CTSU Clinical Trial Service Unit & Epidemiological Studies Unit University of Oxford Nuffield Department of Population Health 27 August 2014 Dr Fiona Godlee Editor, BMJ BMJ Publishing Group Tavistock Square London WC1H 9JR CTSU, Richard Doll Building, Old Road Campus, Roosevelt Drive, Oxford OX3 7LF, UK Tel: +44-(0)-1865-743743 Fax: +44-(0)-1865-743985 Website: www.ctsu.ox.ac.uk Dear Fiona Plans to augment CTT collaborative meta-analyses of the effects of statin therapy on vascular events, mortality and cancer by seeking data on other types of adverse event Thank you for acknowledging in your email that the work done by the Cholesterol Treatment Trialists’ (CTT) collaboration since it was initiated in the mid-1990s has produced evidence about the efficacy and safety of statin therapy which would not otherwise have emerged and that has helped to inform doctors and their patients. The coordination of this project by the CTT secretariat in Oxford and in Sydney over the past 2 decades has been funded entirely by non-pharmaceutical sources (the Medical Research Council and British Heart Foundation in the UK, and equivalent government and charity research funders in Australia). With respect to your specific questions: Data currently available to the CTT collaboration: Your understanding is now correct that the outcome data obtained on individual patients in the trials that have contributed to the CTT meta-analyses are currently limited to cause-specific mortality, major vascular events and site-specific cancers. The rationale for focussing on these events is given in the published protocol for this project which was finalised before any of the results of those trials had been reported [1]. Detailed analyses of these data have been published prominently in a series of articles, often with extensive webtables and figures (as in the most recent CTT collaborators report [2]), in order to help inform treatment decisions. In the debate about side-effects of statins, confusion has been caused by calls for the CTT collaboration to make available data on other types of adverse event (for example, muscle symptoms) that it does not hold and by false allegations that the coordinating team for the CTT collaboration were withholding such data. It is helpful that the BMJ will now correct the record by making it clear that such allegations were without foundation. Basis on which data were provided for the CTT collaboration: Your understanding is correct that these data on deaths, vascular events, and cancers were sought for this collaborative effort from academic investigators who conducted contributing trials and/or pharmaceutical companies that funded them with the explicit assurance that their data would be kept secure and would not be shared with third parties without their permission. By so doing, it has been Co-directors: Sir Rory Collins FMed Sci FRCP BHF Professor of Medicine and Epidemiology Sir Richard Peto FRS Hon FRCP Professor of Medical Statistics and Epidemiology Jane Armitage FFPH FRCP Professor of Clinical Trials & Epidemiology Colin Baigent FFPH FRCP Professor of Epidemiology Louise Bowman MRCP MD Senior Research Fellow Zhengming Chen MBBS DPhil Professor of Epidemiology Michael Clarke DPhil Professor of Clinical Epidemiology Robert Clarke FFPH FRCP Reader in Epidemiology Sarah C Darby PhD Professor of Medical Statistics Christina Davies BMBCh MSc Senior Research Fellow Richard Gray MA MSc Professor of Medical Statistics Alison Halliday FRCS Professor of Vascular Surgery Michael Hill DPhil Laboratory Scientific Director Martin Landray PhD FRCP Reader in Epidemiology Michael Lay DPhil Head of Project Information Science Christine Marsden PhD Unit Administrator Sarah Parish DPhil Senior Research Fellow Max Parkin MD Honorary Senior Research Fellow David Simpson OBE Hon MFPH Director, IATH Alan Young DPhil Director of Information Science possible to obtain individual patient data for almost all of the relevant randomised trials. This has allowed analyses to be conducted that provide more reliable evidence about the effects of statin therapy on these outcomes in a wider range of circumstances (for example, older patients, women, primary prevention) than could be provided by any of the individual trials. Future plans to obtain data on other types of adverse event: Building on the success of this collaborative process, we hope that it will now be possible to obtain data on the other types of adverse event that occurred in these trials. This proposal was included in our submission to the core funders (MRC, BHF and CR-UK) of the Clinical Trial Service Unit in early 2013 and was endorsed at the annual meeting of the CTT collaborative group in November 2013. However, it should be understood that such an effort is likely to be non-trivial: whereas the previous work of the collaboration involved only a limited number of outcome types that are generally defined similarly (i.e. heart attack, stroke, cause of death, site of cancer), what it is now proposed be sought is all the types of adverse event recorded in these trials. In view of the likely heterogeneity and complexity of these data, we intend first to seek detailed tabular data on all adverse events recorded in each trial prior to developing more detailed plans for seeking the relevant individual patient data. In that way, we would hope that the collaborative group will help address the current levels of misinformation about the balance of benefits and risks of statin therapy, and contribute further to providing reliable evidence that doctors can use to advise their patients. Yours sincerely Colin Baigent Jane Armitage Rory Collins 1. Cholesterol Treatment Trialists’ (CTT) Collaboration. Protocol for a prospective collaborative overview of all current and planned randomized trials of cholesterol treatment regimens. Am J Cardiol 1995; 75: 1130-34 2. Cholesterol Treatment Trialists’ (CTT) Collaborators. The effects of lowering LDL cholesterol with statin therapy in people at low risk of vascular disease: meta-analysis of individual data from 27 randomised trials. Lancet 2012; 380: 581-90 2