Browse by author
Lookup NU author(s): Professor John Isaacs,
Professor Fiona Matthews
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0).
Whether described as stratified, precision or personalised medicine, all funders and deliverers of medical research and care are increasingly concerned with ensuring that the right patient receives the right therapy at the right time. The complexity of this evolving discipline requires the application of robust methodologies to disentangle the myriad potential markers and mechanistic pathways that might inform meaningful disease stratification. This document, applying across the related terms of stratified, precision and personalised medicine, attempts to aid investigators in addressing this complexity by providing a methodological framework for the design, conduct, analysis and interpretation of research attempting to divide complex patient groups into sub-classes (or strata), as defined by differences in mechanism, disease course, risk of developing disease or response to therapy. Stratification of complex, heterogeneous patient groups is possible through measurement of traits assessed through any number of modalities – genetic, biochemical, imaging, clinical scores, behavioural/ psychological assessment etc. Strata are increasingly able to be defined by multiple variables, measured through multiple modalities. It is the intent of this Framework to consider stratification by any means, and is not limited to the traditional genetic or biochemical variables. Despite the strategic importance of the field, there remains a lack of clarity in the stratified medicine research community around best practice in study design, analysis and reporting. This Framework will focus on particular issues and methodological challenges in: • The discovery and verification of stratifying biomarkers • The definition of strata by integration of marker information • The methodologies for design and development of diagnostics to facilitate this • Methods for deriving mechanistic insight from these markers/strata • Provision of the requisite level of evidence to inform development of diagnostics and trials able to test new stratifying hypotheses While there are numerous methodological challenges in the design of stratified trials and in the analysis of the health economics of stratification, these issues are not explored in depth in this Framework, but are considered as key downstream drivers. This Framework: Presents a pathway for stratified medicine research that starts with the end in mind and covers stratum discovery, verification and early clinical assessment Highlights to stratified medicine Investigators the critical questions that they should ask themselves when designing studies, and illustrates potential design pitfalls and sources of bias Describes the suite of methodological approaches and resources available to inform study design and analysis, drawing on appropriate guidance and illustrative case studies Enables an investigator working at one phase of the stratified medicine pipeline to understand the outputs and requirements of, and iterative interrelations between, the up- and down-stream phases, such that they can tailor their own work to these requirements 2 Framework for the Development Design and Analysis of Stratified Medicine Research The Framework is structured into six themes: • Theme 1: Framing the Question/Defining the Population • Theme 2: Designing Stratum Discovery Studies; selecting variables, defining response and powering • Theme 3: Assay Design; managing complexity and variability • Theme 4: Defining Strata; data integration, linkage to existing knowledge, linkage to outcome • Theme 5: Stratum Verification • Theme 6: Progression Towards Clinical Utility Ultimately, this Framework aims to illustrate to investigators the various phases of stratified medicine research, the challenges to be faced, the sources of bias and design flaws to be avoided and examples of good practice to be considered. The Framework is directed to enable optimisation of stratified medicine research design in order to increase the likelihood of translation to impact; to ensure that the right patient will receive the right therapy at the right time.
Author(s): Crosby D, Bossutt P, Brockehurst P, Chamberlain C, Dive C, Holmes C, Isaacs J, Kennedy R, Matthews F, Parmer M, Pearce J, Westhead D, Whittaker J, Holgate S
Publication type: Online Publication
Publication status: Published
Acceptance date: 01/11/2017
Publisher: Medical Research Council
Place Published: London
Access Date: 06/02/2018
Type of Medium: Research framework