Biomarkers in clinical trials: what they are and why they matter

What is a biomarker?

A biomarker is a defined characteristic that is measured as an indicator of normal biological processes, pathogenic processes or responses to an exposure or intervention, including therapeutic interventions. Molecular, histologic, radiographic and physiologic characteristics are all types of biomarkers^[1].

A biomarker is not an assessment of how an individual feels, functions or survives^[1]. It is an objective measure that captures what is happening in a cell or an organism at a given moment^[2].

Many biomarkers come from simple measurements made during a routine doctor visit, such as blood pressure or body weight. Others are based on laboratory tests of blood, urine or tissues, and some capture changes at the molecular and cellular level by looking at genes or proteins^[2]. An increased level of cholesterol in the blood, for example, is a biomarker for heart attack risk^[3].

Why do biomarkers matter in research and in trials?

Biomarkers help researchers and doctors track healthy processes, diagnose diseases and other health conditions, monitor responses to medication, and identify health risks in a person. Clinicians use them to detect early brain changes, understand how risk factors are involved, identify participants who meet particular requirements for clinical trials and studies, and track participants' responses to a test drug or other intervention, such as physical exercise^[3].

Scientists also use biomarkers to better understand fundamental biological processes and to turn research findings into practical medical and public health applications^[2]. Biomarkers and target-specific phenotypes are important to targeted drug design and individualised medicine, and constitute an important aspect of modern pharmaceutical research and development^[4].

What are the seven categories of biomarkers?

The Biomarkers, EndpointS and other Tools (BEST) glossary defines seven biomarker categories: susceptibility/risk, diagnostic, monitoring, prognostic, predictive, pharmacodynamic/response, and safety. A full biomarker description includes the biomarker name, the source or matrix, the measurable characteristic(s), and the analytic method used to measure the biomarker^[1].

A biomarker may be a single characteristic or a panel of multiple characteristics^[1].

Category	What it indicates
Susceptibility/risk	Likelihood of developing a condition
Diagnostic	Presence of a condition
Monitoring	Change in a condition over time
Prognostic	Likely course or outcome
Predictive	Likely response to a specific treatment
Pharmacodynamic/response	Biological response to an exposure or intervention[1]
Safety	Risk of harm from an intervention[1]

What is the difference between a prognostic and a predictive biomarker?

Among the seven categories, two are most relevant when you join a study: prognostic and predictive biomarkers. Both belong to the BEST (Biomarkers, EndpointS, and other Tools) glossary framework for describing biomarkers, and both describe the relationship between a measurable characteristic and what happens next^[1].

In studies of environmental disease, scientists classify subgroups of biomarkers within complex disease types, such as cancer, in order to develop and target appropriate treatments. NIEHS supports the development of biomarkers that measure susceptibility – variations in genes that place a person at higher risk to health effects from an environmental exposure^[2].

What are biomarker-driven trials and companion diagnostics?

Biomarker-driven trial designs allow therapies to be tested in participants selected by a measurable characteristic rather than by a clinical diagnosis alone. Effective therapies are likely to be characterised by combination treatments, sequences of treatments, and treatments matched to specific patient subgroups. Innovative trial designs are well suited to measure treatment benefit for such therapies. Innovative clinical trial designs include adaptive, platform and sequential multiple assignment randomised trials^[5].

An example from early-stage research shows how a biomarker can inform drug development: selective calpain-2 inhibitors block a pathway that leads to neuronal damage, and an identified blood biomarker reflects brain activation of this pathway, enabling determination of drug effectiveness^[5].

A companion diagnostic is an assay developed alongside a therapeutic. Co-developing a diagnostic alongside its paired therapeutic, rather than after the drug is already approved, shapes both the clinical trial design and the regulatory timeline^[6].

From a research finding to a clinical-grade test

A discovery-stage assay and a clinical-trial-ready assay are not the same thing; moving between them, "locking down" the assay, is its own distinct development stage^[6].

A spatial biomarker discovered in a research setting is typically optimised iteratively, with a research team free to adjust staining conditions, antibody concentrations and analysis parameters as they refine the finding. A companion diagnostic assay cannot work this way; every parameter has to be fixed, documented and then left alone, precisely because a validated assay's performance characteristics only mean something if the assay itself stops changing^[6].

Research-use-only application of multiplex immunofluorescence in translational research lacks the rigorous controls and standardisation needed to support the stringent reproducibility, sensitivity and specificity requirements that same assay faces once it is intended for clinical trial use. Once an assay is locked down, analytical validation asks whether this exact, fixed protocol produces consistent, accurate results, and whether it keeps doing so across the range of conditions a real clinical laboratory will actually encounter^[6].

The realistic goal of cross-site validation is not perfect agreement between laboratories, but a quantified, documented benchmark for how much variability is acceptable before a result should no longer be trusted^[6].

How are biomarkers qualified for drug development?

When a biomarker is qualified, it means that it has undergone a formal regulatory process to ensure that it can be relied on to have a specific interpretation and application in medical product development and regulatory review, within the stated context of use (COU). A biomarker is qualified, and not the biomarker measurement method^[1].

The qualification process is collaborative, where the Biomarker Qualification Program works with the requestor(s) in guiding biomarker development. Multiple interested parties often work together in working groups or consortia to develop a biomarker for qualification, which allows for shared resources and reduces the burden on individual collaborators^[1].

Qualified biomarkers have the potential to provide valuable information that may reduce uncertainty in regulatory decisions during drug development^[1].

How are new biomarkers discovered?

The discovery of relevant biomarkers is increasingly aided by in silico techniques based on applying data mining and computational chemistry to large molecular databases. An even larger source of information available for such discoveries is repositories constituted by research documents^[4].

A pilot experiment reported discovery of potential novel biomarkers and phenotypes for diabetes and obesity by self-organised text mining of about 120,000 PubMed abstracts, public clinical trial summaries and internal research documents. Recall and precision against established, but different, benchmarks lay in ranges up to 30% and 50% respectively^[4].

This approach, supplemented by appropriate manual curation processes, shows promise as an alternative to vocabulary processes dependent on humans having to read and analyse all the texts. It could impact pharmaceutical research, for example to shorten time-to-market of novel drugs, or speed up early recognition of dead ends and adverse reactions^[4].

What happens when biomarker test accuracy is uncertain?

Not every biomarker test performs well enough to change clinical decisions. A systematic review and economic evaluation assessed three biomarker tests – PartoSure, Actim Partus and the Rapid Fetal Fibronectin 10Q Cassette Kit at thresholds ≠50 ng/ml – for women presenting with signs and symptoms of preterm labour relative to fetal fibronectin at 50 ng/ml^[7].

Twenty studies were identified evaluating diagnostic test accuracy against the reference standard of delivery within 7 days, and seven studies against the reference standard of delivery within 48 hours. The test accuracy results revealed a high level of uncertainty, primarily attributable to substantial methodological, clinical and statistical heterogeneity between studies^[7].

The headline of that assessment states there is a high degree of uncertainty around the test accuracy and cost-effectiveness of these three tests, which may be reduced with the results of four ongoing UK trials^[7].

How are biomarkers used in dementia diagnosis and research?

Before the early 2000s, the only sure way to know whether a person had Alzheimer's disease or another form of dementia was after death through autopsy. Tests are now available to help doctors and researchers see biomarkers associated with dementia in a living person^[3].

When combined with other tests, these biomarkers can help doctors determine whether a person might have or be at risk of developing Alzheimer's or a related dementia. However, no single test can diagnose these conditions, and biomarkers are only part of a complete assessment^[3].

In some cases, these biomarker tests are only available through a specialty clinic or medical research facility. Physicians with expertise in this area include neurologists, geriatric psychiatrists, neuropsychologists and geriatricians. Insurance coverage may be limited, so check with your health insurance plan to find out what is covered^[3].

Volunteers are also needed for clinical trials that are exploring new ways to detect dementia early; by joining one of these studies, you may learn more about the biological signs of dementia and contribute useful information to help diagnose and treat it^[3].

How are biomarkers used to predict cancer risk?

Biomarker assays can be developed to predict future cancer risk. In one research programme, liver tissue-based cancer risk signatures were to be implemented in clinically applicable tissue-based (NanoString) and serum-based (Luminex) assays, and evaluated for technical validity and capability to predict future cancer risk in a cohort of 200 NASH patients with paired serum and liver tissue specimens as well as completed long-term clinical observation for cancer development^[8].

That programme also aims to elucidate molecular dysregulation underlying the carcinogenic milieu in livers affected with nonalcoholic steatohepatitis (NASH), described there as the fastest rising liver cancer aetiology, as clues to refined chemoprevention targets, drugs and biomarker assays. These assays are expected to enable personalised patient management and more cost-effective liver cancer chemoprevention clinical trials^[8].

Where can you find biomarker-driven trials?

Clinical trials that use biomarkers, including trials exploring new ways to detect dementia early, are listed in public registries^[3]. In Europe, trial information is published through CTIS, the EU clinical trials information system, and studies can also be found on ClinicalTrials.gov. Trials recruiting in Europe can also be searched on clinicaltrials.eu.

What should you ask before a biomarker test?

• What is being measured – ask for the biomarker name, the source or matrix, the measurable characteristic and the analytic method used to measure it^[1].
• What the result does and does not mean – a biomarker is not an assessment of how you feel, function or survive^[1].
• Whether the result alone can confirm a diagnosis – no single test can diagnose dementia, and biomarkers are only part of a complete assessment^[3].
• How certain the test result is – some biomarker tests carry a high degree of uncertainty around test accuracy because of heterogeneity between studies^[7].
• Who can order and interpret the test – some biomarker tests are only available through a specialty clinic or medical research facility^[3].
• Whether the test is covered – health insurance plans may cover only certain, limited types of biomarker tests, so check what is covered^[3].

When should you talk to a doctor or pharmacist?

Talk to your treating physician before acting on a biomarker result that was not ordered as part of your standard care. Biomarkers are only part of a complete assessment, and no single test can diagnose a condition such as dementia. If a test is available only through a specialty clinic or medical research facility, ask your physician for a referral to a neurologist, geriatric psychiatrist, neuropsychologist or geriatrician as appropriate. If you are considering a biomarker test outside a clinical trial, check with your health insurance plan which tests are covered before you proceed^[3].

Summary

Start by asking what is actually being measured: the biomarker name, the source or matrix, the measurable characteristic and the analytic method all belong to the description of any biomarker. Then ask which of the seven categories that marker belongs to, because susceptibility, diagnostic, monitoring, prognostic, predictive, pharmacodynamic/response and safety markers answer different questions^[1].

For anyone considering a study, remember that biomarker tests can be used to identify participants who meet particular requirements for a clinical trial and to track their response to a test drug^[3]. Finally, treat an uncertain biomarker result as information to discuss, not as a conclusion: some tests carry a high degree of uncertainty around their accuracy, which is why a result should always be interpreted by a clinician together with other findings^[7].

1. [1] About Biomarkers and Qualification (accessed 2 June 2026) — https://www.fda.gov/drugs/biomarker-qualification-program/about-biomarkers-and-qualification
2. [2] Biomarkers (accessed 2 June 2026) — https://www.niehs.nih.gov/health/topics/science/biomarkers
3. [3] How Biomarkers Help Diagnose Dementia (accessed 2 June 2026) — https://www.nia.nih.gov/health/alzheimers-symptoms-and-diagnosis/how-biomarkers-help-diagnose-dementia
4. [4] Discovery of novel biomarkers and phenotypes by semantic technologies - PMC — PubMed Central (accessed 25.09.2026) — https://pmc.ncbi.nlm.nih.gov/articles/PMC3605201/
5. [5] Examples of Technical Innovation for Traumatic Brain Injury Prevention, Diagnosis, and Care: Proceedings of a Workshop - Treatment (accessed 2 June 2026) — https://www.ncbi.nlm.nih.gov/sites/books/NBK613055/
6. [6] From spatial biomarker to companion diagnostic: The development path (accessed 2 June 2026) — https://www.drugdiscoverynews.com/from-spatial-biomarker-to-companion-diagnostic-the-development-path-17454
7. [7] Three biomarker tests to help diagnose preterm labour: a systematic review and economic evaluation — journalslibrary.nihr.ac.uk (accessed 25.09.2026) — https://www.journalslibrary.nihr.ac.uk/hta/HTA23130
8. [8] Reverse-engineering precision liver cancer chemoprevention (accessed 2 June 2026) — https://prevention.cancer.gov/funding-and-grants/funded-grants/R01CA233794