`
`
`
`
`
`
`
`
`CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`
`APPLICATION NUMBER:
`201655Orig1s000
`
`
`STATISTICAL REVIEW(S)
`
`
`

`

`
`
`
`
`
`
`
`U.S. Department of Health and Human Services
`Food and Drug Administration
`Center for Drug Evaluation and Research
`Office of Translational Science
`Office of Biostatistics
`
`
`Statistical Review and Evaluation
`CLINICAL STUDIES
`
`
`
`NDA/Serial Number:
`Drug Name:
`Study number:
`Applicant:
`Date(s):
`
`Review Priority:
`Biometrics Division:
`Statistical Reviewer:
`
`201-655
`Oxymorphone HCl-ER,
`EN3288-109
`Endo Pharmaceuticals, Inc.
`Filing Mtg: 11/04/10
`PDUFA date: 01/07/11
`Completion date: 12/06/10
`S
`DB VI
`Ling Chen, Ph.D., Mathematical Statistician, Special Project
`Team.
`Concurring Reviewers: Stella Machado, Ph.D., Division Director, and Acting Team
`Leader,
`Controlled Substance Staff
`James Tolliver, Ph.D., Pharmacologist, OD/CSS
`Silvia N. Calderon, Ph.D., Pharmacologist, OD/CSS
`Project Manager:
`Corinne P. Moody, OD/CSS
`Keywords: Crossover design; Drug abuse potential study; Self-reported endpoint;
`Multiple endpoints
`
`Medical Division:
`The CSS Team:
`
`
`Reference ID: 2873052
`
`1
`
`(b) (4)
`
`

`

`Table of Contents
`
`LIST OF TABLES........................................................................................................................................ 2
`LIST OF FIGURES...................................................................................................................................... 2
`1. EXECUTIVE SUMMARY ...................................................................................................................... 3
`2.1 OVERVIEW............................................................................................................................................ 4
`2.1.1 Objectives of the study ................................................................................................................. 4
`2.1.2 Study design ................................................................................................................................. 4
`2.1.3 Abuse Potential Measures............................................................................................................ 5
`2.1.4 Number of subjects....................................................................................................................... 5
`2.1.5 Statistical Methodologies Used in the Sponsor’s Analyses.......................................................... 5
`2.1.5 Sponsor’s results and conclusion................................................................................................. 6
`THE SPONSOR CONCLUDED THAT ............................................................................................................... 7
`2.2 DATA LOCATION .................................................................................................................................. 7
`2.3 REVIEWER’S ANALYSIS ........................................................................................................................ 7
`2.3.1 Descriptive statistics .................................................................................................................... 7
`2.3.2.1 Study model and statistical methodologies.......................................................................................... 15
`2.3.2.2 Results................................................................................................................................................. 15
`3. CONCLUSION....................................................................................................................................... 16
`
`
`List of Tables
`
`
`
`Table 1: Summary Statistics for Emax on “At This Moment” Measures (N=41) .......................................... 8
`Table 2: Summary Statistics for Emax on Overall Measures (N=41) ............................................................ 8
`Table 3: Summary Statistics for Emax on Chewing Experience (N=41) ....................................................... 9
`Table 4: Summary on Least Square Means and Standard Errors (N=41)..................................................... 15
`Table 5: Statistical Analysis Results for Three Comparisons (α=0.05, N=41)............................................. 16
`
`
`
`List of Figures
`
`Figure 1: Mean Time Course Profiles for Drug Liking VAS (N=41) ............................................................ 9
`Figure 2: Mean Time Course Profiles for Good Effects VAS (N=41) ......................................................... 10
`Figure 3: Mean Time Course Profiles for High VAS (N=41) ...................................................................... 10
`Figure 4: Mean Time Course Profiles for ARCI MBG (N=41).................................................................... 11
`Figure 5: Boxplots for Good Effects VAS (N=41)....................................................................................... 11
`Figure 6: Boxplots for High VAS (N=41).................................................................................................... 12
`Figure 7: Mean Responses for Overall Drug Liking VAS (N=41)............................................................... 13
`Figure 8: Mean Responses for Take Drug Again VAS (N=41).................................................................... 13
`Figure 9: Boxplots of Responses at hour 0.5 to Difficulty Chewing VAS by Treatment (N=41)................ 14
`Figure 10: Boxplots of Responses at hour 0.5 to Overall Chewing Experience VAS by Treatment (N=41)14
`
`
`
`Reference ID: 2873052
`
`2
`
`

`

`1. Executive Summary
`
`
`Study EN3288-109 in NDA 201655 was a randomized, single-dose, double-blind, double-
`dummy, four-sequence, four-period, crossover study to evaluate the relative bioavailability and
`subjective effects of EN 3288 40 mg administered intact and after mastication compared with
`OPANA® ER 40 mg administered after mastication and with OPANA® 40 mg (4x10mg)
`administered
`intact
`in healthy nondependent recreational oral prescription opioid user
`experienced in mastication of extended-release opioid formulations.
`
`There were four treatments in the study: EN 3288 40 mg – Intact, EN3288 40 mg - tablet ingested
`after mastication, OPANA® 40 mg IR (4x10 mg) – intact, and OPANA® ER 40 mg - tablet
`ingested after mastication. The comparisons of interest in this study were EN40 3288 40 after
`mastication versus other three treatments on the subjective abuse potential measures: Drug Liking
`VAS, Any Drug Effects, Good Drug Effects VAS, High VAS, Overall Drug Liking VAS, Take
`Drug Again VAS, ARCI MBG, Bad Effects VAS, Sick VAS and Difficulty Chewing VAS as
`well as Overall Chewing Experience VAS. The primary endpoint of interest in this review was
`Emax which was defined as the maximum response during 8 hours after dosing or the maximum
`of change from predose response during 8 hours after dosing if predose response is meaningful,
`for example, High VAS, and ARCI MBG.
`
` A
`
` total of 41 subjects completed the study and were included in this reviewer’s statistical
`analysis.
`
`The reviewer’s analysis showed that
`
`
`• EN 3288 40 mg administered after mastication generated significantly larger drug liking,
`any effects, good effects, high, euphoria effect, and overall drug liking than EN3288 40
`mg administered intact. There was no significant difference on Bad Effects VAS and
`Sick VAS in this comparison. Overall subjects wanted to administer EN 3288 40 mg after
`mastication more than to administer EN 3288 40 mg intact.
`
`• EN 3288 40 mg administered after mastication produced significantly lower any effects,
`good effects and high than OPANA® 40 mg IR (4x10 mg) – intact and OPANA® ER
`40 mg administered after mastication. However, such reduced effects were not seen for
`Drug Liking VAS, Overall Drug Liking VAS, Take Drug Again VAS, Bad Effects VAS
`and Sick VAS in these comparisons, and the least square means of the responses to EN
`3288 40 mg administered after mastication on Good Effects VAS and High VAS are still
`considered large (72.78 ±4.18 and 76.37 ±4.12, respectively) in the unidirectional visual
`analog scale.
`
`• EN 32888 40 mg was significantly more difficult to chew than OPANA® ER 40 mg.
`However, there was no significant difference on overall chewing experience between EN
`32888 40 mg and OPANA® ER 40 mg administered after mastication. Overall, subjects
`disliked the chewing experience for both drugs.
`
`
`
`
`
`
`
`
`Reference ID: 2873052
`
`3
`
`

`

`2. Review Report on Study EN3288-109
`
`2.1 Overview
`
`2.1.1 Objectives of the study
`
`Primary objectives
`
`The primary objective of this study was to evaluate the relative bioavailability (rate and extent of
`absorption) of EN3288 40 mg when administered intact and after mastication compared with
`OPANA ER 40 mg (administered after mastication) and OPANA 40 mg (4×10 mg) (administered
`intact) under fasted conditions in healthy, nondependent, recreational oral prescription opioid
`users experienced in mastication of opioid formulations.
`
`Secondary objectives
`
`The secondary objective of this study was to evaluate the subjective effects of EN3288 40 mg
`administered after mastication compared with EN3288 40 mg administered intact, OPANA ER
`40 mg administered after mastication, and OPANA 40 mg (4×10 mg) administered intact in
`healthy, nondependent, recreational oral prescription opioid users experienced in mastication of
`opioid formulations. In addition, this study evaluated the tamper-resistant qualities of EN3288,
`and explored other potential methods of oral abuse of prescription opioids as described by the
`recreational oral prescription opioid users.
`
`Reviewer’s comment: This review report is only for the secondary objectives of the study.
`
`2.1.2 Study design
`
`This was a randomized, double-blind, double-dummy, 4-sequence, 4-period, single-dose,
`crossover study in healthy, nondependent, recreational oral prescription opioid users experienced
`in mastication of opioid formulations. Each subject participated in a screening visit, a
`qualification phase, and a treatment phase consisting of 4 treatment periods. The washout period
`between two treatments in the treatment phase was at least 72 hours.
`
`There were four treatments in the study. These treatments were
`
`
`A: EN3288 40 mg – intact
`B: EN3288 40 mg – tablet ingested after mastication
`C: OPANA ER 40 mg – tablet ingested after mastication
`D: OPANA 40 mg IR (4x10 mg) – intact (reference product)
`
`
`Four treatment sequences ABCD, BCDA, CDAB, DABC, were used in the study.
`
`Reviewer’s comments: The Sponsor reported that in the treatment phase, subjects were
`randomized to 1 to 4 treatment sequences based on a William’s design (see page 30 on EN3288-
`109 report). However, the design stated in Sponsor’s Table 5 (on page 30 of the study report) is
`not a William’s design.
`
`
`
`Reference ID: 2873052
`
`4
`
`

`

`Data were collected for VAS “at this moment” measures and balanced measures, and ARCI MBG
`at hours -1, 0.5, 1, 1.5, 2, 3, 4, 5, 6, 8, 12, and 24. For Overall Drug Liking VAS, for measures:
`Take Drug Again, and Price Value Assessment, data were collected at hours 8 and 24. Data for
`Pupillometry were collected at hours, -1, 1, 1.5, 2, 3, 4, 5, 6, 8, 12, and 24. Data for Difficult
`Chewing VAS and Overall Chewing Experience VAS were collected at hour 0.5.
`
`
`2.1.3 Abuse Potential Measures
`
`The Sponsor studied the following abuse potential measures:
`
`Visual Analog Scale
`
`Drug Liking (‘at this moment’), Good Effects, Bad Effects, Any Drug Effects, High, Overall
`Drug liking, Take Drug Again, Sick, Difficulty Chewing and Overall Chewing Experience.
`
`ARCI (Addiction Research Center Inventory) short form
`
`ARCI MBG (euphoria effect)
`
`The sponsor also included Pupillometry and Price Value Assessment in the study.
`
`The following summary parameters were calculated for all assessments except for pupillometry
`and VAS Overall Drug Liking and Take Drug Again: Emax (peak effect), tEmax (time of peak
`effect), AUE0-2h (area under the effect curve to 2 hours), AUE0-8h, AUE0-24h.
`
`For VAS Overall Drug Liking, Take Drug Again, and Price Value Assessment, the mean per
`treatment and peak response over all treatments were calculated. For the chewing experience
`VAS, the responses were summarized.
`
`The following summary parameters were calculated for pupillometry: PCmin (apparent minimum
`postdose pupil diameter, PTmin (time to reach the apparent minimum diameter), PT25 (time to
`reach at least 25% reduction in pupil diameter from baseline, PAOC0-2h (the area over the curve
`to 2 hours, relative to the baseline), PAOC0-8h, PAOC0-24h.
`
`2.1.4 Number of subjects
`
`In the qualification phase, 51 subjects were exposed to 30 mg doses of OPANA immediate
`release formulation. Forty-three qualified subjects were randomized into the treatment phase, and
`41 subjects completed all 4 treatment periods of the study. The pharmacodynamic population
`included 41 subjects.
`
`
`2.1.5 Statistical Methodologies Used in the Sponsor’s Analyses
`
`PD variables for each treatment period were derived from the pharmacodynamic assessments. In
`the calculation, actual sample times (hours, relative to the corresponding drug administration
`time) were used instead of planned time points. For each treatment period in the treatment phase,
`the time the subject swallowed the intact tablets was considered time zero. Each PD measure at
`each time during treatment phase was summarized by treatment (A, B, C, and D) using
`
`
`Reference ID: 2873052
`
`5
`
`

`

`appropriate statistics. The derived pharmacodynamic variables were summarized by treatment for
`qualification phase (if applicable) and treatment phase, respectively. All the assessments and the
`derived variables data were presented in the individual listings as well. A linear mixed effects
`model was fit to each endpoint with treatment, period, and sequence as fixed effects, baseline
`(predose) measurements as a covariate where applicable, and subject nested in sequence as a
`random effect. The endpoints were derived pharmacodynamic variables Emax, AUE0-t for VAS
`and ARCI, PCmin, PAOC0-t for Pupillometry, the scores of VAS Overall Drug Liking, Taken
`Drug Again, Chewing Experience and Price Value Assessment Questionnaires.
`
`Reviewer’s Comments: In the reviewer’s analyses, Emax is defined as the maximum response
`during 8 hours after dosing of an abuse potential measure or the maximum of change from
`predose response during 8 hours after dosing if predose response is meaningful, for example,
`High VAS, and ARCI MBG. This reviewer found that the calculation of Emax from the sponsor
`was based on maximum response during 24 hours, and the Emax calculated was not adjusted by
`predose responses, even if the predose responses had been collected.
`
`
`2.1.5 Sponsor’s results and conclusion
`
`The Sponsor reported the following results:
`
`
`• The evaluation of pharmacodynamic assessment was valid as Emax for VAS Drug Liking
`(both ‘At This Moment’ and ‘Overall’) was significantly different between the OPANA
`4×10 mg intact and EN3288 intact treatments.
`• On measures of positive and balance effects, EN3288 masticated induced numerically
`lower Emax and AUE0-2h than OPANA ER masticated and OPANA 4×10 mg intact,
`however, the difference was statistically significant only on some of the “at the moment”
`assessments (VAS Good Effects, VAS High, and VAS Drug Liking [AUE0-2 only]). The
`difference was not significant on any of the “end of the day” measures.
`• On measures of positive and balanced effects, administration of EN3288 intact was
`associated with significantly lower positive effects than administration of EN3288
`masticated, OPANA ER masticated, and OPANA 4×10 mg intact.
`• On measures of positive and balanced effects, the median time to reach Emax was within
`2 hours postdose for all treatments.
`• On the measure of negative effects, all treatments were associated with similar level of
`unpleasant responses and no consistent differentiation between treatments was noted.
`• Administration of all treatments was associated with decreased pupil diameter, with
`EN3288 intact inducing the smallest maximum change from pre-dose in comparison to
`the remaining treatments.
`• Chewing EN3288 was more difficult than chewing OPANA ER or placebo. The overall
`chewing experience for EN3288 and OPANA ER was disliked compared to placebo but
`there was no significant difference between these treatments.
`• On individual interview questionnaires, oxycodone had the highest rate of abuse, with
`subjects clearly preferring that over morphine, codeine, oxymorphone, or other
`prescription opioids.
`• The majority of subjects abuse oxycodone by swallowing (intact and after chewing) and
`chewing as major routes methods of abuse, however, the preferred route is swallowing
`whole.
`
`
`Reference ID: 2873052
`
`6
`
`

`

`The Sponsor concluded that
`• Mastication did partially reduce the extended-release properties of EN3288 and OPANA
`ER. The relative ranking for decreasing Cmax and increasing Tmax is as follows:
`immediate release formulation (intact OPANA 4×10 mg) < OPANA ER 40 mg
`masticated < EN3288 40 mg masticated < EN3288 40 mg intact. Relative to EN3288,
`mastication of EN3288 increased oxymorphone Cmax by 122% in comparison to
`administration of EN3288 intact.
`• There were no new implications from the safety evaluation for the intended uses of
`extended-release EN3288 tablets. Overall, it can be concluded that the mastication of
`EN3288 is associated with subjective and objective drug effects similar to OPANA ER
`and OPANA. However, the new
` formulation was significantly more
`difficult to masticate. Further, based on the results of the interview questionnaire,
`subjects mostly abused oxycodone by swallowing whole and chewing in similar
`proportions. No new methods of tampering became apparent through the results of the
`interview sessions.
`• At least 1 treatment-related TEAE occurred in 92% of subjects. The following TEAEs
`occurred in generally increasing numbers from treatment A to treatment D: pruritus,
`nausea, vomiting, headache, somnolence, and dizziness. There was 1 SAE reported for 1
`subject while taking placebo during the qualification phase. All treatment-related TEAEs
`had been identified in OPANA ER labeling.
`
`
`
`2.2 Data Location
`
`The analysis dataset is located the sponsor’s electronic submission Section 5.3.1.2.25.3.1.
`However, the sponsor did not include original data recorded at scheduled time points. An analysis
`dataset request was sent to the Sponsor on November 9, 2010, and the requested dataset was
`posted on November 12, 2010 at the location:
`
`\\CDSESUB1\EVSPROD\NDA201655\201655.enx
`
`
`2.3 Reviewer’s Analysis
`
`2.3.1 Descriptive statistics
`
`Tables 1-3 summarizes the mean, standard error, minimum, the first quartile (Q1), median, the
`third quartile (Q3), and maximum for the abuse potential measures considered in this reviewer’s
`analysis.
`
`This reviewer categorized abuse potential measures of interest into three categories:
`
`
`1. “At this moment” measures: Drug Liking VAS Any Effects VAS, Good Effects
`VAS, High VAS, ARCI MBG, Bad Effects VAS, and Sick VAS;
`2. Overall measures: Take Drug Again VAS and Overall Drug Liking VAS;
`3. Chewing Experience Measures: Difficult Chewing VAS and Overall Chewing
`Experience VAS.
`
`
`
`
`
`Reference ID: 2873052
`
`7
`
`(b) (4)
`
`

`

`Table 1: Summary Statistics for Emax on “At This Moment” Measures (N=41)
`
`
`Abuse Potential
`Measure
`
`Drug Liking
`VAS
`
`Good Drug
`Effects VAS
`
`High VAS
`
`ARCI MBG
`
`Bad Effects
`VAS
`
`Sick VAS
`
`TRT
`
`Mean StdErr
`
`Min
`
`Q1
`
`Med
`
`Q3
`
`Max
`
`EN40_I
`EN40_M
`O40I I
`O40E M
`EN40_I
`EN40_M
`O40I I
`O40E M
`EN40_I
`EN40_M
`O40I I
`O40E M
`EN40_I
`EN40_M
`O40I I
`O40E M
`EN40_I
`EN40_M
`O40I I
`O40E M
`EN40_I
`EN40 M
`O40I_I
`O40E M
`
`62.08
`79.70
`82.46
`82.04
`45.75
`72.98
`84.44
`83.45
`43.75
`76.49
`88.53
`85.97
`5.05
`7.98
`8.29
`8.59
`15.51
`22.88
`25.64
`24.30
`10.81
`19.28
`20.55
`13.75
`
`3.27
`3.19
`2.43
`2.63
`5.48
`4.64
`2.97
`3.01
`5.55
`4.45
`2.97
`3.25
`0.83
`0.73
`0.78
`0.77
`4.04
`4.12
`4.67
`4.86
`2.90
`4.24
`4.92
`3.87
`
`0
`0
`49.9
`49.4
`0
`0
`26
`22.1
`0
`0
`24.4
`7.8
`0
`0
`0
`0
`0
`0
`0
`0
`0
`-4.1
`-51.1
`0
`
`50.85
`69.9
`72.55
`72.15
`7.9
`57.5
`75.05
`76.3
`9.4
`62.45
`85
`75.6
`0
`4
`4
`4.5
`0
`0
`0
`0
`0
`0
`0
`0
`
`58.6
`85.7
`86.5
`84.4
`51.3
`83
`89.4
`87
`43.9
`87.5
`100
`100
`3
`9
`9
`9
`1.3
`8.5
`9.1
`6.2
`0
`11
`6.7
`0
`
`74.8
`93.6
`97.75
`100
`75.55
`99
`100
`100
`75.65
`100
`100
`100
`9
`11
`12
`13
`19.85
`44.85
`55.6
`52.75
`16.65
`25.9
`34
`16.65
`
`100
`100
`100
`100
`100
`100
`100
`100
`100
`100
`100
`100
`16
`16
`16
`16
`100
`82.3
`93.5
`89.2
`75.1
`100
`100
`100
`
`Table 2: Summary Statistics for Emax on Overall Measures (N=41)
`
`Abuse Potential
`Measure
`
`Overall Drug
`Liking VAS
`
`Take Drug Again
`VAS
`
`TRT
`
`Mean StdErr
`
`Min
`
`Q1
`
`Med
`
`Q3
`
`Max
`
`EN40_I
`EN40 M
`O40I I
`O40E_M
`EN40_I
`EN40 M
`O40I_I
`O40E M
`
`54.98
`69.05
`70.90
`71.53
`55.55
`69.22
`72.56
`71.36
`
`3.56
`3.71
`3.39
`3.69
`3.62
`3.91
`3.40
`4.18
`
`0
`0
`16.2
`15.2
`0
`0
`13.9
`10.7
`
`49.9
`50.3
`53.35
`50
`50
`50.5
`51.9
`50.4
`
`51
`75.1
`71.1
`78
`55.6
`74.1
`73.2
`75
`
`65.65
`85.45
`89.7
`89.55
`71.5
`88.65
`92.25
`98.25
`
`100
`100
`100
`100
`100
`100
`100
`100
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 2873052
`
`8
`
`

`

`Table 3: Summary Statistics for Emax on Chewing Experience (N=41)
`
`Abuse Potential
`
`Difficult Chewing
`
`
`
`100
`70.3
`95.54
`EN40 M
`
`
`III-Inna
`“HI-IE-
`
`
`
`O40E M
`
`16.10
`
`4.00
`
`22.45
`
`83.8
`
`100
`28.4
`0
`4.08
`19.81
`EN40 M
`Overall Chewing
`Expen‘enceVAs mmmmmmmm
`magnum-Elm
`
`Figures 1-4 are mean time course profiles for Drug Liking VAS, Good Drug VAS, High
`VAS, and ARCI MBG. These figures show that the mean time course profiles of EN40_I
`separated from those from EN40_M, O40E_M and O4OI_I, and the mean time course
`profiles from O4OI_I and O40E_M are very close to each other. The profile from
`EN40 M is lower than those of both O4OI_I and O40E_M in early hours, especially for
`Good—Effects VAS and High VAS.
`
`Figure 1: Mean Time Course Profiles for Drug Liking VAS (N=41)
`
`Mean
`
`response 4
`
`Time (hours)
`
`
`—o—EN40_I —I—EN40_M +040I_l —x—O40E_M
`
`
`Note: Drug Liking VAS is on a bipolar scale.
`
`Reference ID: 2873052
`
`9
`
`

`

`Figure 2: Mean Time Course Profiles for Good Eflects VAS (N=41)
`
`
`
`Time (haul!)
`
`—o— EMO_I —I— EMO_M + 040l_l —u— 04(IE_M
`
`-0.5
`
`0.5
`
`1.5
`
`25
`
`3.5
`
`4.5
`
`5.5
`
`6.5
`
`7.5
`
`8.5
`
`Figure 3: Mean Time Course Profiles for High VAS (N=41)
`
`
`
`0.5
`
`1.5
`
`25
`
`3.5
`
`4.5
`
`5.5
`
`6.5
`
`7.5
`
`8.5
`
`Time (hauls)
`
`+ ENMLI + EMOJI +O4G_I +O4(IE_M
`
`Reference ID: 2873052
`
`10
`
`

`

`Figure 4: Mean Time Course Profiles for ARCI MBG (N=41)
`
`
`
`4
`
`5
`
`Time (hours)
`
`+ EN40_I _._ EN40_M +040I_I +040E_M
`
`Figures 5 and 6 present Boxplots for individual responses at hours 0.75, 1.0, 1.5 and 2.0 for Good
`Drug Efi‘ects VAS and High VAS respectively.
`
`Figure 5: Boxplots for Good Effects VAS (N=41)
`
`B
`
`GoodEfiecBathou‘OJS
`
`GoodBfed:VASatl-nour1.0
`
`mm.
`
`32
`
`Good WVASahourts
`
`GoodBfoctsVASat hour 2.0
`
`Impen-
`
`ii
`
`aasssgsgggg
`95888888!
`
`
`Reference ID: 2873052
`
`11
`
`

`

`
`
`Figure 6: Boxplots for High VAS (N=41)
`
`
`
`Note: the middle line in a boxplot represents median of the distribution.
`
`Even though the peak mean response to EN40_M is lower than that of O40I_I and O40E_M on
`Good Effects VAS and High VAS (see Figures 2 and 3), approximately 50% of study subjects
`had a score greater than 72 on Good Effects VAS at hour 1.5 and a score greater than 82 on High
`VAS at hour 1.0. These scores are still considered large good effects and high in the
`unidirectional visual analogy scale.
`
`Figures 7 and 8 present mean responses for overall measures: Overall Drug Liking and Take
`Drug Again at hours 8 and 24. From these two graphs, one may see that the mean responses at
`hour 24 are lower that at hour 8 for all treatments. Mean responses to EN40_I is lower than those
`to the other treatments. There is no much difference in mean response to EN40_M, O40I_I and
`O40E_M for these two measures.
`
`
`
`Reference ID: 2873052
`
`12
`
`Best Available
`Copy
`
`

`

`Figure 7: Mean Responses for Overall Drug Liking VAS (N=41)
`
`
`
`EN40_I
`
`040_l
`EN40_M
`Treatment
`
`I Hour 8 I l-bur 24
`
`0 2OO
`
`.
`
`8 g £
`
`Note: Overall Drug Liking VAS is on a bipolar scale.
`
`Figure 8: Mean Responses for Take Drug Again VAS (N=41)
`
`
`
`EN40_I
`
`040 I
`EN40_M
`Treatment
`
`I Hour 8 l Hour 24
`
`Figures 9 and 10 are boxplots for Emax of individual responses to Difficulty Chewing VAS and
`Overall Chewing Experience.
`
`Reference ID: 2873052
`
`13
`
`

`

`Figure 9: Boxplots of Responses at hour 0.5 to Difficulty Chewing VAS by
`Treatment (N=41)
`
`
`
`
`
`Figure 10: Boxplots of Responses at hour 0.5 to Overall Chewing Experience VAS by
`Treatment (N=41)
`
`
`
`
`
`
`
`
`Note: 0: strong disking; 50: neutral; 100: strong liking.
`
`
`
`This study was designed using double dummy strategy. The responses to EN40_I and O40I_I on
`Difficulty Chewing VAS and Overall Chewing Experience were, in fact, from chewing placebos.
`Difficult Chewing VAS was on a unidirectional scale, while Overall Chewing Experience VAS
`was on a bipolar scale. Chewing EN40 was more difficulty than chewing O40 ER or placebo. The
`overall chewing experience for both EN40_M and O40E_M was disliked compared to placebo.
`
`
`Reference ID: 2873052
`
`14
`
`

`

`
`
`2.3.2 Statistical testing
`
`2.3.2.1 Study model and statistical methodologies
`
`The statistical model used in the reviewer’s primary analysis includes sequence, treatment, and
`period as fixed effects, and subject nested within sequence as a random effect. The model
`assumption of the normality of error terms was checked using Shapiro-Wilk W-test on the
`residuals. If the normal assumption was not satisfied, the rank data (ranking treatment responses
`within subject) were used to obtain the p-value of the test for difference in medians between two
`treatments.
`
`
`2.3.2.2 Results
`
`Table 4 lists the least square mean, and its standard error for each treatment and for each
`abuse potential measure.
`
`
`Table 4: Summary on Least Square Means and Standard Errors (N=41)
`
`
`
`Abuse Potential Measure
`
`O40I I
`O40E M
`EN40 M
`EN40 I
`LSmean StdErr LSmean StdErr LSmean StdErr LSmean StdErr
`62.11
`2.93
`79.63
`2.93
`81.94
`2.93
`82.43
`2.93
`49.70
`3.89
`77.56
`3.89
`87.52
`3.89
`89.35
`3.89
`45.71
`4.17
`72.78
`4.17
`83.33
`4.17
`84.37
`4.17
`43.80
`4.10
`76.37
`4.10
`85.89
`4.10
`88.55
`4.10
`5.05
`0.79
`7.97
`0.79
`8.59
`0.79
`8.29
`0.79
`55.16
`3.65
`69.06
`3.65
`71.64
`3.65
`70.92
`3.65
`55.80
`3.79
`69.40
`3.79
`71.61
`3.79
`72.70
`3.79
`15.38
`4.38
`22.87
`4.38
`24.27
`4.38
`25.59
`4.38
`10.65
`4.08
`19.12
`4.08
`13.66
`4.08
`20.41
`4.08
`10.55
`2.81
`95.52
`2.81
`16.18
`2.81
`9.02
`2.81
`49.35
`3.54
`19.87
`3.54
`26.80
`3.54
`54.31
`3.54
`
`Drug Liking VAS
`Any Effects VAS
`Good Effects VAS
`High VAS
`ARCI MBG
`Overall Drug Liking VAS
`Take Drug Again VAS
`Bad Effects VAS
`Sick VAS
`Difficult Chewing VAS
`Overall Chewing Experience VAS
`
`
`Table 5 lists the difference in the least square means, standard error of the difference, and
`p-value from testing result in the significance of the comparisons EN40_I versus EN40_I,
`EN40_M versus O40E_M, and EN40_M versus O40I_I.
`
`
`
`Reference ID: 2873052
`
`15
`
`

`

`Table 5: Statistical Analysis Results for Three Comparisons (α=0.05, N=41)
`
`
`
`Comparison
`
`Abuse Potential Variable
`
`EN40 M vs. EN40 I
`LSmean
`P-
`diff
`value
`
`StdErr
`
`EN40 M vs. O40E M
`LSmean
`P-
`diff
`value
`
`StdErr
`
`EN40 M vs. O40I I
`LSmean
`P-
`diff
`value
`
`StdErr
`
`Drug Liking VAS
`Any Effects VAS
`Good Effects VAS
`High VAS
`ARCI MBG
`Overall Drug Liking VAS
`Take Drug Again VAS
`Bad Effects VAS
`Sick VAS
`Difficult Chewing VAS
`Overall Chewing Experience VAS
`
`17.52
`27.86
`27.07
`32.57
`2.92
`13.90
`13.60
`7.49
`8.47
`84.97
`-29.48
`
`S+
`2.61
`S+
`4.11
`S+
`4.32
`S+
`4.24
`S+
`0.54
`S+
`3.33
`S+
`3.64
`4.64 NS+
`4.64 NS+
`3.98
`S+
`4.42
`S-
`
`-2.31
`-9.96
`-10.55
`-9.52
`-0.62
`-2.58
`-2.21
`-1.40
`5.46
`79.34
`-6.93
`
`NS-
`2.61
`S-
`4.11
`S-
`4.32
`S-
`4.24
`NS-
`0.54
`NS-
`3.33
`NS-
`3.64
`NS-
`4.64
`4.64 NS+
`3.98
`S+
`4.42
`NS-
`
`-2.80
`-11.79
`-11.59
`-12.18
`-0.32
`-1.86
`-3.30
`-2.72
`-1.30
`86.51
`-34.44
`
`2.61
`4.11
`4.32
`4.24
`0.54
`3.33
`3.64
`4.64
`4.64
`3.98
`4.42
`
`NS-
`S-
`S-
`S-
`NS-
`NS-
`NS-
`NS-
`NS-
`S+
`S-
`
`
`Note: S denotes Significant at α=0.05, NS denotes not significant at α=0.05. “+” (or “-”) sign denotes the
`least square mean in treatment 1 is larger (or smaller) than that in treatment 2.
`
`Table 5 shows that
`
`
`• EN40_M had significantly larger mean response than EN40_I on Drug Liking VAS, Any
`Effects VAS, Good Effects VAS, High VAS, ARCI MBG and Overall Drug Liking VAS.
`There was no significant difference in means on Bad Effects VAS and Sick VAS in this
`comparison. Overall, subjects wanted to take EN40_M more than to take EN40_I
`(significant at α=0.05).
`• EN40_M had significantly lower mean response than O40I_I and O40E_M on Any
`Effects VAS, Good Effects VAS, and High VAS. However, such reduced effects did not
`show on Drug Liking VAS, Overall Drug Liking VAS, Take Drug Again VAS, Bad
`Effects VAS and Sick VAS in these comparisons, and the least square means on Good
`Effects VAS and High VAS are still considered large in the unidirectional scale (72.78
`±4.18 for Good Drug Effects VAS and 76.37 ±4.12 for High VAS).
`• EN40_M had significantly larger mean response than O40E_M on Difficulty Chewing
`VAS. However, there was no significant difference on Overall Chewing Experience VAS
`between EN40_M and O40E_M. Overall subjects disliked Chewing Experience for both
`drugs.
`
`
`
`
`
`3. Conclusion
`
`After evaluating Study EN3288-109, the reviewer concludes that
`
`
`• EN 3288 40 mg administered after mastication generated significantly larger drug liking,
`any effects, good effects, high, euphoria effect, and overall drug liking than EN40 mg
`administered intact. Overall, subjects wanted to administer EN 3288 40 mg after
`mastication more than to administer EN 3288 40 mg intact.
`
`
`Reference ID: 2873052
`
`16
`
`

`

`• EN 3288 40 mg administered after mastication produced significantly lower any effects,
`good effects and high than to OPANA® 40 mg IR (4x10 mg) – intact and OPANA® ER
`40 mg administered after mastication. However, such reduced effects were not seen for
`Drug Liking VAS, Overall Drug Liking VAS, Take Drug Again VAS, Bad Effects VAS
`and Sick VAS, and the least square means of the responses to EN 3288 40 mg
`administered after mastication on Good Effects VAS and High VAS are still considered
`large (72.78 ±4.18 and 76.37 ±4.12, respectively) in the unidirectional visual analog
`scale.
`
`• EN 32888 40 mg was significantly more difficult to chew than OPANA® ER 40 mg.
`However, there was no significant difference on Overall Chewing Experience between
`EN 32888 40 mg and OPANA® ER 40 mg. Overall, subjects disliked the chewing
`experience for both drugs.
`
`
`
`
`Reference ID: 2873052
`
`17
`
`

`

`---------------------------------------------------------------------------------------------------------
`This is a representation of an electronic record that was signed
`electronically and this page is the manifestation of the electronic
`signature.
`---------------------------------------------------------------------------------------------------------
`/s/
`----------------------------------------------------
`
`LING CHEN
`12/06/2010
`
`STELLA G MACHADO
`12/06/2010
`
`Reference ID: 2873052
`
`