Blazing New Paths for Product Introductions

Originally Published MDDI September/October 2003

MEDICAL MANUFACTURER OF THE YEAR

In getting its Cypher stent to market, Johnson & Johnson's Cordis subsidiary not only assumed leadership of the emerging combination product market, but also broke new ground in coordinated FDA-CMS approvals.

Erik Swain

A technology that leads to significantly improved patient outcomes compared with existing treatments. A touchstone for the convergence of two industries. An unprecedented success in attaining Medicare coverage. What do these add up to? A potential revolution.

The Cypher sirolimus-eluting coronary stent, manufactured by Cordis Corp. (Miami Lakes, FL), a Johnson & Johnson (New Brunswick, NJ) company, is not only a leap forward in the treatment of cardiovascular disease; it could change the way the device industry pursues approval and reimbursement for combination products. The Cypher stent also was the first product to have Medicare reimbursement coverage in place before receiving FDA approval. For these and other reasons, the company has earned the distinction of being named MD&DI's Medical Manufacturer of the Year.

The company is not being cited for the success of this product alone. It is also being acknowledged because of the expertise and resources it could muster in the design, development, and manufacturing of this and other products. These qualities combine to produce the excellence that characterizes the device industry at its best.

FDA's approval of Cypher on April 24, 2003, was the first marketing approval for a drug-eluting stent in the United States. And drug-eluting stents so far represent the pinnacle of the combination product field, which harnesses the strengths of the device industry and those of the drug or biologics industries to produce technologies that could not be developed by either sector alone.

"This is a very exciting time for Cordis and the device industry in general, because [the Cypher] is opening the door to the future of the industry—the area of combination products," says Michael Drues, PhD, president, Vascular Sciences (North Grafton, MA). "Many device firms take an overly simplistic view of cardiovascular disease. Partly as a result of this [technology], some of them are starting to change their thinking."

The impact in this case is that the incidence of restenosis, or reblockage of the artery, has been reduced significantly, and almost eliminated among low-risk patients. This breakthrough in cardiovascular care has been the talk of the device industry in 2003.

"Rarely are there instances where a new medical technology and/or surgical procedure is introduced that has such a dramatic improvement on the outcome of that type of procedure," says Patrick Driscoll, president, MedMarket Diligence (Foothill Ranch, CA).

J&J's structure, philosophy, and practices enabled it to take advantage of advanced technologies, research methods, regulatory approaches, and manufacturing techniques. Here is the story of how the company's commitment to excellence may help take an entire segment of the industry in a new direction.

Product Development

The road that led to success in 2003 first began in August 1994, when FDA approved Cordis' bare-metal Palmaz-Schatz stent. It was quickly apparent that stenting solved several problems that angioplasty alone could not. But one challenge, restenosis caused by tissue proliferation, remained unmet.

By 1995, J&J began to explore how to address that problem and turned its attention to drug-based solutions, which reflected an organizational strength. J&J consists of more than 100 individual companies, and has a knowledge base that stretches as deeply into the pharmaceutical industry as it does into the device industry.

The questions, then, became what drug to use and how to apply it. By late 1995, it became clear that oral and catheter-based delivery solutions were not going to achieve the desired effectiveness, but localized, time-release delivery of a product via a stent might. And so began a significant collaboration between Cordis and many of J&J's pharmaceutical experts.

"We knew we wanted a drug that would inhibit proliferation and multiplication of cells into the vessel lumen," recalls Brian Firth, MD, PhD, vice president for medical affairs and health economics at Cordis. "It had to be highly efficient in microgram quantities because there is a small amount of surface area on a stent. And it had to be nontoxic to the vessel walls. So we looked for cytostatic properties that would block cells from multiplying rather than kill them." This differed from competitors' efforts, many of which focused on cytotoxic products.

J&J also made a crucial decision early on to focus efforts on drugs already approved or on track for FDA approval, to make any future combination product application simpler. The corporation's pharmaceutical experts screened more than 800 compounds, settling on a handful to test in animals. From those efforts, sirolimus, a drug developed to prevent rejection of kidney transplants, emerged as the clear choice for testing on humans. Among the points in its favor were the fact that it did not cause cell death nor inhibit endothelial growth, which could raise the risk of thrombosis.

Identifying a potentially winning compound was impressive. But J&J matched that feat by quickly obtaining an exclusive licensing agreement with the maker of sirolimus, Wyeth Pharmaceuticals (Collegeville, PA). Again, J&J's pharmaceutical industry expertise paid off, as the corporation could draw on personnel who knew how to negotiate such an agreement.

At the same time, development proceeded on the device side. The choice to use the already-FDA-approved Bx Velocity stent, whose vessel wall uniformity and uniformity of expansion were well suited to drug delivery, was not a tough one. The bigger challenge was finding a polymer to provide controlled release of the drug without causing irritation or inflammation in the body. Cordis approached SurModics Inc. (Eden Prairie, MN), a company with a long history of developing and producing coatings for the medical device and combination product industries.

"The biocompatibility of the coating that we identified was well established," says Jane Nichols, vice president of marketing for SurModics. "The part that took time was testing the combination of drug and coating, and making sure it was meeting the requirements of the elution profile. We had to make sure [the coating] did not interfere with the drug during the manufacturing process. J&J did a good job of looking at available coating technologies, and they understood analytical method development and drug release. We offered extensive experience [with coatings] interfacing between the body and a device."

Manufacturing the finished product presented considerable challenges because drug-related issues, such as maintaining stability and purity and tightly controlling the rate of release, as well as device-related issues, such as ensuring proper sterilization procedures and proper coating application, all had to be heeded. "This caused [Cordis] to go to other manufacturing sites and use expertise from the drug side at J&J," Firth says. "That's one of the benefits of being part of J&J. You don't normally have those kinds of top-line experts in drug manufacturing at a medical device manufacturer. It was a great help to be assisted by experts outside of Cordis."

By December 1999, a little more than four years after drug-eluting stent research began in earnest, the drug, device, and coating had been chosen, and the product was ready to be tested in humans. "That's an extremely short time given that it was a new concept," Firth says.

Clinical Trials

In the world of combination products, the manufacturer is never sure whether a product might be a winner until the first clinical trials in humans are completed. Often it can take a long time to determine the product's effects. What was remarkable about Cypher's trials was how quickly it became apparent that this was a breakthrough treatment, and how well the results sustained themselves over multiple years.

The patient implants, performed in Brazil in December 1999, were part of a study investigating whether the product had any safety issues. But after four months, to everyone's amazement, "the stents looked as if they had been put in the day before," Firth says. Results for those initial 30 patients have remained excellent, with 93.3% event-free survival reported after three years. "The big question was, while they looked good at four months, would they hold up long term?" says Firth. "At three years there was hardly any change from four months." The drug was, in fact, preventing cells from dividing without destroying them, and the delivery system was, in fact, distributing the drug at the proper rate to the correct localized area.

Initial outstanding results led to the first formal clinical trial, RAVEL, conducted with 238 patients in Europe and Latin America beginning in August of 2000. Six-month results, first announced at the European Society of Cardiology meeting in September 2001, were even more stunning. "It showed zero restenosis, zero thrombosis, and zero life loss," Firth recalls. "That compares to a 26% angiographic restenosis rate for bare-metal stents. Nobody had ever seen results like this." After two years, this group still reported no thrombosis and only three of the 238 patients needed any sort of reintervention.

"RAVEL was the first multicenter trial to show no restenosis at all," says Mario Costa, MD, PhD, an interventional cardiologist who is an assistant professor of medicine and director of research at Shands Jacksonville, a University of Florida hospital. Costa has been involved in four Cypher studies. "But what is really impressive is the longevity of the results. These are not just acute results; they are there beyond the first year."

It was then time for the biggest challenge, a U.S. clinical trial with more patients and more difficult cases of cardiovascular disease. Many of the 1058 patients in this trial, called SIRIUS, had diabetes, and most had larger lesions than the RAVEL patients had had. After one year, 20% of bare-metal patients needed a repeat intervention, compared with 4.9% of Cypher patients.

"Given the preliminary data that had already come in from the European trial, I was not especially surprised by the efficacy results, but the safety results exceeded expectations," says Jeffrey W. Moses, MD, a coprincipal investigator of the SIRIUS trial and the chief of interventional cardiology at Lenox Hill Hospital and Cardiovascular Research Foundation (New York City). "The potential for complications like clotting or aneurysms was lower than I anticipated. There was no tax that you paid for the extraordinary net benefit in terms of restenosis."

Also exceptional, Moses says, was the way in which J&J conducted the data handling. "Not only did we have full access to the data, but it was ours to interpret and further analyze," he says. "They let us run with the ball. It's fairly unique in that industry to have that kind of confidence in the science. I respect them a lot for that."

CMS Approval

While J&J left the scientific data in the hands of the doctors and scientists, its regulatory and marketing experts left nothing to chance with FDA, the Center for Medicare & Medicaid Services (CMS), and hospital organizations.

Particularly notable was the early stage at which the firm met with CMS, and the amount of data it presented to bolster its case for reimbursement beyond existing diagnosis-related groups (DRGs), or incremental reimbursement. Meetings with CMS began in September 2000, after data had come back from the initial trial in Brazil. Firth and other Cordis officials sensed that "it seemed like we were onto something pretty big."

It had taken Cordis three years and three months to get incremental reimbursement for the first bare-metal stents, which meant that for the first three years they were on the market, hospitals had to absorb the increased costs of the procedure. This time, Firth recalls, Cordis was determined that would not happen again. The firm "set out to change the paradigm, and get incremental reimbursement for Cypher as close to FDA approval as humanly possible. We were going to show CMS our data every step of the way, because we wanted unique codes." The company met with a number of large private payers during the process as well. This was the savvy thing to do, Driscoll says, because many industry breakthroughs in the past have been "blown out of the water by reimbursement issues."

Not only was J&J quite open with CMS about all of its clinical trial results, it was also able to provide the center with a persuasive economic study. J&J sponsored a study by David J. Cohen, MD, Director of the Economic and Health Outcomes Research Group at Harvard Clinical Research Institute (Boston). The study showed that even though Cypher was going to cost $2000 more than existing stents, the average 12-month cost of using the device would be $16,813, or only $309 more than for bare-metal stents. Owing to a reduced number of reinterventions, Cypher was shown to be a virtually break-even proposition for payers.

J&J also contended that costs related to bypass surgery have decreased 5 to 10% a year over the last five to 10 years as the popularity of stents and angioplasty have grown, and that drug-eluting stents should create even further savings in that area. It also aggressively solicited support from physician groups and professional societies.

CMS took Cohen's and J&J's data, constructed its own economic models, and agreed with Cohen's conclusion. In April 2002, the center gave drug-eluting stents their own code, ICD9. In August 2002, it approved mapping the code to new DRGs (one for acute myocardial infarction applications, the second for other applications) that would pay more than those for bare-metal stents. These developments came before FDA approval, an unprecedented feat in the history of the device industry.

"I give Cordis tremendous accolades for having the foresight and the chutzpah to spend time, effort, and money up front to get reimbursement pathways in place prior to launch," says Drues. "Not only will it make it easier for physicians to get paid for using the product from the beginning, but it also might make it easier for their competitors' products to be [approved for incremental reimbursement] as well. But management had the foresight to realize how much of a benefit this would be to them."

The costs involved were not insignificant, but J&J understood why they had to be incurred. "The company seems to have conducted significant outcomes research as part of its clinical trials. That's an expensive proposition, but one that is essential to support a higher reimbursement rate," says Steve Halasey, editor of MX magazine (Los Angeles), a sister publication to MD&DI that covers financial aspects of the device industry. "Remember that J&J lost the original bare-metal stent market to competitors, mostly because its pricing was too high. It will be interesting to see if the company has learned its lesson. But other companies that have not conducted equivalent outcomes research could be at a disadvantage when it comes to justifying their pricing."

In fact, J&J's approach with Cypher may indicate a new trend in how device companies deal with reimbursement issues. "The companies most successful in being reimbursed for high-profile technologies can thoughtfully align the technical competencies in their companies with their policy and political competencies," says reimbursement expert Ted Mannen, managing director for Aventor (Washington, DC). "To be successful with a product that presents novel or complex issues, you have to have a more holistic set of skills. Increasingly, companies will make it a best practice to start reimbursement planning very early, and to include reimbursement as a discipline on the product team."

J&J and Cordis have not been content to rest on their laurels once CMS reimbursement was attained. They have engaged in a number of practices that attempt to make reimbursement easier for all involved. The company is continuing to provide CMS with data to get the DRGs adjusted upward. It also has enlisted consulting firm Ernst & Young to do a real-world economic study.

Postlaunch, Cordis has been conducting seminars to show hospitals what procedures to follow to attain proper reimbursement. It is also encouraging some hospitals to change their contracts with private payers. Those that get paid on a per diem basis (often rural hospitals) find themselves not getting reimbursed more for working with more-expensive new technology. Cordis has been encouraging such hospitals to go to a "carve-out" approach, which is based on aggregate percentages and is standard in populous or wealthy areas.

FDA Approval

The path to marketing clearance was fraught with its own challenges. The company was very confident in its clinical trial data, and it had a stent, a drug, and a coating that had been seen in FDA applications before. But it was well aware that FDA had not seen a product like Cypher before, and new product classes invite extra scrutiny.

J&J had to demonstrate a higher standard of manufacturing capability than for a regular device. It had to prove that it would manufacture to a standard of purity and quality consistent with that for drugs. It had to show that the sterilization process did not affect the drug. It had to address shelf life and stability issues that usually don't pertain to devices. It had to show that it had tight control of the quantity and rate at which the drug was coming off the device, so that the delivery process would take about 45 days as specified. It had to show that the polymer coating was compatible with the drug, remained inert in the blood vessel, and did not crack or flake off when applied to the stent. These and other factors were new challenges for a device manufacturer. But Cordis received much assistance from the experts on J&J's drug side in preparing the application and addressing the feedback from the agency.

Because the Cypher stent constituted a drug-device combination product, J&J had to work with multiple FDA centers. While CDRH took the lead in reviewing the PMA application, the Center for Drug Evaluation and Research (CDER) gave ample input, and J&J had to provide many answers to both centers. In fact, Firth says, in many respects the application received more scrutiny from CDER. "The FDA drug [center] tends to see drugs as systemic," he says. "But our product had local drug delivery, with very little spillage into the bloodstream. The drug side raised a lot of questions about the systemic effects of the drug. We had to do quite a bit of work to show that because of the local delivery, things should be viewed differently."

The application represented new territory for FDA, which has been restructuring how it handles applications and inspections for combination products. But Firth says the two centers "worked well together," and the agency completed the review in nine and a half months, which is very quick considering the complexity of the application.

So, in April 2003, Cypher became the first drug-eluting stent to gain U.S. market approval. Acceptance has been swift, and a report by JP Morgan estimates that Cypher will capture 42% of all U.S. stent sales in 2003, including 65% in the fourth quarter. That, the report states, could add up to $1.7 billion to J&J's 2003 revenues, and cause a 110% jump in revenues from the Cordis unit.

Implications for Cardiovascular Care

To maintain that kind of market presence, however, J&J will have to demonstrate that its technology is superior to other drug-eluting stents. Competition could come to the U.S. marketplace as soon as 2004.

J&J appears to have already gained a lot of goodwill with physicians. "I'm very excited about this technology," says Costa. "It's a major revolution in the way we treat coronary disease."

Firth says one thing J&J is most proud of is how the technology has improved patients' quality of life. "Patients are benefiting hugely from a device that reduces the need for repeat interventions," he says. "It is not only making the actual surgery easier, and the clinical benefit better, but it is reducing the anxiety for patients and their families that goes with surgery. It is changing what patients are going to expect."

From a more bottom-line standpoint, Driscoll says, the product has "changed the structure of the marketplace by virtue of being able to increase the potential patient population to which this [stent] approach could be applied, and increasing the clinical effectiveness of the approach."

Moses says that one of the most important aspects of the treatment is that it has been shown to work for both low-risk and high-risk groups. "The percentage reduction of restenosis is fairly common across all groups," he says. "There is no question that this will have a huge impact in the field. This could be the platform for many endovascular technologies, as well as technologies that apply to other organs of the body. It may drive various developments in cardiovascular treatment, as well as other treatments such as oncology." Of particular interest will be the results of the SECURE trial, which is being conducted for patients who have no other recourse.

"The SECURE population is the worst-case scenario," says Costa, who is one of its investigators. "These patients have failed conventional treatments, angioplasty, bare-metal stenting, and bypass surgery. This is the only thing we can do for them. We expect that the technology will be effective, as it has been for all populations. The results may not be as positive for these patients, but they should be better than [those of] anything else they have available."

If drug-eluting stents can be shown to offer hope to patients who had almost none, that could spur even more advances as more firms realize the technology's potential. Drues says the technology is a major advance on the way to making cardiovascular treatment less invasive and more technology based.

"Stents open the artery and restore blood flow. Long term, keeping the artery open and preventing the disease from returning has to go beyond just simply a mechanical approach. Drug-eluting stents are that next step," he says. "Beyond that, we could see biologics-coated stents, which would involve gene therapy. Restenosis is actually a term that encompasses any of four completely different biological problems: atherosclerosis, hyperplasia, inflammation, and thrombosis. These are not one problem that can be completely solved with one solution. We may even see multiple drugs and genes being employed. The future of cardiovascular care will be combining devices, drugs, and biologics to take a more intelligent approach."

As J&J continues to make efforts to improve the reimbursement situation, so is it continuing to push for improvements to the product itself. For example, it is trying out a different metal that is more flexible but still as strong— enabling even more effective and precise drug delivery.

No road to progress is completely smooth, however. Cordis became aware of that when FDA received notice of 47 cases of stent thrombosis during the first months of Cypher's use in the United States. Rather than dismissing the problems as outliers or background noise (more than 50,000 patients received Cypher between launch and early July 2003), Cordis was quick to respond to them. On July 8, the company issued a letter to healthcare professionals reminding them of proper procedures. In particular, the letter asked doctors to make sure they have chosen the appropriate stent size, followed the proper patient-selection protocol, and used the correct antiplatelet regimen and the correct stent-deployment technique. FDA supported the company's actions.

"The problem seems concentrated at a few centers, and does not seem related to the device but to problems with operating technique," Firth says. "We and FDA are looking at it closely. Part of the problem may have been that we rolled out the smaller-size stent first and maybe some used it when the bigger size was necessary."

Implications for the Device Industry

Not only has the initial success of Cypher changed the way cardiovascular care is administered, it could be a harbinger of changes for the device industry as a whole.

"For a long time, device companies just took a mechanical approach to solving biological problems, but that may not make a lot of sense in the long run," Drues says. "The concern longer term is that when you get to the point of using a new drug or a gene, the requirements for the device world will change dramatically, in terms of the length of the development cycle and the number of patients that have to be tested. I don't know if the device companies fully appreciate what is involved. When factoring in medication and biology, problems are not as simple as they may seem if you're just looking at it [in terms of] engineering problem-solving skills."

Driscoll agrees, noting that "medical devices don't cure or eliminate disease. They deal effectively with particular symptoms. Of course, people want cures. Those will ultimately be produced by the biotech arena, unless the drug industry co-opts it. We are now on the road to devices and drugs working to provide treatments of symptoms, and eventually to combinations of devices, drugs, and biologics, and to all biotech. The ability for device-drug hybrids to increase efficacy of procedures is very significant. It is one step closer toward curing disease. They can produce a cure effectively by minimizing the long-term risk of disease recurring. The question is, the clinical results can look great, but do they hold up over time? Over the life of the patient? Or will there be unforeseen consequences of having the drug in the body? That has yet to be seen. Pharmacological science can be predicted but it is not as plain as devices. When you make changes to the biological system, repercussions flow through."

Another question is whether device companies that lack the in-house pharmaceutical and biotechnology expertise of J&J can muster the resources and commit the time to develop combination products properly.

"The barriers to entry are quite high," Firth says. "The R&D costs associated with combination products are high, because you have to evaluate drugs and polymers as well as devices." Given that, will the investment community provide any support for small and medium-sized companies that want to get into these areas, or will it remain the exclusive province of the big firms? That remains to be seen.

It also remains to be seen whether such time and expense can be justified when developing products for markets without the economic potential of cardiovascular care.

"The J&J model worked well in this instance, but it might not work so well if the drug maker were unwilling to license its product exclusively without more-direct involvement, if the market potential were smaller, or if the R&D for a potential product were expected to take much longer than the typical device life cycle," says Halasey.

No Room for Complacency

Drug-eluting stents will likely change the cardiovascular care market, as evidenced by the competing products likely to hit the U.S. market starting in 2004. Similarly, combination products will likely change the device industry, as evidenced by the top firms in the field looking at cross-industry partnerships. Given the talent and drive present in the device sector, the biggest innovation of one year won't stand the test of time if it is not followed by more breakthroughs and improvements. You can expect J&J won't rest on its laurels, nor will the rest of the industry.

Photos courtesy of CORDIS CORP.

Erik Swain is the East Coast editor for MD&DI and contributes frequently to the magazine.

Copyright ©2003 Medical Device & Diagnostic Industry