Definition of biomarkers and efficacy end-points

3. Challenges for Biomarkers in Medicines Development

3.2. Regulatory Challenges

The regulation of medicines is evolving when it comes to new methods such as biomarkers.

A good definition of biomarker for clinical trial purposes is the following one, already formulated by Temple, in 1999: “A biomarker is a laboratory or physical sign that is used in therapeutic trials as a substitute for a clinically meaningful endpoint. The endpoint is a direct measure of how a patient feels, functions, or survives and that is expected to predict the effect of the therapy.”

However, in the regulatory field, biomarkers and surrogate endpoints are not synonymous. For a biomarker to be used as a surrogate endpoint, studies will be done to assess the direct relationship of the biomarker with:

The development of the disease.
A treatment intervention with an important clinical endpoint.

There are advantages of using validated biomarkers as surrogate endpoints, for example:

They can be measured earlier, more easily or frequently and with higher precision than clinical endpoints.
They may be less affected by other treatments
They may reduce the size of the sample required and allow researchers to make faster decisions.
There are important ethical advantages in diseases with poor prognosis.

This is the reason why surrogate endpoints (if validated) are often used in ‘fast-track’ or ‘accelerated’ approval procedures. They will assist the development and speed up the review of new medicines that are intended to treat serious or life-threatening conditions. And/or they can demonstrate the potential to deal with unmet medical needs. In ‘accelerated’ procedures however, data must ‘verify and describe the medicine’s clinical benefit’ after it has been marketed. Post-marketing data are also required to resolve remaining uncertainty on the relation of the clinical benefit to the surrogate endpoint (upon which approval was based). Or the relation of the observed clinical benefit to ultimate outcomes.

What characterises a surrogate endpoint is that it is objectively measured and that it is evaluated as an indicator of:

Normal biologic (physiological) processes - how things work within the body.
Pathogenic processes - how a disease develops.
Pharmacological responses to a therapeutic intervention - how the body responds to the use of a medicine or other therapy.

It should be directly related to clinical outcomes.

For regulatory purposes, surrogate endpoints represent only a ‘sub-set’ of biomarkers. A biomarker can be called a surrogate endpoint if it is expected to predict clinical outcomes. This could be harm, or lack of benefit based on epidemiologic, therapeutic, pathophysiologic or other scientific evidence. Just having a good ‘correlation’ or match does not itself make a good surrogate endpoint. An example is the increase in something called ‘prostate specific antigen’ (PSA). PSA is correlated with prostate cancer, in other words, people who have prostate cancer often also have higher levels of PSA. However, it cannot be relied on as a surrogate endpoint. This is because levels of PSA may be reduced when the patient is treated for prostate cancer, but there is no effect on the long term outcome of the disease.

A strong example is in the development of antiretroviral medicines for HIV and AIDS. Levels of ‘CD4 lymphocytes’ and changes in ‘HIV-RNA’ plasma levels can now be used as surrogate endpoints. Previously, studies would have been based on hard clinical endpoints such as progression of the HIV infection to AIDS and/or patient survival.

Biomarkers are also used for medicines that are developed for a specific sub-population of patients. A medicine that is only going to be prescribed to a small sub-population of patients can be labelled by the European Medicines Agency (EMA) as an ‘orphan’ medicine. Companies that develop such products are given benefits when the medicinal products are authorised, such as ‘market exclusivity’. This means that other companies will not be allowed to market a similar medicine or ‘generic’ for 10 years (market exclusivity). The EMA receives an increasing number of requests for orphan designation for new medicines. This has given the EMA considerable experience in how to assess the possible benefits and limitations of biomarkers used for regulatory purposes.(1)

Medicines developers are encouraged to engage with regulators at an early stage when they consider using new biomarkers. They can submit their plans to use the biomarkers to the EMA. They can present the scientific evidence they have collected to validate specific biomarkers, so that the EMA’s Committee for Medicinal Products for Human Use (CHMP) can give an opinion on their use.

It can be complex and expensive to validate biomarkers so that they can meet regulatory standards. This is especially the case if a biomarker is to be used as a surrogate endpoint. In this case, a dedicated clinical trial is required that is designed to test the link between the biomarker and the clinical endpoint.

Finally, in the EU, medicines and diagnostics are regulated differently. Obtaining a marketing authorisation for a medicine and its companion diagnostic together adds an extra layer of complexity to the approval process.

(1) Tsigkos S, Llinares J, Mariz S, Aarum S, Fregonese L, Dembowska-Baginska B et al. Use of biomarkers in the context of orphan medicines designation in the European Union. Orphanet Journal of Rare Diseases, 2014; 9:13.