As the demand for personalized therapies continues to grow, trial designs must evolve to reflect the biological variability between patients. Traditional models often overlook subtle, yet important, differences in how individuals respond to treatment. Conversely, patient stratification into different subgroups based on shared characteristics can greatly improve trial efficiency and effectiveness.
In precision medicine trials, stratification is often dependent on biomarkers, which could be genetic mutations and variants, molecular and histological subtypes, and other clinical data points that could be associated with disease outcome or treatment response. A meta-analysis of phase II clinical trials in oncology revealed that the use of biomarkers for treatment selection led to significant increases in response rate, progression-free survival, and overall survival.
The use of biomarkers across trial stages and purposes
Biomarkers can be integrated throughout the trial, depending on the stage and study objectives. They can be used as inclusion criteria during patient screening and recruitment, as a tool for allocating patients to specific treatment arms during randomization, and as a way to evaluate treatment response during and/or at the end of the trial.
Biomarkers can also serve distinct functional purposes, each contributing uniquely to trial design and interpretation.
- Diagnostic biomarkers are used to confirm the presence of a disease or condition. For example, evaluating levels of Aβ and phosphorylated tau using PET imaging or cerebrospinal fluid (CSF) analysis can be used to select specific pathologies for Alzheimer’s disease trials.
- Prognostic biomarkers provide information about the likely disease course, regardless of the intervention. Prognostic biomarkers are most commonly used in oncology trials, with high specificity of biomarkers to specific cancers. For instance, high levels of carcinoembryonic antigen (CEA) and prostate-specific antigen (PSA) are associated with poorer prognosis in colorectal cancer and prostate cancer, respectively.
- Predictive biomarkers help identify which patients are more likely to benefit from a particular therapy. Genetic alterations are commonly used to predict response to treatment, especially in oncology trials. This is particularly valuable for allocating patients to the appropriate intervention group, since targeted therapy often entails inhibiting oncogenes that are overly active due to the genetic alteration.
- Pharmacodynamic or response biomarkers measure biological responses to a therapeutic intervention to determine efficacy. This can occur throughout the trial, particularly in adaptive trials where patients could switch to another intervention group if there is a lack of response, or as a trial endpoint.
- Safety biomarkers indicate the likelihood, presence, or extent of adverse effects. Depending on the specific toxicities and adverse effects associated with each intervention, these biomarkers could include testing blood biomarkers for liver and kidney function, among others.
Understanding the purpose and timing of biomarker use is essential for designing trials that are not only more targeted, but also more efficient and ethical for patients.
The rise of digital biomarkers
Interestingly, digital biomarkers have gained traction in recent years. These can include any physiological and behavioral data collected via a digital device, including but not limited to wearables. The use of digital devices allows for real-time, continuous data capture in everyday environments, therefore providing access to an additional layer of information on patient health.
A scoping review of randomized controlled trials (RCTs) that integrated digital biomarkers showed that 77% of the trials used them as interventions and 71% of trials used a wearable device for measurements. The most common therapeutic areas represented by the RCTs were cardiovascular (61%), respiratory (29%), and endocrine, nutritional, or metabolic diseases (23%). This distribution aligns with the primary data types collected, such as heart rate and physical activity, which are particularly relevant in these conditions. Physical activity was also frequently utilized in trials on nervous system diseases which represented 13% of trials, where mobility and functional status are critical indicators of disease progression and treatment response.
By tailoring trial design to patient biology, stratification and biomarker integration can significantly improve both the efficiency and ethical integrity of clinical research. Whether through traditional biomarkers or emerging digital measures, these strategies support more precise patient selection, better outcome assessment, and reduced exposure to ineffective treatments. Their use is key to advancing a more patient-centered and responsive model in precision medicine.
.png)
