Health Affairs, 25, no. 5 (2006): 1309-1317 doi: 10.1377/hlthaff.25.5.1309 © 2006 by Project HOPE	New Online   House Health Reform Bill   Paying for Reform   Vetting AHIP's Report   HIV/AIDS Costs   Brief: Insurance Reform   HA Blog Top 10

This Article Abstract Figures Only Reprint (PDF) Online Technical Appendix Submit a response to this article Alert me when this article is cited Alert me when Comments are posted Alert me if a correction is posted Services E-mail this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Add to My Personal Archive Download to Citation Manager Reprints & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Reed, S. D. Articles by Schulman, K. A. Search for Related Content PubMed PubMed Citation Articles by Reed, S. D. Articles by Schulman, K. A. Related Collections Access To Care Insurance Coverage International Issues Business Of Health Pharmaceuticals Quality Of Care Research And Technology Health Spending Consumer Issues Insurance Market Evidence-Based Medicine

Evolution & Future

How Changes In Drug-Safety Regulations Affect The Way Drug And Biotech Companies Invest In Innovation

Abstract

 
Changes in the economics of product development resulting from heightened safety regulations could have a sizable negative impact on drug and biotechnology companies’ decisions about investing in innovation. We developed a model to compare the potential economic effects of pre- and postmarketing strategies to identify safety problems with new drugs. Although expanding Phase III clinical testing and postmarketing safety surveillance are not perfect substitutes, our findings suggest that even a large increase in funding for the latter will have a relatively small adverse impact on investment decisions by drug companies and venture capital firms, compared with the former.

After new drug products have received approval for marketing, the FDA relies largely on a voluntary spontaneous reporting system to generate signals of a safety problem. The limitations of the current system are numerous and have been widely acknowledged.¹ Thus, another option is to strengthen postmarketing surveillance efforts to allow for more rapid identification of drug-related adverse events after a drug has received FDA approval. Such efforts could include randomized or nonrandomized Phase IV (postmarketing) studies, active surveillance systems, or directed efforts that rely on multiple data sources.²

In today’s complex licensing environment, product development often begins with venture-backed firms. Changes in the economics of product development as a result of stricter safety regulations could have a sizable negative impact on investment in innovation and result in an opportunity cost for patients who would have benefited from the development of new drugs. A full examination of these trade-offs is essential in assessing efforts to improve drug-safety evaluation.

In this paper we focus on the potential economic consequences of implementing changes to the current system to better detect serious adverse drug events. The economic consequences stem from higher direct costs associated with either expanded clinical trials or increased postapproval safety efforts. Although these costs would be borne initially by the individual companies, ultimately they would be passed on to the public through higher prices for prescription drugs or a reduction in the development of innovative therapies. Thus, the economic consequences of improving drug safety are of great interest to society and the pharmaceutical and biotech industries alike. However, the magnitude of these consequences has received little attention.

To contribute to the debate about which general strategy for identifying severe adverse drug events should be favored, we developed a model to estimate and compare potential economic effects of regulations that would require expanding preapproval clinical testing of investigational drugs and the effects of regulations that would require that manufacturers increase funding for postmarketing efforts to identify safety problems associated with marketed drugs.

Study Data And Methods

Top Study Data And Methods Study Results Discussion NOTES

 
We developed a model to calculate the net present value of all costs and revenues realized for a hypothetical drug over a twenty-year period, representing the period of patent protection for new drug compounds.³ Calculation of net present value is relevant from the perspective of the pharmaceutical and biotech industries, given the large outlay of resources for research and development (R&D) that must precede the filing of a new drug application (NDA) with the FDA. Net present value is an investment appraisal tool that companies use to make decisions about investing in new compounds.⁴ When the expected net present value of a new drug compound at the beginning of year 1 is positive, development of the drug is expected to have a positive return for the company. When the net present value is negative, a negative return is expected. All else being equal, firms prefer investments in opportunities with positive net present value, and they will forgo investments in opportunities with negative net present value.

We developed three model scenarios, one to represent the current regulatory environment ("present") and two to represent hypothetical regulatory environments ("expand Phase III" and "expand surveillance"). The "expand Phase III" scenario represents a regulatory environment in which Phase III clinical trials have been expanded so that more people are exposed to the drug before FDA approval. The "expand surveillance" model represents a regulatory scenario in which funding for postmarketing safety systems has been increased to identify adverse drug events more rapidly and reliably after FDA approval.

"Present" scenario. We designed the "present" model to represent cash outflows and revenues received over a twenty-year period.⁵ We derived estimates of the duration and total cost of each phase of drug development from a study by Joseph DiMasi and colleagues, which estimated the preapproval cost of developing a new drug compound.⁶ The cost estimates represent total out-of-pocket spending for FDA-approved drugs. Costs assigned to the period during FDA review of the NDA are based on user fees paid to the FDA by pharmaceutical companies, as mandated by the Prescription Drug User Fee Amendments (PDUFA) of 2002.⁷ Consistent with the study by DiMasi and colleagues, we calculated net present value using a discount rate of 11 percent, which represents the weighted average cost of capital in the pharmaceutical industry.

We estimated total annual sales by multiplying the number of patients expected to be treated with the drug each year and the annual revenue received for each treated patient. We assumed that the company would receive net annual revenues of $800 per patient (net of the cost of goods and marketing expenses). We derived the estimate of the number of patients expected to receive the drug annually using the model itself by setting this number to the value at which the net present value at year 1 in the "present" scenario was equal to $0, or the point at which the company would be indifferent in its decision about whether to develop the compound.⁸

To represent a trajectory of increasing sales of the drug after FDA approval, we assumed that 25 percent of the annual projected number of patients would receive the drug in the first year after approval. In the second and third years after approval, the percentages would increase to 50 percent and 75 percent, respectively. Not until the fourth year after FDA approval would 100 percent of the annual projected number of patients receive the drug.

"Expand Phase III" scenario. In this scenario, we increased the duration or the cost, or both, of Phase III clinical testing associated with the recruitment of additional patients into clinical trials. In one case, we assumed that enrolling more patients in Phase III testing would raise annual costs, but we maintained the duration of Phase III testing at three years, thus inherently assuming that there is no shortage of patients to enroll in clinical trials. In another case, we assumed that the duration of Phase III testing would increase, but we maintained a consistent annual cost. Thus, total costs for Phase III testing increased proportionately as the duration of testing increased. Finally, we also tested the combination of increasing the annual cost and the duration of Phase III testing. For each scenario, we conducted sensitivity analyses by evaluating a range of values corresponding to the cost and duration of Phase III testing.

"Expand surveillance" scenario. In this scenario, we increased the amount of money that companies spend on postmarketing safety efforts by increasing the percentage of annual sales allocated for such activities beyond the estimated 0.3 percent used in the "present" scenario.⁹

Study Results

Top Study Data And Methods Study Results Discussion NOTES

 
In the "present" scenario, undiscounted out-of-pocket expenditures for a new drug were approximately $294 million over 12.5 years. With an estimated $800 in annual per patient net revenues and a market size (694,088 patients) that would provide a break-even net present value estimate at the beginning of year 1, undiscounted revenues totaled approximately $3.3 billion over 7.5 years. This was equal to annual revenues of $553 million in years when 100 percent of the market size was realized (that is, four years after FDA approval).¹⁰

To examine the impact of policies that would require more patients to be enrolled in Phase III trials, we continued to assume that Phase III testing would last three years, but we increased the annual cost of testing from 0 percent to 200 percent of the base-case estimate of $38.4 million. This caused the net present value at the beginning of year 1 to drop marginally by $3.4 million for each twenty-percentage-point increase in the annual cost of Phase III trials.

In comparison, we evaluated the impact of maintaining a constant annual cost for Phase III trials while increasing the duration of testing. An increase from 3 to 3.5 years for Phase III testing resulted in a marginal reduction in net present value at year 1 by $17.2 million. This is equivalent to increasing the annual cost of Phase III testing by 102 percent in the previous case, in which the duration of Phase III trials was set at three years. A one-year increase in Phase III testing (to four years) is equivalent to a 193 percent increase in the annual cost of Phase III trials when the duration is set at three years. If Phase III costs increase beyond 193 percent over the baseline estimate of $38.4 million per year to conduct the study in three years, an equivalent net present value could be realized by increasing the duration of the trial to four years at the baseline estimate of $38.4 million per year.

We also evaluated the impact of increasing both the cost and the duration of Phase III testing. Exhibit 1 illustrates the thresholds at which it becomes more advantageous to increase the duration of clinical trials rather than to increase the annual cost of Phase III testing. However, if annual costs in a four-year trial exceed 123 percent beyond the baseline cost, an equivalent net present value would be realized by conducting a five-year trial while maintaining the baseline annual cost of Phase III testing.

View larger version (14K): [in this window] [in a new window]   EXHIBIT 1 Net Present Value In Year 1 When Varying Both The Cost And The Duration Of Phase III Clinical Trials In The "Expand Phase III" Scenario

Discussion

Top Study Data And Methods Study Results Discussion NOTES

 
Our analysis shows that there are likely to be important economic advantages associated with opting for reform efforts aimed at strengthening postmarketing safety systems as a means of better ensuring the safety of new drug products, as opposed to efforts that would require larger Phase III trials prior to regulatory approval. Although our analyses are based on hypothetical scenarios, the results are useful for placing into context the economic impact of various policy options and their potential effects on the investment decisions of drug companies and venture capital firms.

Revenues necessary to recoup R&D costs. A basic finding from the analysis is that, on average, projected annual revenues close to a half-billion dollars per drug are necessary to recoup the discounted R&D costs for the drug. This finding validates the convention used in the pharmaceutical industry that, on average, a product must generate peak annual sales of $500 million to cover R&D costs.¹² It is possible that the drug development costs reported by DiMasi and colleagues are overestimates.¹³ However, the estimates are consistent with estimates produced by other analysts.¹⁴ Furthermore, the cost estimates used in our model represent out-of-pocket expenses for clinical testing and do not include costs for compounds that did not ultimately gain market approval. Thus, our selection of cost estimates generated relatively conservative results. If we had included additional costs for compounds that failed, estimates of annual revenues required from successful drugs to recoup R&D spending would have been much higher.

Potential for negative economic effects. Our model emphasizes the substantial negative economic effects that could result from modification of premar-keting regulatory requirements to improve the detection of drug-safety problems. Even when we assume that there is an adequate supply of patients who are willing to participate in clinical trials, such that the average duration of Phase III clinical testing remains at three years, each twenty-percentage-point increase in costs results in an additional $3.4 million reduction in net present value in year 1. Our model also demonstrates major negative economic consequences of prolonging the duration of Phase III trials. Each one-month increase beyond three years lowers the net present value in year 1 by approximately $2.9 million. If we wanted to increase premarketing safety requirements while maintaining a neutral impact on investment incentives, we would have to greatly decrease the costs of clinical research by other means. This could be possible if there were concomitant changes in other regulatory requirements involved in clinical testing, such as the need for extensive site monitoring.¹⁵ Another option would be to increase the period of patent protection to allow companies to generate revenue over a longer period of time to recoup the additional costs from preapproval clinical testing. The government could also subsidize the fixed costs of the clinical research infrastructure, as proposed in the National Institutes of Health (NIH) Roadmap Initiative.¹⁶ Finally, companies could increase their prices to recoup the additional costs and maintain a zero net present value.

Our model also shows that sizable increases in spending for postmarketing safety evaluation are likely to have a much less detrimental economic impact on manufacturers.¹⁷ Our conclusion that the expansion of postmarketing safety efforts is preferred over expanded Phase III testing is similar to other recommendations to market new pharmaceuticals with limited FDA approval on the condition that the manufacturer collect extensive safety data for several years after marketing approval.¹⁸

Relevance for biotech firms. Our findings might be particularly relevant for biotech firms. First, biotech companies may have limited amounts of venture capital for the development of new drug compounds. Thus, regulatory changes that lead to increased cost or duration of Phase III clinical testing, or both, could stifle investment in drug development in the biotech sector, as compared to companies with greater resources. Second, drugs developed by biotech firms often target niche markets. With fewer patients available in such settings, expanding Phase III testing requirements might be more likely to prolong Phase III trials for biotech companies targeting small markets compared to companies that develop drugs for larger markets. As our findings suggest, investment decisions appear to be more sensitive to relative changes in the duration of clinical testing than to relative changes in the annual costs of clinical testing. Lastly, if regulatory changes required that companies invest a percentage of sales in postmarketing surveillance efforts, biotech companies that offer high-cost drugs for smaller patient populations would effectively spend more money per treated patient for safety surveillance. Whether this effect would be desirable would need to be determined in light of multiple factors that are considered risk-benefit decisions for new drugs, including the health benefits of the drug, the severity of the condition that the drug is intended to treat, the availability of alternative treatments, and the incidence and severity of various adverse drug events.

Implementing changes to regulations that improve the detection of drug-safety problems in the pre- or postmarketing period could affect drug development decisions. Therefore, although politicians, regulators, and the public might perceive short-run benefits in terms of improved drug safety, there is a risk that innovation in the pharmaceutical and biotech industries could be reduced by the adverse economic consequences of regulatory policy decisions. Previous research has shown that reductions in drug industry profits, achieved through price controls, could have a sizable impact on R&D investment, leading to fewer breakthrough therapies in the future.¹⁹ Conversely, improvements in pre- or postmarketing safety efforts could have beneficial economic effects on companies if the improvements could help protect companies from litigation. This could occur through regulatory relief as a quid pro quo for implementation of improved safety systems and by limiting the number of patients exposed to a drug before its risks are identified and made available to prescribers and the public.

Limitations. Our analysis has some limitations. We limited the analysis to drugs that are ultimately approved for marketing. The results might be overly optimistic in terms of absolute values for required net revenue or market size. However, because our focus is directed at the relative impact of policy options for improving the detection of drug-safety problems, our model should be robust to this concern. Also, it should be recognized that pre- and postmarketing efforts to identify safety problems are not direct substitutes for each other. Postmarketing safety surveillance, whether through active or passive systems, is usually necessary to identify rare adverse drug events.

Another limitation is that the model does not incorporate other variables that affect the ability to identify adverse drug events, such as the underlying rate of events in the patient population, the increased odds of the event to be detected, the level of statistical power desired, and the level of tolerance for false-positive results. Each of these issues would affect the costs and benefits of both policy options. In addition, the nature of the adverse drug event is critical. Potentially fatal events are of more concern than less severe events. However, the severity of the adverse event must also be balanced with regard to the condition for which the drug is indicated, the number of people with the condition, and the availability of alternative treatments.

Because this balance of risks and benefits is a highly contentious issue, stakeholders must be involved in such risk-benefit determinations through transparent dissemination of the necessary information and dialogue among parties.²⁰ Our model provides information for just one of many factors at play in these complex decisions. A more comprehensive model could also include the impact of reduced regulatory burden in the conduct of human clinical trials; expanded animal testing to detect nontarget effects prior to clinical testing; the use of genes, proteins, and metabolites to identify nontarget effects during early human testing; and determinations of which patients benefit and which do not, thereby changing the risk-benefit relationship at the patient level.²¹

EVEN A SIZABLE INCREASE IN FUNDING for postmarketing safety efforts will have a relatively small economic impact on the investment decisions of pharmaceutical and biotechnology companies, compared to additional requirements for Phase III clinical testing. Policy analysts should consider these results in the context of the overall net benefits to society resulting from alternative reform options aimed at improving the FDA’s ability to detect adverse drug events.

Editor's Notes

 
Shelby Reed is an assistant research professor of medicine in the Duke University School of Medicine and a faculty affiliate at the Center for Clinical and Genetic Economics, Duke Clinical Research Institute, Duke University Medical Center, in Durham, North Carolina. Robert Califf is a professor of medicine in the Duke University School of Medicine, director of the Duke Clinical Research Institute, and vice chancellor for clinical research at Duke University. Kevin Schulman (kevin.schulman{at}duke.edu) is a professor of medicine in the Duke University School of Medicine; a professor of management in the Fuqua School of Business, Duke University; and director of the Center for Clinical and Genetic Economics.

The authors received no funding for this study. They thank Damon Seils of Duke University for editorial assistance and manuscript preparation.

NOTES

Top Study Data And Methods Study Results Discussion NOTES

 

1. C. Baum et al., "The Spontaneous Reporting System in the United States," in Pharmacoepidemiology, 2d ed., ed. B.L. Strom (West Sussex, U.K.: John Wiley and Sons, 1994); and P.B. Fontanarosa, D. Rennie, and C.D. DeAngelis, "Postmarketing Surveillance—Lack of Vigilance, Lack of Trust," Journal of the American Medical Association 292, no. 21 (2004): 2647–2650.[Free Full Text]
2. C.L. Bennett et al., "The Research on Adverse Drug Events and Reports (RADAR) Project," Journal of the American Medical Association 293, no. 17 (2005): 2131–2140.[Abstract/Free Full Text]
3. C.P. Stickney and R.L. Weil, Financial Accounting: An Introduction to Concepts, Methods, and Uses, 10th ed. (Cincinnati: South-Western College Publishing, 2003).
4. F.C. Zinkhan, "Combining Net Present Value and Strategic Effectiveness to Evaluate Drug Development Alternatives," Journal of Pharmaceutical Marketing and Management 9, no. 1 (1995): 19–22.
5. For a summary of the parameters and estimates incorporated into the model, see the online technical appendix at http://content.healthaffairs.org/cgi/content/full/25/5/1309/DC1.
6. J.A. DiMasi, R.W. Hansen, and H.G. Grabowski, "The Price of Innovation: New Estimates of Drug Development Costs," Journal of Health Economics 22, no. 2 (2003): 151–185.[CrossRef][Web of Science][Medline]
7. "Establishment of Prescription Drug User Fee Rates for Fiscal Year 2005," Federal Register 69, no. 147 (2004): 46165–46168.
8. The company would not be expected to receive an incremental financial benefit or loss from its investment decision.
9. D.B. Ridley et al., "Spending on Postapproval Drug Safety," Health Affairs 25, no. 2 (2006): 429–436.[Abstract/Free Full Text]
10. Note that this estimate refers to net present value at the beginning of year 1, before preclinical and clinical development of the product. As time passes, the net present value changes, because costs incurred up to that point are considered sunk costs and do not factor into the calculation of net present value.
11. Ridley et al., "Spending on Postapproval Drug Safety."
12. M.D. Rawlins, "Cutting the Cost of Drug Development?" Nature Reviews Drug Discovery 3, no. 4 (2004): 360–364.[CrossRef][Web of Science][Medline]
13. "Drug Costs: Research and Development Costs: The Great Illusion," Prescrire International 13, no. 69 (2004): 32–36[Medline]; and U.S. Congress Office of Technology Assessment, Pharmaceutical R&D: Costs, Risks and Rewards, Pub. no. OTA-H-522, February 1993, http://www.wws.princeton.edu/cgi-bin/byteserv.prl/~ota/disk1/1993/9336/9336.PDF (accessed 15 August 2005).
14. P. Tollman et al., A Revolution in R&D: How Genomics and Genetics Are Transforming the Biopharmaceutical Industry (Boston: Boston Consulting Group, 2001).
15. E.L. Eisenstein et al., "Reducing the Costs of Phase III Cardiovascular Clinical Trials," American Heart Journal 149, no. 3 (2005): 482–488.[CrossRef][Web of Science][Medline]
16. E. Zerhouni, "Medicine: The NIH Roadmap," Science 302, no. 5642 (2003): 63–72.[Abstract/Free Full Text]
17. Baum et al., "The Spontaneous Reporting System."
18. S. Okie, "Safety in Numbers—Monitoring Risk in Approved Drugs," New England Journal of Medicine 352, no. 12 (2005): 1173–1176.[Free Full Text]
19. J.A. Vernon, "Examining the Link between Price Regulation and Pharmaceutical R&D Investment," Health Economics 14, no. 1 (2005): 1–16[CrossRef][Web of Science][Medline]; T.A. Abbott and J.A. Vernon, "The Cost of U.S. Pharmaceutical Price Reductions: A Financial Simulation Model of R&D Decisions," NBER Working Paper no. 11114 (Cambridge, Mass.: National Bureau of Economic Research, 2 August 2005); and F.M. Scherer, "The Link between Gross Profitability and Pharmaceutical R&D Spending," Health Affairs 20, no. 5 (2001): 216–220.[Free Full Text]
20. E.M. Vogt, "Effective Communication of Drug Safety Information to Patients and the Public: A New Look," Drug Safety 25, no. 5 (2002): 313–321.[CrossRef][Web of Science][Medline]
21. J.W. Lee et al., "Method Validation and Measurement of Biomarkers in Nonclinical and Clinical Samples in Drug Development: A Conference Report," Pharmaceutical Research 22, no. 4 (2005): 499–511[CrossRef][Web of Science][Medline]; W. Lee et al., "Cancer Pharmacogenomics: Powerful Tools in Cancer Chemotherapy and Drug Development," Oncologist 10, no. 2 (2005): 104–111[Abstract/Free Full Text]; and A.D. Roses, "Pharmacogenetics and Drug Development: The Path to Safer and More Effective Drugs," Nature Reviews Genetics 5, no. 9 (2004): 645–656.[CrossRef][Web of Science][Medline]

                      What's this?