CENTER FOR DRUG EVALUATION AND RESEARCH APPLICATION NUMBER: 022406Orig1s000 MEDICAL REVIEW(S) M E M O R A N D U M DEPARTMENT OF HEALTH AND HUMAN SERVICES PUBLIC HEALTH SERVICE FOOD AND DRUG ADMINISTRATION CENTER FOR DRUG EVALUATION AND RESEARCH Date: June 13, 2011 From: Kathy M. Robie Suh, M.D., Ph.D. Medical Team Leader, Hematology Division of Hematology Drug Products (HFD-160) Subject: Medical Team Leader Secondary Clinical Review NDA 22-406 resubmission, letter date 1/14/2011; received 1/14/2011 XARELTOR (rivaroxaban) for prophylaxis of deep vein thrombosis (DVT) and pulmonary embolism (PE) in patients undergoing hip or knee replacement surgery To: NDA 22-406 Xarelto (rivaroxaban) Tablets is an orally administered Factor Xa inhibitor being developed for several anticoagulation indications. In this NDA application the sponsor is seeking initial marketing approval of rivaroxaban for the indication: for the prophylaxis of deep vein thrombosis (DVT) and pulmonary embolism (PE) in patients undergoing knee or hip replacement surgery. The proposed dose is 10 mg orally once daily with a treatment duration of 14 days for knee surgery and 35 days for hip surgery. This is the second review cycle for this drug product. See my Medical Team Leader/CDTL review dated 5/27/2009 for background and summary of cycle 1 review findings. Briefly, the database consisted of four trials (the RECORD 1, 2, 3, and 4 studies), each comparing rivaroxaban to enoxaparin (different regimens) with two studies for knee surgery and two studies for hip surgery. All four studies were multinational, randomized (1:1), double-blind, double- dummy, parallel groups design. The studies were conducted by Bayer but the right of reference for use of the studies was transferred to Johnson & Johnson (J & J) just prior to NDA submission and J & J is the sponsor of the NDA. The first review cycle found convincing statistical evidence for efficacy of rivaroxaban in the two proposed clinical settings based on the primary efficacy endpoint total VTE (composite of any DVT [venographically determined], non-fatal PE or death) and concluded that, barring gross irregularities in conduct of the RECORD studies, rivaroxaban has efficacy as an anticoagulant for thromboprophylaxis in the settings of elective hip replacement and knee replacement surgery. The predominant effect appeared to be on venographically-detected DVT. The safety data suggested that bleeding rates may be somewhat greater with rivaroxaban than with enoxaparin. Though the available data did not identify a risk for hepatotoxicity, most of these data were from short-term studies and insufficient data were available to rule out a risk for hepatotoxicity with Reference ID: 2960076 NDA 22-406 Page 2 of 8 longer term use of rivaroxaban. Chronic use is a concern for rivaroxaban, since as an oral agent it can be reasonably expected to have some long-term use in practice for treatment of chronic indications, such as stroke prevention in patients with atrial fibrillation. A meeting of the Cardiovascular and Renal Drugs Advisory Committee on March 19, 2009 voted overwhelmingly that a favorable benefit-risk profile had been demonstrated for use of rivaroxaban in the prophylaxis of venous thromboembolism (VTE) in patients undergoing hip or knee replacement surgery, but voiced some concerns about the strength of the signals for hepatotoxicity and the feasibility of long-term studies to further elucidate the hepatotoxicity potential. Subsequent to the advisory committee meeting, findings of the Division of Scientific Investigations (DSI) inspections of several sites, particularly in RECORD 4, identified deficiencies with regard to compliance with study procedures, completeness in reporting of adverse events and other irregularities during the conduct of the RECORD studies raising questions about the adequacy of study monitoring by Bayer and necessitating further examination of the integrity of the studies by DSI. On May 27, 2010 the Agency issued a Complete Response (CR) letter to Johnson & Johnson (Appendix B) citing results from the DSI clinical investigator inspections indicating that some sites may be unreliable and results from the sponsor (Bayer) inspection revealing that “the sponsor failed to 1)ensure proper monitoring of the study, 2)to ensure that study was conducted in accordance with the protocol and/or investigational plan, and 3)to ensure that FDA and all investigators were promptly informed of significant new adverse effects or risks.” The sponsor was requested to provide a detailed report of their clinical quality assurance (QA) audit plan including plan for securing investigator compliance, audit findings, corrective actions including termination of investigators, oversight of CROs and Bayer handling of review information obtained from the CROs. The sponsor was asked to plan and perform an additional audit and provide a full report. Also, in the CR letter the sponsor was informed that the supplied clinical data were insufficient to fully characterize a potential risk for serious liver toxicity. The sponsor was asked to provide additional long-term safety data from the studies of rivaroxaban in patients with atrial fibrillation (ROCKET studies), post-marketing experience outside the U.S., final reports for other completed long-term treatment studies and summary of post-marketing studies initiated outside the U.S. The CR letter also listed Chemistry, Manufacturing and Controls (CMC) deficiencies that needed to be addressed prior to product approval of the application. In the resubmission the sponsor has provided a full response the CR letter. The submission includes: • detailed information regarding the auditing and monitoring procedures and results for the RECORD studies (including the Bayer QA audit plans, SOPs, audit findings and corrective actions, and list of clinical investigators terminated for noncompliance for the RECORD studies; information on each CRO hired to monitor the clinical sites for the RECORD program, the oversight process for each CRO, and processes to ensure monitoring, as well as a list of noncompliant sites reported by the CROs; and the Bayer audits and the independent third party (b) (4) audits methodology, results and analyses, also including the results of the Sponsor’s RECORD 4 study data verification). This information was reviewed by the 2 Reference ID: 2960076 NDA 22-406 Page 3 of 8 Division of Scientific Investigations (DSI) (See Compliance Review by Susan Thompson, M.D., 5/25/2011). • Background information on the Hepatic Event Assessment Committee (HEAC) and a comprehensive Integrated Summary of Liver Safety • Safety findings from post-marketing exposures outside the United States • The protocol and safety findings from the observational post marketing study XAMOS • The final study report for ATLAS ACS TIMI 46, including electronic datasets • Reference to previous submissions indicating and including information about updated Bayer and J & J DMFs in support of the application. Information to address the deficiencies in drug substance information, drug product specifications, dissolution criteria, stability data and description of container and closure system information as requested in the CR letter. (See review by Chemistry, Manufacturing and Controls (CMC) (Joyce Z. Crich, Ph.D., 6/02/2011). • A study synopsis and proposal to conduct a Phase 1 drug interaction study (RIVAROXACS1001) in patients with mild or moderate renal impairment concomitantly receiving erythromycin, a moderate CYP3A4/moderate P-gp inhibitor to address the Cinical Pharmacology recommendation that the sponsor develop a lower strength formulation for use in dose modification in certain populations of patients. The sponsor’s arguments and information have been reviewed by Clinical Pharmacology (Joseph Grillo, Pharm.D. and Nitin Mehrotra, Ph.D., 6/03/2011). Additional information was provided upon Division’s request during the review cycle regarding occurrences of events of agranulocytosis, Stevens-Johnson syndrome, and neuraxial hematoma within the safety database of rivaroxaban (including all completed and ongoing clinical trials up to date and post-marketing experience). Clinical review of the new data and updated safety information in the resubmission has been completed by Min Lu, M.D. (review signed 6/03/2011). Please see Dr. Lu’s review for detailed presentation of the new and updated safety data and other information in the application. Important in the clinical review of the resubmission were the findings of the DSI review of audit and monitoring information for the four RECORD studies. See Dr. Thompson’s Compliance Review (signed 5/25/2011) for detailed description and analysis of the audit results. All four RECORD studies were conducted by Bayer. Monitoring oversight for RECORD 1, 2 and 3 was done by Bayer while RECORD 4 was monitored by (b) (4) a contract research organization (CRO). After receipt of the CR letter, J & J contracted (b) (4) for an independent third party audit of the RECORD studies. This audit examined 30/617 sites encompassing 945/12729 (7.4%) of enrolled patients across the four studies. (For the individual studies percentage of sites examined ranged from 2.0% for RECORD 3 to 6.9% for RECORD 4 and percentage of patients from 2.8% for RECORD 3 to 9.9% for RECORD 4). DSI review of the audit information found deficiencies in all four studies. The DSI review comments that the audits of the RECORD 1, 2, and 3 studies found deficiencies in drug accountability but did not demonstrate systematic deficiencies in multiple aspects of the conduct of the trials that would bring into question data integrity from all study sites. However, for RECORD 4 the deficiencies in terms of both extent and scope were more pervasive than in the other three studies. Adequacy of monitoring was assessed as ineffective in identifying significant violations or in taking actions towards securing compliance. In the 3 Reference ID: 2960076 NDA 22-406 Page 4 of 8 RECORD 4 study 63.1% of audited patients were found to be inadequately monitored as compared to 25.5%-40.0% of audited patients in the other three studies.