Medical Device Daily Washington Editor

WASHINGTON – On the second day of FDA's “Century of Science“ forum, the agency hosted a breakout session on combination products with a look at the developmental and regulatory path for drug-eluting stents as case studies. Representatives of the FDA, Boston Scientific (Natick, Massachusetts) and Cordis (Miami Lakes, Florida) gave their perspectives on the ins and outs of coping with a class of products that offers unique technological and regulatory challenges.

The agency's three-day forum, formally titled “A Century of FDA Science: Pioneering the Future of Public Health,“ offered attendees a fresh look at issues old and new facing the agency and the industries it regulates, including seminars on nanotechnology, influenza pandemics and natural disasters. Held at the Washington Convention Center April 18-20, the event drew hundreds from private industry, members of overseas regulatory agencies and representatives of other federal agencies, including the Nuclear Regulatory Commission and the Naval Research Laboratory.

In the first case study presented on combination products, Ron Dadino, vice president for pharmaceutical and packaging development at Cordis, described some of the hurdles his employer had to clear to get the Cypher to market. He started his discussion by pointing out that had a meeting on such a topic convened two years earlier, the room might have included nobody from the Center for Drugs. “The business paradigm for combination products has changed,“ he observed.

Developing a product always starts with the needs of the users and proceeds to an analysis of what the company can offer to answer the need. However, according to Dadino, “that was not necessarily well defined“ at the beginning of the product development cycle for the Cypher, which elutes the cell-cycle inhibitor sirolimus under a license from Wyeth (Madison, New Jersey).

One of the tougher elements to overcome was not just a lack of institutional experience at Cordis in developing a combination product, but the lack of any such experience anywhere. Cordis also had to closely watch for interactions between the stent and the drug, not to mention the drug and the polymer coating around the stent, a series of considerations that created a substantial amount of work.

While Cordis dealt almost exclusively with CDRH at the beginning, “by the time we were ready to file [for the PMA], CDER was getting actively involved,“ Dadino related. CDER posed questions that are typically encountered by drug makers in their CMC (chemistry, manufacturing and controls) reviews despite the fact that Wyeth had previously obtained NDA clearance for sirolimus.

Dadino posed the question of “what kind of language do you use to bridge the gap“ between drug GMP standards and the quality systems requirements embodied in 21 CFR 820.

Kathleen Miller, PhD, of Boston Scientific (Natick, Massachusetts) followed Dadino to the microphone to describe some of the headaches her firm encountered in the development of the Taxus, a competitor of the Cypher that employs paclitaxel, an anticarcinogenic, to suppress restenosis.

Alluding to the problems of language that Cordis encountered in bouncing between the centers, Miller observed that “we discovered how very important it is to develop a cross-functional language“ to explain the testing and validation schemes Boston Scientific employed. This work was made more complicated by the fact that Boston Scientific employed two solvents in making the Taxus.

Miller offered some historical perspective on cardiac stent development. She reminded the audience that the first balloon angioplasty procedure took place less than thirty years ago, in 1977, but that restenosis occurred in 30-50% of those patients. Uncoated metal stents arrived on the scene 10 years later, cutting restenosis to 20%-30%, albeit with an occurrence rate of thrombosis on the order of 1%.

According to an article in the March 6, 2006, edition of Advanced Drug Delivery Reviews by W.L. Hunter, drug-eluting stents are responsible for “reducing the failure rate of BMS (bare-metal stents) by about four-fold,“ an assertion based on “[d]ata from initial large-scale, comparable, U.S. pivotal trials“ for the Taxus and the Cypher. Hunter remarks that the outcomes of these studies seems to “warrant the enthusiasm“ of the devices' sponsors.

Ashley Boam, chief of the Interventional Cardiology Devices branch at CDRH, delivered an explanation of how the FDA climbed into the combination product age and how it sees the comparative developmental characteristics of drugs and devices. She pointed out that the rate of technological change in the device universe has been much faster than in the drug universe, but that devices are sensitive to physician experience in ways that drugs are not.

Boam remarked further that the FDA staffers who came together at the Office of Combination Products had to familiarize themselves with somewhat comparable, but by no means identical, ideas. These included the details of stability vs. those of shelf life and the differences between pre-market approvals and new drug applications. When it came to reviews, the agency found it difficult at times “to make sure that we had the people with the right experience on review teams,“ recruiting from CDRH and CDER to get the needed talent to cobble together an understanding of how a combination product might behave in practice.

On the other hand, FDA was not the only entity that found this an odd bridge to cross. Boam pointed out that “most companies tend not to be combination product companies,“ a fact of life that did not ease the agency's attempts to get to the bottom of the details of an application.

“We are still learning how to handle and communicate these issues,“ Boam admitted.

Boam offered a ray of hope to firms interested in getting into or staying in the drug-eluting stent business, however. Data from existing studies can be leveraged in future submissions if a drug, a device or a polymer has appeared in a previous study, but data on comparative in vitro/in vivo elution for the new configuration are still essential. Trial design for future applications will be somewhat sensitive to the novelty of a drug-eluting stent, she suggested, and the Office of Combination Products may allow a single-arm study in the case of minor product modifications to existing, cleared products. The same opportunity may be available for new diameters and lengths of existing stent products as well as for an application to expand an indicated use.

Another trial design issue involves clinical endpoints, which can be covered somewhat by surrogates to reduce the number of study subjects, but Boam remarked that this technique, while helpful for efficacy evaluation, was of little help in examining safety profiles. Companies should also bear in mind that an endpoint-surrogate correlation may degrade when different technologies are in play.

Boam informed attendees that the total product life cycle paradigm was still in force, including a requirement that stent makers track the first 2,000 subjects for five years, including “all comers,“ that is anyone who has had a stent installed even in an off-label use. As for firms looking to put a new product to the test, she noted that “one of the biggest lessons learned is to start talking to FDA early“ in the development cycle, including each center that might be involved.

On the subject of packaging Taxus for shipment as laid out in drug GMP versus device QS requirements, Miller told Medical Device Daily that “we followed the packaging requirement for Part 820 [device regulations], but we also had to comply with Part 211 requirements.“ She added that Boston Scientific had put more than three lots of the Taxus on stability, but that this applied to multiple sizes/configurations of the stent. Dadino stated that Cordis had employed a similar approach to allocating lots to stability and had used the ICH (International Conference on Harmonization) standards for stability testing and retains.