Statistical Review and Evaluation - RECOTHROM
STATISTICAL REVIEW AND EVALUATION
|Date of the review:||1/15/2008|
|Type of submission:||BLA_125248|
|Product / Application:||Recombinant Human Thrombin|
|Title of document:||Validation of Anti-rh Thrombin ELISA for Human Plasma Matrix using Commercial Process rhThrombin|
|From:||Paul B. Hshieh|
|CC:||DCC, Ghanshyam Gupta, Tie-Hua Ng; Steven Anderson; Chron, DCC. Paul Hshieh.|
- Hemostasis are frequently required in surgical practice when diffuse bleeding that is not amenable to control by conventional surgical techniques such as suture and cautery is encountered.
- ZymoGenetics has conducted a clinical program with recombinant human Thrombin (rhThrombin) used in conjunction with a gelatin sponge as a general adjunct to hemostasis in surgery.
- One pivotal Phase 3 study evaluating rhThrombin efficacies in surgical hemostasis have been conducted. This study evaluated the use of rhThrombin in surgical settings where adjuncts to hemostasis are often required: spinal surgery, hepatic resection, peripheral arterial bypass (PAB) surgery, and arteriovenous (AV) graft formation for hemodialysis access. This review is primarily based on the phase III study results.
- Recombinant human Thrombin or control was administered in conjunction with an absorbable gelatin sponge under blinded conditions.
To evaluate the relative efficacy of rhThrombin and bovine thrombin (Thrombin-JMI®), both applied topically with absorbable gelatin sponge, as adjuncts to hemostasis in surgery as measured by the incidence of hemostasis within 10 minutes.
- To evaluate the safety of rhThrombin and bovine thrombin as measured by the incidence and severity of adverse events (AEs) and the incidence and grade of clinical laboratory abnormalities
- To evaluate the immunogenicity of rhThrombin and bovine thrombin as measured by the incidence of anti-product antibodies
Study Design and Conduct:
- Protocol 499E01 was a Phase 3 multiple-site, randomized, double-blind, controlled trial designed to evaluate the comparative efficacy and safety of rhThrombin and bovine thrombin (Thrombin-JMI ® ; King Pharmaceuticals) in subjects undergoing surgery in 1 of 4 surgical indications: spinal surgery, hepatic resection, PAB surgery, and AV graft formation for hemodialysis access.
- Subjects were randomized in a 1:1 ratio to receive either rhThrombin (1000 U/mL) or bovine thrombin (1000 U/mL) in combination with a gelatin sponge.
- Enrollment was monitored to ensure adequate representation of each surgery type, with targets of approximately 30% each in spinal and hepatic surgery and 40% in PAB and AV graft combined.
- Dynamic allocation was used to attain an approximately equal number of subjects randomized to rhThrombin and bovine thrombin within each surgery type.
- During surgery, subjects were treated with blinded study drug in combination with an absorbable gelatin sponge at the evaluation site(s) and TTH was assessed for up to 10 minutes (at intervals of 30 seconds or less). If hemostasis was not achieved after 10 minutes, additional blinded study drug was allowed, followed by other hemostatic methods if needed.
- Fifty-two surgeons at 34 United States study sites enrolled subjects in the study.
- An independent data monitoring committee reviewed study data from 2 interim analyses. The study was planned to enroll between 400 and 600 subjects in a group-sequential design.
- The study was powered to exclude an incidence of hemostasis within 10 minutes of 15% less among subjects treated with rhThrombin compared to those treated with bovine thrombin.
- The time to hemostasis (TTH) is the primary endpoint for evaluating hemostatic effectiveness.
- Hemostasis is defined as occurring when no more blood from the bleeding site is observed seeping through or around the gelatin sponge.
- The start time for TTH is the time when the gelatin sponge(s) with study drug is applied to the bleeding surface.
- The end time for TTH computation is defined as the time blood ceases seeping through or around the gelatin sponge(s).
- Bleeding that re-occurs after hemostasis is achieved and following sponge removal will not be considered a failure to achieve hemostasis.
- Incidence and severity of adverse events
- Incidence and grade of clinical laboratory abnormalities
- Incidence of anti-product antibodies
- The study planned to enroll between 400 and 600 subjects unless it was stopped early for safety or futility concerns.
- In the final, there were 463 patients enrolled and randomly assigned.
- Among 463 patients, 52 patients the study drugs were not given, 3 were dead, five were lost to follow up, one withdrew the consent, and one was the other reason.
- The sponsor used total 401 subjects as their final ITT analysis.
Primary endpoint of hemostatic effect was to be based on the computation of 95% confidence interval of the difference (rhThrombin - bovine Thrombin) and exclusion of -15% to declare non-inferiority of the two treatments. The binary outcome of the primary endpoint was to be modeled using a linear model consisting of treatment group, surgery type, and the interaction between treatment group and surgery type. Confidence intervals were based on normal approximation using robust standard errors. Robust standard errors (Huber, 1967) were derived for the linear regression model to address potential heteroscedasticity. Additional details of the procedure can be obtained from the protocol.
However, this reviewer used the conservative procedure without adjustment to construct the confidence intervals with the assumption that if this procedure provides satisfactory results then the procedure used by the sponsor will also provide satisfactory results.
Demographic Characteristics: Tables 1A-1C provide distribution of patients into the two treatment groups by race, age, and sex for three different sets of data population.
Table 1A. Distribution of patients as randomized. (N=463)
RACE(Race) TRTGRP(Treatment group) Frequency , Row Pct ,bThrombi,rhThromb, Total ,n ,in , fffffffffffffffff^ffffffff^ffffffff^ WHITE , 155 , 160 , 315 , 49.21 , 50.79 , fffffffffffffffff^ffffffff^ffffffff^ BLACK OR AFRICAN , 44 , 46 , 90 AMERICAN , 48.89 , 51.11 , fffffffffffffffff^ffffffff^ffffffff^ HISPANIC , 17 , 18 , 35 , 48.57 , 51.43 , fffffffffffffffff^ffffffff^ffffffff^ ASIAN , 12 , 5 , 17 , 70.59 , 29.41 , fffffffffffffffff^ffffffff^ffffffff^ OTHER , 3 , 1 , 4 , 75.00 , 25.00 , fffffffffffffffff^ffffffff^ffffffff^ NATIVE HAWAIIAN/ , 1 , 1 , 2 PACIFIC ISLANDER , 50.00 , 50.00 , fffffffffffffffff^ffffffff^ffffffff^ Total 232 231 463 Q_AGECAT(Age Category) TRTGRP(Treatment group) Frequency , Row Pct ,bThrombi,rhThromb, Total ,n ,in , ffffffffffff^ffffffff^ffffffff^ < 65 years , 147 , 149 , 296 , 49.66 , 50.34 , ffffffffffff^ffffffff^ffffffff^ >= 65 years , 85 , 82 , 167 , 50.90 , 49.10 , ffffffffffff^ffffffff^ffffffff^ Total 232 231 463