dose per Administration
0
Test Product Dose Units
mg
Test Product Dose Units
mg
Test Product Dose Units
mg
Test Product Dose Units
mg
Test Product Dosing Frequency
QD
Test Product Dosing Frequency
QD
Test Product Dosing Frequency
ONCE
Test Product Dosing Frequency
QD
Route of Administration
ORAL
Route of Administration
ORAL
Route of Administration
ORAL
Route of Administration
ORAL
Trial Length
P56D
Planned Number of Subjects
40
ECG Continuous Monitoring
Y
ECG Read Method
SEMI-AUTOMATIC
ECG Reading Blinded
Y
ECG Twave Algorithm
TANGENT METHOD
ECG Planned Primary Lead
LEAD II
ECG Replicates On-Treatment
Y
ECG Replicates at Baseline
Y
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
...
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
797
798
799
Proposed terminology for Trial Summary Parameter Test Codes and Trial Summary Parameter Test Names is given below:.
CDISC Submission Value
TSPARMCD
EGCTMON
EGRDMETH
EGBLIND
EGTWVALG
EGLEADPR
EGREPLTR
EGREPLBL
EGLEADSM
CDISC Submission Value TSPARM
ECG Continuous Monitoring
ECG Read Method
ECG Reading Blinded
ECG Twave Algorithm
ECG Planned Primary Lead
ECG Replicates On-Treatment
ECG Replicates at Baseline
ECG Used Same Lead
CDISC Definition
Indicate whether the ECG measurements are extracted from continuous monitoring.
The ECG read method, e.g. manual, semi-automatic, automatic.
Indicate whether the ECG readers are blinded to subject details, treatment, day, and time.
Algorithm used to determine the end of Twave.
The ECG lead intended to be used as the primary lead.
Indicate whether replicate ECGs were collected on-treatment.
Indicate whether replicate ECGs were collected at baseline.
Indicate whether the same lead was used for interval measurements for all ECGs used in the
analyses.
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
4.2 Time Point Planning
In clinical studies, the timing and collection of ECGs are guided by the known properties of the drug and its metabolites. ECGs are recorded at baseline *, on the
day on which steady-state or other intended concentration of the experimental drug is reached (and the same day relative to baseline for placebo), and
periodically thereafter. It has been recommended that 13:
• In order to study the peak effect on QTc and the QT interval, ECGs should be recorded several times shortly before and around the time of the
maximum drug concentration (Tmax).
• To evaluate any delayed effects of the drug or its metabolites on cardiac repolarization, ECG recording should continue even after Tmax.
• To demonstrate assay sensitivity, ECGs should also be recorded close to the Tmax of the positive control.
• To ensure that blinding is maintained, ECGs should be recorded on the same days and at the same time points in all treatment groups.
The diagrams below show how ECG data are organized within 10-second ECGs, and how those 10-second ECGs are organized within and across time points
(Figure 13 and Figure 14). Although analysis methods that use all the data from continuous monitoring over a long period of time (e.g., 24 hours) are being
developed, analysis usually assumes that data are organized by time points. ECGs are recorded in replicates (usually three or more), 30-120 seconds apart, to
account for inherent variability. Each replicate lasts 10 seconds. Replicates may be individual recordings or extracted from a continuous recording (typically
lasting 24 hours or a multiple thereof). The ECG record is digitally stored for later processing.
*
Several alternatives for baseline exist, such as: the day before treatment is first administered, or just a 1- or 2-hour period before treatment is first administered.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 32
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
24 hours
Beat 1 (7:59:00.00)
817
818
819
820
821
Beat 2 (7:59:01.30)
Beat 3 (7:59:02.15)
…
Beat 12 (7:59:10.05)
…
Beat 100,800
Extracted 10-second ECG
≈ 12 P-QRS-T complexes
Figure 13: Individual Beats and Relevant Identifying Information Within a 24-hour Continuous Recording
Each cycle, in a normal ECG obtained from a healthy person, consists of a P-QRS-T complex and the subsequent isoelectric activity before the next P-QRS-T
complex. The pertinent information is the sequence number and the time.
Page 33
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
822
823
824
825
826
827
828
829
830
831
832
833
Figure 14: ECG Replicates and Nominal Time Points
The concept map below displays the process and timing of potential parameters (such as vital signs and PK) and their order of collection in a QT study. In a
typical QT study, ECGs are recorded at baseline on the day of treatment (day 1) and at the same pre-defined clock times (time matched) within each of the
treatment groups. A stable heart rate is typically obtained via ECGs taken after subjects have rested (but not while sleeping) for at least 5 to 10 minutes in the
supine position.
Vital signs such as blood pressure and heart rate in the supine and/or standing positions can be collected at the same time points as the ECG recordings,
immediately after collecting the ECGs. However, the assessment of orthostatic challenge is optional and dependent on the study objectives. To investigate
potential exposure-response relationships and confirm the availability of the drug, blood samples for PK measurements can be collected at the same time points
as the ECG recordings. Blood draws are performed after the ECG recordings to avoid confounding stress.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 34
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
834
835
836
837
838
This concept map displays the process and timing of a collection of parameters if used in a QT study. Note that multiple ECG replicates are recorded (the number
can vary but is generally 3 or more), 30-120 seconds apart, to account for inherent variability; each recording lasting 10 seconds.
839
5 Subject Characteristics and Eligibility
840
5.1 Inclusion/Exclusion Criteria
841
842
843
844
845
Typically, TQT studies are conducted in a healthy population unless there are safety and tolerability concerns with administration of the drug to healthy subjects.
Inclusion of healthy subjects minimizes confounding factors of underlying disease, co-morbidities and/or concomitant medications. Thus, standard
inclusion/exclusion criteria for healthy subject studies are often applied in TQT studies. TQT studies are designed to protect subject safety. Furthermore,
consideration may be given to excluding individuals who have a risk for Torsades de pointes (e.g., cardiac disease, hypokalemia, personal history of unexplained
syncopal episodes, or known personal or family history of Long QT Syndrome) and individuals who have repeatedly high QTcF values at baseline. TQT studies
Concept Map 3: Potential Parameters and Their Order of Collection
Page 35
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
846
847
848
849
850
851
are also designed to guard against falsely concluding a lack of effect on QT/QTc and to limit all potential sources of QT/QTc variability such as concomitant
medications or other substances that may prolong QT intervals.
Note that general inclusion and exclusion criteria are also addressed in other CDISC standards. The table below provides example exclusion criteria for studies
evaluating QT. Data collection for these data is already well documented in the current SDTM IG using the IE domain.
Protect
Patient
Safety
Exclusion Criteria
A known or suspected hypersensitivity to
1
study drug, moxifloxacin or any
components of the formulations used.
Subject has any clinically significant
abnormality following the investigator’s
review of the physical examination, and
protocol-defined clinical laboratory tests at
Screening or day -1.
2
a. For the following laboratory analytes, any
result outside reference range, even if NOT
judged clinically significant is reason for
exclusion: sodium, potassium, chloride,
calcium, magnesium
Any prescription medication, over-thecounter medication, herbal medication or
product (except for vitamins supplements
3
without herbal additives and
acetaminophen) within 2 weeks of first
treatment or during the study.
Any use of recreational drugs or a positive
4
drug screen.
Excessive methylxanthine (caffeine, etc.)
5
use, exceeding _____ per day
Smoking >____ cigarette equivalents per
6
day.
Alcohol intake >____ U per day (U = 1.5oz
7
of ≥80 proof spirit equivalent).
Methylxanthine, nicotine, alcohol, or
8
grapefruit juice consumption within 2
weeks of any treatment.
Resting, basal systolic BP ≥140 mmHg or
9
diastolic BP ≥90 mmHg.
10 Orthostatic hypotension or postural
Guard Against Falsely Concluding No Effect and
Eliminate Confounding Influence (confounds
increase within and between subject variance)
Additional Details
X
X
X
X
X
X
X
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
X
X
Examination and lab abnormalities can suggest poor
health that predispose to safety concerns when QTc is
prolonged. This is especially the case for electrolytes.
Alterations in electrolytes can alter the normal
configuration of the T wave.
Substances can both alter metabolism of test drug and
thereby alter any potential influence, and directly alter
the QTc itself and alter autonomic tone that alters T
wave response to change in heart rate and change heart
rate, complicating analyses through correction.
Substances may alter the metabolism of test drug, the
autonomic tone and/or alter heart rate.
X
Alcohol intake may alter the autonomic tone and/or
alter heart rate.
X
Substances may alter the metabolism of test drug, the
autonomic tone and/or alter heart rate.
X
These factors suggest a disturbed autonomic tone.
X
Page 36
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
11
12
Orthostatic Tachycardia
Personal or family history of: syncope;
ventricular tachydysrhythmia; cardiac arrest
/ sudden cardiac death; congenital long QT
syndrome.
Specific ECG exclusionary criteria:
a. Heart rate <40 or >99 bpm
b. PR <120 or >200 ms
c. QRS >110 ms
d. QTcF males >430 ms, females >450 ms
e. Any qualitative / morphological
abnormality except: sinus arrhythmia;
isolated premature atrial complexes /
premature ventricular complexes
f. T-wave / U-wave characteristics making
determination of the end of the T-wave
difficult such as biphasic T-waves; U-waves
of width greater than 1/3 the width of the
preceding T-wave
Protect
Patient
Safety
Guard Against Falsely Concluding No Effect and
Eliminate Confounding Influence (confounds
increase within and between subject variance)
X
X
X
(d only)
X (all)
Additional Details
Suggests a lack of safety margin if QTc prolonged and
potential for prolongation due to factors other than drug
influence
HR: difficulties in correction - confound; PR & QRS:
lack of good normal cardiac electrophysiology - some
mild safety concern and confound; QTc: safety concern
if further prolonged and confound; Other
abnormalities: lack of good normal heart - confound; T
wave: abnormality - confound and difficulty in
measuring T wave
852
853
854
855
In general, a minimum age of 18 years old is an entry criterion, and in order to promote homogeneity in the experimental population, some maximum age
criterion is likely to be in place, which can be variable given enrolment considerations. The list above does not cover all potential exclusion criteria that sponsors
might select.
856
5.2 Pharmacogenetics
857
858
859
860
861
862
863
864
865
866
867
868
Several genetic factors that alter the risk of drug-induced QT prolongation have been reported. Some of these genetic factors are associated with Long QT
Syndrome. Many forms of Long QT Syndrome have been linked to mutations in genes encoding cardiac ion channel proteins. Polymorphisms can affect ion
channels, leading to an increased sensitivity to drugs that affect repolarization. Because of incomplete penetrance, not all carriers of non-wildtype alleles of ion
channel genes will manifest QT/QTc interval prolongation in screening ECG evaluations.
Other genetic factors may affect drug metabolism genes (mainly CYP2D6 and CYP3A4) 14. Poor metabolizers accumulate excess concentrations of drugs. If the
drug blocks relevant cardiac ion channels, this poses a risk of QT interval prolongation. This would be particularly exacerbated when combined with
polymorphisms associated with Long QT Syndrome.
Thus, the presence of either or both types of genetic variations may predispose a subject to a potentially fatal adverse reaction, even at therapeutic drug doses 15.
The ICH E14 2005 guidance on QT prolongation advises considering genotyping patients who experience marked prolongation of QT/QTc or TdP while on drug
therapy 1.
Page 37
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
869
870
871
872
873
874
875
876
877
878
879
The pharmacogenomics/genetics (PGx) team, a sub-team within the CDISC SDS Team (Clinical Data Interchange Standards Consortium-Study Data
Submission), has developed several domains designed to carry PGx data. The PGx findings domain stores key results such as intensity values (both raw and
normalized), p-value, fold-change, ratio, nucleotide variation, amino acid variation, etc. The initial development of these CDISC PGx domains was done in
parallel with the work being done by the Health Level Seven Clinical Genomics Work Group. Please refer to the SDTM Implementation Guide for
Pharmacogenomics/Genetics (STDMIG-PGx)† for these domains.
880
6 Study Assessments
881
6.1 ECG Assessments
882
883
884
885
886
887
Genetic testing may occur prospectively in QT studies to exclude subjects predisposed to genetic Long QT Syndrome or Short QT Syndrome, or it may be
executed post hoc in a case of unusual results. Additionally, prospective genetic testing may be performed either to exclude slow or rapid metabolizers or to
include only slow metabolizers. Note that the inclusion of slow metabolizers is a potential option to achieve increased drug exposure with lower doses in lieu of
using higher doses greater than maximum clinical dose or administering an inhibitor.
Quantitative ECG data (e.g., HR, PR interval, QRS, absolute QT interval, and QTc) represent one interest of any ECG data analysis. Qualitative ECG data are
descriptive findings based on visual interpretation of the waveforms, sometimes referred to as morphological findings. These data include electrical axis ‡, heart
rhythm, conduction, evidence of ischemia, ST segment and T-wave and U-wave shapes. These data also provide extremely important data for analysis (Concept
Map 4). It is important to note that nearly all abnormal, outlier quantitative data are reflected in the qualitative findings. For example, a PR interval greater than
the upper reference limit of 220 msec would lead to the conduction qualitative finding of 1st-degree AV block.
†
Currently in development as of the publication of this document. The Virology Therapeutic Area Data Standards User Guide (VR-UG, available at:
http://www.cdisc.org/stuff/contentmgr/files/0/2356ae38ac190ab8ca4ae0b222392b37/misc/vr_ug_v1_0_prov.pdf) has prototype versions of the domains.
‡
Also measured as a quantitative finding.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 38
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
888
889
890
891
892
Concept Map 4: ECG Quantitative Results and Morphological (Qualitative) Findings Determination
This concept map displays the process for determining the measurements of the PR, QT, RR intervals and QRS complex as well as determining any abnormal
qualitative findings on the ECG.
Page 39
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
6.1.1 Specification for ECG Test Results
The ECG Test Results (EG) domain model is a fully approved and formally published domain model. The table below was modified from the current standard as
defined in the SDTMIG 3.2 to extend the record structure to cover beat-to-beat records as well as conventional 10-second ECGs, and to add two pre-existing
variables (SPDEVID and EPOCH) and two new variables (EGREPNUM and EGBEATNO §) to the specification.
SPDEVID and EPOCH are identifier and timing variables, respectively, which may be added to any domain model. EGBEATNO is used to differentiate between
beats in beat-to-beat records, and EGREPNUM is used to differentiate between multiple repetitions of a test within a given time frame. Because they are new,
they have been highlighted in yellow.
An additional column, BRIDG Mapping, has also been added to the table below. This column contains the BRIDG class and attribute associated with each
variable. Sometimes no mapping exists for a particular variable, in which case a brief comment as to why is given in italics.
For non-individual ECG beat data, and for aggregate ECG parameter results (e.g., QT interval, RR, PR, QRS), EGREFID is populated for all unique ECGs, so
that submitted SDTM data can be matched to the actual ECGs stored in the ECG warehouse. Therefore, this variable is expected for these types of records.
For individual-beat parameter results, waveform data will not be stored in the warehouse, so there will be no associated identifier for these beats.
Note that EGREFID holds the identifier or accession number for the actual ECG stored in the warehouse, rather than its file name and path.
Controlled terminology is used where it exists; however, new terminology such as that for beat-to-beat data has been proposed. Please refer to the current CDISC
Controlled Terminology (http://www.cancer.gov/cancertopics/cancerlibrary/terminologyresources/cdisc).
eg.xpt, ECG Test Results — Findings. One record per ECG observation per replicate per time point or one record per ECG observation per beat per
visit per subject, Tabulation.
Variable Name Variable Label
STUDYID
DOMAIN
USUBJID
SPDEVID
EGSEQ
EGGRPID
§
Study Identifier
Domain
Abbreviation
Unique Subject
Identifier
Sponsor Device
Identifier
Sequence
Number
Group ID
DocumentIdentifier.identifier
PerformedObservation
Controlled
Type Terms, Codelist
Role
or Format
Char
Identifier
Char EG
Identifier
SubjectIdentifier.identifier
Char
Identifier
MaterialIdentifier.identifier
Char
Identifier
implementation specific record
identifier
implementation specific record
grouping variable
Num
Identifier
Char
Identifier
BRIDG Mapping
CDISC Notes
Unique identifier for a study.
Two-character abbreviation for the domain.
Core
Req
Req
Identifier used to uniquely identify a subject across all studies Req
for all applications or submissions involving the product.
Sponsor-defined identifier for a device
Perm
Sequence Number given to ensure uniqueness of subject
records within a domain. May be any valid number.
Used to tie together a block of related records in a single
domain for a subject.
Req
Perm
Details for the new variables are in accordance with SDTM Governance via the SDTM Variable Requests tracking tool as of July 15, 2014, but may be subject to change prior to formal publication.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 40
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Variable Name Variable Label
EGREFID
EGSPID
BRIDG Mapping
ECG Reference implementation specific
ID
Sponsor-Defined implementation specific
Identifier
Controlled
Type Terms, Codelist
Role
or Format
Char
Identifier
Char
Identifier
EGTESTCD
ECG Test or
Examination
Short Name
DefinedObservation.nameCode
Char (EGTESTCD) ! Topic
EGTEST
ECG Test or
Examination
Name
DefinedObservation.nameCode
Char (EGTEST) !
Synonym
Qualifier
EGCAT
Category for
ECG
Subcategory for
ECG
ECG Position of
Subject
ECG Beat
Number
Result or
Finding in
Original Units
DefinedActivity.categoryCode
Char *
Grouping
Qualifier
Grouping
Qualifier
Record
Qualifier
Result
Qualifier
Result
Qualifier
Original Units
PerformedObservationResult.value Char (UNIT)
EGSCAT
EGPOS
EGBEATNO
EGORRES
EGORRESU
Page 41
Draft
DefinedActivity.subcategoryCode Char *
PerformedObservation.
bodyPositionCode
new variable
Char (POSITION)
PerformedObservationResult.
value
Char
Num
Variable
Qualifier
CDISC Notes
Core
Internal or external ECG identifier.
Perm
Example: UUID.
Sponsor-defined reference number. Perhaps pre-printed on the Perm
CRF as an explicit line identifier or defined in the sponsor's
operational database. Example: Line number from the ECG
page.
Short name of the measurement, test, or examination described Req
in EGTEST. It can be used as a column name when converting
a dataset from a vertical to a horizontal format. The value in
EGTESTCD cannot be longer than 8 characters, nor can it start
with a number (e.g., “1TEST”). EGTESTCD cannot contain
characters other than letters, numbers, or underscores.
Examples: PRAG, QTAG
Verbatim name of the test or examination used to obtain the
Req
measurement or finding. The value in EGTEST cannot be
longer than 40 characters. Examples: PR Interval, Aggregate,
QT Interval, Aggregate
Used to categorize ECG observations across subjects.
Perm
Examples: MEASUREMENT, FINDING, INTERVAL.
A further categorization of the ECG.
Perm
Position of the subject during a measurement or examination.
Examples: SUPINE, STANDING, SITTING.
Variable describing ECG measurements of individual beat
data. Examples: 1,2,3
Result of the ECG measurement or finding as originally
received or collected. Examples of expected values are 62 or
0.151 when the result is an interval or measurement, or
“ATRIAL FIBRILLATION” or “QT PROLONGATION”
when the result is a finding.
Original units in which the data were collected. The unit for
EGORRES. Examples: sec or msec.
Perm
Perm
Exp
Perm
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Variable Name Variable Label
EGSTRESC
Character
Result/Finding
in Std Format
EGSTRESN
Numeric
Result/Finding
in Standard
Units
Standard Units
EGSTRESU
Controlled
Type Terms, Codelist
Role
or Format
PerformedObservationResult.value Char
Result
Qualifier
BRIDG Mapping
PerformedObservationResult.
value
Num
Result
Qualifier
PerformedObservationResult.
value
PerformedObservation.
negationIndicator
Char (UNIT)
Variable
Qualifier
Record
Qualifier
EGSTAT
Completion
Status
EGREASND
Reason ECG Not PerformedObservation.
Performed
negationReason
Char
Record
Qualifier
EGXFN
External File
Name
Vendor Name
PerformedObservationResult.
value
Organization.name
Char
Lead Location
Used for
Measurement
Method of ECG
Test
Baseline Flag
complex mapping
Char (EGLEAD)
Record
Qualifier
Record
Qualifier
Record
Qualifier
PerformedObservation.
methodCode
PerformedObservationResult.
baselineIndicator
Char (EGMETHOD)
EGNAM
EGLEAD
EGMETHOD
EGBLFL
Char (ND)
Char
Char (NY)
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Record
Qualifier
Record
Qualifier
CDISC Notes
Core
Contains the result value for all findings, copied or derived
Exp
from EGORRES in a standard format or standard units.
EGSTRESC should store all results or findings in character
format; if results are numeric, they should also be stored in
numeric format in EGSTRESN. For example, if a test has
results of “NONE”, “NEG”, and “NEGATIVE” in EGORRES
and these results effectively have the same meaning, they
could be represented in standard format in EGSTRESC as
“NEGATIVE”. For other examples, see general assumptions.
Additional examples of result data: SINUS BRADYCARDIA,
ATRIAL FLUTTER, ATRIAL FIBRILLATION. .
Used for continuous or numeric results or findings in standard Perm
format; copied in numeric format from EGSTRESC.
EGSTRESN should store all numeric test results or findings.
Standardized unit used for EGSTRESC or EGSTRESN.
Perm
Used to indicate an ECG was not done, or an ECG
Perm
measurement was not taken. Should be null if a result exists in
EGORRES.
Describes why a measurement or test was not performed.
Perm
Examples: BROKEN EQUIPMENT or SUBJECT REFUSED.
Used in conjunction with EGSTAT when value is NOT
DONE.
File name and path for the external ECG Waveform file.
Perm
Name or identifier of the laboratory or vendor who provided
the test results.
The lead used for the measurement. Examples: LEAD V1,
LEAD V6, LEAD aVR, LEAD I, LEAD II, LEAD III
Perm
Method of the ECG test. Examples: 12 LEAD STANDARD.
Perm
Perm
Indicator used to identify a baseline value. The value should be Exp
“Y” or null.
Page 42
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
EGDRVFL
Derived Flag
Defer to Statistical Analysis
Modeling.
Controlled
Type Terms, Codelist
Role
or Format
Char (NY)
Record
Qualifier
EGEVAL
Evaluator
Performer.typeCode
Char *
Record
Qualifier
VISITNUM
Visit Number
Num
Timing
VISIT
Visit Name
Char
Timing
VISITDY
Planned Study
Day of Visit
Epoch
PlannedSubjectActivityGroup.
sequenceNumber
PlannedSubjectActivityGroup.
name
PlannedActivity.studyDayRange
Num
Timing
Epoch.name
Char *
Timing
Variable Name Variable Label
EPOCH
BRIDG Mapping
EGDTC
Date/Time of
ECG
PerformedObservation.dateRange Char ISO 8601
Timing
EGDY
Study Day of
ECG
PerformedObservation.
studyDayRange
Num
Timing
EGTPT
Planned Time
Point Name
PlannedSubjectActivityGroup.
name
Char
Timing
EGTPTNUM
Planned Time
Point Number
Planned Elapsed
Time from Time
Point Ref
PlannedSubjectActivityGroup.
Num
sequenceNumber
PlannedContingentOnRelationship Char ISO 8601
EGELTM
Page 43
Draft
PlannedCompositionRelationship.
pauseQuantity
Timing
Timing
CDISC Notes
Used to indicate a derived record. The value should be Y or
null. Records which represent the average of other records, or
that do not come from the CRF, or are not as originally
collected or received are examples of records that would be
derived for the submission datasets. If EGDRVFL=Y, then
EGORRES could be null, with EGSTRESC, and (if numeric)
EGSTRESN having the derived value.
Role of the person who provided the evaluation. Used only for
results that are subjective (e.g., assigned by a person or a
group). Should be null for records that contain collected or
derived data. Examples: INVESTIGATOR, ADJUDICATION
COMMITTEE, VENDOR.
1. Clinical encounter number.
2. Numeric version of VISIT, used for sorting.
1. Protocol-defined description of clinical encounter.
2. May be used in addition to VISITNUM and/or VISITDY.
Planned study day of the visit based upon RFSTDTC in
Demographics.
Epoch associated with the Element in the planned sequence of
Elements for the ARM to which the subject was assigned
Date/Time of ECG
(note: the next version of the SDTMIG will reflect “date/time
of ECG”)
1. Study day of the ECG, measured as integer days.
2. Algorithm for calculations must be relative to the sponsordefined RFSTDTC variable in Demographics.
1. Text Description of time when measurement should be
taken.
2. This may be represented as an elapsed time relative to a
fixed reference point, such as time of last dose. See
EGTPTNUM and EGTPTREF. Examples: Start, 5 min post.
Numerical version of EGTPT to aid in sorting.
Core
Perm
Perm
Exp
Perm
Perm
Perm
Exp
Perm
Perm
Perm
Planned elapsed time (in ISO 8601) relative to a fixed time point Perm
reference (EGTPTREF). Not a clock time or a date time variable.
Represented as an ISO 8601 duration. Examples: “-PT15M” to
represent the period of 15 minutes prior to the reference point
indicated by EGTPTREF, or “PT8H” to represent the period of 8
hours after the reference point indicated by EGTPTREF.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Variable Name Variable Label
BRIDG Mapping
EGTPTREF
Time Point
Reference
EGRFTDTC
Date/Time of
PerformedActivity.dateRange
Char ISO 8601
Reference Time
Point
Repetition
PerformedObservation.repetitionN Num
Number
umber
EGREPNUM
DefinedActivity.nameCode
Controlled
Type Terms, Codelist
Role
or Format
Char
Timing
Timing
Name of the fixed reference point referred to by EGELTM,
EGTPTNUM, and EGTPT. Examples: PREVIOUS DOSE,
PREVIOUS MEAL.
Date/time of the reference time point, EGTPTREF.
Core
Perm
Perm
Timing
The incidence number of a test that is repeated within a given Perm
timeframe for the same test. The level of granularity can vary,
e.g., within a time point or within a visit. For example,
multiple measurements of blood pressure or multiple analyses
of a sample.
* Indicates variable may be subject to controlled terminology, (Parenthesis indicates CDISC/ NCI codelist code value), ! Indicates terminology to be reviewed by the CDISC
Terminology Team. BRIDG Mapping values in italics are comments on why no mapping exists for that variable.
917
918
919
920
For assumptions for the ECG Tests Results domain model, please refer to the current SDTMIG.
921
6.1.2 Examples for ECG Test Results
922
923
924
925
926
927
928
929
CDISC Notes
Example 1
This example is from 10-second ECG replicates extracted from a continuous recording. The example shows one subject’s (USUBJID=2324-P0001) extracted 10second ECG replicate results. Three replicates were extracted for planned time points 1 HR and 2 HR. Summary mean measurements are reported for the 10
seconds of extracted data for each replicate. EGDTC is the date/time of the first individual beat in the extracted 10-second ECG. In this example, corrections to
QT were performed by the sponsor and are not included; however, this information can be found in the ADaM dataset. In order to save space, some permissible
variables (EGREFID, VISITDY, EGTPTNUM, EGTPTREF, EGRFTDTC) have been omitted, as marked by an ellipsis.
eg.xpt
Row
1
2
3
4
6
6
7
8
9
10
11
12
13
14
STUDYID DOMAIN USUBJID EGSEQ EGTESTCD
EGTEST
STUDY01
EG
2324-P0001
1
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
2
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
3
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
4
QTAG
QT Interval, Aggregate
STUDY01
EG
2324-P0001
5
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
6
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
7
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
8
QTAG
QT Interval, Aggregate
STUDY01
EG
2324-P0001
9
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
10
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
11
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
12
QTAG
QT Interval, Aggregate
STUDY01
EG
2324-P0001
13
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
14
RRAG
RR Interval, Aggregate
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
EGCAT
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
EGPOS EGORRES EGORRESU EGSTRESC EGSTRESN EGSTRESU
SUPINE
176
msec
176
176
msec
SUPINE
658
msec
658
658
msec
SUPINE
97
msec
97
97
msec
SUPINE
440
msec
440
440
msec
SUPINE
176
msec
176
176
msec
SUPINE
679
msec
679
679
msec
SUPINE
95
msec
95
95
msec
SUPINE
389
msec
389
389
msec
SUPINE
169
msec
169
169
msec
SUPINE
661
msec
661
661
msec
SUPINE
90
msec
90
90
msec
SUPINE
377
msec
377
377
msec
SUPINE
176
msec
176
176
msec
SUPINE
771
msec
771
771
msec
Page 44
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row
15
16
17
18
19
20
21
22
23
24
STUDYID DOMAIN USUBJID EGSEQ EGTESTCD
EGTEST
STUDY01
EG
2324-P0001
15
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
16
QTAG
QT Interval, Aggregate
STUDY01
EG
2324-P0001
17
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
18
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
19
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
20
QTAG
QT Interval, Aggregate
STUDY01
EG
2324-P0001
21
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
22
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
23
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
24
QTAG
QT Interval, Aggregate
EGCAT
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
EGPOS EGORRES EGORRESU EGSTRESC EGSTRESN EGSTRESU
SUPINE
100
msec
100
100
msec
SUPINE
379
msec
379
379
msec
SUPINE
179
msec
179
179
msec
SUPINE
749
msec
749
749
msec
SUPINE
103
msec
103
103
msec
SUPINE
402
msec
402
402
msec
SUPINE
175
msec
175
175
msec
SUPINE
771
msec
771
771
msec
SUPINE
98
msec
98
98
msec
SUPINE
356
msec
356
356
msec
930
Row
EGLEAD
EGMETHOD
VISITNUM VISIT
EGDTC
EGTPT … EGREPNUM
2
VISIT 2 2014-03-22T10:00:21 1 HR …
1
1 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:00:21 1 HR …
1
2 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:00:21 1 HR …
1
3 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:00:21 1 HR …
1
4 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:01:35 1 HR …
2
5 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:01:35 1 HR …
2
6 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:01:35 1 HR …
2
7 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:01:35 1 HR …
2
8 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:02:14 1 HR …
3
9 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:02:14 1 HR …
3
10 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:02:14 1 HR …
3
11 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T10:02:14 1 HR …
3
12 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:00:21 2 HR …
1
13 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:00:21 2 HR …
1
14 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:00:21 2 HR …
1
15 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:00:21 2 HR …
1
16 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:01:31 2 HR …
2
17 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:01:31 2 HR …
2
18 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:01:31 2 HR …
2
19 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:01:31 2 HR …
2
20 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:02:40 2 HR …
3
21 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:02:40 2 HR …
3
22 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:02:40 2 HR …
3
23 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2 2014-03-22T11:02:40 2 HR …
3
24 (cont) LEAD II 12 LEAD STANDARD
931
932
933
934
935
936
937
938
939
Example 2
This example is from ECG results where beat-to-beat measurements were recorded. The example shows one subject’s (USUBJID=2324-P0001) continuous beatto-beat EG results. Only 3 beats are shown, but there could be measurements for, as an example, 101,000 complexes in 24 hours. The actual number of
complexes in 24 hours can be variable and depends on average heart rate. The results are mapped to the EG (ECG Test Results) domain using EGBEATNO. The
controlled terminology for EGTESTCD, EGTEST, EGLEAD, and EGMETHOD is represented in the table.
If there is no result to be reported, then the row would not be included.
Page 45
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
940
941
942
943
944
Rows 1-2:
Rows 3-8:
There is no RR measurement for the first beat (EGBEATNO=1), because RR is measured as the duration (time) between the peak of the R wave
in the reported single beat and peak of the R wave in the preceding single beat. EGDTC is the date/time of the individual beat.
There is a 1-hour gap between beats 2 and 3 due to electrical interference or other artifacts that prevented measurements from being recorded.
eg.xpt
Row
1
2
3
4
5
6
7
8
STUDYID DOMAIN USUBJID EGSEQ EGTESTCD
EGTEST
STUDY01
EG
2324-P0001
1
PRSB
PR Interval, Single Beat
STUDY01
EG
2324-P0001
2
QRSSB
QRS Duration, Single Beat
STUDY01
EG
2324-P0001
3
PRSB
PR Interval, Single Beat
STUDY01
EG
2324-P0001
4
RRSB
RR Interval, Single Measurement
STUDY01
EG
2324-P0001
5
QRSSB
QRS Duration, Single Beat
STUDY01
EG
2324-P0001
6
PRSB
PR Interval, Single Beat
STUDY01
EG
2324-P0001
7
RRSB
RR Duration, Single Measurement
STUDY01
EG
2324-P0001
8
QRSSB
QRS Duration, Single Beat
EGCAT
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
EGPOS EGBEATNO EGORRES EGORRESU
SUPINE
1
176
msec
SUPINE
1
97
msec
SUPINE
2
176
msec
SUPINE
2
679
msec
SUPINE
2
95
msec
SUPINE
3
169
msec
SUPINE
3
661
msec
SUPINE
3
90
msec
945
Row EGSTRESC EGSTRESN EGSTRESU EGLEAD
EGMETHOD
VISITNUM
VISIT
VISITDY
EGDTC
176
176
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T14:32:12.3
1 (cont)
97
97
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T14:32:12.3
2 (cont)
176
176
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T14:32:13.3
3 (cont)
679
679
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T14:32:13.3
4 (cont)
95
95
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T14:32:13.3
5 (cont)
169
169
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T15:32:14.2
6 (cont)
661
661
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T15:32:14.2
7 (cont)
90
90
msec
LEAD II 12 LEAD STANDARD
1
SCREENING
-7
2014-02-11T15:32:14.2
8 (cont)
946
947
948
949
950
951
952
953
954
955
956
Example 3
In order to be able to indicate which ECG records were used in the calculation of the population and individual QT correction coefficients, the SUPPEG dataset
was used to flag those records that were used in this evaluation. This example also shows the corresponding EG dataset with pre-dose ECG data for one subject’s
(USUBJID=2324-P0001) extracted 10-second ECG replicate results. Summary mean measurements are reported for the 10 seconds of extracted data (for each
replicate). EGDTC is the date/time of the start of the 10-second ECG segment.
Rows 1-12:
Rows 13-16:
Extract of pre-dose ECG results showing results for PR, RR, QRS and QT measurements.
Extract of post-dose ECG results showing results for PR, RR, QRS and QT measurements. Only the first replicate is included.
eg.xpt
Row
1
2
3
4
5
6
7
8
STUDYID DOMAIN USUBJID EGSEQ EGTESTCD
EGTEST
STUDY01
EG
2324-P0001
1
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
2
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
3
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
4
QTAG
QT Interval, Aggregate
STUDY01
EG
2324-P0001
5
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
6
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
7
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
8
QTAG
QT Interval, Aggregate
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
EGCAT
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
EGPOS EGORRES EGORRESU EGSTRESC EGSTRESN EGSTRESU
SUPINE
176
msec
176
176
msec
SUPINE
658
msec
658
658
msec
SUPINE
97
msec
97
97
msec
SUPINE
343
msec
343
343
msec
SUPINE
176
msec
176
176
msec
SUPINE
679
msec
679
679
msec
SUPINE
95
msec
95
95
msec
SUPINE
345
msec
345
345
msec
Page 46
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row
9
10
11
12
13
14
15
16
STUDYID DOMAIN USUBJID EGSEQ EGTESTCD
EGTEST
STUDY01
EG
2324-P0001
9
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
10
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
11
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
12
QTAG
QT Interval, Aggregate
STUDY01
EG
2324-P0001
13
PRAG
PR Interval, Aggregate
STUDY01
EG
2324-P0001
14
RRAG
RR Interval, Aggregate
STUDY01
EG
2324-P0001
15
QRSAG
QRS Duration, Aggregate
STUDY01
EG
2324-P0001
16
QTAG
QT Interval, Aggregate
EGCAT
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
INTERVAL
EGPOS EGORRES EGORRESU EGSTRESC EGSTRESN EGSTRESU
SUPINE
169
msec
169
169
msec
SUPINE
661
msec
661
661
msec
SUPINE
90
msec
90
90
msec
SUPINE
352
msec
352
352
Msec
SUPINE
176
msec
173
176
msec
SUPINE
658
msec
666
658
msec
SUPINE
97
msec
94
97
msec
SUPINE
343
msec
347
343
msec
957
Row
EGLEAD
EGMETHOD
VISITNUM VISIT VISITDY
EGDTC
EGTPT … EGREPNUM
2
VISIT 2
1
2014-02-11T14:32:12 Pre-dose …
1
1 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:32:12 Pre-dose …
1
2 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:32:12 Pre-dose …
1
3 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:32:12 Pre-dose …
1
4 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:33:13 Pre-dose …
2
5 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:33:13 Pre-dose …
2
6 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:33:13 Pre-dose …
2
7 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:33:13 Pre-dose …
2
8 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:34:14 Pre-dose …
3
9 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:34:14 Pre-dose …
3
10 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:34:14 Pre-dose …
3
11 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T14:34:14 Pre-dose …
3
12 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T15:02:12 30 min post …
1
13 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T15:02:12 30 min post …
1
14 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T15:02:12 30 min post …
1
15 (cont) LEAD II 12 LEAD STANDARD
2
VISIT 2
1
2014-02-11T15:02:12 30 min post …
1
16 (cont) LEAD II 12 LEAD STANDARD
958
959
960
961
962
Rows 1-12:
Links those records from eg.xpt that were used in the calculation of the population and individual QT correction coefficients (i.e. the RR and QT
measurements). The data between suppeg.xpt and eg.xpt is linked via the EGSEQ variable in eg.xpt (IDVAR and IDVARVAL in suppeg.xpt).
suppeg.xpt
Row
1
2
3
4
5
6
7
8
9
10
11
12
STUDYID RDOMAIN USUBJID IDVAR IDVARVAL QNAM
QLABEL
QVAL QORIG QEVAL
STUDY01
EG
2324-P0001 EGSEQ
2
QTCCIFL Individ QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
2
QTCCNFL Populatn QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
4
QTCCIFL Individ QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
4
QTCCNFL Populatn QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
6
QTCCIFL Individ QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
6
QTCCNFL Populatn QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
8
QTCCIFL Individ QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
8
QTCCNFL Populatn QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
10
QTCCIFL Individ QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
10
QTCCNFL Populatn QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
12
QTCCIFL Individ QT Correction Coefficient Flag
Y
Derived
STUDY01
EG
2324-P0001 EGSEQ
12
QTCCNFL Populatn QT Correction Coefficient Flag
Y
Derived
963
Page 47
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
6.2 QT Correction
Building on the information presented in Section 2.3.4, the following is a summary of the correction methods that can be used in TQT studies.
Comparing QT intervals to determine whether a drug has affected them works best when the drug is the only variable to be taken into consideration, which
means correcting the raw data for the other major influencing factor, i.e., heart rate. HR-corrected QT (QTc) intervals may be derived in a variety of ways:
• Historical Population QTc: Derived by fitting a model to data from a historical population of subjects. The formula for correcting QT, including its
coefficient(s), is based on a formula which is known before the study starts, such as Bazett’s or Fridericia’s corrections.
• Study Population QTc: Derived by fitting a model to data from the population of subjects under study. The formula for correcting QT is known before
the study starts, but the coefficient(s) in the correction formula are derived from data collected in the study.
• Multi-model Study Population QTc: Derived by fitting several models to data for the study population, then identifying one or more “best” models and
using a correction method based on the best model(s). The candidate models, the method for choosing the best model(s) and, for methods that use
multiple best models, the way in which the best models are combined, are known before the study starts. The formulas for correcting QT, as well as its
coefficient(s), are derived from data collected in the study.
• Individual QTc: Derived by fitting a model to data for each study subject individually. The formula for correcting QT is known before the study starts,
but the coefficient(s) in the formula are derived from data collected from the individual subject.
• Multi-model Individual QTc: Derived by fitting several models to data for each subject, then identifying one or more “best fit” models for the subject
and using a correction method based on the best fit model(s). The candidate models, the method for choosing the best model(s) and, for methods that use
multiple best models, the way in which the best models are combined, are known before the study starts. The formulas for correcting QT, as well as its
coefficient(s), are derived from data collected from the individual subject.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 48
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
986 This concept map displays the process for determining QT correction
987 coefficients for a pre-specified model for a study population.
988
989
990
Concept Map 5: Population QT Correction for a Pre-specified Model
This concept map displays the process for determining QT correction
coefficients for a pre-specified model for individuals.
Concept Map 6: Individual QT Correction for a Pre-specified Model
991
Page 49
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
992
993
994
995
This concept map displays the process for determining QT correction formula and coefficient(s) for Individual QTc ‘Best Fit’ Model.
Concept Map 7: Individual QT Correction (QTc) ‘Best Fit’ Model
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 50
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
996
997
998
999
1000
1001
1002
1003
1004
1005
1006
1007
1008
1009
1010
1011
1012
1013
1014
1015
1016
1017
1018
1019
1020
1021
For each QT correction method used in the study, values of EGTESTCD and EGTEST are assigned at the study level. CDISC Controlled Terminology includes
EGTESTCD and EGTEST values for QT corrected by Bazett’s method and QT corrected by Fridericia’s method. CDISC Controlled Terminology does not plan
to include any other corrected QT tests in CDISC Controlled Terminology in the near future. This is because of the large and growing number of QT corrections
in use, and the fact that there is not scientific consensus on a best fit QT correction method. A protocol may specify that several different QT correction models
be assessed and that only the best fit model for the individual be used for calculating the individual’s corrected QT intervals. Another alternative is that the
protocol specifies multiple models be computed for each individual irrespective of best fit. There are multiple scenarios of analysis, and the sponsor should
assign values for EGTESTCD/EGTEST appropriately with clear documentation on what each test code represents. EGTESTCD's begin with the fragment 'QTCI'
for individual-based corrected QT intervals, and 'QTCN' for population-based corrected QT intervals. For example, if the protocol calls for computing the top
two best fit models, the sponsor could choose to name the top best fit model QTCIAG1 and the second best fit model QTCIAG2, in rank order.
This domain provides information about the QT corrections used in the study. In that sense, it contains supporting information about tests. The name of the test
about which information is being supplied is stored in QTGRPID.
This domain is provided when coefficients for QT corrections other than Bazett’s or Fridericia’s are used in a study, and those corrections are derived from data
collected in the study (i.e., are not historical), and the derivation was performed by a vendor, rather than the sponsor. Information on historical correction
methods or derived by the sponsor could be submitted in the QT data domain, if agreed by the submitter and recipient (e.g., study sponsor and regulatory
agency).
For both individual and population corrections, the USUBJID will be populated.
6.2.1 Specification for ECG QT Correction
The following table is based on the proposed SDTM Findings QT domain model. An additional column, BRIDG 3.0.3 Mapping, has also been included.
Controlled terminology is used where it exists. Please refer to the current CDISC Controlled Terminology
(http://www.cancer.gov/cancertopics/cancerlibrary/terminologyresources/cdisc). Sponsor-defined controlled terminology should be provided in the Define-XML.
qt.xpt, ECG QT Correction Model Data — Findings. One record per QT correction observation per subject, Tabulation.
Variable Name Variable Label
STUDYID
DOMAIN
USUBJID
QTSEQ
Page 51
Draft
BRIDG 3.0.3 Mapping
Study Identifier DocumentIdentifier.identifier
Domain
PerformedObservation
Abbreviation
Unique Subject SubjectIdentifier.identifier
Identifier
Sequence
Number
Type
Char
Char
Controlled
Terms, Codelist
or Format
Role
CDISC Notes
Core
Identifier
Identifier
Unique identifier for a study.
Two-character abbreviation for the domain.
Char
Identifier
implementation specific record Num
identifier
Identifier
Identifier used to uniquely identify a subject across all Req
studies for all applications or submissions involving the
product.
Sequence Number given to ensure uniqueness of
Req
subject records within a domain. May be any valid
number.
QT
Req
Req
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Variable Name Variable Label
BRIDG 3.0.3 Mapping
QTGRPID
Group ID
QTREFID
QTc Reference implementation specific
ID
Sponsorimplementation specific
Defined
Identifier
QTSPID
Type
Controlled
Terms, Codelist
or Format
implementation specific record Char
grouping variable
Role
Identifier
Char
Identifier
Char
Identifier
QTTESTCD
QTc Test or
Examination
Short Name
DefinedObservation.nameCode Char
Topic
QTTEST
QTc Test or
Examination
Name
DefinedObservation.nameCode Char
Synonym
Qualifier
QTCAT
Category for
QTc
Subcategory for
QTc
Result or
Finding in
Original Units
DefinedActivity.categoryCode Char
*
DefinedActivity.subcategoryCo Char
de
PerformedObservationResult. Char
value
*
QTSCAT
QTORRES
QTSTRESC
Character
PerformedObservationResult.
Result/Finding value
in Std Format
Char
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Grouping
Qualifier
Grouping
Qualifier
Result Qualifier
Result Qualifier
CDISC Notes
Core
Used to tie together a block of related records in a
single domain for a subject. Example: use EGTESTCD
as value for QTGRPID for easier relating of QT dataset
to the EG dataset.
Internal or external QT identifier. Generally not used
for QT dataset.
Sponsor-defined reference number. Perhaps pre-printed
on the CRF as an explicit line identifier or defined in
the sponsor's operational database. Generally not used
for QT dataset.
Short name of the measurement, test, or examination
described in QTTEST. It can be used as a column name
when converting a dataset from a vertical to a
horizontal format. The value in QTTESTCD cannot be
longer than 8 characters, nor can it start with a number
(e.g., “1TEST”). QTTESTCD cannot contain characters
other than letters, numbers, or underscores. Examples:
QTCDESC, QTCFORM.
Verbatim name of the test or examination used to
obtain the measurement or finding. The value in
QTTEST cannot be longer than 40 characters.
Examples: QT Correction Method Description, QT
Correction Formula
Used to categorize QT correction model observations
across subjects. Generally not used in QT dataset.
A further categorization of the QT correction model.
Generally not used in QT dataset.
Result of the QT correction model measurement or
finding as originally received or collected. Examples of
expected values are 0.132 or 0.432 when the result is a
coefficient or measurement, or “LINEAR” or
“PARABOLIC LOG/LOG” when the result is a
formula.
Contains the result value for all findings, copied or
derived from QTORRES in a standard format or
standard units.
QTSTRESC should store all results or findings in
character format; if results are numeric, they should
also be stored in numeric format in QTSTRESN, for
example 0.132 or 0.432 when the result is a coefficient
or measurement, or “LINEAR” or “PARABOLIC
Perm
Perm
Perm
Req
Req
Perm
Perm
Exp
Exp
Page 52
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Variable Name Variable Label
BRIDG 3.0.3 Mapping
Type
Controlled
Terms, Codelist
or Format
Role
CDISC Notes
Core
1022
1023
LOG/LOG” when the result is a formula.
Used for continuous or numeric results or findings in
Perm
standard format; copied in numeric format from
QTSTRESC. QTSTRESN should store all numeric test
results or findings.
QTSTAT
PerformedObservation.
Char
(ND)
Record Qualifier Used to indicate a QT correction model was not done. Perm
negationIndicator
Should be null if a result exists in QTORRES.
QTREASND
PerformedObservation.
Char
Record Qualifier Describes why a measurement or test was not
Perm
negationReason
performed. Used in conjunction with QTSTAT when
correction is NOT DONE.
QTNAM
Vendor Name Organization.name
Char
Record Qualifier Name or identifier of the laboratory or vendor who
Perm
provided the test results.
QTDRVFL
Derived Flag
Defer to Statistical Analysis
Char
(NY)
Record Qualifier Used to indicate a derived record. The value should be Perm
Modeling.
Y or null. Records which represent the average of other
records, or that do not come from the CRF, or are not as
originally collected or received are examples of records
that would be derived for the submission datasets. If
QTDRVFL=Y, then QTORRES could be null, with
QTSTRESC, and (if numeric) QTSTRESN having the
derived value.
* Indicates variable may be subject to controlled terminology (Parenthesis indicates CDISC/ NCI codelist code value). BRIDG Mapping values in italics are comments on why no
mapping exists for that variable.
1024
6.2.2 Examples for ECG QT Corrections
QTSTRESN
1025
1026
1027
1028
1029
Numeric
Result/Finding
in Standard
Units
Completion
Status
Reason QTc
Not Performed
PerformedObservationResult.
value
Num
Result Qualifier
The example consists of two Findings datasets: one for the ECG measurements (eg.xpt) and one for the QTc modeling information (qt.xpt). To tie the EG and QT
datasets together will also require use of RELREC (relrec.xpt) to represent a dataset-to-dataset relationship that holds true for all subjects for all values of
EGTESTCD and QTGRPID.
eg.xpt
Row DOMAIN
USUBJID
EGSEQ EGCAT EGTESTCD
EGTEST
EGORRES EGORRESU …
EG
P384QT204_001
1
INTERVAL QTCIAG1 QTCI Interval, Aggregate 1
345
msec
…
1
EG
P384QT204_001
2
INTERVAL QTCIAG2 QTCI Interval, Aggregate 2
350
msec
…
2
EG
P384QT204_001
3
INTERVAL QTCNAG
QTCN Interval, Aggregate
353
msec
…
3
EG
P384QT204_002
1
INTERVAL QTCIAG1 QTCI Interval, Aggregate 1
372
msec
…
4
EG
P384QT204_002
2
INTERVAL QTCIAG2 QTCI Interval, Aggregate 2
374
msec
…
5
EG
P384QT204_002
3
INTERVAL QTCNAG
QTCN Interval, Aggregate
377
msec
…
6
EG
P384QT204_003
1
INTERVAL QTCIAG1 QTCI Interval, Aggregate 1
401
msec
…
7
EG
P384QT204_003
2
INTERVAL QTCIAG2 QTCI Interval, Aggregate 2
411
msec
…
8
EG
P384QT204_003
3
INTERVAL QTCNAG
QTCN Interval, Aggregate
414
msec
…
9
1030
Page 53
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1031
qt.xpt
Row DOMAIN
USUBJID
QTSEQ QTGRPID
QT
P384QT204_001
1
QTCIAG1
1
QT
P384QT204_001
2
QTCIAG1
2
QT
P384QT204_001
3
QTCIAG1
3
QT
P384QT204_001
4
QTCIAG2
4
QT
P384QT204_001
5
QTCIAG2
5
QT
P384QT204_001
6
QTCIAG2
6
QT
P384QT204_001
7
QTCNAG
7
QT
P384QT204_001
8
QTCNAG
8
QT
P384QT204_001
9
QTCNAG
9
QT
P384QT204_001
10
QTCNAG
10
QT
P384QT204_001
11
QTCNAG
11
QT
P384QT204_002
1
QTCIAG1
12
QT
P384QT204_002
2
QTCIAG1
13
QT
P384QT204_002
3
QTCIAG1
14
QT
P384QT204_002
4
QTCIAG2
15
QT
P384QT204_002
5
QTCIAG2
16
QT
P384QT204_002
6
QTCIAG2
17
QT
P384QT204_002
7
QTCNAG
18
QT
P384QT204_002
8
QTCNAG
19
QT
P384QT204_002
9
QTCNAG
20
QT
P384QT204_002
10
QTCNAG
21
QT
P384QT204_002
11
QTCNAG
22
QT
P384QT204_003
1
QTCIAG1
23
QT
P384QT204_003
2
QTCIAG1
24
QT
P384QT204_003
3
QTCIAG1
25
QT
P384QT204_003
4
QTCIAG2
26
QT
P384QT204_003
5
QTCIAG2
27
QT
P384QT204_003
6
QTCIAG2
28
QT
P384QT204_003
7
QTCNAG
29
QT
P384QT204_003
8
QTCNAG
30
QT
P384QT204_003
9
QTCNAG
31
QT
P384QT204_003
10
QTCNAG
32
QT
P384QT204_003
11
QTCNAG
33
1032
1033
QTTESTCD
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
QTCCOEFB
QTCCOEFC
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
QTCCOEFB
QTCCOEFC
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
QTCCOEFB
QTCCOEFC
QTTEST
QTORRES
…
QT Correction Method Description PARABOLIC LOG/LOG …
QT Correction Formula
QTC=QT/(RR^A)
…
QT Correction Coefficient A
0.432
…
QT Correction Method Description
LINEAR
…
QT Correction Formula
QTC=QT+(A*(1-RR))
…
QT Correction Coefficient A
0.154
…
QT Correction Method Description
RAUTAHARJU COR
…
QT Correction Formula
QTC=QT+A-(B*(e^(C^HR)) …
QT Correction Coefficient A
0.2425
…
QT Correction Coefficient B
0.434
…
QT Correction Coefficient C
-0.0097
…
QT Correction Method Description PARABOLIC LOG/LOG …
QT Correction Formula
QTC=QT/(RR^A)
…
QT Correction Coefficient A
0.374
…
QT Correction Method Description
LINEAR
…
QT Correction Formula
QTC=QT+(A*(1-RR))
…
QT Correction Coefficient A
0.132
…
QT Correction Method Description
RAUTAHARJU COR
…
QT Correction Formula
QTC=QT+A-(B*(e^(C^HR)) …
QT Correction Coefficient A
0.2425
…
QT Correction Coefficient B
0.434
…
QT Correction Coefficient C
-0.0097
…
QT Correction Method Description PARABOLIC LOG/LOG …
QT Correction Formula
QTC=QT/(RR^A)
…
QT Correction Coefficient A
0.412
…
QTc Model Type
LINEAR
…
QT Correction Formula
QTC=QT+(A*(1-RR))
…
QT Correction Coefficient A
0.158
…
QTc Model Type
RAUTAHARJU COR
…
QT Correction Formula
QTC=QT+A-(B*(e^(C^HR)) …
QT Correction Coefficient A
0.2425
…
QT Correction Coefficient B
0.434
…
QT Correction Coefficient C
-0.0097
…
relrec.xpt
Row USUBJID RDOMAIN
IDVAR
IDVARVAL RELTYPE RELID
EG
EGTESTCD
MANY
EGEQ1
1
QT
QTGRPID
MANY
EGEQ1
2
1034
1035
6.3 PK Assessments
1036
1037
An appropriate PK sampling schedule should be designed to adequately characterize the PK/PD concentration/response relationship for QT/QTc prolongation.
The evaluation of the concentration-QT relationship involves individual subject responses instead of averaging the QT response across subjects at each time
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 54
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1038
1039
1040
point. Concentration-QTc analysis helps in predicting QT prolongation at doses beyond those used in TQT studies and may help determine whether excessive QT
prolongation observed in a few subjects is due to high drug concentration or due to differential susceptibility in these subjects.
1041
1042
1043
1044
1045
1046
This concept map displays the sequence of steps involved if performing a PK/PD analysis within a TQT study. At each time point, data on PK drug
concentrations and the relevant PD parameter such as QT are obtained to allow for PK/PD modeling. SDTM standard datasets (PC and PP) will be appropriate
for capturing PK data from TQT study, as well.
1047
6.4 Hemodynamics/Vital Signs
1048
1049
1050
1051
1052
1053
1054
1055
1056
1057
1058
1059
Concept Map 8: PK/PD Analysis Steps and Timing
This section focuses on the assessment of response to orthostatic challenge. Orthostatic, or postural, vital signs are used to assess the body’s response to position
changes when a suspected drug might have the potential for altering autonomic tone and/or directly alter vascular tone or other physiological processes
influencing cardiac output. Conditions leading to hypovolemia and autonomic abnormalities may result in a sudden drop in blood pressure, known as orthostatic
hypotension, and result in impaired perfusion to the upper body. Assessment of response to orthostatic challenge might be conducted for two purposes within a
TQT study: 1) as a screening mechanism to screen out potential subject that display an abnormal hemodynamic response; 2) as an assessment if the TQT study is
being used for the secondary purpose of the careful and well-controlled assessment of the test drug’s influence on hemodynamic parameters. Severe orthostatic
hypotension may exclude a patient from participation in TQT studies.
Heart rate and blood pressure at rest and under orthostatic challenge might be evaluated as screening procedures for TQT studies. Additionally, the TQT study is
well suited to collect information on potential changes in these parameters with drug treatment as a secondary study objective. Please note that resting heart rate
and blood pressure would routinely be assessed during the study; however, assessment of orthostatic effect is optional.
Page 55
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1060
1061
1062
1063
1064
1065
1066
Concept Map 9: Assessment of Orthostasis
This concept map shows that the assessment of orthostasis is made based on changes in vital sign measurements measured within 3 minutes of standing or the
inability to stand because the patient is severely orthostatic symptomatic and/or symptom assessment. This concept map is based on several key publications that
are outlined below. Reference limits for orthostasis are also provided; the sponsor predefines these limits generally in the statistical analysis plan.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 56
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
American Autonomic Society / American Academy of Neurology Consensus Statement, 1996 16
• Provides a definitive definition of orthostasis based on either systolic or diastolic blood pressure decrease (see reference limits in table below)
o Does not suggest frequency of measurement or time of first measurement within the 3 minute interval from standing
• Does not provide definition based on heart rate increase
Low, 1995 17
• Defined postural orthostatic tachycardia syndrome (POTS) as an increase in heart rate ≥30 bpm or absolute value ≥120 bpm within 5 minutes of
standing associated with symptoms of orthostatic intolerance.
Garland, 2007 18
• POTS as an increase in heart rate ≥30 bpm or absolute value ≥120 bpm but measured at 3 and 5 minutes in standing position, for study purposes, in the
absence of orthostatic hypotension (i.e. SBP decrease >20 mmHg, DBP decrease > 10 mmHg)
1067
1068
Reference Limits for Orthostasis (All Ages)
Basic Assessment of Orthostasis (time= 3 min)
Orthostatic hypotension
Decrease in systolic blood pressure when going from 5 minutes supine to 3 minutes standing of ≥20 mmHg or the inability to stand
(systolic)
quickly for the measurements due to symptoms of orthostasis.
Orthostatic hypotension
Decrease in diastolic blood pressure when going from 5 minutes supine to 3 minutes standing of ≥10 mmHg or the inability to stand
(diastolic)
quickly for the measurements due to symptoms of orthostasis.
Orthostatic pulse
Increase in heart rate when going from 5 minutes supine to 3 minutes standing of ≥30 bpm.
(tachycardia)
Detailed Assessment of Orthostasis (time = 0, 1, 2, 3 min)
Orthostatic hypotension
Decrease in systolic blood pressure when going from 5 minutes supine to any point during standing of ≥20 mmHg or the inability to
(systolic)
stand quickly for the measurements due to symptoms of orthostasis.
Orthostatic hypotension
(diastolic)
Decrease in diastolic blood pressure when going from 5 minutes supine to any point during standing of ≥10 mmHg or the inability
to stand quickly for the measurement due to symptoms of orthostasis.
Orthostatic pulse
(tachycardia)
Increase in heart rate when going from 5 minutes supine to standing of ≥30 bpm.
1069
1070
1071
1072
Page 57
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1073
1074
1075
1076
1077
1078
1079
1080
1081
1082
1083
1084
1085
1086
1087
1088
1089
1090
1091
6.4.1 Example for Orthostatic Challenge
Example 1
This example shows blood pressure and heart rate data collected to assess a subject’s response to orthostatic challenge (i.e., standing up quickly). These
measurements are part of a pharmacodynamic profile, which involves repeating the challenge pre-dose and at certain time points post-dose. At each time point,
these vital signs are measured while the subject is supine, and then three times after they stand up. The example shows data for two time points: one before the
Day 1 dose of study treatment, and one an hour after the Day 1 dose. Some variables have been replaced with an ellipsis to save space.
In this study, analysis of the orthostatic challenge data utilizes an approach to “baseline” data that is more complex than can be described by flagging a single
value per subject per test using the VSBLFL variable, so the flag has not been populated. Since the absence of any records with VSBLFL populated would likely
to be noted by validation software, the rationale for not using the flag would be included in the Study Data Reviewer Guide.
Rows 1-12:
Show vital signs for the pre-dose orthostatic challenge. All the data was collected at “PREDOSE” time point. The nominal planned elapsed time
between this time point and the dose is 15 minutes (VSELTM = -P15M), although actual times range over 4 minutes. The replicate number
(VSREPNUM) and the actual time of the measurement (VSDTC) distinguish among the measurements during this nominal time point. The
measurements at the fourth replicate are flagged with the SDTM baseline flag (BLFL) since these were the last measurements before the start of
dosing. The ADaM datasets will describe how measurements are used as baseline measurements in analyses.
Rows 13-24: Show vital signs data for the time point “HOUR 1.” As for the pre-dose time point, the four replicate sets of measurements within the nominal
time point were made over the course of four minutes.
vs.xpt
Row STUDYID DOMAIN USUBJID VSSEQ VSTESTCD
VSTEST
ABC
VS
ABC-001-001
7
SYSBP
Systolic Blood Pressure
1
ABC
VS
ABC-001-001
8
DIABP
Diastolic Blood Pressure
2
VSPOS
SUPINE
SUPINE
VSCAT
RESTING
RESTING
VSORRES
133
77
RESTING
70
3
ABC
VS
ABC-001-001
9
HR
Heart Rate
SUPINE
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
ABC
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
VS
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
ABC-001-001
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
SYSBP
DIABP
HR
SYSBP
DIABP
HR
SYSBP
DIABP
HR
SYSBP
DIABP
HR
SYSBP
DIABP
HR
SYSBP
DIABP
HR
SYSBP
Systolic Blood Pressure
Diastolic Blood Pressure
Heart Rate
Systolic Blood Pressure
Diastolic Blood Pressure
Heart Rate
Systolic Blood Pressure
Diastolic Blood Pressure
Heart Rate
Systolic Blood Pressure
Diastolic Blood Pressure
Heart Rate
Systolic Blood Pressure
Diastolic Blood Pressure
Heart Rate
Systolic Blood Pressure
Diastolic Blood Pressure
Heart Rate
Systolic Blood Pressure
STANDING
STANDING
STANDING
STANDING
STANDING
STANDING
STANDING
STANDING
STANDING
SUPINE
SUPINE
SUPINE
STANDING
STANDING
STANDING
STANDING
STANDING
STANDING
STANDING
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
RESTING
RESTING
RESTING
127
70
74
125
77
78
121
88
76
117
81
95
115
78
80
116
79
80
121
VSORRESU
mmHg
mmHg
BEATS/MIN
BEATS/MIN
mmHg
mmHg
BEATS/MIN
mmHg
mmHg
BEATS/MIN
mmHg
mmHg
BEATS/MIN
mmHg
mmHg
BEATS/MIN
mmHg
mmHg
BEATS/MIN
mmHg
mmHg
BEATS/MIN
mmHg
… VSBLFL
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
…
Page 58
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row STUDYID DOMAIN USUBJID VSSEQ VSTESTCD
VSTEST
ABC
VS
ABC-001-001
29
DIABP
Diastolic Blood Pressure
23
ABC
VS
ABC-001-001
30
HR
Heart Rate
24
VSPOS
STANDING
STANDING
VSCAT
VSORRES
78
82
VSORRESU
mmHg
BEATS/MIN
… VSBLFL
…
…
1092
Row
1 (cont)
2 (cont)
3 (cont)
4 (cont)
5 (cont)
6 (cont)
7 (cont)
8 (cont)
9 (cont)
10 (cont)
11 (cont)
12 (cont)
13 (cont)
14 (cont)
15 (cont)
16 (cont)
17 (cont)
18 (cont)
19 (cont)
20 (cont)
21 (cont)
22 (cont)
23 (cont)
24 (cont)
1093
1094
1095
1096
1097
1098
1099
1100
1101
1102
VISIT VISITNUM
VSDTC
VSTPT VSTPTNUM VSELTM VSTPTREF
VSRFTDTC
VSREPNUM
DAY 1
1
2014-04-01T7:45 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T7:45 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T7:45 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T7:47 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T7:47 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T7:47 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T7:48 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
2
DAY 1
1
2014-04-01T7:48 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
2
DAY 1
1
2014-04-01T7:48 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
2
DAY 1
1
2014-04-01T7:49 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
3
DAY 1
1
2014-04-01T7:49 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
3
DAY 1
1
2014-04-01T7:49 PREDOSE
1
-P15M DAY 1 DOSE 2014-04-01T8:00
3
DAY 1
1
2014-04-01T9:00 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T9:00 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T9:00 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T9:02 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T9:02 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T9:02 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
1
DAY 1
1
2014-04-01T9:03 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
2
DAY 1
1
2014-04-01T9:03 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
2
DAY 1
1
2014-04-01T9:03 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
2
DAY 1
1
2014-04-01T9:04 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
3
DAY 1
1
2014-04-01T9:04 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
3
DAY 1
1
2014-04-01T9:04 HOUR 1
2
P1H
DAY 1 DOSE 2014-04-01T8:00
3
7 Data Analysis
This section contains analysis data metadata and examples of analysis datasets used for the analysis and reporting of ECG data.
Caveats and disclaimers:
• The analysis data metadata examples are not intended to illustrate every possible variable that might be included in the analysis dataset. Sponsors should
consult the ADaMIG for further guidance on additional variables that may be added.
• The Source/Derivation column is for illustration purposes only and not intended to imply universally accepted definitions or derivations of variables.
Algorithms are producer-defined and dependent on trial and analysis design.
• The examples are for illustration purposes only and should not be viewed as a statement of the standards themselves. In addition, the examples are
intended to illustrate content and not appearance.
Page 59
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1103
7.1 ADEG Example
1104
7.1.1 Analysis Data Metadata
1105
1106
1107
1108
1109
1110
1111
1112
1113
1114
1115
1116
1117
1118
1119
The ADEG dataset is based on the ADaM Basic Data Structure (BDS).
Analysis Dataset Metadata for ADEG Example
Dataset Name
Dataset Description
Dataset Location
ADEG
Analysis Dataset for ECG
Tests
adeg.xpt
Dataset Structure
Key Variables of Dataset
One record per subject per
STUDYID, USUBJID,
parameter per analysis
PARAMCD, APERIOD,
period per analysis visit per
AVISIT, ATPT
analysis time point
Class of Dataset
Documentation
BDS
adeg.sas
In the metadata illustrated in the table below, the parameter identifier of “*ALL*” is used for variables expected to be consistent across analysis parameters and
“*DEFAULT*” is used for parameters not otherwise specified in the Parameter Identifier column for that variable. The value of PARAMCD as the parameter
identifier is used for variables that have metadata dependent on the analysis parameter.
In this example, the study has a crossover design schema and ECG measurements were collected as a set of 3 replicates (each replicate approximately 1 minute
apart) at time points 1 hr, 2 hr, etc. on Day -1 (baseline day) and Day 1 for each period, Period 1-4. The average of each set of 3 replicates is derived as a new
record in ADaM and used for analysis. A time-matched baseline is used, in which the baseline for each period is the average of the 3 replicates at each time point
on the baseline day, and corresponds to the average of the 3 replicates at each post-dose time point. QTcF and QTcI are the corrected QT measurements. QTcF
was provided by the vendor and included in SDTM and QTcI is derived by the sponsor.
Analysis Variable Metadata for ADEG Example
Dataset
Name
ADEG
Parameter
Identifier
*ALL*
ADEG
*ALL*
ADEG
ADEG
*ALL*
*ALL*
Variable Name
Variable Label
STUDYID
Study Identifier
Unique Subject
Identifier
USUBJID
TRTSEQA
TRTA
Actual Sequence of
Treatments
Actual Treatment
Variable
Type
text
Display
Format
$12
Codelist / Controlled Terms
Source / Derivation
ADSL.STUDYID
text
$20
ADSL.USUBJID
text
Placebo-Drug A 100 mgMoxifloxacin-Drug A 1 mg,
Moxifloxacin-Placebo-Drug A 1
mg-Drug 1 100 mg,
$50
Drug A 1 mg-Moxifloxacin-Drug A
100 mg-Placebo,
Drug A 100 mg-Drug A 1 mgPlacebo-Moxifloxacin
ADSL.TRTSEQA
text
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
$15
Drug A 1 mg,
Drug A 100 mg,
Placebo,
Moxifloxacin
If APERIOD=1 then
TRTA=ADSL.TRT01A,
If APERIOD=2 then
TRTA=ADSL.TRT02A
If APERIOD=3 then
Page 60
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Dataset
Name
ADEG
ADEG
ADEG
ADEG
ADEG
Parameter
Identifier
*DEFAULT*
QTCIAG
PARAMCD
*ALL*
*ALL*
Variable Name
PARAM
PARAM
PARAMCD
APERIODC
APERIOD
Variable Label
Parameter
Parameter
Parameter Code
Period (C)
Period
Variable
Type
text
text
text
text
integer
Display
Format
Codelist / Controlled Terms
Heart Rate (BEATS/MIN);
PR Interval, Aggregate (msec); RR
$50
Interval, Aggregate (msec); QT
Interval, Aggregate (msec); QTCF
Interval, Aggregate (msec); ….
$50
$8
QTCI Interval, Aggregate (msec)
HR, PRAG,
RRAG,
QTAG, QTCFAG, QTCIAG
$8
PERIOD 1,
PERIOD 2,
PERIOD 3,
PERIOD 4
8
1=PERIOD 1,
2=PERIOD 2,
3=PERIOD 3,
4=PERIOD 4
Source / Derivation
TRTA=ADSL.TRT03A,
If APERIOD=4 then
TRTA=ADSL.TRT04A
For records with a corresponding
record in EG, populate with the value
in EGTEST concatenated with the
unit in EGSTRESU in parentheses.
For records created to contain the QT
Individual Correction result, populate
with “QTCI Interval, Aggregate
(msec)”
One-to-one correspondence with
PARAM. For records with a
corresponding record in EG,
PARAMCD=EG.EGTESTCD.
If PARAM=”QTCI Interval,
Aggregate (msec)”, then
PARAMCD=”QTCIAG”
Populate based on EG.VISITNUM.
If VISITNUM=2 then
APERIODC=”PERIOD 1”,
If VISITNUM=4 then
APERIODC=”PERIOD 2”,
…
Numeric version of APERIODC.
Populate based on EG.VISITNUM.
ADEG
*ALL*
AVISIT
Analysis Visit
text
ADEG
*ALL*
AVISITN
Analysis Visit (N)
integer
Page 61
Draft
$11
8
DAY -1,
DAY 1
-1=DAY -1,
If VISITNUM=2 then
AVISIT=’DAY -1’,
Else if VISITNUM=3 then
AVISIT=’DAY 1’
….
[Note, in this example visit
windowing was not performed for
analysis.]
Numeric version of AVISIT.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Dataset
Name
ADEG
Parameter
Identifier
*ALL*
Variable Name
ATPT
Variable Label
Analysis Time Point
Variable
Type
text
Display
Format
Codelist / Controlled Terms
$40
ADEG
*ALL*
ATPTN
Analysis Time Point
(N)
integer
8
ADEG
*ALL*
ADT
Analysis Date
integer
yymmdd10.
ADEG
*ALL*
ATM
Analysis Time
Integer
time8.
ADEG
*ALL*
ADY
Analysis Relative
Day
Integer
8
ADEG
ADEG
*ALL*
*ALL*
AVISDY
APERDAY
Analysis Visit Day
Analysis Nominal
Period Day
Integer
integer
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
8
8
1=DAY 1
1 HR,
2 HR,
3 HR,
4 HR,
6 HR,
8 HR,
10 HR,
12 HR,
24 HR
1=1 HR,
2=2 HR,
3=3 HR,
4=4 HR,
6=6 HR,
8=8 HR,
10=10 HR,
12=12 HR,
24=24 HR
Source / Derivation
Populate based on EG.EGTPT.
[Note, in this example visit
windowing was not performed for
analysis.]
Numeric version of ATPT.
Numeric date value from
EG.EGDTC.
Numeric time value from
EG.EGDTC.
EG.EGDY
Sort by APERIOD, AVISITN,
ATPTN, ADY.
If first AVISITN then
AVISDY=ADY;
Retain AVISDY until next
AVISITN.
The visit day remains consistent
within a given AVISITN even when
the actual relative day crosses into
the next calendar day.
Populate with the numeric relative
day within the Period based on
Page 62
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Dataset
Name
Parameter
Identifier
Variable Name
Variable Label
Variable
Type
Display
Format
Codelist / Controlled Terms
Source / Derivation
AVISITN. [For example, “Period 2,
Day 1” is APERDAY=1]
ADEG
*ALL*
EGREPNUM
ECG Replicate
Number
integer
8
EG.EGREPNUM.
EG.EGSTRESN.
ADEG
*DEFAULT*
AVAL
Analysis Value
float
best12.
ADEG
QTCIAG,
AVAL
Analysis Value
float
best12.
HR, PRAG, RRAG, QTAG,
QTCFAG, …
ADEG
*ALL*
DTYPE
Derivation Type
text
$15
AVERAGE
ADEG
*ALL*
ABLFL
Baseline Record
Flag
text
$1
Y
ADEG
*ALL*
BASE
Baseline Value
float
best12.
Page 63
Draft
Also create a derived record to
contain the average of the three
individual replicate values for ECG
measurements collected at each time
point.
Calculate the individual QT
correction (QTCIAG) based on the
SAP Section x.x. A new derived
record will be created for each
replicate.
Also create a derived record for the
average of the individual QTcI
replicate measurements at each time
point.
DTYPE=”AVERAGE” for records
containing the average of the three
replicate values for each time point.
Otherwise, DTYPE is null.
If AVISIT=“DAY -1” and
DTYPE=“AVERAGE” then
ABLFL=“Y”.
If AVISIT=”DAY -1” and
ABLFL=”Y” then BASE=AVAL;
Else if AVISIT=”DAY 1” and
DTYPE=”AVERAGE” and
BASETYPE=“PERIOD 1, TIMEMATCH 1 HR” then BASE=AVAL
where ABLFL=“Y” and
APERIODC=”PERIOD 1” and
ATPT=“1 HR”;
Else If AVISIT=”DAY 1” and
DTYPE=”AVERAGE” and
BASETYPE=“PERIOD 1, TIMEMATCH 2 HR” then BASE=AVAL
where ABLFL=”Y” and
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Dataset
Name
ADEG
ADEG
1120
1121
1122
1123
Parameter
Identifier
*ALL*
*ALL*
Variable Name
CHG
Variable Label
Change from
Baseline
Variable
Type
float
text
Display
Format
Codelist / Controlled Terms
APERIODC=”PERIOD 1” and
ATPT=“2 HR”,
….
If AVISIT ne ‘DAY -1’ and
DTYPE=AVERAGE, then
CHG=AVAL – BASE.
best12.
$25
PERIOD 1, TIME-MATCH 1 HR,
PERIOD 1, TIME-MATCH 2 HR
PERIOD 1, TIME-MATCH 3 HR
PERIOD 1, TIME-MATCH 4 HR
….
BASETYPE
Baseline Type
text
$1
Y
Y
ADEG
QTAG, RRAG
QTCCIFL
Individ QT
Correction
Coefficient Flag
ADEG
*ALL*
ECGPCFL
ECG Matching PK
Flag
text
$1
ADEG
*ALL*
EGSEQ
Sequence Number
integer
8
ADEG
*ALL*
EGLEAD
Lead Location Used
for Measurement
text
$10
ADEG
*ALL*
EGREFID
ECG Reference ID
text
$10
Source / Derivation
Populate BASETYPE with the
APERIODC, “TIME-MATCH”, and
ATPT.
Populate with “Y” if the QTAG or
RRAG value was used to calculate
the QTCCOEFx for
PARAMCD=”QTCIAG”
Populate with “Y” if there is a
matching nominal pk sample time
point for the nominal ECG time point
EG.EGSEQ.
Set to missing if
DTYPE=“AVERAGE”or
PARAMCD=”QTCIAG”.
EG.EGLEAD
Set to missing if
DTYPE=“AVERAGE” or
PARAMCD=”QTCIAG”.
EG.EGREFID
Set to missing if
DTYPE=“AVERAGE” or
PARAMCD=”QTCIAG”.
7.1.2 Analysis Dataset
The following example illustrates the analysis dataset (ADEG) defined above. To save space, not all variables in the dataset are included. The SDTM dataset
(EG) in Section 6.1.2 Example 1 contains the input source records for the ADaM dataset.
Rows 1-3, 5-7, Show records from SDTM containing the three individual replicate values for QTAG, collected at approximately 1-minute intervals.
EGREPNUM identifies the replicate number. These records support traceability of the derived average record used for analysis. EGSEQ values
9-11, 13-15:
from SDTM are included and support traceability to records in SDTM EG.
Rows 4, 8, 12: Shows the average of the three replicates. In this example, the average is derived in ADaM so DTYPE is assigned “AVERAGE”. If the sponsor
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 64
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Rows 9-12:
Row 10:
Row 16:
1124
1125
derived the average in SDTM, then include the SDTM derived flag, EGDRVFL in ADaM so average records can be identified when both
individual and average records are submitted in the dataset.
ABLFL=“Y” indicates these records are used as baseline values. For these time-matched baselines, BASETYPE identifies the definition of
baseline and is different for each Period and time point. When BASETYPE is used for any PARAM within a dataset, it should be non-null for
all records of that dataset.
The AVAL at PERIOD 1 DAY -1, 1 HR (row 4), is used as the BASE for the post-baseline measurement at PERIOD 1 DAY 1, 1 HR (row 16).
Similarly, the AVAL at PERIOD 1 DAY -1, 2 HR (row 8), is used as the BASE for the post-baseline measurement at PERIOD 1 DAY 1, 2 HR
(row not shown). Though not illustrated, ABLFL=“Y” would also be assigned similarly for all PERIOD 1-4, DAY -1 time points.
Shows difference in ADY, AVISDY for the 24 HR time point. The variable AVISDY is different from ADY. The variable AVISDY is the
nominal relative study day for the Period Day, and remains consistent regardless of whether the actual relative day crosses into the next
calendar day. For AVISIT=DAY -1, the AVISDY=-1 for 1 HR, 2 HR, … 24 HR, but the actual relative day ADY=1 for the 24 HR time point
because it crossed into the next calendar day. Similarly, for Period 2 Day -1 (not shown), the AVISDY=3 for all time points, but ADY=4 for
the 24 HR time point.
Shows a record where the EGLEAD used is not the planned primary lead. The planned primary lead is LEAD II, however this measurement was
taken from LEAD V. The planned primary lead is specified in the proposed terminology for Trial Summary parameters in Section 4.1.4.
Shows a derived record for the average of the three post-baseline replicates at PERIOD 1 DAY 1, 1 HR. The change from baseline is calculated
as CHG = AVAL – BASE. This record is flagged as having a matching PK concentration for this time point (ECGPCFL=“Y”).
adeg.xpt
Row STUDYID USUBJID
1
2
3
4
5
6
7
8
9
10
11
12
TRTSEQA
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Additional records for 3 HR, 4 HR, etc….
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
STUDY01 2324-P0001
A 100 mg – Moxifloxacin
Page 65
Draft
TRTA
PARAMCD
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
PARAM
APERIODC APERIOD AVISIT AVISITN ATPT ATPTN
ADT
QT Interval,
PERIOD 1
1
DAY -1
-1
1 HR
1
2014-03-22
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY -1
-1
1 HR
1
2014-03-22
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY -1
-1
1 HR
1
2014-03-22
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY -1
-1
1 HR
1
2014-03-22
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY -1
-1
2 HR
2
2014-03-22
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY -1
-1
2 HR
2
2014-03-22
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY -1
-1
2 HR
2
2014-03-22
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY -1
-1
2 HR
2
2014-03-22
Aggregate (msec)
QT Interval,
Aggregate (msec)
QT Interval,
Aggregate (msec)
QT Interval,
Aggregate (msec)
QT Interval,
Aggregate (msec)
PERIOD 1
1
DAY -1
-1
24 HR
24
2014-03-23
PERIOD 1
1
DAY -1
-1
24 HR
24
2014-03-23
PERIOD 1
1
DAY -1
-1
24 HR
24
2014-03-23
PERIOD 1
1
DAY -1
-1
24 HR
24
2014-03-23
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row STUDYID USUBJID
13
STUDY01 2324-P0001
14
STUDY01 2324-P0001
15
STUDY01 2324-P0001
16
STUDY01 2324-P0001
TRTSEQA
Drug A 1 mg – Placebo - Drug
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
A 100 mg – Moxifloxacin
Drug A 1 mg – Placebo - Drug
A 100 mg – Moxifloxacin
TRTA
PARAMCD
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
Drug A 1 mg
QTAG
PARAM
APERIODC APERIOD AVISIT AVISITN ATPT ATPTN
ADT
QT Interval,
PERIOD 1
1
DAY 1
1
1 HR
1
2014-03-23
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY 1
1
1 HR
1
2014-03-23
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY 1
1
1 HR
1
2014-03-23
Aggregate (msec)
QT Interval,
PERIOD 1
1
DAY 1
1
1 HR
1
2014-03-23
Aggregate (msec)
1126
Row
ATM
ADY AVISDY APERDAY EGREPNUM AVAL
1 (cont) 10:00:21
-1
-1
-1
1
440
2 (cont) 10:01:35
-1
-1
-1
2
389
3 (cont) 10:02:14
-1
-1
-1
3
377
4 (cont)
-1
-1
-1
5 (cont) 11:00:21
-1
-1
-1
1
379
6 (cont) 11:01:31
-1
-1
-1
2
402
7 (cont) 11:02:40
-1
-1
-1
3
356
8 (cont)
-1
-1
-1
402
379
DTYPE
AVERAGE
AVERAGE
ABLFL BASE CHG
Y
Y
BASETYPE
ECGPCFL EGSEQ EGLEAD EGREFID QTCCIFL
PERIOD 1, TIME4
LEAD II 1234933
Y
MATCH 1 HR
PERIOD 1, TIMEMATCH 1 HR
8
LEAD II 1289032
Y
PERIOD 1, TIMEMATCH 1 HR
PERIOD 1, TIMEMATCH 1 HR
PERIOD 1, TIMEMATCH 2 HR
PERIOD 1, TIMEMATCH 2 HR
PERIOD 1, TIMEMATCH 2 HR
PERIOD 1, TIMEMATCH 2 HR
402
379
12
LEAD II
1343131
Y
Y
16
LEAD II
1342003
Y
20
LEAD II
1340323
Y
24
LEAD II
1343459
Y
Y
Additional records for 3 HR, 4 HR, etc….
9 (cont) 09:00:07
1
-1
-1
1
386
10 (cont) 09:01:38
1
-1
-1
2
387
11 (cont) 09:02:34
1
-1
-1
3
388
12 (cont)
1
-1
-1
13 (cont) 10:01:33
1
1
1
1
397
14 (cont) 10:02:35
1
1
1
2
403
15 (cont) 10:03:34
1
1
1
3
377
16 (cont)
1
1
1
387
392
AVERAGE
AVERAGE
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Y
387
402
-10
PERIOD 1, TIMEMATCH 24 HR
PERIOD 1, TIMEMATCH 24 HR
PERIOD 1, TIMEMATCH 24 HR
PERIOD 1, TIMEMATCH 24 HR
PERIOD 1, TIMEMATCH 1 HR
PERIOD 1, TIMEMATCH 1 HR
PERIOD 1, TIMEMATCH 1 HR
PERIOD 1, TIMEMATCH 1 HR
100
LEAD II
1347893
Y
104
LEAD V
1348832
Y
108
LEAD II
1348814
Y
Y
112
LEAD II
1234933
116
LEAD II
1289032
120
LEAD II
1343131
Y
Page 66
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1127
7.2 ADQT Example
1128
1129
1130
1131
1132
1133
1134
The ADQT dataset contains information about the QT correction model(s) used in the study. It is created when the QT correction formula and coefficients are
derived by the sponsor or vendor, and not created for commonly used historical correction models such as Bazett or Fridericia. The model information could be
for one or more specific models specified in the protocol, or one or more ‘best fit’ models from a series of models. Additional details regarding the kinds of QT
correction models can be found in Section 2.3.4 and Section 6.2. If the sponsor derives the QT corrections, the calculations would usually be performed at the
ADaM stage, and the model information would be submitted in ADaM ADQT and not in SDTM QT. If the vendor derives the QT corrections, the model
information would be stored in the SDTM QT domain as specified in Section 6.2, and may be carried forward into ADQT, depending on the sponsor’s needs.
The ADQT dataset is based on the ADaM Basic Data Structure (BDS).
1135
7.2.1 Analysis Data Metadata
1136
Analysis Dataset Metadata for ADQT Example
1137
1138
1139
1140
1141
1142
1143
1144
1145
1146
1147
Dataset Name
Dataset Description
Dataset Location
ADQT
Analysis Dataset for
ECG QTc Model Data
adqt.xpt
Key Variables of
Dataset
Dataset Structure
Class of Dataset
Documentation
BDS
adqt.sas
One record per subject,
STUDYID, USUBJID,
per QT model, per QT
AGRPID, PARAMCD
model parameter
In the metadata illustrated in the table below, the parameter identifier of “*ALL*” is used for variables expected to be consistent across analysis parameters,
“*DEFAULT*” is used for parameters not otherwise specified in the Parameter Identifier column for that variable. The value of PARAMCD as the parameter
identifier is used for variables that have metadata dependent on the analysis parameter.
In this example, the ADQT dataset contains the top two ‘best fit’ QT correction models for each subject and a historical model used for all subjects. For each
subject, the baseline QT/RR data was fit into twelve different protocol-specified potential models, from which the top two best fit models will be used for
calculating QTcI corrections. For the historical model, the QT formula and coefficients are the same across all subjects; however, the sponsor chose to include
the model information in the dataset for each subject.
Analysis Variable Metadata for ADQT Example
Dataset
Name
ADQT
Parameter
Identifier
*ALL*
ADQT
*ALL*
USUBJID
ADQT
QTCDESC
PARAM
ADQT
QTCFORM
PARAM
Page 67
Draft
Variable Name
Variable Label
STUDYID
Study Identifier
Unique Subject
Identifier
Variable
Type
text
Display
Format
Codelist / Controlled
Terms
Source / Derivation
$12
ADSL.STUDYID
text
$20
ADSL.USUBJID
Parameter
text
$50
QT Correction Method
Description
Parameter
text
$50
QT Correction Formula
For records created to contain the QT
Correction method, populate with “QT
Correction Method Description”
For records created to contain the QT
Correction formula, populate with “QT
Correction Formula”
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Dataset
Name
1148
1149
1150
1151
1152
1153
1154
1155
1156
1157
1158
1159
1160
Parameter
Identifier
Variable Name
Variable Label
Variable
Type
Display
Format
Codelist / Controlled
Terms
ADQT
QTCCOEFx
PARAM
Parameter
text
$50 QT Correction Coefficient x
ADQT
*ALL*
PARAMCD
Parameter Code
text
$8
ADQT
QTCCOEFx
AVAL
Analysis Value
float
best12.
ADQT
QTCDESC
AVALC
Analysis Value (C)
text
$50
ADQT
QTCFORM
AVALC
Analysis Value (C)
text
$50
ADQT
*ALL*
AGRPID
Group ID
text
$8
QTCDESC, QTCFORM,
QTCCOEFx
QTCIAG1, QTCIAG2,
QTCNAG
Source / Derivation
For records created to contain the QT
correction coefficient x, populate with
“QT Correction Coefficient x”
One-to-one correspondence with
PARAM
Create one or more records containing
the individual or historical correction
coefficient used to calculate the QTc
values specified in AGRPID
Set AVALC to the model description
used to calculate the QTc values
specified in AGRPID
Set AVALC to the model formulas
used to calculate the QTc values
specified in AGRPID
Based on the protocol, assign AGRPID
to the name of the QTc parameters to
be calculated for each subject
7.2.2 Analysis Dataset
In this example, the variable AGRPID identifies the QT corrected parameter code (PARAMCD) in ADEG whose values are based on the model information.
Variables that do not apply for the AVAL are not included in the dataset (e.g. AVISIT, ADT, ATM, etc.).
Rows 1-3:
Rows 4-6:
Rows 7-11:
Rows 12-22:
Row
1
2
3
4
5
6
Shows records containing model information for the first best fit model for USUBJID=2324-P0001. This model contains one correction
coefficient 0.266, which was used in the LINEAR model formula to calculate the subject’s QTc values for PARAMCD=”QTCIAG1” in ADEG.
Shows records containing model information for the second best fit model for this subject. This model contains one correction coefficient 0.532,
which was used in the PARABOLIC model formula to calculate the subject’s QTc values for PARAMCD=”QTCIAG2” in ADEG.
Shows records containing model information for the historical model, RAUTAHARJU COR, that contains three correction coefficients. The
coefficients 0.2425, 0.434, and -0.0097 are constant values across all subjects and used to calculate PARAMCD=QTCNAG in ADEG.
Shows records for USUBJID=2324-P0002. The top two best fit models, LOGARITHMIC and LINEAR, and the correction coefficients are
different than for USUBJID=2324-P0001. However, the historical model information is the same.
STUDYID
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
USUBJID
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
PARAMCD
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
PARAM
QT Correction Method Description
QT Correction Formula
QT Correction Coefficient A
QT Correction Method Description
QT Correction Formula
QT Correction Coefficient A
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
AVAL
AVALC
LINEAR
QTC=QT+(QTCCOEFA*(1-RR))
0.266
PARABOLIC
QTC=QT/(RR^QTCCOEFA)
0.532
AGRPID
QTCIAG1
QTCIAG1
QTCIAG1
QTCIAG2
QTCIAG2
QTCIAG2
Page 68
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row
7
STUDYID
STUDY01
USUBJID
2324-P0001
PARAMCD
QTCDESC
PARAM
QT Correction Method Description
8
STUDY01
2324-P0001
QTCFORM
QT Correction Formula
9
10
11
12
13
14
15
16
17
18
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
STUDY01
2324-P0001
2324-P0001
2324-P0001
2324-P0002
2324-P0002
2324-P0002
2324-P0002
2324-P0002
2324-P0002
2324-P0002
QTCCOEFA
QTCCOEFB
QTCCOEFC
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QTCFORM
QTCCOEFA
QTCDESC
QT Correction Coefficient A
QT Correction Coefficient B
QT Correction Coefficient C
QT Correction Method Description
QT Correction Formula
QT Correction Coefficient A
QT Correction Method Description
QT Correction Formula
QT Correction Coefficient A
QT Correction Method Description
2324-P0002
QTCFORM
QT Correction Formula
2324-P0001
2324-P0001
2324-P0001
QTCCOEFA
QTCCOEFB
QTCCOEFC
QT Correction Coefficient A
QT Correction Coefficient B
QT Correction Coefficient C
19
20
21
22
STUDY01
STUDY01
STUDY01
AVAL
AVALC
RAUTAHARJU COR
QTC=QT+QTCCOEFA(QTCCOEFB*(e^(QTCCOEFC^HR))
0.2425
0.434
-0.0097
LOGARITHMIC
QTC=QT-QTCCOEFA*ln(RR)
0.116
LINEAR
QTC=QT+(QTCCOEFA*(1-RR))
0.387
RAUTAHARJU COR
QTC=QT+QTCCOEFA(QTCCOEFB*(e^(QTCCOEFC^HR))
0.2425
0.434
-0.0097
AGRPID
QTCNAG
QTCNAG
QTCNAG
QTCNAG
QTCNAG
QTCIAG1
QTCIAG1
QTCIAG1
QTCIAG2
QTCIAG2
QTCIAG2
QTCNAG
QTCNAG
QTCNAG
QTCNAG
QTCNAG
1161
7.3 Baseline Alternatives and Definition of Treatment Differences
1162
1163
1164
1165
In this section, four different baseline alternatives, including the possibility of no baseline, are described. For each baseline method, the resulting definition of
treatment differences is described. Note that this is not an exhaustive list of possibilities. For example, triple-delta (∆∆∆QTc) definitions are possible where both
lead-in day ECGs are collected at matched time points to the time points of collection on the treatment day and an ECG replicate set is collected immediately
before treatment administration (and at the same time point on the lead-in day).
1166
7.3.1 Time-Matched Baseline; Double-Delta Treatment Difference
1167
1168
1169
1170
For time-matched baseline, the baseline for each period is the average of values at a time point on the baseline day (Day -1) that corresponds to the post-dose
time point. ECGs are collected or extracted from continuous recording in replicate sets (usually 3 replicates about a minute or so apart) at each bj and Xij. The
average of the replicates is presented in the table below.
1171
1172
1173
1174
For crossover design, ΔΔQTcij is computed for each subject: ΔΔQTcij = �Xij − bj �
and n=time point.
Page 69
Draft
Drug A
− �Xij − bj �
Placebo
where i=1, 2, … d, j=1, 2, … n; d=days postdose
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1175
1176
1177
1178
1179
1180
1181
1182
1183
1184
1185
1186
1187
1188
1189
1190
1191
Row USUBJID
TRTA
PARAMCD APERIOD AVISIT ATPT AVAL* ABLFL BASE*
1
DAY -1 1 HR b1(Drug A)
Y
b1(Drug A)
1 2324-P0001 Drug A 1 mg QTCFAG
1
DAY -1 2 HR b2(Drug A)
Y
b2(Drug A)
2 2324-P0001 Drug A 1 mg QTCFAG
1
DAY -1 3 HR b3(Drug A)
Y
b3(Drug A)
3 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1 1 HR X11(Drug A)
b1(Drug A)
4 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1 2 HR X12(Drug A)
b2(Drug A)
5 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1 3 HR X13(Drug A)
b3(Drug A)
6 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2 1 HR X21(Drug A)
b1(Drug A)
7 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2 2 HR X22(Drug A)
b2(Drug A)
8 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2 3 HR X23(Drug A)
b3(Drug A)
9 2324-P0001 Drug A 1 mg QTCFAG
Placebo
QTCFAG
2
DAY -1 1 HR b1(Placebo)
Y
b1(Placebo)
10 2324-P0001
Placebo
QTCFAG
2
DAY -1 2 HR b2(Placebo)
Y
b2(Placebo)
11 2324-P0001
Placebo
QTCFAG
2
DAY -1 3 HR b3(Placebo)
Y
b3(Placebo)
12 2324-P0001
Placebo
QTCFAG
2
DAY 1 1 HR X11(Placebo)
b1(Placebo)
13 2324-P0001
Placebo
QTCFAG
2
DAY 1 2 HR X12(Placebo)
b2(Placebo)
14 2324-P0001
Placebo
QTCFAG
2
DAY 1 3 HR X13(Placebo)
b3(Placebo)
15 2324-P0001
Placebo
QTCFAG
2
DAY 2 1 HR X21(Placebo)
b1(Placebo)
16 2324-P0001
Placebo
QTCFAG
2
DAY 2 2 HR X22(Placebo)
b2(Placebo)
17 2324-P0001
Placebo
QTCFAG
2
DAY 2 3 HR X23(Placebo)
b3(Placebo)
18 2324-P0001
*Values in these columns are mathematical expressions. See the next table for example submission values.
����������
For a parallel design, (Xij – bj) would be averaged across subjects: ΔΔQTcij = �X
ıȷ − bȷ �
7.3.1.1 Analysis Dataset Example for Time-Matched Baseline
CHG*
(X11-b1)Drug A
(X12-b2)Drug A
(X13-b3)Drug A
(X21-b1)Drug A
(X22-b2)Drug A
(X23-b3)Drug A
(X11-b1)Placebo
(X12-b2)Placebo
(X13-b3)Placebo
(X21-b1)Placebo
(X22-b2)Placebo
(X23-b3)Placebo
Drug A
BASETYPE
PERIOD 1, TIME-MATCH 1 HR
PERIOD 1, TIME-MATCH 2 HR
PERIOD 1, TIME-MATCH 3 HR
PERIOD 1, TIME-MATCH 1 HR
PERIOD 1, TIME-MATCH 2 HR
PERIOD 1, TIME-MATCH 3 HR
PERIOD 1, TIME-MATCH 1 HR
PERIOD 1, TIME-MATCH 2 HR
PERIOD 1, TIME-MATCH 3 HR
PERIOD 2, TIME-MATCH 1 HR
PERIOD 2, TIME-MATCH 2 HR
PERIOD 2, TIME-MATCH 3 HR
PERIOD 2, TIME-MATCH 1 HR
PERIOD 2, TIME-MATCH 2 HR
PERIOD 2, TIME-MATCH 3 HR
PERIOD 2, TIME-MATCH 1 HR
PERIOD 2, TIME-MATCH 2 HR
PERIOD 2, TIME-MATCH 3 HR
����������
− �X
ıȷ − bȷ �
Placebo
The below table illustrates time-matched baseline and post-baseline records with example values.
Rows 1-3:
Row 4:
Row 4, 8:
Row 12, 16:
Row
1
2
3
4
5
6
USUBJID
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
Shows records from SDTM containing the three individual replicate values for QTCFAG, collected at approximately 1-minute intervals.
EGREPNUM identifies the replicate number. These records support traceability of the derived average record used for analysis.
Shows a derived record with the average of the three replicates.
ABLFL=”Y” indicates these records are used as baseline values. For these time-matched baselines, BASETYPE identifies the definition of
baseline, and is different for each Period and time point. The AVAL at PERIOD 1 DAY -1, 1 HR (row 4), is used as the BASE for the postbaseline measurement at PERIOD 1 DAY 1, 1 HR (row 12). Similarly, the AVAL at PERIOD 1 DAY -1, 2 HR (row 8), is used as the BASE for
the post-baseline measurement at PERIOD 1 DAY 1, 2 HR (row 16). Though not illustrated, ABLFL=”Y” would also be assigned similarly for all
other PERIOD 1-4, DAY -1 time points.
The change from baseline, CHG = AVAL – BASE, is calculated for the average of the replicate values. BASETYPE identifies which baseline
value corresponds to the value of BASE in that row.
TRTA
PARAMCD APERIOD AVISIT
Drug A 1 mg QTCFAG
1
DAY -1
Drug A 1 mg QTCFAG
1
DAY -1
Drug A 1 mg QTCFAG
1
DAY -1
Drug A 1 mg QTCFAG
1
DAY -1
Drug A 1 mg QTCFAG
1
DAY -1
Drug A 1 mg QTCFAG
1
DAY -1
ATPT EGREPNUM AVAL DTYPE ABLFL BASE CHG
BASETYPE
1 HR
1
387
PERIOD 1, TIME-MATCH 1 HR
1 HR
2
357
PERIOD 1, TIME-MATCH 1 HR
1 HR
3
364
PERIOD 1, TIME-MATCH 1 HR
1 HR
369 AVERAGE
Y
369
PERIOD 1, TIME-MATCH 1 HR
2 HR
1
361
PERIOD 1, TIME-MATCH 2 HR
2 HR
2
376
PERIOD 1, TIME-MATCH 2 HR
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 70
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row
7
8
9
10
11
12
13
14
15
16
1192
1193
1194
1195
1196
1197
1198
1199
1200
1201
USUBJID
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
TRTA
PARAMCD APERIOD AVISIT
Drug A 1 mg QTCFAG
1
DAY -1
Drug A 1 mg QTCFAG
1
DAY -1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
ATPT EGREPNUM AVAL DTYPE ABLFL BASE CHG
BASETYPE
2 HR
3
409
PERIOD 1, TIME-MATCH 2 HR
2 HR
382 AVERAGE
Y
382
PERIOD 1, TIME-MATCH 2 HR
1 HR
1
400
PERIOD 1, TIME-MATCH 1 HR
1 HR
2
388
PERIOD 1, TIME-MATCH 1 HR
1 HR
3
379
PERIOD 1, TIME-MATCH 1 HR
1 HR
389 AVERAGE
369
20 PERIOD 1, TIME-MATCH 1 HR
2 HR
1
396
PERIOD 1, TIME-MATCH 2 HR
2 HR
2
384
PERIOD 1, TIME-MATCH 2 HR
2 HR
3
402
PERIOD 1, TIME-MATCH 2 HR
2 HR
394 AVERAGE
382
12 PERIOD 1, TIME-MATCH 2 HR
7.3.2 Time-Averaged Baseline; Double-Delta Treatment Difference
For time-averaged baseline, the baseline for each period is the average of all values on the baseline day (e.g. Day -1, 1 hr, 2 hr, 3 hr, 4 hr, etc.). ECGs are
collected or extracted from continuous recording in replicate sets (usually 3 replicates about a minute or so apart) at each bj and Xij. The average of the replicates
is presented in the table below.
For crossover design, ΔΔQTcij is computed for each subject: ΔΔQTcij = �Xij − bavg �
d = days postdose and n = time point.
Drug A
Row USUBJID
TRTA
PARAMCD APERIOD AVISIT
ATPT
AVAL* ABLFL
1
DAY -1 BASELINE bavg(Drug A)
Y
1 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1
1 HR
X11(Drug A)
2 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1
2 HR
X12(Drug A)
3 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1
3 HR
X13(Drug A)
4 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2
1 HR
X21(Drug A)
5 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2
2 HR
X22(Drug A)
6 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2
3 HR
X23(Drug A)
7 2324-P0001 Drug A 1 mg QTCFAG
Placebo
QTCFAG
2
DAY -1 BASELINE bavg (Placebo)
Y
8 2324-P0001
Placebo
QTCFAG
2
DAY 1
1 HR
X11(Placebo)
9 2324-P0001
Placebo
QTCFAG
2
DAY 1
2 HR
X12(Placebo)
10 2324-P0001
Placebo
QTCFAG
2
DAY 1
3 HR
X13(Placebo)
11 2324-P0001
Placebo
QTCFAG
2
DAY 2
1 HR
X21(Placebo)
12 2324-P0001
Placebo
QTCFAG
2
DAY 2
2 HR
X22(Placebo)
13 2324-P0001
Placebo
QTCFAG
2
DAY 2
3 HR
X23(Placebo)
14 2324-P0001
*Values in these columns are mathematical expressions. See the next table for example submission values.
Page 71
Draft
− �Xij − bavg �
BASE*
bavg(Drug A)
bavg(Drug A)
bavg(Drug A)
bavg(Drug A)
bavg(Drug A)
bavg(Drug A)
bavg(Drug A)
bavg(Placebo)
bavg(Placebo)
bavg(Placebo)
bavg(Placebo)
bavg(Placebo)
bavg(Placebo)
bavg(Placebo)
Placebo
CHG*
(X11-bavg)Drug A
(X12-bavg)Drug A
(X13-bavg)Drug A
(X21-bavg)Drug A
(X22-bavg)Drug A
(X23-bavg)Drug A
(X11-bavg)Placebo
(X12-bavg)Placebo
(X13-bavg)Placebo
(X21-bavg)Placebo
(X22-bavg)Placebo
(X23-bavg)Placebo
where bavg = ∑ bj /n; i=1, 2, … d, j=1, 2, … n;
BASETYPE
PERIOD 1, TIME-AVG
PERIOD 1, TIME-AVG
PERIOD 1, TIME-AVG
PERIOD 1, TIME-AVG
PERIOD 1, TIME-AVG
PERIOD 1, TIME-AVG
PERIOD 1, TIME-AVG
PERIOD 2, TIME-AVG
PERIOD 2, TIME-AVG
PERIOD 2, TIME-AVG
PERIOD 2, TIME-AVG
PERIOD 2, TIME-AVG
PERIOD 2, TIME-AVG
PERIOD 2, TIME-AVG
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1202
1203
1204
1205
1206
1207
1208
1209
1210
1211
1212
1213
1214
1215
�������������
For a parallel design, (Xij – bavg) would be averaged across subjects: ΔΔQTcij = �X
ıȷ − bavg �
Drug A
The below table illustrates time-averaged baseline and post-baseline records with example values.
Rows 1-3:
Rows 4, 8:
Rows 9:
Row 13, 17:
Row
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
1216
1217
1218
1219
1220
USUBJID
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
�������������
− �X
ıȷ − bavg �
Placebo
Shows three individual replicate values for QTCFAG, collected at approximately 1-minute intervals. EGREPNUM identifies the replicate
number.
Shows a derived record with the average of the three replicates.
The baseline for PERIOD 1 is based on the average of the replicate averages for 1 HR (row 4), 2 HR (row 8), and 3 HR, 4 HR, 6 HR, 8 HR, 10
HR, 12 HR, and 24 HR replicate averages (not shown). ABLFL=”Y” indicates this contains the baseline value for this BASETYPE. The AVAL
(“370”) is the value of BASE (row 9) and the value of BASE for all rows that have the same BASETYPE (row 13, 17, and additional rows not
shown). The records for BASETYPE=”PERIOD 2, TIME-AVG” are not shown.
The change from baseline, CHG = AVAL – BASE, is calculated for the average of the replicate values.
TRTA
PARAMCD APERIOD AVISIT
ATPT
EGREPNUM AVAL DTYPE ABLFL BASE CHG
BASETYPE
Drug A 1 mg QTCFAG
1
DAY -1
1 HR
1
387
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1
1 HR
2
357
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1
1 HR
3
364
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1
1 HR
369 AVERAGE
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1
2 HR
1
361
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1
2 HR
2
376
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1
2 HR
3
409
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1
2 HR
382 AVERAGE
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY -1 BASELINE
370 AVERAGE
Y
370
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
1
400
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
2
388
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
3
379
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
389 AVERAGE
370
19 PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
1
396
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
2
384
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
3
402
PERIOD 1, TIME-AVG
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
394 AVERAGE
370
24 PERIOD 1, TIME-AVG
7.3.3 Predose Averaged Baseline; Double-Delta Treatment Difference
For predose averaged baseline, ECGs are collected or extracted as replicate sets (usually 3 replicates about a minute or less apart) at predose in close temporal
proximity to treatment administration (e.g., 15 min intervals and immediately before treatment administration on the same day of treatment administration) and
as replicate sets (usually 3 replicates about a minute or so apart) at each Xij post dose. The average of the replicates is presented in the table below.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 72
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1221
1222
1223
1224
1225
1226
1227
1228
1229
1230
1231
1232
1233
1234
For crossover design, ΔΔQTcij is computed for each subject: ∆∆QTcij = �Xij − b0 �
days postdose, k = number of predose replicates, n = time point.
Drug A
− �Xij − b0 �
Placebo
Row USUBJID
TRTA
PARAMCD APERIOD AVISIT
ATPT
AVAL* BASE* ABLFL
CHG*
1
DAY 1 BASELINE b0(Drug A) b0(Drug A)
Y
1 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1
1 HR
X11(Drug A) b0(Drug A)
(X11-b0)Drug A
2 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1
2 HR
X12(Drug A) b0(Drug A)
(X12-b0)Drug A
3 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 1
3 HR
X13(Drug A) b0(Drug A)
(X13-b0)Drug A
4 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2
1 HR
X21(Drug A) b0(Drug A)
(X21-b0)Drug A
5 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2
2 HR
X22(Drug A) b0(Drug A)
(X22-b0)Drug A
6 2324-P0001 Drug A 1 mg QTCFAG
1
DAY 2
3 HR
X23(Drug A) b0(Drug A)
(X23-b0)Drug A
7 2324-P0001 Drug A 1 mg QTCFAG
Placebo
QTCFAG
2
DAY 1 BASELINE b0(Placebo) b0(Placebo)
Y
8 2324-P0001
Placebo
QTCFAG
2
DAY 1
1 HR
X11(Placebo) b0(Placebo)
(X11-b0)Placebo
9 2324-P0001
Placebo
QTCFAG
2
DAY 1
2 HR
X12(Placebo) b0(Placebo)
(X12-b0)Placebo
10 2324-P0001
Placebo
QTCFAG
2
DAY 1
3 HR
X13(Placebo) b0(Placebo)
(X13-b0)Placebo
11 2324-P0001
Placebo
QTCFAG
2
DAY 2
1 HR
X21(Placebo) b0(Placebo)
(X21-b0)Placebo
12 2324-P0001
Placebo
QTCFAG
2
DAY 2
2 HR
X22(Placebo) b0(Placebo)
(X22-b0)Placebo
13 2324-P0001
Placebo
QTCFAG
2
DAY 2
3 HR
X23(Placebo) b0(Placebo)
(X23-b0)Placebo
14 2324-P0001
*Values in these columns are mathematical expressions. See the next table for example submission values.
�����������
For a parallel design, (Xij – bj) would be averaged across subjects: ΔΔQTcij = �X
ıȷ − b0 �
Drug A
where b0 = ∑ bj /k; i=1, 2, … d, j=1, 2, … n; d =
BASETYPE
PERIOD 1, PREDOSE AVG
PERIOD 1, PREDOSE AVG
PERIOD 1, PREDOSE AVG
PERIOD 1, PREDOSE AVG
PERIOD 1, PREDOSE AVG
PERIOD 1, PREDOSE AVG
PERIOD 1, PREDOSE AVG
PERIOD 2, PREDOSE AVG
PERIOD 2, PREDOSE AVG
PERIOD 2, PREDOSE AVG
PERIOD 2, PREDOSE AVG
PERIOD 2, PREDOSE AVG
PERIOD 2, PREDOSE AVG
PERIOD 2, PREDOSE AVG
�����������
− �X
ıȷ − b0 �
Placebo
The below table is an example ADEG dataset illustrating predose averaged baseline and post-baseline records.
Rows 4, 8, 12: Shows the average of three replicate QTCFAG measurements taken within one hour prior to dose which are used to calculate the baseline.
Rows 13:
Shows a derived record with the average of rows 4, 8, and 12. This value is the predose average baseline (ABLFL=”Y”).
Rows 17, 21: The change from baseline, CHG = AVAL – BASE, is calculated for the average of the replicate values.
Row
1
2
3
4
5
6
7
8
USUBJID
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
Page 73
Draft
TRTA
PARAMCD APERIOD AVISIT
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
Drug A 1 mg QTCFAG
1
DAY 1
ATPT
-45 MIN
-45 MIN
-45 MIN
-45 MIN
-30 MIN
-30 MIN
-30 MIN
-30 MIN
EGREPNUM AVAL DTYPE BASE ABLFL CHG
BASETYPE
1
380
PERIOD 1, PREDOSE AVG
2
381
PERIOD 1, PREDOSE AVG
3
382
PERIOD 1, PREDOSE AVG
381 AVERAGE
PERIOD 1, PREDOSE AVG
1
378
PERIOD 1, PREDOSE AVG
2
379
PERIOD 1, PREDOSE AVG
3
380
PERIOD 1, PREDOSE AVG
379 AVERAGE
PERIOD 1, PREDOSE AVG
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Row
9
10
11
12
13
14
15
16
17
18
19
20
21
1235
1236
1237
1238
1239
1240
1241
1242
1243
1244
1245
1246
USUBJID
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
2324-P0001
TRTA
PARAMCD APERIOD AVISIT
ATPT
EGREPNUM AVAL DTYPE BASE ABLFL CHG
BASETYPE
Drug A 1 mg QTCFAG
1
DAY 1
-15 MIN
1
395
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
-15 MIN
2
393
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
-15 MIN
3
394
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
-15 MIN
394 AVERAGE
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1 BASELINE
385 AVERAGE 385
Y
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
1
400
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
2
388
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
3
379
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
1 HR
389 AVERAGE 385
4
PERIOD 1, PREDOSE AVG
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
1
396
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
2
384
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
3
402
PERIOD 1, PREDOSE AVG
Drug A 1 mg QTCFAG
1
DAY 1
2 HR
394 AVERAGE 385
9
PERIOD 1, PREDOSE AVG
7.3.4 No Baseline; Single-Delta Treatment Difference
ECGs are collected or extracted from continuous recordings in replicate sets (usually 3 replicates about a minute or so apart) at each Xij. The average of the
replicates is presented in the table below.
This definition of a treatment difference is rare and is not recommended, particularly not recommended for parallel studies.
For crossover design, ΔQTcij is computed for each subject: ∆QTcij = Xij(Drug A) − Xij(Placebo) where i=1, 2, … d, j=1, 2, … n; d=days postdose and n=time point.
USUBJID
TRTA
PARAMCD
APERIOD
2324-P0001
Drug A 1 mg QTCFAG
1
2324-P0001
Drug A 1 mg QTCFAG
1
2324-P0001
Drug A 1 mg QTCFAG
1
2324-P0001
Drug A 1 mg QTCFAG
1
2324-P0001
Drug A 1 mg QTCFAG
1
2324-P0001
Drug A 1 mg QTCFAG
1
2324-P0001
Placebo
QTCFAG
2
2324-P0001
Placebo
QTCFAG
2
2324-P0001
Placebo
QTCFAG
2
2324-P0001
Placebo
QTCFAG
2
2324-P0001
Placebo
QTCFAG
2
2324-P0001
Placebo
QTCFAG
2
*Values in this column are mathematical expressions.
AVISIT
DAY 1
DAY 1
DAY 1
DAY 2
DAY 2
DAY 2
DAY 1
DAY 1
DAY 1
DAY 2
DAY 2
DAY 2
ATPT
1 HR
2 HR
3 HR
1 HR
2 HR
3 HR
1 HR
2 HR
3 HR
1 HR
2 HR
3 HR
AVAL*
X11(Drug A)
X12(Drug A)
X13(Drug A)
X21(Drug A)
X22(Drug A)
X23(Drug A)
X11(Placebo)
X12(Placebo)
X13(Placebo)
X21(Placebo)
X22(Placebo)
X23(Placebo)
X ij(Drug A) − �
X ij(Placebo)
For a parallel design, the Xijs would be averaged across subjects: ∆QTcij = �
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 74
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1247
Appendices
1248
Appendix A: Project Proposal
1249
1250
1251
1252
1253
1254
1255
1256
1257
1258
The project began in Q2 2013, with a target completion of the TAUG-QT in Q3 2014.
1259
Appendix B: CFAST Organizations
1260
1261
1262
1263
1264
1265
1266
1267
1268
1269
1270
1271
1272
1273
1274
1275
1276
1277
1278
1279
1280
1281
1282
1283
1284
1285
1286
1287
1288
1289
1290
1291
1292
1293
1294
1295
CFAST is proposing development of v1.0 of the CDISC QT therapeutic area data standard. This standard will build
on the existing SDTM to facilitate the collection and use of data relevant to QT clinical trials.
The workgroup proposes developing a CDISC therapeutic area user guide, including concept maps, examples and
controlled terminology. It is clear that the existing CDISC standards are sufficient to cover the data standard
requirements. The user guide is expected to focus on how these existing standards are to be used in the QT setting
via examples and guidelines. Some investigation into the current ECG terminology and FDA subject level analysis
dataset requirements related to QT may be required.
CDISC
CDISC is a global, open, multidisciplinary, non-profit organization that has established standards to support the
acquisition, exchange, submission and archive of clinical research data and metadata. The CDISC mission is to
develop and support global, platform-independent data standards that enable information system interoperability to
improve medical research and related areas of healthcare. CDISC standards are vendor-neutral, platformindependent and freely available via the CDISC website.
Critical Path Institute
An independent, non-profit organization established in 2005 with public and private philanthropic support from the
Arizona community, Science Foundation Arizona, and the U.S. Food and Drug Administration, C-Path’s mission is
to improve human health and well-being by developing new technologies and methods to accelerate the
development and review of medical products. An international leader in forming collaborations, C-Path has
established global, public-private partnerships that currently include 1,000+ scientists from government regulatory
agencies, academia, patient advocacy organizations, and dozens of major pharmaceutical companies.
Association of Clinical Research Organizations
The Association of Clinical Research Organizations (ACRO) represents the world's leading clinical research
organizations. ACRO members provide specialized services that are integral to the development of drugs, biologics
and medical devices. ACRO advances clinical outsourcing to improve the quality, efficiency and safety of
biomedical research. Each year, ACRO’s members conduct thousands of clinical trials and provide related drug
development services in more than 115 countries while ensuring the safety of nearly 2 million research participants.
Innovative Medicines Initiative
The Innovative Medicines Initiative (IMI) is Europe's largest public-private partnership aiming to improve the drug
development process by supporting a more efficient discovery and development of better and safer medicines for
patients. IMI supports collaborative research projects and builds networks of industrial and academic experts in
Europe that will boost innovation in healthcare. Acting as a neutral third party in creating innovative partnerships,
IMI aims to build a more collaborative ecosystem for pharmaceutical research and development. IMI supports
research projects in the areas of safety and efficacy, knowledge management and education and training.
National Cancer Institute Enterprise Vocabulary Services
Since 1997, NCI Enterprise Vocabulary Services (EVS) has provided terminology content, tools, and services to
accurately code, analyze and share cancer and biomedical research, clinical and public health information. EVS
works with many partners to develop, license and publish terminology, jointly develop software tools, and support
harmonization and shared standards. EVS provides the foundational layer for NCI's informatics infrastructure, and
plays an important role in federal and international standards efforts.
Page 75
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1296
1297
1298
1299
1300
1301
1302
1303
1304
1305
1306
1307
1308
1309
U.S. Food and Drug Administration
The FDA is an agency of the United States Department of Health and Human Services. FDA is responsible for
protecting the public health by assuring the safety, efficacy and security of human and veterinary drugs, biological
products, medical devices, our nation’s food supply, cosmetics, and products that emit radiation. FDA is also
responsible for advancing the public health by helping to speed innovations that make medicines more effective,
safer, and more affordable and by helping the public get the accurate, science-based information they need to use
medicines and foods to maintain and improve their health.
1310
Appendix C: Workgroup
1311
TransCelerate BioPharma, Inc.
Launched in September 2012, TransCelerate BioPharma, Inc. aims to simplify and accelerate the delivery of
innovative medicines to patients. The TCB mission is to collaborate across the global biopharmaceutical R&D
community in order to identify, prioritize, design and facilitate implementation of solutions designed to drive the
efficient, effective and high quality delivery of new medicines.
Name
Rhonda Facile, Team Leader
Charles Beasley
Cathy Bezek
Natalie Boone
Katie Carothers
Marty Cisneroz
Julie Evans
Bala Hosmane
Donna Kowalski
Jordan Li
Erin Muhlbradt
Jaya Natarajan
John Owen
Dianne Reeves
Pamela Rinaldi
Klaus Romero
Darcy Wold
Diane Wold
Fred Wood
Sherry Wunder
Institution/Organization
CDISC
Eli Lilly and Company
Astellas Pharma Global Development, Inc.
Astellas Pharma Global Development, Inc.
CDISC
C-Path
CDISC
AbbVie
Astellas Pharma Global Development, Inc.
NIH
NCI EVS
Janssen Research & Development
Janssen Research & Development
NCI EVS
Boehringer Ingelheim
C-Path
CDISC
GlaxoSmithKline
Accenture
Janssen Research & Development
Ron Fitzmartin
FDA
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 76
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1312
Appendix D: Glossary and Abbreviations
Term/Abbreviation
ADaM
BRIDG
CDASH
CDISC
CDSA
CFAST
C-Path
Collected
Controlled
Terminology
CRF
CRO
CV
CV SAC
Domain
ECG/EKG
eCRF
FDA
Foundational
Standards
GLS
HR
MedDRA
mm
msec
mV
Patient
NCI EVS
NIH
QTc
QTcB
QTcF
SDS
SDTM
SDTMIG
SHARE
Subject
TCB
UG
UML
1313
1314
Definition
Analysis Data Model
Biomedical Research Integrated Domain Group
Clinical Data Acquisition Standards Harmonization Project
Clinical Data Interchange Standards Consortium
Clinical Diagnostic Services Associate
Coalition for Accelerating Standards and Therapies
Critical Path Institute
“Collected” refers to information that is recorded and/or transmitted to the sponsor. This
includes data entered by the site on CRFs/eCRFs as well as vendor data such as core lab
data. This term is a synonym for “captured”.
A finite set of values that represent the only allowed values for a data item. These values
may be codes, text, or numeric. A codelist is one type of controlled terminology.
Case Report Form (sometimes called a Case Record Form). A printed, optical, or electronic
document designed to record all required information to be reported to the sponsor for each
trial subject.
Clinical research organization
Cardiovascular
Cardiovascular Safety Advisory Committee
A collection of observations with a topic-specific commonality about a subject.
Electrocardiogram
Electronic Case Report Form
U.S. Food and Drug Administration
“Foundational standards” is the term used to refer to the suite of CDISC standards that
describe the clinical study protocol (Protocol), design (Study Design), data collection
(CDASH), laboratory work (Lab), analysis (ADaM), and data tabulation (SDTM and
SEND). See http://www.cdisc.org/ for more information on each of these clinical data
standards.
Generic laboratory system
Heart rate
Medical Dictionary for Regulatory Activities. Global standard medical terminology
designed to supersede other terminologies (such as COSTART and ICD9) used in the
medical product development process.
millimeter
millisecond
microvolt
A recipient of medical attention.
National Cancer Institute (NCI) Enterprise Vocabulary Services
National Institutes of Health
QT interval corrected for heart rate
QT interval corrected for heart rate using Bazett’s formula
QT interval corrected for heart rate using Fridericia’s formula
Submission Data Standards. Also the name of the team that created the SDTM and SDTMIG
Study Data Tabulation Model
SDTM Implementation Guide (for Human Clinical Trials)
CDISC’s metadata repository that is currently under development.
A participant in a study.
TransCelerate BioPharma Inc.
User Guide
Unified Modeling Language
See also Section 2.2, Explanation of Common Terms.
Page 77
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
1315
Appendix E: References
1. International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for
Human Use. E14: The Clinical Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential for
Non-antiarrhythmic Drugs. ICH. May 12, 2005. Available at:
http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E14/E14_Guideline.pdf.
Accessed April 11, 2014.
2. Committee for Medicinal Products for Human Use. ICH topic E14: the clinical evaluation of QT/QTc interval
prolongation and proarrhythmic potential for non-antiarrhythmic drugs questions and answers
(R1)(EMA/CHMP/ICH/310133/2008). EMA. May 2012. Available at:
http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500002878.pdf.
Accessed March 18, 2014.
3. Committee for Medicinal Products for Human Use. ICH topic E14: the clinical evaluation of QT/QTc interval
prolongation and proarrhythmic potential for non-antiarrhythmic drugs questions and answers (R2)
(EMA/CHMP/ICH/310111/2008). ICH. March 21, 2014. Available at:
http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/E14/E14_QAs_R2_Step4.pdf.
Accessed July 24, 2014.
4. Garnett CE, Zhu H, Malik M, et al. Methodologies to characterize the QT/corrected QT interval in the presence
of drug-induced heart rate changes or other autonomic effects. Am Heart J. 2012;163(6):912-930. doi:
10.1016/j.ahj.2012.02.023.
5. Darpo B. The thorough QT/QTc study 4 years after the implementation of the ICH E14 guidance. Br J
Pharmacol. 2010;159(1):49-57. doi: 10.1111/j.1476-5381.2009.00487.x.
6. Couderc JP, McNitt S, Hyrien O, et al. Improving the detection of subtle I(Kr)-inhibition: assessing
electrocardiographic abnormalities of repolarization induced by moxifloxacin. Drug Saf. 2008;31(3):249-260.
7. Shah RR, Hondeghem LM. Refining detection of drug-induced proarrhythmia: QT interval and TRIaD. Heart
Rhythm. 2005;2(7):758-772. doi: 10.1016/j.hrthm.2005.03.023.
8. United States Food and Drug Administration. Guidance for Industry: E14 Clinical Evaluation of QT/QTc Interval
Prolongation and Proarrhythmic Potential for Non-Antiarrhythmic Drugs. Questions and Answers (R1). U.S.
Food and Drug Administration. October 2012. Available at:
http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/ucm073161.pdf.
Accessed March 15, 2014.
9. Morganroth J. Design and conduct of the thorough phase I ECG trial for new bioactive drugs. In: Morganroth J,
Gussak I, eds. Cardiac Safety of Noncardiac Drugs. Totowa, NJ: Humana Press; 2005.
10. Couderc JP, Xiaojuan X, Zareba W, Moss AJ. Assessment of the stability of the individual-based correction of
QT interval for heart rate. Ann Noninvasive Electrocardiol. 2005;10(1):25-34. doi: 10.1111/j.1542474X.2005.00593.x.
11. Dmitrienko A, Beasely C, Mitchell M. Design and Analysis of Thorough QT Studies. BioPharmaceutical
Network. April 29, 2008. Available at: http://www.biopharmnet.com/doc/2008_04_29_report.pdf. Accessed July
6, 2014.
12. Tsong Y. On the designs of thorough QT/QTc clinical trials. J Biopharm Stat. 2013;23(1):43-56. doi:
10.1080/10543406.2013.735762.
13. Salvi V, Karnad DR, Panicker GK, Kothari S. Update on the evaluation of a new drug for effects on cardiac
repolarization in humans: issues in early drug development. Br J Pharmacol. 2010;159(1):34-48. doi:
10.1111/j.1476-5381.2009.00427.x.
14. Judson RS, Salisbury BA, Reed CR, Ackerman MJ. Pharmacogenetic issues in thorough QT trials. Mol Diagn
Ther. 2006;10(3):153-162.
15. van Noord C, Eijgelsheim M, Stricker BH. Drug- and non-drug-associated QT interval prolongation. Br J Clin
Pharmacol. 2010;70(1):16-23. doi: 10.1111/j.1365-2125.2010.03660.x.
16. Kaufmann H. Consensus statement on the definition of orthostatic hypotension, pure autonomic failure and
multiple system atrophy. Clin Auton Res. 1996;6(2):125-126.
17. Low PA, Opfer-Gehrking TL, Textor SC, et al. Postural tachycardia syndrome (POTS). Neurology. 1995;45(4
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 78
Draft
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Suppl 5):S19-S25.
18. Garland EM, Raj SR, Black BK, Harris PA, Robertson D. The hemodynamic and neurohumoral phenotype of
postural tachycardia syndrome. Neurology. 2007;69(8):790-798. doi: 10.1212/01.wnl.0000267663.05398.40.
19. Malik M. Problems of heart rate correction in assessment of drug-induced QT interval prolongation. J Cardiovasc
Electrophysiol. 2001;12(4):411-20.
1317
1318
1319
Appendix E1: Figures
The following figures have been reproduced, with permission:
Figure 3
Creative Commons Attribution 2.5 (CC-BY 2.5) license, 2007
Diagram of the Heart’s
Heuser J. Electrical conduction system of the heart. Based on: Lynch PJ, Jaffe CC.
Conduction System
Heart anterior view coronal section.
http://commons.wikimedia.org/wiki/File:RLS_12blauLeg.png
Figure 4
Public domain.
Sequence of Heart
http://en.ecgpedia.org/wiki/File:Conduction_ap.svg.
Excitation and the
Associated ECG Waveforms
Figure 6
(Left) Public domain.
Standard Electrode/Lead
http://en.wikipedia.org/wiki/File:Limb_leads.svg
Configuration for a 12(Right) Creative Commons CC0 1.0 Universal (CC0 1.0) Public Domain Dedication
Lead/Waveform ECG
http://en.wikipedia.org/wiki/File:Precordial_leads_in_ECG.png
1320
1321
All other figures in this document were created or donated by the team, or originated within CDISC.
1322
Appendix E2: Further Reading
1323
1324
1325
1326
1327
1328
1329
1330
1331
1332
1333
1334
1335
1336
1337
1338
1339
1340
The following works are of interest to this document, but not actively referenced within it.
•
•
•
•
•
Florian JA, Tornoe CW, Brundage R, Parekh A, Garnett CE. Population pharmacokinetic and concentration-QTc models for moxifloxacin: pooled analysis of 20 thorough QT studies. J Clin Pharmacol. 2011;51(8):11521162. doi: 10.1177/0091270010381498.
Garnett CE, Beasley N, Bhattaram VA, et al. Concentration-QT relationships play a key role in the evaluation
of proarrhythmic risk during regulatory review. J Clin Pharmacol. 2008;48(1):13-18. doi:
10.1177/0091270007307881.
ICH Topic S 7 B The nonclinical Evaluation of the Potential for delayed Ventricular Repolarization (QT
Interval Prolongation) by Human Pharmaceuticals. (2005, November). European Medicines Agency. Available
at:
http://www.ema.europa.eu/docs/en_GB/document_library/Scientific_guideline/2009/09/WC500002841.pdf.
Accessed 28 March 2014.
Low PA, Sandroni P, Joyner M, et al. Postural tachycardia syndrome (POTS). J Cardiovasc Electrophysiol.
2009;20(3):352-358.
Tornoe CW, Garnett CE, Wang Y, Florian J, Li M, Gobburu JV. Creation of a knowledge management system
for QT analyses. J Clin Pharmacol. 2011;51(7):1035-1042. doi: 10.1177/0091270010378408.
Page 79
Draft
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
CDISC Therapeutic Area Data Standards: User Guide for QT (Version 1.0 Draft)
Appendix F: Representations and Warranties, Limitations of
Liability, and Disclaimers
CDISC Patent Disclaimers
It is possible that implementation of and compliance with this standard may require use of subject matter covered by
patent rights. By publication of this standard, no position is taken with respect to the existence or validity of any
claim or of any patent rights in connection therewith. CDISC, including the CDISC Board of Directors, shall not be
responsible for identifying patent claims for which a license may be required in order to implement this standard or
for conducting inquiries into the legal validity or scope of those patents or patent claims that are brought to its
attention.
Representations and Warranties
“CDISC grants open public use of this User Guide (or Final Standards) under CDISC’s copyright.”
Each Participant in the development of this standard shall be deemed to represent, warrant, and covenant, at the time
of a Contribution by such Participant (or by its Representative), that to the best of its knowledge and ability: (a) it
holds or has the right to grant all relevant licenses to any of its Contributions in all jurisdictions or territories in
which it holds relevant intellectual property rights; (b) there are no limits to the Participant’s ability to make the
grants, acknowledgments, and agreements herein; and (c) the Contribution does not subject any Contribution, Draft
Standard, Final Standard, or implementations thereof, in whole or in part, to licensing obligations with additional
restrictions or requirements inconsistent with those set forth in this Policy, or that would require any such
Contribution, Final Standard, or implementation, in whole or in part, to be either: (i) disclosed or distributed in
source code form; (ii) licensed for the purpose of making derivative works (other than as set forth in Section 4.2 of
the CDISC Intellectual Property Policy (“the Policy”)); or (iii) distributed at no charge, except as set forth in
Sections 3, 5.1, and 4.2 of the Policy. If a Participant has knowledge that a Contribution made by any Participant or
any other party may subject any Contribution, Draft Standard, Final Standard, or implementation, in whole or in
part, to one or more of the licensing obligations listed in Section 9.3, such Participant shall give prompt notice of the
same to the CDISC President who shall promptly notify all Participants.
No Other Warranties/Disclaimers. ALL PARTICIPANTS ACKNOWLEDGE THAT, EXCEPT AS PROVIDED
UNDER SECTION 9.3 OF THE CDISC INTELLECTUAL PROPERTY POLICY, ALL DRAFT STANDARDS
AND FINAL STANDARDS, AND ALL CONTRIBUTIONS TO FINAL STANDARDS AND DRAFT
STANDARDS, ARE PROVIDED “AS IS” WITH NO WARRANTIES WHATSOEVER, WHETHER EXPRESS,
IMPLIED, STATUTORY, OR OTHERWISE, AND THE PARTICIPANTS, REPRESENTATIVES, THE CDISC
PRESIDENT, THE CDISC BOARD OF DIRECTORS, AND CDISC EXPRESSLY DISCLAIM ANY
WARRANTY OF MERCHANTABILITY, NONINFRINGEMENT, FITNESS FOR ANY PARTICULAR OR
INTENDED PURPOSE, OR ANY OTHER WARRANTY OTHERWISE ARISING OUT OF ANY PROPOSAL,
FINAL STANDARDS OR DRAFT STANDARDS, OR CONTRIBUTION.
Limitation of Liability
IN NO EVENT WILL CDISC OR ANY OF ITS CONSTITUENT PARTS (INCLUDING, BUT NOT LIMITED
TO, THE CDISC BOARD OF DIRECTORS, THE CDISC PRESIDENT, CDISC STAFF, AND CDISC
MEMBERS) BE LIABLE TO ANY OTHER PERSON OR ENTITY FOR ANY LOSS OF PROFITS, LOSS OF
USE, DIRECT, INDIRECT, INCIDENTAL, CONSEQUENTIAL, OR SPECIAL DAMAGES, WHETHER
UNDER CONTRACT, TORT, WARRANTY, OR OTHERWISE, ARISING IN ANY WAY OUT OF THIS
POLICY OR ANY RELATED AGREEMENT, WHETHER OR NOT SUCH PARTY HAD ADVANCE NOTICE
OF THE POSSIBILITY OF SUCH DAMAGES.
Note: The CDISC Intellectual Property Policy can be found at
http://www.cdisc.org/about/bylaws_pdfs/CDISCIPPolicy-FINAL.pdf
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
July 31, 2014
Page 80
Draft
Supplementary Material
SDTMIG Draft Domain: ECG QT Correction Model Data (QT)
6 Domain Models Based on the General
Observation Classes
6.3 Findings
ECG QT Correction Model Data (QT)
QT – Description/Overview for ECG QT Correction Model Data Domain Model
Data describing the description, correction formula and the coefficients of the correction formula used in correction of QT values. CDISC controlled terminology handles standard
correction factors such as Bazett’s and Fredericia’s; however, due to the large and growing number of correction methods used, will not develop controlled terminology for those
alternative correction factors. Therefore, this new findings domain was proposed to store the correction formula information.
The ECG QT Correction Model Data domain is a draft domain at the time of this publication, but it fully conforms to the SDTM findings structure for versions 1.0 through 1.4,
and could be used ould be used by sponsors under this name as a custom domain. No CDISC controlled terminology definition exists for the domain yet.
QT – Specification for ECG QT Correction Model Data Domain Model
qt.xpt, ECG QT Correction Model Data — Findings. One record per QT correction observation per subject, Tabulation.
Controlled
Role
CDISC Notes
Variable Name
Variable Label
Type Terms, Codelist
or Format
STUDYID
Study Identifier
Char
Identifier
Unique identifier for a study.
DOMAIN
Domain Abbreviation
Char
QT
Identifier
Two-character abbreviation for the domain.
USUBJID
Unique Subject Identifier
Char
Identifier
Identifier used to uniquely identify a subject across all studies for all
applications or submissions involving the product.
QTSEQ
Sequence Number
Num
Identifier
Sequence Number given to ensure uniqueness of subject records within a
domain. May be any valid number.
QTGRPID
Group ID
Char
Identifier
Used to tie together a block of related records in a single domain for a
subject. Example: use EGTESTCD as value for QTGRPID for easier
relating of QT dataset to the EG dataset.
QTREFID
QTc Reference ID
Char
Identifier
Internal or external QT identifier.Generally not used for QT dataset.
QTSPID
Sponsor-Defined Identifier Char
Identifier
Sponsor-defined reference number. Perhaps pre-printed on the CRF as an
explicit line identifier or defined in the sponsor's operational database.
Generally not used for QT dataset.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
Core
Req
Req
Req
Req
Perm
Perm
Perm
QT - Page 1
July 7, 2014
SDTMIG Draft Domain: ECG QT Correction Model Data (QT)
Variable Name
QTTESTCD
Variable Label
QTc Test or Examination
Short Name
Type
Controlled
Terms, Codelist
or Format
Char
Role
Core
Short name of the measurement, test, or examination described in
QTTEST. It can be used as a column name when converting a dataset
from a vertical to a horizontal format. The value in QTTESTCD cannot be
longer than 8 characters, nor can it start with a number (e.g., “1TEST”).
QTTESTCD cannot contain characters other than letters, numbers, or
underscores. Examples: QTCDESC, QTCFORM.
QTTEST
QTc Test or Examination
Char
Synonym
Verbatim name of the test or examination used to obtain the measurement
Name
Qualifier
or finding. The value in QTTEST cannot be longer than 40 characters.
Examples: QT Correction Method Description, QT Correction Formula
QTCAT
Category for QTc
Char
*
Grouping
Used to categorize QT correction model observations across subjects.
Qualifier
Generally not used in QT dataset.
QTSCAT
Subcategory for QTc
Char
*
Grouping
A further categorization of the QT correction model. Generally not used
Qualifier
in QT dataset.
QTORRES
Result or Finding in
Char
Result
Result of the QT correction model measurement or finding as originally
Original Units
Qualifier
received or collected. Examples of expected values are 0.132 or 0.432
when the result is a coefficient or measurement, or “LINEAR” or
“PARABOLIC LOG/LOG” when the result is a formula.
QTSTRESC
Character Result/Finding
Char
Result
Contains the result value for all findings, copied or derived from
in Std Format
Qualifier
QTORRES in a standard format or standard units.
QTSTRESC should store all results or findings in character format; if
results are numeric, they should also be stored in numeric format in
QTSTRESN, for example 0.132 or 0.432 when the result is a coefficient
or measurement, or “LINEAR” or “PARABOLIC LOG/LOG” when the
result is a formula.
QTSTRESN
Numeric Result/Finding in Num
Result
Used for continuous or numeric results or findings in standard format;
Standard Units
Qualifier
copied in numeric format from QTSTRESC. QTSTRESN should store all
numeric test results or findings.
QTSTAT
Completion Status
Char
(ND)
Record
Used to indicate a QT correction model was not done. Should be null if a
Qualifier
result exists in QTORRES.
QTREASND
Reason QTc Not
Char
Record
Describes why a measurement or test was not performed. Used in
Performed
Qualifier
conjunction with QTSTAT when correction is NOT DONE.
QTNAM
Vendor Name
Char
Record
Name or identifier of the laboratory or vendor who provided the test
Qualifier
results.
QTDRVFL
Derived Flag
Char
(NY)
Record
Used to indicate a derived record. The value should be Y or null. Records
Qualifier
which represent the average of other records, or that do not come from the
CRF, or are not as originally collected or received are examples of records
that would be derived for the submission datasets. If QTDRVFL=Y, then
QTORRES could be null, with QTSTRESC, and (if numeric)
QTSTRESN having the derived value.
* Indicates variable may be subject to controlled terminology (Parenthesis indicates CDISC/ NCI codelist code value)
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
Topic
CDISC Notes
Req
Req
Perm
Perm
Exp
Exp
Perm
Perm
Perm
Perm
Perm
QT - Page 2
July 7, 2014
SDTMIG Draft Domain: ECG QT Correction Model Data (QT)
QT – Assumptions for ECG QT Correction Model Data Domain Model
1.
The possible kinds of QT corrections are:
a.
2.
3.
4.
5.
6.
7.
8.
Historical Population QTc: Derived by fitting a model to data from a historical population of subjects. The formula for correcting QT, including its
coefficient(s), is based on a formula which is known before the study starts. For example Bazett’s and Fridericia’s corrections are historical correction
methods.
b. Population QTc: Derived by fitting a model to data from the population of subjects in this study. The formula for correcting QT is known before the study
starts, but the coefficient(s) in the correction formula are derived from data collected in the study.
c. Multi-model population QTc: Derived by fitting several models to data for the study populating, then identifying one or more “best” models and using a
correction method based on the best model(s). The candidate models, the method for choosing the best model(s) and, for methods that use multiple best
models, the way in which the best models are combined, are known before the study starts. The formula for correcting QT, as well as its coefficient(s), are
derived from data collected in the study.
d. Individual QTc: Derived by fitting a model to data for each study subject individually. The formula for correcting QT is known before the study starts, but the
coefficient(s) in the formula are derived from data collected from the individual subject.
e. Multi-model Individual QTc: Derived by fitting several models to data for each subject, then identifying one or more “best fit” models for the subject and
using a correction method based on the best fit model(s). The candidate models, the method for choosing the best model(s) and, for methods that use multiple
best models, the way in which the best models are combined, are known before the study starts. The formula for correcting QT, as well as its coefficient(s), are
derived from data collected from the individual subject.
For each QT correction method used in the study, values of EGTESTCD and EGTEST which are assigned at the study level.
CDISC Controlled Terminology includes EGTESTCD and EGTEST values for QT corrected by Bazett’s method and QT corrected by Fredericia’s method. CDISC
Controlled Terminology does not plan to include any other corrected QT tests in CDISC Controlled Terminology in the near future. This is because of the large and
growing number of QT corrections in use, and the fact that there is not scientific consensus on a best fit QT correction method.
A protocol may specify that several different QT correction models be assessed and that only the best fit model for the individual be used for calculating the individual’s
corrected QT intervals.
Another alternative is that the protocol specifies multiple models be computed for each individual irrespective of best fit. There are multiple scenarios of analysis, and
the sponsor should assign values for EGTESTCD/EGTEST appropriately with clear documentation on what each test code represents. For example, if the protocol calls
for computing the top two best fit models, the sponsor could choose to name the top best fit model QTCIAG1 and the second best fit model QTCIAG2, in rank order.
This domain provides information about the QT corrections used in the study. In that sense it contains findings about tests. The name of the test about which
information is being supplied is stored in QTGRPID
This domain is provided when coefficients for QT corrections other than Bazett’s or Fredericia’s are used in a study, and those corrections are derived from data
collected in the study (i.e., are not historical), and the derivation was performed by a vendor, rather than the sponsor. Information on historical correction methods or
derived by the sponsor could be submitted in the QT data domain, if agreed by the submitter and recipient (e.g., study sponsor and regulatory agency).
For both individual and population corrections the USUBJID will be populated.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
QT - Page 3
July 7, 2014
SDTMIG Draft Domain: ECG QT Correction Model Data (QT)
QT – Examples for ECG QT Correction Model Data Domain Model
Example 1
This example consists of two Findings datasets: one for the ECG measurements (eg.xpt) and one for the QTc modeling information (qt.xpt). To tie the EG and QT datasets together
will also require use of RELREC (relrec.xpt) to represent a dataset-to-dataset relationship that holds true for all subjects for all values of EGSEQ and QTGRPID.
Rows 1-9:
Shows calculated QTc values for three subjects. Each subject has three calculated QTc values. For each subject the first two corrections are based on an individual
correction and the third for a population based correction.
eg.xpt
Row
1
2
3
4
5
6
7
8
9
STUDYID DOMAIN USUBJID EGSEQ EGCAT EGTESTCD
EGTEST
EGORRES EGORRESU
STUDY01
EG
1
1
INTERVAL QTCIAG1 QTCI Interval, Aggregate 1
345
msec
STUDY01
EG
1
2
INTERVAL QTCIAG2 QTCI Interval, Aggregate 2
350
msec
STUDY01
EG
1
3
INTERVAL QTCNAG
QTCN Interval, Aggregate
353
msec
STUDY01
EG
2
1
INTERVAL QTCIAG1 QTCI Interval, Aggregate 1
372
msec
STUDY01
EG
2
2
INTERVAL QTCIAG2 QTCI Interval, Aggregate 2
374
msec
STUDY01
EG
2
3
INTERVAL QTCNAG
QTCN Interval, Aggregate
377
msec
STUDY01
EG
3
1
INTERVAL QTCIAG1 QTCI Interval, Aggregate 1
401
msec
STUDY01
EG
3
2
INTERVAL QTCIAG2 QTCI Interval, Aggregate 2
411
msec
STUDY01
EG
3
3
INTERVAL QTCNAG
QTCN Interval, Aggregate
414
msec
The example below contains the QT correction model information used for the three subjects mentioned in the eg.xpt file above.
Rows 1-3:
Row 1:
Row 2:
Row 3:
Shows the correction method information for QTCIAG1.
Shows a description of the QT Correction method.
Shows the formula that was used to calculate the QT correction for that subject.
Shows the QT correction co-efficient that was used in the formula for row 2. Note that this is an example of an individual correction model therefore the co-efficient
used differs between subjects. Also note that for this example there is only 1 co-efficient used in the formula.
Rows 4-6:
Shows the same information used to calculate the linear correction model QTc value. Note that this is an example of an individual correction model therefore the coefficient used differs between subjects.
Rows 7-11: Shows the same information used to calculate the Rautaharju correction model information. Note that this is a population based correction model and therefore the
three co-efficients that are used in this calculation are the same for each subject. Also note that this model shows an example of how to represent three co-efficients
used in a formula.
Rows 12-22: Shows the same correction information used for subject 2.
Rows 23-33: Shows the same correction information used for subject 3.
qt.xpt
Row
1
2
3
4
5
6
7
8
STUDYID DOMAIN USUBJID QTSEQ QTGRPID
STUDY01
QT
1
1
QTCIAG1
STUDY01
QT
1
2
QTCIAG1
STUDY01
QT
1
3
QTCIAG1
STUDY01
QT
1
4
QTCIAG2
STUDY01
QT
1
5
QTCIAG2
STUDY01
QT
1
6
QTCIAG2
STUDY01
QT
1
7
QTCNAG
STUDY01
QT
1
8
QTCNAG
QTTESTCD
QTTEST
QTORRES
QTCDESC QT Correction Method Description PARABOLIC LOG/LOG
QTCFORM
QT Correction Formula
QTC=QT/(RR^A)
QTCCOEFn
QT Correction Coefficient 1
0.432
QTCDESC QT Correction Method Description
LINEAR
QTCFORM
QT Correction Formula
QTC=QT+(A*(1-RR))
QTCCOEFn
QT Correction Coefficient 1
0.154
QTCDESC QT Correction Method Description
RAUTAHARJU COR
QTCFORM
QT Correction Formula
QTC=QT+A-(B*(e^(C^HR))
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
QT - Page 4
July 7, 2014
SDTMIG Draft Domain: ECG QT Correction Model Data (QT)
Row
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
STUDYID DOMAIN USUBJID QTSEQ QTGRPID
STUDY01
QT
1
9
QTCNAG
STUDY01
QT
1
10
QTCNAG
STUDY01
QT
1
11
QTCNAG
STUDY01
QT
2
1
QTCIAG1
STUDY01
QT
2
2
QTCIAG1
STUDY01
QT
2
3
QTCIAG1
STUDY01
QT
2
4
QTCIAG2
STUDY01
QT
2
5
QTCIAG2
STUDY01
QT
2
6
QTCIAG2
STUDY01
QT
2
7
QTCNAG
STUDY01
QT
2
8
QTCNAG
STUDY01
QT
2
9
QTCNAG
STUDY01
QT
2
10
QTCNAG
STUDY01
QT
2
11
QTCNAG
STUDY01
QT
3
1
QTCIAG1
STUDY01
QT
3
2
QTCIAG1
STUDY01
QT
3
3
QTCIAG1
STUDY01
QT
3
4
QTCIAG2
STUDY01
QT
3
5
QTCIAG2
STUDY01
QT
3
6
QTCIAG2
STUDY01
QT
3
7
QTCNAG
STUDY01
QT
3
8
QTCNAG
STUDY01
QT
3
9
QTCNAG
STUDY01
QT
3
10
QTCNAG
STUDY01
QT
3
11
QTCNAG
Rows 1-2:
QTTESTCD
QTCCOEFn
QTCCOEFn
QTCCOEFn
QTCDESC
QTCFORM
QTCCOEFn
QTCDESC
QTCFORM
QTCCOEFn
QTCDESC
QTCFORM
QTCCOEFn
QTCCOEFn
QTCCOEFn
QTCDESC
QTCFORM
QTCCOEFn
QTCDESC
QTCFORM
QTCCOEFn
QTCDESC
QTCFORM
QTCCOEFn
QTCCOEFn
QTCCOEFn
QTTEST
QTORRES
QT Correction Coefficient 1
0.2425
QT Correction Coefficient 2
0.434
Correction Coefficient 3
-0.0097
QT Correction Method Description PARABOLIC LOG/LOG
QT Correction Formula
QTC=QT/(RR^A)
QT Correction Coefficient 1
0.374
QT Correction Method Description
LINEAR
QT Correction Formula
QTC=QT+(A*(1-RR))
QT Correction Coefficient 1
0.132
QT Correction Method Description
RAUTAHARJU COR
QT Correction Formula
QTC=QT+A-(B*(e^(C^HR))
QT Correction Coefficient 1
0.2425
QT Correction Coefficient 2
0.434
Coefficient 3
-0.0097
QT Correction Method Description PARABOLIC LOG/LOG
QT Correction Formula
QTC=QT/(RR^A)
QT Correction Coefficient 1
0.412
QTc Model Type
LINEAR
QT Correction Formula
QTC=QT+(A*(1-RR))
QT Correction Coefficient 1
0.158
QTc Model Type
RAUTAHARJU COR
QT Correction Formula
QTC=QT+A-(B*(e^(C^HR))
QT Correction Coefficient 1
0.2425
QT Correction Coefficient 2
0.434
Coefficient 3
-0.0097
Shows the RELREC information used to join the data in the eg.xpt dataset and the qt.xpt dataset.
relrec.xpt
Row
1
2
3
4
5
6
STUDYID RDOMAIN USUBJID
IDVAR
IDVARVAL RELTYPE RELID
STUDY01
EG
1
EGTESTCD
MANY
EGEQ1
STUDY01
QT
1
QTGRPID
MANY
EGEQ1
STUDY01
EG
2
EGTESTCD
MANY
EGEQ1
STUDY01
QT
2
QTGRPID
MANY
EGEQ1
STUDY01
EG
3
EGTESTCD
MANY
EGEQ1
STUDY01
QT
3
QTGRPID
MANY
EGEQ1
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
QT - Page 5
July 7, 2014
SDTMIG-MD Draft Dataset Specification: Device-Device Relationships (RELDEV)
4 Device Domains
4.8 Device-Device Relationships
Some devices have components and it is important to record those relationships. In some cases relationship between a device and a component device may be
physical and relatively permanent. In other cases, the top-level device may be a logical device, consisting of a devices used together to perform a test or
intervention.
reldev.xpt, Device-Device Relationships — Version x.x. One record per relationship parameter
Variable
Name
Variable Label
Controlled
Definition
Type Terms, Codelist
or Format
Char
Unique identifier for a study.
Char
Sponsor-defined identifier for a device.
STUDYID
SPDEVID
Study Identifier
Sponsor Device
Identifier
PARENT
Parent Sponsor
Device Identifier
Char
DEVLVL
Device Level
Num
PARMCD
Relationship
Parameter Short
Name
Relationship
Parameter
Char
Relationship
Parameter Value
Char
PARM
VAL
Char
Implementation Notes
Core
Unique identifier for a study.
It must be unique for each tracked unit of the device under
study, and can be at whatever level of granularity the device
should be identified (e.g., model or serial number, or
combination of identifiers) as defined in DI.
Must be a value of SPDEVID as defined in DI.
Req
Req
The sponsor-defined identifier for the
Exp
“parent” device, the device of which the
device named in SPDEVID is a component.
Represents the hierarchy level of the device Should be a positive integer. The device at the top level has a
Req
named in SPDEVID, where 1 is the highest level of 1. A direct component of a level 1 device has a level of
level.
2, etc.
Short name for the relationship parameter.
Perm
A parameter that describes properties of the Examples: Quantity, Connection Type, Connection Location
relationship between the component device
named in SPDEVID and the parent device
named in PARENT.
The value of the parameter described in
PARM.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
Perm
Perm
Page 1
April 4, 2014
SDTMIG-MD Draft Dataset Specification: Device-Device Relationships (RELDEV)
4.8.1 Assumptions for Device-Device Relationships Dataset
1.
2.
3.
4.
5.
6.
The Device-Device Relationships Dataset is a relationship dataset that links device components to “parent” devices.
As a relationship dataset, this does not have a Topic variable.
As a relationship dataset, this dataset does not include variables such as DOMAIN and SEQ which are present in all general observation class domains.
RDOMAIN is not needed, since it is expected that all records in this domain will link back to records in DI.
PARENT is not populated for a device at the Level 1.
PARM, PARMCD, and VAL are populated only if there are properties of the relationship that need to be represented.
If the relationship between a device and a component has multiple properties, then the dataset will contain a record for each of these as a parameter.
4.8.2 Examples for Device-Device Relationships Dataset
Example 1:
This example shows the types of devices used to obtain ECG data for a particular study. Although multiple ECG machines were used, they were all of the same
type, so identifiers were assigned to the type of machine, rather than to individual devices. There were two separate devices used to obtain the ECG data: an ECG
machine that recorded data from the subject, and a second device that performed further processing of the ECG data from the first machine, and also allowed
manual over-reading of the ECG data. Because these two machines were used together to produce the results in the ECG domain, the combination of the two
devices is treated as a composite device, and the combination is given its own unique sponsor ID.
Row 1:
Rows 2-5:
Rows 6-10:
Shows the sponsor device identifier for the composite ECG Device. This is the device whose identifier appears in the EG dataset where subject
data are represented. Further information about what this device consisted of is provided in the Device-Device Relationships dataset, below.
Show the characteristics of the ECG machine, which has been given an SPDEVID value of "2
Show the characteristics of the machine which performs further processing of the ECG data from the first machine. This device has been given
an SPDEVID value of “3.”
di.xpt
Row STUDYID DOMAIN SPDEVID DISEQ DIPARMCD DIPARM
DIVAL
ABC-123
DI
1
1
TYPE
Device Type Composite ECG Device
1
ABC-123
DI
2
1
TYPE
Device Type
ECG Machine
2
ABC-123
DI
2
2
MANUF
Manufacturer
Acme
3
ABC-123
DI
2
3
MODEL
Model
XYZ 2000
4
ABC-123
DI
3
1
TYPE
Device Type
ECG Analyzer
5
ABC-123
DI
3
2
MANUF
Manufacturer
Acme
6
ABC-123
DI
3
3
MODEL
Model
ECG Wizard
7
The relationships between the two ECG machines and their combination is a simple one, which requires no additional parameters to describe the relationship, so
the permissible variables PARMCD, PARM, and VAL are not included in the Device-Device Relationship dataset.
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
Page 2
April 4, 2014
SDTMIG-MD Draft Dataset Specification: Device-Device Relationships (RELDEV)
Row 1:
Row 2:
Row 3:
Shows that the device with SPDEVID = 1 has no parent and thus is at Level 1.
Shows that the device with SPDEVID = 2 is a part of the composite device with SPDEVID = 1. Since the device with SPDEVID = 1 is at Level
1, this device is at Level 2.
Shows that the device with SPDEVID = 3 is also a part of the composite device with SPDEVID = 1 and thus is also at Level 2.
reldev.xpt
Row STUDYID SPDEVID PARENT LEVEL
ABC
1
1
ABC
2
1
2
2
ABC
3
1
2
3
Example 2
This example shows the components of a cardiac pacing device. There is information about the nature of the relationship between a device and its
components that is needed, so the Device-Device Relationship Dataset includes the parameters PARMCD, PARM, and VAL.
Row 1:
Row 2-3:
Row 4:
Shows that the device with SPDEVID = LSKKDH23 has no parent and thus is at Level 1.
Show that the device with SPDEVID = 237YALU is a part of the composite device with SPDEVID = LSKKDH23. There are two
records for this relationship because there are two properties of the relationship of this component to its parent that are considered to be
important in describing the relationship, and have been recorded. Row 3 shows that ten of the type of device with SPDEVID =
237YALU are components of the parent device. Row 4 shows that the connections of these components to the parent have “Connection
Type” = “Active.”
Shows that the device with SPDEVID = 29384LHS is also a part of the composite device with SPDEVID = LSKKDH23 and thus is
also at Level 2. This relationship has one property, “Connection Location,” which is “Slot 3.”
reldev.xpt
Row STUDYID SPDEVID
PARENT LEVEL PARMCD
PARM
VAL
ABC
LSKKDH23
1
1
ABC
237YALU LSKKDH23
2
QTY
Quantity
10
2
ABC
237YALU LSKKDH23
2
CONNTYPE Connection Type Active
3
ABC
29384LHS LSKKDH23
2
CONNLOC Connection Location Slot 3
4
© 2014 Clinical Data Interchange Standards Consortium, Inc. All rights reserved
Draft
Page 3
April 4, 2014