Kurzfassung: Abstract 2099 Poster Board II-76 Rivaroxaban is a new, oral, direct Factor Xa inhibitor with a linear dose-relationship between Factor Xa inhibition and drug concentrations in plasma. It also affects global clotting tests such as prothrombin time (PT), particularly when Neoplastin® is used for testing. Rivaroxaban was investigated in four randomized, double-blind phase III studies (RECORD1, 2, 3, and 4) for the prevention of venous thromboembolism (VTE) after total hip or knee replacement surgery. This post hoc, explorative analysis evaluated the relationship between PT and bleeding events from pooled RECORD1–4 data. Patients were randomized to receive oral rivaroxaban 10 mg once daily (od) starting 6–8 hours after surgery, or subcutaneous enoxaparin 40 mg od starting the evening before surgery (RECORD1–3), or 30 mg twice daily starting 12–24 hours after wound closure (RECORD4). RECORD1, 3, and 4 were head-to-head comparisons of rivaroxaban and enoxaparin; RECORD2 compared extended-duration rivaroxaban with short-duration enoxaparin. Pooled analysis of the results of these studies showed that rivaroxaban significantly reduced the incidence of symptomatic VTE and death compared with enoxaparin regimens. The relationship between PT (measured by Neoplastin reagent) and adjudicated bleeding events was evaluated in 6,040 subjects valid for the safety analysis who underwent surgery and had taken at least one dose of rivaroxaban. The day of the surgery was defined as day 1. PT was measured on day 0 (baseline) and day 6 (trough and peak, and grouped according to quartiles). All adjudicated bleeding events were analyzed with respect to the day 6 measurements. The analysis focused on treatment-emergent adjudicated bleeding events up to 12±2 days after the initiation of active rivaroxaban administration. Bleeding events were grouped as major and clinically relevant non-major bleeding (composite) or any bleeding. A total of 5,912 patients had PT measurements at baseline; 345 experienced any bleeding event, of which 149 events were major or clinically relevant non-major bleeding events. Baseline mean PT ranged from 13.5 (with no bleeding) to 14.0 seconds (with any bleeding). Mean trough values on day 6 ranged from 14.3 (no bleeding) to 15.3 seconds (any bleeding). Mean peak values for PT on day 6 (measured at 2–4 hours after dosing) ranged from 18.9 (no bleeding) to 20.5 seconds (any bleeding). The peak/baseline PT ratios were 1.42 (no bleeding events), 1.46 (with major and clinically relevant non-major bleeding events), and 1.49 (any bleeding events). Of the patients with bleeding events, most events occurred in those with PT values in the extreme (Q4) peak and trough quartiles (Table). This explorative analysis indicated that PT values (seconds) in patients with no bleeding or with different types of bleeding were similar for each time point (baseline, day 6 trough and peak). The reproducible effects of rivaroxaban on PT are consistent with its known pharmacokinetics in plasma. Although extreme PT trough and peak values were associated with an increased risk of bleeding, this analysis suggests that PT over a wide range is not an accurate predictor of bleeding events in individual patients undergoing total hip or knee replacement surgery receiving 10 mg rivaroxaban. Refinement of PT testing methods has focused specifically on vitamin K antagonist (VKA) monitoring; therefore PT testing may be of greater utility for monitoring VKAs than Factor Xa inhibitors. Table. Percentage of patients with bleeding events grouped by prothrombin time peaks and troughs according to quartiles in rivaroxaban-treated patients from the RECORD1–4 studies Prothrombin time range (seconds) Patients with bleeding events (%) 95% confidence interval Any bleeding Prothrombin time day 6 trough Q1 9.1–12.9 3.00 2.11–4.10 Q2 〉 12.9–13.6 3.43 2.40–4.74 Q3 〉 13.6–14.7 4.89 3.66–6.38 Q4 〉 14.7–36.4 8.17 6.60–9.96 Prothrombin time day 6 peak Q1 10.0–16.0 3.70 2.66–4.98 Q2 〉 16.0–18.4 3.48 2.47–4.74 Q3 〉 18.4–21.1 4.11 3.01–5.45 Q4 〉 21.1–70.4 8.09 6.52–9.87 Major and clinically relevant non-major bleeding Prothrombin time day 6 trough Q1 9.1–12.9 1.54 0.92–2.39 Q2 〉 12.9–13.6 1.86 1.12–2.89 Q3 〉 13.6–14.7 1.92 1.17–2.94 Q4 〉 14.7–36.4 3.71 2.66–5.02 Prothrombin time day 6 peak Q1 10.0–16.0 1.62 0.96–2.55 Q2 〉 16.0–18.4 2.20 1.41–3.25 Q3 〉 18.4–21.1 1.73 1.04–2.69 Q4 〉 21.1–70.4 3.44 2.43–4.70 Day 6 trough: value at pre-dose on day 6. Day 6 peak: value at 2–4 hours post-dose on day 6. Q, quartile. Disclosures: Haas: Bayer Healthcare: Honoraria; BMS: Honoraria; Sanofi-Aventis: Honoraria; CSL-Behring: Honoraria; Novartis: Honoraria. Turpie:Astellas: Consultancy; Takeda: Consultancy; GSK: Consultancy; sanofi-aventis: Consultancy; Bayer Schering Pharma AG: Consultancy, Speakers Bureau; Johnson & Johnson Pharmaceutical Research & Development, L.L.C: Consultancy; Portola: Consultancy. Lassen:Bayer Schering Pharma: Consultancy, Honoraria; sanofi-aventis: Consultancy; Astellas Pharma: Consultancy; Bristol-Myers Squibb: Consultancy; Pfizer Inc.: Consultancy. Kakkar:Aryx therapeutics: Consultancy; Eisai: Consultancy, Honoraria, Research Funding, Scientific Advisory Board; Bristol–Myers Squibb: Consultancy, Honoraria, Research Funding, Scientific Advisory Board; Pfizer: Consultancy, Honoraria, Research Funding, Scientific Advisory Board; Boehringer Ingelheim: Consultancy, Honoraria, Research Funding, Scientific Advisory Board; Sanofi-Aventis: Consultancy, Honoraria, Research Funding, Scientific Advisory Board; Bayer HealthCare: Consultancy, Honoraria, Research Funding, Scientific Advisory Board; GSK: Honoraria. Eriksson:Bayer Schering Pharma: Consultancy, Honoraria. Vogtlaender:Bayer Schering Pharma AG: Employment. Kubitza:Bayer Schering Pharma AG: Employment. Mueck:Bayer Schering Pharma AG: Employment. Misselwitz:Bayer Schering Pharma AG: Employment, Patents & Royalties. Berkowitz:Bayer HealthCare Pharmaceuticals : Employment.

Kurzfassung: Abstract 169 Introduction: Venous thromboembolism (VTE) is a common complication after hip or knee replacement and is associated with significant morbidity and mortality. Anticoagulants reduce the risk of these complications, but can also result in increased bleeding, thus requiring an assessment of benefit–risk. Patients and Methods: To obtain more precise estimates of treatment effects on low frequency events, data were pooled from 4 phase 3 randomized clinical trials (RECORD1–4) of rivaroxaban vs enoxaparin regimens (or enoxaparin/placebo combination in 1 study) for the prevention of deep vein thrombosis (DVT) and pulmonary embolism (PE) in total hip and knee replacement. Although the studies were designed to answer slightly different questions and differed in treatment duration and comparator dose, pooling was supported by similar study designs, identical endpoints, identical event ascertainment methods, and the same independent central blinded adjudication committees. Benefit–risk was assessed by comparing the excess number of outcome events for benefits vs that for harms (‘risks'), occurring over the treatment period. Excess number of events was defined as the number of events in a hypothetical population of 10,000 patients treated with enoxaparin minus the number of events in such a population treated with rivaroxaban. A positive value indicates that fewer events occur in the population treated with rivaroxaban. The analysis was undertaken for several clinically comparable pairs of composite benefit and harm outcomes: total VTE (any DVT, non-fatal PE, or death from any cause) vs major and clinically relevant non-major (CRNM) bleeding; major VTE (proximal DVT, non-fatal PE, or VTE-related death) vs major bleeding; and symptomatic VTE/all-cause mortality vs major bleeding. For each pair, pooled Mantel–Haenszel weighted risk differences were used to compute the excess number of benefit and harm events, and the difference between excess numbers of events was used to evaluate net clinical benefit (NCB; Table). In all calculations, benefits and risks were weighted equally. An additional assessment was performed using all treatment-emergent serious adverse events (SAEs) as reported by investigators. Results: Rixaroxaban is associated with statistically significantly fewer total VTE, major VTE, and symptomatic VTE/all-cause mortality events than enoxaparin, whereas enoxaparin is associated with a smaller number of different bleeding events, although no bleeding endpoints, other than the composite of major + CRNM bleeding, were statistically significantly different. In each pairwise comparison, the excess number of bleeding events is less than the excess number of VTE-related events by a factor ranging from 4 to 10 (Table). Enoxaparin was also associated with an excess of 194 treatment-emergent SAEs compared with rivaroxaban out of 10,000 patients. In all cases, there is a positive NCB for rivaroxaban vs enoxaparin with 95% confidence intervals (CIs) excluding 0, suggesting that the benefits of rivaroxaban exceed the risks when compared with enoxaparin. Conclusions: This quantitative benefit–risk approach provides a comparison of interventions in clinically relevant population terms. Using the net clinical benefit approach, for a variety of endpoints defining benefits and harms, the benefit–risk profile for rivaroxaban is consistently improved compared with enoxaparin for patients after elective hip and knee replacement. Disclosures: Levitan: Johnson & Johnson Pharmaceutical Research & Development, L.L.C.: Employment, Equity Ownership. Yuan:Johnson & Johnson Pharmaceutical Research & Development, L.L.C.: Employment, Equity Ownership. Turpie:Johnson & Johnson Pharmaceutical Research & Development, L.L.C.: Consultancy; Bayer Schering Pharma AG: Consultancy, Speakers Bureau; sanofi-aventis: Consultancy; GSK: Consultancy; Astellas: Consultancy; Takeda: Consultancy; Portola: Consultancy. Friedman:Boehringer Ingelheim : Consultancy, Research Funding; Johnson & Johnson : Consultancy; Astellas US: Consultancy, Research Funding; Sanofi-Aventis: Consultancy. Homering:Bayer Schering Pharma AG: Employment. Berlin:Johnson & Johnson Pharmaceutical Research & Development, L.L.C.: Employment, Equity Ownership. Berkowitz:Bayer HealthCare Pharmaceuticals: Employment. DiBattiste:Johnson & Johnson Pharmaceutical Research & Development, L.L.C.: Employment.

Kurzfassung: Rivaroxaban is an oral, direct Factor Xa inhibitor that has been evaluated in the RECORD phase III clinical trial program for the prevention of venous thromboembolism (VTE) in major orthopaedic surgery. RECORD3 (Lassen M, et al. N Engl J Med2008;358:2776–2786) showed that oral rivaroxaban 10 mg once daily (od) given post-operatively significantly reduced VTE after total knee replacement (TKR), compared with subcutaneous (s.c.) enoxaparin 40 mg od initiated pre-operatively, with similar rates of bleeding. RECORD4 was designed to determine the efficacy and safety of rivaroxaban compared with enoxaparin 30 mg administered twice daily after TKR. Patients (N=3,148) were randomized to receive either oral rivaroxaban 10 mg od (initiated 6–8 hours after surgery) or s.c. enoxaparin 30 mg every 12 hours (initiated 12 to 24 hours after surgery) for 10 to 14 days. Patients underwent mandatory, bilateral venography between day 11 and day 15. The primary efficacy endpoint was the composite of any deep vein thrombosis (DVT), non-fatal pulmonary embolism (PE), and all-cause mortality up to day 17. The primary efficacy analysis was a test for non-inferiority in the per-protocol population (n=1,702), followed by a test for superiority in the modified intention-to-treat population (n=1,924) (if non-inferiority was established in the per-protocol population). The main secondary efficacy endpoint was major VTE: the composite of proximal DVT, non-fatal PE, and VTE-related death. Treatment-emergent major bleeding observed no later than two days after the last intake of study was the main safety endpoint. The results are shown in the table. Rivaroxaban significantly reduced the incidence of the primary efficacy outcome compared with enoxaparin (6.9% vs 10.1%, respectively; p=0.012; relative risk reduction 31%). Rivaroxaban was non-inferior to enoxaparin for the prevention of major VTE in the per-protocol population (p & lt;0.001). The observed incidences of major VTE and symptomatic VTE in those receiving rivaroxaban or enoxaparin were 1.2% vs 2.0% (p=0.124), and 0.7% vs 1.2% (p=0.187), respectively, and the rates of major bleeding were 0.7% vs 0.3% (p=0.110) respectively, and major and clinically relevant non-major bleeding 3.0% vs 2.3% (p=0.179) in the rivaroxaban and enoxaparin treated groups, respectively. The data demonstrate that rivaroxaban has superior efficacy to enoxaparin 30 mg administered every 12 hours for the prevention of VTE after TKR, without significantly increasing the risk of bleeding. Endpoint Rivaroxaban10 mg od % (n/N) Enoxaparin30 mg q12h % (n/N) p-value for differenced aModified intention-to-treat population bModified intention-to-treat population valid for major VTE analysis cSafety population dCalculated for the absolute risk difference DVT, non-fatal PE, and all-cause mortalitya 6.9% (67/965) 10.1% (97/959) 0.012 Major VTEb 1.2% (13/1,122) 2.0% (22/1,112) 0.124 Symptomatic VTEc 0.7% (11/1,526) 1.2% (18/1,508) 0.187 Major bleedingc 0.7% (10/1,526) 0.3% (4/1,508) 0.110 Any non-major bleedingc 10.2% (155/1,526) 9.2% (138/1,508) – Major and clinically relevant non-major bleedingc 3.0% (46/1526) 2.3% (34/1,508) 0.179

Kurzfassung: The RECORD program comprised four pivotal phase III trials comparing the oral direct Factor Xa inhibitor, rivaroxaban, with subcutaneous enoxaparin for the prevention of venous thromboembolism (VTE) after total hip and knee replacement (THR and TKR). A total of 12,729 patients were randomized to receive rivaroxaban 10 mg once daily (od) starting 6–8 hours after surgery, or enoxaparin 40 mg od (RECORD1–3) starting the evening before surgery, or 30 mg every 12 hours (RECORD4) starting 12–24 hours after surgery. In RECORD1, patients undergoing THR received extended prophylaxis for 31–39 days. In RECORD2, patients undergoing THR received rivaroxaban for 31–39 days, or enoxaparin for 12±2 days followed by placebo up to 35 days. Patients undergoing TKR in RECORD3 and 4 received prophylaxis for 12±2 days. In all four trials the rivaroxaban regimens showed superior efficacy to the enoxaparin regimens with no significant increase in bleeding. The RECORD1–4 pooled analysis showed that rivaroxaban was significantly superior to enoxaparin for the primary endpoint, symptomatic VTE and all-cause mortality (0.8% vs 1.6%, respectively; p & lt;0.001), with similar bleeding rates in both groups. Pooled subgroup analyses were performed to determine the influence of age, weight, gender, or renal function (calculated creatinine clearance [CrCl] at baseline) on the studies primary and main secondary efficacy outcomes, and any adjudicated bleeding events in RECORD1–4. Treatment effect (rivaroxaban vs enoxaparin regimens) was expressed on the odds-ratio (OR) scale for total VTE (the composite of any deep vein thrombosis, non-fatal pulmonary embolism and all-cause mortality, modified-intention-to-treat [mITT] population) and major VTE (the composite of proximal deep vein thrombosis, non-fatal pulmonary embolism and VTE-related death, mITT population for major VTE) at the end of the planned medication period. For any bleeding (major and non-major bleeding after the start of study medication up to 2 days after the last dose, safety population) the hazard ratio (HR) was used. Separate analyses were done for each subgroup. These pooled analyses show consistently superior efficacy to enoxaparin, irrespective of patients’ age ( & lt;65, 65–75, & gt;75 yrs), weight (≤70, & gt;70–90, & gt;90kg), gender, or renal function (CrCl & gt;80, 50–80, & lt;50 ml/min). Interaction testing in logistic regression models provided no indication of treatment effect differences in these subgroups, except for total VTE and the CrCl subgroups, where treatment effect with rivaroxaban appeared to be larger for the & gt;80 and & lt;50 ml/min group compared with the 50–80 ml/min group. With respect to any bleeding, the risk on rivaroxaban relative to enoxaparin regimens was comparable across gender and the different age, body weight and renal function ( & gt;30 ml/min CrCl) groups studied. Similar trends were seen in the smaller subgroups of body weight extremes ( & gt;110 kg subgroup - total VTE: 5/100 [5.0%] vs 8/119 [6.7%] , respectively; ≤50 kg subgroup - any bleeding: 12/147 [8.2%] vs 8/162 [4.9%] , respectively). These results suggest that age, body weight, gender and renal function ( & gt;30 ml/min CrCl) have no clinically relevant effect on the efficacy or safety of rivaroxaban as thrombroprophylaxis following major orthopedic surgery. Subgroup Total VTE Odds ratio (rivaroxaban vs enoxaparin regimens) (95% CI Major VTE Odds ratio (rivaroxaban vs enoxaparin regimens) (95% CI) Any major or non-major bleeding Hazard ratio (rivaroxaban vs enoxaparin regimens) (95% CI) *Because of study exclusion criteria, few patients had CrCl & lt;30 ml/min, CI = confidence interval Age & lt;65 years 0.40 (0.28–0.55) 0.22 (0.10–0.44) 1.05 (0.86–1.29) 65–75 years 0.39 (0.29–0.52) 0.18 (0.08–0.34) 1.11 (0.90–1.38) & gt;75 years 0.54 (0.36–0.81) 0.51 (0.21–1.15) 1.10 (0.79–1.52) Weight ≤70kg 0.43 (0.31–0.60) 0.25 (0.11–0.52) 1.17 (0.91–1.51) & gt;70–90kg 0.45 (0.34–0.59) 0.21 (0.11–0.39) 0.95 (0.77–1.17) & gt;90kg 0.38 (0.25–0.57) 0.32 (0.13–0.69) 1.22 (0.95–1.58) Gender Male 0.35 (0.25–0.48) 0.21 (0.10–0.40) 1.12 (0.93–1.36) Female 0.47 (0.37–0.59) 0.27 (0.16–0.45) 1.06 (0.87–1.28) CrCl (ml/min) & gt;80 0.35 (0.27–0.46) 0.22 (0.12–0.38) 1.12 (0.94–1.35) 50–80 0.60 (0.45–0.81) 0.30 (0.15–0.56) 0.99 (0.79–1.24) & lt;50* 0.32 (0.15–0.64) 0.23 (0.02–1.15) 1.07 (0.64 – 1.79)