The Neurological Devices Panel (the Panel) met on Monday, February 23, 2004 to review the information in a premarket notification (510(k)) submission for the Concentric Medical, Inc. MERCI Retriever, K033736. The MERCI Retriever is intended to restore blood flow in the neurovasculature by removing thrombi in patients experiencing ischemic stroke.
The sponsor presented the results of the MERCI (Mechanical Embolus Retrieval in Cerebral Ischemia) trial. This was followed by an FDA presentation, two Panel member presentations, and a Panel discussion. The Panel considered FDA’s questions. These questions addressed: 1) The safety of the device, relating to the adverse events occurring in the clinical trial; 2) Whether the clinical data demonstrated successful revascularization; 3) The validity of revascularization as a surrogate endpoint for clinical outcomes (the NIHSS and modified Rankin Score); and 4) Any changes to the proposed indication for use and labeling of the device. The Panel’s responses to the FDA questions as follows, constituted its recommendation for this submission:
Q1. The majority of the Panel members believed that the results of the MERCI trial were inadequate to determine the safety of the device for the new indication of thrombus removal in patients suffering from ischemic stroke. Several members were concerned about the device failures (tip fracture) and others about the mortality experienced in the treated patients. One Panel member believed safety had been demonstrated.
Q2. Ten of the thirteen Panel members agreed that the primary endpoint of revascularization was achieved. They commented the device was clearly capable of removing thrombi from the cerebral vasculature. One Panelist believed the statistical validity of the comparison to a literature control for this endpoint was inappropriate. Another believed that revascularization was an inappropriate efficacy endpoint.
Q3. The Panel’s consensus was that revascularization as a surrogate endpoint was not supported by the data in this study. Several Panel members expressed dissenting views stating that revascularization was an appropriate goal of therapy in this patient population.
Q4. Due to several concerns about the results of the MERCI trial, the majority of the Panel believed no specific recommendations about labeling could be made based on the available data. One Panel member suggested that a warning be included about tip fractures, and another suggested that the device be labeled only for embolic occlusions.
The Panel did not vote on a recommendation regarding the determination of substantial equivalence for this 510(k) submission. The consensus of the Panel was that the safety of the device has yet to be established, and that additional studies would be necessary. Some Panel members believed that the device showed promise and would be a useful tool in the armamentarium of physicians treating acute stroke.
The Panel membership for this meeting consisted of three neurosurgeons, three interventional neuroradiologists, four neurologists, a statistician, a consumer representative, and an industry representative.
Contact: Janet L. Scudiero, Executive Secretary, at 301-594-1184 ext. 176 or firstname.lastname@example.org.
Transcripts of this meeting may be purchased from:
Neal R. Gross and Company
1323 Rhode Island Ave., N.W.
Washington, D.C. 20005
202-234-4433 or 800-473-1433
Food and Drug Administration
Freedom of Information Staff (HFI-35)
5600 Fishers Lane, Rockville, MD 20857