`
`CENTER FOR DRUG EVALUATION AND
`RESEARCH
`
`
`APPLICATION NUMBER:
`022406Orig1s000
`
`STATISTICAL REVIEW(S)
`
`
`
`
`
`
`
`
`

`

`
`
`
`
`
`
`U.S. Department of Health and Human Services
`Food and Drug Administration
`Center for Drug Evaluation and Research
`Office of Translational Sciences
`Office of Biostatistics
`
`
`S TAT I S T I C A L R E V I E W A N D E VA L U AT I O N
`CLINICAL STUDIES
`
`NDA/BLA Serial
`Number:
`Drug Name:
`Indication(s):
`
`Applicant:
`Date(s):
`
`Review Priority:
`
`22-406
`
`XARELTO (rivaroxaban) Film-coated 10 mg tablet for oral intake
`Prophylaxis of deep vein thrombosis (DVT) and pulmonary embolism (PE) in
`patients undergoing hip or knee replacement surgery (10 mg orally, once daily)
`Johnson and Johnson Pharmaceutical Research and Development, LLC
`01/03/2011 (NDA re-submission); 07/03/2011 (PDUFA goal date);
`02/16/2011 (Consult received from DHP)
`Standard
`
`
`
`
`
`Division of Biometrics 7
`
`Biometrics Division:
`Statistical Reviewer:
`John Stephen Yap, PhD
`Concurring Reviewers: LaRee Tracy, MA, PhD (Team Leader)
`Aloka Chakravarty, PhD (Division Director)
`
`
`Division of Hematology Products
`
`
`Medical Division:
`Clinical Team:
`
`Min Lu, MD (Clinical Reviewer); Kathy Robie Suh, MD, PhD (Team Leader);
`Anne Farrell, MD (Division Director)
`Tyree Newman
`
`
`
`Project Manager:
`
`
`
`Keywords: Venous Thromboembolism, Thrombosis, DVT, Pulmonary Embolism, Liver Toxicity, Safety
`
`
`
`
`
`Reference ID: 2959006
`
`
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`Table of Contents
`
`
`
`LIST OF TABLES ........................................................................................................................ 3
`
`LIST OF FIGURES ...................................................................................................................... 4
`
`1. EXECUTIVE SUMMARY.................................................................................................... 5
`
`2. INTRODUCTION.................................................................................................................. 7
`2.1 OVERVIEW.......................................................................................................................... 7
`2.2 DATA SOURCES ................................................................................................................ 11
`3. STATISTICAL EVALUATION......................................................................................... 11
`3.1 DATA AND ANALYSIS QUALITY........................................................................................ 11
`3.2 EVALUATION OF EFFICACY............................................................................................... 12
`3.3 EVALUATION OF SAFETY ...................................................................................................... 12
`3.3.1
`Study Design and Endpoints ..................................................................................... 12
`3.3.2
`Study Populations and Extent of Exposure............................................................... 14
`3.3.3 Demographic and Baseline Characteristics ............................................................. 18
`3.3.4 Endpoints and Statistical Methodologies.................................................................. 19
`3.3.5 Applicant’s Results and Statistical Safety Reviewer’s Findings and Comments...... 21
` FINDINGS IN SPECIAL/SUBGROUP POPULATIONS .............................................. 32
`
`4.
`
`5. SUMMARY AND CONCLUSIONS................................................................................... 33
`5.1 STATISTICAL ISSUES AND COLLECTIVE EVIDENCE ........................................................... 33
`5.2 CONCLUSIONS AND RECOMMENDATIONS ......................................................................... 35
`APPENDIX I: HEPATIC ADVERSE EVENTS BY MEDDRA PREFERRED TERM...... 37
`
`A. ROCKET AND J-ROCKET................................................................................................ 37
`
`B. EINSTEIN DVT/PE .............................................................................................................. 40
`
`C. EINSTEIN EXTENSION..................................................................................................... 42
`
`SIGNATURES/DISTRIBUTION LIST.................................................................................... 43
`
`
`
`
`
`
`
`
`
`
`Reference ID: 2959006
`
`2
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`LIST OF TABLES
`
`Table 1: Summary of Laboratory Findings..................................................................................... 6
`Table 2: Summary of Clinical Studies.......................................................................................... 10
`Table 3: Study Populations for Rocket, J-Rocket and Einstein Studies ....................................... 15
`Table 4: Subject Years of Exposure for Rocket, J-Rocket and Einstein Studies.......................... 15
`Table 5: Number of Subjects in the Safety Population in Record 1-4 Studies............................. 16
`Table 6: Duration of Treatment of Active Study Medication by Study in Record 1-4 Studies.... 16
`Table 7: ALT Measurements by Study Week in Rocket, J-Rocket and Pooled........................... 17
`Table 8: TBL Data Measurements by Study Week in Rocket, J-Rocket and Pooled................... 17
`Table 9: Study Medication Completion/Withdrawal Information During the Double-Blind Period
`(Rocket Study) .............................................................................................................................. 18
`Table 10: Demographic and Baseline Characteristics (All Randomized Subjects)...................... 19
`Table 11: ALT (SGPT) Post Baseline Data*................................................................................ 22
`Table 12: Total Bilirubin (based on central laboratory) Post-Baseline Data................................ 23
`Table 13: Incidence of Prespecified Laboratory Abnormalities with Hazard Ratios (Combined
`ALT>3xULN and TBL>2xULN) (Based on Central and Local laboratory assessments) Rocket
`and J-Rocket: Safety Analysis Set ................................................................................................ 25
`Table 14: Total Bilirubin Post-Baseline Data Corresponding to ALT>3xULN........................... 27
`Table 15: ALT>3x ULN Occurrence by Study Window.............................................................. 28
`Table 16: TBL>2x ULN Occurrence by Study Window.............................................................. 28
`Table 17: ALT>3x, TBL>2x ULN (Hy’s Law Cases) Occurrence by Study Window................ 29
`Table 18: Post-baseline Hepatic-Related AEs in Rocket and J-Rocket Studies*......................... 30
`Table 19: Hepatic adverse events among identified Hy's Law Cases........................................... 31
`Table 20: Summary of Post-baseline Adverse Events in Einstein DVT and PE Studies* ........... 31
`Table 21: Analysis of ALT>3x ULN by Gender and Race in the Rocket Study ......................... 32
`Table 22: Subgroups for Hy's Law Cases in the Rocket Study .................................................... 33
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 2959006
`
`3
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`LIST OF FIGURES
`
`Figure 1: Scatter Plot of Baseline and Postbaseline Maximum ALT Levels with Maximum Total
`Bilirubin Levels (Based on Central and Local laboratory assessment) Rocket and J-Rocket:
`Safety Analysis Set ....................................................................................................................... 24
`Figure 2: Hazard Ratios and 95% CIs for the Time from the First Study Drug Administration to
`the First Post-baseline Occurrence of Hy's Law Cases Either on the Same Day or Within 30 Days
`(Based on Central and Local laboratory assessments).................................................................. 26
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Reference ID: 2959006
`
`4
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`EXECUTIVE SUMMARY
`
`1.
`
`This statistical safety review assesses liver toxicity using data from five randomized, phase 3
`trials of rivaroxaban (Rocket, J-Rocket, Einstein DVT, Einstein PE and Einstein Extension).
`These five trials evaluated chronic use (i.e. treatment durations of >35 days and up to 4 years) of
`rivaroxaban at doses ranging from 10-30 mg daily in patients with atrial fibrillation, deep venous
`thrombosis or pulmonary embolism. Assessments of liver toxicity were performed on specific
`liver function tests using local and central laboratory results and reported hepatic adverse events.
`A total of 24,134 subjects in the safety populations across the five studies were available for
`assessment: 12,077 rivaroxaban, 7,764 warfarin, 3,703 enoxaparin, and 590 placebo. Warfarin
`was used as the active control in both Rocket studies whereas enoxaparin was used in the
`Einstein DVT and PE studies and placebo in the Extension study.
`
`The proportions of subjects with an elevated alanine aminotransferase (ALT) using predefined
`clinical thresholds (>3, 5, 8, 10 and 20 times the upper limit of normal (ULN)) were generally
`balanced between the rivaroxaban and warfarin arms in the pooled Rocket and J-Rocket studies.
`In the pooled Einstein DVT/PE studies (comprising 31% of data from all five studies), the
`proportion of patients with an ALT >3x ULN or >5x ULN was significantly lower in the
`rivaroxaban arm compared to the enoxaparin arm with no difference noted at the higher ALT
`elevations (Table 1). In the Einstein Extension study (comprising 5% of data from all five
`studies) the proportion of patients with an ALT>3x ULN was numerically larger in the
`rivaroxaban arm compared to placebo.
`
`The proportions of subjects with an elevated total bilirubin (TBL) using predefined clinical
`thresholds (>1.5, 2, 3, 5 and 8x ULN) were generally balanced between treatment groups across
`all five studies. However, results from the pooled Rocket studies suggested a lower proportion
`of elevated TBL>1.5 and 2x ULN in the rivaroxaban arm compared to warfarin. In contrast,
`results from the pooled Einstein DVT/PE studies, resulted in slightly more elevations of
`TBL>1.5x ULN in the rivaroxaban arm compared to enoxaparin. In the Einstein Extension study,
`there were no statistically significant differences in the comparisons of TBL elevations between
`treatment arms.
`
`Models to estimates hazard ratios for high thresholds, e.g. ALT>5x ULN or TBL>3x ULN, may
`be underpowered to detect any differences between treatment arms due to low event rates by
`threshold level.
`
`Analysis by study time windows showed that the proportion of patients with ALT>3x ULN and
`TBL>2x ULN were similar per treatment arm in the Rocket and J-Rocket studies. However, for
`the Einstein DVT and PE studies there was a large proportion of subjects (77%) who had
`ALT>3x ULN from baseline up to week 2 in the enoxaparin arm; after week 2, there were more
`subjects in the rivaroxaban arm (65%) than in the enoxaparin arm (24%) with ALT>3x ULN.
`
`Concurrent (same day) values of ALT>3x ULN and TBL>2x ULN (Hy’s Law cases) occurred in
`approximately 0.5% of subjects (34 and 36 in the rivaroxaban and warfarin groups, respectively)
`in the pooled Rocket and J-Rocket studies. The average time (standard deviation) from first dose
`
`
`
`Reference ID: 2959006
`
`5
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`administration until the occurrence of Hy’s Law cases (among the Hy’s Law cases only) was
`longer in the pooled rivaroxaban group [412 (260) days] compared to the warfarin group [329
`(206) days] in the Rocket studies. The occurrences of the Hy’s Law cases in the rivaroxaban arm
`were roughly uniform over 2.5 years while those in warfarin occurred mostly within 1 year
`(78%) post-randomization. The estimated hazard ratio in a time-to-event analysis was 0.95 with
`a 95% CI of (0.60, 1.52), which suggest no statistically significant difference. However, given
`the low incidence of Hy’s Law cases, the hazard ratios may be underpowered to detect
`differences between treatment arms. Additional time-to-event analyses considering concurrent
`and non-concurrent cases, direct bilirubin ≥0.5 TBL, and treatment emergent cases resulted in no
`statistically significant differences between treatment arms. The numbers of Hy’s Law cases in
`the pooled Einstein studies (follow-up to one-year post-randomization) were 6 and 6 in the
`rivaroxaban and enoxaparin arms, respectively (approximately less than 0.2% of subjects) and
`there were no Hy’s Law cases observed in the Einstein Extension study (follow-up to one-year
`after entry into the extension study). These results are summarized in Table 1.
`
`Table 1: Summary of Laboratory Findings
`Studies/Criteria
`Treatment Arms
`Rocket, J-Rocket
`Rivaroxaban
`Warfarin
`TBL>1.5x ULN
`512/7,618 (6.72)
`561/7,646 (7.34)
`TBL>2x ULN
`157/7,618 (2.06)
`201/7,646 (2.63)
`Hy’s Law Cases
`34/7,618 (0.45)
`36/7,650 (0.47)
`Einstein DVT/PE
`Rivaroxaban
`Enoxaparin
`ALT>3x ULN
`58/3,556 (1.63)
`111/3,489 (3.18)
`ALT>5x ULN
`15/3,556 (0.22)
`33/3,489 (0.95)
`TBL>1.5x ULN
`71/3,559 (1.99)
`55/3,491 (1.58)
`Hy’s Law Cases
`6/3,560 (0.17)
`6/3,487 (17)
`Einstein Extension
`Rivaroxaban
`Enoxaparin
`ALT>3x ULN
`11/591 (1.86)
`3/586 (0.51)
`Hy’s Law Cases=ALT>3x ULN and TBL>2x ULN; HR=hazard ratio; R=rivaroxaban
`NOTE: HR in bold are statistically significant
`
`HR (R vs Control)
`
`0.92 (0.81,1.03)
`0.79 (0.64,0.97)
`0.95 (0.60,1.52)
`
`0.50 (0.37,0.69)
`0.44 (0.24,0.81)
`1.26 (0.89,1.79)
`0.97 (0.31,3.01)
`
`3.56 (0.99,12.76)
`
`
`For liver assessments of ALT and TBL in the Rocket studies, there was a large proportion of
`missing data after 1 year (~20%) and 2 years (~56%) in all treatment groups. Subject disposition
`in the Rocket study showed that there was around 35% early study medication discontinuation
`mainly due to adverse events and withdrawal of consent. The rate of early study medication
`discontinuation in the J-Rocket study was approximately 25% and was mostly due to adverse
`events (11%). Hence although subjects in the Rocket study were followed up to 4 years, most of
`the laboratory data were available up to 1 year with limited data available out to 2 years post-
`randomization.
`
`The numbers of reported hepatic adverse events (AEs) were generally balanced across all five
`studies except for ALT increased, aspartate aminotransferase (AST) increased, cholelithiasis,
`international normalized ratio (INR) increased, and liver function tests abnormal (based on
`MedDRA preferred terms (PTs)). In the Rocket study, the proportion of patients with either
`increased ALT or cholelithiasis was higher in the rivaroxaban arm compared to the warfarin arm.
`In contrast, the proportions of patients with increased ALT and increased AST were lower in the
`
`6
`
`Reference ID: 2959006
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`rivaroxaban arm compared to the control in the J-Rocket and Einstein DVT/PE studies,
`respectively; the proportion of patients with increased INR was lower in the rivaroxaban arm
`compared to control in the Rocket and Einstein DVT/PE studies; and the proportion of patients
`with abnormal liver function tests was lower in the rivaroxaban arm compared to control in the
`Einstein DVT study.
`
`The odds ratios (ORs) (95% confidence intervals) for the comparison of rivaroxaban versus
`active control in the proportion of patients with increased INR was 0.25 (0.20, 0.32) for the
`pooled Rocket and J-Rocket studies and 0.10 (0.05, 0.20) for the pooled Einstein DVT/PE
`studies, favoring rivaroxaban. In the pooled Rocket and J-Rocket studies, the time to onset of
`increased INR (among INR increased subjects only) was shorter in the warfarin arm compared to
`rivaroxaban (median days of 169 and 577 for warfarin and rivaroxaban, respectively).
`
`There are some inconsistencies in the results from the Rocket and Einstein studies, which might
`be associated with differences in randomized control treatments (warfarin for Rocket/J-Rocket
`and enoxaparin for Einstein DVT/PE, placebo for Einstein Extension), treatment durations (up to
`2-4 years for Rocket/J-Rocket and 1 year for the Einstein studies), rivaroxaban doses (10-30 mg
`daily) or study design (double-blind for Rocket and Einstein Extension and open-label for
`Einstein DVT/PE). In addition, none of the trials evaluated were designed primarily to assess
`liver toxicity. Note that the pooling of the studies in this review was not pre-specified but a
`rational for the pooling approach is discussed in section 3.3.4.
`
`In addition, findings from this review are consistent with results presented in an earlier statistical
`assessment (statistical review by Dr. Chava Zibman) that evaluated serious liver toxicity based
`on the RECORD 1-4 trials. It is important to note, however, that the dose (10 mg/daily) and
`duration (up to 35 days) of rivaroxaban studied in the RECORD 1-4 trials was lower and shorter,
`respectively, compared to the dose and duration of rivaroxaban studied in the Rocket and
`Einstein studies.
`
`The results of this review are subject to certain interpretation limitations. Please see section 5 for
`detailed discussion of these issues.
`
`In conclusion, findings from this review suggest that the liver toxicity profile based on
`assessment of evaluated ALT, TBL, and of reported hepatic events, of rivaroxaban is comparable
`to active controls, warfarin and enoxaparin, studied across four randomized clinical trials. No
`notable differences in these outcomes were identified in a single placebo-controlled trial versus
`rivaroxaban. Lastly, the incidence of Hy’s Law cases were similar between the rivaroxaban and
`warfarin arms in the Rocket trials; however, these events occurred earlier (within the first year)
`in the warfarin arm compared to the rivaroxaban arm.
`
`2.
`
`
`INTRODUCTION
`2.1 Overview
`
`
`NDA 22-406 for rivaroxaban (XARELTO) was originally submitted to the Division of
`Hematology Products (DHP) on July 28, 2008. The proposed indication is for prophylaxis of
`
`7
`
`Reference ID: 2959006
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`deep vein thrombosis (DVT) and pulmonary embolism (PE) in patients undergoing hip or knee
`replacement surgery. The proposed dose is 10 mg for daily oral intake for a recommended
`duration of 35 and 14 days for hip replacement surgery and knee replacement surgery
`respectively.
`
`During the original review cycle, a statistical safety review was completed by Dr. Chava Zibman
`on February 26, 2009 (available in DARRTS) on liver toxicity based on data from the RECORD
`1-4 randomized clinical trials submitted in support of the efficacy and safety of rivaroxaban. The
`safety population of these studies consisted of 6,183 adult patients receiving rivaroxaban and
`6,200 receiving enoxaparin as a prophylaxis for DVT and PE following total hip or total knee
`replacement surgery. The review concluded that a clear liver toxicity signal in rivaroxaban when
`compared to the randomized active control (enoxaparin) could not be determined.
`In addition,
`this review concluded that there was no clear excess of Hy’s Law cases in the rivaroxaban arm
`compared to control.
`
`On May 26, 2009, FDA issued a complete response (CR) letter (available in DARRTS) for NDA
`22—406 citing clinical and product quality deficiencies. A clinical deficiency noted was
`insufficient clinical data to fully characterize a potential risk for serious liver toxicity. The CR
`letter requested that the applicant provide an assessment of the potential for severe liver toxicity
`in the applicant’s ongoing clinical studies in patients with atrial fibrillation (the “Rocket”
`studies). The letter further requested information on patients from the ongoing clinical trials
`with reported elevations of serum ALT values greater than three times ULN and serum total
`bilirubin (TBL) values greater than twice ULN.
`
`The Rocket studies were included in rivaroxaban NDA
`
`(but) submission to the Division of
`(b) (4)
`
`Cardiovascular and Renal Products (DCRP). At the time of this review, NDA
`under review for the proposed indication of the prevention of stroke and systemic embolism in
`subjects with non-valvular atrial fibrillation.
`(m4)
`
`Unlike the RECORD studies (submitted in the original NDA 22—406), which evaluated
`rivaroxaban for 12 or 35 days, the Rocket studies were chronic dosing studies, i.e. >35 days of
`planned dosing, with treatment periods of up to 4 years for the Rocket study and up to 2 years for
`the J—Rocket study. Also, the RECORD studies evaluated rivaroxaban at a lower dose (10 mg
`daily) for a shorter treatment duration (at most 35 days) while the Rocket studies evaluated
`rivaroxaban at a higher dose (20 mg daily) for a longer treatment duration (>35 days and up to 4
`years).
`
`In response to a consult (received on February 11, 2011) received from DI-IP, this statistical
`safety review assessed data from the five rivaroxaban chronic dosing studies included in NDA
`(m4) (and referenced in the complete response for NDA 22-406): Rocket, J—Rocket, Einstein
`DVT, Einstein PE, and Einstein Extension. These studies were completed phase 3 studies except
`for Einstein PE which was ongoing at the time of this review. Table 2 provides a summary of
`the studies and descriptions for each study are as follows:
`
`Rocket (Study 11630) was a prospective, randomized, double-blind, double-dummy, parallel-
`group, active-controlled, multi-center, event-driven phase 3 study that compared the efficacy and
`8
`
`Reference ID: 2959006
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`safety of rivaroxaban versus warfarin for the prevention of stroke and non-central nervous
`system (CNS) systemic embolism in subjects with non-valvular atrial fibrillation. This study
`was conducted in North and Latin America, Europe, and Asia Pacific. Treatment duration was up
`to 4 years post-randomization.
`
`J-Rocket (Study 12620) was a prospective, randomized, double-blind, double-dummy, parallel-
`group, active-controlled, multi-center phase 3 study conducted in Japan that compared the
`efficacy and safety of rivaroxaban versus warfarin for the prevention of stroke and non-CNS
`systemic embolism in subjects with non-valvular atrial fibrillation. Treatment duration was up to
`2.5 years post-randomization.
`
`Einstein DVT (Study 11702) was part of a multi-center, randomized, open-label, assessor-blind,
`event-driven, non-inferiority program that evaluated the efficacy of rivaroxaban with study
`treatment durations of 3, 6, or 12 months compared with enoxaparin/Vitamin K anatagonist
`(VKA) therapy. The program consisted of two independent evaluations: (1) on subjects with
`confirmed acute symptomatic deep venous thrombosis (DVT) without symptomatic pulmonary
`embolism (PE) (Einstein DVT) and (2) on subjects with confirmed acute symptomatic PE with
`or without symptomatic DVT (Einstein PE).
`
`Einstein Extension (Study 11899) was a multi-center, randomized, double-blind, placebo-
`controlled, phase 3, event-driven, superiority study that evaluated the efficacy and safety of
`once-daily oral Factor Xa inhibitor rivaroxaban in the long-term prevention of recurrent
`symptomatic venous thromboembolism in subjects with symptomatic deep-vein thrombosis or
`pulmonary embolism after 6 months of anticoagulant therapy.
`
`
`
`
`
`Reference ID: 2959006
`
`9
`
`

`

`Study
`ROCKET
`(11630)
`J-ROCKET
`(12620)
`EINSTEIN
`DVT (11702)
`EINSTEIN PE
`(11702)
`Safety: 1,188 (598 R/590 P)
`EINSTEIN
`Rand: 1,197 c (602 R/ 595 P)
`Extension
`(11899)
`
`SE=systemic embolism; NVAF=non-valvular atrial fibrillation; DVT=deep venous thrombosis; PE=pulmonary embolism; VKA=vitamin K antagonist; E=enoxaparin
`Rand=all randomized subjects; R=rivaroxaban; W=warfarin; E=enoxaparin; P=placebo
`aIn Rocket, subjects with moderate renal impairment on entry to the study received 15 mg rivaroxaban; bIn J-Rocket, subjects with moderate renal impairment on entry to the study receive 10 mg
`rivaroxaban; cIn Einstein Extension, 632 subjects valid for safety were previously enrolled in the Einstein DVT/PE
`*Warfarin doses were titrated to a target INR range of 2.0-3.0, inclusive.
`
`Study Design
`Double-blind,
`active-controlled
`Double-blind,
`active-controlled
`Open label,
`active-controlled
`Open label,
`active-controlled
`
`Active
`Control
`Warfarin*
`
`Warfarin*
`
`Enoxaparin
`(1 mg/kg)/VKA
`Enoxaparin
`(1 mg/kg)/VKA
`
`Scheduled
`Treatment
`Study Population
`Duration
`Safety: 14,236 (7,111 R/7,125 W)
`Up to 4 yrs
`Rand: 14,269 (7,133 R/7,136 W)
`(exp. 32 mos)
`Up to 2.5 yrs Safety: 1,278 (639 R/ 639 W)
`Rand: 1,280 (640 R/ 640 W)
`Safety: 3,429 (1,718 R/1,711 E)
`Rand: 3,429 (1,718 R/1,711 E)
`Safety: 4,003 (2,011 R/1,992 E)
`Rand: 3,997 (2,010 R/1,987 E)
`
`3, 6, or 12
`months
`3, 6, or 12
`months
`
`Double-blind,
`placebo- controlled
`
`Placebo
`
`6 or 12
`months
`
`Table 2: Summary of Clinical Studies
`Proposed
`Rivaroxaban
`Indication (Subject
`Total Daily
`Population)
`Dose
`20 mg/15 mga
`Prevention of stroke and
`non-CNS SE (NVAF)
`Prevention of stroke and
`non-CNS SE (NVAF)
`Symptomatic DVT
`without symptomatic PE
`Symptomatic PE with or
`without symptomatic
`DVT
`DVT/PE after 6 months
`of anticoagulant therapy
`
`15 mg/10 mgb
`
`30 mg for 3 wks;
`then 20 mg
`30 mg for 3 wks;
`then 20 mg
`
`20 mg
`
`
`
`
`
`
`
`
`
`Reference ID: 2959006
`
`

`

`
`
`2.2 Data Sources
`
`
`Tabulation and analysis datasets were submitted for the Rocket and J-Rocket studies and only
`analysis datasets for the Einstein DVT, PE, and Extension studies.
`
`Pooled analysis datasets for both Rocket studies were available for laboratory liver function tests
`and liver-specific adverse events.
`
`For the Einstein studies, laboratory liver function tests were available but the datasets did not
`include a variable to indicate the laboratory in which the liver function tests were analyzed (i.e.
`central or local). A pooled Rocket and Einstein time-to-event dataset (adttelbp.xpt) that
`included a variable for the laboratory in which the sample was assessed was available and
`considered by the reviewer for analyses. Adverse events datasets were requested by the reviewer
`and submitted by the sponsor under NDA 22-406 on March 6, 2011.
`
`The clinical study reports and datasets analyzed in this review, including applicant responses to
`FDA information requests, are located in the CDER Electronic Document Room (EDR) at the
`links shown below. Note that only the pooled adverse events dataset (adae.xpt) was submitted to
`NDA 22-406; the laboratory and all other adverse events dataset were submitted to NDA
`
`
`
`\\CDSESUB1\EVSPROD\NDA022406
`5.3.5.3 ISS – Integrated Summary of Safety (ISS) for adverse events analysis dataset for
`Einstein DVT, Einstein PE, and Einstein Extension studies (adae.xpt)
`
`\\CDSESUB1\EVSPROD\NDA202439:
`5.3.5.3 isls - Integrated Summary of Liver Safety (ISLS) for summary clinical study reports
`(Rocket, J-Rocket, Einstein DVT, Einstein PE, and Einstein Extension studies), laboratory
`(adlbl.xpt) and adverse events (adae.xpt) analysis datasets for Rocket and J-Rocket, and pooled
`time-to-event analysis datasets (adttelbp.xpt) for all five studies.
`5.3.5.4 11702-einstein dvt for laboratory analysis dataset for Einstein DVT (liver.xpt)
`5.3.5.4 11702-einstein pe for laboratory analysis dataset for Einstein PE (liver.xpt)
`5.3.5.4 11899-einstein extension for laboratory analysis dataset for Einstein Extension
`(liver.xpt)
`
`3.
`
`
`STATISTICAL EVALUATION
`
`3.1 Data and Analysis Quality
`
`
`Original laboratory values for each liver function test parameter from centers in different regions
`were reported in varying units. The original laboratory values were converted to standardized
`values using a documented scheme. However, for the purpose of liver function tests
`assessments, a derived variable (“RATIO”) based on the ratio of the original laboratory value
`and the ULN of that laboratory value was created and used in the sponsor and reviewer analyses.
`
`The reviewer identified the following data-related issues or limitations:
`
`
`Reference ID: 2959006
`
`(b) (4)
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`• For original laboratory values reported with an inequality sign (‘<’ or ‘>’), only the numeric
`portion was used (e.g. for ‘<0.3 mg/dl’, the value 0.3 mg/dl was used) and original values
`reported with a ‘,’ were converted to ‘.’ (e.g. 34,77 was converted to 34.77). Upon visual
`inspection of the data, the reviewer did not identify many conversions from the original
`values with inequality signs or ‘,’ into standardized values for laboratory values of ALT>2x
`ULN (Note that elevations of ALT>3x ULN are of primary interest). That is, most
`conversions were performed by the sponsor for values of ALT<2x ULN. Furthermore, for
`those values that were converted, there did not appear to be any bias with regard to treatment
`assignment i.e. there were as many conversions in the rivaroxaban arm as there were in the
`active control arms for converted values of ALT>2x ULN.
`
` •
`
`
`
`In all datasets included in this review, the variable ‘TRTP’ for ‘Planned Treatment’ was the
`only variable available to distinguish treatment groups and was therefore used throughout all
`analyses in this report. Metadata files state that ‘TRTP’ represents actual treatment for the
`Einstein DVT, PE and Extension studies, and planned treatment for Rocket and J-Rocket
`studies. If the actual numbers of subjects who were assigned to treatment arms in the Rocket
`studies are very different from the planned treatment, then the results of the analyses using
`actual treatment will differ from those provided in this report which was based on planned
`treatment. In summary, the reviewer found it difficult to easily identify a common variable
`specific to treatment assignment among all available datasets.
`
`
`
`3.2 Evaluation of Efficacy
`
`This review focuses only on specific safety parameters measured in phase 3 studies. No
`assessment of efficacy was performed in this review.
`
`3.3 Evaluation of Safety
`
`3.3.1 Study Design and Endpoints
`
`Rocket (Study 11630) was a prospective, randomized, double-blind, double-dummy, parallel-
`group, active-controlled, multi-center, event-driven phase 3 study that compared the efficacy and
`safety of rivaroxaban with warfarin for the prevention of stroke and non-CNS systemic
`embolism in subjects with non-vulvar atrial fibrillation. The primary objective of this study was
`to demonstrate that the efficacy of rivaroxaban is non-inferior to the efficacy of dose-adjusted
`warfarin for the prevention of thromboembolic events in subjects with non-valvular AF as
`measured by the composite of stroke and non-CNS systemic embolism. The principal safety
`objective of this study was to demonstrate that rivaroxaban is superior to dose-adjusted warfarin
`as assessed by the composite of major and non-major clinically relevant bleeding events.
`
`Eligible subjects included those who had prior stroke (ischemic or unknown type), transient
`ischemic attack (TIA) or non-CNS systemic embolism, or who had at least 2 of the following
`risk factors: age ≥75 years, hypertension, heart failure and/or left ventricular ejection fraction
`≤35%, or diabetes mellitus. Subjects were randomized to one of the following two treatment
`groups:
`
`
`
`Reference ID: 2959006
`
`12
`
`

`

`NDA 22-406 (XARELTO (rivaroxaban))
`Statistical Safety Review of Potential risk for serious liver toxicity
`
`• Oral rivaroxaban 20 mg once daily + oral warfarin placebo once daily titrated to a target
`sham international normalized ratio (INR) of 2.5 (range 2.0-3.0, inclusive). Subjects
`with moderate renal impairment at screening (defined as calculated creatinine clearance
`between 30 to 49 mL/min, inclusive) had a dose adaptation to oral rivaroxaban 15 mg
`once daily.
`• Oral warfarin once daily titrated to a target INR of 2.5 (range