NCT02867709

Study ID: UBR-MD-02

Title: A Phase 3, Multicenter, Randomized, Double-Blind, Placebo-Controlled Single Attack Study to Evaluate the Efficacy, Safety, and Tolerability of Oral Ubrogepant in the Acute Treatment of Migraine
Statistical Analysis Plan Amendment 1 Date: 21-Mar-2018

1. Title Page

STATISTICAL ANALYSIS PLAN

A Phase 3, Multicenter, Randomized, Double-Blind, Placebo-Controlled Single Attack Study to Evaluate the Efficacy, Safety, and Tolerability of Oral Ubrogepant in the Acute Treatment of Migraine

Final: 2017-05-17
Amendment #1: 2018-03-21
Protocol Number:
UBR-MD-02 Amendment 3
Development Phase:
3
Product Name:
Ubrogepant
Study Statistician:
Sponsor:
This document is the property of Allergan, Inc. and may not, in full or part, be passed on, reproduced, published, distributed to any person, or submitted to any regulatory authority without the express written permission of Allergan, Inc.

2. Table of Contents

  1. Title Page ..... 1
  2. Table of Contents ..... 2
    2.1 List of Tables ..... 3
    2.2 List of Figures ..... 5
  3. List of Abbreviations and Definition of Terms ..... 6
  4. Introduction ..... 8
    4.1 Study Design Summary ..... 8
    4.2 Study Objectives and Endpoints ..... 9
  5. Statistical Methodology and Study Endpoints ..... 18
    5.1 Statistical Methods Planned in the Protocol and Determination of Sample Size ..... 18
    5.1.1 Statistical and Analytical Plans ..... 18
    5.1.1.1 Common Conventions ..... 18
    5.1.1.2 Demographics ..... 31
    5.1.1.3 Efficacy and
    5.2 Changes in the Conduct of the Study or Planned Analyses ..... 54
    5.2.1 Changes in the Conduct of the Study ..... 54
    5.2.2 Changes to Analyses Prior to Database Lock ..... 54
  6. Data Handling and Analysis Conventions ..... 55
    6.1 Study Treatment Conventions ..... 55
    6.1.1 Analysis Days ..... 55
    6.1.2 Missing/Incomplete Treatment End Date ..... 55
    6.2 Analysis Visit Windows ..... 55
    6.2.1 Efficacy ..... 55
    L
D
□

□

                □

6.4 Imputed Value Listing Conventions ..... 62
7. References ..... 63
2.1 List of Tables
Table 3-1 Abbreviations and Definitions of Terms ..... 6
Table 4-1 Study Objectives and Corresponding Endpoints ..... 9
Table 5-1 Analysis Populations ..... 18
Table 5-2 Statistical Methodology ..... 19
Table 5-3 Missing Data Handling by Endpoint Type ..... 21
Table 5-4 Analysis Population Summaries ..... 31
Table 5-5 Participant Disposition Summaries ..... 31
Table 5-6 Protocol Deviation Summary ..... 32
Table 5-7 Demographic Summaries ..... 32
Table 5-8 Baseline Characteristics Summaries ..... 32
Table 5-9 Medical History Summary ..... 33
Table 5-10 Medical History Summary ..... 33
Table 5-11 Medication Summaries ..... 34
Table 5-12 Efficacy Assessments ..... 35
Table 5-13 Efficacy Endpoint Baseline Definitions ..... 36
Table 5-14 US Analyses ..... 36
Table 5-15 EU Analyses ..... 40
Table 5-16 Multiple Comparisons Procedure Definitions for the US ..... 44
□ □
U.
□□□



Table 5-27 Assumed Response Rates and Estimated Power for Primary and Secondary Efficacy Endpoints ..... 53
Table 6-1 Analysis Day Definitions ..... 55
Table 6-2 Efficacy Analysis Visit Definitions ..... 55





2.2 List of Figures







Figure 5-5 Determination of sustained pain freedom from 2 to 24 hours after the initial dose ..... 27

Figure 5-6 Determination of sustained pain relief from 2 to 24 hours after the initial dose ..... 28
Figure 5-7 Determination of sustained pain freedom from 2 to 48 hours after the initial dose ..... 29
Figure 5-8 Determination of sustained pain relief from 2 to 48 hours after the initial dose .. ..... 30
Figure 5-9 Multiple Comparisons Procedure for the US ..... 44

3. List of Abbreviations and Definition of Terms

Table 3-1 Abbreviations and Definitions of Terms
Abbreviation/Term Definition
AE adverse event
ALP alkaline phosphatase
ALT alanine aminotransferase
ANCOVA analysis of covariance
AST aspartate aminotransferase
ATC Anatomical Therapeutic Chemical
BMI body mass index
C-SSRS Columbia-Suicide Severity Rating Scale
CFB change from baseline
CHD coronary heart disease
CSR clinical study report
CV cardiovascular
DBS dry blood spot
eCRF electronic case report form
ECG electrocardiogram, electrocardiographic
EQ VAS European Quality of Life Visual Analogue Scale
EQ-5D-5L European Quality of Life 5 Dimensional - 5 Level
EU European Union
FDS Functional Disability Scale
FWER familywise error rate
HEOR Health Economics and Outcomes Research
HR hazard ratio
INR international normalized ratio
IP investigational product
IWRS interactive web response system
kg kilogram(s)
KM Kaplan-Meier
LOCF last observation carried forward
LS least squares
m meter(s)
MedDRA Medication Dictionary for Regulatory Activities
mITT modified intent-to-treat
PCS potentially clinically significant
PF pain free
PGIC Participant Global Impression of Change
PO Primary Objective
PR pain relief
PT preferred term
QTc QT interval corrected for heart rate
QTcB QT interval corrected for heart rate using the Bazett formula ( )
QTcF QT interval corrected for heart rate using the Fridericia formula ( )
Abbreviation/Term Definition
SAE serious adverse event
SAP statistical analysis plan
SAS Statistical Analysis Software
SD standard deviation
SE standard error
SI Le Système International d'Unités (International System of Units)
SOC system organ class
SPF sustained pain freedom
SPR sustained pain relief
TBL total bilirubin
TEAE treatment-emergent adverse event
ULN upper limit of normal
US United States of America
WHO World Health Organization

4. Introduction

This statistical analysis plan (SAP) details comprehensive, technical specifications of the statistical analyses of the efficacy and safety data outlined and/or specified in the protocol amendment #3 dated 17May2017 of Study UBR-MD-02. Specifications of tables, figures, and data listings are contained in a separate document.
This document is organized into 3 main sections:
  1. Study overview
  2. Statistical Methodology and Study Endpoints
  3. Data Handling and Analysis Conventions

4.1 Study Design Summary

Structure: Multicenter, randomized, double-blind, placebo-controlled, parallel-group, single attack study; randomization to placebo, ubrogepant 25 mg , or ubrogepant 50 mg .
Duration: The study includes a screening period of up to 14 days prior to randomization, a 60-day period in which to treat a single migraine attack, and a 4-week follow-up period.
Study Treatment Groups: Ubrogepant 25 mg , ubrogepant 50 mg
Controls: Ubrogepant placebo
Dosage/Dose Regimen: Study participants will have up to 60 days to treat a single qualifying migraine attack of moderate or severe headache pain intensity at home. Participants have the option to take a second dose of investigational product (IP) or rescue medication if the participant has either a nonresponding migraine or a migraine recurrence. Participants who are randomized to ubrogepant arms will be randomly assigned at the Randomization Visit (Visit 2) to active treatment or placebo ( ) for the blinded optional second dose. Participants randomized to the placebo arm will receive placebo for their blinded optional second dose.
Randomization Participants will be randomized to 1 of the following 3 treatment groups: placebo, ubrogepant 25 mg , or ubrogepant 50 mg .
Number of Participants: Approximately 1650 participants will be randomized ( 550 per treatment arm).

4.2 Study Objectives and Endpoints

Each study objective is presented with corresponding endpoint(s) below:
Table 4-1 Study Objectives and Corresponding Endpoints
Objectives Endpoints
Primary
- [PO1] To compare the efficacy, safety, and tolerability of 2 doses of ubrogepant ( 25 and 50 mg ) with placebo in participants with a single migraine attack
Primary Efficacy Endpoints
The coprimary efficacy parameters for the United States of America (US) are as follows:
- [P1] Pain freedom (PF) at 2 hours after the initial dose, defined as a reduction in headache severity from moderate/severe at baseline to no pain, at 2 hours after the initial dose
- [P2] Absence of the most bothersome migraine-associated symptom (the most bothersome migraine-associated symptom will be identified at baseline for each participant) at 2 hours after the initial dose.
Secondary Efficacy Endpoints
The secondary efficacy parameters for the US are:
- [S1] Pain relief (PR) at 2 hours after the initial dose, defined as the reduction of a moderate/severe migraine headache to a mild headache or to no headache, at 2 hours after the initial dose
- [S2] Sustained pain relief (SPR) from 2 to 24 hours after the initial dose, defined as pain relief with no administration of either rescue medication or the second dose of IP, and with
Objectives Endpoints
no occurrence thereafter of a moderate/severe headache during the relevant number of hours after dosing with the IP
- [S3] Sustained pain freedom (SPF) from 2 to 24 hours after the initial dose, defined as pain freedom with no administration of either rescue medication or the second dose of IP, and with no occurrence thereafter of a mild/moderate/severe headache during the relevant number of hours after dosing with the IP
- [S4a] Absence of photophobia at 2 hours after the initial dose
- [S4b] Absence of phonophobia at 2 hours after the initial dose
- [S4c] Absence of nausea at 2 hours after the initial dose



5. Statistical Methodology and Study Endpoints

5.1 Statistical Methods Planned in the Protocol and Determination of Sample Size

This statistical analysis plan (SAP) will be approved prior to database lock. The SAP expands the statistical section of the protocol and contains a detailed description of methods to analyze data collected in the study. The text portion of the SAP will be included in the clinical study report (CSR) report as Appendix 16.1.9.

5.1.1 Statistical and Analytical Plans

Statistical analyses will be conducted using SAS Version 9.3 or newer.

5.1.1.1 Common Conventions

5.1.1.1.1 Analysis Populations

The analysis populations will consist of participants as defined below:
Table 5-1 Analysis Populations
Population Definition Study Treatment
Screened All screened participants who signed informed consent and received a participant identification (PID) number -
Intent-to-Treat (ITT) All randomized participants Randomized assignment
Modified Intent-toTreat (mITT) All randomized participants who received at least 1 dose of study treatment, recorded a baseline migraine headache severity measurement, and had post dose migraine headache severity or migraine-associated symptom measurement at or before the 2hour timepoint. Randomized assignment
Safety All participants who received dose of study treatment Actual received

5.1.1.1.2 Study Treatments

The following treatment groups are defined for this study:
  • Placebo
  • Ubrogepant 25 mg
  • Ubrogepant 50 mg

5.1.1.1.3 Statistical Methodology

The methodologies defined below apply as specified to individual endpoints defined in this SAP. All statistical tests will be 2 -sided hypothesis tests performed at the level of significance for
main effects. All confidence intervals will be 2 -sided confidence intervals, unless stated otherwise.
Table 5-2 Statistical Methodology
Methodology Description
M1 Categorical counts
- Number of participants in individual categories
- Participants with qualifying event counted once per individual category
M2 Categorical descriptives
- Number and percentage of participants in individual categories
- Participants with qualifying event counted once per individual category
- N1 if proportion denominator number of participants in the population (standard percentage denominator)
- N1 = participants with non-missing baseline value
M5 Continuous descriptives
- N1, mean, standard deviation (SD), median, minimum, maximum
- N1 = participants with non-missing value
M6 CFB descriptives
- Continuous descriptives for baseline, postbaseline, and change from baseline (CFB) values
- N1 = participants with non-missing values at both baseline and the specified postbaseline analysis visit
M8 Responder
- Categorical descriptives using proportions for responders and nonresponders
- Nonresponders include:
- Participants who do not meet responder criteria
- N1 = all participants unless otherwise specified
M10 Logistic regression model
- Measures the relationship between the binary categorical dependent variable (responder or nonresponder) and independent variables
- Independent variables for initial dose:
- treatment group (placebo, ubrogepant 25 mg , ubrogepant 50 mg )
- historical triptan response (triptan responder, triptan insufficient responder, or triptan naïve)
- use of medication for migraine prevention (yes, no)
- baseline headache severity (moderate or severe)

change from baseline; analysis of covariance.
Raw and derived data listings will be provided, and will be fully defined in the table, figure, and data listing specification document.

5.1.1.1.4 Missing Data

General missing data handling conventions are specified for methodologies in Section 5.1.1.1.3 and summarized as follows:
Table 5-3 Missing Data Handling by Endpoint Type
Parameter type Timing Missing Data Handling
Responder Treatment Period
- If missing headache severity, migraine-associated symptoms, satisfaction with study medication, or functional disability scale at scheduled postdose time points, use LOCF
- Sensitivity analysis for the primary efficacy endpoints is to impute participants with missing data at 2 hours as nonresponders, provided that the participant has at least 1 postdose value before 2 hours after the initial dose
A conservative approach will used to resolve the incompatibility between the answers to the headache recurrence questions at the 24 - and 48 -hour time points by setting the answer to the recurrence question at the 48-hour time point the same as the answer to the recurrence question at the 24 -hour time point, when the 24 -hour time point recurrence question indicates headache recurrence between 2 and 24 hours but the 48-hour time point recurrence question indicates either no or a less severe headache recurrence between 2 and 48 hours.
For sustained efficacy endpoints (sustained pain freedom and sustained pain relief from 2 to 24 and 48 hours after the initial dose), the primary analysis will only include participants for whom the sustained efficacy endpoint in question can be determined based on all available data on headache severity, headache recurrence, use of rescue medication, and use of optional second dose. Figure 5-5 to Figure 5-8 show the diagrams for determining the sustained efficacy endpoints based on all available data.
Figure 5-5 Determination of sustained pain freedom from 2 to 24 hours after the initial dose
Figure 5-6 Determination of sustained pain relief from 2 to 24 hours after the initial dose
Figure 5-7 Determination of sustained pain freedom from 2 to 48 hours after the initial dose
Figure 5-8 Determination of sustained pain relief from 2 to 48 hours after the initial dose
DU



5.1.1.2 Demographics

5.1.1.2.1 Analysis Populations

The distribution of participants within the analysis populations will be summarized as follows:
Table 5-4 Analysis Population Summaries
Population Description Timing Methodology
Screened Population Distribution overall and within sites in total - Categorical counts
ITT, mITT, and Safety populations Distribution overall and within sites in total and by treatment group - Categorical counts

5.1.1.2.2 Participant Disposition

Participant disposition encompasses the distribution of participants who enter, complete, and discontinue each specified analysis period, along with eCRF-reported discontinuation reasons from each respective analysis period. Participant disposition will be summarized as follows:
Table 5-5 Participant Disposition Summaries
Parameter Description Timing Methodology
Screening disposition Distribution in the Screened Population in total Screening Period Categorical descriptives
Double blind disposition Distribution in the Safety Population and ITT Population in total and by treatment group Double Blind Period Categorical descriptives
4 Week Safety Followup disposition Distribution in the Safety Population and ITT Population in total and by treatment group Post-treatment Period Categorical descriptives

5.1.1.2.3 Protocol Deviations

Protocol deviations will be defined in Protocol Deviation Requirement Specification, including importance classification. Protocol deviations will be summarized as follows:
Table 5-6 Protocol Deviation Summary
Parameter Description Timing Methodology
Major protocol
deviations
Distribution in the ITT Population in total
and by treatment group
-
Categorical
descriptives
Protocol deviations will be listed.

5.1.1.2.4 Demographics

Demographics will be summarized in total and by treatment group for the ITT, Safety, and mITT populations, as follows:
Table 5-7 Demographic Summaries
Parameter Description Timing Methodology
Age Age (years) relative to informed consent date Informed consent Continuous descriptives
Age group
- <20
- 20 to 29
- 30 to 39
- 40 to 49
- 50 to 59
- 60 to 69
- >=70
Informed consent Categorical descriptives
Sex, race, and ethnicity
- eCRF categories
- Race group
- White
- Non-white
Screening Period Categorical descriptives
Participant demographics will be listed.

5.1.1.2.5 Baseline Characteristics

Baseline characteristics will be summarized in total and by treatment group for the ITT, Safety, and mITT populations as follows:
Table 5-8 Baseline Characteristics Summaries
Parameter Description Timing Methodology
Baseline characteristics
- Height (m)
- Weight (kg)
- Body mass index (BMI) - Weight (kg) / height (m)
Latest assessment in Screening Period Continuous descriptives
Randomization strata - Previous response to triptans (Triptan Responder, Triptan Insufficient Responder, Triptan Naïve) Randomization date Categorical descriptives
Parameter Description Timing Methodology
- Current use of prophylactic concomitant medication for migraine (Yes, No)
Baseline efficacy
Endpoints and timing fully described in Section 5.1.1.3
- migraine headache severity
- migraine-associated symptom
- most bothersome migraineassociated symptom by symptom
Summary for mITT Population only
Predose Categorical descriptives
Cardiovascular risk
- Cardiovascular risk factor subgroup (low risk, moderate risk, high risk)
Summary for Safety Population only
Randomization date Categorical descriptives
Participant baseline characteristics will be listed.
Participant randomization scheme and codes will be listed.

5.1.1.2.6 Medical History

Medical history, encompassing abnormalities and surgeries reported as occurring before the Screening Visit, will be coded using the Medical Dictionary for Regulatory Activities (MedDRA), version 19.0 or newer. Unique participants who report medical history events will be summarized by MedDRA system organ class (SOC) and preferred term (PT) in total and by treatment group for the Safety Population as follows:
Table 5-9 Medical History Summary
Parameter Description Timing Methodology
Medical history
Abnormalities and surgeries occurring
before the Screening Visit
Screening Period
Categorical
descriptives
SOCs will be sorted alphabetically; PTs will be sorted in descending frequency in the highest dose group.
Participant medical history will be listed.

5.1.1.2.7 Migraine History

Migraine history, including diagnosis, duration of disorder, previous use of prophylaxis treatment, average frequency of moderate to severe migraines per month in past 3 months, and acute treatments will be reported in total and by treatment group for the Safety Population as follows:
Table 5-10 Medical History Summary
Parameter Description Timing Methodology
Migraine Diagnosis With Aura, Without Aura, Both Screening Period Categorical descriptives
Previous Prophylaxis Migraine Treatment Yes or No Screening Period Categorical descriptives
Acute Migraine Treatment Categorize as Yes or No, and subcategorize the Yes by: Screening Period Categorical descriptives
Parameter Description Timing Methodology
- Triptan
- Ergot or Ergot Combinations
- NSAID
- Opiate or Opiate Combination
- Antiemetic Agent
- Barbiturates
- Other
Migraine Disorder Duration In the Table summarize in Years, in the Listing show original data in Years and Months Screening Period Continuous descriptives
Average Frequency of Moderate to Severe Migraines per Month in Last 3 Months N/A Screening Period Continuous descriptives
Participant migraine history will be listed.

5.1.1.2.8 Prior and Concomitant Medications

Medications will be coded using the World Health Organization (WHO) Drug Dictionary, version March 2016 or newer. Unique participants who reported medications will be summarized by Anatomical Therapeutic Chemical (ATC) 4 class and PT in total and by treatment group for the Safety Population as follows:
Table 5-11 Medication Summaries
Parameter Description Timing Methodology
Prior medications Medications taken time before the study treatment start date, regardless of medication end date Screening Period Categorical descriptives
Concomitant medications Medications taken time on or after the study treatment start date, regardless of medication start date Treatment Period Categorical descriptives
ATC4 classes will be sorted alphabetically; PTs will be sorted in descending frequency in the highest dose group.
Participant prior and concomitant medication will be listed.

5.1.1.3 Efficacy and Pharmacokinetic Analyses

Efficacy analyses will be based on the mITT Population.
The following efficacy assessments and terms are defined:
Table 5-12 Efficacy Assessments
Assessment/Term Description
Rating of Headache Severity
Headache severity will be subjectively rated by the participant at predefined timepoints (predose and , and 48 hours after the initial dose) on a scale from no pain to severe pain:
- No pain
- Mild pain
- Moderate pain
- Severe pain
Use of Rescue Medication
Any rescue medication taken within 48 hours after treating their migraine attack with IP, in addition documenting the date and time that the rescue medication was taken.
- Recorded by participants in e-diary
Use of Optional Second Dose and Recurrence of Headache Pain
Optional second dose of IP due to inadequate response to their initial dose of IP. Date and time of the second dose will be reported, as well as pain severity and absence or presence of migraine-associated symptoms at the time the second dose is taken and 2 hours after taking the second dose. The incidence of recurrence in participants who had pain relief and pain freedom at 2 hours after the initial dose will be collected.
- Recorded by participants in e-diary
Migraine-associated symptoms
Absence or presence of migraine-associated symptoms: photophobia, phonophobia, nausea, and vomiting
- Completed by the participant in e-diary
Baseline assessments for applicable efficacy endpoints are defined as follows:
Table 5-13 Efficacy Endpoint Baseline Definitions
Endpoint Description Timing
Baseline rating of headache severity Headache severity rating (Moderate, Severe) pre-dose
Baseline migraineassociated symptom
Migraine-associated symptom
- Photophobia (Yes, No)
- Phonophobia (Yes, No)
- Nausea (Yes, No)
- Vomiting (Yes, No)
pre-dose

5.1.1.3.1 Endpoints for US

The efficacy endpoints for the United States analyses are described as follows.
Table 5-14
US Analyses
Endpoint Description Timing Methodology
P1 Pain freedom (PF) at 2 hours after the initial dose, defined as a reduction in headache severity from moderate/severe at baseline to no pain, at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model
P2 Absence of the most bothersome migraine-associated symptom (the most bothersome migraine-associated symptom will be identified at baseline for each participant) at 2 hours after the initial dose. 2 hours after the initial dose Logistic regression model
S1 Pain relief (PR) at 2 hours after the initial dose, defined as the reduction of a moderate/severe migraine headache to a mild headache or to no headache, at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model
S2 Sustained pain relief (SPR) from 2 to 24 hours after the initial dose, defined as pain relief with no administration of either rescue medication or the second dose of IP, and with no occurrence thereafter of a moderate/severe headache during the relevant number of hours after dosing with the IP 2 to 24 hours after the initial dose Logistic regression model
S3 Sustained pain freedom (SPF) from 2 to 24 hours after the initial dose, defined as pain freedom with no administration of either rescue medication or the second dose of IP, and with no occurrence thereafter of a mild/moderate/severe headache during the relevant number of hours after dosing with the IP 2 to 24 hours after the initial dose Logistic regression model
S4a Absence of photophobia at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model
S4b Absence of phonophobia at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model
S4c Absence of nausea at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model





Table 5-15 EU Analyses

A summary of the number and percentage of participants who took both the optional second dose and rescue medication within 24 (48) hours after the initial dose will also be provided by treatment sequence (placebo/placebo, ubrogepant , ubrogepant placebo, ubrogepant , ubrogepant placebo . The denominator is the number of participants who took the optional second dose.

5.1.1.3.4 Multiple Comparisons Procedure for Primary and Secondary Endpoints

The overall familywise error rate (FWER) will be controlled at for the set of primary and secondary endpoint comparisons between each dose level of ubrogepant vs. placebo both for the US analyses and for the EU analyses.
A graphical approach by Bretz et al (2009) will be used to control the overall type I error rate for multiple comparisons across the ubrogepant doses and the primary and secondary efficacy endpoints. For the US analyses, the coprimary efficacy endpoints will serve as the gatekeepers of the secondary endpoints.
□ The secondary endpoints will be tested in the same order as they appear in the list of secondary endpoints, except for the 2 migraine-associated symptoms, photophobia and phonophobia, which will be treated at the same level to allow the recycling of weights among the 2 symptom endpoints. Recycling of weights between the 2 doses from nausea to pain freedom is also allowed.
Using graphical approach with the weighted Bonferroni-based closed test procedure, the endpoints are represented by circles with associated weights inside the circle. The weight is the fraction of , representing local significance levels. The fraction in the rectangle, associated with a line connecting two circles, indicates the fraction of the local significance level of the circle at the beginning of the line which is added to the local significance level of the circle at the end of the line, if the null hypothesis at the beginning circle is rejected.
Figure 5-9 Multiple Comparisons Procedure for the US
Table 5-16
Circle Alternative Hypothesis Objective Weight Local Significance Level
25 mgP 1 Primary Efficacy Endpoint 1 for 25 mg ubrogepant is significantly different from placebo PO1 1/2
25 mgP 2 Primary Efficacy Endpoint 2 for 25 mg ubrogepant is significantly different from placebo PO1 0
25 mgS 3 Secondary Efficacy Endpoint 1 for 25 mg ubrogepant is significantly different from placebo PO1 0
25 mgS 1 Secondary Efficacy Endpoint 2 for 25 mg ubrogepant is significantly different from placebo PO1 0
25 mgS 2 Secondary Efficacy Endpoint 3 for 25 mg ubrogepant is significantly different from placebo PO1 0
25 mgS 4 a Secondary Efficacy Endpoint 4a for 25 mg ubrogepant is significantly different from placebo PO1 0
25 mgS 4 b Secondary Efficacy Endpoint 4b for 25 mg ubrogepant is significantly different from placebo PO1 0
25 mgS 4 c Secondary Efficacy Endpoint 4c for 25 mg ubrogepant is significantly different from placebo PO1 0
50 mgP 1 Primary Efficacy Endpoint 1 for 50 mg ubrogepant is significantly different from placebo PO1 1/2
Primary Efficacy Endpoint 2 for 50 mg ubrogepant is significantly different from placebo PO1 0
50 mgS 3 Secondary Efficacy Endpoint 1 for 50 mg ubrogepant is significantly different from placebo PO1 0
50 mgS 1 Secondary Efficacy Endpoint 2 for 50 mg ubrogepant is significantly different from placebo PO1 0
50 mgS 2 Secondary Efficacy Endpoint 3 for 50 mg ubrogepant is significantly different from placebo PO1 0
50 mgS 4 a Secondary Efficacy Endpoint 4a for 50 mg ubrogepant is significantly different from placebo PO1 0
50 mgS 4 b Secondary Efficacy Endpoint 4b for 50 mg ubrogepant is significantly different from placebo PO1 0
50 mS 4 c Secondary Efficacy Endpoint 4c for 50 mg ubrogepant is significantly different from placebo PO1 0

5.1.1.4.1 Study Treatment Exposure and Compliance

Study treatment exposure will be summarized and listed for the Safety Population.
The summary of treatment exposure will include the number and percentage of participants who took the initial dose only, the number and percentage of participants who took both the initial dose and the optional second dose, and the number and percentage of participants who took the PK dose for each treatment group.
The listing of treatment exposure will indicate whether the participant took the optional second dose and in addition to the first dose.
Treatment compliance to the first dose of study medication will not be calculated as participants are only to take 1 tablet.




l

5.1.1.5 Subgroup Analyses

The following subgroup analyses will be conducted by historical triptan response. PF and absence of the most bothersome migraine-associated symptom at 2 hours after initial dose will be analyzed using similar models as the primary endpoints with the additional treatment group by historical triptan response interaction. Time to PF and absence of the most bothersome migraine-associated symptom within 2 hours after the initial dose will be analyzed by historical triptan response separately.
  • PF at 2 hours after initial dose
  • Absence of the most bothersome migraine-associated symptom at 2 hours after initial dose (US only)
  • Time to PF within 48 hours after the initial dose
  • Time to absence of the most bothersome migraine-associated symptom within 48 hours after the initial dose (US only)
A pooled analysis among triptan insufficient responders across ubrogepant pivotal studies will be conducted to demonstrate statistically significant efficacy in the triptan insufficient responders' population.

5.1.1.6 Interim Analyses

Not applicable.

5.2 Changes in the Conduct of the Study or Planned Analyses

Prior to database lock, there were no changes in study conduct or planned analyses from what was described in the protocol and detailed in the SAP.

5.2.1 Changes in the Conduct of the Study

Not applicable.

5.2.2 Changes to Analyses Prior to Database Lock

Not applicable.

6. Data Handling and Analysis Conventions

6.1 Study Treatment Conventions

6.1.1 Analysis Days

Treatment day is defined as follows:
Table 6-1 Analysis Day Definitions
Term Description
Treatment Day
Relative to treatment start date
If analysis date treatment start date:
- Day analysis date - treatment start date +1
- Day 1 = treatment start date
If analysis date < treatment start date:
- Day analysis date - treatment start date
- Day day before treatment start date
- There is no Day 0

6.1.2 Missing/Incomplete Treatment End Date

If the investigator is unable to provide the treatment end date, treatment end date will be imputed to the last available dosing record date.

6.2 Analysis Visit Windows

6.2.1 Efficacy

The analysis visit windows for efficacy endpoints are defined as follows:
Table 6-2 Efficacy Analysis Visit Definitions
Analysis Phase Analysis Visit (Derived) Study Hour (eDiary) Window
Pretreatment Baseline Pre-dose Based on nominal timepoints recorded in eDiary
Treatment 0.5 hour post-dose 0.5 hour post-dose
1 hour post-dose 1 hour post-dose
1.5 hours post-dose 1.5 hours post-dose
2 hours post-dose 2 hours post-dose
3 hours post-dose 3 hours post-dose
4 hours post-dose 4 hours post-dose
6 hours post-dose 6 hours post-dose
8 hours post-dose 8 hours post-dose
24 hours post-dose 24 hours post-dose
48 hours post-dose 48 hours post-dose







6.4 Imputed Value Listing Conventions

In general, listings will present the actual partial or missing values rather than the imputed values that may be used in endpoint derivation. In instances where imputed values will be presented, imputed values will be flagged. Actual rules will be fully defined in the table, figure, and data listing specification document.

7. References

Bretz F, Maurer W, Brannath W, Posch M. A graphical approach to sequentially rejective multiple test procedures. Stat Med. 2009;28:586-604.
Firth D. Bias reduction of maximum likelihood estimates. Biometrika.1993;80:27-38.
Grundy SM, Cleeman JI, Bairey Merz CN, Brewer Jr HB, Clark LT, Hunninghake DB, Pasternak RC, Smith Jr SC, and Stone NJ. Implications of recent clinical trials for the National Cholesterol Education Program Adult Treatment Panel III Guidelines. J Am Coll of Cardiology. 2004;44:720-732.
Heinze G, Schemper M. A solution to the problem of separation in logistic regression. Stat Med. 2002;21:2409-2419.
National Cholesterol Education Program Adult Treatment Panel III Guidelines. NIH Publication No. 01-3670, May 2001. http://www.scymed.com/en/smnxdj/edzr/edzr9610.htm
Harborside Financial Center, Plaza V
Jersey City, NJ 07311
UBR-MD-02

A Phase 3, Multicenter, Randomized, Double-Blind, Placebo-Controlled Single Attack Study to Evaluate the Efficacy, Safety, and Tolerability of Oral Ubrogepant in the Acute Treatment of Migraine

STATISTICAL ANALYSIS PLAN AMENDMENT
SUMMARY OF CHANGES

Original SAP Date: 17 May 2017
Amendment #1: 21 March 2018

Confidentiality Statement

This document is the property of Allergan, Inc. and may not, in full or part, be passed on, reproduced, published, distributed to any person, or submitted to any regulatory authority without the express written permission of Allergan, Inc.

1 AMENDMENT #1 TO STATISTICAL ANALYSIS PLAN for UBR-MD-02

1.1 Introduction

Amendment #1 specifies the following changes to the original Statistical Analysis Plan (SAP) for Study UBR-MD-02, dated 17 May 2017:
  • Adding 9 other efficacy endpoints in Table 4-1 Study Objectives and Corresponding Endpoints
  • Adding proportional odds model analysis in Table 5-2 Statistical Methodology
  • Updating missing data handling for headache recurrence endpoints
  • Adding 9 endpoints in Table 5-14 US Analyses and 4 endpoints in
  • Updating Figure 5-9 Multiple Comparisons Procedure for the US
  • Updating Section 5.1.1.4.1 Study Treatment Exposure and Compliance
  • Adding subgroup analyses in Section 5.1.1.5 Subgroup Analyses
  • Minor editorial changes

1.2 Global Changes

None.

1.3 Sections Deleted

None.

1.4 Sections Added

None.

1.5 Revisions

1.5.1 Table 4-1, Study Objectives and Corresponding Endpoints (Pages 11-14)

Rationale: This section has been amended to reflect adding 9 other efficacy endpoints to the list of other efficacy endpoints.

1.5.2 Table 5-2, Statistical Methodology (Page 20-21)

Rationale: This table has been amended to add the proportional odds model for the analysis of pain severity at 2 hours after the initial dose.

1.5.3 Section 5.1.1.1.4, Missing Data (Pages 21-31)

Rationale: This section has been amended to update the missing data handling of sustained efficacy endpoints.

The missing data section now reads as follows:

General missing data handling conventions are specified for methodologies in Section 5.1.1.1.3 and summarized as follows:
Table 5-3 Missing Data Handling by Endpoint Type
A conservative approach will used to resolve the incompatibility between the answers to the headache recurrence questions at the 24 - and 48 -hour time points by setting the answer to the recurrence question at the 48-hour time point the same as the answer to the recurrence question at the 24-hour time point, when the 24-hour time point recurrence question indicates headache recurrence between 2 and 24 hours but the 48-hour time point recurrence question indicates either no or a less severe headache recurrence between 2 and 48 hours.
Figure 5-51 Determination of sustained pain freedom from 2 to 24 hours after the initial dose
Figure 5-6x2 Determination of sustained pain relief from 2 to 24 hours after the initial dose
Figure 5-73 Determination of sustained pain freedom from 2 to 48 hours after the initial dose
Figure 5-84 Determination of sustained pain relief from 2 to 48 hours after the initial dose

1.5.4 Table 5-14, US Analyses

(Pages 37-44)
Rationale: These tables have been amended to reflect the addition of 9 other efficacy endpoints.

The US Analyses and

tables now reads as follows:

Table 5-14 US Analyses
Endpoint Description Timing Methodology
P1 Pain freedom (PF) at 2 hours after the initial dose, defined as a reduction in headache severity from moderate/severe at baseline to no pain, at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model
P2 Absence of the most bothersome migraine-associated symptom (the most bothersome migraine-associated symptom will be identified at baseline for each participant) at 2 hours after the initial dose. 2 hours after the initial dose Logistic regression model
S1 Pain relief (PR) at 2 hours after the initial dose, defined as the reduction of a moderate/severe migraine headache to a mild headache or to no headache, at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model
S2 Sustained pain relief (SPR) from 2 to 24 hours after the initial dose, defined as pain relief with no administration of either rescue medication or the second dose of IP, and with no occurrence thereafter of a moderate/severe headache during the relevant number of hours after dosing with the IP 2 to 24 hours after the initial dose Logistic regression model
S3 Sustained pain freedom (SPF) from 2 to 24 hours after the initial dose, defined as pain freedom with no administration of either rescue medication or the second dose of IP, and with no occurrence thereafter of a mild/moderate/severe headache during the relevant number of hours after dosing with the IP 2 to 24 hours after the initial dose Logistic regression model
S4a Absence of photophobia at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model
S4b Absence of phonophobia at 2 hours after the initial dose 2 hours after the initial dose Logistic regression model




1.5.5 Section 5.1.1.3.4 Multiple Comparisons Procedure for Primary and Secondary Endpoints (Pages 46-48)

Rationale: This section has been modified to enhance the robustness of the testing procedure for efficacy at the ubrogepant 50 mg dose, in order to better reflect the relative clinical importance and likelihood of demonstration of effect for each endpoint.

The section now reads as follows:

A graphical approach by Bretz et al (2009) will be used to control the overall type I error rate for multiple comparisons across the ubrogepant doses and the primary and secondary efficacy endpoints. For the US analyses, the coprimary efficacy endpoints will serve as the gatekeepers of the secondary endpoints.
  • The secondary endpoints will be tested in the same order as they appear in the list of secondary endpoints, except for the 子 migraine-associated symptoms, photophobia and phonophobia, which will be treated at the same level to allow the recycling of weights among the 2 子 symptom endpoints. Recycling of weights between the 2 doses from nausea to pain freedom is also allowed.
Figure 5-95 Multiple Comparisons Procedure for the US
Figure 5-95 Multiple Comparisons Procedure for the US

1.5.6 Section 5.1.1.4.1 Study Treatment Exposure and Compliance (Page 48)

Rationale: This section has been amended to add the summary of treatment exposure.

This section now reads as follows:

Study treatment exposure will be summarized and listed for the Safety Population.
The summary of treatment exposure will include the number and percentage of participants who took the initial dose only, the number and percentage of participants who took both the initial dose and the optional second dose, and the number and percentage of participants who took the PK dose for each treatment group.
The listing of treatment exposure will indicate whether the participant took the optional second dose and in addition to the first dose.
Treatment compliance to the first dose of study medication will not be calculated as participants are only to take 1 tablet.



D

1.5.10Section 5.1.1.5 Subgroup Analyses

The section now reads as follows:
The following subgroup analyses will be conducted by historical triptan response. PF and absence of the most bothersome migraine-associated symptom at 2 hours after initial dose will be analyzed using similar models as the primary endpoints with the additional treatment group by historical triptan response interaction. Time to PF and absence of the most bothersome migraine-associated symptom within 2 hours after the initial dose will be analyzed by historical triptan response separately.
  • PF at 2 hours after initial dose
  • Absence of the most bothersome migraine-associated symptom at 2 hours after initial dose (US only)
  • Time to PF within 48 hours after the initial dose
  • Time to absence of the most bothersome migraine-associated symptom within 48 hours after the initial dose (US only)






  1. Participant disposition will be listed and participants who prematurely discontinued will be listed.
  2. Participant efficacy parameters will be listed.