Today's FDA
http://www.100md.com
《新英格兰医药杂志》
The Food and Drug Administration (FDA) is responsible for the safety and efficacy of most food products and all human and veterinary drugs, biologic products, medical devices, cosmetics, and products emitting radiation that are sold within U.S. borders — a list that accounts for an estimated 20 percent of consumer spending, valued at approximately $1.5 trillion.1,2 Although the FDA is often at the center of controversy, the agency has consistently commanded the respect of more than two thirds of American adults.3 Given the magnitude of the FDA's responsibility and the scope of its public trust, I will examine the performance of the agency, with particular attention given to the FDA's relationship with its major counterpart, the pharmaceutical industry, especially in the decade since the enactment of the Prescription Drug User Fee Act in 1992. I do this from my perspective as a former senior vice president of Merck Research Laboratories (until 2000) and as a former assistant secretary for health at the Department of Health and Human Services.
The pressure to approve drugs more or less rapidly ebbs and flows, largely as a consequence of external events. The publication of The Jungle, Upton Sinclair's exposé of the meatpacking industry, helped to create the FDA in 1906.4 The crisis over the elixir sulfanilamide in the 1930s and the thalidomide tragedy in the 1960s prompted legislation to expand the FDA's authority. In the 1980s, the AIDS epidemic focused attention on the need for faster approval of drugs, especially for potentially lifesaving therapies. For the most part, current convention has demanded rapid review coupled with scientifically sound judgment by a strong agency. The Prescription Drug User Fee Act was passed primarily to improve the efficiency of the FDA, specifying that sponsors pay user fees to provide resources to allow the FDA to be more efficient in reviewing applications. The act's subsequent extension, the FDA Modernization Act of 1997, facilitated certain evidentiary standards for priority drugs. Amendments to the Prescription Drug User Fee Act that are contained in the Public Health Security and Bioterrorism Preparedness and Response Act of 2002 renewed prior provisions, added several more efficiency measures, and provided modest funding for post-marketing surveillance.5,6,7,8,9,10
The FDA currently employs approximately 10,000 people, an increase of 85 percent in the decade since the Prescription Drug User Fee Act was enacted, and commands a budget of $1.7 billion, of which $300 million (18 percent) is projected to derive from user fees that are charged to companies when they submit applications.1,2,11 By comparison, the FDA budget is less than 1/50 that of the Department of Agriculture, which has 10 times the workforce of the FDA.12,13 The FDA's budget did not increase in tandem with the recent doubling of the annual budget of the National Institutes of Health, to more than $27 billion. However, as a consequence of the terrorist attacks of September 11, 2001, additional money (more than $500 million) has been earmarked primarily for food-safety surveillance.1,2,14,15
Under the Prescription Drug User Fee Act, the FDA has made appreciable progress in the efficiency of its review process, especially for drugs that are granted priority status, and has enhanced the scientific quality of its actions. This improvement is largely attributable to the growth in the number of its employees, as well as to the maturing of regulatory science. In 2003, the FDA's Center for Drug Evaluation and Research approved 21 new molecular entities, defined as drugs that differ from others on the market, with a total median time for approval of 16.6 months (6.7 months for priority drugs and 23.1 months for standard review, as compared with 14.9 months and 27.2 months, respectively, a decade ago).1,16,17 Since 1992, the average time that is required for clinical development and regulatory review has been reduced by 25 percent, from 9.2 to 6.9 years, and since 1997, the average time for clinical development of fast-track drugs (as compared with standard drugs) has been shortened by 2.0 to 2.5 years.18,19,20 On the other hand, response times for requests that remain outside tracking mandated by the Prescription Drug User Fee Act have continued to suffer from inefficiencies that are extraneous to science. In important progress that is largely attributable to the Food and Drug Administration Modernization Act, the agency has adopted a more streamlined approach to clinical development — for example, through the use of surrogate markers for accelerated approval or approval on the basis of a single multicenter clinical trial.21 Among the provisions of the 2002 amendments to the Prescription Drug User Fee Act are pilot programs involving continuous marketing applications, in which data are submitted and reviewed in continuous fashion, as was done for the first protease inhibitors to treat AIDS, and retrospective performance reviews.
It should be noted that the failure rate for compounds during development has remained fairly constant; however, the overall time from the discovery of a drug to its commercialization has increased from approximately 10 years to 12 years22 (and Walker S: personal communication). The reasons are that the time that is required for preclinical development has been increasing in some cases and that although priority drugs are coming to market sooner, the review time for standard agents has changed less dramatically. Presumably, the Prescription Drug User Fee Act has not resulted in a reduction in the cost of the development of drugs, a figure that has been increasing rather briskly, according to industry statistics.23,24 Furthermore, the sequential legislative actions have placed certain pressures on the agency. According to the Government Accountability Office, performance goals have created overwork for many reviewers.25 This observation, coupled with the fact that more than 30 percent of the FDA's workforce is eligible for retirement within five years, is cause for concern.2,26
The Need for Equipoise
Evolving FDA regulations have prompted the pharmaceutical industry to develop a powerful infrastructure in order to allow timely compliance in a highly competitive environment, whereas the growth in infrastructure at the FDA has lagged behind. This has often resulted in an imbalance when the FDA has attempted to counter industry assertions during negotiations over the approval or labeling of drugs. The problem has been seemingly compounded by several interruptions in FDA leadership. Since 1980, the FDA commissioner has been appointed by the president, subject to confirmation by the Senate. Almost two years have elapsed between the tenures of the past three commissioners, each with distinguished credentials, and the position has been vacant since March 2004 — a problem that is reflective of the political process and stands in contrast to certain overseas regulatory bodies in which heads of agencies serve fixed terms. Thus, the FDA has lacked a continuous voice to articulate policy, advise Congress, and inform the public.
This article proposes that progress since the passage of the Prescription Drug User Fee Act has been insufficient and that the FDA must evolve beyond satisfying the appetite of industry for faster approval of drugs. It is unfortunate that the FDA is measured by the binary standard of review time (too fast or too slow), a measure that often overshadows the FDA's other major responsibilities in public health. The FDA must strengthen its post-marketing safety surveillance and its oversight of the advertising and promotion of drugs. It must continually develop and sustain its capability in emerging areas of science. Perhaps most important, its expertise must be brought to bear in the control of surging drug costs.
Post-Marketing Surveillance
The concept of bringing new drugs and devices to market sooner and more safely must depend on a strong system of post-marketing surveillance. The approval of a drug on the basis of shorter exposure in fewer patients is associated with greater risk. The number of withdrawals of drugs from the market appeared to increase in the period from 1997 to 1998, a time of peak approvals. However, the FDA has maintained that the rates of withdrawals were fairly constant in periods both before and after the passage of the Prescription Drug User Fee Act in 1992.25,26,27 Despite the increasingly large number of patients who have been included in preapproval clinical trials, it is often impossible to detect certain adverse events until drugs have reached the market because such problems are rare, exposure in a more heterogeneous population outside of controlled trials reveals unanticipated risk, or extended use reveals long-term toxicity. Thus, angioneurotic edema, a potentially fatal mechanism-based side effect of angiotensin-converting–enzyme inhibitors that occurs at a rate of less than 1 in 1000, surfaced only after widespread use of the drug. The increased concentrations of terfenadine that resulted from coadministration of drugs that inhibited drug metabolism produced torsades de pointes and sudden death. Protease inhibitors have had as-yet-unexplained metabolic effects in patients with AIDS lipodystrophy. Rofecoxib has been associated with an increased risk of vascular events after prolonged use.27,28,29,30 The voluntary withdrawal of this drug — and other recently recognized drug-safety issues, such as the apparent increased risk of adolescent suicide from certain antidepressants — only serves to highlight the deficiencies inherent in the current surveillance system.
The current reporting and review of post-marketing data are fraught with opportunities for human error.31 Sponsors are obliged to promptly provide the FDA with detailed reports of serious adverse events, regardless of the suspicion of causality, in both the pre- and post-approval phases, along with periodic reports of nonserious events. Simultaneously, the FDA reviews voluntary reports of spontaneous events from consumers and health care professionals.32,33,34 Causality is generally established when the signal event is determined to have occurred at a frequency that was greater than that expected in an analogous population. Such a determination thus depends on the prompt estimation of the rate of events in treated patients and a comparison with rates in similar untreated patients on the basis of epidemiologic databases.
In the current situation, reports can reside in separate repositories (in databases of pre-approval clinical trials, post-approval FDA information, and post-approval data of sponsors and in the medical literature), and the review and collation of those reports allow room for delay and human error. Thus, we are in urgent need of electronic migration of reports, merged databases, and real-time screening with the use of mathematical signal-detection models, executed by a larger complement of FDA professionals.35
Moreover, the FDA has available to it a wealth of information from placebo control groups, which, if compiled, could provide one of the largest comparative epidemiologic databases.36,37 The announced transition to nationwide computer-assisted prescribing of bar-coded medications must proceed promptly to provide an electronic template for the rapid linking of events to prescriptions.38 Finally, the FDA must more readily prosecute lapses in the reporting of adverse events and enforce penalties for failure to complete post-approval studies.
Communication of Information about Risks of Drugs and Devices
An adequate assessment of risk must be accompanied by effective communication about the risk. Recent history is replete with examples of incomplete understanding, such as in the past and present conundrum regarding breast implants.39,40 Much of the information about new agents is communicated by the media in anecdotal reports, at medical symposia (including many that are supported by pharmaceutical companies), and directly by representatives of pharmaceutical companies ("drug detailers") to health care professionals. Since 1997, information has also been disseminated by direct-to-consumer advertising, which exceeded $2.4 billion in 2004.41,42 Federal regulations require that information conveyed by pharmaceutical representatives and advertising conform to data contained in the product label that is approved by the FDA. Nonetheless, many exceptions occur, owing largely to the limited resources allocated for surveillance to the FDA's Division of Drug Marketing, Advertising, and Communications.
Controlling the Cost of Drugs
The direct regulation of drug pricing lies beyond the reach of current laws governing food and drugs. However, public demand to curb escalating costs will probably place more pressure on the FDA to play a role. Pending the outcome of the drug reimportation debate, the FDA may become responsible for ensuring the efficacy and safety of large additional numbers of imported agents.43 Other means to control costs are currently available, such as the acceleration of conversion to over-the-counter status and more thorough attention to the development and approval of generic drugs.44 Nationwide, computer-assisted prescribing of bar-coded medications will reduce medical errors, especially if it is coupled with the establishment of a system of electronic record-keeping for personal medical information (including allergies and concomitant medications) that is similar to the systems implemented by the Department of Veterans Affairs. These systems would, in turn, improve the quality of health care and reduce costs.45,46,47,48 The issue of how to generate more data to determine the "value" of an agent (e.g., by studies of comparative effectiveness) is more difficult than it appears, since proper measurement of benefits and risks often requires large and long-term studies, and enrollment criteria can always be challenged post hoc as not being representative of the "real world." Nevertheless, other countries have succeeded in developing systems to assess the relative value of drugs. The United Kingdom's National Institute for Clinical Excellence makes recommendations regarding which drug costs the National Health Service should reimburse.49
There will always be those who believe that the FDA is overly subject to political influence, is in the pocket of industry, or, conversely, remains an impediment to innovation. Contrasting opinions often appear concurrently in the lay and academic press.50,51,52 There is no doubt that the implementation of the Prescription Drug User Fee Act has resulted in an intensification of the relationship between the FDA and industry sponsors. Fair balance is presumably provided by the appropriate concern of reviewers for safety and the transparency authorized in the Freedom of Information Act. On the other hand, the FDA has become increasingly dependent on guidance from external experts, especially in areas of emerging science, and therefore must be vigilant in evaluating input from such experts who have a potential conflict of interest.53,54 Debate about the relationship between the FDA and the pharmaceutical industry is healthy, but the discussion should expand to consider the true potential of the FDA. It is not merely a matter of too fast or too slow, of too many or too few regulations, or of Democrat or Republican solutions. We must envision the FDA as more than a counterpart to the pharmaceutical industry. It is time for the agency to realize its full potential as both protector and promoter of the public health.
Dr. Slater served as assistant secretary for health at the Department of Health and Human Services from January 2002 until her resignation in February 2003; the FDA was not in her purview in this post. She reports serving on the board of directors of Vertex Pharmaceuticals and having received consulting fees from Celgene Cellular Therapeutics Division, AnorMED, and Warburg Pincus.
References
McClellan MB. The Food and Drug Administration's strategic action plan: protecting and advancing America's health: responding to new challenges and opportunities. August 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/mcclellan/strategic.html.)
FY 2004 Justification of Estimates for Appropriations Committees. Promoting public health through patient, food, and consumer safety. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/oms/ofm/budget/2004/bib1.htm.)
Taylor H. CDC, NIH, FAA, FDA get highest ratings of eleven federal government agencies. Harris Poll no. 73. (Accessed December 30, 2004, at http://www.harrisinteractive.com/harris_poll/index.asp?PID=422.)
Sinclair U. The jungle. New York: Doubleday, Page, 1906:102.
Prescription Drug User Fee Act of 1992, Pub. L. No. 102-571, 21 U.S.C. 379, 106 Stat. 4491 (Oct. 29, 1992).
Food and Drug Administration Modernization Act of 1997, Pub. L. No. 105-115, 21 U.S.C. 355a, 111 Stat. 2296 (Nov. 21, 1997).
Public Health Security and Bioterrorism Preparedness and Response Act of 2002, Pub. L. No. 107-188 (June 12, 2002).
PDUFA III: five-year plan. Rockville, Md.: Food and Drug Administration, July 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/pdufa3/2003plan/default.htm.)
Office of Planning. PDUFA III commitments: new goals and approaches. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/pdufa/report2002/2002-PDUFAIIICommitments.html.)
Department of Health and Human Services. Washington, D.C.: Office of Management and Budget, 2004. (Accessed December 30, 2004, at http://www.whitehouse.gov/omb/budget/fy2005/print/hhs.html.)
FDA's budget proposal for FY 2004. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/bbs/topics/ANSWERS/2003/ANS01195.html.)
Department of Agriculture. Washington, D.C.: Office of Management and Budget, 2004. (Accessed December 30, 2004, at http://www.whitehouse.gov/omb/budget/fy2005/agriculture.html.)
Hilts PJ. Protecting America's health: the FDA, business, and one hundred years of regulation. New York: Alfred A. Knopf, 2003:xv.
McClellan MB. Remarks of the Commissioner of Food and Drugs. Food Drug Law J 2003;58:191-204.
Prior notice of imported food under the Public Health Security and Bioterrorism Preparedness and Response Act of 2002. Washington, D.C.: General Accounting Office, October 23, 2003. (GAO-04-194R.) (Accessed December 30, 2004, at http://www.gao.gov/decisions/majrule/d04194r.pdf.)
Approval times for priority and standard NMEs: calendar years 1993-2003. Rockville, Md.: Food and Drug Administration, 2004. (Accessed December 30, 2004, at http://www.fda.gov/cder/rdmt/NMEapps93-03.htm.)
The pink sheet: FDA clears 21 NMEs in 2003; average approval time is 16.6 months. Pink Sheet 2004;66:1-3.
Drug companies will de-emphasize blockbuster development strategies, according to the Tufts CSDD. Boston: Tufts Center for the Study of Drug Development, January 6, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=24.)
FDA's fast track initiative cut total drug development time by three years, according to Tufts CSDD. Boston: Tufts Center for the Study of Drug Development, November 13, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=34.)
Office of Planning. Report on PDUFA goals: original new product applications. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/pdufa/report2002/2002-onpa.html.)
Temple R. Policy developments in regulatory approval. Stat Med 2002;21:2939-2948.
McAuslane N, Anderson C, Walker S. The changing regulatory environment: reality and perception. R&D briefing no. 42. Surrey, England: Centre for Medicines Research, August 2004.
Total cost to develop a new prescription drug, including cost of post-approval research, is $897 million. Boston: Tufts Center for the Study of Drug Development, May 13, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=29.)
DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug development costs. J Health Econ 2003;22:151-185.
Food and Drug Administration. Effect of user fees on drug approval times, withdrawals, and other agency activities. September 17, 2002. Washington, D.C.: General Accountability Office, 2002. (GAO-02-958.) (Accessed December 30, 2004, at http://www.gao.gov/docdblite/summary.php?recflag=&accno=A05090&rptno=GAO-02-958.)
Major management challenges and program risks. Washington, D.C.: General Accountability Office, January 2003:46-75. (GAO-03-101.)
Friedman MA, Woodcock J, Lumpkin MM, Shuren JE, Hass AE, Thompson LJ. The safety of newly approved medicines: do recent market removals mean there is a problem? JAMA 1999;281:1728-1734.
Slater EE, Merrill DD, Guess HA, et al. Clinical profile of angioedema associated with angiotensin converting-enzyme inhibition. JAMA 1988;260:967-970.
Benson JO, McGhee K, Coplan P, et al. Fat redistribution on indinavir-treated patients with HIV infection: a review of postmarketing cases. J Acquir Immune Defic Syndr 2000;25:130-139.
Merck announces voluntary worldwide withdrawal of Vioxx. Whitehouse Station, N.J.: Merck, 2004. (Accessed December 30, 2004, at http://www.merck.com/newsroom/press_releases/product/2004_0930.html.)
Gale EA. Lessons from the glitazones: a story of drug development. Lancet 2001;357:1870-1875.
Guidance for industry: postmarketing safety reporting for human drug and biological products including vaccines: draft guidance. Rockville, Md.: Food and Drug Administration, March 2001.
Managing the risks from medical product use: creating a risk management framework. Rockville, Md.: Food and Drug Administration, May 1999.
Trontell A. Expecting the unexpected -- drug safety, pharmacovigilance, and the prepared mind. N Engl J Med 2004;351:1385-1387.
DuMouchel W. Bayesian data mining in large frequency tables, with an application to the FDA spontaneous reporting system. Am Stat 1999;53:177-190.
Bennett DA, Jull A. FDA: untapped source of unpublished trials. Lancet 2003;361:1402-1403.
Bone HG. Presented at the 37th Annual Meeting of the Drug Information Association, Denver, July 8–12, 2001. abstract.
FDA rule requires bar codes on drugs and blood to help reduce errors. 2004. (Accessed December 30, 2004, at http://www.fda.gov/oc/initiatives/barcode-sadr/.)
Angell M. Shattuck Lecture -- evaluating the health risks of breast implants: the interplay of medical science, the law, and public opinion. N Engl J Med 1996;334:1513-1518.
Temple RJ, Himmel MH. Safety of newly approved drugs: implications for prescribing. JAMA 2002;287:2273-2275.
Prescription drugs. FDA oversight of direct-to-consumer advertising has limitations. Washington, D.C.: General Accounting Office, October 28, 2002. (GAO-03-177.) (Accessed December 30, 2004, at http://www.gao.gov/new.items/d03177.pdf.)
Wilde Mathews A, Martinez B. Celebrex drama may finally prompt changes at the FDA. Wall Street Journal. December 20, 2004:B1.
Zuger A. Rx: Canadian drugs. N Engl J Med 2003;349:2188-2190.
Rising drug spending is spurring switches of prescription drugs to OTC status in U.S. and Europe, according to CSDD. Boston: Tufts Center for the Study of Drug Development, September 3, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=32.)
McClellan MB. Protecting and advancing America's health through 21st century patient safety: speech before the Urban Institute. November 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/speeches/2003/urbaninstitute1112.html.)
Advanced technologies to lower health care costs and improve quality. Westborough: Massachusetts Technology Collaborative, 2003.
Halvorson GC, Isham GJ. Epidemic of care: a call for safer, better, and more accountable health care. San Francisco: Jossey-Bass, 2003.
Walter LC, Davidowitz NP, Heineken PA, Covinsky KE. Pitfalls of converting practice guidelines into quality measures: lessons learned from a VA performance measure. JAMA 2004;291:2466-2470.
Rawlins MD. NICE work -- providing guidance to the British National Health Service. N Engl J Med 2004;351:1383-1385.
Abraham J. The pharmaceutical industry as a political player. Lancet 2002;360:1498-1502.
Horton R. Lotronex and the FDA: a fatal erosion of integrity. Lancet 2001;357:1544-1545.
Galson S, Kweder S, Houn F, Raczowski V, Honig PK. The FDA and the Lancet: an exchange. Lancet 2001;358:415-415.
Department of Health and Human Services. Financial relationships and interests in research involving human subjects: guidance for human subject protection. Fed Regist 2004;69:26393-26397.
Henderson JA, Smith JJ. Financial conflict of interest in medical research: overview and analysis of institutional controls. Food Drug Law J 2003;58:251-268.(Eve E. Slater, M.D.)
The pressure to approve drugs more or less rapidly ebbs and flows, largely as a consequence of external events. The publication of The Jungle, Upton Sinclair's exposé of the meatpacking industry, helped to create the FDA in 1906.4 The crisis over the elixir sulfanilamide in the 1930s and the thalidomide tragedy in the 1960s prompted legislation to expand the FDA's authority. In the 1980s, the AIDS epidemic focused attention on the need for faster approval of drugs, especially for potentially lifesaving therapies. For the most part, current convention has demanded rapid review coupled with scientifically sound judgment by a strong agency. The Prescription Drug User Fee Act was passed primarily to improve the efficiency of the FDA, specifying that sponsors pay user fees to provide resources to allow the FDA to be more efficient in reviewing applications. The act's subsequent extension, the FDA Modernization Act of 1997, facilitated certain evidentiary standards for priority drugs. Amendments to the Prescription Drug User Fee Act that are contained in the Public Health Security and Bioterrorism Preparedness and Response Act of 2002 renewed prior provisions, added several more efficiency measures, and provided modest funding for post-marketing surveillance.5,6,7,8,9,10
The FDA currently employs approximately 10,000 people, an increase of 85 percent in the decade since the Prescription Drug User Fee Act was enacted, and commands a budget of $1.7 billion, of which $300 million (18 percent) is projected to derive from user fees that are charged to companies when they submit applications.1,2,11 By comparison, the FDA budget is less than 1/50 that of the Department of Agriculture, which has 10 times the workforce of the FDA.12,13 The FDA's budget did not increase in tandem with the recent doubling of the annual budget of the National Institutes of Health, to more than $27 billion. However, as a consequence of the terrorist attacks of September 11, 2001, additional money (more than $500 million) has been earmarked primarily for food-safety surveillance.1,2,14,15
Under the Prescription Drug User Fee Act, the FDA has made appreciable progress in the efficiency of its review process, especially for drugs that are granted priority status, and has enhanced the scientific quality of its actions. This improvement is largely attributable to the growth in the number of its employees, as well as to the maturing of regulatory science. In 2003, the FDA's Center for Drug Evaluation and Research approved 21 new molecular entities, defined as drugs that differ from others on the market, with a total median time for approval of 16.6 months (6.7 months for priority drugs and 23.1 months for standard review, as compared with 14.9 months and 27.2 months, respectively, a decade ago).1,16,17 Since 1992, the average time that is required for clinical development and regulatory review has been reduced by 25 percent, from 9.2 to 6.9 years, and since 1997, the average time for clinical development of fast-track drugs (as compared with standard drugs) has been shortened by 2.0 to 2.5 years.18,19,20 On the other hand, response times for requests that remain outside tracking mandated by the Prescription Drug User Fee Act have continued to suffer from inefficiencies that are extraneous to science. In important progress that is largely attributable to the Food and Drug Administration Modernization Act, the agency has adopted a more streamlined approach to clinical development — for example, through the use of surrogate markers for accelerated approval or approval on the basis of a single multicenter clinical trial.21 Among the provisions of the 2002 amendments to the Prescription Drug User Fee Act are pilot programs involving continuous marketing applications, in which data are submitted and reviewed in continuous fashion, as was done for the first protease inhibitors to treat AIDS, and retrospective performance reviews.
It should be noted that the failure rate for compounds during development has remained fairly constant; however, the overall time from the discovery of a drug to its commercialization has increased from approximately 10 years to 12 years22 (and Walker S: personal communication). The reasons are that the time that is required for preclinical development has been increasing in some cases and that although priority drugs are coming to market sooner, the review time for standard agents has changed less dramatically. Presumably, the Prescription Drug User Fee Act has not resulted in a reduction in the cost of the development of drugs, a figure that has been increasing rather briskly, according to industry statistics.23,24 Furthermore, the sequential legislative actions have placed certain pressures on the agency. According to the Government Accountability Office, performance goals have created overwork for many reviewers.25 This observation, coupled with the fact that more than 30 percent of the FDA's workforce is eligible for retirement within five years, is cause for concern.2,26
The Need for Equipoise
Evolving FDA regulations have prompted the pharmaceutical industry to develop a powerful infrastructure in order to allow timely compliance in a highly competitive environment, whereas the growth in infrastructure at the FDA has lagged behind. This has often resulted in an imbalance when the FDA has attempted to counter industry assertions during negotiations over the approval or labeling of drugs. The problem has been seemingly compounded by several interruptions in FDA leadership. Since 1980, the FDA commissioner has been appointed by the president, subject to confirmation by the Senate. Almost two years have elapsed between the tenures of the past three commissioners, each with distinguished credentials, and the position has been vacant since March 2004 — a problem that is reflective of the political process and stands in contrast to certain overseas regulatory bodies in which heads of agencies serve fixed terms. Thus, the FDA has lacked a continuous voice to articulate policy, advise Congress, and inform the public.
This article proposes that progress since the passage of the Prescription Drug User Fee Act has been insufficient and that the FDA must evolve beyond satisfying the appetite of industry for faster approval of drugs. It is unfortunate that the FDA is measured by the binary standard of review time (too fast or too slow), a measure that often overshadows the FDA's other major responsibilities in public health. The FDA must strengthen its post-marketing safety surveillance and its oversight of the advertising and promotion of drugs. It must continually develop and sustain its capability in emerging areas of science. Perhaps most important, its expertise must be brought to bear in the control of surging drug costs.
Post-Marketing Surveillance
The concept of bringing new drugs and devices to market sooner and more safely must depend on a strong system of post-marketing surveillance. The approval of a drug on the basis of shorter exposure in fewer patients is associated with greater risk. The number of withdrawals of drugs from the market appeared to increase in the period from 1997 to 1998, a time of peak approvals. However, the FDA has maintained that the rates of withdrawals were fairly constant in periods both before and after the passage of the Prescription Drug User Fee Act in 1992.25,26,27 Despite the increasingly large number of patients who have been included in preapproval clinical trials, it is often impossible to detect certain adverse events until drugs have reached the market because such problems are rare, exposure in a more heterogeneous population outside of controlled trials reveals unanticipated risk, or extended use reveals long-term toxicity. Thus, angioneurotic edema, a potentially fatal mechanism-based side effect of angiotensin-converting–enzyme inhibitors that occurs at a rate of less than 1 in 1000, surfaced only after widespread use of the drug. The increased concentrations of terfenadine that resulted from coadministration of drugs that inhibited drug metabolism produced torsades de pointes and sudden death. Protease inhibitors have had as-yet-unexplained metabolic effects in patients with AIDS lipodystrophy. Rofecoxib has been associated with an increased risk of vascular events after prolonged use.27,28,29,30 The voluntary withdrawal of this drug — and other recently recognized drug-safety issues, such as the apparent increased risk of adolescent suicide from certain antidepressants — only serves to highlight the deficiencies inherent in the current surveillance system.
The current reporting and review of post-marketing data are fraught with opportunities for human error.31 Sponsors are obliged to promptly provide the FDA with detailed reports of serious adverse events, regardless of the suspicion of causality, in both the pre- and post-approval phases, along with periodic reports of nonserious events. Simultaneously, the FDA reviews voluntary reports of spontaneous events from consumers and health care professionals.32,33,34 Causality is generally established when the signal event is determined to have occurred at a frequency that was greater than that expected in an analogous population. Such a determination thus depends on the prompt estimation of the rate of events in treated patients and a comparison with rates in similar untreated patients on the basis of epidemiologic databases.
In the current situation, reports can reside in separate repositories (in databases of pre-approval clinical trials, post-approval FDA information, and post-approval data of sponsors and in the medical literature), and the review and collation of those reports allow room for delay and human error. Thus, we are in urgent need of electronic migration of reports, merged databases, and real-time screening with the use of mathematical signal-detection models, executed by a larger complement of FDA professionals.35
Moreover, the FDA has available to it a wealth of information from placebo control groups, which, if compiled, could provide one of the largest comparative epidemiologic databases.36,37 The announced transition to nationwide computer-assisted prescribing of bar-coded medications must proceed promptly to provide an electronic template for the rapid linking of events to prescriptions.38 Finally, the FDA must more readily prosecute lapses in the reporting of adverse events and enforce penalties for failure to complete post-approval studies.
Communication of Information about Risks of Drugs and Devices
An adequate assessment of risk must be accompanied by effective communication about the risk. Recent history is replete with examples of incomplete understanding, such as in the past and present conundrum regarding breast implants.39,40 Much of the information about new agents is communicated by the media in anecdotal reports, at medical symposia (including many that are supported by pharmaceutical companies), and directly by representatives of pharmaceutical companies ("drug detailers") to health care professionals. Since 1997, information has also been disseminated by direct-to-consumer advertising, which exceeded $2.4 billion in 2004.41,42 Federal regulations require that information conveyed by pharmaceutical representatives and advertising conform to data contained in the product label that is approved by the FDA. Nonetheless, many exceptions occur, owing largely to the limited resources allocated for surveillance to the FDA's Division of Drug Marketing, Advertising, and Communications.
Controlling the Cost of Drugs
The direct regulation of drug pricing lies beyond the reach of current laws governing food and drugs. However, public demand to curb escalating costs will probably place more pressure on the FDA to play a role. Pending the outcome of the drug reimportation debate, the FDA may become responsible for ensuring the efficacy and safety of large additional numbers of imported agents.43 Other means to control costs are currently available, such as the acceleration of conversion to over-the-counter status and more thorough attention to the development and approval of generic drugs.44 Nationwide, computer-assisted prescribing of bar-coded medications will reduce medical errors, especially if it is coupled with the establishment of a system of electronic record-keeping for personal medical information (including allergies and concomitant medications) that is similar to the systems implemented by the Department of Veterans Affairs. These systems would, in turn, improve the quality of health care and reduce costs.45,46,47,48 The issue of how to generate more data to determine the "value" of an agent (e.g., by studies of comparative effectiveness) is more difficult than it appears, since proper measurement of benefits and risks often requires large and long-term studies, and enrollment criteria can always be challenged post hoc as not being representative of the "real world." Nevertheless, other countries have succeeded in developing systems to assess the relative value of drugs. The United Kingdom's National Institute for Clinical Excellence makes recommendations regarding which drug costs the National Health Service should reimburse.49
There will always be those who believe that the FDA is overly subject to political influence, is in the pocket of industry, or, conversely, remains an impediment to innovation. Contrasting opinions often appear concurrently in the lay and academic press.50,51,52 There is no doubt that the implementation of the Prescription Drug User Fee Act has resulted in an intensification of the relationship between the FDA and industry sponsors. Fair balance is presumably provided by the appropriate concern of reviewers for safety and the transparency authorized in the Freedom of Information Act. On the other hand, the FDA has become increasingly dependent on guidance from external experts, especially in areas of emerging science, and therefore must be vigilant in evaluating input from such experts who have a potential conflict of interest.53,54 Debate about the relationship between the FDA and the pharmaceutical industry is healthy, but the discussion should expand to consider the true potential of the FDA. It is not merely a matter of too fast or too slow, of too many or too few regulations, or of Democrat or Republican solutions. We must envision the FDA as more than a counterpart to the pharmaceutical industry. It is time for the agency to realize its full potential as both protector and promoter of the public health.
Dr. Slater served as assistant secretary for health at the Department of Health and Human Services from January 2002 until her resignation in February 2003; the FDA was not in her purview in this post. She reports serving on the board of directors of Vertex Pharmaceuticals and having received consulting fees from Celgene Cellular Therapeutics Division, AnorMED, and Warburg Pincus.
References
McClellan MB. The Food and Drug Administration's strategic action plan: protecting and advancing America's health: responding to new challenges and opportunities. August 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/mcclellan/strategic.html.)
FY 2004 Justification of Estimates for Appropriations Committees. Promoting public health through patient, food, and consumer safety. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/oms/ofm/budget/2004/bib1.htm.)
Taylor H. CDC, NIH, FAA, FDA get highest ratings of eleven federal government agencies. Harris Poll no. 73. (Accessed December 30, 2004, at http://www.harrisinteractive.com/harris_poll/index.asp?PID=422.)
Sinclair U. The jungle. New York: Doubleday, Page, 1906:102.
Prescription Drug User Fee Act of 1992, Pub. L. No. 102-571, 21 U.S.C. 379, 106 Stat. 4491 (Oct. 29, 1992).
Food and Drug Administration Modernization Act of 1997, Pub. L. No. 105-115, 21 U.S.C. 355a, 111 Stat. 2296 (Nov. 21, 1997).
Public Health Security and Bioterrorism Preparedness and Response Act of 2002, Pub. L. No. 107-188 (June 12, 2002).
PDUFA III: five-year plan. Rockville, Md.: Food and Drug Administration, July 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/pdufa3/2003plan/default.htm.)
Office of Planning. PDUFA III commitments: new goals and approaches. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/pdufa/report2002/2002-PDUFAIIICommitments.html.)
Department of Health and Human Services. Washington, D.C.: Office of Management and Budget, 2004. (Accessed December 30, 2004, at http://www.whitehouse.gov/omb/budget/fy2005/print/hhs.html.)
FDA's budget proposal for FY 2004. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/bbs/topics/ANSWERS/2003/ANS01195.html.)
Department of Agriculture. Washington, D.C.: Office of Management and Budget, 2004. (Accessed December 30, 2004, at http://www.whitehouse.gov/omb/budget/fy2005/agriculture.html.)
Hilts PJ. Protecting America's health: the FDA, business, and one hundred years of regulation. New York: Alfred A. Knopf, 2003:xv.
McClellan MB. Remarks of the Commissioner of Food and Drugs. Food Drug Law J 2003;58:191-204.
Prior notice of imported food under the Public Health Security and Bioterrorism Preparedness and Response Act of 2002. Washington, D.C.: General Accounting Office, October 23, 2003. (GAO-04-194R.) (Accessed December 30, 2004, at http://www.gao.gov/decisions/majrule/d04194r.pdf.)
Approval times for priority and standard NMEs: calendar years 1993-2003. Rockville, Md.: Food and Drug Administration, 2004. (Accessed December 30, 2004, at http://www.fda.gov/cder/rdmt/NMEapps93-03.htm.)
The pink sheet: FDA clears 21 NMEs in 2003; average approval time is 16.6 months. Pink Sheet 2004;66:1-3.
Drug companies will de-emphasize blockbuster development strategies, according to the Tufts CSDD. Boston: Tufts Center for the Study of Drug Development, January 6, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=24.)
FDA's fast track initiative cut total drug development time by three years, according to Tufts CSDD. Boston: Tufts Center for the Study of Drug Development, November 13, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=34.)
Office of Planning. Report on PDUFA goals: original new product applications. Rockville, Md.: Food and Drug Administration, 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/pdufa/report2002/2002-onpa.html.)
Temple R. Policy developments in regulatory approval. Stat Med 2002;21:2939-2948.
McAuslane N, Anderson C, Walker S. The changing regulatory environment: reality and perception. R&D briefing no. 42. Surrey, England: Centre for Medicines Research, August 2004.
Total cost to develop a new prescription drug, including cost of post-approval research, is $897 million. Boston: Tufts Center for the Study of Drug Development, May 13, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=29.)
DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug development costs. J Health Econ 2003;22:151-185.
Food and Drug Administration. Effect of user fees on drug approval times, withdrawals, and other agency activities. September 17, 2002. Washington, D.C.: General Accountability Office, 2002. (GAO-02-958.) (Accessed December 30, 2004, at http://www.gao.gov/docdblite/summary.php?recflag=&accno=A05090&rptno=GAO-02-958.)
Major management challenges and program risks. Washington, D.C.: General Accountability Office, January 2003:46-75. (GAO-03-101.)
Friedman MA, Woodcock J, Lumpkin MM, Shuren JE, Hass AE, Thompson LJ. The safety of newly approved medicines: do recent market removals mean there is a problem? JAMA 1999;281:1728-1734.
Slater EE, Merrill DD, Guess HA, et al. Clinical profile of angioedema associated with angiotensin converting-enzyme inhibition. JAMA 1988;260:967-970.
Benson JO, McGhee K, Coplan P, et al. Fat redistribution on indinavir-treated patients with HIV infection: a review of postmarketing cases. J Acquir Immune Defic Syndr 2000;25:130-139.
Merck announces voluntary worldwide withdrawal of Vioxx. Whitehouse Station, N.J.: Merck, 2004. (Accessed December 30, 2004, at http://www.merck.com/newsroom/press_releases/product/2004_0930.html.)
Gale EA. Lessons from the glitazones: a story of drug development. Lancet 2001;357:1870-1875.
Guidance for industry: postmarketing safety reporting for human drug and biological products including vaccines: draft guidance. Rockville, Md.: Food and Drug Administration, March 2001.
Managing the risks from medical product use: creating a risk management framework. Rockville, Md.: Food and Drug Administration, May 1999.
Trontell A. Expecting the unexpected -- drug safety, pharmacovigilance, and the prepared mind. N Engl J Med 2004;351:1385-1387.
DuMouchel W. Bayesian data mining in large frequency tables, with an application to the FDA spontaneous reporting system. Am Stat 1999;53:177-190.
Bennett DA, Jull A. FDA: untapped source of unpublished trials. Lancet 2003;361:1402-1403.
Bone HG. Presented at the 37th Annual Meeting of the Drug Information Association, Denver, July 8–12, 2001. abstract.
FDA rule requires bar codes on drugs and blood to help reduce errors. 2004. (Accessed December 30, 2004, at http://www.fda.gov/oc/initiatives/barcode-sadr/.)
Angell M. Shattuck Lecture -- evaluating the health risks of breast implants: the interplay of medical science, the law, and public opinion. N Engl J Med 1996;334:1513-1518.
Temple RJ, Himmel MH. Safety of newly approved drugs: implications for prescribing. JAMA 2002;287:2273-2275.
Prescription drugs. FDA oversight of direct-to-consumer advertising has limitations. Washington, D.C.: General Accounting Office, October 28, 2002. (GAO-03-177.) (Accessed December 30, 2004, at http://www.gao.gov/new.items/d03177.pdf.)
Wilde Mathews A, Martinez B. Celebrex drama may finally prompt changes at the FDA. Wall Street Journal. December 20, 2004:B1.
Zuger A. Rx: Canadian drugs. N Engl J Med 2003;349:2188-2190.
Rising drug spending is spurring switches of prescription drugs to OTC status in U.S. and Europe, according to CSDD. Boston: Tufts Center for the Study of Drug Development, September 3, 2003. (Accessed December 30, 2004, at http://155.212.10.127/NewsEvents/RecentNews.asp?newsid=32.)
McClellan MB. Protecting and advancing America's health through 21st century patient safety: speech before the Urban Institute. November 2003. (Accessed December 30, 2004, at http://www.fda.gov/oc/speeches/2003/urbaninstitute1112.html.)
Advanced technologies to lower health care costs and improve quality. Westborough: Massachusetts Technology Collaborative, 2003.
Halvorson GC, Isham GJ. Epidemic of care: a call for safer, better, and more accountable health care. San Francisco: Jossey-Bass, 2003.
Walter LC, Davidowitz NP, Heineken PA, Covinsky KE. Pitfalls of converting practice guidelines into quality measures: lessons learned from a VA performance measure. JAMA 2004;291:2466-2470.
Rawlins MD. NICE work -- providing guidance to the British National Health Service. N Engl J Med 2004;351:1383-1385.
Abraham J. The pharmaceutical industry as a political player. Lancet 2002;360:1498-1502.
Horton R. Lotronex and the FDA: a fatal erosion of integrity. Lancet 2001;357:1544-1545.
Galson S, Kweder S, Houn F, Raczowski V, Honig PK. The FDA and the Lancet: an exchange. Lancet 2001;358:415-415.
Department of Health and Human Services. Financial relationships and interests in research involving human subjects: guidance for human subject protection. Fed Regist 2004;69:26393-26397.
Henderson JA, Smith JJ. Financial conflict of interest in medical research: overview and analysis of institutional controls. Food Drug Law J 2003;58:251-268.(Eve E. Slater, M.D.)