SECTION 4

FDA SUMMARIES

 

1.0            Lead Reviewer & Biocompatibility Summary

 

Date:               October 23, 2003

 

From:              Jennifer Goode, B.S., Biomedical Engineer

                        CDRH/ODE/DCD

 

Subject:           TAXUS™ Express2™ (SR) Paclitaxel-Eluting Monorail and Over-the-Wire Coronary Stent Systems

 

1.1.            PMA Chronology

June 19, 2003 – P030025 Received by CDRH/ODE

August 1, 2003 – P030025 Filed

September 15, 2003 – Major Deficiency Letter sent to sponsor

September 29, 2003 – Response to Major Deficiencies Received by CDRH/ODE

November 20, 2003 – P030025 Scheduled for Review by Circulatory Systems Devices Panel

 

1.2.            Summary

This PMA has been submitted to seek marketing approval for the TAXUS™ Express2™ Paclitaxel-Eluting (Monorail and Over-the-Wire) Coronary Stent Systems (CSS).  The TAXUS™ Express2™ Paclitaxel-Eluting CSS is a combination product intended for “improving luminal diameter and reducing restenosis for the treatment of de novo lesions < 28 mm in length in native coronary arteries > 2.5 to < 3.75 mm in diameter.” 

 

Note:  The stent that is used for the substrate, the Express™ Coronary Stent was approved on September 11, 2002 (P020009), but does not include an indication for improving restenosis.

 

The TAXUS™ Express2™ Paclitaxel-Eluting CSS is a combination product for which the primary jurisdictional responsibility has been delegated to CDRH with consultation to CDER. 

 

The TAXUS™ Express2™ Paclitaxel-Eluting CSS consists of a balloon-expandable, 316L stainless steel coronary stent, coated with a drug/polymer coating, mounted on a delivery catheter.  Both slow release and moderate release formulations were used for testing conducted throughout the development of the TAXUS™ stent.  The proposed commercial product has a 1µg/mm2 slow release (SR) formulation.  No other SR formulation dose/unit area quantities were tested.  MR has been used to designate all stents coated with the moderate release formulation, and as appropriate, the relevant dosage has been identified (i.e., 1, 2, or 4 μg/mm2). Additional specifics about the drug and polymer coating configuration are provided in subsequent sections. 

 

Device description, bench testing, animal testing, and pharmacokinetic information provided to support the clinical use of the TAXUS™ Express2™ Paclitaxel-Eluting CSS are summarized below and in the Animal Studies Summary provided by Stephen Hilbert, M.D., Ph.D., CDRH/ODE/DCD, with incorporation of comments from Belay Tesfamariam, Ph.D., CDER/OND/DCRDP, and Steven Woods, Ph.D., CDRH/OST/DLS. 

 

The Chemistry, Manufacturing and Controls (CMC) of the drug substance and polymer coating were reviewed by Nallaperumal Chidambaram, Ph.D., CDER/OPS/DNDCI.  Additional comments were provided by W. Leroy Schroeder, Ph.D., CDRH/OST/DMMS, and Joyce Whang, Ph.D., CDRH/ODE/DRARD, with regards to the polymer chemistry.  The in vitro elution method and specifications and human PK study were reviewed by Angelica Dorantes, Ph.D., CDER/OPS/DPEI.  The engineering was reviewed by Carolyn Vaughan, B.S., CDRH/ODE/DCD.  As appropriate, summarized information from these reviews have been included within this summary. 

 

The sponsor has conducted a multi-center, double-blind, randomized, clinical investigation, referred to as the TAXUS IV Trial.  This trial evaluated the safety and effectiveness of the TAXUS™ Express™ Paclitaxel-Eluting CSS in reducing target vessel revascularization (TVR) in de novo coronary artery lesions ranging in diameter from 2.5mm to 3.75mm (inclusive) and lengths from 10mm to 28mm versus the uncoated Express™ CSS, both mounted on the Express™ Monorail stent delivery system.  A total of 1328 subjects were enrolled in the study at 76 US sites.  Twelve (12) patients were “deregistered” from the study (i.e., randomized, but with no attempt to treat), and two (2) patients were excluded from the efficacy analysis because they were not randomized using the study protocol randomization scheme.  Of the remaining 1314 patients, 662 were randomized to the treatment arm and 652 were randomized to the control.  All patients in the treatment group received the 1 µg/mm2 slow release formulation (SR).  The primary endpoint for this study was 9-month ischemia-driven target vessel revascularization (TVR).  Follow-up angiography was completed on a subset of 559 patients (76% of the pre-selected subgroup of 732) at 9-months with an IVUS subset of 268++ patients enrolled, with 2-D and 3-D measurements available on smaller subsets (66% and 53%, respectively) due to analysis constraints.

 

The initial clinical feasibility study, the TAXUS I Study, was initiated at 3 sites in Germany.  The TAXUS I Study was a multicenter, prospective, double-blind, two-arm randomized, controlled trial designed to assess the 30-day safety and performance of the TAXUS™ NIRx™ Paclitaxel-Eluting CSS compared to the bare metal NIRx™ stent (control), both mounted on the Express™ Monorail stent delivery system, in patients with de novo or restenotic coronary artery lesions ranging in diameter from 3.0mm to 3.5mm (inclusive) and lengths Ł 12mm.  A total of 61 patients were included in the study (31 in the treatment group and 30 in control).  All patients in the treatment group received the 1 µg/mm2 slow release (SR) formulation.  The primary endpoint was 30 day safety, defined as freedom from MACE (i.e., cardiac death, Q-wave and non-Q-wave myocardial infarction, and target vessel revascularization).  Patients are being followed clinically at 1, 3, 6, 12, and 24 months, with yearly telephone follow-up (planned out to 5 years post-procedure).  Angiography and IVUS were performed for all patients at the 6 month follow-up visit. Repeat angiography and IVUS will be performed at the 2-year follow-up for patients who are in the TAXUS™ NIRx™ (SR) treatment group.

 

Prior to the initiation of the U.S. pivotal trial, the sponsor conducted the TAXUS II trial at 38 sites in 15 countries outside of the U.S.  The TAXUS II Study was a multicenter, prospective, double-blind, randomized, controlled trial involving two cohorts designed to assess the safety and effectiveness of the TAXUS™ NIRx™ Paclitaxel-Eluting CSS SR (1µg/mm2) compared to the bare metal NIRx™ stent, and the TAXUS™ NIRx™ Paclitaxel-Eluting CSS MR (1µg/mm2) compared to the bare metal NIRx™ stent.  Both the TAXUS™ SR and MR CSS were mounted on the Express™ Monorail stent delivery system and were implanted in patients with documented angina pectoris and a single de novo coronary artery lesion ranging in diameter from 3.0mm to 3.5mm (inclusive) and lengths Ł 12mm.  A total of 536 patients were included in the study [131 in the SR cohort (Cohort 1), 136 in the matched bare metal control group for Cohort 1, 135 in the MR cohort (Cohort 2), and 134 in the matched bare metal control stent group for Cohort 2].  The primary endpoint was percent in-stent net volume obstruction, as measured by IVUS.  Patients were followed clinically at 1, and 6 months, with yearly telephone follow-up (planned out to 5 years post-procedure) to assess MACE clinical parameters (i.e., cardiac death, Q-wave and non-Q-wave myocardial infarction, and repeat revascularization procedures).  Angiography and IVUS were performed for all patients at the 6 month follow-up visit.

 

These data provide the basis for the analyses presented in this panel pack.  The clinical investigation information is summarized in the Clinical Summary provided by John Stuhlmuller, M.D., CDRH/ODE/DCD, and the Statistical Summary provided by Heng Li, Ph.D., CDRH/OSB/DBS. 

 

To date, there remain some unresolved issues pertaining to the non-clinical testing and final manufacturing specifications.  These issues have been communicated to the sponsor in a Major Deficiency Letter, sent on September 15, 2003.  The sponsor has responded to the majority of issues in amendments dated September 26, October 1, October 2, and October 15, 2003.  The Agency is working interactively with the sponsor to ensure that the outstanding issues are resolved in a timely manner. 

 

1.3.            Device Description

The TAXUS™ Express2™ Paclitaxel-Eluting CSS is a combination product, consisting of a permanently-implanted stent with a conformal drug/polymer coating and the TAXUS™ Express2™ Delivery Catheter System(s).

1.3.1.      TAXUS™ Stent

The TAXUS™ Express™ Paclitaxel-Eluting Stent is a combination product, consisting of the Express™ balloon-expandable 316L stainless steel coronary stent with the addition of a conformal drug/polymer coating adhered to the entire surface (i.e., luminal and abluminal) of the stent.  The coating formulation is designated as a slow release formulation with a 1μg/mm2 dose per unit area. 

 

The Express™ stent is a 316L stainless steel, laser-cut, slotted-tube, pre-mounted balloon expandable stent.  The Express™ stent is currently approved for improving coronary luminal diameter in patients with symptomatic ischemic disease associated with stenotic lesions in native coronary arteries (Ł18mm in length) with reference diameters from 3.0mm to 5.0mm (P020009). 

 

Note: The Express™ is an uncoated stent, with neither a polymer nor drug coating.  Also note that the proposed indication for the TAXUS™ Express2™ Stent represents a change from the bare Express2TM Coronary Stent System indication because it includes the phrase “reducing restenosis.” 

 

For the TAXUS™ Express™ stent, the sponsor has currently requested approval for the device sizes designated in Table 1. 

 

Table 1:  TAXUS™ Express™ (SR) Stents proposed for approval

 

Stent Diameter

Stent Length

8mm

12mm

16mm

20mm

24mm

28mm

32mm

2.50mm

D: 50μg

P: 519μg

D: 79μg

P: 819μg

D: 108μg

P: 1119μg

D: 137μg

P: 1419μg

D: 151μg

P: 1570μg

 

 

2.75mm

D: 50μg

P: 519μg

D: 79μg

P: 819μg

D: 108μg

P: 1119μg

D: 137μg

P: 1419μg

D: 151μg

P: 1570μg

D: 180μg

P: 1870μg

D: 209μg

P: 2170μg

3.00mm

D: 50μg

P: 519μg

D: 79μg

P: 819μg

D: 108μg

P: 1119μg

D: 137μg

P: 1419μg

D: 151μg

P: 1570μg

D: 180μg

P: 1870μg

D: 209μg

P: 2170μg

3.50mm

D: 50μg

P: 519μg

D: 79μg

P: 819μg

D: 108μg

P: 1119μg

D: 137μg

P: 1419μg

D: 151μg

P: 1570μg

D: 180μg

P: 1870μg

D: 209μg

P: 2170μg

 

The 2.5-3.5mm diameter stent is one stent design.  The same stent is crimped on various size delivery catheter balloons, which are sized from 2.5 to 3.5mm.  Because the identical stent component is used for the entire 2.5-3.5mm range, the total drug and polymer per stent are a function of the stent length, irrespective of stent diameter.

 

1.3.2.      TAXUS™ Express2™ Delivery Systems

The TAXUS™ Express™ (SR) Paclitaxel-Eluting stent will be available pre-mounted on either the Express2 Over-the-Wire (OTW) delivery system or the Express2™ Monorail delivery system.  Both delivery systems are sterile, single use catheters on which the TAXUS™ Paclitaxel-Eluting stent is crimped.  The stent should not be removed from the delivery system as removal may damage the stent and/or lead to stent embolization.  The stent is centered on a DynaLeap™ balloon, nominally 0.3mm longer than the stent, between two radiopaque marker bands to aid in positioning the stent fluoroscopically and also aid in accurately positioning the delivery system.  The delivery systems provide the mechanism of action for deployment of the stent to the treatment site (i.e., coronary artery). 

 

1.3.2.1.                        Express2™ Over-the-Wire (OTW)

The Express2™ OTW delivery system has been previously approved for deployment of the uncoated Express™ stent in P020009 (approved September 11, 2002).  The Express2™ OTW version has a working length of 135cm.  It is compatible with guide catheters having an internal diameter of ł 0.066” (ł 6F). 

 

The Express2™ catheter has a y-connector, where the side arm allows for access to the balloon inflation/deflation lumen.  The straight arm is continuous with the shaft inner lumen, which is designed for guidewires Ł 0.014”.  The proximal shaft of the catheter has an outside diameter of 3.2F (0.042”), which tapers to an outside diameter of 2.7F (0.035”) at the distal shaft. 

 

 

1.3.2.2.                        Express2™ Monorail

The Express2™ Monorail delivery system has been previously approved for deployment of the uncoated Express™ stent in P020009 (approved September 11, 2002).  The Express2™ Monorail version has a working length of 140cm.  It is compatible with guide catheters having an internal diameter of ł 0.058” (ł 4.5F). 

 

The Express2™ Monorail catheter has a single access port to the inflation lumen and a guidewire exit port located 25cm proximal to the distal tip, which is designed for guidewires Ł 0.014”.  The proximal shaft of the catheter has an outside diameter of 1.8F (for all sizes, except the 3.5mm in 24, 28, and 32mm lengths, which is 2.0F) in comparison to the distal shaft, which has an outside diameter of 2.7F. 

 

Note:  The proposed Express2 Delivery Catheters were not used in the TAXUS IV (SR) trial.  The TAXUS IV (SR) trial used the TAXUS™ Express™ Delivery Catheters.  The only differences between the previous (Express™) design, and the proposed (Express2™) version of the delivery catheters are that: (1) the Express2™ catheter changed one of the shaft materials to another material that was used in other components of the original Express™ delivery catheter shaft; and (2) the Express2™ manifold was molded slightly differently to allow better retention in the packaging.  Both the Express™ and Express2™ OTW and Monorail delivery catheters were approved for use with the bare Express™ stent (P020009, approved September 11, 2002).  This PMA is requesting approval of only the Express2™ OTW and Monorail delivery systems for use in delivering the TAXUS™ (SR) stent.

 

1.3.3.      Drug/Polymer Coating

The TAXUS™ Paclitaxel-Eluting Stent is a combination product, consisting of the Express™ balloon-expandable 316L stainless steel coronary stent (approved in P020009) coated with a drug/polymer coating.  According to the sponsor, the coating is conformal and approximately 18 to 20 mm thick. 

 

1.3.3.1.                        Polymer Coating

The coating process consists of the polymer, --------------------------------------------------, mixed with solvent and the drug paclitaxel. The combined solution is then sprayed onto the stents.  There is no primer or topcoat layer.

 

Note:  The polymer coating formulation and process is proprietary, trade secret information.  For the remainder of this memo, and for all discussions at the open public meeting, the polymer coating will be referred to by its commercial name:  Translute™

 

Release of paclitaxel from the polymer follows a biphasic profile.  The early burst phase represents dissolution and mediated burst release of paclitaxel particles from the polymer surface.  The slower, low-level, sustained release occurs as complete dissolution of the surface drug particles enables biological fluid to access sub-surface particles, which then dissolve, and diffuse out of the polymer through the tortuous pathways in the polymer matrix.

 

1.3.3.2.                        Drug Substance

Text Box: Figure 1.  Chemical structure of paclitaxel.
 
The drug substance, Paclitaxel, is manufactured by Indena, S.p.A., Milan Italy.  Paclitaxel is a white to off-white powder, originally isolated from the bark of the Pacific Yew tree (Taxus brevifolia), but currently isolated from a spectrum of Taxus species and hybrids cultivated specifically to provide a biomass for the manufacture of the drug.  Paclitaxel has the molecular formula of C47H51NO14 (Figure 1) and a molecular weight of 853.92. 

 

The sponsor has provided a letter from Indena S.p.A. authorizing the Agency the right to access the Drug Master File (DMF) for paclitaxel in support of this PMA application.  Indena manufactures a generic form of the drug Taxol®, a Bristol Myers Squibb drug that was approved by the FDA for injection (NDA 20-262; approved December 29, 1992) for the following:

·         As first- line and subsequent therapy for the treatment of advanced carcinoma of the ovary.

·         In combination with cisplatin, as first-line therapy, for the adjuvant treatment of node- positive breast cancer administered sequentially to standard doxorubicin containing combination chemotherapy. [In the clinical trial, there was an overall favorable effect on disease free and overall survival in the total population of patients with receptor-positive and receptor-negative tumors, but the benefit has been specifically demonstrated by available data (median follow up 30 months) only in the patients with estrogen and progesterone receptor-negative tumors.]

·         For the treatment of breast cancer after failure of combination chemotherapy for metastatic disease or relapse within 6 months of adjuvant chemotherapy. [Prior therapy should have included an anthracycline unless clinically contraindicated.]

·         In combination with cisplatin, for the first- line treatment of non-small cell lung cancer in patients who are not candidates for potentially curative surgery and/ or radiation therapy.

·         As second line treatment for AIDS-related Kaposi’s sarcoma.

 

Paclitaxel is not indicated for the treatment of restenosis or for use in coronary arteries. 

 

In 2002, six generic products containing paclitaxel (PACLITAXEL Injection) were approved by the FDA.  However, Indena does not make a generic paclitaxel product for U.S. market.  The drug substance information provided in Indena’s DMF was available for review by FDA, and both Indena (the drug substance manufacturer) and Boston Scientific are interactively working with FDA to resolve any outstanding issues.

 

1.4.            Bench Testing

1.4.1.      Stent/Delivery System Integrity Testing

The following bench testing, in accordance with the FDA “Guidance for the Submission of Research and Marketing Applications for Interventional Cardiology Devices: Intravascular Stents”, May 1995, was conducted on the TAXUS™ Express™ (SR) Paclitaxel-Eluting Coronary Stent System (Monorail and Over-the-Wire Delivery Catheters): 

·         Specification Conformance Testing

o        Material analysis (stent: raw material, elemental analysis, surface contamination)

o        Tensile testing

o        Elongation testing

o        Corrosion

·         Stent Integrity Testing

o        Foreshortening

o        Stent material to artery ratio

o        Stent coating integrity

o        Fatigue analysis (FEA)

o        Fatigue testing (up to 400 million cycles)

o        Stent recoil

o        Radial hoop strength

o         Stent expansion uniformity

o        Dimensional verification

o        Stent flexibility/conformability

·         Stent/Delivery System Integrity Testing

o        Stent deployment

o        Maximum burst pressure

o        Compliance

o        Bond strength

o        Balloon/catheter diameter and profile

o        Repeat balloon inflations (fatigue)

o        Stent retention

o        Crossing profile

 

The sponsor also provided appropriate rationales for use of testing from the uncoated Express™  and Express2™ CSS (P020009, Approved September 11, 2002) to characterize the following properties of the TAXUS™ Express™ CSS (i.e., coating of stent will not modify these properties of the stent or delivery catheters):

·         Stent Integrity Testing

o        Radiopacity

o        Magnetic resonance imaging

·         Stent/Delivery System Integrity Testing

o        Balloon deflatability

o        Balloon inflation/deflation time

 

The nominal inflation pressure for the TAXUS™ (SR) product is 9 atm, and the rated burst pressure (RBP) for all sizes is 18 atm. 

 

Fatigue testing showed no evidence of stent cracks or breakages, coating fatigue, or corrosion after 400 million cycles (equivalent to 10 years). 

 

Clarifications have been requested of the sponsor regarding the in vitro bench testing results.  The FDA is working interactively with the sponsor to ensure that these issues are resolved in a timely fashion. 

 

1.4.2.      Coating Integrity

The Chemistry, Manufacturing and Controls (CMC) of the drug substance and polymer coating were reviewed by Chidambaram Nallaperumal, Ph.D., chemist, CDER/OPS/DNDCI.  Additional comments were provided by Leroy Schroeder, Ph.D., polymer and materials scientist, CDRH/OST/DMMS, and Joyce Whang, Ph.D., chemist, CDRH/ODE/DRARD, with regards to the polymer chemistry. 

 

To ensure coating integrity of the finished TAXUS™ product, the sponsor conducted the following tests:

·         Drug content

·         Degradation impurities

·         Residual solvents

·         Particulates/Coating integrity

·         Drug in vitro elution

 

According to the sponsor, all samples passed the product specifications for each test. 

 

The in vitro elution method (e.g., kinetic drug release) and specifications were reviewed by Angelica Dorantes, Ph.D., pharmacologist, CDER/OPS/DPEI.  According to her review, the currently proposed in vitro elution method is appropriate from a quality control perspective.  The FDA is working interactively with the sponsor to finalize release specification for the manufactured product. 

1.5.            Biocompatibility Testing

Biocompatibility tests in accordance with ISO 10993-1, “Biological Evaluation of Medical Devices,” were conducted on polymer-coated stainless steel coupons, or cast polymer films (where the drug substance, Paclitaxel, was not included in the polymer coating).  The sponsor has been asked to provide clarification on how the test articles compare to the final sterilized device with respect to formulation, manufacturing and sterilization. 

 

The following tests were conducted on ethylene oxide sterilized polymer coated stainless steel coupons:  hemolysis (direct contact method), cytotoxicity (MEM elution method), genotoxicity (Bacterial Reverse Mutation - Ames Assay,), sensitization (Guinea Pig Maximization Method), pyrogenicity (Materials-Mediated), acute systemic toxicity, and intracutaneous reactivity.  The results of this testing demonstrated that the tested systems were non-toxic, non-hemolytic, and non-irritating.

 

The following tests were conducted on polymer cast films:  genotoxicity (Chromosome Aberration Assay, Mouse Lymphoma Mutagenesis Assay), chronic toxicity in mice, and demonstrated that the tested systems were non-toxic.

 

Since the sponsor did not conduct ISO-10993 testing on the finished product with drug substance, sub-chronic toxicity, thrombosis, implantation, and biodegradation of the final product, containing all components and processing, were evaluated via porcine implant studies, as summarized in Section 2. 

 

The paclitaxel carcinogenicity, genotoxicity, and reproductive toxicity information were reviewed by Belay Tesfamariam, Ph.D., pharmacologist, CDER/OND/DCRDP. Genotoxicity testing was performed across a range of paclitaxel concentrations (0.06 to 0.6 ng/mg) representing the levels found in animal tissues following stenting with the TAXUS™ CSS.  The results of this testing did not raise any questions.  However, because a majority of the loaded paclitaxel is intended to remain in the stent over the implant life, Dr. Tesfamariam has recommended that a summary of the genotoxicity and reproductive toxicity information from the drug label be included in the label for the TAXUS™ Express2™ Paclitaxel-Eluting CSS.

 

The rationale for omission of carcinogenicity testing was reviewed by Raju Kammula, D.V.M., Ph.D., toxicologist, CDRH/ODE/DRARD.  The sponsor provided a scientifically valid rationale for why carcinogenicity testing of the drug substance, paclitaxel, is unnecessary.  However, the sponsor has been asked to provide additional information regarding the carcinogenic potential effects of the entire stent (to include the polymer carrier, manufacturing additives, etc.).  The FDA is working interactively with the sponsor to ensure that these issues are resolved in a timely fashion. 

 

The sponsor provided an appropriate rationale for use of biocompatibility testing on the Express™ Delivery Catheters (P020009, approved September 11, 2002) to support the biocompatibility of the TAXUS™ Express2™ Monorail and Over-the-Wire delivery catheters.

1.6.            Pharmacology/Toxicity Testing

The sponsor submitted results from two in vitro preclinical pharmacology studies on the effect of paclitaxel on human smooth muscle cells, as summarized in the panel pack (Volume 1, Non-clinical and Clinical Testing Summary, p.25).  The panel pack also notes that more recent studies have been conducted by Boston Scientific, but these additional studies have not been provided for FDA review.  The proposed labeling (Section 6.4, Carcinogenicity, Genotoxicity, and Reproductive Toxicity, p.12) includes the following statement:

 

The effects of paclitaxel on human coronary smooth muscle cells… were evaluated over a broad range of paclitaxel concentrations.  The data showed no evidence of apoptosis and provided evidence for a cytostatic mechanism in the inhibition of proliferation of human coronary smooth muscle cells.

 

Sufficient information to support the claim of a cytostatic versus a cytotoxic effect has not been submitted to FDA for review.

 

The pre-clinical (animal) and clinical in vivo release studies were reviewed by Angelica Dorantes, Ph.D., pharmacologist, CDER/OPS/DPEI.  Three animal studies were conducted to investigate in vivo release rates of the product.  The paclitaxel levels adjacent to the stent were low in the ng to µg range for all the formulations tested (i.e., 1µg/mm2 SR, 1µg/mm2 MR, 1µg/mm2 fast release (FR), and 4µg/mm2 FR), with practically no paclitaxel detected in the proximal, distal, or more remote tissues (including the carotid artery and liver).  There were also no detectable systemic blood levels of paclitaxel.  From the tissue studies, it appears that there is biphasic drug release profile, with an initial burst followed by a constant (very slow and low release of the drug with time).

 

The sponsor demonstrated in 3 clinical studies, that the systemic levels of paclitaxel (at all times after implantation of the TAXUS™ SR and MR stents) are below the detection limit of 10ng/ml.  Therefore, the pharmacokinetic profile of paclitaxel in humans could not be assessed.

 

The sponsor was unable to use the animal and clinical in vivo pK studies to establish a relationship between the in vitro elution rate characteristics and the in vivo performance of the product.  However, a commercial in vitro elution assay has been developed by the sponsor that has confirmed that the commercially manufactured product has the same rate of drug release as the product that was implanted in the clinical studies.

 

The pre-clinical in vivo release studies in animals were also reviewed by Belay Tesfamariam, Ph.D., pharmacologist, CDER/OND/DCRDP. Relevant toxicity information from Dr. Tesfamariam’s review has been included in the Animal Studies Summary provided by Steve Hilbert, M.D., Ph.D., CDRH/ODE/DCD. 

 

 

1.7.            Sterility Testing and Package Integrity

The product is loaded into a stent protector and tubing, and then sealed in a Tyvek pouch, and boxed (same package used for Express™ CSS, P020009, approved September 11, 2002).

 

The package is sterilized with ethylene oxide (EtO) to achieve an SAL of 10-6.  The packaging and all sterilization procedures have been appropriately validated (and meet specifications for EtO residue limits). 

1.8.            Shelf Life/Stability

The sponsor has requested a 1 year expiration date for the TAXUS™ Express2™ Paclitaxel-Eluting (SR) CSS.  The FDA is working actively with the sponsor to finalize the stability protocols and specifications prior to assigning an expiration date for the product.  The sponsor indicated that appropriate numbers of commercial lots have been placed on stability according to ICH guidelines, and will be tested once protocols and specifications are finalized with the Agency.

 

1.9.            Device Failures and Malfunctions - Clinical Use

Sections 10.4.4.2 and 10.5 of the TAXUS IV Study Report (Panel Pack, Volume 2) identify the device failure and malfunction events that occurred during the TAXUS IV study.   This information indicates that the incidence of delivery system failures (26 reported) are within the range of what would be expected for study of this size where the patients are undergoing a procedure such as coronary catheterization with stent placement. 

 

With regard to device failures, there were 9 patients with stent delivery failure (3 TAXUS, 6 control), 1 broken stent tip (TAXUS), and 1 bent stent tip (Control).  There were 20 incidents (11 TAXUS, 8 control) in which the assigned device was never delivered, and 1 patient had treatment with a cutting balloon and placement of a commercially available stent after failed predilatation (1 TAXUS).  The TAXUS IV study was conducted using the TAXUS™ Express™ Paclitaxel-Eluting stent or the uncoated Express ™ stent pre-mounted on the Express™ Monorail delivery system.

 

In the TAXUS I study, all patients received the assigned device, and there were no reports of technical failures.

 

In the TAXUS II 6 month clinical reports, there were 11 incidents in which the assigned device was never delivered (3 SR, 3 MR, 5 Control), which were resolved with placement of another non-study stent (2 SR, 1 MR, 1 Control), or PTCA (1 SR, 2 MR, 3 Control).  There were 5 reported cases of technical failures (3 SR, x MR, 1 Control).  Details were not provided.  [See CD-ROM: TAXUS II 6 Month Clinical Report - Cohort I (SR), p. 9-10; and TAXUS II 6 Month Clinical Report - Cohort II (MR),  p.4-5]


1.10.        Conclusions

·         To date, there remain some unresolved issues pertaining to the non-clinical testing and final manufacturing specifications.  These issues have been communicated to the sponsor in a Major Deficiency Letter, sent on September 15, 2003.  The sponsor has responded to the majority of issues in amendments dated September 26, October 1, October 2, and October 15, 2003.  The Agency is working interactively with the sponsor to ensure that the outstanding issues are resolved in a timely manner. 

·         No data have been presented that indicate a clear safety concern in the clinical setting regarding mechanical device failure or malfunction. 


2.0            Animal Studies Review  

 

Date:          October 21, 2003

 

To:             Jennifer Goode

                  DCD/ODE/CDRH

 

From:          Stephen L Hilbert, MD, PhD

                  Experimental Pathologist

                  DCD/ODE/CDRH

 

Subject:       Taxus™ Paclitaxel Eluting Coronary Stent: Preclinical In Vivo Explant Pathology

 

2.1.            Introduction

Preclinical in vivo explant pathology studies of drug-eluting stents are recommended to address a number of safety issues, typically before allowing human clinical trials to begin.  One issue to be evaluated is the chronic effect of the polymer coated metal stent and the drug eluting stent on vessel wall healing following one and six months of implantation as compared to a bare metal stent.  Another critical issue is the evaluation of possible toxicity of the highest drug dosage intended for clinical use and the determination of the safety margin (local, regional, systemic) at this dosage (by testing stents with greater than the clinical dosage, 3- 10X is recommended).  Animal studies also provide insight into possible adverse effects of the eluted drug on myocardial tissue “downstream” from the stent and whether the drug effect and vessel healing vary along the length of the stent (e.g., proximal, mid- and distal regions).  Finally, if the clinical use of the stent will include overlapping stents or treatment of in-stent restenosis, the effect of these conditions on vessel healing should be evaluated in animals following a minimum one month and six months of implantation.

 

Although safety and effectiveness can only be demonstrated through human clinical trials, preclinical in vivo studies can provide important information concerning the handling characteristics of the device (delivery and deployment), device-related pathology (safety) as well as an initial assessment of short-term effectiveness.

 

2.2.            Animal Model Justification

The preclinical in vivo evaluation of the TAXUS drug eluting stent in farm swine is justified based on vascular and cardiac anatomy and vessel diameters suitable for the delivery and deployment of the clinical version of the device and comparison of findings to historical data.  In the absence of a suitable animal model which mimics human coronary artery disease, implantation in normal farm swine is appropriate for the evaluation of device handling characteristics (delivery and deployment), device-tissue responses (vessel wall injury and healing, inflammatory response) and the assessment of device-related pathology (safety).  The assessment of efficacy and long-term device safety can only be determined based on experience gained by use in general patient populations with advanced coronary artery disease.

 

 

2.3.            Methods

Test and control stents were delivered and deployed using a randomized assignment method to three coronary artery sites (LAD, LCX, RCA).   Stents were expanded to a targeted artery-to-balloon ratio of 1:1-1:1.2.  The stents were explanted following 28, 90, 180 and 360 days depending on the study design.  The graded histopathology endpoints consisted of: mural thrombus, neointima, luminal narrowing, apparent endothelialization, inflammatory response at struts, IEL disruption, EEL disruption, medial alterations, para-strut amorphous material (PAM, also referred to as drug effect, thrombus, strut fibrin).  The following morphometric measurements were made: luminal area, intimal thickness, IEL area, EEL area, neointimal area, medial area, lumen equivalent diameter, stent profile, % stenosis (lumen/IEL based) and % stenosis (lumen/stent profile based).

 

Statistical analyses (SAS Version 8.02) were undertaken comparing control (bare metal stent, polymer-coated metal stent) to the TAXUS (1ug/ mm2) drug eluting stent.

 

 

2.4.            Study A242-GX-00 (Bare Metal Stent and Translute-coated Metal Stent)

 

Device Description:    NIR Conformer (referred to as bare metal stent or BMS); Translute-coated NIR Conformer (referred to as Translute-coated stent); stent dimensions:  3.0 x 16 mm or 3.5 x 16 mm

 

Translute stents (without paclitaxel) were implanted in 14 swine (2 stents per animal) and a bare metal stent implanted in 13 swine (2 stents per animal) for duration of 28, 90 and 180 days. There were two deaths (embolized stent; dissected vessel) reported in the bare metal stent group (90 day cohort) and one death (bleeding at access site) in the Translute-coated stent group (180 day cohort).

 

Histopathology and Morphometric Findings:

 

28 Days.

Bare metal stent.  Endothelialization was essentially complete without evidence of mural thrombus formation.  The lumen was widely patent and lined by a thin neointimal consisting of smooth muscle cells and fibroblasts embedded in a collagen-rich extracellular matrix.  IEL was rarely disrupted by the stent struts and the ELL intact.  The media was well defined.  The inflammatory response consisted of macrophages and an occasional multinucleated foreign body giant cell.  Eosinophils were noted in the inflammatory response in two explants (four stent segments).

 

Translute-coated stent.  Similar histologic findings as described for the bare metal stent with the following exceptions: one small mural thrombus; no eosinophils were present in the inflammatory response.

 

90 Days.

Bare metal stent and Translute-coated stent.  Histologic findings comparable to the 28 day results; however, the apparent endothelialization is now complete.  Eosinophils were present in the inflammatory response in two Translute-coated stent explants.

 

180 Days.

Bare metal stent.  Histologic findings comparable to the 28 and 90 day results; however, more fibrous tissue ingrowth was present in the media. Eosinophils were present in the inflammatory response in two explants.

 

Translute-coated stent. Comparable to bare metal stent findings with the exception of a slight luminal narrowing ( two stents), greater disruption of the IEL and considerable disruption of the media with fibrous tissue ingrowth. Eosinophils were noted in two explants.

 


Comparison of Formulation A and B

 

Formulation A

                                           

 

Intima Area

Strut-based Intima Thickness

Lumen Area

Stenosis        (%)

Lumen Diameter          Mean

28 Days

 

 

 

 

 

BMS (8)

1.34

229

5.3

19.1

2.58

Translute (10)

1.88

351

5.6

25.6

2.64

 

 

 

 

 

 

90 Days

1.47

211

5.7

20.9

2.68

BMS (5)

2.28

416

4.63

32.1

2.4

Translute(6)

 

 

 

 

 

 

 

 

 

 

 

180 Days

 

 

 

 

 

BMS (8)

1.56

224

5.39

22.7

2.6

Translute (8)

2.00

312

5.37

27.3

2.6

t-test, no significant difference between means

neointimal area NR

 

 

Formulation B

                                                          

 

Intima Area

Strut-based Intima Thickness

Stenosis        (%)

Lumen Diameter          Mean

28 Days

 

 

 

 

BMS (8)

5.3

229

1.34

19

Translute (7)

5.0

403

1.8

26

 

 

 

 

 

90 Days

 

 

 

 

BMS (5)

5.7

211

1.47

21

Translute(8)

5.55

248

1.30

19

 

 

 

 

 

180 Days

 

 

 

 

BMS (8)

5.39

224

1.56

23

Translute (6)

5.25

257

1.67

25

Intimal area NR, no statistical comparison reported

 

 

Summary of Study Pathologist Observations and Comments

 

Both bare metal stents and Translute-coated stents demonstrated a marked chronic inflammatory reaction which included eosinophils surrounding the struts following 28 days, 90 days and 180 days of implantation.

 

The study pathologist hypothesized that a contaminant, inducing an allergic reaction heralded by eosinophils, was introduced during hand mounting of the stents before implantation.  The study pathologist was also involved in another study conducted by this same group of investigators in which all the stents were pre-mounted before implantation.  Interestingly, no eosinophils were observed in the inflammatory response to these five NIR Conformer metal stents (the control stent used in this study) implanted for 28 days.

 

No histologic differences were noted between Formulation A and B.

 

 

2.5.            RVF01-049 and 050 (Translute Moderate Release, 1ug/mm2 paclitaxel)

 

Device Description:    Express (referred to as bare metal stent or BMS); Translute-coated moderate release formulation + 1ug/mm2 (referred to as Translute MR+ paclitaxel); stent dimensions:  3.0 x 16 mm or 3.5 x 16 mm

 

Translute MR +paclitaxel stents were implanted in 23 swine (1 or 2 stents per animal) and Express bare metal stents implanted in 21 swine (1 or 2 stents per animal) for duration of 28, 90, 180 and 360 days. There was one death (euthanized- dermatitis) reported in the bare metal stent group (180 day cohort) and one death (aortic laceration) in the Translute-coated stent group (28 day cohort).

 

Histopathology and Morphometric Findings:

 

28 Days.

Bare metal stent. Complete endothelialization without evidence of mural thrombus.  Widely patent lumen with thin neointima.  IEL generally intact with a few focal disruptions. Media well defined and EEL intact.  Inflammatory response consisting of macrophages and occasional foreign body giant cells.  No para-strut amorphous material (PAM also referred to as drug effect, thrombus or strut fibrin).

Translute MR + paclitaxel stent. Endothelialization partially complete.  Thin mural thrombus variably present in close proximity to the struts. Lumen widely patent with thin neointima.  IEL and EEL with few disruptions.  Media alterations consisting of significant loss of smooth muscle cells and replacement by fibrous tissue.  Moderate amounts of PAM were present in the majority of the sections.  Focal regions of dystrophic calcification were also noted associated with PAM.

 

90 Days.

Bare metal stent.  Similar histologic findings as described for the 28 days explants.

Translute MR + paclitaxel stent. Similar findings as described for the 28 day explants except that mural thrombi were rarely present and the majority of the smooth muscle cells lost in the media were replaced by fibrous tissue.  Moderate amounts of PAM were present and dystrophic calcification was occasionally present in PAM and the media.  Eosinophils were noted in the inflammatory response in two of the explants.

 

180 Days.

Bare metal stent.  Similar histologic findings as reported for the 28 and 90 day explants; however, PAM was present in small amounts.

Translute MR +paclitaxel stent.  Endothelialization complete with a small thin focal mural thrombus present.  Widely patent lumen with thin neointima IEL and EEL intact with few focal disruptions.  Extensive alterations (loss of smooth muscle cells) involving 75-100% of the media including the presence of dystrophic calcification.  Inflammatory response ranged from mild to moderate with eosinophils noted in one explanted stent.

 

360 Days.

Bare metal stent.  Similar histologic findings as reported for the 28, 90 and 180 day explants except mild to moderate alterations of the media were noted (smooth muscle cell loss, neovascularization, increased fibrous tissue).

Translute MR+ paclitaxel stent.  Endothelialization complete without evidence of mural thrombus. Extensive changes involving 75-100% of the media (smooth muscle cell loss, neovascularization, increased fibrous tissue) including the presence of dystrophic calcification.  Moderate amounts of PAM were present.

 


Morphometric Comparisons

 

 

Lumen Area

Strut-based Intima Thickness

Neointimal Area

Stenosis  (%)

28 Days

 

 

 

 

BMS (10)

    5.81

207

1.05

15.4

Translute MR +paclitaxel (10)

6.03

196

1.14

16.3

 

 

 

 

 

90 Days

 

 

 

 

BMS (10)

      6.82

193

1.15

14.6

Translute MR+paclitaxel (10)

6.02

184

1.34

18.3

 

 

 

 

 

180 Days

 

 

 

 

BMS (10)

6.6

188

1.38

17.0

Translute MR +paclitaxel (9)

5.31

270

1.96

27.0

 

 

 

 

 

360 Days

 

 

 

 

BMS (5)

7.13

191

1.41

16.0

Translute MR +paclitaxel (5)

6.34

194

1.64

20.0

t-test, no significant difference between means

 

Study Pathologist Comments

 

Medial Changes.  Substantially greater histologic alterations were observed in the Translute + paclitaxel explants (e.g., inflammation, neovascularization, basophilic debris, dystrophic calcification) as compared to control stents.  The observed alterations in the media progressed from focal areas of necrosis (28 days), to regions of smooth muscle cell loss and fibro-cellular remodeling (repair) in the media (90, 180 and 360 days).

 

Dystrophic Calcification.  Focal dystrophic calcification occurred more frequently in the Translute +paclitaxel stents (100%) as compared to the bare metal control (80%).  The regions of dystrophic calcification, commonly seen in the 360 day Translute +paclitaxel explant series, were quite small.  Due to the small size of these sites of dystrophic calcification the study pathologist was of the opinion the structural integrity of the coronary artery and stent would not be compromised.

 

Study Pathologist Observations: PAM (Thrombus)

 

The following comments concerning PAM (M03002/M3Appendix 3802.00):

 

                Carstairs stain demonstrated fibrin around stent struts in both control bare metal stents and Taxus stents.

                               

                Fibrin staining was more prominent in the Translute + paclitaxel stents.

 

                Less fibrin stained material was noted in the Taxus series of explants at 360 day versus 28 days, while no fibrin staining was observed in control bare metal series following 360 days of implantation.

 

                The transition from a confluent mass of fibrin stained acellular eosinophilic material around the struts to smaller fibrin staining regions was associated with foamy macrophages suggesting that the fibrin is being cleared by the macrophages. 

 

                Paclitaxel does not cause fibrin deposition, but may inhibit the migration of macrophages required for the removal of the thrombotic material.

 

 

2.6.            BS5P Overlapping Stent (Translute Moderate Release, 1ug/mm2 paclitaxel)

 

Device Description:             Express (referred to as bare metal stent, BMS); Translute moderate release formulation + 1ug/mm2; total loaded dose- 50 ug (referred to as Translute MR+ paclitaxel)

 

Stent dimensions:  3.0 x 8 mm or 3.5 x 8 mm; overlapping 4 mm

 

Translute MR +paclitaxel stents were implanted in 36 swine (28 days- 9 overlapping pairs; 90 days- 9 overlapping pairs; 180 days- 12 overlapping pairs) and Express bare metal stents implanted in 36 swine (28 days- 6 overlapping pairs; 90 days- 7 overlapping pairs; 180 days- 4 overlapping) for duration of 28, 90, 180 and 360 days. There was one death (intraoperative- air embolism) reported in the bare metal stent group (180 day cohort) and one death (intraoperative- air embolism) in the Translute MR + paclitaxel stent group (180 day cohort).

 

Histopathology and Morphometric Findings:

 

IEL, EEL and medial cell loss descriptive observations were not reported.

 

28 Days.

Bare metal stent.  Endothelialization was complete without evidence of mural thrombus.  A mild inflammatory response was present.  No PAM or dystrophic calcification. Similar findings were reported for the proximal, overlapping and distal stent segments.

Translute MR + paclitaxel stent. Endothelialization was partially complete in the proximal and overlapping stent segments and complete in the distal segment.  No mural thrombus formation was present and a moderate inflammatory response was present without evidence of eosinophils.  Extensive and moderate amounts of PAM were present in the overlapping stent segments, while mild amounts were noted in the proximal and distal segments.

 

90 Days.

Bare metal stent.  Similar findings as reported for the 28 day explants.

Translute MR + paclitaxel stent.  Similar findings as reported for the 28 day explants; however, greater amounts of PAM were present in the overlapping stent segments.

 

180 Days.

Bare metal stent. Similar findings as reported for the 28 and 90 day explants.

Translute MR + paclitaxel stent. Similar findings as reported for the 28 and 90 day explants, except the amount of PMA is generally decreasing in all stent segments.

 


Morphometric Comparisons: 28 Days

 

Intima Area

Strut-based Intima Thickness

Neointimal Area

Lumen Area

Stenosis        (%)

Proximal

 

 

 

 

 

BMS (6)

NR

0.37

NR

5.05

29.26

Translute MR +paclitaxel (9) 

NR

0.25

NR

 5.55

18.32*

 

 

 

 

 

 

Overlapping

 

 

 

 

 

BMS (7)

NR

0.39

NR

5.73

33.79

Translute MR +paclitaxel (7, 9)

NR

0.21

NR

6.54

19.26*

 

 

 

 

 

 

Distal

 

 

 

 

 

BMS (4)

NR

0.30

NR

5.73

18.46

Translute MR +paclitaxel (9)

NR

0.24

NR

     5.34

15.47

* significant,  p<0.05

 

Morphometric Comparisons: 90 Days

 

 

Intima Area

Strut-based Intima Thickness

Neointimal Area

Lumen Area

Stenosis        (%)

Proximal

 

 

 

 

 

BMS (6)

NR

0.34 

NR

5.58

29.61

Translute MR +paclitaxel (9) 

NR

0.30  

NR

6.66

27.51

 

 

 

 

 

 

Overlapping

 

 

 

 

 

BMS (7)

NR

0.34

NR

5.37

34.65

Translute MR +paclitaxel (7, 9)

NR

0.28

NR

6.64

34.22

 

 

 

 

 

 

Distal

 

 

 

 

 

BMS (4)

NR

0.32

NR

5.43

25.22

Translute MR +paclitaxel (9)

NR

0.34

NR

6.11

31.73

*significant, p<0.05

 


Morphometric Comparisons: 180  Days

 

 

Intima Area

Strut-based Intima Thickness

Neointimal Area

Lumen Area

Stenosis        (%)

Proximal

 

 

 

 

 

BMS (6)

NR

0.22

NR

6.73

19.92

Translute MR +paclitaxel (9) 

NR

0.24

NR

6.06

24.65

 

 

 

 

 

 

Overlapping

 

 

 

 

 

BMS (7)

NR

0.26

NR

6.43

25.79

Translute MR +paclitaxel (7, 9)

NR

0.24

NR

5.00

40.59*

 

 

 

 

 

 

Distal

 

 

 

 

 

BMS (4)

NR

0.26

NR

6.48

20.78

Translute MR +paclitaxel (9)

NR

0.37

NR

4.95

31.82

*significant,  p<0.05

 

 

2.7.            RVF01-069 (Translute Slow Release, 1ug/mm2 paclitaxel)

 

Device Description:             Express (referred to as bare metal stent, BMS); Translute slow release formulation + 1ug/mm2; total loaded dose- 108 µg (referred to as Translute SR+ paclitaxel)

 

Stent dimensions:  3.0 x 16 mm or 3.5 x 16 mm; overlapping 8 mm

 

Translute SR +paclitaxel stents were implanted in 10 swine (2 overlapping pairs per animal) and Express bare metal stents implanted in 10 swine (2 overlapping pairs per animal) for duration of 180 days. There were no deaths in either group.

 

Histopathology and Morphometric Findings:

 

No edge effects noted in the proximal and distal coronary arteries.  No histologic changes noted in kidney, liver or myocardium distal to the stent.

 

180 Days.

Bare metal stent.  Endothelialization complete without evidence of mural thrombus. Medial smooth muscle cell loss was minimal (ranging from none to < 25%). IEL and EEL were essentially intact with focal disruptions noted in the IEL.  The inflammatory response ranged from none to mild.  No PAM was observed in any stent segments. 

Translute SR + paclitaxel stents.  Endothelialization complete without evidence of mural thrombus.  Significant medial cell loss was noted in all stent segment regions.  The inflammatory response ranged from mild to moderate. IEL and EEL were essentially intact with focal disruptions noted in the IEL. PAM was rarely present (proximal- 5/8 none; overlapping- 5/8; distal- 6/8).

 


Morphometric Comparisons: 180 Days

 

 

Intima Area

Strut-based Intima Thickness

Neointimal Area

Lumen Area

Stenosis        (%)

Proximal

 

 

 

 

 

BMS (7)

NR

0.22

1.83

7.18

20.07

Translute SR +paclitaxel (8) 

NR

0.25*

1.76

  6.61

21.65

 

 

 

 

 

 

Overlapping

 

 

 

 

 

BMS (7)

NR

0.27

2.17

6.08

26.76

Translute SR +paclitaxel (8)

NR

0.35*

3.00*

5.78

34.43

 

 

 

 

 

 

Distal

 

 

 

 

 

BMS (7)

NR

0.20

1.29

6.93

15.56

Translute SR +paclitaxel (8)

NR

0.29*

2.20

 5.57

27.13*

* significant,  p<0.05

 

 

 

2.8.            A203 Moderate Release Formulation Dose Response-2 and 4ug/mm2 paclitaxel (28,90 and 180Days)

 

Bare metal stent and paclitaxel, 1 ug/mm2 findings not reported (refer to previous

studies).  The Translute moderate release formulation stents contained either 2 ug/mm2 or  4 ug/mm2 paclitaxel (n=12).

 

28 Days.  Endothelialization was nearly complete.

Two microthrombi noted in the 4 ug/mm2 group.  The inflammatory response was mild

to moderate.  Medial changes were described as mild to moderate.  PAM was present in mild to moderate amounts. 

90 Days.  Similar findings as the 28 days explants, except marked smooth muscle loss occurred in the media and dystrophic calcification was present in many sections. PAM was present in mild to moderate amounts.

180 Days.  Similar findings as the 28 and 90 day explants; however, no microthrombi were observed. The media was thin and at times unrecognizable similar to 90 day explant findings. Less PAM (mild- 2ug/mm2 > 4 ug/mm2) was present as compared to the 28 and 90 day explants.

 

2.9.            Reviewer’s Comments

 

PAM (fibrin, thrombus) associated with the struts of the Translute+ paclitaxel stents is a unique device-related pathology finding that may be related to the pharmacology of paclitaxel (i.e., stabilization of microtubules) resulting in a retarded rate of macrophage phagocytosis and clearance of the thrombotic material.   While PAM was also present in both the bare metal stent and Translute (polymer only) coated stent explants; the magnitude and frequency of PAM was less than that observed in the Translute+paclitaxel series of explants.

 

Translute + paclitaxel produced marked changes in the media (i.e., smooth muscle cell loss, in growth of fibrous tissue, revascularization, dystrophic calcification) of the vessel wall.  Extensive and severe medial changes were observed in the 2ug/mm2 and 4 ug/mm2 studies. However, in the 1ug/mm2 series of explants none of these changes resulted in structural deterioration (e.g., dissection, aneurysm formation) of the coronary artery wall or stent as observed in the preclinical in vivo studies in normal swine. 

 

Long-term clinical studies will be necessary to ensure the safety and effectiveness of this device, due to the limitations of preclinical in vivo studies such as an accelerated rate of healing including re-endothelialization of the luminal surface and implantation of the device in a non-diseased coronary artery.  The limitations of studies conducted in normal swine would be expected to underestimate the potential for 1) acute (< 30 days) and late (> 30 days) thrombosis associated with unstable coronary artery plaque; 2) adverse events such as vessel wall dissection and/or aneurysm formation secondary to slower rates of healing associated with paclitaxel induced vessel wall injury and the subsequent loss of smooth muscle cells in the media; and 3) structural deterioration associated with the presence dystrophic calcification resulting from the incomplete clearance of strut-related thrombus (fibrin) and cellular remnants (e.g., platelets, RBCs, WBCs and smooth muscle cell remnants in the media).

 

Eosinophils have been observed in the inflammatory response to both BMS and Translute +paclitaxel stents.  This observation suggests that the presence of eosinophils may be a response to an allergen introduced during implantation. Stratification of the data based on the use of hand loaded versus preloaded stents may clarify this issue.

 

Edge effect data was collected in a single study (Translute SR, 1ug/mm2 paclitaxel).  No edge effects were observed in the proximal and distal coronary arteries

 

No evidence of myocardial injury was observed in myocardium immediately adjacent to the Translute + paclitaxel stents or myocardium sampled downstream of stented coronary arteries.


3.0             Clinical Summary

 

Date:               October 23, 2003

 

From:              John E. Stuhlmuller, M.D.

                        CDRH/ODE/DCD/ICDB

                       

Subject:           TAXUS™ Express2 Paclitaxel-Eluting Coronary Stent System

 

3.1       Introduction

 

The abbreviations used in this summary are the same ones used in the sponsor’s report in the section entitled Nonclinical and Clinical Testing Summary.

 

3.1.1    Device Description

 

The device to be marketed is the TAXUS™ Express Paclitaxel-Eluting Stent mounted on the Express2 Monorail and Express2 OTW delivery systems.  The Express stent is laser-cut from a 316L stainless steel tube into a specific geometric pattern.  The bare metal stent is coated with paclitaxel that has been incorporated into an elastomeric polymer that serves as a carrier.  The formulation of the paclitaxel and polymer can be varied based on the weight to weight ratio of the paclitaxel to polymer.  The sponsor has characterized the release characteristics as slow, moderate and fast based on the weight to weight ratio of the drug and polymer as outlined in the following table.

 

Release rate

Formulation

Paclitaxel Weight

Polymer Weight

Slow

8.8%

8.8%

91.8%

Moderate

25%

25%

75%

Fast

35%

35%

65%

 

The coating is sprayed on the abluminal (internal) and adluminal (external) stent surfaces. 

 

The nominal total loaded drug dose by weight and total polymer by weight per stent is outlined in the following table.

 

 

Nominal Total Loaded Drug (TLD) Dose per Nominal Stent Length

 

Nominal Stent

Length

8

mm

12

mm

16

mm

20

mm

24

mm

28

mm

32

mm

TLD

Paclitaxel/Stent

(µg)

 

50

 

79

 

108

 

137

 

151

 

180

 

209

Total Polymer

/Stent

(µg)

 

519

 

819

 

1119

 

1419

 

1570

 

1870

 

2170

Total Coating

Weight*/Stent

(µg)

 

569

 

898

 

1227

 

1556

 

1721

 

2050

 

2379

*Total Coating Weight= TLD Paclitaxel/Stent (µg) + Total Polymer/Stent (µg)

 

The stent coating is applied to the stent at a dose density of 1 µg/mm2.  The same stent design is used for implantation in arteries with reference vessel diameters of 2.5 to 3.75 mm.  The stent is crimped onto different size delivery balloons.  Therefore, the total drug dose per stent as a function of the stent length remains the same across the diameter range for any given stent length.

 

The stent matrix that the sponsor is requesting approval to market using the slow-release formulation is outlined in the following table.

 

 

Proposed Stent Matrix for Marketing

 

 

Stent Lengths

Stent

Diameters

8

mm

12

mm

16

mm

20

mm

24

Mm

28

mm

32

mm

2.50 mm

X

X

X

X

X

na

na

2.75 mm

X

X

X

X

X

X

X

3.00 mm

X

X

X

X

X

X

X

3.50 mm

X

X

X

X

X

X

X

na-not available

 

3.1.2    Intended Use Statement

 

The proposed indication for use is the following:

 

The TAXUS Express2 Paclitaxel-Eluting Coronary Stent System is indicated for improving luminal diameter and reducing restenosis for the treatment of de novo lesions Ł 28 mm in length in native coronary arteries ≥ 2.5 and ≤ 3.75 mm in diameter.

 

3.1.3    Overview of Clinical Studies

 

In their PMA application, the sponsor has provided data for the following clinical studies that are applicable to the evaluation of safety and effectiveness of the TAXUS™ stent for the treatment of discrete lesions in de novo coronary artery lesions:

 

 

Data from the TAXUS IV study is considered the pivotal data set.  Data from the TAXUS I, TAXUS II and TAXUS V studies are considered supporting data sets.

 

Data from the TAXUS III and TAXUS V ISR (in-stent restenosis) studies are not considered directly applicable to proposed intended use and are not reviewed.  However, they are being acknowledged for completeness.

 

As noted in Section 9.4 of the sponsor’s Nonclinical and Clinical Testing Summary, the TAXUS I and II studies used the 15 mm NIR stent platform.  Single de novo lesions with lesion length of ≤ 12 mm and reference vessel diameters (RVD) of 3.0 to 3.5 mm were treated.  QCA and IVUS imaging were completed at 6 months. 

 

The TAXUS IV study used the Express™ stent platform with lesion lengths to ≤ 28 mm and the RVD from 2.5 to 3.75 mm.  The TAXUS V study further extended the lesion length to ≤ 46 mm and the RVD from 2.25 to 4.0 mm using the Express stent platform.  The TAXUS IV and V studies have been conducted in the United States.  In addition, the TAXUS IV and V studies allowed multiple vessel treatment with one vessel randomized as part of the study and the second vessel being treated with standard percutaneous coronary interventions.

 

Follow-up through five years for the TAXUS I, II, IV and V studies is currently being conducted by the sponsor.


3.2       The TAXUS IV Trial

 

The TAXUS IV study is considered the pivotal marketing study completed by the sponsor for evaluation of safety and effectiveness for the TAXUS™ Paclitaxel-Eluting Coronary Stent System.

 

The sponsor’s full clinical report and the investigational plan for the TAXUS IV study have been provided as appendices in Volume 2 of the Panel Pack (hard copy). 

 

3.2.1    Trial Design

 

This was a domestic, multi-center, prospective, randomized, double-blind comparison of TAXUS™ drug-eluting stent to a bare (uncoated) metal control stent (Express™ stent) in patients with de novo coronary artery lesions.  The minimum projected sample size was 1172 subjects.  The primary clinical endpoint was target vessel revascularization (TVR) at 9 months.

 

Informed consent was obtained based on the general patient selection criteria.  A 1:1 randomization to either the TAXUS™ stent or the bare metal control stent (Express™) was completed after completion of the coronary angiogram and assessment of the angiographic selection criteria and “prior to pre-dilation” using an Interactive Voice Randomization System.  Randomization was stratified by clinical site, vessel size diameter (< 3.0 mm and ≥ 3.0 mm) and the presence or absence of medically treated diabetes.  At the time of randomization a designation was also made as to whether the patient would complete angiographic and/or IVUS follow-up.

 

The following table interpolates the stent matrix and lesion characteristics (small versus large diameter and short versus long lesions).  The bold lines delineate the characterization of lesion length and vessel diameter.  The intent of the stratified randomization was to ensure that an adequate number of patients were enrolled in the various cohorts to allow assessment of whether a treatment effect was noted that would affect the poolability of the data.  The study was not adequately powered to reach any specific conclusions regarding safety and effectiveness in the subgroups such as medically treated diabetes.

 

 

Stent Length

(mm)

Stent diameter

(mm)

2.50

3.00

3.50

16

 

 

 

 

 

 

24

 

 

 

32

 

 

 

 

 

 

Legend

Color

Single Stent Use

 

Small diameters      (< 3.0 mm)

 

Long lesions            (32 mm stent -28 mm lesion length)

 

Large diameters

 

Not available

Color

Multiple Stent Use

 

Provisional overlapping for bail-out stenting

 
3.2.1.1 Eligibility

 

Patients had to meet the following highlighted criteria for inclusion:

 

 

 

Patients were excluded from the study by the following criteria:

 

 

 

Staged procedures were permitted and the criteria are outlined in the appendix containing the investigational plan provided in Volume 2 of the Panel Pack (hard copy).

 

3.2.1.2 Treatment

 

Balloon angioplasty was the only pretreatment allowed.  The target stent/vessel ratio was 1.1:1 and re-dilation was permitted to achieve that result.  Additional study stents were allowed for incomplete coverage or to treat dissection.  Direct stenting was not permitted.

 

Concomitant medical therapy consisted of ASA and either clopidogrel or ticlopidine.  The clopidogrel loading dose (300 mg) was to be administered pre-procedure or within 2 hours post-procedure.  Use of IIb/IIIa inhibitors in conjunction with the procedure was permitted at the discretion of the investigator.  Use of direct thrombin inhibitors as an alternative to heparin was not included in the investigational plan.  Following the procedure, patients were to use ASA indefinitely and either clopidogrel or ticlopidine for a minimum of 6 months.

 

3.2.1.3 Follow-up

 

At discharge, all patients had evaluation of clinical status and laboratory evaluation including a complete blood count.

 

At 30 days (± 7 days) all patients returned for an office visit that included clinical assessment, review of current medications, evaluation for adverse events, an ECG and phlebotomy for measurement of paclitaxel levels.

 

At 4 months (± 14 days) all patients returned for an office visit or telephone follow-up that included clinical assessment, review of current medications, evaluation for adverse events and an ECG.

 

At 9 months (± 14 days), patients returned for an office visit that included clinical assessment, review of current medications, evaluation for adverse events and an ECG (office visits).  Angiography and or IVUS were completed in those patients enrolled in the respective sub-studies.

 

At 12 months through 60 months in 12 month increments (± 30 days), telephone follow-up evaluating clinical status adverse events and aspirin use is scheduled.  This follow-up is in progress.

 

Angiograms were evaluated at the QCA core laboratory at Brigham and Women's Angiographic Core Laboratory in Boston, MA.  IVUS images were evaluated in the IVUS core laboratory at the Cardiovascular Research Institute in Washington, D.C.  The data coordinating center was PAREXEL.  The study was monitored by a DSMB, and a CEC adjudicated clinical events.  The sponsor provided FDA periodic blinded safety updates throughout the study. 

 

3.2.1.4 Endpoints & Analyses

 

The primary endpoint was designated as adjudicated ischemia-driven target vessel revascularization (TVR) at 9-months.  TVR is defined as any ischemia-driven repeat percutaneous intervention of the target vessel or bypass surgery of the target vessel.

 

Secondary endpoints included MACE (defined as all cardiac deaths versus deaths limited to the target vessel, MI, and TVR) at 9-months.   QCA assessment included in-stent and in-lesion (analysis segment) minimum luminal diameter (MLD) and binary restenosis from the 9-month angiography.  IVUS assessment included volumetric measurements (for example, luminal area and neointimal area) and evaluation of stent apposition.  Summaries of the QCA and IVUS core laboratory procedures are outlined in Appendices to the sponsor’s report.

 

3.2.2    Description of Patients

 

The study enrolled 1326 patients at 73 study sites in the U.S. beginning in March 2002.  Nine-month follow-up was completed in April 2003.  A total of 12 patients were deregistered due to failed pre-dilation, consent withdrawal or lack of appropriate study-stent size.  Deregistration was balanced between the study arms. 

 

All patients who were implanted with at least one study stent were considered registered in the study.  Follow-up information is based on the 1324 patients considered registered in the study.  Although the sponsor uses the term intent-to-treat analysis, FDA considers the analysis to be based on the number of patients actually treated. 

 

The table below highlights patient characteristics at enrollment of the two study arms. The groups are balanced with respect to patient characteristics. 

 

TAXUS IV Patient Characteristics

 

 

TAXUS™

Control

Number of Patients

662

652

Age

62 ± 11

62 ± 11

% Male

72%

72%

Diabetes

23%

25%

Hypertension

71%

69%

Hyperlipidemia

65%

66%

Current smoker

23%

20%

Prior MI

31%

30%

Prior PTCA

29%

32%

Prior CABG

10%

10%

Ejection Fraction

0.55

0.55

CHF

5%

7%

Angina CSS III or IV

25%

26%

Silent Ischemia

13%

15%

 

The table below highlights QCA lesion characteristics of the two study arms. The groups are balanced with respect to lesion characteristics.

 

TAXUS IV Lesion Characteristics

 

 

TAXUS™

Control

Number of Patients

662

652

Lesion in LAD

40%

41%

Thrombus

1%

2%

Aneurysm

2%

1%

Total Occlusion

2%

2%

RVD (QCA)

2.8 ± 0.5

2.8 ± 0.5

MLD (QCA)

0.92 ± 0.3

0.95 ±  0.3

% Stenosis

67± 11

66 ± 11

Lesion Length (QCA)

  • < 10 mm
  • 10 - < 20 mm
  • 20 – 28 mm
  • > 28 mm

 

33%

54%

11%

3%

 

35%

50%

12%

3%

 

 

3.2.3 Stent Use by Stent Diameter and Length

 

The following table summarizes the number of TAXUS stents implanted by stent diameter and length.  The bold lines delineate the characterization of lesion length and vessel diameter shown in the table above in Section 3.2.1.

 

 

Stent Length

(mm)

Stent diameter

(mm)

2.50

3.00

3.50

Total

16

104

209

125

438

24

40

57

43

140

32

N/A

81

38

119

Total

144

347

206

697

 

The following table summarizes the number of control stents implanted by stent diameter and length.

 

 

Stent Length

(mm)

Stent diameter

(mm)

2.50

3.00

3.50

Total

16

117

184

134

435

24

27

66

48

141

32

N/A

72

36

108

Total

144

322

218

684

 

 

The tables indicate that stent implantation by stent diameter and length was similar between the two study groups.

 

3.2.4    Clinical Results

 

3.2.4.1 Study Endpoint Analysis

 

The primary endpoint was designated as adjudicated ischemia-driven target vessel revascularization (TVR) at 9-months.  TVR was defined as any ischemia-driven revascularization of the target vessel, myocardial infarction (Q- or non-Q-wave) related to the target vessel or cardiac death related to the target vessel.  In the event that it could not be determined with certainty whether an MI or death was related to the target vessel, it was considered a TVR.

 

TAXUS IV Primary Endpoint: TVR at 9 Months

 

 

TAXUS™

Control

Difference

[95% C.I.]

p-value

TVR

(crude rate)

4.7%

(31/662)

12.0%

(78/652)

-7.3%

[-10.2%, -4.3%]

0.0001

 

These results demonstrate a drug effect of paclitaxel in reducing TVR at 9 months.  The absolute reduction in TVR is 7.3% and the relative reduction is 61%.  The difference in TVR is due to a reduction in the number of target lesion revascularizations versus differences in the rates of death or myocardial infarction in the TAXUS™ arm versus the control arm.

 

The following two tables highlight the principal effectiveness and safety results.  The analysis segment includes the five mm margins proximal and distal to the stent.

 

 

 

TAXUS IV Principal Effectiveness Measures at 9 Months

 

 

TAXUS™
(N=662)

Control
(N=652)

Difference
[95% CI]

Clinical Procedural Success

97.3%

(643/661)

97.4%

(635/652)

-0.1%

[ -1.9%, 1.6%]

Technical Success

97.9%

(648/ 662)

98.2%

(640/ 652)

-0.3%

[ -1.8%, 1.2%]

9-month Target Vessel Revascularization

4.7%

(31/ 662)

12.0%

(78/ 652)

-7.3%

[-10.2%, -4.3%]

9-month

In-stent restenosis

5.5%

(16/ 291)

24.4%

(65/ 266)

-18.9%

[-24.7%,-13.1%]

9-month

Analysis segment restenosis

7.9%

(23/ 291)

26.6%

(71/ 267)

-18.7%

[-24.8%,-12.5%]

MLD (mm),  In-Stent

     Post-Procedure

2.65 ± 0.42

2.67 ± 0.41

-0.01

[ -0.07,0.05]

     9-Month

2.26 ± 0.58

 

1.75 ± 0.65

 

0.51

[  0.41,0.61]

MLD (mm), Analysis Segment

     Post Procedure

2.26 ± 0.48

2.29 ± 0.50

-0.03

[ -0.10,0.04]

     9-Month

2.03 ± 0.55

1.68 ± 0.61

0.35

[0.26, 0.45]

Diameter Stenosis, In-Stent (%)

      Post Procedure

4.21 ± 10.84

5.16 ± 11.41

-0.95

[ -2.57,  0.67]

     9-Month

17.43 ±17.71

 

37.24 ± 19.76

 

-19.82

[-22.93,16.70]

Diameter Stenosis, Analysis Segment (%)

   Post Procedure

19.16 ± 9.67

19.33 ± 10.45

-0.17

[ -1.63,  1.29]

   9-Month

26.29 ± 15.45

39.79 ± 18.45

-13.50

[-16.31,-10.68]

Late Loss, In-Stent (mm)

0.39 ± 0.50

0.92 ± 0.58

-0.53

[ -0.62, -0.44]

Late Loss, Analysis Segment (mm)

0.23 ± 0.44

0.61 ± 0.57

-0.38

[ -0.47, -0.30]

 

The effectiveness data demonstrates that the principal benefit is through a reduction in target vessel revascularization in the TAXUS™ arm compared to the control arm as a function of reducing restenosis.

 

 

TAXUS IV Principal Safety Measures at 9 Months

 

 

TAXUS™
(N=662)

Control
(N=652)

Difference
[95% CI]

In-hospital MACE

2.4%

(16/662)

2.1%

(14/ 652)

0.3%

[ -1.3%,  1.9%]

Out-of-Hospital MACE to 9-months

6.2%

(41/ 662)

13.0%

(85/ 652)

-6.8%

[-10.0%, -3.7%]

MACE to 9-months

8.5%

(56/ 662)

15.0%

(98/ 652)

-6.6%

[-10.0%, -3.1%]

TVR to 9-months (Primary Endpoint)

4.7%

(31/ 662)

12.0%

(78/ 652)

-7.3%

[-10.2%, -4.3%]

Stent Thrombosis (to 30 days)

0.3%

(2/ 662)

0.3%

(2/ 652)

-0.0%

[ -0.6%, 0.6%]

Stent Thrombosis (to 9 months)

0.6%

 (4/ 662)

0.8%

(5/ 652)

-0.2%

[ -1.1%, 0.7%]

CVA to 9 months

1.5%

(10/ 662)

0.8%

(5/ 652)

0.7%

[ -0.4%, 1.9%]

Serious Bleeding Complications

2.6%

(17/ 662)

1.8%

(12/ 652)

0.7%

[ -0.9%, 2.3%]

Serious Vascular Complications

1.5%

(10/ 662)

1.8%

(12/ 652)

-0.3%

[ -1.7%, 1.1%]

Platelet Disorders

0.6%

(4/ 662)

0.8%

(5/ 652)

-0.2%

[ -1.1%, 0.7%]

Hematological Dyscrasia to 9 months

1.5%

(10/ 662)

0.5%

(3/ 652)

1.1%

[ -0.0%, 2.1%]

 

The safety data demonstrates that the occurrence of stent thrombosis and other adverse events are similar in the TAXUS arm compared to the control arm through 9 months.

 

Acute results (peri-procedure and hospital period) are highlighted in the table below. 

 

TAXUS IV Acute Results

 

 

TAXUS™

Control

Difference

[95% C.I.]

Number of patients

662

652

 

Acute Procedural Success

97%

97%

-0.1%

[-1.9%, 1.6%]

Acute Technical Success

98%

98%

-0.3%

[1..8%, 1.2%]

Stent Failure

0.1%

0%

0.1%

[-0.1%, 0.4%]

Non-Target Vessel Treated During Index Procedure

21%

18%

3%

[-1.2%, 7.3%]

Procedural Medications

  • Aspirin pre-procedure

 

  • Clopidogrel/Ticlopidine pre-procedure

 

  • GP IIb/IIIa During Procedure

 

99%

 

99%

 

 

58%

 

98%

 

98%

 

 

57%

 

 

0.3%

[-1.0%, 1.7%]

0.8%

[-0.4%, 2.0%]

 

1.0%

[-4.4%, 6.35]

In-Hospital Events

MACE

2.4%

2.1%

-0.3

[-1.3%, 1.9%]

Stent Thrombosis

0.0%

0.3%

-0.3

[-0.7, 0.1]

 

Clopidogrel or ticlopidine was administered pre-procedure in 99% of the patients in the TAXUS arm and 98% in the control arm.  After discharge, patients were to take ASA indefinitely and clopidogrel or ticlopidine for 6 months.  At four months after discharge, 97% of patients were compliant with ASA in both study arms.  Compliance with either clopidogrel or ticlopidine was 93% in the TAXUS arm and 91% in the control arm.

 

 


 

3.2.5    Angiographic Results

Forty percent of the patients enrolled were assigned to have angiographic follow-up at 9 months.  The following table highlights the angiographic measurements in the analysis segment at baseline, post-procedure and at 9 months.

 

TAXUS IV Angiographic Results (Analysis Segment)

Baseline, Post-procedure and at 9 Months

 

 

TAXUS™

Control

Difference

[95% C.I.]

Baseline Measurements

n = 374

n = 356

 

RVD

(mm)

2.76 ± 0.48

2.79 ± 0.48

-0.03

[-0.10, 0.04]

In-Lesion, MLD

(mm)

0.93 ± 0.33

0.95 ± 0.34

-0.03

[-0.07, 0.02]

Diameter Stenosis

(%)

66 ± 11

66 ± 11

0.5

[-1.04, 2.08]

Post-Procedure Measurements

 

 

 

RVD

(mm)

2.79 ± 0.48

2.83 ± 0.48

-0.04

[-0.11, 0.03]

Analysis Segment, MLD

(mm)

2.26 ± 0.48

2.29 ± 0.50

-0.03

[-0.10, 0.04]

Diameter Stenosis, Analysis Segment

(%)

19 ± 10

19 ± 10

-0.17

[-1.63, 1.29]

9-Month Follow-up Measurements

n = 291

n = 267

 

RVD

(mm)

2.75 ± 0.46

2.77 ± 0.47

-0.02

[-0.09, 0.06]

Analysis Segment, MLD

(mm)

2.03 ± 0.55

1.68 ± 0.61

-0.35

[0.26, 0.45]

Diameter Stenosis, Analysis Segment

(%)

26 ± 15

40 ± 18

-13.45

[-16.31, -10.68]

Binary Restenosis, Analysis Segment

(%)

8%

27%

-18.7

[-24.8%, -13.1%]

 

Assessment of the angiographic measurements in the analysis segment at 9 months demonstrates a beneficial treatment effect in the TAXUS™ arm compared to the control arm by a larger mean MLD, reduced percent diameter stenosis and reduced binary restenosis rate.  

 

3.2.6    IVUS Results

 

Twenty percent of the patients were assigned to have IVUS follow-up at 9 months.  The following table highlights the IVUS measurements post-procedure and at 9 months.  While the IVUS follow-up was 975, the number of IVUS measurements was limited to 66% for 2-D analysis and 53% for 3-D (volumetric) analysis.  Denominators are provided for each of the measurements shown below.

 

TAXUS IV IVUS Results – Post-procedure and at 9 Months

 

 

TAXUS™

Control

Difference

[95% C.I.]

Post-Procedure Measurements     Mean ± SD                         Mean ± SD

                                                                (n)                                       (n)

Vessel Volume

(mm3)

283 ± 91

(47)

285 ± 125

(54)

-2.37

[-45.47, 40.73]

Stent Volume

(mm3)

155 ± 61

(88)

151 ± 56

(82)

3.95

[-13.70, 21.60 ]

Lumen Volume

(mm3)

155 ± 61

(87)

151 ± 56

(82)

-0.17

[-1.63, 1.29]

Neointimal Volume

(mm3)

0.03 ± 0.24

(88)

0.00 ± 0.00

(82)

0.03

[-0.02, 0.08]

Incomplete Stent Apposition (%)

11.6%

(13/112)

6.4%

(7/109)

5.2%

[-2.3%, 12.7%]

9-Month Follow-up Measurements

Vessel Volume

(mm3)

280 ± 89

(60)

283 ± 97

(62)

-2.42

[-35.52, 30.68, ]

Stent Volume

(mm3)

142 ± 47

(81)

145 ± 51

(80)

-2.96

[-18.12, 12.21 ]

Lumen Volume

(mm3)

123 ± 43

(80)

104 ± 44

(80)

19.55

[6.07, 33.03]

Neointimal Volume

(mm3)

18 ± 18

(81)

42 ± 23

(80)

-23.92

[-30.33, -17.51]

Incomplete Stent Apposition (%)

4%

(4/99)

3%

(3/100)

1%

[-4.1%, 6.2%]

 

Assessment of the IVUS measurements demonstrate that the vessel wall remains structurally intact as assessed by the vessel and stent volumes in both the TAXUS and control groups.  Larger lumen volumes as a result of lower neointimal volumes are noted in the TAXUS group compared to the control group at 9-month follow-up.  Although a higher incidence of post-procedure incomplete stent apposition is noted in the TAXUS group compared to the control group, the rates are equivalent at the 9-month follow-up.

 

3.2.7    Overlapping Stents

 

In the TAXUS IV study, overlapping stenting was permitted when clinically indicated for sub-optimal results.  A total of 45 patients received more than one stent in the TAXUS IV study.  Data for this subgroup had clinical event rates that tended to be slightly higher overall, but the treatment effects were very similar to those of the entire study.  Angiographic endpoint treatment effects were also very similar. 

 

Table 21 in Section 9.6.1 of the sponsor’s report entitled Nonclinical and Clinical Testing summarizes multiple stent use for all the TAXUS studies. 

 

No specific safety issues are noted in the patients with overlapping stents.

 

 

3.3       The TAXUS I Trial

 

The TAXUS I trial is considered a clinical feasibility study and supporting data set in the evaluation of safety and effectiveness of the TAXUS™ stent. 

 

The sponsor’s full clinical report and the investigational plan for the TAXUS I trial have been provided as appendices on CD-ROM located at the back of Volume 2 of the Panel Pack.

 

3.3.1    Trial Design

 

A total of 61 patients were enrolled in a prospective, blinded, randomized study in patients with de novo native coronary artery lesions at 3 investigational centers in Germany from October 2000 to March 2001. 

 

3.3.1.1 Eligibility

 

Patients had to meet the following highlighted criteria for inclusion:

 

 

 

Patients were excluded from the study for the following criteria:

 

 

 

3.3.1.2 Treatment

 

Balloon angioplasty was the only pretreatment allowed.  The target stent/vessel ratio was 1.1:1, and re-dilation was permitted to achieve that result.  Additional study stents were allowed for incomplete coverage or to treat dissection.  Direct stenting was not permitted.

 

Concomitant medical therapy included aspirin and ticlopidine or clopidogrel.  The protocol specified that the aspirin and ticlopidine or clopidogrel be administered prior to the procedure.  Post-procedure, the aspirin was to be continued indefinitely and the ticlopidine or clopidogrel for a minimum of 6 months.  Procedural use of glycoprotein IIb/IIIa inhibitors was at the discretion of the treating physician.

 

3.3.1.3 Follow-up

 

At discharge, all patients had evaluation of clinical status and laboratory evaluation including a complete blood count.

 

At 30 days (± 7 days) all patients returned for an office visit that included clinical assessment, review of current medications, evaluation for adverse events, an ECG and phlebotomy for measurement of paclitaxel levels.

 

At 6 months (± 30 days) all patients returned for an office visit that included clinical assessment, review of current medications, evaluation for adverse events and an ECG.  In addition, follow-up angiography with IVUS evaluation was conducted at this time. 

 

At 9 and 12 months (± 30 days), patients returned for an office visit that included clinical assessment, review of current medications, evaluation for adverse events and an ECG (office visits).

 

At 24 months through 60 months in 12 month increments (± 30 days), telephone follow-up evaluating clinical status and adverse events is scheduled.  This follow-up is in progress.  An amendment was added incorporating QCA and IVUS evaluation at 24 months in the patients receiving a TAXUS™ stent.

 

Angiograms were evaluated in a QCA core laboratory.  IVUS images were evaluated in an IVUS core laboratory.  The study was monitored by a DSMB, and a CEC adjudicated clinical events.

 

3.3.2    Description of Patients

The patient and lesion characteristics are similar to those seen in patients with coronary artery disease and to those seen in the other TAXUS studies.

 

3.3.3    Study Endpoint Analysis

 

The primary safety endpoint was designated as MACE at 30 days.   The primary efficacy endpoint was angiographic restenosis at 6 months. 

 

The effectiveness and safety outcome measures through 2 years are highlighted in the following table. 

 

TAXUS I Effectiveness and Safety Outcome Measures Through 2 Years

 

Safety Measures and Other Clinical Events

TAXUS NIRxTM

N=31

NIRTM Control

N=30

MACE (30-day)

0%

0%

    Cardiac Death

0%

0%

    Q-Wave MI

0%

0%

    TVR (CABG and/or PCI)

0%

0%

MACE (12-Month )

3%

10%

     Cardiac Death

0%

0%

     Q-Wave MI

0%

0%

     TVR (CABG and/or PCI)

3%

10%

MACE (2-Year)

3%

10%

     Cardiac Death

0%

0%

     Q-Wave MI

0%

0%

     TVR (CABG and/or PCI)

3%

10%

Stent Thrombosis to 2 years

0%

0%

Pre-procedure

     RVD, mm

2.99±0.46

2.94±0.52

     MLD, mm

1.30±0.43

1.23±0.43

     %DS

56.51±12.26

57.82±13.24

     Lesion length, mm

10.70±3.27

11.89±4.93

Post-procedure

     MLD, mm

 

 

     %DS

6.12±9.49

2.95±0.34

6-Month Follow-up

     RVD, mm

3.02±0.47

3.01±0.53

     MLD, mm

2.60±0.49

2.19±0.65

     %DS

13.56±11.77

27.23±16.69

     Restenosis Rate >50%

0%

10%

     Late lumen loss, mm

0.36±0.48

0.71±0.47

     Loss index

0.22±0.29

0.45±0.29

 

Within the limitations of the sample size, no safety issues are noted through 2-year follow-up.

 

The following table summarizes the angiographic results for those patients in the TAXUS™ arm that agreed to completion of an angiogram and/or IVUS evaluation at 2 years.  The baseline and 6 months results are summarized in the preceding table that outlines the principal safety and effectiveness results at 6 months.


TAXUS I Angiographic Results at 6 and 24 Months in the TAXUS™ Arm

 

Lesion Characteristic

NIRx

6-Month Follow-up

 

MLD (mm)

2.640.63

RVD (mm)

2.950.45

% Diameter Stenosis

10.9413.81

2-Year Follow-up

 

MLD (mm)

2.680.41

RVD (mm)

2.920.38

% Diameter Stenosis

8.254.61

>50% Restenosis

0%

Net Gain (mm)

1.460.60

Late Lumen Loss (mm)

0.250.31

Loss Index

0.160.18

2-Year Follow-up – 6-Month Follow-up

 

MLD (mm)

0.040.48

RVD (mm)

-0.030.34

% Diameter Stenosis

-2.6912.86

 

Within the limitations of the sample sizes, the angiographic measurements through 2 years demonstrate a persistent drug effect.  No safety issues are noted.

 

The following two tables summarize the IVUS results.  The first table summarizes the results for the baseline and 6 month evaluations.  The second table summarizes the results for those patients in the TAXUS™ arm that agreed to completion of an IVUS evaluation at 2 years compared to the evaluation at 6 months. 

 

TAXUS I IVUS Results at Baseline and 6 Months

 

Lesion Characteristic

TAXUS NIRx

NIR

Pre-Procedure MLA (mm2)

3.20 ± 0.64

2.92 ± 0.55

Pre-Procedure Vessel Volume (mm3)

232.94 ± 65.40

196.95 ± 69.84

Pre-Procedure Lumen Volume (mm3)

86.86 ± 20.82

74.08 ± 24.01

Post-Procedure MLA (mm2)

6.97 ± 1.69

6.36 ± 1.55

Post-Procedure Maximum Stent Diameter (mm)

3.43 ± 0.38

3.39 ± 0.39

Post-Procedure Normalized Vessel Volume (mm3)

282.06 ± 70.13

261.77 ± 70.45

Post-Procedure Normalized Lumen Volume (mm3)

119.08 ± 27.67

113.05 ± 25.19

Post-Procedure Normalized Stent Volume (mm3)

123.36 ± 28.09

117.04 ± 26.25

Post-Procedure Normalized Vessel – Stent Volume (mm3)

158.70 ± 47.84

144.72 ± 52.07

Post-Procedure Normalized Neointimal Hyperplasia (mm3)

4.28 ± 2.73

4.00 ± 2.64

6-Month Follow-up MLA (mm)

5.57 ± 1.21

4.77 ± 1.31

6-Month Follow-up Maximum Stent Diameter (mm)

3.45 ± 0.28

3.44 ± 0.37

6-Month Follow-up Normalized Vessel Volume  (mm3)

285.97 ± 50.33

270.67 ± 61.54

6-Month Follow-up Normalized Lumen Volume (mm3)

107.67 ± 19.16

97.96 ± 26.39

6-Month Follow-up Normalized Stent Volume (mm3)

122.49 ± 19.49

119.53 ± 26.26

6-Month Follow-up Normalized Vessel – Stent Volume (mm3)

163.48 ± 38.60

151.95 ± 40.88

6-Month Follow-up Normalized Neointimal Hyperplasia (mm3)

14.82 ± 10.83

21.57 ± 10.71

 

TAXUS I IVUS Results at 6 and 24 Months in the TAXUS™ Arm

 

Lesion Characteristic

NIRx

6-Month Follow-up

MLA (mm)

6.301.36

Vessel Volume (mm3)

260.9079.13

Lumen Volume (mm3)

108.8831.26

Stent Volume (mm3)

117.1333.96

Vessel - Stent Volume (mm3)

152.0254.59

Neointimal Hyperplasia (mm3)

8.255.92

2-Year Follow-up

 

MLA (mm)

5.671.08

Vessel Volume (mm3)

237.1253.9

Lumen Volume (mm3)

103.8922.53

Stent Volume (mm3)

113.5427.09

Vessel - Stent Volume (mm3)

133.2336.04

Neointimal Hyperplasia (mm3)

9.658.47

2-Year Follow-up – 6-Month Follow-up

 

MLA (mm)

-0.240.84

Vessel Volume (mm3)

-16.6739.53

Lumen Volume (mm3)

-5.5324.82

Stent Volume (mm3)

-2.8623.45

Vessel - Stent Volume (mm3)

-11.1521.60

Neointimal Hyperplasia (mm3)

2.665.44

 

Within the limitations of the sample sizes, the IVUS measurements demonstrate a persistent drug effect.  The vessel wall remains structurally intact in the TAXUS arm through 24 months.  Larger lumen volumes as a result of lower neointimal volumes are noted in the TAXUS group compared to the control group at 6-month follow-up.  Larger lumen volumes and lower neointimal volumes are maintained through 24-month follow-up.   No safety issues are noted.

 

3.4       The TAXUS II Trial

 

The TAXUS II study is considered a supporting data set in the evaluation of safety and effectiveness.  This study evaluated the slow- and moderate-release formulations on the NIR stent platform in two separate independent patient cohorts referred to in the sponsor’s report as either TAXUS II Cohort I or TAXUS II-SR cohort and TAXUS II Cohort II or TAXUS II-MR cohort. 

 

Data from the moderate-release formulation provides clinical data that allows assessment of safety of a stent platform with the same total drug dose and a different (faster) dose release profile.

 

 

 

3.4.1    Trial Design for TAXUS II SR and TAXUS II MR

 

The TAXUS II-SR cohort enrolled a total of 267 patients at 28 centers in 12 countries randomized to receive treatment with either the TAXUS™ stent with the slow release formulation or a control stent. 

 

The TAXUS II-MR cohort enrolled a total of 270 patients randomized to receive treatment with either the TAXUS™ stent with the moderate release formulation or a control stent. 

 

Enrollment began in May 2001.

 

3.4.1.1 Eligibility for TAXUS II-SR and TAXUS II-MR

 

Patients had to meet the following criteria for inclusion:

 

 

 

Patients were excluded from the study for the following criteria:

 

 

 

3.4.1.2 Treatment in TAXUS II-SR and TAXUS II-MR

 

Direct stenting was not permitted.  Pre-dilation was required.  The target stent/vessel ratio was 1.1:1, and re-dilation was permitted to achieve that result.  Additional study stents were allowed for incomplete coverage or to treat dissection.  

 

Concomitant medical therapy included aspirin and ticlopidine or clopidogrel.  The protocol specified that the aspirin and ticlopidine or clopidogrel be administered a minimum of 4 hour and preferably 24 hours prior to the procedure.  Post-procedure, the aspirin was to be continued indefinitely and the ticlopidine or clopidogrel for a minimum of 6 months.  Procedural use of glycoprotein IIb/IIIa inhibitors was at the discretion of the treating physician.

 

3.4.1.3 Follow-up in TAXUS II-SR and TAXUS II-MR

 

At discharge, all patients had evaluation of clinical status and laboratory evaluation including a complete blood count.

 

At 30 days (± 7 days) all patients returned for an office visit that included clinical assessment, review of current medications, evaluation for adverse events, an ECG and phlebotomy for measurement of paclitaxel levels.

 

At 6 months (± 30 days) all patients returned for an office visit that included clinical assessment, review of current medications, evaluation for adverse events and an ECG.  In addition, follow-up angiography with IVUS evaluation was conducted at this time. 

 

At 12 months (± 30 days), patients could either returned for an office visit or complete telephone evaluation that included clinical assessment, review of current medications, evaluation for adverse events and an ECG (office visits).

 

At 24 months through 60 months in 12 month increments (± 30 days), telephone follow-up evaluating clinical status and adverse events is scheduled.  This follow-up is in progress.

 

Angiograms were evaluated in a QCA core laboratory.  IVUS images were evaluated in an IVUS core laboratory.  The study was monitored by a DSMB, and a CEC adjudicated clinical events.

 

3.4.2    Description of Patients

The patient characteristics in the TAXUS I- SR and TAXUS II-MR patient cohorts are similar to those seen in patients with coronary artery disease and to those seen in the other TAXUS studies.

3.4.3        Study Endpoint Analysis – TAXUS II-SR Cohort

 

The following table highlights the principal effectiveness and safety results for the TAXUS II-SR cohort at 6 months.

 

TAXUS II-SR Cohort Principal Effectiveness and Safety Results at 6 Months

 

Efficacy Measures

TAXUS SR
(N=131)

Control
(N=136)

Difference
[95% CI]

Clinical Procedural Success

95.4%

93.4%

2.0%

[-3.5%, 7.5%]

30-Day MACE

2.3%

4.4%

-2.1%

[-6.4%, 2.2%]

6-Month MACE

8.5%

19.5%

-11.1%

[-19.4%, -2.8%]

  Death

0.0%

0.8%

-0.8%

[-2.2%, 0.7%]

 Q-Wave MI

0.0%

1.5%

-1.5%

[-3.6%, 0.6%]

 Non-Q-Wave MI

1.5%

3.8%

-2.2%

[-6.1%, 1.6%]

 TVR, Overall

7.7%

14.3%

-6.6%

[-14.1%, 0.9%]

 TVR, non-TLR

3.1%

2.3%

0.8%

[-3.1%, 4.7%]

 TVR, TLR

4.6%

12.0%

-7.4%

[-14.0%, -0.8%]

 TVR, CABG

0.8%

0.8%

0.0%

[-2.1%, 2.1%]

6-Month Restenosis (QCA)

Analysis Segment

5.5%

20.1%

-14.7%

[-22.5%, -6.8%]

Stented Segment

2.3%

17.9%

-15.6%

[-22.6%, -8.6%]

Safety Measures

Stent Thrombosis

 

 

 

Ł 1 Day

0.8%

0.0% (0/136)

0.8%

[-0.7%, 2.3%]

Ł 210 Days

0.8%

0.0% (0/136)

0.8%

[-0.7%, 2.3%]

Periprocedural MACE

1.5%

4.4% (6/136)

-2.9%

[-6.9%, 1.2%]

In-Hospital MACE

1.5%

4.4% (6/136)

-2.9%

[-6.9%, 1.2%]

Out-of-Hospital MACE

6.9%

15.6%

-8.7%

[-16.2%, -1.2%]

Serious Bleeding Complications to 210 Days

3.1%

4.4%

-1.4%

[-5.9%, 3.2%]

Serious Vascular Complications to 210 Days

3.1% (4/131)

0.0% (0/136)

3.1%

[0.1%, 6.0%]

 

A reduction in MACE events at 6 months is seen in the TAXUS arm compared to the control arm primarily due to a reduction in target vessel revascularizations.  This clinical benefit is maintained through 1-year follow-up.

 

The following table highlights the IVUS analysis for the TAXUS II-SR cohort at baseline and 6 months.

 

TAXUS II SR IVUS Results at Baseline and 6 Months

 

Measure

TAXUS SR
(N=131)

Control
(N=136)

Difference
[95% CI]

Post-Procedure

 

 

 

Vessel Volume (mm3)

254.67  67.57

248.51  63.27

6.16

[-12.69, 25.02]

Lumen Volume (mm3)

131.14  40.86

129.57  35.76

1.56

[-8.11, 11.23]

Stent Volume (mm3)

131.19  40.99

129.68  35.80

1.51

[-8.18, 11.20]

MLD, Stented Segment (mm)

2.99  0.33

2.98  0.34

0.01

[-0.08, 0.09]

MLA, Stented Segment (mm2)

7.10  1.58

7.06  1.63

0.04

[-0.36, 0.45]

Mean Stent Area (mm2)

8.47  1.76

8.43  1.96

0.04

[-0.43, 0.51]

Mean Lumen Area (mm2)

8.47  1.76

8.42  1.96

0.05

[-0.42, 0.52]

Stent-Lumen Area (mm2)

0.00  0.02

0.01  0.06

-0.00

[-0.02, 0.01]

Stent-Lumen Volume (mm3)

0.06   0.43

0.11  0.84

-0.05

[-0.22, 0.11]

Percent Stented Segment Net Volume Obstruction

0.03  0.18

0.08  0.57

-0.05

[-0.16, 0.06]

6-Month

 

 

 

Vessel Volume (mm3)

266.79  89.66

253.60  71.93

13.19

[-8.60, 34.98]

Lumen Volume (mm3)

118.55  40.53

99.41  38.05

19.14

[9.26, 29.02]

Stent Volume (mm3)

128.80  42.44

129.49  40.03

-0.68

[-11.05, 9.69]

MLD, Stented Segment (mm)

2.83  0.42

2.47  0.53

0.35

[0.23, 0.47]

MLA, Stented Segment (mm2)

6.42  1.84

5.03  2.01

1.39

[0.90, 1.87]

Mean Stent Area (mm2)

8.57  1.85

8.56  1.91

0.01

[-0.47, 0.48]

Mean Lumen Area (mm2)

7.90  1.94

6.62  2.21

1.28

[0.76, 1.80]

Stent-Lumen Area (mm2)

0.67  0.84

1.94  1.50

-1.28

[-1.58, -0.97]

Stent-Lumen Volume (mm3)

10.25  13.26

30.07  25.75

-19.82

[-25.02, -14.63]