Improving clinical trial efficiency using a machine learning‐based risk score to enrich study populations

Abstract

Aims Prognostic enrichment strategies can make trials more efficient, although potentially at the cost of diminishing external validity. Whether using a risk score to identify a population at increased mortality risk could improve trial efficiency is uncertain. We aimed to assess whether Machine learning Assessment of RisK and EaRly mortality in Heart Failure (MARKER‐HF), a previously validated risk score, could improve clinical trial efficiency. Methods and results Mortality rates and association of MARKER‐HF with all‐cause death by 1 year were evaluated in four community‐based heart failure (HF) and five HF clinical trial cohorts. Sample size required to assess effects of an investigational therapy on mortality was calculated assuming varying underlying MARKER‐HF risk and proposed treatment effect profiles. Patients from community‐based HF cohorts (n = 11 297) had higher observed mortality and MARKER‐HF scores than did clinical trial patients (n = 13 165) with HF with either reduced ejection fraction (HFrEF) or preserved ejection fraction (HFpEF). MARKER‐HF score was strongly associated with risk of 1‐year mortality both in the community (hazard ratio [HR] 1.48, 95% confidence interval [CI] 1.44–1.52) and clinical trial cohorts with HFrEF (HR 1.41, 95% CI 1.30–1.54), and HFpEF (HR 1.74, 95% CI 1.53–1.98), per 0.1 increase in MARKER‐HF. Using MARKER‐HF to identify patients for a hypothetical clinical trial assessing mortality reduction with an intervention, enabled a reduction in sample size required to show benefit. Conclusion Using a reliable predictor of mortality such as MARKER‐HF to enrich clinical trial populations provides a potential strategy to improve efficiency by requiring a smaller sample size to demonstrate a clinical benefit.

Artificial intelligence and heart failure: A state‐of‐the‐art review

“…ML algorithms can also help identify patients at various levels of risk for endpoints used in clinical trials which can improve trial efficiency and enrich study populations. 88 In contrast to trials examining interventions in HF, trials evaluating AI algorithms in HF patients are relatively new and often require prospective validation in external datasets. Hence, the interpretation of trial results must be made with caution.…”

“…This can also enable trialists to take prompt decisions about altering or stopping the trial to reduce the risk of potential harm to participants. ML algorithms can also help identify patients at various levels of risk for endpoints used in clinical trials which can improve trial efficiency and enrich study populations 88 . In contrast to trials examining interventions in HF, trials evaluating AI algorithms in HF patients are relatively new and often require prospective validation in external datasets.…”

August 2023 at a glance: Focus on epidemiology and medical therapy

“…Machine learning is now often used for risk stratification and possibly improve clinical trials' efficiency. [16][17][18] This technique was applied to echocardiographic parameters to predict the effects of spironolactone in patients at risk for HF investigated in the HOMAGE trial. 19 Different echocardiographic phenotypes were identified.…”

December 2023 at a glance: Focus on medical therapy in chronic and acute heart failure

“…Machine‐learning procedures have been successfully used to predict prognosis 36 . The InterTAK‐ML is a machine‐learning model developed to predict in‐hospital mortality in 3482 patients with takotsubo syndrome.…”