《学术英语（医学）》拓展阅读资料：randomized clinical trials removing unnecssary obstacles

The NEW ENGLAND JOURNAL Of MEDICINE SOUNDING BOARD Randomized clinical trials Removing Unnecessary Obstacles Christina Reith, M.B., Ch B, Martin Landray, M.B., Ch B, P J. Devereaux, M.D., Ph D Jackie Bosch, M.Sc, Christopher B. Granger, M D, Colin Baigent, B.M., BCh Robert M. Califf, M.D., Rory Collins, M.B., B.S., and Salim Yusuf, M.D., D Phil. Since their widespread introduction in the mid- large increases in cost and effort caused by the dle of the 20th century, randomized trials of current regulatory system, many existing and new sufficiently large size have provided reliable as- interventions are not being evaluated, and the sessments of the safety and efficacy of treat- trials that are conducted are smaller and less ments that have produced substantial improve- formative than they might otherwise be 7,10 Trials ments in health. During the past decade, comparing widely accepted treatments are also however, increasingly onerous regulation and re- being inhibited because of the trend toward re- lated bureaucracy have made trials much more quiring excessively detailed informed consent difficult and costly to conduct, slowing further and related litigation (as in the case of recent improvements. 2,3 This adverse regulatory envi- neonatal intensive care trials 1, 12). The negative ronment hinders important research and urgent- effect these obstacles have on efforts to obtain ly needs to be changed for the benefit of patients reliable evidence regarding the safety and effi- cacy of treatments affects the care of people not As one example of the current problems just in developed countries but also in develop (Table 1), the requirement to obtain approval ing countries, where resources are more limited from many different bodies before starting a and the burden of disease is large. 13 Hence, there trial results in substantial delays and costs. 4 is an urgent need for major changes in proce Even where centralized regulatory and ethics re- dures for the initiation, conduct, monitoring, views have been adopted (e.g, in the United and safety reporting of clinical trials, such that Kingdom), hurdles remain when permission is they are more proportionate to the likely hazards required from each study site, resulting in dupli- of the trials. Otherwise, researchers may be in- cated effort and delays. For multicenter trials, hibited from conducting such trials at all, which this can involve hundreds of separate approvals, will ultimately place patients at much greater risk. take more than a year to complete, and cost Many regulators acknowledge the seriousness hundreds of thousands of dollars. 5 The situation of these problems, but their attempts to resolve is exacerbated when trials involve more than them have often further hampered research. For one country, each requiring separate approval instance, the European Union 2001 Clinical Trials from multiple organizations. For example, in Directive(Directive 2001/20/EC)4 was intended China, it typically takes up to a year to obtain to facilitate the performance of trials across Eu- regulatory approval. Such inefficiencies are in- rope and better protect the public. However, it is creasingly recognized as damaging to medical now widely accepted including by the Euro- progress, as noted in reports from the U.S. pean Commission itself -that this directive, President's Council of Advisors on Science and which was incorporated into legislation differ Technology and the Institute of Medicine. ently in different countries, failed to achieve Previous randomized trials have shown that either aim. ,10 As a consequence, trials that in number of widely used treatments are not ef- volve European Union sites, including those con- fective or safe; for example, treatment with anti- ducted by U.S. investigators, have been impeded, arrhythmic drugs after heart attacks causes car- and patient care has suffered. After public diac arrests, and routine glucocorticoid use for sultation, the European Commission isst head injury reduces survival. 8,9 Because of the proposal in July 2012 to replace the Clinical N ENGLJMED 369: 11 NEJMORC SEPTEMBER 12, 2013 1061

n engl j med 369;11 nejm.org september 12, 2013 1061 s o u n d i n g b o a r d T h e n e w e ngl a nd j o u r na l o f m e dic i n e Randomized Clinical Trials — Removing Unnecessary Obstacles Christina Reith, M.B., Ch.B., Martin Landray, M.B., Ch.B., P.J. Devereaux, M.D., Ph.D., Jackie Bosch, M.Sc., Christopher B. Granger, M.D., Colin Baigent, B.M., B.Ch., Robert M. Califf, M.D., Rory Collins, M.B., B.S., and Salim Yusuf, M.D., D.Phil. Since their widespread introduction in the mid￾dle of the 20th century, randomized trials of sufficiently large size have provided reliable as￾sessments of the safety and efficacy of treat￾ments that have produced substantial improve￾ments in health.1 During the past decade, however, increasingly onerous regulation and re￾lated bureaucracy have made trials much more difficult and costly to conduct, slowing further improvements.2,3 This adverse regulatory envi￾ronment hinders important research and urgent￾ly needs to be changed for the benefit of patients and public health. As one example of the current problems (Table 1), the requirement to obtain approval from many different bodies before starting a trial results in substantial delays and costs.4 Even where centralized regulatory and ethics re￾views have been adopted (e.g., in the United Kingdom), hurdles remain when permission is required from each study site, resulting in dupli￾cated effort and delays. For multicenter trials, this can involve hundreds of separate approvals, take more than a year to complete, and cost hundreds of thousands of dollars.5 The situation is exacerbated when trials involve more than one country, each requiring separate approval from multiple organizations. For example, in China, it typically takes up to a year to obtain regulatory approval. Such inefficiencies are in￾creasingly recognized as damaging to medical progress, as noted in reports from the U.S. President’s Council of Advisors on Science and Technology6 and the Institute of Medicine.7 Previous randomized trials have shown that a number of widely used treatments are not ef￾fective or safe; for example, treatment with anti￾arrhythmic drugs after heart attacks causes car￾diac arrests, and routine glucocorticoid use for head injury reduces survival.8,9 Because of the large increases in cost and effort caused by the current regulatory system, many existing and new interventions are not being evaluated, and the trials that are conducted are smaller and less in￾formative than they might otherwise be.7,10 Trials comparing widely accepted treatments are also being inhibited because of the trend toward re￾quiring excessively detailed informed consent and related litigation (as in the case of recent neonatal intensive care trials11,12). The negative effect these obstacles have on efforts to obtain reliable evidence regarding the safety and effi￾cacy of treatments affects the care of people not just in developed countries but also in develop￾ing countries, where resources are more limited and the burden of disease is large.13 Hence, there is an urgent need for major changes in proce￾dures for the initiation, conduct, monitoring, and safety reporting of clinical trials, such that they are more proportionate to the likely hazards of the trials. Otherwise, researchers may be in￾hibited from conducting such trials at all, which will ultimately place patients at much greater risk. Many regulators acknowledge the seriousness of these problems, but their attempts to resolve them have often further hampered research. For instance, the European Union 2001 Clinical Trials Directive (Directive 2001/20/EC)14 was intended to facilitate the performance of trials across Eu￾rope and better protect the public. However, it is now widely accepted — including by the Euro￾pean Commission itself — that this directive, which was incorporated into legislation differ￾ently in different countries, failed to achieve either aim.2,10 As a consequence, trials that in￾volve European Union sites, including those con￾ducted by U.S. investigators, have been impeded, and patient care has suffered. After public con￾sultation, the European Commission issued a proposal in July 2012 to replace the Clinical Trials

The NEW ENGLAND JOURNAL Of MEDICINE Table 1. Problems with the Clinical Trial Environment and Possible Solutions. t for cancer prevention(Table 1). The proposed regulation also does not address many of the Problem other problems; its changes are directed chiefly inical-trial toward expediting trial initiation(e.g, approval complex, authorization with defined timelines processes). In particular, there is still inappro- for approval priate emphasis on safety assessments that rely ts of individual adverse events, as we of low-risk trial ss burdensome rules and shorter as on inefficient approaches to the conduct and 邮一2m Clinical Practice(ICH-GCP) guidelines The intention of the ICH-GCP guidelines disproportionate focus on retro- toring of clinical trials, with increased to ensure the safety and rights of participants in trials and also to ensure the reliability of trial Monitoring of drug safety involves Greater empha erging safety data by independent results so that the safety of future patients would be protected. 19 Despite these well-intend- of adverse- event rates in con ees in the context of the efficacy re- ed aims, the ICH-GCP guidelines have often been trol groups sults, with appropriate sharing of this interpreted and implemented in ways that have information with regulatory authorities been unnecessarily obstructive and have not The ICH-GCP guidelines are in. Propagation of risk-based approaches yielded the intended outcomes. 0Regulatory ternational Conference on Har. agencies have drafted guidance on the ICH-GCP monization, by issuing appropriate guidelines, but these documents are often lengthy of key quality aspects guidelines, a revised version of the(some several hundred pages long),illustrating H-GCP guidelines, the difficulties in its interpretation. The ICH-GCP guidelines were developed jointly between the guidelines by experts in clinical trials. pharmaceutical industry and various regulatory authorities, without the involvement of academ- ic experts in trial design and conduct. Perhaps ICH-GCP denotes International Conference on Harmonization Good Clinical as a consequence, it is not based on a clear under Practice standing of the key principles that underlie trials involving randomization and control groups (in Directive with a regulation that would be uni- particular, their robustness for unbiased assess form among all European Union countries. 10 ments of safety and efficacy). 1, 20, 21 Instead, the This proposal does include some improvements, ICH-GCP guidelines often place undue emphasis such as a single portal for authorizing trials on less important aspects of trials at the expens conducted in the European Union, more flexi- of critical aspects -for example, they focus to bility for obtaining consent in emergency situa- an inappropriate extent on ensuring the com- tions, and measures to decrease indemnity costs. pleteness and accuracy of each piece of data that It also proposes less burdensome rules and is recorded, even though minor errors occurring "low-intervention"(i. e trials testing marketed should not materially affect the findings. 2 pg shorter approval times for trials described as with similar frequency in the treatment gro treatments used in accordance with their author- Approaches derived from the ICH-GCP guide ized uses or standard practice and involving ad- lines for monitoring the conduct of trials often ditional procedures that pose no more than a involve frequent site visits, even though they are minimal risk or burden to participants). This costly and time consuming and evidence of their proportionate, risk-based approach will be appre- usefulness is lacking.20-22 The emphasis at such ciated by researchers who have been advocating visits, reinforced by the extensive list of docu it for many years. 15-18 However, the definition of ments classed by the ICH-GCP guidelines as low-intervention trials used by the European "essential " for trial conduct, is typically on Union should be extended to trials in which es- things that are easy to check- for example tablished treatments with good safety profiles curricula vitae and drug-storage records and are tested for novel uses-for example, aspirin temperature logs. However, it is now widely ac- N ENGLJMED 369: 11 NEJM.ORG SEPTEMBER 12, 2013

T h e n e w e ngl a nd j o u r na l o f m e dic i n e 1062 n engl j med 369;11 nejm.org september 12, 2013 Directive with a regulation that would be uni￾form among all European Union countries.10 This proposal does include some improvements, such as a single portal for authorizing trials conducted in the European Union, more flexi￾bility for obtaining consent in emergency situa￾tions, and measures to decrease indemnity costs. It also proposes less burdensome rules and shorter approval times for trials described as “low-intervention” (i.e., trials testing marketed treatments used in accordance with their author￾ized uses or standard practice and involving ad￾ditional procedures that pose no more than a minimal risk or burden to participants). This proportionate, risk-based approach will be appre￾ciated by researchers who have been advocating it for many years.15-18 However, the definition of low-intervention trials used by the European Union should be extended to trials in which es￾tablished treatments with good safety profiles are tested for novel uses — for example, aspirin for cancer prevention (Table 1). The proposed regulation also does not address many of the other problems; its changes are directed chiefly toward expediting trial initiation (e.g., approval processes). In particular, there is still inappro￾priate emphasis on safety assessments that rely on reports of individual adverse events, as well as on inefficient approaches to the conduct and monitoring of trials that derive from the Inter￾national Conference on Harmonization Good Clinical Practice (ICH-GCP) guidelines. The intention of the ICH-GCP guidelines was to ensure the safety and rights of participants in trials and also to ensure the reliability of trial results so that the safety of future patients would be protected.19 Despite these well-intend￾ed aims, the ICH-GCP guidelines have often been interpreted and implemented in ways that have been unnecessarily obstructive and have not yielded the intended outcomes.20 Regulatory agencies have drafted guidance on the ICH-GCP guidelines, but these documents are often lengthy (some several hundred pages long), illustrating the difficulties in its interpretation. The ICH-GCP guidelines were developed jointly between the pharmaceutical industry and various regulatory authorities, without the involvement of academ￾ic experts in trial design and conduct. Perhaps as a consequence, it is not based on a clear under￾standing of the key principles that underlie trials involving randomization and control groups (in particular, their robustness for unbiased assess￾ments of safety and efficacy).1,20,21 Instead, the ICH-GCP guidelines often place undue emphasis on less important aspects of trials at the expense of critical aspects — for example, they focus to an inappropriate extent on ensuring the com￾pleteness and accuracy of each piece of data that is recorded, even though minor errors occurring with similar frequency in the treatment groups should not materially affect the findings.20,21 Approaches derived from the ICH-GCP guide￾lines for monitoring the conduct of trials often involve frequent site visits, even though they are costly and time consuming and evidence of their usefulness is lacking.20-22 The emphasis at such visits, reinforced by the extensive list of docu￾ments classed by the ICH-GCP guidelines as “essential” for trial conduct, is typically on things that are easy to check — for example, curricula vitae and drug-storage records and temperature logs. However, it is now widely ac￾Table 1. Problems with the Clinical Trial Environment and Possible Solutions.* Problem Solution The approval process is complex, costly, heterogeneous, and time-consuming Single submission point for clinical-trial authorization with defined timelines for approval A one-size-fits-all approach is used, with regulation of low-risk trials of well-understood drugs that is similar to regulation of trials of completely new drugs, for which the risks are unknown Adoption of risk-based approach, with less burdensome rules and shorter approval time for low-risk trials (e.g., a marketed drug with a good safety profile being tested for non￾standard uses) Monitoring of trial conduct involves disproportionate focus on retro￾spective data verification Adoption of risk-based approach to moni￾toring of clinical trials, with increased use of centralized monitoring Monitoring of drug safety involves undue focus on individual case reports without consideration of adverse-event rates in con￾trol groups Greater emphasis on regular review of emerging safety data by independent data and safety monitoring commit￾tees in the context of the efficacy re￾sults, with appropriate sharing of this information with regulatory authorities The ICH-GCP guidelines are in￾flexible and frequently over￾interpreted and place undue emphasis on relatively unim￾portant aspects of trials at the expense of key quality aspects Propagation of risk-based approaches to streamlining clinical trials by the International Conference on Har￾monization, by issuing appropriate interpretations of the ICH-GCP guidelines, a revised version of the ICH-GCP guidelines, or both; de￾velopment of authoritative and in￾formed “good clinical trial practice” guidelines by experts in clinical trials, with input from regulators * ICH-GCP denotes International Conference on Harmonization Good Clinical Practice

SOUNDING BOARD cepted that monitoring should focus on those those involved in the conduct of clinical trials aspects of trials that are of most relevance to are being confronted by an overwhelming vol the rights and safety of participants (e.g, the ume of reports on suspected unexpected serious consent procedures and ascertainment of serious adverse reactions25 and, despite a substantial ex- adverse events) and on the reliability of the study penditure of money and effort, this approach to results(e.g, the integrity of the randomization pharmacovigilance has rarely led to useful in process and completeness of follow-up). 20, 21,23 sights or improved safety Monitoring should be designed to detect impor- By contrast, moderate adverse effects of treat tant problems early in a trial so that they can be ment on common outcomes, such as an increase Dressed, as opposed to discovering problems in cardiovascular events with coxib treatment, 26 retrospectively by auditing, when there is no may well be of much greater relevance to public longer an opportunity to rectify them. Conse- health, but their reliable detection requires evi- quently, the use of centralized statistical moni- dence from studies that are both randomized toring as part of a risk-based approach to ensur- and controlled. 1 Consequently, a more effective ing trial quality has been advocated by the strategy for safety monitor n randomize Clinical Trials Transformation Initiative(www controlled trials would be based on the regular ctti-clinicaltrials. org)24 and endorsed by the review of the emerging safety data that is cus Food and Drug Administration(FDA). 6 European tomarily conducted by independent data and regulators hav ly recommended risk- safety monitoring committees, with the study based quality management, 17 18 and the proposed treatment assignments revealed and considered Union regulation allows the extent and in the context of the efficacy results. The degree nature of monitoring to be modified depending to which such safety information is shared with on the characteristics of the trial. However, be- regulatory authorities could be agreed on before cause the current draft of the regulation refers the start of the trial and could depend or to the ICH-GCP guidelines as a quality standard ber of factors, such as the amount of (despite their inherent problems), the successful experience with the treatment being implementation of a risk-based approach in Eu- The cogency of these concerns about the present rope will require either a fundamental shift approach to safety monitoring has been recog away from the current rigid interpretation of the nized by the FDA, which has issued revised guidelines or a substantial modification of the guidance for reporting requirements. 25,27 In par- guidelines themselves ticular, to reduce current levels of overreporting a risk-based approach should also be applied of alleged serious drug reactions, the FDA seeks safety monitoring during trials. However, the rapid reporting of such an event only when there ICH-GCP guidelines and related regulations cur- is considered to be a reasonable possibility that it ently require rapid reporting of all serious ad- was caused by the study drug. The FDa guidance verse events thought to be related to the study also distinguishes between the rare circumstanc- treatment and not previously documented with es in which it is appropriate to submit individual that treatment to all relevant regulatory authori- case reports and the more common circumstanc- ties,ethics committees, and site investigators. es in which cases should be aggregated and com- In accordance with the regulatory require- pared with those in a control group. However, ments, these suspected unexpected serious ad- the European Commission guidance and the pro- verse reactions"are typically reported on a case- posed regulation from the European Union have by-case basis (rather than as grouped reports not yet made this important distinction. with a meaningful denominator) only for the In conclusion, some regulatory agencies have participants in the active-treatment group rath- responded positively to the need for impro er than with the corresponding event rates for ments in the regulatory environment for clinical the control group. The use of this type of un- trials. However, the effect of those responses controlled data can reasonably be expected to may be limited by the complex system of regula- detect only large effects of drug exposure on tion and related bureaucracy that applies to clin- rare outcomes, such as the Stevens-Johnson ical trials at local, national, and international syndrome, hepatic failure, or angioedema. Cur- levels. Those responsible for the ICH-GCP guide- rently, however, regulatory authorities and all lines have been resistant to engaging adequately N ENGLJMED 369: 11 NEJMORC SEPTEMBER 12, 2013 106

n engl j med 369;11 nejm.org september 12, 2013 1063 sounding board cepted that monitoring should focus on those aspects of trials that are of most relevance to the rights and safety of participants (e.g., the consent procedures and ascertainment of serious adverse events) and on the reliability of the study results (e.g., the integrity of the randomization process and completeness of follow-up).20,21,23 Monitoring should be designed to detect impor￾tant problems early in a trial so that they can be addressed, as opposed to discovering problems retrospectively by auditing, when there is no longer an opportunity to rectify them. Conse￾quently, the use of centralized statistical moni￾toring as part of a risk-based approach to ensur￾ing trial quality has been advocated by the Clinical Trials Transformation Initiative (www .ctti-clinicaltrials.org)24 and endorsed by the Food and Drug Administration (FDA).16 European regulators have also recently recommended risk￾based quality management,17,18 and the proposed European Union regulation allows the extent and nature of monitoring to be modified depending on the characteristics of the trial. However, be￾cause the current draft of the regulation refers to the ICH-GCP guidelines as a quality standard (despite their inherent problems), the successful implementation of a risk-based approach in Eu￾rope will require either a fundamental shift away from the current rigid interpretation of the guidelines or a substantial modification of the guidelines themselves. A risk-based approach should also be applied to safety monitoring during trials. However, the ICH-GCP guidelines and related regulations cur￾rently require rapid reporting of all serious ad￾verse events thought to be related to the study treatment and not previously documented with that treatment to all relevant regulatory authori￾ties, ethics committees, and site investigators. In accordance with the regulatory require￾ments, these “suspected unexpected serious ad￾verse reactions” are typically reported on a case￾by-case basis (rather than as grouped reports with a meaningful denominator) only for the participants in the active-treatment group, rath￾er than with the corresponding event rates for the control group. The use of this type of un￾controlled data can reasonably be expected to detect only large effects of drug exposure on rare outcomes, such as the Stevens–Johnson syndrome, hepatic failure, or angioedema.1 Cur￾rently, however, regulatory authorities and all those involved in the conduct of clinical trials are being confronted by an overwhelming vol￾ume of reports on suspected unexpected serious adverse reactions25 and, despite a substantial ex￾penditure of money and effort, this approach to pharmacovigilance has rarely led to useful in￾sights or improved safety. By contrast, moderate adverse effects of treat￾ment on common outcomes, such as an increase in cardiovascular events with coxib treatment,26 may well be of much greater relevance to public health, but their reliable detection requires evi￾dence from studies that are both randomized and controlled.1 Consequently, a more effective strategy for safety monitoring in randomized, controlled trials would be based on the regular review of the emerging safety data that is cus￾tomarily conducted by independent data and safety monitoring committees, with the study￾treatment assignments revealed and considered in the context of the efficacy results. The degree to which such safety information is shared with regulatory authorities could be agreed on before the start of the trial and could depend on a num￾ber of factors, such as the amount of previous experience with the treatment being studied. The cogency of these concerns about the present approach to safety monitoring has been recog￾nized by the FDA, which has issued revised guidance for reporting requirements.25,27 In par￾ticular, to reduce current levels of overreporting of alleged serious drug reactions, the FDA seeks rapid reporting of such an event only when there is considered to be a reasonable possibility that it was caused by the study drug. The FDA guidance also distinguishes between the rare circumstanc￾es in which it is appropriate to submit individual case reports and the more common circumstanc￾es in which cases should be aggregated and com￾pared with those in a control group. However, the European Commission guidance and the pro￾posed regulation from the European Union have not yet made this important distinction. In conclusion, some regulatory agencies have responded positively to the need for improve￾ments in the regulatory environment for clinical trials. However, the effect of those responses may be limited by the complex system of regula￾tion and related bureaucracy that applies to clin￾ical trials at local, national, and international levels. Those responsible for the ICH-GCP guide￾lines have been resistant to engaging adequately

The NEW ENGLAND JOURNAL Of MEDICINE with the research community about the incorpora- and McMaster University- both in Hamilton, ON, Canada tion of risk-based approaches and other evidence-(- D. J-B SY ) and the Duke Clinical Research Institute,Duke University, Durham, NC based revisions. Certain entities have benefited quests to Dr. Reith at the Clinical Trial Service Unit and m the complexity of the current regulatory ological Studies Unit, Richard Doll Bldg,old Road Roosevelt Dr, Oxford OX3 7LF, United Kingdom, or at environment-not just contract research organi- reith@ctsu. x ac uk. zations and companies providing training in the ICH-GCP guidelines, but also regulatory groups 1. Collins R, MacMahon S Reliable assessment of the effects in pharmaceutical companies and other institu- of treatment on mortality and major morbidity, I: clinical trials tions, which have seen their revenue and influ- 2. McMahon AD, Conway Dl, Macdonald TM, McInnes GT. The ence increase substantially -and they too may unintended consequences of clinical trials regulations. PLoS Med ppose streamlining. Ultimately, rather than the 2009: 3011): e1000131. piecemeal changes that are currently being made 3. calit Mi, linical trials bureaucrac unintended conse- to the regulations and bureaucracy governing 4. Duley L, Antman K, Arena J, et al. Sp clinical trials, comprehensive reform of the whole conduct of randomized trials. Clin Trials 2008: 5:40- system is required. This should involve experts 3. Petersen LA, Simpson K, Sorelle R,Urech T, Chitwood low variability in the institutional review board review process in clinical trials developing authoritative "good affects minimal-risk multisite health services research. Ann In- clinical trial practice"guidelines founded on the tern Med 2012: 156: 72, Advisors on Science and Technology key principles that underpin the reliable assess-(PCAST). Report to the President on propelling innovation in ment of the safety and efficacy of treatments. At drug discovery, development, and evaluation. Washington, DC thesametimeprocessesthatareoflittleprovenPcAst,September2012(httpillwww.whitehouse-gowlsites/defaultl value should be strongly 7. Institute of Medicine. Envisaging a transformed clinical tr With support from all the relevant stakehold- terprise in the United States: establishing an agenda for rs-including regulators, academics, those in 2020: workshop summary. Washington, DC: National Acad industry, and patient representatives -and, 8. The Cardiac Arrhythmia Suppression Trial (CAST) Investiga crucially, a better understanding of these issues tors. Preliminary report: effect of encainide and flecainide on provide a more appropriate basis for the devel myocardi in a randomize trait of arhythmia suppressionafter at the governmental level, such guidelines could morta opment and interpretation of regulations for clin. 9. Roberts I, Yates D, Sandercock B, et al. Effect of inuravewith orticosteroids on death within 14 days in 10 008 adult ical trials. It is becoming increasingly clear that clinically significant head injury (MRC CRASH triaD): random- more extensive use of health records and infor- hd placebo- controled ran dancer matics platforms, 2 along with more refined the council on clinical trials on medicinal products for human ethical approaches characterized by the expecta- use oneill on tt ilec.europa. eu/health/files/clinicaltrials/ tion that participation in clinical trials is the 2012_07pr norm rather than the exception, 29 could support ll.Drazen JM, Solomon CG, Greene MEInformed consent and a dramatic increase in the emergence of defini- SUPPORT. N Engl J Med 2013: 368: 1929-31 tive evidence about treatments. Unless radical 12. Magnus D, Caplan AL Risk, consent, and SUPPORT. N Engl improvements are made to the regulatory envi- 13. Lang T, Cheah PY, White NJ. Clinical research: time for sen- ronment, the potential of clinical trials to assess sible global guidelines. Lancet 2011-377:1553-5 the safety and efficacy of new and existing treat- council of 4 April 2001 on the approximation of the laws, regula- ments and, thereby, to produce substantial im- tions and administrative provi f the Member States relat. provements in health care and public health will ing to the implementation of good clinical practice in the conduct not be fulfilled urnaloftheEuropeanCommunities(http://eur-lex.europa.eu All authors are members of the Sensible Guidelines Group, LexUriServ/LexUriServ do?uri=OJ: L: 2001: 121: 0034: 0044: en: PD Clip Hananized jointly by the Population Health Research1Bam, ouben P, Ihrig K,世Rm山改 ercial clinical trials inical Trial Service Unit and Epidemiological Studies Unit, Clin Trials 2009: 6: 585-96 Nuffield Department of Population Health, University of Oxford; 16. Guidance for Industry: oversight of clinical investigations and the Duke Clinical Research Institute, Duke University. a risk-based approach to monitoring. Rockville, MD: Food inclosureformsprovidedbytheauthorsareavailablewithandDrugAdministrationAugust2013(http://www.fda.govl the full text of this article at nEJM. org. downloads/ Drugs/ Guidance Compliance Regulatorylnformation/ From the Clinical Trial Service Unit and Epidemiologic ncil/Department of Health/Medicine ield Department of Population Health, care Products Regulatory of Oxford, Oxford, United Kingdom(CR, ML,CB adapted approaches to the manageme nical trials Population Health Research Institute, Hamilton Health vestigational medicinal products 011(http N ENGLJMED 369: 11 NEJM.ORG SEPTEMBER 12, 2013

T h e n e w e ngl a nd j o u r na l o f m e dic i n e 1064 n engl j med 369;11 nejm.org september 12, 2013 with the research community about the incorpora￾tion of risk-based approaches and other evidence￾based revisions. Certain entities have benefited from the complexity of the current regulatory environment — not just contract research organi￾zations and companies providing training in the ICH-GCP guidelines, but also regulatory groups in pharmaceutical companies and other institu￾tions, which have seen their revenue and influ￾ence increase substantially — and they too may oppose streamlining. Ultimately, rather than the piecemeal changes that are currently being made to the regulations and bureaucracy governing clinical trials, comprehensive reform of the whole system is required. This should involve experts in clinical trials developing authoritative “good clinical trial practice” guidelines founded on the key principles that underpin the reliable assess￾ment of the safety and efficacy of treatments. At the same time, processes that are of little proven value should be strongly discouraged. With support from all the relevant stakehold￾ers — including regulators, academics, those in industry, and patient representatives — and, crucially, a better understanding of these issues at the governmental level, such guidelines could provide a more appropriate basis for the devel￾opment and interpretation of regulations for clin￾ical trials. It is becoming increasingly clear that more extensive use of health records and infor￾matics platforms,28 along with more refined ethical approaches characterized by the expecta￾tion that participation in clinical trials is the norm rather than the exception,29 could support a dramatic increase in the emergence of defini￾tive evidence about treatments. Unless radical improvements are made to the regulatory envi￾ronment, the potential of clinical trials to assess the safety and efficacy of new and existing treat￾ments and, thereby, to produce substantial im￾provements in health care and public health will not be fulfilled. All authors are members of the Sensible Guidelines Group, which is organized jointly by the Population Health Research Institute, Hamilton Health Sciences and McMaster University; the Clinical Trial Service Unit and Epidemiological Studies Unit, Nuffield Department of Population Health, University of Oxford; and the Duke Clinical Research Institute, Duke University. Disclosure forms provided by the authors are available with the full text of this article at NEJM.org. From the Clinical Trial Service Unit and Epidemiological Stud￾ies Unit, Nuffield Department of Population Health, University of Oxford, Oxford, United Kingdom (C.R., M.L., C.B., R.C.); the Population Health Research Institute, Hamilton Health Sciences and McMaster University — both in Hamilton, ON, Canada (P.J.D., J.B., S.Y.); and the Duke Clinical Research Institute, Duke University, Durham, NC (C.B.G., R.M.C.). Address reprint re￾quests to Dr. Reith at the Clinical Trial Service Unit and Epidemi￾ological Studies Unit, Richard Doll Bldg., Old Road Campus, Roosevelt Dr., Oxford OX3 7LF, United Kingdom, or at christina .reith@ctsu.ox.ac.uk. 1. Collins R, MacMahon S. Reliable assessment of the effects of treatment on mortality and major morbidity, I: clinical trials. Lancet 2001;357:373-80. 2. McMahon AD, Conway DI, Macdonald TM, McInnes GT. The unintended consequences of clinical trials regulations. PLoS Med 2009;3(11):e1000131. 3. Califf RM. Clinical trials bureaucracy: unintended conse￾quences of well-intentioned policy. Clin Trials 2006;3:496-502. 4. Duley L, Antman K, Arena J, et al. Specific barriers to the conduct of randomized trials. Clin Trials 2008;5:40-8. 5. Petersen LA, Simpson K, Sorelle R, Urech T, Chitwood SS. How variability in the institutional review board review process affects minimal-risk multisite health services research. Ann In￾tern Med 2012;156:728-35. 6. President’s Council of Advisors on Science and Technology (PCAST). Report to the President on propelling innovation in drug discovery, development, and evaluation. Washington, DC: PCAST, September 2012 (http://www.whitehouse.gov/sites/default/ files/microsites/ostp/pcast-fda-final.pdf). 7. Institute of Medicine. Envisaging a transformed clinical tri￾als enterprise in the United States: establishing an agenda for 2020: workshop summary. Washington, DC: National Acade￾mies Press, 2012. 8. The Cardiac Arrhythmia Suppression Trial (CAST) Investiga￾tors. Preliminary report: effect of encainide and flecainide on mortality in a randomized trial of arrhythmia suppression after myocardial infarction. N Engl J Med 1989;321:406-12. 9. Roberts I, Yates D, Sandercock P, et al. Effect of intravenous corticosteroids on death within 14 days in 10 008 adults with clinically significant head injury (MRC CRASH trial): random￾ised placebo-controlled trial. Lancet 2004;364:1321-8. 10. Proposal for a regulation of the European parliament and of the council on clinical trials on medicinal products for human use, and repealing Directive 2001/20/EC. Brussels: European Commission, 2012 (http://ec.europa.eu/health/files/clinicaltrials/ 2012_07/proposal/2012_07_proposal_en.pdf). 11. Drazen JM, Solomon CG, Greene MF. Informed consent and SUPPORT. N Engl J Med 2013;368:1929-31. 12. Magnus D, Caplan AL. Risk, consent, and SUPPORT. N Engl J Med 2013;368:1864-5. 13. Lang T, Cheah PY, White NJ. Clinical research: time for sen￾sible global guidelines. Lancet 2011;377:1553-5. 14. Directive 2001/20/EC of the European parliament and of the council of 4 April 2001 on the approximation of the laws, regula￾tions and administrative provisions of the Member States relat￾ing to the implementation of good clinical practice in the conduct of clinical trials on medicinal products for human use. Official Journal of the European Communities (http://eur-lex.europa.eu/ LexUriServ/LexUriServ.do?uri=OJ:L:2001:121:0034:0044:en:PDF). 15. Brosteanu O, Houben P, Ihrig K, et al. Risk analysis and risk adapted on-site monitoring in noncommercial clinical trials. Clin Trials 2009;6:585-96. 16. Guidance for Industry: oversight of clinical investigations — a risk-based approach to monitoring. Rockville, MD: Food and Drug Administration, August 2013 (http://www.fda.gov/ downloads/Drugs/GuidanceComplianceRegulatoryInformation/ Guidances/UCM269919.pdf). 17. Medical Research Council/Department of Health/Medicines and Healthcare Products Regulatory Agency Joint Project. Risk￾adapted approaches to the management of clinical trials of in￾vestigational medicinal products. October 2011 (http://www

SOUNDING BOARD mhra. gov. uk/home/groups/l-ctuldocuments/websiteresources Sherman RE. Clinical trials: rethinking how we ensure c Drug Inf)2012;46:657-60 18. European Medicines Agency Reflection paper on risk based 25. Sherman RB, Woodcock Norden J, Grandinetti C, uality managementin clinical trials: EMA/INS/GCP)394194/2011. R. New FDA regulation to improve safety reporting in Gust4,2011(httpllwww.ema.europa.eu/ b/document trials. N Engl J Med 2011: 365: 3-5 26. Coxib and traditional NSAID Trialists'(CNT) Collaboration. 9. International Conference on Harmonisation Guideline for Vascular and upper gastrointestinal effects of non-steroidal GoodClinicalPracticeE6(ri).June1996(http://www.ich.organti-inflammatorydrugsmeta-analys fileadmin/public_web_site/ich_products/guidelines/efficacyldatafromrandomisedtrialsLancet2013(http://ww om/journals/lancet/article/PIISo140-6736(13)60900-9/fulltext). 20. Grimes DA, Hubacher D, Nanda K, Schulz KF, Moher D, Alt- 27. Guidance for Industry and Investigators. Safety reporting nan DG. The Good Clinical Practice guideline: a bronze stan- requirements for INDs and BA/BE studies. Rockville, MD: Food dard for clinical research. Lancet 2005:36 indDrugAdministrationDecember2012(http://www.fda.govl 21. Baigent C, Harrell FE, Buyse M, Emberson JR, Altman DG. downloads/ Drugs/Guidance Compliance Regulatorylnformation nsuring trial validity by data quality assurance and diversifica- 28. A new pathway for the regulation and governance of health 22. Morrison BW, Cochran CL, White JG, et al. Monitoring the research. London: Academy of Medical Sciences, January 2011. quality of conduct of clinical trials: a survey of current practices. 29. Kass NE, Faden RR, Goodman SN, Pronovost P, Tunis S Clin Trials 2011: 8: 342-9 Beauchamp TL. The research-treatment distinction: a probler 23. Little R, D'Agostino R, Cohen ML, et al. The prevention and atic approach for determining which activities should have ethi- eatment of missing data in clinical trials. N Engl J Med 2012: cal oversight. Hastings Cent Rep 2013; Spec No: S4-S15. Do10.1056/ NEJMsb1300760 24. Landray M), Grandinetti C, Kramer JM, Morrison BW, Ball L, Copyright O 2013 Massachusetts Medical Society N ENGLJMED 369: 11 NEJM ORC SEPTEMBER 12, 2013 1065

n engl j med 369;11 nejm.org september 12, 2013 1065 sounding board .mhra.gov.uk/home/groups/l-ctu/documents/websiteresources/ con111784.pdf). 18. European Medicines Agency. Reflection paper on risk based quality management in clinical trials: EMA/INS/GCP/394194/2011. August 4, 2011 (http://www.ema.europa.eu/docs/en_GB/document _library/Scientific_guideline/2011/08/WC500110059.pdf). 19. International Conference on Harmonisation Guideline for Good Clinical Practice E6(R1). June 1996 (http://www.ich.org/ fileadmin/Public_Web_Site/ICH_Products/Guidelines/Efficacy/ E6_R1/Step4/E6_R1__Guideline.pdf). 20. Grimes DA, Hubacher D, Nanda K, Schulz KF, Moher D, Alt￾man DG. The Good Clinical Practice guideline: a bronze stan￾dard for clinical research. Lancet 2005;366:172-4. 21. Baigent C, Harrell FE, Buyse M, Emberson JR, Altman DG. Ensuring trial validity by data quality assurance and diversifica￾tion of monitoring methods. Clin Trials 2008;5:49-55. 22. Morrison BW, Cochran CJ, White JG, et al. Monitoring the quality of conduct of clinical trials: a survey of current practices. Clin Trials 2011;8:342-9. 23. Little RJ, D’Agostino R, Cohen ML, et al. The prevention and treatment of missing data in clinical trials. N Engl J Med 2012; 367:1355-60. 24. Landray MJ, Grandinetti C, Kramer JM, Morrison BW, Ball L, Sherman RE. Clinical trials: rethinking how we ensure quality. Drug Inf J 2012;46:657-60. 25. Sherman RB, Woodcock J, Norden J, Grandinetti C, Temple RJ. New FDA regulation to improve safety reporting in clinical trials. N Engl J Med 2011;365:3-5. 26. Coxib and traditional NSAID Trialists’ (CNT) Collaboration. Vascular and upper gastrointestinal effects of non-steroidal anti-inflammatory drugs: meta-analyses of individual participant data from randomised trials. Lancet 2013 (http://www.thelancet .com/journals/lancet/article/PIIS0140-6736(13)60900-9/fulltext). 27. Guidance for Industry and Investigators. Safety reporting requirements for INDs and BA/BE studies. Rockville, MD: Food and Drug Administration, December 2012 (http://www.fda.gov/ downloads/Drugs/GuidanceComplianceRegulatoryInformation/ Guidances/UCM227351.pdf). 28. A new pathway for the regulation and governance of health research. London: Academy of Medical Sciences, January 2011. 29. Kass NE, Faden RR, Goodman SN, Pronovost P, Tunis S, Beauchamp TL. The research-treatment distinction: a problem￾atic approach for determining which activities should have ethi￾cal oversight. Hastings Cent Rep 2013;Spec No:S4-S15. DOI: 10.1056/NEJMsb1300760 Copyright © 2013 Massachusetts Medical Society