How to Evaluate Medical Device Technologies

Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI August 1998 Column

Taking a close look at clinical value, barriers to entry, and market size can help investors determine where to put their money.

Whether you're a senior medical device industry executive or a professional money manager, you're looking for the winning technology, product, or company that will help you develop your business, career, and portfolio. A parade of exciting technologies passes before you each year. But what happens to them? Why do so many fizzle at commercialization? Although several important new technologies have driven unprecedented industry growth, young medical device stocks, as a group, have been a resounding disappointment. Since June 1995, initial public offerings (IPOs) are up only 8%, versus 108% for the much lower-risk Standard & Poors (S&P) stocks. What happened?

Clearly, investors underestimated the risks associated with these technologies. They may have also overestimated their potential rewards. An analysis of the medical device companies that went public from June 1995 to July 1997 reveals patterns of success and failure. Understanding these patterns can help an investor capitalize on the many new medical technologies that are invented.

For this analysis, public market values—shareholder value—will be considered a measure of success. Although imperfect, the stock market is probably the most impartial and quantitative reflection of future expectations for risk and reward that we have available.

OBSERVATIONS

The following observations can be made about the industry's public-market performance during this period:

The implications of these observations are much the same for large and small companies—both will benefit more from breakthrough products by building new, long-term franchises. The difference, of course, is that larger companies can also make money selling me-too products, whereas start-ups usually cannot.

ASSESSING TECHNOLOGIES

The preceding observations can be used as the basis for assessing technologies using three variables: clinical value, barriers to entry, and market size.

The first two, clinical value and barriers to entry, determine the probability of success for a company entering a new technology area. Clinical value, or the extent to which a product meets an unmet clinical need, drives market demand and allows value pricing and high profit margins. Barriers to entry, or the technological, regulatory, patent, and commercial hurdles that others must surmount before bringing a similiar device to market, determine the risk of competition and the potential for market share loss and price erosion. Success is broadly defined as achieving market adoption and a leading market share—in other words, the creation of a franchise, a business that cannot be easily replicated.

Table I. Slightly more PMA-track IPOs showed positive returns, but both types were disappointing overall.

Table II. More PMA-track investments showed high returns than 510(k)-track investments, and there were more PMA home runs.

These two success variables also help determine the value of the investment opportunity. Investors put a high premium on the predictability of growth and earnings and an even higher premium on companies with leading market positions—franchise value. At the end of the day, investors will pay much more for a dollar of profit from a secure franchise than from a volatile, competitive industry.

Price-to-earning (P/E) ratios reflect the amount investors will pay (the share price) for a dollar of anticipated future earnings. On average as of May 1998, the market traded at 22x 1999 earnings. P/E ratios are a function of the expected long-term earnings growth rate and the perceived risk associated with that growth. The P/E ratio can be normalized for growth by dividing it by the expected long-term growth rate. This yields the relative ratio of perceived risk to growth, permitting comparison of one stock to another. For example, the best franchise in the industry, Medtronic (Minneapolis), trades at 1.9 times its growth rate; whereas Guidant (Indianapolis), a good company but one exposed to the volatile stent market, trades at 1.5 times its growth rate.

Figure 1. A technology's likelihood of success can be evaluated by plotting clinical value (horizontal axis) and barriers to entry (vertical axis).

A new technology's desirability can be evaluated by plotting it on a 2 x 2 matrix using these variables (Figure 1).

"High" and "low" are hardly specific measurements, of course, and defining them is the hard part of this exercise. For this analysis, solidly above-average clinical value was characterized as follows:

The scale for evaluating barriers to entry is based on the following assumptions:

The most desirable quadrant for a new technology is the high-clinical-value/high-barrier-to-entry quadrant, #1. Yet, although this quadrant may indicate that a technology has the maximum probability of success, it does not indicate how much the technology may be worth. It is market size (price x potential market units x market share), the third variable, that determines a new technology's value. Investors and managements tend to underestimate the risks (variables one and two—clinical value and barriers to entry) and overestimate the potential rewards (variable three—market size) of most new medical technologies.

TESTING THE FRAMEWORK

To test the validity of this investment framework, proxies for the high-clinical-value/high-barrier-to-entry technologies (Quadrant #1) and low-clinical-value/low-barrier-to-entry technologies (Quadrant #4) were chosen—devices that take the premarket approval (PMA) and 510(k) regulatory paths, respectively. The PMA path requires clinical trials, generates quantified claims for clinical value, and confers a lead to market that is a barrier to entry for competitors. The 510(k) confers no claims for clinical value (only equivalence to a preexisting device) and poses no barrier to entry to competitors because it takes only 90 days to obtain.

The sample consisted of all medical device IPOs between June 1995 and September 1997 that were based around technologies in early stages of development or commercialization. This definition included almost every medical device IPO under $700 million in valuation. Consolidation plays and companies that predominantly offered services (such as HMOs) were eliminated from the sample, leaving a total of 57 companies, 22 with products covered by a PMA and 35 with products covered by a 510(k). The total market capitalization today of these companies is approximately $9 billion.

As of the end of May 1998, on average since their IPO, the PMA companies had gained 38.5% in share price, whereas the 510(k) companies showed —9% return. "Home runs" (providing 100% or more return) were marginally more frequent and were much bigger successes among the PMA companies. IPO valuations were higher for PMA companies than 510(k) companies, at $185 million versus $145 million. These findings are summarized in Tables I and II. The divergence in performance between early-stage PMA and 510(k) companies has held up over the past year.

What do these results indicate? In aggregate, more than 60% of medical device IPOs went south—a resoundingly poor record. Even though they did better than the 510(k) companies, returns for the PMA companies were still far behind those of the lower-risk S&P index. Early-stage PMA companies continue to be a disappointment, bringing the percentage of winners closer to 510(k) companies, but the home runs keep getting bigger, offsetting the disappointments in the average return.

In the PMA group, the home runs were Spine-Tech, Inc. (Minneapolis; 472% return); Arterial Vascular Engineering (AVE; Santa Rosa, CA; 235% return); and Closure Medical (Raleigh, NC; 191% return). Winners include the home runs plus Instent (Eden Prairie, MN); Endovascular Technologies (Menlo Park, CA); Novoste Corp. (Norcross, GA); Aradigm Corp. (Hayward, CA); and Endocardial Solutions, Inc. (St. Paul, MN). In the 510(k) group, the home runs were MiniMed Technologies (Sylmar, CA; 292% return) and Ventana Medical Systems, Inc. (Tucson, AZ; 166% return). Other winners include Arthrocare Corp. (Sunnyvale, CA); Bionx Implants, Inc. (Blue Bell, PA); Biopsys Medical, Inc. (Irvine, CA); ESC Medical Systems, Inc. (Yokneam, Israel); SelfCare, Inc. (Waltham, MA); Xomed Surgical Products (Jacksonville, FL); Biosite Diagnostics, Inc. (San Diego); Micro Therapeutics, Inc. (San Clemente, CA); Schick Technologies, Inc. (Long Island City, NY); SpectRx, Inc. (Norcross, GA); and Computer Motion (Santa Barbara, CA). Most of the 510(k) winners were 1997 deals made at valuations 20% below the average.

So, is there magic to a PMA? No, but there is magic to the clinical value and barriers to entry that the PMA represents. Even in the 510(k) group, the home runs were characterized by high barriers to entry and at least some clinical value for the patient.

What about the onerous development time for PMAs? Oddly, this appears to be more positive than it is negative. Think of these stocks, for a moment, as options that trade on the basis of the size of the reward at the end of the development process, the probability of making it through the development process (i.e., past FDA), and the threat of competition. The long PMA development process provides visibility for two of these three major elements of value. The threat of competition is the number-one valuation killer and, therefore, long-term certainty about the nature of the competition adds value. For PMA-track IPOs, the reductions in risk appear to outweigh the long development time.

FITTING DEVICES INTO THE FRAMEWORK

Where do current technologies fall in these quadrants? Quadrant #1 clearly includes implantable cardioverter defibrillators (ICDs), pacemakers, spinal fusion cages, and vision surgery lasers. These products exhibit high clinical value and are limited to few players—ICDs/pacemakers by intellectual property and fusion cages by regulatory timing. Although there will eventually be more competition in the fusion cage arena, the fact that two new companies were able to establish sustainable franchises with these products indicates a successful new start-up technology.

Technologies on the border of Quadrants #1 and #3 might include abdominal aortic aneurysm (AAA) repair products and intracoronary radiation. The companies manufacturing these products are likely to be absorbed by a bigger company (as AAA repair has), but an early start and significant differentiation among the leaders can be hard for bigger players to surmount, leading them to acquire the companies rather than compete. From a management and investor standpoint, this can be considered a success if the acquisition is at a premium to previous financings. For a technology to fall into Quadrant #1, barriers to entry should allow the leader, at least, to establish a franchise.

Quadrant #3 represents technologies with high value but low barriers to entry, which allow prices to degrade and reduce market share assumptions for any given player. Almost all specialty disposable products fall into Quadrant #3, such as interventional cardiology catheters, transmyocardial revascularization/percutaneous myocardial revascularization (PMR) catheters and products, benign prostatic hyperplasia (BPH) ablation treatments, and ablation catheters. Mature implanted products, such as hip, knee, and pedicle screw spinal implants, may also fall into Quadrant #3. Even at maturity, these products tend to have high profit margins (e.g., 60–70%) because they address high-stakes indications, making quality and reputation very important.

Where do stents fall? The universal assumption is that they fall into Quadrant #3—a big-company product line. AVE stands out as a notable exception because of its remarkable success. However, its valuation at a P/E ratio of 15x 1999 earnings per share reflects the assumption that it cannot maintain this success. With fewer competitors (more barriers to entry), the stock would be double today's value. Stents have the short product life-cycle common to products in Quadrant #3. Product differentiation in this quadrant rapidly regresses to the mean and, therefore, companies rely on critical mass for success with these products in this market. Whether AVE is an exception to the rule remains to be seen.

Quadrants #1 and #3 are the primary purview of the major, high-tech medical device companies such as Medtronic, Guidant, and Boston Scientific. Medtronic has strategically targeted Quadrant #1, with almost all of its successful products falling into this space. It is also the most highly valued device company on our P/E-to—growth rate measure. The company has been notably unsuccessful with products in other quadrants, however. Guidant shares Medtronic's attributes on the cardiac rhythm management side but has struggled at times to keep up with the rapid product cycles on the interventional side. Guidant's recent successes clearly indicate that it is recovering, and its rich pipeline in AAA repair and myriad interventional technologies (stents, radiation, e.g.) indicate that it has both business models in hand—no small feat. Boston Scientific lives and breathes in Quadrant #3, targeting critical mass and leadership as an end in itself, introducing hundreds of incremental, high-quality products. The company leaves no new technology unsurveyed and often hedges its bets with multiple partners in a given developing area. In the long term, Boston Scientific will probably maximize growth in this realm but offer lower profitability than Medtronic. As a result, its sales dollars and growth will continue to be valued lower. Orthopedic implant companies also generally fall into Quadrant #3.

In Quadrant #4 fall US Surgical, Ethicon, Baxter, Becton Dickinson, and similar companies. Some of these companies are beginning to extend into the higher-margin Quadrant #3 through acquisitions and internal development, recognizing the tough growth and economic prospects of Quadrant #4. They may be late to market but—presumably—gain market share because of their critical mass. These are the players whose threatening presence has made Quadrant #3 so difficult for start-ups. PMR and BPH treatments are good examples of this.

Quadrant #2 is an odd duck that includes products of marginal or real but poorly quantified clinical value that face a significant barrier to entry. This quadrant, too, can be a healthy environment for start-ups, but is highly situation- or company-specific. Companies in this quadrant include Biopsys, Bionx, and SafeSkin. Biopsys co-opted the critical mass of a partner to lock up access to biopsy tables, and SafeSkin has a new way of making latex gloves that reduces allergic reactions to them, a minor clinical but economically valuable product benefit.

CONCLUSION

So, what should be done now? Is starting a new medical device venture too difficult? Are the odds of success too low? No. Recent experience indicates that great successes are possible and investors are oriented toward highly rewarding opportunities in the long term. However, the one lesson that should be taken away from this recent era in medical device start-ups is that novel proprietary technologies that address important and previously unmet clinical needs are the secret to success. If the medical device industry focuses on such high-value opportunities, the industry's growth, profitability, and value should increase from today's already heady pace.

Robert C. Faulkner is senior medical products analyst at Hambrecht & Quist, LLC (New York City).

Copyright ©1998 Medical Device & Diagnostic Industry