ReviewPost screenDefining drug response for stratified medicine
Introduction
Stratified medicine is the differential prescribing of medications, or treatment programs, to groups of individuals based on attributes other than the symptoms resulting from their disease [1] (Fig. 1). The term is often considered synonymous with ‘personalised’, ‘precision’, or even ‘P4’ medicine. Stratification could be seen as less ambitious, and more immediately realisable, than the other approaches because of its focus on identifying groups needing particular treatments rather than the direct optimisation of treatment for each individual patient [2]. Its purpose is to improve outcomes by refining drug dosages or by administering more appropriate treatments to specific groups of patients. It requires the identification of appropriate subgroups of individuals, through the use of biomarkers, and suitable measures of the benefits and costs of the potential treatments for each group. The advantages of stratified medicine flow from the existence of subgroups, or strata, of patients, such that individuals within the same subgroup tend to have similar responses to treatment, whereas those in different strata respond differently and require different treatment.
The evaluation of variability in individuals’ responses to treatments, in terms of both efficacy and safety (Fig. 2), is central to assessing potential benefits from stratified medicine. Although this requirement is obvious, fulfilling it is more difficult than is generally appreciated. A lack of clarity about the extent and causes of variability in the effects of a particular treatment can result in efforts being focussed on investigations of inappropriate or irrelevant biomarkers. Therefore, in this position paper, we consider issues relating to three areas with respect to drug response: (i) the variability of drug outcomes (in terms of both efficacy and adverse effects); (ii) the identification of subgroups that show differential efficacious responses; and (iii) the identification of subgroups most at risk of adverse drug outcomes.
This position paper was developed out of discussions during a workshop on ‘Defining Drug Response for Stratified Medicine’ held by the UK Pharmacogenetics and Stratified Medicine Network.
Section snippets
Variability in drug response
In 1997, Sir Richard Sykes predicted:
‘it will soon be possible for patients in clinical trials to undergo genetic tests to identify those individuals who will respond favourably to the drug candidate, based on their genotype. This will translate into smaller, more effective clinical trials with corresponding cost savings and ultimately better treatment in general practice. Individual patients will be targeted with specific treatment and personalised dosing regimens to maximise efficacy and
Efficacy and drug response
If efficacy is the beneficial response from treatment, then trials and retrospective studies provide estimates of the efficacy of specific doses of drugs for particular sets of individuals. The use of their results in traditional, ‘non-stratified’, medicine requires two sets of decisions: (i) what treatment would be most appropriate for individuals similar to those that were studied?; and (ii) what wider group of patients can be considered sufficiently similar to the study population to be
Defining drug efficacy
Hard endpoints, such as all-cause mortality or rates of cardiovascular death, are the simplest outcomes of drug action to interpret. However, very large studies with lengthy follow-up can be necessary to produce clear results from such outcomes. Therefore, surrogate endpoints are often used to evaluate the efficacy of drugs. Many of these surrogates, such as blood pressure, glycosylated haemoglobin and low-density lipoprotein (LDL) cholesterol, are measured on continuous scales. The use of such
The use of efficacy in stratification
As well as the issues that concern the definition of drug efficacy per se, there are further ones that affect the identification of strata, the groups of individuals who respond to treatment in distinct ways. Most of the issues around the definition of efficacy are exactly the same for stratified medicine as for any other decisions about treatment. Stratification can almost be considered as an add-on, taking the clinically relevant definition of efficacy and applying it to the splitting of the
Stratification to reduce adverse reactions to drugs
Medicines are not always effective in improving health: some patients have adverse reactions to drugs. These cause 6.5% of admissions into hospital [17], including 2.5% emergency admissions [18]. 14.7% of inpatients develop an adverse drug reaction (ADR) in hospital [19], and 25% of primary care consultations involve patients seeking help for an ADR [20].
ADRs can be classified as on target (predictable from the known primary or secondary pharmacology of the drug, and with a clear
Sources of variability in drug safety studies
Many of the issues discussed above with regard to drug efficacy studies are also important for drug safety. In addition, other specific areas also need to be considered for drug safety.
Concluding remarks
Many areas, such as cancer, and inflammatory and rare diseases, have made great progress in stratifying patients’ treatment to deliver ‘the right drug, at the right time, at the right dose’. Grouping patients into strata based on their genetic profile, or molecular basis of their disease, then treating them based on their strata (rather than the symptoms they present) has already improved healthcare. Not only do patients recover more quickly, but there is also the potential to improve drug
Acknowledgements
The article is based on a workshop held by the UK Pharmacogenetics and Stratified Medicine Network (www.uk-pgx-stratmed.co.uk/), which was funded by the Knowledge Transfer Network and Innovate UK at the time of the workshop. M.P. is an NIHR Senior Investigator and acknowledges the support of the MRC Centre for Drug Safety Science and Wolfson Centre for Personalised Medicine, both in Liverpool. R.S. is also an NIHR Senior Investigator, and acknowledges the support of the NIHR Liverpool Pancreas
References (35)
A review of blood pressure measurement protocols among hypertension trials: implications for ‘evidence-based’ clinical practice
J. Am. Soc. Hypertens.
(2014)A regression model for longitudinal change in the presence of measurement error
Ann. Epidemiol.
(2002)- et al.
Adverse drug reactions: definitions, diagnosis, and management
Lancet
(2000) Phenotype standardization for drug-induced kidney disease
Kidney Int.
(2015)Realising the Potential of Stratified Medicine
(2013)Stratified, Personalised or P4 Medicine: A New Direction for Placing the Patient at the Centre of Healthcare and Health Education
(2015)The Pharmaceutical Industry in the New Millennium: Capturing the Scientific Promise
(1997)Individual response to treatment: is it a valid assumption?
BMJ
(2004)How many repeated measures in repeated measures designs? Statistical issues for comparative trials
BMC Med. Res. Methodol.
(2003)Mastering variation: variance components and personalised medicine
Stat. Med.
(2016)
Individual therapy: new dawn or false dawn?
Drug Inf. J.
When a randomised controlled trial is needed to assess drug safety. The case of paediatric ibuprofen
Drug Saf.
HLA-B*5701 screening for hypersensitivity to abacavir
N. Engl. J. Med.
Determining the optimal dose in the development of anticancer agents
Nat. Rev. Clin. Oncol.
Medication adherence with diabetes medication: a systematic review of the literature
Diabetes Educ.
Statistical Issues in Drug Development
Genome-wide association of lipid-lowering response to statins in combined study populations
PLoS ONE
Cited by (22)
Precision antiplatelet therapy
2023, Research and Practice in Thrombosis and HaemostasisThe gut microbiome–Does stool represent right?
2023, HeliyonDoes understanding endotypes translate to better asthma management options for all?
2019, Journal of Allergy and Clinical ImmunologyCitation Excerpt :The logical consequence of asthma heterogeneity is that the efficacy and safety of different asthma drugs will vary considerably between groups of patients, and that the treatment that might be effective in one group may have no effect, or may even be harmful in other groups. However, the stepwise approach recommended by most asthma guidelines is an archetype of the traditional approach to patient treatment (Fig 4).21 Failure to gain symptom control usually leads to a rightward shift in asthma treatment, with an increase in dose and/or number of drugs prescribed to a patient.
Omics in precision medicine
2023, Oncology: Genomics, Precision Medicine and Therapeutic Targets