Drug Prescription Based on Predicted Treatment Responsiveness: An Application to the PROSPER Randomised Controlled Trial

Background: Clinical guidelines are often based on risk stratified outcomes of randomised controlled trials, assuming that patients with a similar risk respond to the treatment alike. There is a need to move beyond this paradigm, as patients at similar risk may respond differently to the same treatment due to differences in underlying risk factors. Methods: We reanalysed the PROSPER randomized trial, which showed an overall benefit of pravastatin treatment compared to placebo in old age individuals at risk of vascular diseases. We derived and validated a model predicting individualised responsiveness to pravastatin on the primary outcome composed of coronary death, non-fatal myocardial infarction, and fatal or non-fatal stroke. We evaluated the efficacy of three treatment strategies where pravastatin was prescribed to (i) all older individuals; (ii) older individuals at elevated risk given a previous history of vascular diseases; or (iii) older individuals with a favourable prediction of treatment response. Findings: According to the prediction model, 86.1% of individuals were expected to benefit from pravastatin. Individuals with a favourable treatment response were more likely to be male, non-smoking, hypertensive, and to have a history of vascular diseases but no diabetes. The hazard ratios associated with the three treatment strategies were: 0.85 (95% CI: 0.75 to 0.98) when prescribing pravastatin to all older individuals; 0.88 (95% CI: 0.80 to 0.97) when treating only individuals with an elevated vascular risk; and 0.81 (95% CI: 0.71 to 0.91) when restricting pravastatin to individuals with a favourable prediction of treatment response. Interpretation: Prescribing pravastatin based on predicted treatment responsiveness may be superior to a strategy where pravastatin is prescribed to all older individuals or based on vascular disease risk. Trial Registration: Trial registration: ISRCTN40976937 (https://www.isrctn.com/ISRCTN40976937) Funding Information: University of Copenhagen; Netherlands Organisation for Health Research and Development; EU Horizon 2020; British Heart Foundation Centre of Research Excellence; Novo Nordisk Fonden. Declaration of Interests: N.S. has consulted for Afimmune, Amgen, AstraZeneca, Boehringer Ingelheim, Eli Lilly, Hanmi Pharmaceuticals, Merck Sharp & Dohme, Novartis, Novo Nordisk, Pfizer, and Sanofi; and received grant support paid to his University from AstraZeneca, Boehringer Ingelheim, Novartis, and Roche Diagnostics outside the submitted work. All the other authors declare no conflicts of interest. Ethics Approval Statement: The PROSPER trial complied with the Declaration of Helsinki; written informed consents were obtained. The ethics committees of all centres approved the original study (Greater Glasgow Community/Primary Care Local Research Ethics Committee; Dumfries and Galloway Health Board Local Research Ethics Committee; Argyll and Clyde Health Board Local Research Ethics Committee [reference LREC 85/97]; Lanarkshire Research Ethics Committee [reference ER/9/97/37]; Research Ethics Committee of the Cork Teaching Hospitals [CREC]; Medical Ethical Committee (METc) of the Leiden University Medical Center.