Asthma is a heterogeneous disease with diverse etiological drivers. Precision medicine (PM) trials aim to demonstrate differential clinical efficacy for specific endophenotypes. Such approaches are reliant on well-powered biomarker-defined subgroup analyses. To Identify and enroll eligible patients, based on a biomarker of interest, studies need to screen a large number of prospective patients.

Many sites participate in multiple asthma studies simultaneously. The objective of this work was to assess the potential impact a PM study could have towards recruitment times, and the statistical power, of other non-PM studies at sites that seek to recruit asthma patients into both PM and non-PM studies.  

Modelling and simulations were used to approximate the impact of preferentially selecting biomarker-positive individuals for PM-studies. Both Forced Expiratory Volume (FEV1) and Severe Exacerbation rates (SevEx) were considered as endpoints and the impact on statistical power was quantified with respect to a hypothetical Phase 3 trial.

Based on parametrizing the model against in-house respiratory trial data, we provide a framework that quantifies the relationship between the reduction in statistical power of a Phase 3 trial and the number of participants that can be preferentially selected from non-PM studies. In these analyses we show SevEx to be more sensitive to biomarker*treatment response interactions than FEV1. Assuming a Ph3 biomarker*treatment response interaction <10%, a Phase 3 trial could tolerate (i.e. <5% drop in power) up to a 15% loss of biomarker-positive patients.

This approach provides a framework to help quantify the impact PM studies may have on the recruitment and statistical power of other non-PM studies.