|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 25, No 19 (July 1), 2007: pp. 21e-23 © 2007 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.11.1930
Industry, Clinical Trials, and the Cost of Cancer Drugs: An Investor's PerspectiveTeaneck, NJ To the Editor: The January 10, 2007, issue of the Journal of Clinical Oncology (JCO) examined the cost of cancer care and featured a number of excellent articles from multiple viewpoints; however, one major perspective was not offered—that of the investors who support much of the industry-sponsored research in oncology. As a physician and fund manager who invests in biotechnology and pharmaceutical companies, I would like to offer some thoughts specifically related to the cost of drug development, which is the primary factor cited by companies when defending their drug pricing decisions. A fundamental difference must be noted in the approach to drug development taken by profitable pharmaceutical and biotechnology firms compared with the approach taken by most unprofitable, development-stage biotechnology companies. For the former, research and development are financed by current drug sales, whereas for the latter, research and development costs are provided by investors—both venture capitalists and public equity investors. Profitable companies are primarily valued by investors on the basis of their earnings, thus every research and development dollar frivolously wasted on a drug unlikely to make it to market is a tax-adjusted dollar less of earnings and has negative consequences for a company's stock price. Not surprisingly, these companies tend to take a rigorous approach to assessing candidate drugs for advancement into expensive phase III trials. In stark contrast, the development-stage company is often crudely valued on the basis of its product candidates, both the number of molecules and their stage in clinical development. Phase III compounds are highly valued by the market because they are thought to have been "de-risked" (ie, phase III compounds are presumed to have successfully navigated earlier clinical trials that provided good evidence of safety and efficacy). This creates very strong incentive for development-stage companies to push products along into phase III, an incentive magnified manifold for companies with only one or two drugs in clinical development, which is the case for the majority of nonprofitable biotechnology companies. This factor often leads development-stage companies to make very poor assessments with their own product candidates and to radically misjudge their likelihood of success. Indeed, if the fortunes of the entire company depend on the fate of a single phase II compound, and the interests of those deciding whether or not to enter phase III are tied entirely to the ongoing viability of the company, it would hardly seem surprising that companies push forward with the development of drugs when to objective outside observers further development seems futile. Indeed the market is likely to punish correct decision making by development-stage biotechnology companies. Given a set of questionable phase II data, the stock price of a company would suffer far more if management concluded it would be improper to expend shareholder capital on a phase III program likely to fail than if management decided to forge ahead into phase III on the basis of some dubious, post hoc subgroup analyses. An important element that permits such poor decision making to occur is a relatively easy financing environment. For most companies, the barriers to raising capital are virtually nonexistent—there are seemingly countless money managers willing to gamble their investors' money for the potential upside no matter how long the odds. Some of these managers simply plan to exit long before the phase III data are in, while others lack the knowledge and analytic skills to properly assess the risks; some get caught up in the hype and spin of a highly confident management team, and still others just make analytic mistakes. From among these groups of investors and money managers, companies are almost always able to raise capital and keep moving undeserving compounds forward in development. The terms may not always be to management's liking, but the capital is there waiting. Thus, the company remains afloat for a few more years, salaries are paid, stock options granted, and the phase III trial is funded, and if they get lucky, the phase III trial just might work and then there will be a lot more money for everyone involved. Thus, market forces do not produce efficient drug development; at least for the biotechnology industry, they may actually hinder it. This is particularly true in oncology drug development, where a set of unique circumstances conspire to make drug development more difficult and increase the likelihood that drug candidates are advanced too quickly. Zia et al1 documented a high rate of phase III failures in oncology, even when the phase III protocol uses a regimen identical to what was used in phase II. In particular, the lack of reliable surrogate markers and the common practice of looking for response rates in single arm trials make phase II oncology trials unreliable. Most troubling, in my view (which is admittedly the view of a battle-scarred skeptic), oncology clinical development programs often appear to be designed specifically not to provide insight into the likelihood of success in phase III. Most of the people involved in the development of a particular drug believe it will work, often based on strong preclinical data or anecdotal experience with one or a handful of patients. In addition, we have already seen that those directing the drug development path have multiple, profound incentives to move programs into phase III. Under such circumstances, it is easy for even a highly scrupulous management team to run phase II programs unlikely to generate reliable data but likely to generate a false-positive signal (eg, small, single-arm trials in hand-picked patient populations or in combination with other active agents). The rationalizations for such programs abound. One can claim this represents the most sensitive assay for assessing the activity of the experimental compound. More commonly, the belief that the drug will ultimately work in phase III becomes an assumed end that justifies the means (in this case, the inadequate phase II program). For the unscrupulous, these same types of phase II programs are an easy path to a "promising" phase III cancer drug that can mean a high stock price and enormous personal riches. In my opinion, having followed the industry for a decade, this phenomenon is more common than anyone would like to admit. Another factor in this somewhat perverted set of incentives is the proverbial pot of gold sitting at the end of the phase III rainbow. As the industry has increasingly discovered, insurers face too potent a political and public relations backlash for denying or limiting access to cancer drugs, and as investors have come to expect six figure price tags for drugs that add a few weeks or months of survival (or perhaps an even less robust benefit), all pretense of altruism seems to have disappeared from the industry. It is hard to believe that from the 1990s through 2003, an enormous brouhaha (including Senate hearings and investigations) stirred over the cost of taxol at $10,000 per treatment. Although that was only 4 years ago, that price seems as quaint and antiquated as talking about 5-cent sodas and gasoline at 25 cents per gallon. Furthermore, the problem is not merely the lack of incentives (and the many disincentives) to running phase II oncology trials in which patients are randomly assigned to standard of care plus or minus the experimental agent. Participation in clinical trials has become big business for many academic medical centers, oncology clinics, and group practices. This financial windfall, in combination with the derivative benefits, including consulting and speaking fees, and career advancement associated with abstract and publication authorships, undermines the ability of oncologists to objectively push back on poor clinical trial design and phase III trials doomed to failure. The increasing direct influence of Wall Street on oncology trialists (through intermediary firms putting investors in touch with those involved in trials or through direct compensation for consulting services from stock brokers and money managers) is a pervasive but insidious influence that creates further financial incentives for oncologists to be involved in clinical trials regardless of the quality, and to advocate on behalf of experimental drugs no matter how poor the chances of success. A recent JCO article addressed this topic from the perspective of the naïve oncologist facing wily investors willing to resort to outright deception to obtain confidential information about clinical trials.2 However, the question yet to be addressed is how the objectivity of oncologists (who are often already receiving payments of various kinds directly from companies for whom they consult and conduct research) is affected by payments from Wall Street firms and frequent interactions with investors. In my experience, physician involvement with companies and investors creates serious biases in data interpretation and analysis that generally go completely unrecognized. These inefficiencies in oncology drug development are manifest in various ways. As was amply detailed in the January 10, 2007, JCO issue, the cost for cancer therapy has dramatically increased in recent years with companies citing the costs of drug development as a major reason. Yet the research and development cost of $1 billion per oncology drug mentioned by Adams and Brantner3 and frequently cited by the industry is quite misleading because it is an aggregate number generated by estimating the probability of drugs moving forward through each phase of development. So the company apparently seeking to recoup that $1 billion in research and development on sales of its new drug did not spend anything close to $1 billion on its development. Let's not forget, it only takes 10,000 to 40,000 patients at current oncology drug prices to reach cumulative sales of $1 billion and conservatively assuming a 33% net margin, only 30,000 to 120,000 patients treated to achieve cumulative net profits of $1 billion. The typical development-stage oncology-focused biotechnology company has two or three drugs in development, and will spend only a fraction of that $1 billion figure developing its pipeline of drugs. Most will fail, yet the handful of companies that are skilled and/or fortunate enough to develop winners will be able to use the research and development costs of all those failed companies (by citing this $1 billion per cancer drug figure) to justify pricing that allows it to generate outsized returns on its research and development investment. Ironically, the investors who funded the development of those failed drugs see no returns from sales of the successful drug despite the fact that their failed investment is indirectly responsible for the elevated price of the successful drug. Moreover, how much of that $1 billion per cancer drug estimate can be attributed to needless expenditures on drugs that had almost no shot of success but were advanced into expensive phase III trials after showing promise in phase II studies that were little more than uncontrolled case series? Why should patients, insurers, and taxpayers have to pay the bill for bad decisions about experimental drugs driven by a system that fosters poor drug development while creating disincentives to proper drug development? Finally, the rewards of successful drug development coupled with the extra risk in oncology drug development create disincentives for innovation. While there are many companies conducting high-quality research and development on novel therapeutics and novel targets, there is a seemingly equal number working on modest modifications of existing drugs that have less development risk but have the potential for premium pricing. The number of oncology products in development that are liposomal, pegylated, controlled-release, or prodrug versions of existing compounds that have gone or will soon go generic is staggering. While I do not mean to understate the potential advantages of better dosing schedules or modestly improved adverse effect profiles, one has to balance this against the added costs to the system of replacing a generic product with a premium-priced one and the opportunity cost of devoting capital and brainpower to the development of such "me-too" drugs often with marginal incremental benefits. From my perspective, there are few obvious solutions to the problem. Investors and money managers will always be there to fund the next idea, no matter how preposterous. It seems to me that the burden falls on the oncology community to apply feedback to the industry, to reject trial designs that are destined to yield no useful information, and to decline to enroll patients in phase III trials in which the data supporting drug efficacy is weak. I certainly do not mean to suggest that every single-arm trial ought to be rejected or that a controlled phase II trial of an experimental agent plus or minus standard of care can or need be run in every circumstance. Clearly drug development requires a substantial amount of risk taking under the best of circumstances and many phase III trials will fail. However, a higher degree of skepticism and objectivity from oncologists is needed to balance the multiple and complex impediments to good drug development often faced by company sponsors of clinical trials of experimental drugs. AUTHOR'S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST The author(s) indicated no potential conflicts of interest. REFERENCES
1. Zia MI, Siu LL, Pond GR, et al: Comparison of outcomes of phase II studies and subsequent randomized control studies using identical chemotherapeutic regimens. J Clin Oncol 23:6982-6991, 2005 2. Berlin J, Bruinooge SS, Tannock IF: Ethics in oncology: Consulting for the investment industry. J Clin Oncol 25:444-446, 2007 3. Adams CP, Brantner W: Estimating the cost of new drug development: Is it really 802 million dollars? Health Aff (Millwood) 25:420-428, 2006 Related Articles
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2007 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
