A meeting of the Circulatory System Devices Panel was held on March 18, 2009, to discuss and vote on the final clinical data presented in the premarket approval application PMA P080005) for the TherOx Aqueous Oxygen (AO) System. The Panel heard the Company and FDA presentations, discussed the clinical data presented, addressed the FDA questions, and finally voted (9-5-0) to recommend that the PMA application for the TherOx AO System be found “Not Approvable.”
The TherOx AO System is a first-of-a-kind device intended as an adjunct therapy following successful PCI and stent placement in LAD AMI patients. The AO System continually combines (extracorporeally) the patient’s own blood (72ml/min) with a super saturated oxygen solution (3 ml/min) and delivers it, via an infusion catheter, for 90 minutes, directly to the injured site – i.e., the infusion catheter is placed just proximal to the newly placed stent. In theory, the delivery of the super saturated oxygen, directly to the injured site, would reduce the final infarct size as compared to the controls (PCI and stenting alone).
Indications for Use: “The TherOx® Downstream® AO System, Downstream® AO Cartridge, and MI-Cath™ Infusion Catheter are indicated for the preparation and delivery of SuperSaturated Oxygen Therapy (SSO 2 Therapy) to targeted ischemic regions of the patient’s coronary vasculature immediately following revascularization by means of percutaneous coronary intervention (PCI) with stenting that has been completed within 6 hours after the onset of anterior acute myocardial infarction (AMI) symptoms.”
The clinical data presented to the Panel included data from the AMIHOT II study (Acute Myocardial Infarction with HyperOxymic Therapy) , and the use of Bayesian Statistical methods in order to borrow information/strength from the first (failed) AMIHOT I study. Both studies were randomized, unblinded, controlled studies, with the patient cohort for AMIHOT II being a focused subgroup of the wider AMIHOT I patient cohort. This subgroup of patients was identified (post-hoc) from the AMIHOT I study as AMI patients that may benefit from the AO Therapy, and post-hoc analyses of the results from the AMIHOT I study were used to generate hypotheses for the AMIHOT II study.
AMIHOT I (269 evaluable patients at 23 centers, 1:1 randomization)
Patient Cohort : Anterior and non-anterior AMI patients receiving successful
PCI/stenting within 24 hours of initial symptoms
Safety Endpoint : MACE (death, reinfarction, TVR, stroke) with a non-inferiority delta
Effectiveness Endpoint :
- Reduction of Infarct Size
- Improvement of RWMSI at 3 months
- ST-Segment improvement during the first 3 hours
AMIHOT II (301 evaluable patients at 22 centers, 2.8:1 randomization)
Patient cohort : LAD AMI patients receiving successful PCI/stenting within 6 hours
of initial symptoms Safety : same MACE with a non-inferiority delta of 6%
- Reduction of Infarct Size (Primary)
- St-Segment improvement during the first 3 hours (secondary)
Although the data presented at the Panel demonstrated that the study met both the composite primary safety and effectiveness endpoints for the AMIHOT II trial, the Panel voted (9-5-0) to recommend that the PMA application for the TherOx AO System be found “Not Approvable”. This “Not Approvable” recommendation was primarily based on a still undefined potential for risk with only marginal benefit (3.9% mean and 6.5% median absolute difference in infarct size as compared to controls). Concerns included the fact that many of the adverse events (including deaths) trended in the negative direction for the AO Therapy arm, and although the study was not powered to show statistical significance in any of the individual events, the trends were concerning. The Panel was also concerned with the following: 1) the lack of animal data to help understand the mechanism of action for the device, as well as possible physiochemical effects from the therapy and this relationship on the adverse events seen (e.g., stent occlusions and myocardial rupture); 2)the lack of precise safety data - since the safety endpoint was a composite of MACE, the magnitude of the actual effect for each of the individual endpoints could not be established because the study was not powered to demonstrate statistical significance for any of the individual endpoints; and 3) the use of a surrogate (infarct size) as the effectiveness endpoint.
The Panel also suggested ideas for a future study and made some recommendations regarding how the sponsor can focus the data set to better address the safety profile of the device, including additional long-term animal studies to address potential oxygen toxicity and to better understand the physiochemical mechanism of action, as well as better end user training.
Contact: James Swink, Executive Secretary,
(240) 276-4050 James.Swink@fda.hhs.gov
Transcripts may be purchased from: (written requests only)
Free State Reporting, Inc.
1378 Cape St. Claire Road
Annapolis, MD 21409
410-974-0947 or 800-231-8973 Ext. 103
Food and Drug Administration
Freedom of Information Staff (FOI)
5600 Fishers Lane, HFI-35
Rockville, MD 20851
(301) 827-6500 (voice), (301) 443-1726 (fax)