Accuray: Building the Radiosurgery Market

Building a new product market is never easy, and the chances for a start-up in capital equipment are especially slim. Accuray has the scars to prove it. But after a long haul, it appears to be succeeding in building a market in radiosurgery.

Building a new product market is never easy, and the chances for a start-up in capital equipment are especially slim. Accuray has the scars to prove it. But after a long haul, it appears to be succeeding.

by Stephen Levin

Radiosurgery—the use of a high dose of radiation to ablate a tumor—was first developed to exclusively treat brain tumors.

John Adler, a Stanford neurosurgeon, was convinced that the benefits of radiosurgery—it is non-invasive and highly accurate—could be applied to tumors in other parts of the body and launched Accuray with that goal.

Recognizing the company’s need for professional management, Adler chose his successor, Euan Thomson, who shares Accuray’s founder’s zeal for radiosurgery and the company’s mission to drive adoption for a broad range of clinical applications.

Following a complete reorganization, Thomson has transformed Accuray from an engineering project into a true medical capital equipment company that this year had one of the top device IPOs and installed its 100th system.

Albert Einstein, for all of his achievements, apparently missed his true calling—CEO of a medical device start-up company. One of Einstein’s notable quotes can serve as the mantra for many successful device CEOs. "It’s not that I’m so smart," the physicist said, "it’s just that I stay with problems longer." For no company does that ring truer for than Sunnyvale, CA-based Accuray Inc. and its founder, John Adler, MD, neurosurgeon and radiosurgery pioneer.

Upon being introduced to radiosurgery in 1985, during a fellowship at Stockholm’s Karolinska Institute with the field’s founder, Swedish neurosurgeon Lars Leksell, MD, Adler was immediately convinced that the benefits of this therapy, which involves delivering a high dose of radiation with great precision to ablate a tumor, could be extended beyond brain cancers, the limits of Leksell’s approach, to treat solid tumors elsewhere in the body. Adler came up with the idea of a frameless system that used what he calls "x-ray image-to-image correlation," (using computers to compare, in real time, digital x-rays taken of the patient at the time of treatment with the patient’s prior CT scan), which is the underlying concept behind Accuray’s CyberKnife system. Adler returned from Sweden to his training at the Harvard Medical School and Massachusetts General Hospital committed not only to a career in radiosurgery, but also dedicated to expanding the clinical applications of this treatment. Broadening the clinical scope of radiosurgery at that time meant developing a new system that could be used to treat not only brain tumors but also extra-cranial cancers. Adler accepted a position in 1987 at the Stanford University’s Stanford School of Medicine, where he specialized in neurosurgical targeting and started the process that led to the founding of Accuray five years later.

Adler admits that if he knew then what he knows now about how difficult it would be to found and build Accuray, things might have turned out differently. His lack of business experience was, by his own admission, an issue that contributed to the company’s early problems. But his arrival at Stanford, although primarily a move made for clinical reasons, would serendipitously also help Accuray with some of its business issues, if only by putting him in the midst of a thriving device start-up community. Though it took a while for the venture capital establishment on Sand Hill Road to take a chance on Accuray, by being located in the medical technology hotbed around Palo Alto, the company was able to take advantage of other aspects of the fertile device development environment, including experienced engineers and other personnel, and access to local suppliers of advanced technology, most notably miniature linear accelerators, needed to build the new system.

Chief among the reasons why the venture community showed no interest in Accuray can be summed up in two words: capital equipment. Venture investors historically focus on business models that involve disposable products that result in recurring revenue streams, similar to the razor/razorblade model that was so successful in consumer products. For many VCs, capital equipment amounts to the kiss of death, particularly for a small start-up looking to compete against the large companies like Siemens AG and Varian Medical Systems Inc. that dominate the radiation therapy market, against which radiosurgery is most frequently compared.

Launched in 1990, Accuray started with the odds stacked against it, even more so than most small device companies. Those odds continued to increase during the company’s first decade when, as Adler points out, the company went through just about every travail that could befall a device start-up. (See "Accuray: Tightly Targeting Tumors," IN VIVO, April 2001. (Also see "Accuray: Tightly Targeting Tumors" - In Vivo, 1 Apr, 2001.)) Indeed, so desperate was Accuray’s plight that Adler even found himself forced to take the reins as CEO for several years, despite, he says, not having any business experience or any desire to become an entrepreneur.

To his credit, Adler kept Accuray going during a time when other companies might simply have given up. But even Adler recognized that his greatest value to Accuray was in serving as the company’s clinical ambassador to physicians to grow the radiosurgery market, not in running the company. Accuray’s technology development was much farther ahead of its commercial development: by 1994 the company had a crude but viable first-generation CyberKnife that had been sold to several hospitals.

But as an indication of the company’s lack of business acumen, the early systems were sold for less than they cost to make. Adler’s decision to look for a new CEO and focus his efforts on building the clinical side of Accuray’s business turned out to be a turning point for the company. Adler chose Euan S. Thomson, PhD, to succeed him in 2002 as president and CEO. Thomson, a young executive with a background in radiosurgery who shared Adler's excitement about the future of this market, reorganized Accuray, raising its commercial competence to match its technological capability. The company’s recent performance speaks to the success of the new management team’s efforts: this year Accuray enjoyed one of the most successful recent device IPOs and placed its 100th CyberKnife system.

Accuray has emerged as the clear leader in the radiosurgery market, but continuing this pattern of growth depends upon convincing physicians, primarily surgeons, to change their practice patterns and adopt radiosurgery, instead of traditional open surgical resection, for patients with a variety of solid tumors. The adoption challenge for Accuray is not insignificant. Not only is the company looking to build a market in a specialty—radiosurgery—that is not currently in surgeons’ armamentarium for treating solid tumors, but it is looking to drive adoption among physician customers who are historically conservative adopters of any type of new technology. Indeed, surgeons can be particularly loathe to embrace any kind of robotic technology, such as that employed by the CyberKnife, that does not rely on the tactile skills that surgeons traditionally spend their entire careers developing.

Go Frameless

John Adler was introduced to radiosurgery during his work with Lars Leksell. Leksell developed the first radiosurgery system, the Gamma Knife, made by Sweden’s Elekta AB , a company founded by his son, based upon the invention of the stereotactic frame. (See "Elekta: Growing From the Head Down," IN VIVO, May 2006. (Also see "Elekta: Growing from the Head Down" - In Vivo, 1 May, 2006.)) When Lars Leksell was doing his research in the 1940s, he of course didn’t have the benefit of modern imaging technology. Instead, physicians relied on brain atlases to precisely locate critical structures. But to pinpoint treatment locations in the brain, physicians needed a fixed frame of reference, one that required the patient’s head to remain motionless during the procedure.

Leksell’s answer was the stereotactic frame, a rigid metal frame that is bolted to the patient’s skull, which remains the basis for the application of the Gamma Knife since the system’s first prototype was introduced in 1951. Although the Gamma Knife became commercially available in 1968, it wasn’t until the 1980s, when modern imaging technology enabled neurosurgeons to precisely locate critical brain structures, that radiosurgery began to gain acceptance as a treatment for certain brain tumors; in fact, as a practicing neurosurgeon, John Adler was among the earliest adopters of this new technology. (The first Gamma Knife in the US was installed in 1987.)

The use of radiation has been increasing in recent years as a treatment for various types of cancer, although surgery remains the primary therapy for eradicating solid tumors. Approximately 90% of cancer patients have solid tumors, such as brain, breast, prostate, and lung cancers, as opposed to blood-borne cancers like leukemia. Despite advances in prevention and treatment, the number of patients with solid cancers in the US continues to increase, in part due to an aging population and in part to improved diagnosis, particularly of breast and prostate cancer. The good news is that better diagnosis is producing higher cure rates.

Roughly half of all localized cancers are treated with some form of radiation, generally in the form of radiation therapy as opposed to radiosurgery. (Certain solid tumors do not respond well to radiation treatment, so-called radio-resistant cancers.) The typical treatment path for a cancer patient with a solid tumor calls for the surgeon to attempt to remove as much of the tumor as possible through an open resection procedure. In addition to the risks that accompany any surgical procedure, such as the pain and trauma associated with an invasive procedure, infection risk, hospital stay, and postoperative recovery time, tumor resections present additional obstacles. The most obvious arises when the tumor is located close to a critical organ, nerve, or artery, or when the tumor is very large; both may make surgery a risky option. Further, whereas endoscopic and laparoscopic techniques have enabled surgeons to perform tumor resections less invasively, these techniques are primarily designed for procedures within body cavities and thus are generally not well suited for the removal of solid tumors, which are often located within organs themselves.

The challenges facing traditional tumor resection has resulted in physicians and entrepreneurs exploring other potential treatment options employing less invasive approaches, such as radiofrequency ablation and cryotherapy. (See, e.g., "Rita Medical: Can a Device Pure Play Compete in Oncology?" IN VIVO, February 2006. (Also see "Rita Medical: Can a Device Pure Play Compete in Oncology?" - In Vivo, 1 Feb, 2006.)) Because these energy sources are not specifically designed to kill only cancer cells, they can present a risk when the tumor is located adjacent to vital structures or blood vessels, making them a less attractive option for certain patients.

In contrast, chemotherapy uses drugs that are designed to kill only cancer cells, either systemically or regionally (delivered directly to the tumor). However, chemotherapy is almost never the primary therapy used to eradicate solid tumors. More often, chemo is used in conjunction with traditional surgery as a means of eliminating cancer cells that have been missed or were too widely dispersed to be surgically removed.

The appeal of radiation, in part, is that it generates fewer complications and lower treatment costs than surgery, and with certain cancers, it is showing comparable cure rates. This has resulted in radiation’s role in cancer treatment increasing recently. Radiation has a long history of being used as a cancer treatment, either postsurgically or to alleviate certain symptoms for inoperable patients. More recently, radiation therapy is also being used as a first-line therapy to treat early-stage curable cancers such as prostate and breast.

Radiation therapy is generally administered using a large linear accelerator (linac) mounted on a rotating arm that travels around the patient delivering radiation. Euan Thomson points out that although this process delivers radiation to the tumor, it also irradiates a large amount of the surrounding healthy tissue. "Normal tissue can recover from small doses of radiation, whereas tumor cells can’t," Thomson notes. But to allow for that recovery, the radiation must be delivered in lesser doses with time in between treatments for the noncancerous cells to heal. As a result, he explains, radiation therapy is typically administered in a series of relatively small doses given daily over a period of several weeks, a process called fractionation, with a typical course of therapy running 30 to 40 fractions administered over the course of six or seven weeks.

The role of radiation therapy is usually to supplement surgery, the purpose of the radiation being to ensure that no tumor cells have been left behind in the tumor bed after the surgery is complete. Surgeons typically refer a patient to a radiation oncologist following surgery to begin the therapy.

Advances in radiation therapy and imaging technology have also enabled these treatments to be delivered with greater precision. Beginning with the introduction of the CT scanner in 1972, and continuing with the development of MRI and PET scanners, along with the linking of CT scanners to advanced computer technology, radiation oncologists now have tools that significantly improve their ability to plan individualized patient treatments. Further improvements in radiation dose calculation techniques coupled with advances in treatment delivery equipment made possible what was called integrated conformal therapy in the early 1980s, which led to intensity modulated radiation therapy (IMRT) in the mid-1990s, and more recently the use of image-guided radiation therapy (IGRT). Euan Thomson notes that each of these improvements enabled oncologists to more precisely conform the radiation beams to the shape of the tumor, and thereby begin to reduce damage to adjacent healthy tissue and critical structures. Radiation therapy is dominated by large companies such as Siemens, Varian Medical Systems, and Elekta, with TomoTherapy Inc. being one of the rare, recent start-ups in this space. Despite technology improvements, however, there have been few accompanying changes in clinical practice. Referral patterns remain essentially unchanged, and radiation therapy is still delivered in 30 to 40 fractions.

Radiosurgery essentially looks to combine the best elements of surgery and radiation therapy. By using an intense dose of high-energy radiation, radiosurgery looks to destroy a tumor in a small number of treatments, rather than with surgery followed by weeks of radiation therapy. Because of the high dosage of radiation being administered, radiosurgery must be much more precise than other radiation treatments, or one runs the risk of severely damaging surrounding healthy tissue and critical structures. Targeting accuracy for radiosurgery must be below one millimeter. By contrast, conventional radiation therapy has targeting accuracies that are typically on the order of one centimeter.

If accuracy is critical in radiosurgery, one of the drawbacks of traditional radiosurgery instruments has been their lack of image guidance technology to ensure greater precision. This has necessitated the use of the stereotactic metal frame screwed into the patient’s skull to accurately target the tumor and immobilize the patient’s head during treatment to ensure precise delivery of the radiation. Attaching the frame is actually the only invasive part of the procedure; however, adding to the patient’s discomfort is the fact that the frame must remain in place both while the surgeons are developing a computerized treatment plan and during the actual treatment. The need for a frame has thus limited traditional radiosurgery to intracranial tumors because using a frame elsewhere in the body is problematic. The result is that, until Accuray introduced the CyberKnife, radiosurgery was only an option for a small group—around 10%--of solid tumor patients.

For those patients for whom radiosurgery is a viable treatment option, John Adler explains that this therapy can destroy a tumor in a single 30- to 90-minute procedure. The entire treatment process, including attaching the frame and treatment planning, takes four to six hours and can be performed in an outpatient setting.

Adler concedes that IMRT and IGRT, using advanced computer technology, improve treatment plans and attack the tumor with multiple beams of radiation from different angles that do a better job than traditional radiation of conforming more closely to the shape of the tumor. But he argues, "What has not changed is the inability of these systems to know exactly where the tumor is in space." These systems still rely on what Adler calls "relatively primitive methods such as laser triangulation." If, however, "the patient moves, breathes, or hiccoughs," he goes on, "there’s no ability to calculate that movement and determine the exact location of the tumor at that precise moment."

John Adler’s original vision was of a technology that could address the shortcomings of both traditional radiosurgery and radiation therapy, while maintaining the advantages of both treatment modalities, thereby enabling patients to have a viable alternative to surgical resection. Specifically, by using a frameless design, Accuray expands the potential patient population that can be treated with radiosurgery to include those with most types of solid tumors. And by combining robotic image-guidance technology with x-ray image-to-image correlation, Accuray believes the CyberKnife provides the precise real-time targeting that radiosurgery requires. According to Adler, the overall success rate of radiosurgery is approximately 85%, comparable to that of surgical resection, but with the added benefits of less pain, no infection risk, shorter hospital stays (because most are outpatient procedures), and no use of general anesthesia, resulting in a cost that is up to 65% lower than open surgery.

A Reluctant Entrepreneur

When John Adler accepted a position at Stanford in 1987, it was with the goal of building a frameless radiosurgical device based on his concept of x-ray image to CT image correlation. But Adler had little other than that goal and lacked not only the business experience but also the technological background in the areas required for such a project. He was a pioneer in the emerging specialty of radiosurgery and ultimately, the radiosurgery program that he built and continues to direct at Stanford proved critical in the eventual clinical validation of Accuray’s technology, but it still left him ill-prepared for the commercial venture he was about to launch.

Indeed, Adler’s early efforts were focused on developing a product, not building a company. Only after being repeatedly turned down in his efforts to secure grants from the NIH and other sources (including industrial partners who rejected the project as requiring too much basic research) did he accept the suggestion of a retired venture capitalist, who had come upon Adler’s idea while looking through rejected federal grant applications that had been submitted by Stanford faculty, and who advised Adler that this could be the basis for a medical device company.

However, Accuray encountered more than its share of setbacks, mostly around fundraising; setbacks that would last for the better part of a decade. The "capital equipment" label alone was enough to discourage the venture capital community, which had not done well with previous such investments, and which by the early 1990s was focused on business models tied to disposable/single-use devices as the interventional cardiology market was beginning to develop, fueled by the growth of angioplasty (pre-stenting).

After striking out with venture investors, Adler went the angel financing route. In 1990, however, there were few, if any, of the organized angel investor groups that are now popular; Accuray’s angels were truly friends, family, and colleagues, which for Adler meant primarily fellow neurosurgeons. Starting with a goal of raising $2 million from physicians who would invest $100,000 each, Adler’s efforts ran out of steam at around $800,000.

However, several physicians who declined to invest personally were willing to try to convince their hospitals to buy the system (then called the Neurotron 1000) if the company would build one for them. That then became the strategy for funding the company’s operations. With that incentive, Accuray targeted several hospitals, including Stanford, as the first prospective customers for this frameless radiosurgical device with a sales pitch based only on a set of concept drawings. Nevertheless, Stanford agreed to buy the first CyberKnife.

John Adler acknowledges that it sounds crazy, given today’s hospital financial environment, to think that major academic medical centers would consider buying a yet-to-be-created multimillion-dollar piece of capital equipment from an unknown device company with no track record and no credible financing. "But 1991-92 was a very different time," he points out. Accuray was at such an early stage then that it wasn’t until 1992 that the company actually had its own office space, and it didn’t even have the money for a radiation-shielded facility in which to test a completed system. Instead, component parts, which had only undergone preliminary testing, were assembled for the first time at Stanford, where the first patient was treated on June 8, 1994. Unfortunately, the patient did not survive long enough for a follow-up MRI to determine the success of the CyberKnife procedure.

The complexity and unreliability of the early system meant that it was only used sparingly—Adler estimates that Stanford treated only about one patient per month initially, hardly an efficient use of such an expensive piece of equipment, which some at Stanford took to calling "Adler’s folly." Indeed, radiosurgery, particularly the Gamma Knife, became a symbol of the high-priced, rapidly increasing costs of medical technology during the health care debate that was raging during the first term of the Clinton administration; a famous front-page article in The Wall Street Journal from 1994 singled out this $3.5 million system that hospitals only used a few hours per week as an example of unnecessary, costly technologies.

Still, despite all of the concerns regarding the company, the device, and the cost, radiosurgery was gaining acceptance among neurosurgeons throughout the early to mid-1990s, and Accuray began to see increased interest from hospitals in the CyberKnife. Driven largely by neurosurgeon demand, by 1994, Accuray had five sales on the books. But there was a problem with Accuray’s efforts to finance the company using this build-as-you-go plan. "The trouble was that we were selling systems for anywhere from $1.3 to $1.6 million, when they were costing us $2.2 million to make, and we didn’t have the money to cover the shortfall," Adler recalls. (The current list price of the CyberKnife is $4 million, plus the institution has to build a concrete room to house the unit, which can cost anywhere from $500,000 to $1 million.) Making matters worse, the company was not immediately aware that this strategy was producing such a shortfall.

Back from the Brink

By late 1994, Accuray was in desperate financial straits and started looking for an acquirer. Negotiations with Picker International Inc. progressed to the point where Accuray received a letter of intent from Picker to buy the company. However, the deal fell apart right before Christmas because Picker executives were concerned it would jeopardize their relationship with radiation therapy leader Varian for the sale of CT simulators.

By January 1995, Accuray was out of cash and forced to downsize considerably, cutting its workforce by two-thirds, down to 13 employees. Over the next three years, the company fought to stay afloat financed by a bridge loan and income from the sale of four additional CyberKnife systems. Unfortunately, the limited resources proved insufficient to meet customers’ needs, and John Adler acknowledges that the system performed poorly. "There just wasn’t enough money to finish much needed engineering work," he says. Small numbers of patients were treated successfully under an IDE from the FDA, and incremental improvements enabled Adler to perform the first image-guided spine radiosurgery procedure at Stanford.

The solutions to the company’s immediate financial issues came unexpectedly from Asia. First, in 1996, Accuray’s Japanese distributor, Meditec Corp., the wholly owned health care subsidiary of the Marubeni Corp., a major trading firm, obtained approval to sell the CyberKnife in Japan. "This was critical to the company’s survival," Adler notes, "because our distributor was able to put more substantial sales, marketing, and service resources behind the product in Japan than we could dream of doing in the US." Meditec’s sales effort sparked enthusiasm among Japanese physicians, generating revenue that was critical in Accuray’s efforts to rebuild its reputation in the US.

Shortly thereafter, in 1997, James Doty, MD, a neurosurgeon, philanthropist, and friend of Adler’s who was a strong believer in Accuray’s technology and an early investor, stepped in to run and help fund the company. In 1998, Doty introduced Accuray officials to venture capitalists from Pacific Republic Capital, a venture firm based in Newport Beach, CA, but funded out of Taiwan. These investors were impressed with the CyberKnife and provided an initial loan. A year later, they followed with a $9 million equity investment in the company, along with another $3 million raised from angel investors in Asia, mostly surgeons. At the same time, sales in Japan were increasing, generating a consistent source of revenue. "For the first time in five years, we were able to make a meaningful investment in product development and service," Adler recalls.

The infusion of additional resources enabled Accuray to refine the CyberKnife and address some customers’ concerns, for what clearly was proving to be an effective radiosurgery system. In 1999, the CyberKnife received FDA 510(k) clearance to treat tumors in the head and base of the skull, and in 2001, Accuray received clearance for tumors anywhere in the body, fulfilling John Adler’s original vision for the company.

Handing Over the Reins

Product development at Accuray continued, however, to be a rear-guard action. The lack of any significant US sales deprived the company of its major potential revenue source, and while Asian sales revenue enabled the company to achieve the important milestone of gaining FDA clearance, soon after, Accuray was once again strapped for cash. Jim Doty stepped down as CEO, having overseen a period that, while still tumultuous, resulted in some of the company’s most significant accomplishments, most notably FDA clearance. According to John Adler, "While the operational progress of Accuray during the two years of Jim’s tenure was relatively modest, the company still experienced the most significant momentum in its history."

At that point, the management of Accuray underwent another unexpected twist: in September 1999, John Adler took a leave of absence from Stanford and assumed the role of chairman and CEO. When Adler took over, the company had little cash left, no serious prospects for new US sales, and the company was still receiving complaints from its few customers that the system, although promising, remained unreliable and therefore was underutilized. Further, the FDA clearance covered the next-generation CyberKnife, a product that had yet to be sold or shipped, so treatments could only be performed under the original IDE, which created a paperwork nightmare for users.

Adler admits that employee reaction to his taking over the company’s top spot was "frosty at best," and that morale was low. "The general feeling was that, founder or not, another neurosurgeon had no business running a company with 60-plus employees," he says. "In their eyes, a professional CEO was long overdue."

John Adler did not necessarily disagree, but in his mind, the company’s business was unusually complex, and although it was critical to find the right outside person to lead the company, that would be a long and difficult process. Adler did not believe that was the right time in the company’s development to try to attract a talented new senior executive. "What capable CEO would climb aboard our sinking ship?" Adler recalls thinking. Although he knew he lacked management experience, Adler also feared an exodus of talent from which Accuray might not recover, and says the fact that he knew the technology and market better than anyone else made him the logical choice to head the company until it was more stable and could attract an experienced executive.

Indeed, while Accuray was facing significant challenges, Adler believed, the company’s outlook remained promising. "Most of our customers remained believers in the CyberKnife concept, even if they were disillusioned by the company’s execution," he says. And the product continued to be well-received in Japan, where procedures were being successfully performed and additional sales were in the pipeline.

Reviving Accuray’s US sales proved harder than Adler expected. The company started with its first customer—Stanford. After lengthy negotiations, Stanford Hospital agreed to buy an upgraded, FDA-approved CyberKnife but only at a heavily discounted price. Adler says the discount was necessary to restart US sales, but adds that this would be the last time the company would use this sales strategy.

Accuray also went back to Georgetown University Hospital, which had signed on early to buy a CyberKnife, but rescinded the contract and asked for a refund of its money when the company failed to deliver the system on time. Accuray returned the money, although at the time, the company was hard-pressed to do so. John Adler later found out that returning the down payment preserved Accuray’s reputation with Georgetown, so that when the company re-approached the hospital (with the support of the chief of neurosurgery, who had been a resident with Adler), the administration agreed to buy a CyberKnife. "The Stanford and Georgetown sales were important because they were clear signs that Accuray was finally back in the US market," Adler says.

The resumption of US sales not only marked an important milestone for the company, but it also was significant for John Adler because it was a sign that Accuray had reached a point where it could attract an experienced business executive to head the company. This would allow Adler to do what he does best: serve as a clinical ambassador for radiosurgery and the CyberKnife.

In 2001, Adler advised Accuray’s board of directors of his intention to step down. While the board hired an executive search firm to find his successor, Adler already had someone in mind: Euan Thomson. The board agreed and Thomson became Accuray’s new president and CEO in March 2002.

Adler and Thomson were only casual acquaintances, but they have certain key things in common. Both share a longstanding interest in radiosurgery, with Thomson having received his PhD in physics from the University of London with a project based on stereotactic brain radiotherapy. Thomson also has a clinical background, having worked at several UK hospitals, where he conducted research into high-precision radiation oncology treatments. And like Adler, Thomson also initially had no interest in a business career, although he eventually decided that that is where his skills were best suited.

Thomson’s introduction to the commercial side of health care came through a consulting business that he started when radiation therapy companies such as Varian and Radionics sought his expertise. He found he enjoyed working with industry, and when Photoelectron, a Boston-based manufacturer of miniature x-ray units (that has since gone out of business) for which Thomson had done consulting work for several years, approached him about joining the company in 1998, he accepted the post. "I reached a point in my life where I decided I had to choose one of the three areas I was then working in," he says. "What attracted me into industry full-time was that, where working in a single academic environment, you could be a small component of making a difference, a good company with an active R&D program can actually be working with many of the top academic centers in the world to develop new ideas into products that are being utilized and have a broader impact."

Photoelectron recruited Thomson initially to set up and head the company’s European operations. However, within a few months, he was asked to run the entire company from its Boston headquarters. Photoelectron was a member of the ill-fated group of device companies that went public prematurely in the mid-1990s, only to crash soon after their IPOs as their disappointing performances caused public investors to flee the industry. "I went into this with my eyes open, knowing they were a struggling company that had probably gone public too early, so that when I took over, they only had about three months worth of cash left," he says.

Thomson ran Photoelectron for three and a half years, directing a major restructuring of the company that included significant downsizing, and he spent much of his time looking for financing, raising a total of around $30 million. He also did deals with both Carl Zeiss AG and Cordis Corp. , a Johnson & Johnson operating company, with the former establishing a strategic relationship under which Zeiss took over sales and marketing of Photoelectron’s primary system, allowing Thomson to restructure the company as more of an OEM-type supplier. [See Deal]

Although Photoelectron was still struggling, Euan Thomson says he was not looking to change jobs when at the 2001 ASTRO (American Society for Therapeutic Radiology and Oncology) meeting, John Adler approached him about succeeding him at Accuray. Asked about his initial reaction to the job offer, Thomson recalls, "I didn’t know John very well, but I knew of Accuray and loved the technology and the field. It seemed that everything I’d done previously in my career came together in that position, so it was almost inevitable that I should be there running the company. And once I’d made the decision to accept the job, I’ve never regretted it."

Thomson was the youngest of all the candidates for the Accuray job and was competing against more experienced, traditional medical device executives. What convinced John Adler that Thomson was the right person to succeed him were his communication skills and his radiosurgery background. "I had gotten to know Euan informally at the major national meetings and found him to be a very effective communicator, and when you combine that with his background in radiosurgery, I thought he’d be able to tell Accuray’s story very well," says Adler.

A Rebuilding Project

John Adler, in his tenure running Accuray, provided needed stability for the company’s operations, but more importantly, he completed development of the second-generation, FDA-approved CyberKnife, and jump-started the company’s US sales efforts. For Euan Thomson, the challenge was to transition the company from one focused on engineering a product into a true commercial enterprise, while continuing to refine the product to enable the company to execute its mission to build the radiosurgery market.

One of the first things Thomson did upon joining Accuray was to go out and visit the company’s handful of existing customers. Not surprisingly, given the recent history, he found that they weren’t happy. "Our customers didn’t feel that the company supported them," Thomson recalls. In assessing Accuray’s infrastructure, it didn’t take long for him to see there were significant organizational problems. "While we had a very dedicated, hard-working staff who over the years had worked through a lot of hard times to keep the company alive, they weren’t organized to work in a way that effectively supported our customer base," he explains. One example was the engineering group, which had to handle both customer support and R&D. "That meant that when they were pulled out into the field to address a customer’s problem, R&D projects were necessarily delayed," Thomson points out.

As a result, Thomson spent most of his first six months on the job fundamentally restructuring Accuray. He recalls advising his senior management team that he expected the company would undergo a 60% personnel turnover rate during the first 18 months. It wasn’t that he was looking to let people go, he explained; rather, he wanted to create a different environment at Accuray, and if successful, the personnel changes would be self-fulfilling. Thomson’s turnover estimate turned out to be pretty accurate, but a bit low.

Thomson’s looked to restructure not only the company’s internal organization and relationships with customers, but also its relationships with its investors. "I came in with a clear program in my own mind to get the company cash flow positive in as short a space of time as possible, in order to make sure we were keeping our existing investors happy and supportive of our efforts," he explains. "My sense is that investors are generally loyal and the time they lose confidence is when they see continuing successive financings." Within six months of joining the company, Thomson visited one of Accuray’s major investors and rather than ask for more money as the investor expected, Thomson asked him to set up a line of credit with the commitment that the company would not draw on it unless absolutely necessary. The investor wound up making an additional investment in the company and setting up the credit line, which Accuray has not used.

Part of Thomson’s internal restructuring also included implementing solid financial management tools, which were lacking. This was all a part of his effort to change the profile that the company had developed in the industry. "It was clear that if radical things weren’t done, the process of struggling for financing was just going to continue indefinitely because investors look for certain things, particularly in an organization that has something of a tarnished history as Accuray did at the time," he points out.

In Thomson’s view, not only were investors looking for responsible financial management, they also wanted to see that the technology was being developed, that the market was being built, that there was a clear direction to the sales and marketing effort, and that customers were satisfied with the product and level of service. "The bottom line was that many of those things were not in place at Accuray, and it really required dismantling the organization and re-building it from the ground up in order to create the kind of organization that would make customers and investors happy, and that could move the company forward," he says.

Building a Market

From its customers’ perspective, Accuray does not see itself as competing with the large radiation therapy players. According to Thomson, Accuray views radiosurgery and radiation therapy as adjunct, complementary treatments, depending on the patient. "What’s unique about the CyberKnife is that we are not focused on the radiation therapy market, and our positioning is quite different from the companies in that space," Thomson explains.

One of the big differences between the two markets is their respective physician customers. Elekta, in fact, maintains two separate divisions within the company: one focused on the Gamma Knife and one focused on radiation therapy. That is because radiosurgery, using the Gamma Knife or the CyberKnife, is performed primarily by neurosurgeons, whereas radiation therapy is primarily the province of radiation oncologists and medical physicists. That is not to say that oncologists are not involved in radiosurgery, but that is more a matter of regulatory requirements than medical necessity. Indeed, John Adler points out that, "Radiosurgery is performed as a partnership between surgeons and oncologists because that’s what the law requires. Even though I know more about the CyberKnife system than anyone in the hospital, I am not permitted to push the buttons to operate the system; legally, that can only be done by an oncologist."

Just as the introduction of radiosurgery expanded the customers for radiation-based oncology treatment to include neurosurgeons, Accuray’s strategy is to expand that customer base further to include different surgical specialties. Aiding the growth of this market, Euan Thomson notes, is that the precise accuracy and frameless design of the CyberKnife have combined to change radiosurgery beyond what was originally possible with the Gamma Knife. Not only has Accuray’s technology enabled radiosurgery to be used to treat solid tumors throughout the body, as opposed to just those in the brain, but whereas treatment with the Gamma Knife is limited to a single treatment because of the risks inherent with attempting to reattach the frame in the exact same location for subsequent treatments, the CyberKnife’s precision accuracy allows for repeated treatments to the same tumor location. "The CyberKnife has enabled radiosurgery to evolve to where it now can be delivered in what are called multiple stages, to distinguish this approach from radiation therapy’s fractionated treatments," he explains.

This evolution has largely been enabled by the accuracy of the CyberKnife, which is due primarily to the use of intelligent robotic technology. Thomson sees the system’s robotic components as an important differentiator for Accuray. "We not only have the only dedicated whole body radiosurgery system, but it is also the only radiation delivery system using intelligent robotic technology," he claims. In looking at the CyberKnife, Thomson says most people see the robotic arm and think that is the sole robotic component, when in fact, the entire system operates robotically.

The imaging detectors are the eyes of the system, determining where the tumor is located within the dedicated CyberKnife treatment room; the computer system is the brains, operating the system without user intervention to guide the robotic arm so that the radiation beam can be fired directly at the tumor from any angle or orientation, adjusting throughout the course of the treatment to track, correct, and compensate for respiration or patient movement to continue to precisely irradiate the tumor. In Thomson’s view, Accuray is approaching radiation treatment in a very different way than other companies. "We’re approaching radiation delivery in the same way as a manufacturer of high-precision products approaches a robotic manufacturing process, which involves much greater accuracy than other forms of radiation delivery," he argues.

The CyberKnife’s capabilities have allowed physicians to begin exploring different radiosurgical treatment protocols for different types of tumors. For example, surgeons at the University of Pittsburgh Medical Center, one of the pioneering US hospitals in radiosurgery, treats spinal tumors in one treatment, whereas Georgetown surgeons use three to five stages. Generally, tumors outside of the central nervous system (CNS), that is, other than the brain and spine, tend to be treated in more stages than CNS tumors.

Accuray’s technology has also enabled surgeons to begin using radiosurgery as a first-line therapy for patients who otherwise would have been candidates for traditional surgical resections. That also represents a change in practice patterns because radiosurgery has traditionally been used as a treatment (often palliative) for non-operable patients. "We’ve definitely reached an inflection point where surgeons are getting more comfortable and familiar with CyberKnife radiosurgery, enough that they are starting to use it as a primary therapy for otherwise operable patients who would have previously undergone resections," Thomson says.

The battle to convince surgeons to adopt radiosurgery started in the head and is moving down the rest of the body. When it comes to brain tumors, "the war has been fought and won, and the fights are only at the margins," says John Adler. "Neurosurgeons now generally accept radiosurgery, even though many do so reluctantly because it undermines the basic skills that they dedicated a lifetime to developing."

"But the radiosurgical revolution that I’ve spent much of my professional life proselytizing on behalf of is not going to be won until all other surgeons embrace this therapy, and there we’re really still in the early days," Adler goes on. "Now many thoracic and general surgeons at least are starting to recognize the terminology and accept that radiosurgery could be an option for certain patients, even if they’re ill-prepared to perform the procedure themselves." As a first-line technique to be used for many lung, prostate, pancreatic, and other solid cancers, Adler admits the use of radiosurgery has a long way to go, adding, "That is Accuray’s biggest challenge." For every lung cancer patient now being treated primarily with radiosurgery, Adler estimates there are 50 or more still undergoing traditional surgical resections, and for prostate cancer, the ratio is probably more like 100 to 1, in his view.

Expanding into Soft Tissues

As Adler notes, the radiosurgical revolution started in the brain and then moved down to treat spinal tumors. Spine was a logical next step because many of these tumors are CNS tumors, much like in the brain, and they were also treated primarily by neurosurgeons, who were already familiar with radiosurgery.

One of the challenges in treating spinal tumors is that the cancer is often wrapped around the spine or intertwined with critical structures, so the CyberKnife’s ability to deliver radiation conforming precisely to the tumor, using small, highly collimated beams delivered from a variety of different angles and orientations, makes it well-suited to treat these often difficult cases. Euan Thomson estimates that the US radiosurgery market opportunity for Accuray for brain and spine tumors is around 150,000 patients annually (84,000 brain tumors and 66,000 spine tumors). (See Exhibit 1.)

The real growth, however, for radiosurgery is going to come from expansion of the therapy into soft tissue tumors, primarily in the lung, prostate, pancreas, liver, nasopharynx, and breast. Some of these patients are currently being treated with radiation therapy, so there may be some competition between these treatment modalities, Thomson acknowledges, but generally these are surgical resection cases, so the surgeons are already treating these patients but are generally not as familiar with radiosurgery as their neurosurgeon colleagues. Accuray hopes that the ability to use the CyberKnife to treat patients previously deemed inoperable may further encourage surgeons to adopt radiosurgery to expand their prospective patient population.

The growth opportunities in these other markets dwarf those in brain and spine. Accuray estimates that the potential annual US radiosurgery market for lung and prostate tumors is around 264,000 patients (176,000 lung and 88,000 prostate), and for other soft tissue tumors, approximately 340,000 patients per year. Euan Thomson expects lung and prostate tumors to be the next areas of rapid growth in radiosurgery, noting that thoracic and urologic surgeons have only been using this therapy for about three years and are "still first-generation users, who are coming out of the early adopter phase and entering the early acceptance phase" in terms of CyberKnife utilization.

Even in these early years, the growth of radiosurgery using the CyberKnife has been quite dramatic. In the first quarter of 2001, 87% of US CyberKnife treatments were intracranial, 6% were spinal, and only 7% treated tumors elsewhere in the body. By the first quarter of this year, only 46% were intracranial, 13% were spinal, and 41% treated other tumors, indicated that more than half of all CyberKnife treatments are now extracranial. (See Exhibit 3.) Lung cases are growing particularly rapidly, having increased by 30% each of the last several quarters, to the point where lung is the second most common form of radiosurgery performed in the US with the CyberKnife, moving ahead of spine.

Hitting Milestones

Along with Accuray’s clinical success in expanding radiosurgery has come progress on the sales and financial fronts, resulting in major milestones occurring this year in both areas. In May, the company placed its 100th CyberKnife system, and in February, Accuray went public in one of the most successful recent device IPOs. [See Deal]

The company has clearly come a long way from the days when they were inadvertently selling systems at well below cost. Currently, there are more than 100 units installed worldwide (roughly 65% are in the US, 25% in Asia (mostly Japan), and 10% are in Europe). Euan Thomson estimates the US market size to be in excess of 3,500 systems, with the global market totaling twice that amount. "We know that typical utilization of the CyberKnife is around 200 patients per year, and as hospitals approach 300 annual patients, they’re starting to buy a second system," he notes. Stanford and Georgetown are among the hospitals that already have two systems, and other institutions have also ordered a second. Despite the high cost, Thomson says, "We have found that once a hospital gets to the point where they need another system, if they don’t buy a second CyberKnife, another area hospital is likely to buy a system because local demand tends to increase once a system is in place."

In turn, the company’s sales and revenue growth had Accuray well-positioned to take advantage of the re-emergence of a public market for device companies in the past couple of years. For Accuray, this was actually its second effort at an IPO, and not surprisingly, things went much better this time around.

In late 2000, badly in need of investment, Accuray’s board of directors decided to try to take the company public, looking to take advantage of a slight resurgence in interest in device companies among public investors. But like so many of Accuray’s early-stage decisions, John Adler recalls, this one was also driven "not by what was best for the business, but rather by the absence of a credible alternative." The company and its underwriters put together the S-1 prospectus, but in the spring of 2001, just two days before Accuray’s SEC filing date, Meditec reported unexpectedly lower Japanese CyberKnife sales. At the same time, the bottom was falling out of the US stock market, and as a result, the board decided to withdraw the IPO.

Fast-forward to February 8, 2007: Accuray’s original range for its IPO was $14 to $16, but the company ended up going out at $18 and also increased the size of the offering and was still oversubscribed, raising $174 million. The company’s stock price has been as high as $31 and at press time was around $22 with a $1.2 billion market cap.

What is particularly promising in analyzing the various indicators of Accuray’s recent growth is that the company has achieved these increases on a relatively small installed base of 100 systems. Accuray has staked out a position as the clear leader in the emerging stereotactic radiosurgery market, a space that medtech analyst Mark Richter of Jefferies & Co., who follows the company, believes can be as large as the radiation therapy market.

And although Varian Medical is the radiation therapy market leader, Richter points out that their most recent system, the Trilogy, which is designed as an all-in-one solution for delivering both radiation therapy and radiosurgery, is used for radiosurgery only 25% of the time. In Richter’s view, this means hospitals that are buying Trilogy are doing so to upgrade their radiation therapy capability, not to deliver radiosurgery. "If Accuray continues on its current course, by FY 2008, the company will outperform Varian," Richter predicts. (See "How Varian Medical Stays on Top," IN VIVO, December 2003. (Also see "How Varian Medical Stays On Top" - In Vivo, 1 Dec, 2003.))

The challenge for Accuray remains differentiating the radiation therapy and radiosurgery markets, and convincing the different surgery specialties of the benefits that radiosurgery offers to patients. The company’s accomplishments, particularly in the last five years, have placed them in a position to execute on that strategy. In Euan Thomson’s words, "Accuray and the CyberKnife have matured from being an engineering project with an interesting concept to a medical device company with a compelling product."

Patient Growth = Sales Backlog

This rapid recent growth of radiosurgery beyond the brain is a clear indication that Accuray’s strategy of expanding the treatment’s clinical applications to fulfill John Adler’s original vision for the company is generating results. More than 30,000 patients have now been treated worldwide with the CyberKnife, which means Accuray has successfully doubled the size of the global radiosurgery patient market in the last four years. "Our vision is to make radiosurgery an option for every cancer patient," Thomson says. "By that, we don’t mean that every cancer patient should receive radiosurgery, but that it should be explored as an option that is part of the overall continuum of treatment."

This patient growth is also fueled by the fact that reimbursement for CyberKnife procedures is covered by existing codes for radiation treatment. The average treatment reimbursement is $19,300, a level that users consider well-reimbursed.

Accuray, like most capital equipment companies, has a program to help buyers who lack the capital to purchase a CyberKnife outright to acquire a system by helping with financing. For Accuray, this takes the form of shared ownership agreements, and the company’s goal is to use this program to finance 15% of all units sold. Thomson estimates that currently 10% of its installed base is shared ownership deals. Accuray calculates that the break-even point on a CyberKnife is 54 patients the first year, and then somewhere between 60 and 70 patients the second year, when the hospital bears the additional cost of service contracts that kick in after the expiration of the initial warranty. These high-level service contracts cost $460,000 annually and constitute an important recurring revenue stream for the company: more than 90% of the company’s US customers have purchased these service contracts, which also include significant system upgrades.

The company uses a mixture of direct sales and distributors to cover 45 countries. Accuray sells direct in North America, the UK, Germany, and France, with a 30-person sales force, and it uses distributors elsewhere. One of the challenges in selling capital equipment is the extended sales cycle required to close a deal. Euan Thomson estimates that it takes six to nine months to sell a CyberKnife, a process that is getting shorter as more customers become familiar with the technology.

In addition to the longer sales cycle, the CyberKnife also requires a hospital to construct a special room to house the system, which can take about a year from when an institution places an order. This results in Accuray having a backlog of sales contracts, which enables the company to have a pretty accurate idea of how many units it will be installing a year ahead. This pattern provides excellent guidance into future revenue.

Accuray’s recent revenue growth has been dramatic. After generating $52.9 million in 2006, the company has nearly doubled that through the first three quarters of this year, producing $96.5 million in revenue. And Accuray has $559 million of contracts in backlog from customers who will be installing a CyberKnife at some point in the future.