A new application of magnetic resonance imaging (MRI) can now look at blood vessels directly to find subtle changes in the walls, determining if a patient will develop atherosclerosis, which can lead to stroke or heart attack. "With this non-invasive procedure, we can do it better, faster and easier," said Bruce Wasserman, MD, assistant professor of radiology at Johns Hopkins University School of Medicine (Baltimore, Maryland), during an American Medical Association (AMA; Chicago, Illinois) medical imaging media briefing at the Marriott Marquis in New York last month.

The current standard cardiac diagnostic method is an angiogram, in which a needle is injected and catheterization performed in order to diagnose atherosclerotic disease in more advanced stages. In noting the reliance on angiography to determine if plaque is present, Wasserman said, "the problem is that early atherosclerosis doesn't cause narrowing of the lumen in the early stages of the process." In fact, he said, "the blood vessel compensates when plaque starts to form, and it widens. An angiogram simply looks at the inside of the lumen, so if the vessel is compensating, the angiogram will register it as normal." Wasserman was the first researcher to use an intravenous contrast agent to highlight the characteristics of atherosclerosis on MRI, and that continues to be the main focus of his research.

Computed tomography (CT) scans and ultrasound imaging can detect the presence of plaque but cannot determine the severity or composition of plaque buildup, he said. Calcification scoring measures plaque buildup as well, but the plaque may be stable and "non-vulnerable." Earlier detection begins with a simple C-reactive protein blood test to check for inflammation of blood vessels, with inflammation being an indicator of plaque formation. Next, a surface coil is attached to the MRI scanner and laid over the neck area, scanning the carotid artery much like a magnifying glass. It takes "five to 10 minutes to get a good assessment" of plaque forming on the artery walls, Wasserman said. In addition to finding lesions or thickening of the walls, the scan breaks down the plaque composition. "Plaque composition is the key to determining stroke or heart attack vulnerability," he added.

While all forms of plaque cause narrowing of the vessel walls, one plaque "does not behave like the other," Wasserman said. Wasserman noted that with the knowledge ascertained through MRI, treatment and therapy against plaque buildup can begin at a much earlier stage. He said that MRI "may be able to identify changes in the blood vessel that are specific to plaque formation and not from other causes," and best of all, "do so at a very early stage."

The MRI surface coil scan is done on the carotid artery, both because it is big and superficial – amenable to imaging, unlike the aorta, which is dense and harder to see – and because it is the area of greatest stroke risk. This technique is currently being used in a select number of medical centers in the U.S., with Wasserman serving as principal investigator of the Multi-ethnic Study of Atherosclerosis, a trial following 6,000 participants.

Wasserman said he sees this technique as "another screening tool" for plaque composition, with the potential of becoming a favored diagnostic test to be followed by angioplasty as the therapy. In advanced cases, he said, physicians can take what they learn through MRI scanning and "pre-emptively strike against plaque ... and reduce stroke risk" by surgical removal of plaque. "The risk of surgery is far outweighed by the benefit of the reduction in stroke risk," Wasserman noted.

Prioritizing combination products

Panelists from the FDA listened as device and drug manufacturers congratulated the agency on tackling a sticky subject: combination products. Most of the speakers at last month's public hearing – attended by about 150 – supported a philosophy of reviewing and regulating combination products based on the amount of risk. If a device portion of the product was deemed more risky, then the Center for Devices and Radiological Health (CDRH) should get jurisdiction over the product, and to the Center for Drug Evaluation and Research (CDER) if the drug is deemed as having the greater risk.

Michael Gross from Aventis Behring (King of Prussia, Pennsylvania) said, "Combination products should not be referred to as a drug, device or biologic. They should be called 'combination products' because it otherwise adds to a general environment of confusion about the products. It makes the hair on the back of my neck rise when I hear the term misused." Gross's solution: "Maybe you should consider a new application process, something that is not a PMA [premarket approval application] or a 510(k)."

The agency created a combination products program within the Office of the Ombudsman to provide support to both CDER and CDRH "to address concerns about consistency, predictability, and jurisdiction issues related to the management and timeliness when two or more centers have review responsibilities for a combination product," according to the Federal Register announcement of the meeting.

The combination of different components brings new development issues to the forefront, Guy Chamberland, vice president of regulatory affairs and drug development at Angiogene (Montreal, Canada), told the panelists. "The issues involve drugs released from a polymer coating, local safety issues of the drug and polymer, new drug stability issues and drug-device interactions," he said. Criteria used to select the premarket regulatory authority should be based on assuring the safety of patients and not on the primary mode of action, Chamberland noted. "If the risks of the drug outweigh the risks of the device, it should go to CDER. If the risks of the device outweigh the risks of the drug, it should go to CDRH," he said.

Chamberland proposed that the FDA develop a single file for combination products even when one or both of the components are not approved. The team should review the application from the point of view that safety and efficacy is entirely dependent on the combination of the two components, he explained. A premarket review mechanism – which he used as an example – for a drug-device combination product would consist of preclinical studies, nonclinical safety, biocompatibility, physical testing, chemistry, manufacturing and controls, submission of investigational new drug or investigational drug exemption, clinical data and submission of new drug application or PMA.

Expressing a consensus of many speakers was Stewart Portenoy, MD, an independent device consultant and former FDA employee. The FDA should not, he said, create a new office with personnel from the current offices of CDRH and CDER to deal with combination products. "Reviewers need to stay within their current offices of expertise. There could certainly be cross-review teams and administrative support in a new office, but the actual reviewers should remain where they are," he said.

The clinical trial design should determine which office has jurisdiction over a product, said Portenoy, an FDA employee during the early PMA process for the drug-eluting stent from Cordis/Johnson & Johnson (Miami Lakes, Florida). "The trial's design can help determine which division has the best experience. In the Cordis/J&J example, the trial involved a coated and uncoated stent, and the product stayed in CDRH, which had the most experience dealing with the stents," Portenoy said.

The agency will accept written and electronic comments until Jan. 24, 2003, on issues such as intercenter agreements, primary mode of action, selection of premarket regulatory authorities, dual applications, good manufacturing practices and adverse-event reporting.