Bureaucratic obstacles to clinical trials: in the public interest? Rory Collins BHF Professor of Medicine & Epidemiology Clinical Trial Service Unit & Epidemiological Studies Unit University of Oxford

Stated aim of EU Directive on Clinical Trials Article 1: "Compliance with this good practice provides assurance that the rights, safety and well-being of trial subjects are protected, and that the results of the clinical trials are credible." What is the evidence for this claim?

Criteria for a good trial Ask an IMPORTANT question Answer it RELIABLY

Unanticipated consequences of good intentions MODERATE effects of treatments can have LARGE effects on public health RELIABLE assessment of MODERATE effects on mortality and major morbidity requires LARGE RANDOMISED trials Consequently, bureaucratic obstacles to LARGE RANDOMISED trials may well inadvertently REDUCE public safety

ISIS-2: 2 x 2 “factorial” study of iv streptokinase and of oral aspirin in acute MI (17,000 patients)

Meta-analysis of small fibrinolytic trials ( )

Pharmacovigilance in randomised trials But, reliable assessment of MODERATE effects on COMMON outcomes does need LARGE-SCALE RANDOMISED evidence – but this does not require randomisation LARGE effects on RARE outcomes may be detected by the reporting of Suspected Unexpected Serious Adverse Reactions (SUSARS) required in trials by regulators

RELIABLE assessment of MODERATE hazards: large-scale randomised meta-analysis of the effects of COX-2 inhibitors on major vascular events Coxib vs. No. of trials Events/person-years Allocated CoxibOther Ratio of annual event rates Coxib : Other Placebo / /8689 (1·8%/y)(1·2%/y) 1·41 ( SE 0·14) Naproxen / /10755 (1·1%/y)(0·7%/y) 1·56 ( SE 0·17) Non-naproxen NSAID / /12098 (0·9%/y)(1·0%/y) 0·86 ( SE 0·12) Allocated 0·250·51·02·55·010 Coxib betterOther better Heterogeneity:  2 2 = 12·9; p = 0·002 0·1

CR-UK assessment of impact of Clinical Trials Directive on UK non-commercial cancer trials (European Journal of Cancer 2006) Doubling in costs of running non-commercial cancer trials and 6-12 month delays to starting Major concerns about correct interpretation due to lack of central guidance, lack of clarity regarding interpretation of guidance notes, and increased documentation Clinical trial units unable or unwilling to start in non-UK centres due to different interpretations in different European countries

UK NHS R&D: Increasing trial approval times Days Measure Notification to R&D sign off R&D sign off to first patient visit First submission to first patient visit Report for the Ministerial Industry Strategy Group (2008)

Over-interpretation of NHS Research Governance requirements for pharmacovigilance “… you must report to the Research Management & Governance Unit any serious adverse event occurring during the study….” quote from the Research Governance lead of a local Primary Care Trust

Cost per patient in phase III trial with 10 visits over 10 weeks Report for the Ministerial Industry Strategy Group (2008) Quote from Bob Temple (US FDA): “EU Directive on Clinical Trials is Europe’s gift to the USA”

Global patient enrolment: 2000 vs 2006 EU Core Europe Non-Core EU Accession % 8% 4% % 13% 7% North America % % Latin America % % ME & Africa % % SE Asia & W Pacific % % UK % % Report for the Ministerial Industry Strategy Group (2008)

ICH GCP: Guidance on monitoring “… extent and nature of monitoring should be based on considerations such as the objectives, purpose, design, complexity, blinding, size and endpoints of the trial. In general there is a need for on-site monitoring before, during and after the trial; …[but] central monitoring…can assure appropriate conduct of the trial in accordance with GCP” ICH GCP Section

Central monitoring by coordinating centre Record checks for: Patient eligibility (eg, pathology report to substantiate diagnosis) Patient existence (eg, ONS flagging or imaging investigation) Outcome (eg, ONS flagging for death; investigation results) Statistical checks for: Missing or invalid data (eg, range checks) Calendar checks (eg, dates of recruitment) Unusual patterns (eg, digit preference, rounding or unusual frequency distribution) Reporting rates (eg, frequency of adverse events or missing data) Repeated measures (eg, variability and within-individual changes) MRC/DH joint project (

ESPS-2: Potential problem at one site identified by central statistical monitoring Unusually rapid recruitment Regularity of follow-up visits (inc. weekends & holidays) Reduced variability of data (eg, BP) Better study treatment compliance Lower incidence of adverse events Poor correlation between measurements at different follow-up visits

ESPS-2: Confirmation of problem by central assay of blood samples Both DIP & ASA were in all samples (rather than expected 25%) DIP levels were either trace or very high (& different from other centres) ASA levels were incompatible with low-dose aspirin regimen Plasma protein polymorphism measurement showed each of 90 samples was from same mixture of plasma from several persons

ESPS-2: Failure to detect fraud during special audit site visit by Good Clinical Research Practices Ltd “The auditors were not able to uncover major anomalies in the data collected by this centre.” ESPS-2 report, June 1996

Claims of risks to patients in clinical trials “Trials of new medicines ….. are so badly flawed that they endanger the health of the patients ….. according to scientists who have been auditing them in confidence for 10 years ….. Good Clinical Research Practices is called in by pharmaceutical companies to establish whether trials meet international standards” Guardian July 1999

“Possible” conflicts of interest Petitions for compulsory winding-up Week ending 18 June 1999 ( 14/07/99 Good Clinical Research Practices Ltd

“….. fraud in clinical trials is so rare and ….. generally inconsequential, that the public may be far more misguided by studies that are poorly designed, wrongly analysed and inappropriately reported than by fraud” ISCB subcommittee on fraud Stat Med 1999

Report of the ISCB subcommittee on misconduct in clinical trials (Statistics in Medicine 1999) Misconduct is unlikely to affect study results if any of the following conditions hold: The misconduct is limited to a few investigators (eg, one centre in a multicentre setting) and/or to a few data items; The misconduct bears on secondary variables that have little or no effect on the primary endpoint of the study; or The misconduct affects all treatment groups equally, and hence does not bias the results of the study. NB: Misconduct committed without regard to the treatment assignments (for example, prior to randomization or in double blind trials) generates noise but no bias.

“The growing number of regulations ….. may also have the unintended consequence of making trials ever more complex ….. such complexity may be counterproductive and may pave the way to fraud” ISCB subcommittee on fraud Stat Med 1999

Proliferation of laws and guidelines may make clinical research LESS reliable (and so HARM, not help, patients) Second Conference on Sensible Guidelines for the Conduct of Clinical Trials 5-6 September 2009 St Anne’s College, Oxford For details, contact: Proceedings of First Conference: Clinical Trials 2008; 5 (1): What is now needed urgently is a more streamlined, risk-based, approach to the design, conduct and monitoring of trials