The Lucrative Elution
Boston Scientific’s blockbuster medical device–and the novel way it was developed
In 1996, Johnson & Johnson was the undisputed king of bare-metal stents. Stents are the mesh tubes that prevent arterial collapse after balloon angioplasty, the principal treatment for atherosclerosis: A balloon is inserted into an artery to clear away plaque and is removed. Then a stent containing another balloon is inserted into the artery. The balloon is inflated to open the blocked artery and push the stent against the arterial walls; this balloon is then deflated and removed. J&J held a strong patent portfolio that gave it dominance in the U.S. stent market. It also led in Europe, where it faced stiffer competition.
Stents revolutionized the treatment of atherosclerosis in coronary and peripheral arteries, but they did little to address one of the chief problems with balloon angioplasty. In about 30 percent of cases, scar tissue formed around the site of the injury, causing the artery to close again, a setback called restenosis. Stents reduced the restenosis rate slightly, but it was still high.
Today, restenosis in coronary arteries afflicts less than 10 percent of patients thanks to the development of the drug-eluting stent (DES), which slowly releases a drug that inhibits the growth of scar tissue. Drug-eluting stents now command more than 90 percent of the $3 billion U.S. coronary-stent market, according to the Millennium Research Group. DESs have not been approved for peripheral arteries.
Johnson & Johnson pioneered the new generation of stents, but the $50 billion company lost its dominant market position to a partnership between medical-device company Boston Scientific of Natick, MA, and Angiotech Pharmaceuticals of Vancouver, BC. The two companies signed a pact in 1997 that led to the development of Boston Scientific’s Taxus stent, which was introduced in the U.S. in March 2004.
Taxus was arguably the most successful new medical product in history, netting more than $1.4 billion in sales in its first nine months in the U.S. alone. And that’s despite the divergent business models of the companies that created it.
The project dates to 1996, when Bill Hunter, cofounder and chief scientific officer of Angiotech Pharmaceuticals, approached J&J and other stent makers with his own solution to the restenosis problem. His company had obtained a license to produce paclitaxelbetter known by its brand name, Taxolan anticancer drug derived from the Pacific yew tree. Approved as an anticancer agent in 1992, it is marketed by Bristol-Myers Squibb. Stents coated with the drug worked remarkably well in animals, keeping rat arteries clearer than uncoated control stents did. Hunter made the rounds of the stent manufacturers, including J&J, Medtronic, Guidant, Boston Scientific, and Cook.
Angiotech and J&J engaged in discussions, though J&J was already working on a DES that would utilize sirolimus, an immunosuppressant marketed by Wyeth. Hunter talked with the other companies while keeping an eye on Europe, where J&J was also a market leader, but nothing was settled. New stents entered the market often, and “other companies were taking market share from J&J,” says Hunter.
As he pondered his options, Hunter received an unusual offer. Cook and Boston Scientific were longtime competitors, but in order to make a more attractive offer to Angiotech, they had decided to band together, proposing a joint agreement that would allow both to develop paclitaxel-coated stents. The financial terms for both companies would be identical.
“They said, ‘We understand that if you want to deal with one company, it would be the market leader [J&J], but would you be more interested in dealing with the number two and number three companies?’ We thought it would be a phenomenally good idea,” says Hunter, especially in light of the situation in Europe, where, he says, “cardiologists were switching brands almost monthly.
It became very difficult to predict who would have the best stent.” And no matter how good the drug, if it were matched with a lousy stent, it wouldn’t have a chance. “We felt with two horses, we doubled our chances that we would be competitive.” In the summer of 1997, the three companies signed a pact.
Boston Scientific’s Forward Thinker
Using a drug was a novel approach to the problem of restenosis. Stents work on a simple principle: a balloon expands the stent to support the arterial wall and is then deflated and removed. Companies had tinkered with stent designs, trying to prevent restenosis, but with little success. “It was surprising that [only] a few companies were forward-thinking enough to look at biological approaches,” Hunter says.
Most of Boston Scientific’s products were typical low-risk medical devices, but it was open to unusual approaches. After a stint in Pfizer’s orthopedics division, Jim Barry joined Boston Scientific in 1992 to work on using angioplasty balloons to deliver drugs or even to assist in gene therapy. But progress was slow.
Barry found out about paclitaxel when he met Angiotech cofounder and consultant Lindsay Machan at a 1996 radiology meeting in Vancouver. Because restenosis is a problem with a number of implanted medical devices, and Boston Scientific manufactures devices for gastrointestinal, urological, and other uses, Barry realized that paclitaxel was “something we could leverage across all our divisions,” he says. He also liked that the drug was well known and had an extensive clinical history as a cancer treatment. “I thought it might reduce the regulatory burden,” says Barry, who is now Boston Scientific’s vice president of corporate research and advanced technology development.
Soon after he returned from Vancouver, Barry began to campaign for an agreement with Angiotech. Company managers were receptive, he recalls, but leery of the drug-eluting stent’s potentially long development time: they knew they were looking at much more than the 12 to 18 months it usually takes to bring a new medical device to market. Boston Scientific also had little in-house expertise in the clinical-trial and regulatory-approval process for drugs, which is different fromand often more stringent than – that for medical devices. At the same time, it knew it might have a blockbuster on its hands.
And Angiotech was an attractive partner. Earlier in 1996, its paclitaxel-coated stents had been tested for the treatment of patients with cancer of the esophagus. As Angiotech had hoped, the drug prevented new tissue from growing over the top of the stents. Subsequent studies have since shown that paclitaxel works by inhibiting cell migration, markedly slowing the accumulation of scar tissue.
Ultimately, Boston Scientific decided in favor of partnering with Angiotech and entered into its novel alliance with Cook. For his part, Angiotech’s Hunter believes that Boston Scientific and Cook were ideal partners. Companies that trail the market leaders are typically more amenable to taking risks on new technologies.
“When you go to a market leader with a disruptive technology, most of the time they aren’t interested,” says Hunter. “If they already have a dominant position in a lucrative business, taking a risk on something that will cannibalize their own product is not something they’re likely to do. We talked to [the companies that trailed in market share], and those are the kinds of companies that will look up and say, ‘How can I get a piece of that pie?’”
Hunter notes that J&J was unusual in that it led the bare-metal-stent field and yet had a DES program. “They deserve a lot of credit for that,” he says. Angiotech has since signed deals with Broncus Technologies for the use of paclitaxel in an implanted device for emphysema patients and with CABG Medical for coronary artery grafts for use in bypass surgery. Companies interested in Angiotech’s expertise are “the ones trying to claw their way up. It’s not because we don’t want to work with the market leader,” says Hunter with a laugh. “It’s not like we have a predilection for the underdog.”
With their deal signed, Boston Scientific and Cook set to work on DESs. Cook made it to the market first, introducing the paclitaxel-coated V-Flex Plus PTX coronary stent in Europe in September 2002. However, after some disappointing clinical trial results and a failed merger with Guidant, the company withdrew from that market in 2004 to focus instead on making DESs for the femoral artery. “They were the first to launch a paclitaxel-coated coronary stent, and that was a big high…but ultimately they didn’t have the impact on cardiology that we had hoped,” says Hunter.
“It Felt Like a Vise”
It would be Boston Scientific, instead, that would have that impact. But creating the Taxus stent was not easy. A critical technical hurdle was the development of a coating for the stent that would release the paclitaxel slowly for about six months after surgery, after which the scarring response dampens. Angiotech had already tackled the problem, but Boston Scientific preferred to develop its own technology.
That left Angiotech on the sidelines as Boston Scientific methodically designed and developed its new stent. The biggest questions were about drug dosage and rate of release. Barry started with a large dose just to show that the drug-coated stent would indeed work in pig arteries. Satisfied that it had the desired effect, he performed a series of follow-up studies, each time halving the drug dose until he found a minimum effective dose.
The next step was to tinker with the polymer to find the optimum rate of release. Restenosis is essentially a case of the healing process getting carried away, with smooth-muscle cells migrating to the site of the injury with such exuberance that the artery is once again blocked. Paclitaxel inhibits that process, but Barry didn’t want to shut it down completely. He used trial and error to find a polymer with a drug release rate that would allow just enough cell migration to promote healing yet prevent it from going overboard.
Each study took three to six months to complete. Some of them could be done concurrently, but other studies had to be done consecutively, and the development dragged out. Hunter became anxious. He watched as J&J took over the lead in DES development. “At some point, I think we were even or a little ahead of J&J in terms of discovery,” Hunter recalls. But with every experiment Barry’s team performed, its chances of being first to market dwindled. J&J’s Cypher stent was approved in April 2003, and in October of that year, J&J published a study that reported impressive research results: Cypher had a restenosis rate of 5.9 percent, compared with 42.3 percent for bare-metal stents.
As it became clear that J&J would beat them to market, Boston Scientific and Angiotech grew worried. “If you go back to the financial-analyst reports at the time, they thought that J&J having a one-year lead would be hard for Boston Scientific to overcome. [They felt that] doctors would become very comfortable with J&J’s stent,” Hunter recalls.
Barry’s methodical approach drew criticism from market analysts, and even from Hunter. “I remember telling [Boston Scientific CEO] Jim Tobin, ‘We’ve got the polymer, we know the drug release [characteristics], we should go forward….As a scientist, as soon as you have positive animal data, you can’t understand why you’re not treating a patient the next day. But Boston Scientific had to work to get the polymer formulated just right. They had to get uniform drug delivery. There are so many of those steps. Things you think should be solvable in a week end up taking a year.”
Hunter was not alone in his discomfort. Barry felt pressure from both Hunter and Tobin. “I would be sitting in this big conference room, alone with [Tobin]. One day he looked at me and said, ‘Jim, we’re fifth in a three-horse race.’ I had Bill on one side of me and Jim Tobin on the other side. It [felt] like a vise,” Barry says.
In the end, though, both Tobin and Hunter deferred to Barry’s drive to get the release kinetics just right, and in retrospect, it seems it was a wise decision. J&J did indeed ultimately beat Boston Scientific to market, but it had supply problems. This resulted in a shortfall that angered patients and cardiologists.
Moment of Truth
The Taxus stents passed through the early clinical trials, performing well. In 2003, Taxus was launched in Europe, but a final trial dubbed Taxus 4 remained before the FDA would approve the stent for the U.S. market. To much press coverage and fanfare, it was announced that the results of the pivotal Taxus 4 trial would be presented at the Transcatheter Cardiovascular Therapeutics meeting in September 2003.
The night before, Hunter, Tobin, and others from the two companies gathered together. Hunter recalls his trepidation. “I hadn’t seen the data, Boston management hadn’t seen it, and there were lots of rumors of failure. I remember Tobin: he’d look at a slide and laugh and hand it to me, and then he’d look at the next slide and laugh again.”
The trial had been a big success, with results on a par with J&J’s. Restenosis occurred in 7.9 percent of patients receiving the Taxus stent, compared with 26.6 percent of patients receiving bare-metal stents.
Hunter now believes that his angst was unnecessary. “Boston decided this could be a blockbuster, and they wanted to make sure it was bulletproof before they went forward. In the end, you could see the benefit. When the product launched, and there was all this demand, they were able to meet it right away,” Hunter explains. The Taxus stents did suffer some difficulties. Boston Scientific recalled about 85,000 stents from the market in July 2004 because of malfunctions in the catheter system that delivers the stent, but those problems were solved without incurring a significant loss of market share.
Taxus overtook J&J’s Cypher stent quickly. Boston Scientific sold about $42 million worth of Taxus stents in the first 10 selling days alone. A little more than a month after launch, the company estimated the Taxus accounted for 70 percent of DES sales. Today, Taxus sales constitute 30 percent of Boston Scientific’s income. So why did the Taxus so rapidly displace the Cypher? After all, the stents’ clinical trials were approximately equivalent, says Rui Avelar, senior vice president of medical affairs and communications at Angiotech. “Despite the fact that we beat them, I think they’re both very good, comparable stents. You’d be hard pressed to say there’s a medical difference.”
In fact, Boston Scientific may have its competition to thank for the Taxus’s ascendance. J&J made an operational mistake when it introduced the Cypher. It aggressively distributed the results of its trial, generating a great deal of press coverage and buzz among cardiologists. As launch day approached, J&J had to deal with a setback: the FDA told J&J that it couldn’t use stents more than six months old. This forced J&J to discard thousands of stents just a few weeks before the launch. According to Fortune magazine, when the Cypher hit the market, 100,000 patients were waiting for stents, and J&J had 40,000. That made the Cypher vulnerable when Boston Scientific introduced the Taxus.
Partnerships with pharmaceutical companies are now the most common form of collaboration in the medical-device industry, and cross-fertilization between industries is likely to continue. It’s a natural trend, says Hunter. Most medical devices were initially developed decades ago, and years of improvements and fine-tuning have narrowed the gap between competing devices, leaving companies casting about for new ways to distinguish their products from competitors’. “Companies are looking to biology to provide product differentiation,” says Hunter.
That’s something that Angiotech is counting on. After its success with the Taxus stents, the company entered into similar agreements with Broncus, CABG Medical, and other companies. Cook also continues to work with paclitaxel-eluting stents; its new Zilver system is intended for use in the leg. Angiotech sees the possibility of building more business on the kind of arrangement it made with Boston Scientific. Its goal is to become “effectively the pharmaceutical arm of the medtech industry,” says Avelar. Angiotech, and the trend its partnerships represent, are both worth watching.
Become an MIT Technology Review Insider for in-depth analysis and unparalleled perspective.Subscribe today