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, 2013T 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