Ekos' Long, Hard Road

After running into just about every problem a small device company can face, Ekos is turning things around by borrowing some tricks from the pharmaceutical industry.

by David Cassak

Few medical device executives understand the clinical trials process--in all of its twists and turns and, more importantly, its strategic implications--like Peter Rule. In the late 1990s, Rule took embolic protection pioneer PercuSurge Inc. to a robust exit, a $225 million buy out by Medtronic Inc. , largely, if not exclusively on the basis of its ground-breaking SAFER trial, which tracked the clinical effectiveness of embolic protection devices in saphenous vein grafts (SVG). [See Deal] (PercuSurge's device had yet to be FDA-approved at the time of the sale.)

SAFER was a large randomized trial that convincingly demonstrated that the use of embolic protection devices in SVG procedures dramatically reduces the incidence of major adverse cardiac events (MACE) in patients. The results of the trial went far to establish the clinical value of a therapy--embolic protection--that many interventionalists felt, before SAFER, was unnecessary and did much to set the stage for a host of embolic protection companies to follow.

Still, for all of its success, the trial itself was, at the time, controversial and, in some respects, unnecessary—few device companies would have used a clinical trial to prove the clinical value of their technology. But Rule remained committed to a rigorous clinical trials process, even when others were telling him he was crazy. In fact, when, at the last minute, the FDA decided that the PercuSurge product could get clearance as a 510(k) device, rather than require the more rigorous PMA approval process, making a scheduled FDA panel meeting moot, Rule insisted that the panel meeting go on. "It was a very important panel, not just for the clinical consensus of the doctors, but because I knew it would set the subsequent competitor [regulatory] pathways," he says. Indeed, what came out of that panel meeting meant that "all potential competitors in vein grafts would be required to do comparison trials to PercuSurge," notes Rule. "My view was, ‘OK, this can be a 510(k), but it has to be a 510(k) with a randomized trial,' " something he calls "one of the more unusual events in the annals of medical devices."

More importantly, SAFER's compelling clinical trials data helped to establish a robust valuation for a company that had yet to sell a single product in the US in a new and somewhat controversial area. Rule insists that it's difficult to say precisely how the SAFER data translated into PercuSurge's $225 million price tag. But, he goes on, "I remember talking to Bill George [Medtronic's CEO at the time] the day after the acquisition was announced, and it was very clear that the SAFER trial was the thing that persuaded Medtronic executives that PercuSurge was an important strategic acquisition."

At the time, Medtronic was busily looking for new technology to fill out the line of Advanced Vascular Engineering (AVE), the innovative stent company Medtronic bought three years earlier. And certainly PercuSurge was only one of a number of plays in interventional cardiology that Medtronic made in the years following the AVE deal. Still, notes Rule, "the thing that really set PercuSurge apart, I think, was the fact that here was a small company that did a de novo randomized trial that matched anything that the major companies had done in terms of impact and power," and would remain so until the drug-eluting stent trials that came later.

By all accounts, PercuSurge's aggressive clinical trials strategy proved a winner. But at the time, other industry executives were critical of the company's approach. One executive at a company that PercuSurge was talking to about a potential acquisition told Rule he was "crazy" to take the risk. "His point was, there was already value on the table for our shareholders; we should take it and move on," says Rule, because if the SAFER trial failed, much if not all of that value would drain away. In fact, the SAFER trial "did work out," Rule goes on. "But it was an extremely difficult decision and a nervous time for us."

But the challenges PercuSurge ran into are nothing compared to Rule's latest company, Seattle-based Ekos Inc., where he serves as chairman and CEO. A stroke therapy company that combines interventional tools, ultrasound, and pharmaceuticals, Ekos has, in its short history, run into just about every problem a medical device start-up can face, from product failures to difficult financings, aborted partnerships to clinical trials that had to be stopped. Ekos isn't quite out of the woods, but it's getting there, and doing so with a strategy that, particularly in its approach to clinical trials, borrows a lot from the pharmaceutical industry.

A Struggling Company

Founded in 1995 in Seattle, Ekos was the brainchild of three executives: Doug Hansmann, a mechanical engineer and talented device developer, having worked at PhysioControl, among other places, Joseph Eichinger, a Seattle entrepreneur, and a Japanese physician, Katsuro Tachibana. The three spent their first several years working on Ekos' technology, signing some licensing deals, and raising the company's first monies from a group of venture capital firms that included, among others, Weiss Peck & and Greer, MedVenture Associates, and Bessemer.

Fast forward to February 2001. Ekos had run through much of its early capital and was struggling on a number of fronts. Annette Campbell-White, a principal at MedVenture and a key figure in Ekos' development, notes that company executives and its backers "completely underestimated the challenge" Ekos represented. "Not only did we have to get three different technologies right—the catheter, the ultrasound, and the box—but we were trying to develop two different applications at the same time: stroke and peripherals," she recalls.

Moreover, Ekos' board had known as early as 1999 that the then-CEO was the wrong person for the job, unable to reach out to investors. "We were in a muddle," she says. "Nothing was working out." Ekos' first catheter had, she goes on, "developed some tantalizing data, but we couldn't prove anything statistically, and the regulatory path wasn't clear." Ekos' investors were ready to give up, Campbell-White continues. "We had to get someone who had a reputation as a money-maker into the business. That was Peter."

Campbell-White, who had been an early investor in both PercuSurge and TheraSense Inc. , a company Rule had helped establish (which has since been bought by Abbott Laboratories Inc. ), called Rule at his home in California to see if he would be interested in working with Ekos. Her pitch, she recalls: "This is the technology of the future." Rule says that at the time he was "vaguely aware" of Ekos, but only in the way that well-connected executives hear about companies in the close-knit medical device start-up community south of San Francisco.

At first glance, the appeal wasn't obvious. Ekos' CEO had left two months earlier, and the company was being run by its VP of R&D it had also recently downsized from around 50 employees to 15. Moreover, while the company had arranged bridge financing to cover the previous 12 months, it was having trouble finding additional venture backing. Campbell-White wasn't asking Rule to take over as CEO, but simply to help the company with some operational issues in order to position it to raise more money.

Rule was intrigued by Ekos' potential, but wary. "I really wasn't interested in being a full-time CEO since I had just sold PercuSurge," he notes. He said he'd work with Ekos for a couple of months—but on his own dime because he didn't want to create any false hopes or a sense of obligation—and reassess at the end of that time.

From the beginning, Rule felt Ekos' technology, which he calls "a platform technology," had promise. Using ultrasound embedded at the tip of a catheter, the company had created two product lines, both designed to enhance the action of lytics in dissolving clots: one for the treatment of ischemic stroke, using a microcatheter, standard in the brain, as a delivery vehicle; the other for the treatment of lower limb occlusions using a lytic delivery sheath, standard in the periphery.

From his days at PercuSurge, Rule had gotten to know several of the country's leading interventional neuroradiologists—as the result of a protected carotid stent research program the company had in place--with whom he could vet Ekos' stroke technology. For the most part, Rule found that physicians who had seen the technology liked it, but hadn't seen enough cases to be convinced.

But those early positive responses masked several problems. For one thing, there were important design problems. With the stroke device, notes Rule, "We thought we could access the brain 80% of the time, but we found that we couldn't access it anywhere near that often." Redesigning the microcatheter so that it worked with a 0.14 thousandths of an inch diameter guidewire—the first-generation design had worked only with a 0.10 thousandths wire—became an important priority, says Rule. And, perhaps more importantly, the company's clinical trials strategy was geared for a medical device and didn't take into account the pharmaceutical component of the product—about which, more later.

On the peripheral side, Rule also had gotten to know many of the opinion leaders in interventional radiology, who routinely treat clots in the periphery with lytics. Here, too, while the feedback was positive, the physicians to whom Rule spoke felt Ekos simply hadn't done enough cases. And, here, too, as well, there were design problems, most notably a size limitation in the initial iteration: with a working length of only 5cm, the catheter was too short to deal effectively with clots in the peripheral vasculature, where some clots can be as long as 50cm. "Physicians had to take what I called a ‘caterpillar approach,' to lysing the clot," says Rule. "And you really can't ask a physician in a busy cath lab to take the time to advance through a 50cm clot with a 5cm device. It's a non-starter."

In addition, in the periphery, as Campbell-White noted, the device's early regulatory path wasn't clear. "It would have been nice to think that the product was a 510(k)," Rule explains. But the fact that the early peripheral work was done under an investigational device exemption (IDE) "clouded the regulatory pathway," making the more difficult PMA path seem likely. Still, for all of its challenges, Rule decided to join Ekos, if not full-time at least on a continuing basis, devoting several days a week to the company. "They had done a couple of studies in both ischemic stroke and peripheral arterial occlusion for leg clots," he recalls. "I didn't know if they would stand up to scrutiny. But if they did, the potential opportunity was enormous. Stroke is the biggest untapped market in interventional medicine, and I thought that if you really had an innovative approach to stroke therapy, something like ultrasound-enhanced drug delivery, you'd have something that could be a sustainable billion-dollar market—something that was more like a drug franchise than a device franchise."

The fact that there was an ancillary opportunity in the periphery only made Ekos' potential that much greater. "We had all of these interesting but tiny slices of data in stroke, enough to suggest that we had something, but not enough focus to suggest that we had proven anything," he says. "And we had the same thing in the periphery, plus some interesting animal studies. Was it enough to convince physicians? No, because a stroke doctor doesn't care what happens in the periphery, for the most part. But the question was, ‘Does this technology deserve to live?' and I thought, ‘Yes.' I just hated the idea that it would die."

A Combination Therapy

Ekos' MicroLysUS device is a micro-catheter based system that dissolves clots in the brain by delivering a lytic agent using ultrasound energy to ensure that the agent is delivered to the clot. (Between an inner and outer lining in the micro-catheter are wires that lead to a ceramic element, about one-third of a centimeter long, that delivers the ultrasound; one key to Ekos' design is to make the element rigid enough to deliver energy and flexible enough to navigate the cerebral vasculature.) "Whenever you deliver a drug through a catheter into a clot, mechanical resistance will cause the drug to go right out the back," explains Rule. With the Ekos device, however, the ultrasound creates what Rule calls "an acoustic gate" to keep the drug in the area around the clot. In addition, the ultrasound loosens the fibrin structure within the clot, allowing the drug to penetrate and act on a greater surface area. The device also uses a thermocoupler to measure catheter temperature on a continuous basis and automatically control power to ensure safety while delivering the ultrasound energy.

Critical to Ekos' approach is the use of lytics, a substance that occurs naturally in the body and can liquefy clots. Rule insists that early clinical studies have established lytics as the standard of care in stroke therapy in the zero-to-three and three-to-six hour windows following the onset of stroke. For all of the advances in stroke treatment, particularly in mechanical approaches, he notes, "I don't think there's anyone today who believes we're going to abandon lytics in the brain."

In fact, Rule argues that lytics are preferable to the strictly mechanical approaches that many companies are taking to stroke because mechanical devices are often "emboli generators," as he calls them, releasing small chunks even as they break up or remove the bulk of the clot. In the brain, such emboli can lead to diminished speech or motor function; in the periphery, it can result in foot amputations. "Mechanical approaches are going to cause chunks," he says. "It just happens."

Even with the Ekos approach, some smaller particles may flow downstream, Rule acknowledges, but the lytic drug flows with them and continues the liquification process. In addition, mechanical approaches that expand or distend the cerebral artery in the process of removing clots also run the risk of rupturing the artery, which almost certainly results in death.

With the rationale for Ekos' technology—if not the technology itself—firmly grounded, Rule's first major task was to raise a Series B round of financing [See Deal]. An innovative clinical approach that builds off established device technology and a well-accepted pharmaceutical regimen would seem an attractive investment opportunity. Rule eventually raised $22 million, but like so much else at Ekos, the financing wasn't easy.

For one thing, notes Rule, in 2001, few investors understood the billion-dollar opportunity that stroke represented—and for those that did, the failure of previous device-based approaches had made them wary of the space. "Part of my pitch was to point out to investors that a lot of the big companies already had identified stroke as a priority and would soon be building franchises in the periphery," he says. "But that was a harder sell then than it would be today. A lot of people looked at me and said, ‘Thanks very much, but no.' "

Even Ekos' existing investors were reluctant to put more money in, recalls Annette Campbell-White. Campbell-White, herself, who has remained committed to Ekos throughout its history, offered to step down from the board and sat out the next round of financing. "I told Peter he should start with some new investors," she recalls. "I wanted to let a whole new set of eyes look at this company."

(Campbell-White would later, in Ekos' Series C financing, do her first cross-fund investment ever in order to co-lead Ekos' third round. Though still rare and sometimes frowned upon by limited partners, cross-fund financings are becoming more common, especially on the device side as venture funds have to support and sustain their companies longer and public investors remain demanding. [In cross-fund investments, VCs invest monies from a later fund in a company that had received an investment from an earlier fund, sometimes making the limited partners of the later fund feel as if they are bailing out the LPs of the earlier fund.] "I could do it if I got the approval of my board, which said it was okay, provided we didn't set the price," she says.)

Ekos' Series B investors included EGS, Morgan Stanley, and Prism Venture Partners, and the financing was accompanied by a shake-up of the board. Campbell-White was supportive of the board shake-up even though it meant that she would have to step down. "My job was to get Peter into Ekos and then get out of his way," she says. "I didn't want to pollute the company with old thinking."

Not everyone on the board agreed, however, though Rule's success at PercuSurge and his wariness in joining Ekos gave him the leverage to make changes. "There was some emotion" among the displaced board members, he acknowledges. "But I was in an unusual position," recalls Rule, who, in effect, let the Board know what he felt needed to be done; if they didn't agree, he'd gladly go back to a life of full-time tennis. "That's a very different position to be in than most CEOs find themselves," he goes on. "They can't just walk away."

A Tale of Two (or Three) Partners

With the Series B financing done, Ekos had to re-charge its technology development efforts so that it could get into clinical trials. Rule's first order of business was to turn to Boston Scientific Corp. for an OEM arrangement to provide off-the-shelf micro-catheters for Ekos to use. "At that point, I didn't want to create my own micro-catheter because I wanted to focus on the ultrasound technology," he explains.

Rule's relationship with Boston Scientific dated from his PercuSurge days, where he had collaborated with BSC on an early protected carotid stent trial. Under the Ekos arrangement, BSC would provide the company with a standard microcatheter, which Ekos would then modify, for up to five years, at a price that was slightly better than the market price for the BSC device. "I didn't want to start designing our clinical trials and then have somebody come in and force us to stop because we were using an adulterated version of their product," he says. "I wanted to get all of that out of the way up front."

In return, Boston Scientific made no investment in Ekos nor did it get any distribution or other rights. But it does get early glimpses of all of Ekos' clinical trials data, for as long as Ekos uses its microcatheter. For BSC, whose Target Therapeutics Inc. is the market leader in mechanical devices treating hemorrhagic stroke, the insight into the ischemic stroke market, the focus of Ekos' technology, is valuable—hemorrhagic stroke has been an early target of many device companies, but it accounts for less than 20% of all strokes. And BSC is somewhat protected from any competitive messiness: they have a right to discontinue the OEM relationship over time if Ekos is acquired by a Target competitor.

Rule calls the relationship with Boston Scientific "special." "It's been mutually beneficial," he says. "They've seen our data and watch our progress on a more intimate basis than any other company." Still, he goes on, it doesn't preclude Ekos from talking to other companies about relationships and collaborations.

But an OEM relationship was only one of the partnerships Ekos needed in 2001. "The other thing I realized was that we had to have a pharma partner," Rule recalls, not so much to source the lytic agent, but rather to get access to the regulatory files and other data that would be a critical underpinning of Ekos' clinical trials. "One of the things that device executives don't realize is that if you're going to put a pharmaceutical company's drug in play [in your clinical trials], you have to have their consent," he says. Indeed, unlike other drug/device plays—such as drug-eluting stents—where device companies look for widely available, sometimes even generic compounds to use with their technology, Ekos's approach would have to be different. A generic strategy "works as long as there's an existing market and you're trying to create an enhancement for your device," Rule notes. "But if you're dealing with a developing market like stroke, you need a specific drug."

Ekos' first pharma partner was Abbott Laboratories, whose Phase III ProAct II trial had done much to establish lytics as an effective stroke therapy in the three-to-six hour treatment window, using pro-urokinase (Prolyse), a thrombolytic, administered intra-arterially directly to the clot. Ekos had negotiated a deal to establish a feasibility trial with Abbott and had even begun the trial when Abbott suddenly in early 2002 decided to cancel production of its drug. (The ProAct trials established the benefits of intra-arterial (IA) thrombolysis but the FDA wanted Abbott to do another trial to definitively establish the safety and efficacy of IA thrombolysis. Abbott decided not to, in part because of the cost of an additional trial, and in part because the company ran into ethical problems when stroke centers refused to use placebo and deny patients IV tPA tPA (tissue plasminogen activator). The ProAct data provided the basis for neuroradiologists to begin administering IA thrombolysis as an off-label stroke therapy.)

As an alternative, Ekos turned to Genentech Inc. , whose tPA drug had been proven to work in the zero-to-three hour window and had been the subject of a very successful National Institutes of Health -sponsored trial called NINDS, enabling worldwide approval of the drug. At the time, tPA was being studied in another major NIH stroke trial, IMS I, whose principal investigators, Tom Tomsick, MD, and Joseph Broderick, MD, both of the University of Cincinnati , would eventually become the principal investigators using Ekos' technology in a new trial.

IMS I was designed to demonstrate that combination IA and intravenous (IV) delivery of lytic was superior to IV delivery of the lytic alone for the treatment of ischemic stroke. The trial starts a patient, who presents to the hospital in the zero-to-three hour window, on IV tPA and then transitions that patient to a cath lab. In the cath lab, an angiogram is performed to determine if occlusive clot remains—it does 70-80% of the time—in which case IA tPA is administered. IMS I was run using traditional microcatheters to deliver the IA tPA and showed improved outcomes relative to historical data for IV tPA alone.

While IMS I was being run, Ekos published the results of its first cases, using its first-generation microcatheters in US and Canadian centers in patients in the three-to-six hour window. The results were encouraging compared to historical controls, notes Rule, but not statistically valid. Nonetheless, the results caught the attention of Tomsick who called Ekos and invited company officials to talk about additional studies.

Tomsick and Broderick proposed a new study, IMS II, with the same inclusion and exclusion criteria, the same centers, the same drug, but using the Ekos ultrasound microcatheter instead of the traditional ones used in IMS I. With NIH funds running out, Ekos offered to help fund the trial. "It was nirvana for us," says Rule.

The Drug/Device Divide

The Abbott retreat was another setback for Ekos, which had had more than its share of disappointments, and another black eye, particularly among industry observers. "Had Abbott let that trial continue," notes Rule, "I believe we'd now have the data that would allow us to run ProAct III at half the projected cost of ProAct II," a savings he pegs at between $10-15 million.

But the loss of Abbott has proven to be a felix culpa for Ekos, because the subsequent shift to Genentech was a watershed in its evolution. As noted, an Abbott collaboration would have meant Ekos' clinical trial strategies would have built off of the ProAct II trial. But with Genentech as a partner, Ekos' trial strategy would build off of the first IMS trial and its ties to the successful NINDS trial with the associated involvement of the NIH. In the process, it would also allow Ekos to establish the clinical value of IA therapy as an adjunct to IV therapy—ironically, something Abbott had pioneered in its ProAct study.

While some executives might have been alarmed by Abbott's sudden pull out, Rule was philosophical. "It wasn't fun," he says. "But that stuff happens." In fact, that kind of thing happens all the time in drug trials, he notes—and it was that awareness that was one of the most valuable things that Rule brought to Ekos. One of the things that appealed most to Ekos' investors in February 2001 was what Rule calls his "pharmaceutical mentality," rare among device executives but a mentality he picked while working with insulin companies at insulin-pump pioneer MiniMed. "Most device executives think about their businesses in terms of a three-to-six month period, maybe one year at the most" notes Rule. "Outside of an annual planning process, it's rare to think in terms of two to three years, but that's exactly how a pharmaceutical executive looks at his business."

That's in part, he goes on, because of the scope of the opportunity. While device executives think in terms of limited markets and incremental technology enhancements, drug company executives "place enormous dollar bets in hopes of getting enormous returns." It's not that drug companies are more conservative, Rule argues, but they tend to be more deliberative, both because they have longer time frames in which to act and because they like to see more data before deciding what to do next. "When you're putting a lot of money into play and it's hard to switch gears, you're not as quick to make decisions" as device executives who can more easily adapt to new conditions, says Rule.

Rule says that the differences between device- and drug-company approaches are easy to explain but difficult to fully grasp until you've experienced them yourself. At MiniMed, he notes, "It soon became obvious that if I were going to work with a pharmaceutical partner, it would take three to five years just to earn their respect, and it wouldn't be clear at the beginning which of several drug companies I was talking with would eventually be the one we worked with, so I had to talk to all of them. That's not a device mentality." Rather, he notes, "in the device world, you choose a partner, whether Boston Scientific, Johnson & Johnson, Medtronic, or some other company, and they remain your partner."

Indeed, as noted, Ekos has already had two alliances with drug companies to find its drug agent, one, with Abbott Laboratories, that didn't work out; one, with Genentech, that did. "And if you had asked me at the beginning of the process about both of those relationships," says Rule, "I would have argued that the Abbott relationship would be most likely to work, because Genentech had been on record saying that stroke had not been a priority to them."

That's another difference between drug and device companies: says Rule, because of their longer time frames, drug companies often work with several companies at once; device companies, pressured to meet short-term performance metrics, rarely do so. Overall, he goes on, "Mentalities are different, timelines are different, and earning their trust takes a lot more time. If you go to a pharmaceutical company and say, ‘I'd like to form a relationship in three months,' they think you're crazy."

Such considerations notwithstanding, Rule soon realized that the drug-industry influence on Ekos would be felt most strongly in the company's evolving clinical trials process, a process that would make PercuSurge's aggressive approach to data look modest by comparison. "In the pharmaceutical world, the clinical trials process is very rigid," he says. For example, when device companies look at probability values in the replication of data, they think a P-value of 0.05, or 95% probability, is excellent. "At that level, everybody salutes, including the FDA," he says. "But on the pharmaceutical side, no one salutes at that level; they look for P-values of 0.01 or better."

More to the point, says Rule, drug companies structure their trials much more precisely than device companies do: a Phase I trial for safety is clearly distinguished from a Phase II feasibility, which, in turn, is quite different from a Phase III efficacy trial, and all have very specific purposes and each builds toward the next trial.

In devices, these various purposes or goals are often confused and rarely connected. "Most device companies are good at Phase I trials and some are good at Phase II studies, but they rarely think in terms of Phase III trials," Rule goes on. "They know how to get from animals to humans and how to validate features and benefits, but they rarely think three or four years out. And that's becoming important because even if the FDA doesn't demand this right now, physicians will," particularly in a clinical space like stroke. Ekos' own experience had underscored these differences. "My feeling was that if Ekos' trials had been more orderly, the company would have driven much more value and would have had no trouble getting funded," Rule says. Indeed, he concludes, it was precisely the haphazard approach to clinical trials "that had inhibited" the company's development.

A Start then a Stop

For Rule, Ekos' clinical trial process was different not just because of the pharmaceutical influence, but because trials in stroke are, subtly but in important ways, different from those done for cardiovascular devices. One reason is the unpredictability and urgency of stroke therapy. Rule notes that at PercuSurge, his director of clinical trials could almost schedule the initial cases in which the device was used. "She'd get a call from a doctor saying that he had a case and she'd fly out to see it," he notes.

But you can't do that with stroke patients, especially those that have to be treated within a three-hour window. "You have to have a very different, more flexible mindset," Rule goes on. "Doctors absolutely understand that and as long as you act ethically, it's okay." Indeed, he notes, "you have to think like a physician and ask yourself, ‘During these trials, am I acting in the best interests of the physician and his patient?' If you ever question that, you have to think about stopping the trial, even if there are nasty consequences [for the company]."

And, in fact, just over six months into IMS II, in July of 2003, Ekos had to stop its clinical trial because its second-generation device wasn't working well enough. "We were going along fine, and then we shot our selves in the foot," says Rule. As noted, Ekos officials had earlier believed that its first-generation catheter would achieve an 80% access rate; in the early IMS II cases, it failed to gain access in its first five cases. The second-generation catheter was better—it gained access in two of five cases—but still wasn't good enough.

The problem was that the device could get into the cerebral vasculature, but not far enough to be placed into the clot on a consistent basis. As frustrated as Tomsick and Broderick were with the Ekos device, Rule insists it was Ekos, not its investigators, who decided to halt the trial. "One of the things I understand about trials is that they have a certain momentum," says Rule. Ekos' PIs were willing to continue, but Rule felt it was best to stop.

"It wasn't fun," says Rule, with a sigh. "But as I told the board, that was why we had re-built the engineering team—we already had a third-generation microcatheter coming." Today, Ekos' third-generation device has achieved a 90% success rate in its trials, and the company has a fourth-generation design (using its own microcatheter rather than Boston Scientific's) almost completed. MedVenture's Annette Campbell-White calls Ekos' early catheter development efforts, the "Goldilocks" effect: "The first catheter was too flexible, the second was too rigid, the third was just right."

Peter Rule says that the period from July to November of 2003 "was only four months, but it was four months of hell for everyone." Ekos' investors, in particular, grew restive. "There was talk about returning money," says Rule, who adds, "No one's talking about that anymore." By November, Ekos' new device was ready. But, notes Rule, "once you put a trial on hold, when you begin again, you have to start all over." That meant going through the institutional review board (IRB) process again. It took until June 2004, for Ekos to get back to 12 enrolling sites. "We had to rebuild credibility with the physicians," says Rule.

A Drug-Like Clinical Trial

Ekos is far from the only device start-up to struggle with a stop-and-start clinical trial process. Rule insists he told his board from the very beginning that the process was fraught with uncertainty. "I know about trials, and the first time you start a human trial with a new device, there's a chance you'll fail," he remembers saying--even PercuSurge's early trials ran into problems. In Ekos' case, no one had ever tried to develop a microcatheter to do what Ekos' would. But the company faced another challenge: because of Ekos' use of lytics to dissolve clots, "physicians were expecting a drug-like performance," says Rule. "They expect something that really works without changes." Indeed, in many ways, Ekos' subsequent clinical trial strategy has resembled much more that of a pharmaceutical than a medical device therapy.

At this year's American Heart Association's stroke meeting held in February, the University of Cincinnati's Tom Tomsick presented the first results on the Ekos device in the IMS II trial, looking just at the experience of patients treated with the third-generation micro-catheter. "That's the way a drug trial works," notes Rule. "You look only at the patients treated with the latest formulation of the product, because if you make changes to the drug, you don't get to include the previous cases."

But device trials typically include all cases, even ones using earlier iterations of the device, notes Rule. Indeed, in its early clinical trial work, not only had Ekos not done enough cases, it had done those few cases incorrectly from a drug perspective. "They were done with different drugs in multiple anatomical environments," he notes. "That's a device mentality. But if you're going to work in the drug/device world, you have to consider the drug constraints. We needed a consistent, relevant clinical trial pathway."

And drug trials use only the latest formulation, argues Rule, who at MiniMed had been on the industry advisory board for the largest NIH-sponsored clinical trial ever, because they tend to be more rigorous in terms of establishing the scientific bona fides of the product being tested. "I knew the rules were different," he notes. But in Ekos' case, he says, that was okay: "If you're going to crack a market like stroke, you want unassailable scientific credibility," particularly in company-sponsored trials. Much as he would with a drug, Ekos' PI therefore looked only at the results of the third-generation Ekos device. And the results, argues Rule, were just about everything Ekos could have hoped for.

Approved for 72 patients, Tomsick reported on 68 patients at 10 sites, though because the trial mimicked the IMS bridging strategy, treating patients both intravenously and intra-arterially, only 69% of the patients received IA therapy. (Another difference between Ekos' trial and that of the typical device trial, the company had to pay for all patients enrolled, at a cost of $10,000 each, whether they were treated with the Ekos device or not. "We had to pay for everything," notes Rule. "That's something drug companies take for granted, but for device companies, it's a new twist. But when you're trying to prove you have a breakthrough in a billion-dollar market, you have to ante up.")

Of those treated with Ekos' third-generation catheter, the device showed a 90% access rate in cases where access could have been achieved, a rate at the high end of Ekos' goal of 80-90% for that device. (The company has a fourth-generation device coming out this year and hopes to achieve a 95% access rate with that one, eliminating one huge barrier for any stroke device: the issue of accessing the clot.) Tomsick did not report on outcomes measures—which hadn't yet been adjudicated, notes Rule—but performance criteria such as re-canalization, temperature, and lysus, all of which were adjudicated, he notes.

And in both re-canalization and temperature correlation with flow restoration, the trial proved the device effective, says Rule. (For example, in 84% of the cases, temperature changes measured by the catheter accurately predicted changes in flow rate as evidenced by angiograms.) But it was the third metric, that of lysus, measured in terms of blood flow through the cerebral vasculature, that, says Ekos, was most important. Comparing results in this trial to those in the first IMS trial, Tomsick found 20 cases that could be referenced. The IMS II results showed that in 70% of those cases, usage of the Ekos catheter resulted in TIMI 3 blood flow in the brain.

A Fair Criticism

Rule calls the IMS II TIMI 3 flow data "very exciting," noting that the IMS I TIMI 3 flow rate was 52%. Ekos still has to wait for the validated outcomes data, but, Rule notes, other studies have suggested a strong correlation between target vessel flow and outcomes. Ekos' preliminary TIMI 3 data, he argues, "really say ‘Wow.' "

Ekos' affinity for drug-like trials and strategies is appropriate because the device incorporates a drug therapy. Rule notes that other, more mechanical device approaches to stroke, while promising, are really only "niche opportunities," because they focus not on dissolving clots, but on extracting them from the vasculature. "We know that congealed clots are a minority event," he notes. "Fresh clots are much larger in number." That's why Ekos is betting that lysus will be a front-line treatment strategy. "The majority of cases are not organized thrombus," he goes on. Moreover, he argues that even where physicians prefer a clot retrieval technology to a pharmacological dissolution approach, those devices will require micro-catheters to gain access. "That means we'll be complementary to them, not competitive," says Rule.

Given the fitful progress of Ekos' clinical trials progress, company officials say the results validate their perseverance, though Rule acknowledges that Ekos still has much to prove. "We know that 20 cases is a very limited cohort," he says. "We accept that criticism; it's fair. But if that 20 becomes 50, then we really have something."

Indeed, if Ekos' preliminary data holds up, Rule argues, "this is a revolution in the treatment of stroke." He calls Ekos' 70% TIMI 3 flow rate "a huge issue," in stroke treatment since other device-oriented approaches—particularly those that focus on extracting the clot rather than dissolving it--often don't break out TIMI 3 numbers. Ekos is currently working with Tomsick and Broderick to find more cases to test the sustainability of the early findings. Says Rule, "That 70% TIMI 3 flow rate is going to look a lot better when the number of cases is 50 rather than 20."

And the funding for that additional clinical trial work will come from the $18.3 million Series C financing that Ekos raised last November from a syndicate of investors that includes NGN Capital and MedVentures—money that should last Ekos through 2006 and its initial commercialization efforts. Rule notes that many device companies, particularly venture-backed start-ups, would conclude that the preliminary data Ekos has developed is sufficient. "A lot of VCs would get to the equivalent of a Phase II feasibility trial and say, ‘That's good enough; let's go with this data and not gather any more' as we are," he says. And Rule acknowledges that more cases will extend Ekos' time lines, making Ekos' trial look more like a drug-company trial. "Are those timelines attractive from a device standpoint? No," he notes. "But from a drug company standpoint, they're not only acceptable, they're fast," because it can take a drug company seven years to go from validated compound to approved product.

Re-Thinking Stroke

Rule defends his strategy of pushing harder on and investing more in Ekos' clinical trials process. Pursuing the differences between drug and device trials, he notes, "Why do drug companies do Phase II trials? In effect, it's to power their Phase III trial—to get enough clinical data and do enough cases that you really nail the results and the outcomes associated with those results." Rule worries that anything less in terms of a clinical trials process could result in "low market traction" once Ekos is ready to bring its device to market. "Doctors want to see the data," he insists. "And when the procedure costs $15,000-20,000 per patient, aren't they justified?"

More to the point, Rule argues that stroke is one of the few device opportunities that can justify an aggressive, expensive investment in clinical trials. "One of the problems device companies have is that there aren't many truly billion-dollar markets," he notes, citing only drug-eluting stents and cardiac resynchronization therapy (CRT) for congestive heart failure as comparable markets. Moreover, stroke is the only one of those that have been largely venture-backed--both DES and CRT are, arguably, the province of large device companies—making the issue that much more challenging for stroke start-ups and their investors.

But for stroke to realize its potential as a billion-dollar market, Ekos also has to count on the structure of stroke therapy "to change dramatically," as Rule puts it. He notes that of the 600,000 cases of ischemic stroke in the US each year, only about 120,000 go to a hospital stroke center of excellence. Even there, only about 20% of the patients wind up being treated; overall, that means only about 4% of ischemic stroke patients get treated at a stroke center of excellence.

For the 480,000 patients who wind up, in the first moments following the onset of a stroke, going to a local community hospital, treatment rates are much lower, and few if any receive aggressive therapy. Ekos is betting that as new approaches to stroke are promulgated and new devices developed, more and more cases will wind up at centers of excellence. "Once that happens, with only a slight increase in the percentage of patients who get treated at a center of excellence--say, to 23%--the market explodes," says Rule.

But why will more care shift to centers of excellence? For one thing, argues Ekos, new treatment approaches will change referral patterns, nearly all of which now steer patients to community hospitals, in large measure because the public and even many health care professionals aren't trained to recognize stroke in the same way that they recognize heart attack, for example. As a result, the public is slow to respond when stroke symptoms occur and EMTs and ER physicians are slow to diagnose stroke in patients. That's why Ekos' clinical trial strategy has focused on building off of the original IMS I bridging protocol—even though the initial treatment window of zero-to-three hours is potentially restrictive since the IMS approach would allow any stroke victim to start on IV tPA in a community hospital before transferring to IA tPA in a center of excellence.

Indeed, so critical is the IMS II treatment approach, with its implicit endorsement of IA as an important adjunct to IV therapy, that Ekos wants to make sure that IMS II is strongly validated. In effect, the Ekos trial strategy is designed not just to prove the efficacy of the company's micro-catheter lysus technology, but also to validate a new approach to the treatment of stroke—because doing so will, company officials hope, go far to making Ekos' device a standard of care in stroke. "The IMS strategy was a stroke of genius," says an appreciative Annette Campbell-White. "The way Peter conducted [the clinical trials process] may be the most creative thing he's ever done."

Putting Big Pharma Behind You

Ekos' device isn't the only thing pushing stroke therapy in new directions. Rule notes that the emergence of biomarkers to make diagnosis of stroke quick and easy will also change the treatment paradigm by identifying stroke victims earlier. "This is a test requiring a one-quarter of a milliliter blood sample," he says. "It takes about 15 minutes and can be done in the ambulance. You won't know if the patient had a hemorrhagic or ischemic stroke, but you'll know with 90% predictability whether he or she had a stroke." Important, too, are other therapeutic and diagnostic advances in stroke. For example, Danish drug giant Novo Nordisk AS 's Factor VIIa, currently approved for hemophilia, has been shown to have promise as a treatment for hemorrhagic stroke in the zero-to-three hour window, the first drug therapy for that type of stroke. Expected to be approved in Europe and to begin Phase III trials in the US this year, the advent of a drug therapy for hemorrhagic stroke will be "a major force on the system," he argues.

So will diagnostic imaging advances such as MRI or color CT to better diagnose ischemic stroke. Rule notes that "the new wave in stroke therapy" focuses on who can benefit from therapy, "instead of the old paradigm that ‘time is brain,' which suggested a limitation in the time window within which ischemic stroke could be treated." Rule notes that, "If we can see the perfusion using MRI or color CI, we can get a better idea of who's going to benefit from therapy and treat regardless of when the patient presented."

The new therapeutic and diagnostic tools will streamline treatment protocols and push patients toward centers of excellence, key to Ekos' ultimate strategy, Rule argues: "We'll now be able to better capture the incidence of ischemic stroke in the 0-3 hour window and to transport patients to centers where they can get Ekos, because clinical studies will show a significant improvement from intra-arterial therapy," he says.

But just as importantly, it brings more muscle into the game. As stroke treatment paradigms change, says Rule, "suddenly, you have some pretty powerful forces trying to change stroke therapy," forces like Big Pharma (Novo Nordisk will be the first promoting new approaches to stroke), diagnostic imaging giants like GE Medical and Phillips, and biomarker companies. "That's how we think one can create a revolution in therapy," he says.

A Play in Peripherals

Stroke is clearly Ekos' home-run play, but it is in peripheral applications that company officials hope to make a quick, early strike. But, just as in stroke, Ekos stumbled in its initial development efforts in peripherals. Anticipating a product launch in December 2002, Ekos "spent a lot of money, put some creative people to work, and at the last minute, found that the product validation was inconsistent," notes Rule. "Some of them worked, some didn't."

To a degree, Ekos could piggyback its peripheral efforts on its stroke program—Rule calls them "similar, though stroke is more challenging in catheter design while the peripheral product is more so in terms of length of ultrasound coverage." Still, with the company in the midst of trying to right its stroke initiative, it made dramatic cuts throughout the organization. Focus and efficiency became paramount, and some Ekos board members recommended eliminating the peripheral program altogether in an effort to preserve cash.

Rule persuaded the Board not to shut down the peripheral program, though it was downsized to seven people. Rule believed that mechanical devices that removed clot had peaked in the periphery and that there would be a significant role for Ekos' technology.

"It's not that I didn't respect mechanical devices," he says. "But by 2001, I'd seen them come and go and thought they'd had their day." But a pure lytic therapy has limitations as well, most notably, that it can take too long to work and sometimes causes complications. Ekos' innovation in the periphery is to combine mechanical and pharmaceutical approaches. What Ekos had was a mechanical device with the potential to eliminate the complications of lytics, says Rule. "And if you could do that, why wouldn't everybody want to use a lytic?"

Still, with the cut backs, Ekos' peripheral effort "went back to an R&D phase program," notes Rule. By mid 2004, the peripheral program had begun to turn around. The device passed the design verification: "That was our first really good bit of news," Rule recalls. "The second came in the third quarter when the device got 510(k) clearance." Within four months, Ekos had enrolled 20 patients at six sites in a clinical trial.

Today, Ekos' peripheral program is "on a roll," says Rule. The company did a live case at the recent Arizona Heart Institute meeting, following the failure of another company's device to clear peripheral clot. While a panel of physicians recommended tackling the clot with a laser, meeting organizer, Ted Dietrich, MD, chose an Ekos catheter and cleared the clot easily. To date, Ekos has done more than 25 cases in the periphery and is currently doing around two cases a week; Rule notes that physician enthusiasm for the Ekos system is so great, the company hopes to get to around four cases a week for its clinical research program within two months and initiate sales soon.

Ekos is also beginning to do peripheral cases using duplex ultrasound Doppler technology to validate the predictive ability of its technology to sense when clot has been lysed in the peripheral vasculature. And, as in stroke, Ekos is counting on a significant shift in treatment therapies to help power the peripheral market—in Ekos' direction, i.e., away from surgery to percutaneous treatment. Rule notes that between 40,000 and 50,000 catheter-directed cases, using lytics to dissolve the clot, are performed in the US each year. That's around 30% of the peripheral occlusions treated; the rest are all treated surgically.

Riding the Wave

Rule argues that the number of peripheral catheter-based cases is ready to explode—in part because the growing incidence of things like Type II diabetes will expand the patient base, but, more importantly, because there will be a dramatic shift from surgery to interventional approaches, with the latter going from around 30% of all cases to just under 60%. "And that's pretty much what we've seen whenever interventional approaches and surgery go head to head," he says. "Generally what you see is that intervention takes two-thirds of all cases, surgery one-third."

Part of the shift to interventional procedures has already been set in motion by the growth of AAA and carotid stenting—vascular surgeons have already demonstrated that, rather than fighting the trend toward percutaneous approaches, they're willing to learn and adopt them. (Mortality rates range between 5-20% for surgical treatment of clots in the periphery; for interventional procedures, they're under 1%.)

But Ekos doesn't just want to ride the wave of more interventional cases, it wants to help the wave along. Rule notes that one thing that has held lytic-based approaches back in peripherals is that lytic resolution of clots has its own disadvantages, including unpredictable and lengthy time to lysus, cost, time to initial recanalization, and complications. He argues that Ekos technology can assist lytic therapy in making progress in all of these areas. Thus, the device's faster time to lysus and temperature management help make the process shorter and more predictable, as well as improving initial recanalization to the point that it's faster than surgery. Rule argues that Ekos can help ensure that the vast majority of clots are cleared in less than six hours—compared to the current 18 hours that lytics often take to dissolve clots—"we can make a lytic-based procedure virtually event-free."

Rule argues for other benefits of the Ekos system in peripherals: because they're converting a surgical procedure to a non-surgical approach, Ekos can help the hospital save money—more than $2,000 per case, says Rule, over and above the cost of the Ekos system. And he claims that Ekos has benefits over other percutaneous approaches, either because it requires less time in the cath lab—increasing the amount of cath-lab time today is, he says, "a non-starter" in promoting adoption—or because the interventional approach requires pre-treatment with lytics, as in the case of some thrombectomy devices. "If you're going to pre-treat with a lytic, we think doctors will soon realize that they should be using Ekos from the start," he says.

Patience is a Virtue

Annette Campbell-White notes that "one of Peter's great contributions to Ekos was to see that peripherals would be our first money-maker. I had always believed in the stroke opportunity, but Peter understood how important peripherals would be. He had a vision for Ekos that turned out to be exactly right."

And if Ekos' opportunity in peripherals will never match the potential that stroke has, Rule insists it doesn't have to. "Ideally in building a device company what you want is a 510(k) product with a revenue line with a blockbuster behind it," he says. "That's what we did at MiniMed and Therasense, and that's what we can do at Ekos."

For all of its promise, Ekos is still far from having proven its case. Indeed, it has raised over $50 million but its most recent post-money valuation was $38 million, and some industry observers argue that anything short of a revolution and Ekos will have a hard time making a return on its investment. But Rule believes that Ekos has only begun to make its case and that the opportunities are enormous. He argues that stroke is one of three blockbuster opportunities in cardio- and vascular medicine today—along with drug-eluting stents and cardiac resynchronization therapy (CRT) for congestive heart failure (CHF)—and the only one of the three that will be driven in part, if not entirely, by small start-ups rather than the big cardiovascular device giants.

That alone makes—or could make—Ekos, for all of its trials and tribulations until now, a different kind of company than most device start-ups, says Rule: "I believe that if we can get into an NIH-sponsored Phase III trial in stroke, we can take Ekos public." He argues that the reason most device companies and their investment bankers insist that you can't go public until you've reached certain commercialization milestones--$40 million in revenues, for example—is that because they face such limited markets, public investors want the security of a proven track record. "There's no such thing as a device IPO today where you can convince the public investor that the revenue line could be $1 billion," says Rule. "It just won't happen."

But Ekos, he insists, is different. Stroke is "a billion-dollar product line," he goes on. "The only question is, Can we execute? We still have a lot of work to do and a lot more cases need to be done. But I truly believe that the entry into a Phase III NIH trial would cause Ekos to be a public company, giving us the capital necessary to assist in building the ischemic stroke market."

Rule argues that ultimately what Ekos represents is "an opportunity to change medicine in a way very few companies ever have the chance to." And not just Ekos: he notes that stroke is now the number-one priority of the NIH and a handful of companies, Ekos among them, have begun to produce clinical data to suggest that the hope for a therapy for stroke isn't unfounded. "We spend $40 billion a year treating stroke in this society, and we really don't have an effective therapy, except in a very small niche," he notes. Rule concedes that Ekos has had "a hard, challenging road"—product failures, clinical trial cancellations, difficult financings, board upheavals, and aborted partnerships. But, he goes on, the company has always had one thing: the promise of an effective therapy to treat stroke. "Without that, I don' think I would have done this," he says. "The challenges would have been too great."

For Annette Campbell-White, who has been with Ekos from the beginning, the fact that Ekos is still standing, let alone beginning to generate promising data, is amazing. "I know better than anyone how close this company was to going out of business," she says. "At so many times, it would have been so easy just to walk away. But our investors stuck with it." That perseverance, for her, is the lesson of Ekos: "It would have been easy to become pessimistic and say, ‘This will never work,'" she goes on. "But in device investing, it's so often the case that you have to be willing to say, ‘OK, it didn't work when we tried this; let's try something else.' You have to have the conviction that you can make something work."

Peter Rule agrees. He insists that Ekos' strategy "has always been to participate in one of the biggest blockbusters in the history of medicine, and I don't think there's anyone who doesn't think stroke has that potential." But, he goes on, even with blockbusters—and, arguably, particularly with products of that potential—you have to have patience: to get the technology right, to get the data to prove it works, and to use that data to convince physicians of the technology's value. "MiniMed had some horrible revenue years in its first five or so years; what VC would ever have funded that company if he had known it would take so long," he recalls. Target Therapeutics, too, took a long time to get traction in the market. "Emerging markets are, by their nature, unpredictable," he says.

That's why peripherals are so important to Ekos—if the company can establish a solid revenue line in PVD, it can wait out its stroke play. "If we can establish our peripheral business," says Rule, "we'll be able to afford more than enough patience to be what we want: a dominant force in ischemic stroke."