Examining statistical methods that leverage the hierarchical structure of adverse events for signal detection in randomised controlled trials

Examining statistical methods that leverage the hierarchical structure of adverse events for signal detection in randomised controlled trials

BackgroundWhen new drugs or medical interventions are developed, they are evaluated through clinical trials. During this process, data on the adverse events experienced by the trial participants is collected and can be analysed to help identify potential harms associated with the intervention. This is a difficult task due to the complex nature of the data : for each participant and each event type a multitude of outcomes are collected, including the presence of an event, the total number of occurrences, their duration, timing, and severity. But these rich datasets are often not used to their full potential. In addition, the current analysis practices used for adverse events outcomes are quite poor: previous reviews have identified overly simplistic approaches (such as simple frequency tables) and inadequate use of statistical methods. More sophisticated methods have been developed to address theses issues, including methods using groupings of adverse events (e.g. grouping all events linked with the cardiovascular system) instead of considering them individually. But there is a lack of evidence surrounding their use and they are not adopted in practice.Aims and objectivesWe plan to address this gap by appraising and comparing some of these methods. We previously conducted a methodological review to identify existing methods leveraging groupings of events [7]. We identified a total of 18 methods but we found little evidence that these methods were being adopted into practice. We concluded that further research was needed in the form of objective evaluations of these methods with recommendations for use for trial statisticians. We have undertaken a large-scale simulation study to objectively appraise and compare the selected methods by generating scenarios with known 'true' signals for adverse reactions. After evaluating the methods in a wide-range of scenarios with simulated data, we now plan to analyse a series of case-studies to test the selected methods on real-world data from completed clinical trials. This proposal aims to evaluate the performance of these methods on real-world data from completed clinical trials that evaluated the medicine Gilenya (fingolimod) to treat patients with relapsing multiple sclerosis. Gilenya was authorised in the European Union in 2011 and subsequently investigated by the European Medicines Agency because it was linked to increased risks of cardiovascular events, serious infections and a type of skin cancer. The objectives of the proposed research are to:(i) Re-analyse adverse events datasets from completed randomised controlled trials using the different statistical methods being investigated,(ii) For each trial considered, compare the performances of the different methods at the task of detecting signals for known adverse reactions, and(iii) Compare the results obtained from the new analyses with the results obtained in the original analyses and interpret the results in light of published evidence around the target drug.ImpactThe proposed project will produce evidence on the use of statistical methods for the detection of signals in final analyses of adverse events data from randomised controlled trials, as well as on the comparative performances of these methods. The results from these case studies will also be used to create recommendations for best practices in adverse events data analysis, which have the potential to improve the analysis of adverse events data in practice and reduce potential harms caused to future trial participants and patients.

[{ "PostingID": 4084, "Title": "NOVARTIS-CFTY720D2301", "Description": "A 24-month, Double-blind, Randomized, Multicenter, Placebo-controlled, Parallel-group Study Comparing the Efficacy and Safety of Fingolimod 1.25 mg and 0.5 mg Administered Orally Once Daily Versus Placebo in Patients With Relapsing-remitting Multiple Sclerosis" },{ "PostingID": 4085, "Title": "NOVARTIS-CFTY720D2302", "Description": "A 12-month Double-blind, Randomized, Multicenter, Active-controlled, Parallel-group Study Comparing the Efficacy and Safety of 0.5 mg and 1.25 mg Fingolimod (FTY720) Administered Orally Once Daily Versus Interferon ß-1a (Avonex) Administered im Once Weekly in Patients With Relapsing-remitting Multiple Sclerosis With Optional Extension Phase" },{ "PostingID": 4089, "Title": "NOVARTIS-CFTY720D2309", "Description": "24-month Double-blind, Randomized, Multicenter, Placebo-controlled, Parallel-group Study Comparing the Efficacy and Safety of 0.5 mg and 1.25 mg Fingolimod (FTY720) Administered Orally Once Daily Versus Placebo in Patients With Relapsing-remitting Multiple Sclerosis With Optional Extension Phase" },{ "PostingID": 4594, "Title": "NOVARTIS-CFTY720D2306", "Description": "A Double-blind, Randomized, Multicenter, Placebo-controlled, Parallel-group Study Comparing the Efficacy and Safety of 0.5mg FTY720 Administered Orally Once Daily Versus Placebo in Patients With Primary Progressive Multiple Sclerosis" },{ "PostingID": 4899, "Title": "NOVARTIS-CFTY720D1201", "Description": "A 6-month, Double-blind, Randomized, Placebo-controlled, Parallel-group, Multicenter Study Comparing Efficacy and Safety of FTY720 0.5 mg and 1.25 mg Administered Orally Once Daily in Patients With Relapsing Multiple Sclerosis" }]