Biotech's Hidden Stepsister

hen Amir Belson, an Israeli pediatric surgeon, came to Stanford University in 1998 for a fellowship in pediatric nephrology, in his pocket he carried a creased piece of paper on which were scrawled, in tiny Hebrew writing, 64 ideas for tools that he thought could help clinicians do their jobs better. One of the entries on Belson's sheet of inventions was a "smart" endoscope that changed its shape continuously as it followed its tip - essentially, a snake-like device that avoids painful bumps against the body's inner walls by generating a map of the tip's path that the rest of the scope follows. "I was not a good endoscopist," he says. "I invented this technology to make colonoscopy much, much easier."

Naysayers told him that such shape-shifting "was against physics" and would never work, but Belson persisted. In 2001, he enrolled in the one-year Biodesign program at Stanford University, which trains inventors in the business side of the medical device industry. The program is directed by Paul Yock, a cardiologist, bioengineer and entrepreneur who developed several foundational cardiovascular devices over the past three decades. During the course of the program, Belson secured $3.5 million in first-round funding from venture capitalists and assembled a team of engineers to launch the start-up Neoguide Systems to develop the technology.

In 2003, the company hired a president to raise a second round of funds. It was a tough time for fundraising: "Sixty-five VC's told us 'No'," Belson recalls. The board of directors put the company "in hibernation," and he took over as president. Along with one of the other two cofounders, Belson started pounding the pavement in search of the $5 million they needed to get the company running again. "I did not have any financial plan, I did not have any team," he says. "I just had an idea."

With persistence, and within three months, he raised $15 million from eight different investors, and in 2006 he raised $25 million more. Neoguide is now supported by nine VC firms. Last year, the company redesigned its endoscopic device to make it usable not just for colonoscopy but to enable endoscopic surgery through any orifice; so far, it has been tested in animals and human cadavers, and Belson soon hopes to test it in living humans. Neoguide is still the company that Belson considers his baby, but more than 200 patents later, he's now launching a handful of other companies, covering areas as diverse as therapeutic hypothermia and peripheral IV catheters.

Belson's story is as close to a classic inventor's tale as tales come, and the field of medical devices, certainly more so than drugs and biologics development, has long been known as one where such success is possible. Venture capital investment in medical technologies has boomed over the last decade, jumping 45% in a single year, from $2.9 billion in 2006 to $4.1 billion in 2007, according to figures from the National Venture Capital Association. (Investment in the biopharma sector grew at a much less dramatic rate during that period: 13%- from $4.6 to $5.2 billion.) Still, some industry experts suggest inventors like him face an uncertain future.

As recently as 10 years ago, an inventor with a great idea could expect to be able to raise tens of millions of dollars to finance a project, and see it tested and commercialized within four to six years. Since then, however, some industry experts contend that both the time and money required to bring a product to market in the United States have, on average, doubled. "Now it's taking 5, 7, 9 years" to bring a product to market, says Ross Jaffe of the venture capital firm Versant Ventures. "Where it used to take $20 to $40 million in capital, now it's taking $70-100 million."

Medical devices are getting increasingly more complex, the Food and Drug Administration requirements for testing them are becoming increasingly more stringent, and, with health care costs growing at an alarming rate, ensuring that new devices will be reimbursed by insurance companies is becoming increasingly difficult. "It certainly used to be much easier to get a device approved than a drug - although that difference is still there, it's not as wide as it used to be," says Janice Hogan, managing partner of Philadelphia-based law firm Hogan and Hartson, whose practice focuses on regulatory issues in medical devices. "I wonder how much this is going to squeeze the entrepreneurial spirit out of the device business."

n the United States, Federal attempts to regulate the safety of medicines reach all the way back to the early 20th century, but the FDA's first serious efforts to regulate devices began only in 1976, with the passing of the Medical Device Amendments to the Food, Drug and Cosmetic Act. The following year, German cardiologist Andreas Grüntzig conducted the first balloon angioplasty, curing angina in his patient and effectively sparking a technological revolution in interventional medicine.

According to David Cassak, managing partner of Windhover Information, medical technology's modern age dawned in the mid-1980s, as these new technologies gained acceptance and other innovations in joint replacement and minimally invasive surgery took hold. In the late 1980s, investors began to take interest, and their cash infusion helped turn companies such as Medtronic, a mom-and-pop shop since 1949, and Boston Scientific, founded in 1979, into significant players in medical technology innovation. (The two companies reported $12.3 billion and $8.4 billion in sales, respectively, for 2007; the industry's No. 1 seller, Johnson & Johnson, reported $21.7 billion in device sales for 2007.) "Really, the willingness of investors to back innovative technology companies helped shape a lot of the current industry," Cassak says. In the late 1990s, though, the investors' pendulum swung the other way as the first crop of companies went public, but didn't quite deliver on revenues. Now, in the last five or six years, the money is again flowing, with investors actually expanding their investment in devices in light of biotech and pharma's pipeline woes.

The rising costs of bringing a medical device to market, however, often limits how far that money can go. Initial public offerings are all but impossible in the current market; the other main option for an exit strategy is acquisition by a larger company, but the handful of big fish in the med-tech pond are looking to minimize their own risk by buying start-ups at more mature stages than they have in the past. That means small companies have to stay afloat longer, which requires more investment. From an investor's perspective, says Jaffe, "that means it's harder to generate the same level of returns," leaving VCs less willing to take risks on an innovative but unproven idea than they might have been in recent years. According to Mir Imran, who runs InCube Labs and InCube Ventures, a med-tech incubator and venture fund, respectively, with the timeframe and cost on the rise, "There will be fewer and fewer [investors] who have the stomach to invest $100 million or $200 million over a 10 year period."

Probably the biggest driver of increasing costs for device manufacturers is an increasingly stringent regulatory climate at the FDA. Starting in the 1990s (see timeline), the agency began to reassess trial standards for devices, moving the process of testing a device closer to that of testing drugs. Many industry observers say devices have also gotten caught up in the agency's increasingly risk-averse climate, engendered in part by several high-profile drug safety snafus.

This creeping conservatism is having an effect. Steve Anderson is president of St. Paul, Minneapolis-based Acorn Cardiovascular, which is developing a mesh heart restraint that provides additional ventricular support to failing hearts. Four years ago, based on extensive discussions with the FDA, Acorn conducted a 300-patient randomized trial - a sizable trial by device standards, and the largest pre-market randomized study ever done for a chest-opening procedure called a sternotomy - to demonstrate the safety and efficacy of its mesh heart cap, approved in Europe since 2000. Although the trial met the primary endpoints which the company had negotiated with the agency, the device was denied approval, Anderson says, because of new concerns raised with the study design. The FDA demanded more confirmatory data, which Acorn is now collecting. "We literally had people on the panel saying the bar had been raised," Anderson said. "I feel like right now, every year, the bar for a small company goes up."

Daniel Schultz, director of the FDA's Center for Devices and Radiological Health, refused to comment on Acorn's approval process, but stresses that with devices getting increasingly complex, the amount of information the agency needs to assess them has simply gone up. He also concedes that the recent history of high-profile adverse events has spurred stronger post-market monitoring. That means devices similar to ones that have already made it to market may now require more testing, "even if we didn't ask your predecessor, because we weren't smart enough to ask [before]," says Schultz. "We're not just going to put blinders on and say we're going to ignore that due to consistency." As for the "shifting goal post" complaint, he says, "it's always something we try to avoid," but if critical new considerations arise while a product is in development, the agency must address them, even if it means asking a company to do more testing. What's more, he adds, the turnover of reviewers during the product's development can also shift the requirements.

To some extent, says Cassak, increased regulatory oversight is natural: As a field grows, regulations surrounding product approval will become stricter and more codified. In fact, adds inventor Imran, those changes may actually be good for the industry, reflecting a maturing level of standards that protects both consumers and investors. Plus, the very notion of what a device is has changed. Traditionally, medical devices were mechanical innovations, most often in the fields of cardiology (stents) or orthopedics (joints). But many recent biological advances that researchers are now trying to develop into therapeutics - such as stem cell or nanotech-based therapies - will require devices to deliver them. That may make things more complicated and more expensive to develop, notes Andrew Farquharson, a colleague of Imran's, but it also means that there's "more fertile ground in medical devices than before."

But for small companies, the regulatory flux can be a real problem. One company contacted for this article estimates that the gradual shift in regulatory requirements has cost them five years in delays and "easily north of $100 million," according to its CEO, who stresses that he doesn't think his company's woes are unique. (Because the product is still not approved, the CEO requested that the company not be named.) Says Hogan, "If you're caught by surprise, it changes the whole economics of how much money you need to raise, and how long it will take."

ut if you build it, will they pay? With skyrocketing healthcare costs in the United States, it is becoming increasingly more difficult to convince the public and private insurers - the biggest of which is the US federal agency, Centers for Medicare and Medicaid Services (CMS) - to cover new procedures. Device companies rarely achieve reimbursement in under a year after FDA approval, says Imran, and often, the time lag is closer to five years.

This further increases the capital companies need to survive, because it extends the amount of time before their product can be widely sold. "CMS has no incentive to approve anything unless there's a huge amount of data," says Bill Starling, a partner of VC firm Synergy Life Science Partners. "As a result, venture dollars are going into later and later stage deals." "I think the venture and angel communities in particular are much more attuned to the fact that the reimbursement pathway in particular has to be well understood and reasonable before putting any money in," says Stanford's Yock. "That is remarkably different than it was ten years ago."

Case in point is a company called Electrical Geodesics Inc (EGI) in Portland, Oregon, which makes high-resolution brain mapping using electroencephalography. When Donald Tucker, a psychologist at the University of Oregon, developed the technology in the early 1990s, "it didn't seem like the new Gatorade," says Ann Bunnenberg, who cofounded the company with Tucker in 1992. They envisioned primarily a research market, which doesn't need FDA oversight, and launched the business with two government Small Business Innovation Research grants worth about $400,000, and a comparable level of private investment. But by about 2002, neurological interventions had become so widespread that EGI suddenly had a new market in clinical diagnosis. That meant revamping their technology, and seeking regulatory approval in the United States and Europe - a process they funded internally.

Unlike Acorn, EGI had no trouble gaining regulatory approval in just a couple years. "There's a fairly well-trodden path for devices like ours within the FDA," says Bunnenberg. But the company is still struggling to get insurance companies to agree to reimburse doctors who use their brain scanning system. Insurers generally set aside a specific amount for treating a certain condition, and assign codes to treatments that have been deemed cost-effective which clinicians can bill with each use. "There's a fascinating Catch-22," says Bunnenberg. "You've got to get enough of your product out onto the market to get the data to make the case for a new code." But without a code, few clinicians can afford to use it. "It didn't preclude getting our product to market, but it's a very serious problem for widespread adoption," she says.

In addition, the industry is fending off panic about recent changes to the patent system proposed both by the patent office and by Congress, the most significant of which would limit the number of amendments that could be made to a patent filing. The biomedical community has almost universally rallied against the changes, but medical devices may have more to fear than biotech and pharma. Unlike a drug, a device is rarely something completely new - more often, it's a new iteration of an old idea, and it's always a work in progress. "You cannot come up with a device in its entirety," says Imran. "You innovate, you build it, you test it, you realize there are deficiencies, you innovate some more," which could require amendments to the initial patents. There's no way to predict from the outset what a patent portfolio should look like, he says.

In addition, some recent landmark cases (see timeline) have raised standards for whether an invention could be deemed obvious, and thus not patentable, which further raises development costs. To ease the risk on inventors, a start-up company today must spend double the money than five years ago on a watertight patent strategy, says John Caldwell, a patent attorney at the firm of Woodcock Washburn in Philadelphia, which specializes in med-tech.

For his part, Imran believes the trick to avoiding the pitfalls of the current climate is to stay away from heavily trodden spaces. "We go after poorly developed areas where patient outcomes are not good - where, in some cases, they are terrible," Imran says. That way, they can keep trial size down, while maintaining the FDA's interest in helping bring the product through approval. The challenges are real, say Cassak, but they are more a reflection of increasing maturity in the industry, rather than a threat to innovation. And through it all, those who are persistent still survive. "It's tough," says Acorn's Anderson, "But we're still here."

1938 Federal Food Drug and Cosmetic Act for the first time gives the FDA jurisdiction over medical devices. The medical device portion of the law was created for the few, simple devices of the time, and by the 1960s, it was clearly outdated. 1958 English orthopedic surgeon John Charnely creates the first widely used low friction implant for total hip replacement, consisting of a metal ball and a polymer socket. 1974 John Insall, an English orthopedic surgeon working in New York, implants the first modern total knee replacement. 1976 Congress enacts the Medical Device Amendments to the Food, Drugs and Cosmetics Act, which establish the first modern regulatory requirements for medical devices. 1977 Andreas Grüntzig, a German radiologist, conducts the first balloon angioplasty procedure. Positive results of the first four such surgeries, which he presents at that year's American Heart Association meeting, lay the foundation for a new generation of medical devices. 1986 Jacques Puel and Ulrich Sigwart insert the first metal stent into a human carotid artery to prevent an artery's reblocking after opening with balloon angioplasty. 1990 Congress enacts the Safe Medical Devices Act, and two years later, the 1992 Medical Device Amendments, which expand the requirements for reporting adverse effects caused by a medical device, and gives the FDA the authority to require device makers to conduct post-marketing studies. • Eli Lilly and Co. vs. Medtronic - The Supreme Court ruled that damages from patent infringement could not be awarded for products that were not commercialized. This allowed companies to infringe on patents during a product's testing phase, allowing the tweaking that makes up the iterative nature of device innovation. 1992 In the wake of safety problems with silicone breast implants, FDA commissioner David Kessler appoints a Committee of Clinical Review to examine device oversight. The group identifies several inadequacies in device trials that Kessler promises to amend, bringing regulation more in line with approval of drugs and biologics. • Launch of the Global Harmonization Task Force, an ongoing international effort to create a single set of regulations for medical technologies around the world. 1994 The FDA approves Johnson & Johnson's Palmaz-Schatz stent (named after its two US developers), clearing the way for widespread use of the technology. Late 1990's A large crop of med tech companies goes public, but shows disappointing revenues; investors pull back from med tech financing. 2002 The passing of the Medical Device User Fee and Modernization Act establishes the Center for Devices and Radiological Health at the FDA, including the Office of Combination Products, to coordinate the regulation of products that combine drugs, biologics and devices. 2003 The FDA approves the first drug-eluting stent, manufactured by Johnson and Johnson. It was the first major combination product to enter the market. Drug-eluting stents are now a $5 billion market. 2005 Boston Scientific acquires Guidant for $26.7 billion - the biggest medical device deal to date. 2007 KSR vs Teleflex - The Supreme Court determined that KSR's combination of an adjustable automotive pedal with an electric sensor was "obvious," and therefore did not infringe on Teleflex's patent. The ruling raised the standard for obviousness in patents - important because devices are generally iterative in nature, building on products or ideas that already exist.

By The Numbers Total market capitalization Med tech: $240 billion Biotech: $365 billion Pharmaceuticals: $1,094 billion Source: Yahoo Finance Top medical device deals 2005: Guidant acquired by Boston Scientific, $26.7 billion 2006: Biomet acquired by a private equity group, $11.3 billion 2007: Cytic and Hologic merge, $6.5 billion Source: Windhover Information, Inc VC investment 1998: Biopharma: $1.6 billion, med-tech $1.2 billion 2007: Biopharma: $5.2 billion, med-tech $4.1 billion Source: the National Venture Capital Association Private companies that have received VC money 707: Medical device sector 808: Biopharma sector Source: Dow Jones

The headline is indeed way out of step, when the chart shows investment has quadrupled in 10 years.

Insurers are not "the axis of evil." Can you make a lightly studied doodad without a clear added value and say, here it is, it's $2000, and if the insurer is doubtful - claim the insurer is a neanderthal and Scrooge. There are two sides to every exchange - whether a house, a car, or a $2000 device or service - and the system would collapse if the insurer was a "stamping machine" churning out any payment at any price for anything. The alternative is some level of review and hesitation which, yes, has its own problems but it's a fantasyland and not thought through if you expect it to disappear.

I was the founder and CTO of Avocet Medical, a medical device start-up producing a hand held meter to prevent strokes by monitoring proper level of oral anticoagulants. We raised $38M in VC financing between 1991 and 2001.

We were helped by a 1995 Government study showing that 50,000 strokes per year could be saved by proper anticoagulant monitoring. Remember that each stroke can potentially devastate both a life and a family.

Although we obtained FDA approval, the FDA changed the rules on us by insisting at the last minute, after all clinical trials were done, that users do additional (and redundant) safety checks each time they used our product. These redundant checks pushed the cost per use from $6 to $21, and made the product uneconomic.

At the same time, Medicare (now CMS) conducted secret meetings and concluded, in a non-transparent manner, that they would not fund this class of device. When challenged by prominent medical associations, CMS entered a 5-year period in which they announced that they had no policy regards to this type of device, but that they would not fund it.

We had to exit the field. Although this hurt our investors and employees, the much bigger damage was probably to the thousands of patients per year that we might have been able to save. These are the silent victims.

Here, the big problem is that a medical device failure is conspicuous, while preventable deaths are not because they are the status quo. In order to avoid these conspicuous failures, regulatory agencies are killing new technology in the cradle.

"Are Devices Dead?" I can't believe how anyone could splash such a ludicrous title across such a prestigious journal. The medical device industry is alive, well, and birthing new baby companies every day. It is probably one of the healthiest industries around, with many members being generously sprinkled within the Fortune 500, with major layoffs (think the auto industry) being uncommon.

Sure, the regulatory bodies are a bit unkind nowadays, but that could change with the political wind. The medical device industry has a long history of presenting mankind with fixes for what were either unmet or poorly treated medical needs. It will continue to do so for a glorious future.

Non-invasive electrotherapy medical devices with
specific waveforms offer a viable alternative
solution to pain relief and treatment is without
any known negative side effects such as those
with drugs. This industry is not dead but does
need more funding. Certain segments within the
industry have grown at a very slow pace due to
the virtual monopoly position they enjoy. Device
prices are too high and with no new competition
thousands of needy patients are unable to obtain
treatment. FDA, I believe, knows these devices are safe for use and work but has been unable to
provide for more suppliers due to politics and clever actions by the three approved suppliers

I applaud the authors here for this great article. There are a lot of factors that go into why medical devices grew so quickly in their earlier years, and much less so in recent years. If I had to pick one issue it is with the FDA's inability to properly "quantify specific risks" with each and every medical device, and that as a result each ends up "boxed" into either Class 2 or Class 3 (PMA) submission and oversight.

Sure, modern science has already addressed many disorders thru medical device use, that one might call "pent up demand." But there are many other disorders, as well as improvements to existing technology, that stand to gain. Surely reduced reimbursement is not helping progress. But between Wall Street, Congress, FDA's failures, and adverse marketplace influences, we don't have a vibrant R&D system today for medical devices. I shed some additional thought in my COMMENT to The Scientist ODA article:

www.the-scientist.com/article/display/55041/