Introduction

Biomarkers are measurable indicators of normal or disease processes in the body. This might include an individual’s response to a medicine.

For example body temperature is measured to indicate a fever. The concept of biomarkers is not new. Some examples of biomarkers are:

• Biological substances (‘biochemicals’), such as enzymes or hormones - these may be found in the blood or in tissue samples (biopsies).
• Gene changes.
• Images from Magnetic Resonance Imaging (MRI).

Biomarkers can be any measurements that can reliably be made and that can tell us something about a person’s health or disease state. This might include their response to a therapy. It has to be considered that levels of specificity and sensitivity of a given marker may vary widely:

• Specificity - the proportion of ‘negatives’ which are correctly identified.  
• Sensitivity - the proportion of ‘positives’ which are correctly identified.

In other words, the accuracy of biomarkers varies and therefore not all markers are suitable for medicines development. Even more challenging is the use of biomarkers for regulatory purposes (see below).

The main focus of this topic is molecular biomarkers in medicines development. There is a special emphasis on recent genomic technologies, i.e. those that look at genes.

Aims of biomarker use

Biomarkers can be used to measure:

• Normal biological processes in the body.
• Pathological (disease) processes in the body.
• A person’s response to a treatment or medicine.

The two main aims of biomarkers in medicines development are:

• Streamlining the process of medicines development - ideally in a clinical trial you measure the patient’s response to a treatment, however when this is not directly possible, biomarkers may provide an alternative way of measuring an outcome (surrogate endpoint). In addition, biomarkers can help select which patients are best placed to take part in a clinical trial.
• Improving the tailoring of treatment to individuals - biomarker research is helping to improve how well we can predict a person’s risk of disease, how a disease might progress once it is diagnosed, and how an individual will respond to a medicine. This is starting to help with decision-making by healthcare professionals, for safer and more effective treatment.^[1]

There are many established biomarkers that are known to have clinical use. That is, they are proven to give reliable measures of underlying biological processes. Therefore, their use in medicines development is well accepted. For example, a clinical endpoint in a trial (such as relief of symptoms, survival, or disappearance of a tumour) can in some cases be replaced with an established biomarker. This is known as a surrogate endpoint. For the medicine to receive marketing authorisation, there must be good evidence that the biomarker in question is a valid substitute for the clinical endpoint.

Example

• Glucose levels in a patient’s blood (blood sugar level) can be used to monitor if an individual is responding to diabetes treatment - uncontrolled glucose levels are one of the major problems in diabetes patients who are not treated properly.

• Magnetic resonance imaging (MRI) of a patient’s brain to assess disease status in multiple sclerosis - this might be monitored instead of clinical disease progression, or the patient experiencing relapses.

In addition, many new exploratory biomarkers are being discovered and used during the development of new medicines. Many of these use ‘omics’ technologies: genomics, proteomics and metabolomics:

• Genomics is a branch of genetics that applies various methods to sequence, assemble, and analyse the function and structure of genomes.
• Proteomics is the large-scale study of proteins, particularly their structures and functions and quantity. Proteins are vital parts of living organisms, as they are the main components of the metabolic pathways of cells.
• Metabolomics is the scientific study of chemical processes involving ‘metabolites’. Metabolites are small molecules which are left behind after a chemical process has taken place in a cell - they are the ‘product’ of the process. 

Pharmacogenetics or pharmacogenomics specifically uses genetic or genomic information (i.e. genetic or genomic biomarkers) in the development and use of medicines.

Biomarkers in the pharmaceutical industry today

Currently nearly a third of the medicines in development have some form of a genomic or proteomic marker. There has been a rapid increase over recent years (see here). This varies across disease areas, however cancer (oncology) research was one of the first areas where the use of such biomarkers was adopted.

• Biomarkers are being used to make exploratory trials (i.e. early trials, Phase I) of medicines more efficient.
• Only a limited number of biomarkers can be used as a substitute for a clinical endpoint in a confirmatory trial (i.e. late stage trials, Phase III). They may however be used in late stage trials in combination with clinical endpoints.
• For medicines that are being designed for specific molecular targets, only a sub-group of patients might respond. It is important to identify these patients for clinical trials, using biomarker measures.

Companion diagnostics

A validated (proven) test for each biomarker is needed. This is to accurately measure whether a biomarker is present and at what level. It may be a laboratory test or a test kit. The test may help to:

• Select patients likely to respond to a medicine.
• Exclude those likely to have an adverse reaction.
• Determine the best dose for a patient.

‘Companion diagnostics’ are tests that are validated and approved for marketing alongside a new medicine. They can be:

• Developed once a medicine is on the market.
• Developed alongside a medicine that is still under development.

Many companies developing targeted therapies, especially for cancer, have also begun to consider the potential benefits of developing a diagnostic to pair with that treatment. The trend is to develop medicines and companion diagnostics together (co-development) rather than have both developments happen in isolation. A recent example, given by a DIA report about China, highlights the (regulatory) framework of co-development (see here).