Alan Nelson, Ph.D., President and CEO of NeoPath Inc.

Alan Nelson, Ph.D., planned to be a rock and roll star when he was growing up outside Phoenix. Following his parents' divorce, Nelson became "pretty individualistic," playing in a rock band and reading a great deal. Because "one of my nightmares was that I'd end up in a dusty little town in Arizona for the rest of my life," he sold his drums to help pay tuition, and headed to southern California. At USC, Nelson earned a B.S. in physics, which led him to work at NASA's jet propulsion laboratory. "That," he said, "is where my interest in imaging started." At NASA, Nelson helped design an imaging spectrometer for Jupiter satellite Pioneer 10. He then joined Xerox's Medical Diagnostics Division (now defunct) as lead engineer on a mammography system, his first experience working in cancer prevention. Nelson left Xerox to pursue a graduate degree at the University of California at Berkeley, and earned a Ph.D. in biophysics, specializing in medical imaging. Based on his engineering, science, and medical background, Nelson was appointed jointly to the Harvard and MIT faculties, and he held an endowed chair at MIT. He created a Radiological Sciences program for MIT and Harvard; it was the country's first to offer Ph.D.s in imaging and other aspects of radiology, as well as MD/Ph.D.s. On sabbatical in Europe, he was named visiting scholar to the Royal Society, and worked to break through departmental barriers in Europe's academic system, to effect collaboration among radiologists, pathologists, electrical engineers, physicists, and computer scientists. Recruited in 1987 by the University of Washington, Nelson constructed and chaired the Center for Imaging Systems Optimization, for which he gained funding from companies such as IBM, GE, and Siemens. There, he developed a Ph.D. program that ultimately competed with the one he'd built at MIT/Harvard. Nelson joined forces with venture capitalists in 1988; together they founded NeoPath in January 1989 in Redmond, WA. Nelson later resigned from UW to run NeoPath. University colleagues questioned why he would give up his tenured position to head a company that might fail. "Taking a risk meant that, if I was successful, I could help save some lives — and that was important to me," Nelson explained. "I never had a sense of entitlement -- I always thought that you had to continually earn your own way and find a way to contribute," he said. When not at NeoPath, Nelson, along with his wife Sharon Everson, likes to restore old houses; they've lived in seven different homes in the last 12 years — from Victorians to barns. Nelson learned to install plumbing and electricity and to refinish floors. His parents, who were both farmers at one time, taught him much about hard work.

In the following interview, Nelson, 47, discusses his work at NeoPath: developing the AutoPap system, a fully automated instrument approved by the FDA to re-screen normal Pap smears to find abnormal cells that humans may have missed.

Focus on the Economics of the Diagnostics Business

Bauer: How has the diagnostics business changed in the last five years, and how has this change affected start-up companies?

Nelson: One change is a tremendous focus on the economies surrounding the diagnostics business. The financial picture for the clinical laboratory business, particularly in the last few months, is almost a disaster. The largest laboratories have huge pressure to improve profitability — both gross revenues and cost containment. Organizations like Laboratory Corporation of America, the largest clinical laboratory in the world, are struggling from the merging of two cultures — Roche Biomedical and National Health. Their stock has hit an all-time low.

Corning, another of the laboratory giants, is now being spun off as an independent company. SmithKline, which I believe is the healthiest of them all -- is probably looking at further consolidation, driven principally by economics. It's becoming more difficult for a clinical laboratory to operate in this environment; by the end of this decade there will be a handful of centers of excellence that have figured out how to conduct business in this environment — and they will dominate the market. In terms of start-ups, NASDAQ issues recently took a beating, and a lot of IPOs were pulled off the market. I believe that many pending IPOs did not deserve to be public companies. There's a very high possibility that a lot of those IPOs, had they been successful, would have left shareholders very disappointed. All in all, even though it hurts, the shakeup in the IPO industry has probably been a good thing.

Challenges Facing the IVD Start-Up

Bauer: What are the most critical challenges facing IVD start-up companies today?

Nelson: Sourcing cash for the business. There are fewer and fewer investment firms willing to invest in seed capital for the earliest stages of a startup. Fortunately, they are still willing to make smaller investments to nurture companies. We're going to need more sustaining capital in the pre-IPO stage to support companies, because I think IPOs are going to be further down the road. Before they have a successful IPO, companies will have to be more mature and less of an investment risk. To start a company, you might have to consider financing it yourself to get it underway, then raise capital. Previously, capital was available to organizations that had a compelling vision but hadn't demonstrated any value -- I think those days are gone.

Bauer: As you've noted, some companies have had very difficult IPOs; some have been pulled back. But Wall Street was good to you in your $62 million effort and also to Cytyc, in its $44 million effort. What was the attraction?

Nelson: For NeoPath the attraction is that Pap smear screening is viewed as a potential breakaway market. A successful company could be come a billion-dollar enterprise. That's exciting to Wall Street, as is the possibility of revolutionizing an industry that today is conducted very much the way it was 40 years ago. NeoPath has continuously delivered on milestones — we lay them out years in advance and hit them on target or exceed expectations. Investors can look at what we've promised and see that we did it. We've been successful with the FDA process, which impresses investors who would have expected that we ' d still be waiting for FDA approval on our first product. Our clinical studies were very thorough and convincing, and we got PMA approval from the FDA in only seven months — the fastest ever for a Class III medical device.

Cytyc also now has a favorable track record with the FDA, and they do have a very nice technology. I think they're viewed as a complement to NeoPath: Cytyc builds a better Pap smear, and NeoPath provides a more accurate and economical analysis of the smear.

Automated Pap Smear Screening Economics

Bauer: Your recent FDA approval for AutoPap for re-screening normal Pap smears was a landmark in this field. How did it come about?

Nelson: In the late 1980s, there was a very insightful article in The Wall Street Journal that identified shortcomings in Pap testing. The problem is that the Pap test is very difficult. The cervix is scraped, cells are spread onto a slide and it is sent to a clinical laboratory; it's stained, then viewed by a cytotechnologist with a college degree and one year of specialized training. The cytotechnologist looks for very subtle changes— early indications of cancer— in many different kinds of cells. The cytotechnologists do find most of the disease cases and identify most women at risk of getting cancer, but they also make mistakes— which can become tragic mistakes. They occasionally miss a cancer (false negative) — the patient is told she's normal and the next thing she knows her life is threatened. This is a particularly difficult industry because a Pap smear's cost is already at rock bottom. At a few dollars, it's the cheapest cancer prescreening test available in the clinical lab business— it's very undervalued. It is one of the few tests that has been successful in the prevention of any cancers. In the U.S. it has resulted in a 70 percent reduction in death rate just by early detection. But for a product to be successful in this business, it has to improve the standard of care and curb cost.

Bauer: In the case of re-screening negatives, aren't you adding cost to the test procedure?

Nelson: Yes — we're trying to minimize that but there's almost a law that if you want more quality, you have to pay more. There are many ways to get quality. One that requires no technology is simply to re-screen the Pap smear. That will definitely raise the quality but will double your cost. Our instrument is not as expensive as re-screening -- we come in at about half that cost. We charge customers $3 to $4 for the QC of their normal slides. It feels like a cost increase but laboratories look at it two different ways: One is incremental cost— over and above their conventional re-screening. The second is value: AutoPap detects many cases they've missed — which for big laboratories creates unfavorable exposure. Some of our largest customers took the amount of their settlement costs for a year for misread Pap smears, divided that by the number of Pap smears, and came up with a figure between $1 and $3 that they pay in litigation. When we charge $4 to $5 and recover most of their serious misses, they're willing to give us their $1 to $3 — something that will pay for itself in the long run. So it feels to them like a $1 to $2 increase for quality, which most of the clinical laboratories are willing to embrace because they want to deliver the highest possible quality. And in an integrated health care system that insures patient lives, this $1 to $2 increase can easily be recovered in downstream medical costs.

Bauer: Is there a cost savings to the laboratory to use AutoPap as the primary screener?

Nelson: Yes. That's what makes this all work. The principal cost in Pap smear screening is labor, so if our machine can call a test normal, and the cytotechnologist doesn't have to review it, the lab will save about $5 to $6 for that Pap smear, minus our charge. Our application for primary screening is now before the FDA; it shows very compelling clinical evidence that we can remove up to 30 percent of the workload and also offer a much higher quality standard to the lab. Removing up to 30 percent of the workload, averaged over all slides, saves the lab roughly $2 on each slide — and if, using our QC system, they're al ready paying a dollar or two more, it's either a break even or a cost savings of perhaps 50 cents. Primary screening would allow the lab to reduce its cost per slide or grow its business or both, and still retain the quality of the re-screening capability to detect human mistakes. We think we have the only solution to both of those issues: cost containment and increased accuracy. Right behind the primary screener we plan studies to increase the sort rate, so our objective is to continue to provide software upgrades on the system and to further reduce its costs.

Editor's note: The FDA Hematology and Pathology Devices Panel recommended not to approve the AutoPap system as a primary screener at this time. The advisory panel asked NeoPath to present additional, on-site data that use of AutoPap as a primary screener of microscopic images from Pap smear slides is superior to the current standard of review — by a cytotechnologist followed by quality control review. Nelson said that NeoPath will "work vigilantly to answer all of the panel's questions and resubmit this matter to the FDA's Division of Clinical Laboratory Devices. "

Bauer: There's been concern about taking the cytotechnologist out of the process and having a "black box" What must happen to change that in practice?

Nelson: Actually, I don't think a lot has to be done. At first, we thought that if we developed a black box it might create some discomfort in an industry where people are used to looking at things, but we have not found that to be a significant issue. In fact, we were at an analyst panel recently in New York that included Dr. Wilbur, a very well-known cytopathologist at the University of Rochester, and he was asked point blank: 'The AutoPap is a black box that doesn't show you pictures; does that bother you?' He said, 'Why should it? If the clinical data are good science and show strong evidence that I don't have to look at pictures, then I don't want to look at pictures.' He used as his analogy the complete revolution that took place in bloods, when Technicon, more than 15 years ago, introduced a black box to analyze blood. At that time, bloods were all done manually by a hematech looking under a microscope; that doesn't happen today. All bloods are analyzed on instruments that provide analytical data, and no human looks at a blood sample. I believe that the same thing will happen to Pap smear screening and other related specimens. Technology will provide higher accuracy, a better cost basis, with total automation. NeoPath remains unique in the industry — we made the decision to completely automate this business and therefore we did not make the cytotechnologist a necessary part of the system. There are no pictures to be viewed— the instrument really is a black box that doesn't depend on a human. We do, however, have the capability of showing pictures, and we use it internally to train on.

Development of Automated Pap Smear Screening

Bauer: As you look back on this technology's development, what lessons did you learn?

Nelson: A Pap smear has hundreds of thousands of cells on it, and we reasoned — and pretty well proved — that we could generate statistics that would let us analyze a subsample of that specimen and get the right answer. But when we talked to doctors, they said, 'Wait a second — you're not going to look at all of the Pap smear?' That's what sampling means — we'd be looking at enough to get the correct answer. But the doctors needed assurance that we'd look at the whole Pap smear, because they're haunted by cases where there may be only one or two abnormal cells; if the lab misses them, and the instrument only samples, it may miss them as well. That was a wake-up call, which fortunately we got before we'd invested heavily in a sampling scheme. That instantly raised a technical barrier, and increased our investment: There are four gigabytes of data on one complete Pap smear under high magnification, compared with a mammogram, which has 1/500th of that data. We had to develop technologies including a unique computer — it's patented and we own the chip design — specialized to handle massive morphology computations.

Bauer: How critical are third-party payers to the success of your first-phase product?

Nelson: They haven't been critical— all of our customers have made the decision to go forward with quality control even though third-party payers are not necessarily covering it. I think they realize it is better medicine and in the end will be better business, so they're willing to bite the bullet. At the same time, they've used this as an argument for third party payers: The industry has changed — this really is a much better test and it is our best opportunity to get a fair reimbursement. There comes a point where if reimbursement is too low it will stop women from getting a vital test. Some of those arguments fall on deaf ears, while other organizations are finding some success.

Bauer: What tenets have guided your approach to managing NeoPath's employees?

Nelson: I'm a very fair person. I'm a strong leader, but I define leadership as perhaps the most democratic process there is — you're only a leader because people want to be led by you. That's my philosophy and I let everybody know it, and therefore I have to earn my leadership every day. I strive to inspire people to exceed their personal expectations. I did that as a professor and I do it at NeoPath. One of the greatest inspirations for people is to know at the end of the day that they've done something even they didn't think they could do. It's created an amazing sense of vitality. At NeoPath, we have the largest assemblage of world experts in the industry. We get the best, and we make no compromises. If we have the best team in the world we have the best advantage, and we'll create the best technology and the best product.

Bauer: What does the future hold for NeoPath?

Nelson: We intend to do many things besides Pap smears in the future: sputums, for early detection of lung cancer, and probe analysis . There's a wonderful opportunity to combine with a strategic partner -- a biotech firm making gene or antibody probes — so that we can provide accurate and inexpensive analysis of those probes. We're also looking at histology, determining whether a surgical biopsy is cancerous. We'll start investing in a development program next year, which could result in a product on the market in a few years.