Dose–response study of recombinant human soluble thrombomodulin (ART-123) in the prevention of venous thromboembolism after total hip replacement

Authors


C. Kearon, Hamilton Health Sciences, Henderson Division, 711 Concession Street, Hamilton, Ontario, L8V 1C3, Canada.
Tel.: +1 905 383 2252; fax: +1 905 575 7320; e-mail: kearonc@mcmaster.ca

Abstract

Summary. Background: Recombinant human soluble thrombomodulin (ART-123) is composed of the active, extracellular, domain of thrombomodulin. ART-123 binds to thrombin and this complex converts protein C into the natural anticoagulant activated protein C. This study was performed to identify an effective and safe dose of ART-123 for prevention of venous thromboembolism after elective, unilateral total hip replacement. Methods and results: An open-label, sequential, dose-ranging study was performed in which 312 patients received either 0.3 mg kg−1 or 0.45 mg kg−1 of ART-123, subcutaneously, 2–4 h after surgery (day 1). Those who received 0.3 mg kg−1 were given a second dose of 0.3 mg kg−1 on day 6, and the first 29 of these patients also used intermittent pneumatic compression devices. Those who received 0.45 mg kg−1 were not given a second dose. Primary efficacy outcome was all deep vein thrombosis on mandatory bilateral venography performed on day 9 ± 2 and symptomatic venous thromboembolism up to day 11. Primary safety outcome was major bleeding up to day 11. Among patients who did not use intermittent pneumatic compression, venous thromboembolism occurred in 3.4% of 116 evaluable patients in the 0.3 mg kg−1 group and 0.9% of 111 patients in the 0.45 mg kg−1 group. Major bleeding occurred in 1.4% of 139 patients in the 0.3 mg kg−1 group and 6.3% of 144 patients in the 0.45 mg kg−1 group. Conclusion: ART-123 is a highly effective antithrombotic agent that should be directly compared with current methods of prophylaxis in patients who have major orthopedic surgery.

Introduction

Thrombomodulin is a mostly endothelial cell, membrane-bound glycoprotein that has extracellular, transmembrane, and cytoplasmic domains [1,2]. Thrombin binds to the extracellular domain of thrombomodulin and undergoes a conformational change that converts thrombin from a procoagulant enzyme to one that can activate protein C and thereby inhibit coagulation. The extracellular moiety of thrombomodulin that binds thrombin to the endothelial cell surface has been made using recombinant DNA technology and is termed soluble thrombomodulin (ART-123) [3–5]. Like naturally occurring thrombomodulin, ART-123 binds to thrombin, blocks thrombin's procoagulant activity, and converts thrombin to an activator of the protein C pathway, leading to inhibition of coagulation through inactivation of factor Va and factor VIIIa [6]. ART-123 has a long half-life (about 48–72 h) and high bioavailability after subcutaneous injection (about 66%), is active against clot-bound thrombin, and has been shown to have a favorable antithrombotic to hemorrhagic profile in animal studies [3–6]. These qualities suggest that it may be suitable as an antithrombotic agent.

Venous thromboembolism (VTE) is a common complication in patients who undergo elective hip replacement. Without prophylaxis, approximately 55% of patients will develop deep vein thrombosis (DVT), 1.5% will develop symptomatic pulmonary embolism (PE), and 0.5% will die from PE [7]. This study was designed to determine if ART-123 could reduce the frequency of VTE after hip replacement and to identify a suitable dosing regimen for this purpose.

Methods

Patients

Patients 18 years or older undergoing elective, unilateral total hip replacement were potentially eligible. Preoperative exclusion criteria were: previous VTE; active bleeding or previous history of abnormal bleeding; uncontrolled hypertension; clinically significant abnormalities in hematological or biochemical measurements; significant renal dysfunction; known FV Leiden, or protein C or S deficiency; inability to undergo venography (including contrast allergy); requirement for anticoagulation (e.g. atrial fibrillation); known to be pregnant; receiving another investigational drug; and weight of > 135 kg. Postoperative exclusions were: use of an epidural catheter for anesthesia or analgesia, traumatic spinal anesthesia, and failure to achieve primary hemostasis.

All patients provided written informed consent, and each was considered to be enrolled after he or she received one dose of ART-123. The study was conducted according to the International Good Clinical Practice Guidelines and was approved by Research Ethics Boards of each clinical center.

Study intervention and sequential dose-adjustment

The original intention of the study was to perform a Phase II, open-label, sequential, dose-ranging study evaluating four dosing regimens without a comparator group. Patients in the first group were to be treated with ART-123 at a dose of 0.03 mg kg−1, given by subcutaneous injection, 2–4 h after surgery (day 1) and after 5 days (day 6). Depending on the observed frequency of VTE and major bleeding in this group, the following escalating dosing regimens would then be tested: 0.1 mg kg−1 on days 1 and 6; 0.3 mg kg−1 on days 1 and 6; and 0.45 mg kg−1 on day 1 only. In response to a request from a regulatory agency, the dosing schedule was changed to start with the 0.3 mg kg−1 regimen, initially with concomitant use of bilateral intermittent pneumatic compression devices. The rationale for this modification was concern that ART-123 might be ineffective, particularly using the two lowest dose regimens. Once efficacy of the 0.3 mg kg−1 and intermittent pneumatic compression regimen was demonstrated by a low frequency of VTE, use of intermittent pneumatic compression was permanently stopped and subsequently enrolled patients only received ART-123 for prophylaxis.

By use of the principle of maximizing efficacy with acceptable safety, dose-adjustment decisions were made by the Steering Committee on the basis of regular review of the clinical center and centrally adjudicated rates of overall DVT and major bleeding. These rates were interpreted in the context of expected rates of DVT (i.e. 10–15%) and major bleeding (i.e. 2–3%) when low-molecular-weight heparin is used for prophylaxis [7,8]. Decisions regarding dose adjustment were guided by a Standardized Operational Procedure that was approved by participating Research Ethics Boards.

With the exceptions of graduated (non-pneumatic) compression stockings (acceptable but discouraged), and intermittent pneumatic compression devices in the first group of patients, additional prophylactic methods were not permitted. Patients without symptomatic thrombosis who received any anticoagulant therapy other than ART-123 before having mandatory venography were excluded from the efficacy analysis. Concomitant antiplatelet therapy was discouraged but allowed at the discretion of the attending physician.

Patient surveillance

While in hospital, patients were examined daily for evidence of thrombosis and bleeding, and laboratory investigations were performed to monitor patient safety. Before discharge, patients were instructed in the signs and symptoms of DVT, PE and bleeding, and were asked to contact the investigator if any of these symptoms developed. Appropriate investigations would then be performed. A final clinic visit was completed for all patients at day 30 to day 35.

Outcome measures

Primary efficacy outcome was a composite of overall DVT diagnosed on mandatory bilateral and technically adequate venography on day 9 ± 2, and confirmed symptomatic DVT or PE up to day 11. Confirmed symptomatic DVT or PE that occurred after day 11 up to day 35 was considered a secondary efficacy outcome event. For suspected symptomatic DVT and PE, venous ultrasonography, ventilation-perfusion lung scanning, spiral computed tomography, venography and/or pulmonary angiography, were performed as appropriate. Findings were interpreted by the Independent Central Adjudication Committee using standard criteria [9,10].

Primary safety outcome was major bleeding up to day 11. A bleeding event was defined as clinically overt bleeding in excess of the usual amount expected for the surgical procedure. To qualify as a major bleeding episode, one or more of the following criteria also had to be satisfied: a fall in hemoglobin of ≥ 20 g L−1; packed red-cell transfusion of ≥ 2 units with a bleeding index of ≥ 2.0; or bleeding in a critical site, such as retroperitoneal, intraocular, spinal, or pericardial bleeding. The bleeding index was calculated as: number of units of red cells transfused + [hemoglobin prebleed (g dL−1) − hemoglobin postbleed (g dL−1)]. All other clinically overt bleeding events were considered to be minor. Details of unusual bleeding were reported by the investigators and were adjudicated by the Independent Central Adjudication Committee.

Routine assays of coagulation, hematological and biochemical variables were performed before surgery and serially after surgery. This included a bleeding index calculation for each patient using hemoglobin measurements that were obtained on day 1 (before first dose of ART-123) and day 6, which served as a routine assessment of postoperative blood loss. Testing for antibodies to ART-123 was performed on blood obtained at the final visit (day 30 to day 35) using a validated ELISA-based assay.

Avoidance of bias

All mandatory venograms, suspected DVT or PE events, and bleeding events were reviewed by a Independent Central Adjudication Committee, whose members were blinded to the regimen of ART-123 and to local interpretation of the events by the clinical centers.

Statistical analysis

The proportion of patients with VTE, proximal DVT or PE, major bleeding, and all bleeding, with their 95% confidence intervals using the score interval method for binomial proportions [11], was calculated for each regimen and for all patients who did not receive intermittent pneumatic compression (two groups combined).

Results

Patients were recruited from 20 clinical centers from June 2002 through April 2003. A total of 345 patients met the preoperative eligibility criteria and provided informed consent. Of these, 23 were excluded postoperatively for the following reasons: placement of an epidural catheter (five patients); excessive bleeding during or immediately after surgery (five patients); traumatic spinal anesthesia (two patients); intercurrent illness (two patients); did not have surgery (one patient); late discovery of allergy to contrast media (one patient); need for leg traction (one patient); missed 4-h treatment window (one patient); study closed (one patient); and the reason was not specified in four patients. Another 10 patients withdrew consent postoperatively, leaving 312 patients who were included in the study (Fig. 1).

Figure 1.

Patient disposition.

Enrollment sequence

Initially, patients were enrolled to receive ART-123 at a dose of 0.3 mg kg−1 on days 1 and 6 in conjunction with intermittent pneumatic compression devices. A low frequency of VTE was documented in the first 29 patients and use of intermittent pneumatic compression devices was permanently stopped. Of this group of 29 patients, two did not receive ART-123 on day 6 due to adverse events (Fig. 1).

Patients were then consecutively enrolled to receive ART-123 at the same dose of 0.3 mg kg−1 on days 1 and 6 but without use of intermittent pneumatic compression. After 97 patients were enrolled, based on an interim review of outcome rates, enrollment to this group was suspended and subsequent patients received a single injection of ART-123 on day 1 at a dose of 0.45 mg kg−1 without intermittent pneumatic compression. After 144 patients were studied at this dose, enrollment to the 0.45 mg kg−1 regimen was permanently stopped based on another interim analysis and enrollment to the second regimen (0.3 mg kg−1 without intermittent pneumatic compression) was resumed. Forty-two additional patients were enrolled to receive 0.3 mg kg−1 on days 1 and 6 without intermittent pneumatic compression, resulting in a total of 139 patients in this group (Fig. 1). Of these 139 patients, five did not receive ART-123 on day 6 due to adverse events.

Baseline and surgical characteristics

Three hundred and twelve patients underwent total hip arthroplasty and received at least one dose of ART-123 (Table 1). Mean age of all patients was 64 years, 51% were male, and mean weight was 84 kg. Surgery lasted a mean of 136 min, mean blood loss during surgery was 426 mL, and the mean immediate postoperative hemoglobin was 11.5 g dL−1.

Table 1.  Baseline characteristics
CharacteristicART-123 Regimen, mg kg−1 0.3 days 1 and 60.45 day 1 No IPC
(n = 144)
All (n = 312)
IPC* (n = 29)No IPC (n = 139)
  • *

    IPC, Intermittent pneumatic compression.

  • a

    n = 28;

  • b

    n = 136;

  • c

    n = 143;

  • d

    n = 307.

Mean age, years (SD)62.4 (12.1)62.7 (12.12)65.2 (12.1)63.8 (12.2)
Female, n (%)12 (41)72 (52)68 (47)152 (49)
Mean weight, kg (SD)84.1 (18.9)82.7 (16.4)85.5 (18.0)84.2 (17.4)
Previous venous thromboembolism, n (%)002 (1.4)2 (0.6)
Active cancer, n (%)02 (1.4)02 (0.6)
Anesthesia, n (%)
 General22 (76)84 (60)103 (72)209 (67)
 Regional/spinal7 (24)53 (38)41 (28)101 (32)
 General and regional/spinal02 (1)02 (1)
Mean length of surgery, min (SD)129 (31)138 (47)135 (52)136 (48)
Mean operative blood loss, mL (SD)400 (254)437 (367)422 (315)426 (333)
Mean immediate postoperative hemoglobin, g dL−1 (SD)11.2a (2.0)11.5b (1.6)11.5c (1.5)11.5d (1.6)

DVT and symptomatic PE

Of the 312 patients who received study drug, 61 (20%) could not be evaluated for the primary efficacy analysis. Seventeen patients did not have their mandatory bilateral venograms due to patient refusal (five patients), failed venous access (seven patients), or medical contraindications (five patients), and 35 patients had venograms performed that were considered to be inadequate for evaluation by the Independent Central Adjudication Committee because only one leg was successfully studied (16 patients; 11 of the failures were of non-operated leg), not all deep veins were well visualized (16 patients), or for an unspecified reason (three patients). Another nine patients were not included in the efficacy analysis because they received additional anticoagulants (none for treatment of confirmed VTE) before mandatory venography (none had VTE). On average, mandatory venography was performed on day 8.7 (days 8.6–8.8 for each of the three groups).

Table 2 summarizes rates of thromboembolic outcomes for each of the three groups. In total, there were five episodes of confirmed VTE, of which four were asymptomatic distal DVT diagnosed by mandatory venography, and one was a symptomatic, non-fatal, PE on day 2. There were no episodes of VTE between day 11 and day 35. Among patients who did not receive intermittent pneumatic compression, the rate of VTE was somewhat lower in the group that received 0.45 mg kg−1 on day 1 only [0.9%; 95% confidence interval (CI) 0.2, 4.9] than in the group that received 0.3 mg kg−1 on days 1 and 6 (3.4%; 95% CI 1.3, 8.5). The overall frequency of VTE among the two groups of patients who received ART-123 without intermittent pneumatic compression was 2.2% (95% CI 0.9, 5.0).

Table 2.  Rates of venous thromboembolism with each regimen in patients evaluable for efficacy analysis
 ART-123 regimen, mg kg−1 0.3 days 1 and 60.45 day 1 No IPC
(n = 111)
IPC (n = 25)No IPC (n = 116)
  1. IPC, Intermittent pneumatic compression. *Symptomatic pulmonary embolism.

All venous thromboembolism, n (%, 95% CI)0 (0, 0.0, 13.3)4 (3.4, 1.3, 8.5)1 (0.9, 0.2, 4.9)
Proximal deep vein thrombosis or pulmonary embolism, n (%, 95% CI)0 (0, 0.0, 13.3)1* (0.9, 0.2, 4.7)0 (0, 0.0, 3.3)

Bleeding

Table 3 summarizes rates of major bleeding and all bleeding up to day 11, and the routinely calculated bleeding index from day 1 to day 6, in each group. Among patients who did not receive intermittent pneumatic compression, the rate for bleeding appeared to be higher in the group that received 0.45 mg kg−1 on day 1 [major bleeding, 6.3% (95% CI 3.3, 11.5); all bleeding, 8.3%] than in the group that received 0.3 mg kg−1 on day 1 and day 6 [major bleeding, 1.4% (95% CI 0.4, 5.1); all bleeding, 2.2%]. One episode of bleeding occurred in a patient who had received additional anticoagulation; this patient, who was in the 0.45 mg kg−1 group, had a major bleed 4 days after starting therapeutic doses of low-molecular-weight heparin for a presumptive symptomatic DVT that was not confirmed by the Independent Central Adjudication Committee.

Table 3.  Bleeding with each regimen
 ART-123 regimen, mg kg−1
0.3 day 1and day 60.45 day 1
IPC (n = 29)No IPC (n = 139)No IPC (n = 144)
  1. IPC, Intermittent pneumatic compression; SD, standard deviation for individual values. Bleeding index = [hemoglobin (g dL−1) before first dose of ART-123 on day 1 − hemoglobin on morning of day 6] + number of units of blood transfused between the two hemoglobin measurements.

Major bleeds, n (%, 95% CI)1 (3.4, 0.6, 17.2)2 (1.4, 0.4, 5.1)9 (6.3, 3.3, 11.5)
All bleeds, n (%, 95% CI)2 (6.9, 1.9, 22.0)3 (2.2, 0.7, 6.2)12 (8.3, 4.8, 14.0)
Bleeding index (SD)1.8 (2.0)1.7 (1.7)1.9 (1.7)

There was no apparent difference in postoperative blood loss among the three groups as estimated by the routinely calculated bleeding index from day 1 to day 6.

Deaths

No deaths occurred during the course of the study or within 30 days of last study drug administration.

Laboratory results

Compared with preoperative values, mean prothrombin time or activated partial thromboplastin time on day 1 (before first ART-123 dose), day 2 (morning after first dose) or day 6 (fifth morning after first dose) did not decrease by more than 1.5 s or increase by more than 2.2 s for any of the three groups. There were no episodes of hepatic, renal or hematological toxicity attributable to ART-123. Of 298 patients who were assessed, two patients (0.7%) developed antibodies to ART-123; in both cases, these antibodies did not inhibit activation of protein C by complexes of ART-123 and thrombin, and were not associated with clinical manifestations.

Discussion

This study has a number of important findings. First, the rates of VTE of 3.4% (95% CI 1.3, 8.5) and major bleeding of 1.4% (95% CI 0.4, 5.1) among patients who received ART-123 at a dose of 0.3 mg kg−1 without intermittent pneumatic compression compare very favorably with rates of these outcomes that have recently been observed after elective hip replacement with use of low-molecular-weight heparin [12–17], warfarin [12], fondaparinux [13,15], or ximelagatran [14,16,17] (Table 4). Therefore, this dosing regimen of ART-123 appears to be very effective and safe. Second, whereas bleeding was infrequent with 0.3 mg kg−1 of ART-123 on days 1 and 6, the rate of bleeding was high after a dose of 0.45 mg kg−1 on day 1. This suggests that doses of ART-123 > 0.3 mg kg−1 should be avoided if therapy is started 2–4 h after surgery. Although there was a suggestion that the dosing regimen of 0.45 mg kg−1 on day 1 may have been more effective than that of 0.3 mg kg−1 on days 1 and 6, there were too few episodes of venous thrombosis in both groups to identify a difference in efficacy. Third, ART-123 was well tolerated, without evidence of hepatic, renal, or hematological toxicity.

Table 4.  Venous thromboembolism and major bleeding after elective hip replacement in recent trials evaluating venous thrombosis prophylaxis
 InterventionVTEMajor bleeding
  1. In the trials by Turpie et al. [13], Lassen et al. [15] and Colwell et al. [17] outcomes were adjudicated using similar criteria and by the same group as in the current study. s.c., Subcutaneous; o.d., once-daily; b.i.d., twice daily; Pre, preoperative; Post, postoperative; INR, International Normalization Ratio. *Unilateral venography was performed (bilateral venography was performed in all other studies).

Eriksson 2003 [14]Melagatran 3 mg s.c.Post and
Ximelagatran 24 mg p.o. b.i.d. Post
25.4% (205/806)1.7% (16/966)
Enoxaparin 40 mg s.c. o.d. Pre+Post19.4% (153/789)1.8% (17/975)
Colwell 2003 [17]Ximelagatran 24 mg p.o. b.i.d. Post7.9% (62/782)*0.8% (7/906)
Enoxaparin 30 mg s.c. b.i.d. Post4.6% (36/775)*0.9% (8/910)
Eriksson 2003 [16]Melagatran 2 mg s.c. Pre + 3 mg s.c. Post and
Ximelagatran 24 mg p.o. b.i.d. Post
12.9% (99/765)4.0% (37/915)
Enoxaparin 40 mg s.c. o.d. Pre+Post18.2% (146/801)1.1% (10/942)
Lassen 2002 [15]Fondaparinux 2.5 mg s.c. o.d. Post4.1% (37/908)3.7% (42/1140)
Enoxaparin 40 mg s.c. o.d. Pre+Post9.3% (85/919)2.6% (29/1133)
Turpie 2002 [13]Fondaparinux 2.5 mg s.c. o.d. Post6.1% (48/787)1.6% (18/1128)
Enoxaparin 30 mg s.c. b.i.d. Post8.3% (66/797)0.7% (8/1129)
Hull 2000 [12]Dalteparin 2500 IU s.c. Pre+Post, 5000 IU s.c. Post10.7% (36/337)8.9% (44/496)
Dalteparin 2500 IU Post, 5000 IU s.c. o.d. Post13.1% (44/336)6.6% (32/487)
Warfarin INR 2.0–3.0 Post24.0% (81/338)4.5% (22/489)

A number of methodological features of the present study are worthy of note. First, lack of a concurrent control group precludes firm conclusions about the efficacy of ART-123 compared with other antithrombotic agents. Second, an open-label, sequential, dose-ranging design was used rather than a double-blind parallel design in order to reduce the likelihood that patients would receive doses of ART-123 that were very ineffective or associated with a high risk of bleeding. Bias associated with the open-label design was reduced by having all reported events adjudicated by the blinded Independent Central Adjudication Committee. Lastly, only two dosing regimens of ART-123 were studied, which limits our ability to describe a dose–response. After it was determined that a dose of ART-123 of 0.45 mg kg−1 on day 1 was associated with an unacceptably high risk of bleeding, a third, lower, dosing regimen of 0.1 mg kg−1 on days 1 and 6 was considered. However, it was decided that this option would yield less clinically useful information than enrolling more patients to the 0.3 mg kg−1 on the days 1 and 6 regimen, since we were unlikely to demonstrate less bleeding with the 0.1 mg kg−1 dose than with the 0.3 mg kg−1 dose given that bleeding was so uncommon with the 0.3 mg kg−1 dose (major bleeding of 1.4%). Increasing the number of patients in the 0.3 mg kg−1 group provided a more precise estimate of the rates of bleeding and VTE with this regimen.

The present study provides the first clinical evidence that recombinant soluble thrombomodulin is effective at preventing postoperative VTE. Rates of VTE with the two dosing regimens (0.3 mg kg−1 on days 1 and 6; 0.45 mg kg−1 on day 1) appear to be as low as, or lower than, those that have been observed after hip replacement with other antithrombotic agents (Table 4). An acceptable rate of major bleeding was associated with the lower of the two doses. ART-123 therefore appears to be a promising long-acting antithrombotic agent that is suitable for further evaluation in clinical trials.

Conflict of interest

C.K. received salary support from Asahi Kasei America, Inc. for his role as Principal Investigator. This support was approved by, and paid through, the Department of Medicine, McMaster University, Hamilton. He has no other financial interests relating to the outcome of this study. P.C. has no related conflicts of interest. J.D. has no related conflicts of interest. R.R. is the Medical Director of Theradex, the CRO which managed the above trial on behalf of Asahi Kasei America, Inc. K.Y. is an employee of Asahi Kasei America, Inc. M.G. received consultation fees from Asahi Kasei America, Inc. during the course of the study.

Acknowledgements

C.K. and J.D. were supported by the Heart and Stroke Foundation of Canada. Sponsor: Asahi Kasei America, Inc., New York, NY, USA. C.K. is supported by the Canadian Institutes of Health Research.

Appendices

Appendix

Steering Committee

C. Kearon (Chair), P. Comp, J. Douketis, M. Gent, R. Royds, K. Yamada.

Central Adjudication Committee

M. Gent, J. Ginsberg, J. Hirsh, M. Levine, J.Weitz.

Clinical centers (patients enrolled)

D. Puskas, Thunderbay Regional Hospital, Thunder Bay, ON, Canada (42); T. McCoy, Charlotte Orthopedic, Charlotte, NC, USA (35); M. Jove, Atlanta Knee & Sports Medicine, Decatur, GA, USA (33); M. Mant, Walter MacKenzie Center, Edmonton, Alberta, Canada (32); P. Peters, WB Carell Clinic, Dallas, TX, USA (31); J. Muntz, The Methodist Hospital, Houston, TX, USA (24); P. Comp, Oklahoma Thrombosis Research Group, Oklahoma City, OK, USA (19); D. Stevens, Grand River Hospital, ON, Canada (17); C. Kearon, Henderson Hospital, Hamilton Health Sciences, Hamilton, ON, Canada; (14); D. Tapadiya, Shreenath Clinical Service, Fountain Valley, CA, USA (12); W. Navigato, Shreenath Clinical Service, Riverside, CA, USA (11); F. Burke, Bluegrass Musculoskeletal Research, Lexington, KY, USA (10); S. Siff, St Luke's Episcopal Hospital, Houston, TX, USA (10); L. Vickars, St Paul's Hospital, Vancouver, BC, Canada (8); M. Kimball, Saddleback Medical Research Services, San Diego, CA, USA (5); P. Hanson, Grossmonth Orthopedics, La Mesa, CA, USA (3); Y. K. Chan, Niagara Falls Medical Center, Niagara Falls, Ontario, Canada (2); W. Lanzer, Orthopedic International, Seattle, WA, USA (2); J. Douketis, St Joseph's Hospital, Hamilton, Ontario, Canada (1); J. Kassis, Hosp. Maisonneuve-Rosemont, Montreal, Quebec, Canada (1).

Study coordination

Theradex®, Princeton, NJ, USA: K. Cook, R. Royds, J. DeChamplain, D. Ionata, D. Kimball, C. Eberhardt.

Sponsor

Asahi Kasei America, Inc.: K. Yamada.

Ancillary