Interview: How to validate a biomarker

SW: What motivated you to become Biomarker Validation Scientist at IBBL?

MM: Before joining IBBL, I had spent 10 years in Canada. First as a researcher in the field of neurodegeneration and then building-up a lab completely from scratch for the Canadian Longitudinal Study on Aging (CLSA). While I enjoyed both of those roles, I was really looking for a new challenge. So, when I came across the opportunity to develop this new biomarker validation service at IBBL, I just went for it. Firstly, because this is a field that I have always found very interesting and validation has been part of my work throughout my career. Secondly, because within the biobanking community IBBL has a very good reputation for being one of the leaders as far as quality and validation are concerned. Therefore, I was very much looking forward to joining this multi-disciplinary team and contributing to the introduction of new biomarkers into the clinic.

SW: Why is it important to validate a biomarker?

Biomarkers have long been hailed as the key to better patient care and lower cost to the healthcare system. But, while there have been thousands of papers related to biomarker discovery, very few of these biomarkers have been validated for use in the clinic. So, there is clearly a problem and the scientific community is aware of it. It is actually a multi-facetted and multi-factorial problem. There are multiple stages during the discovery and validation of a biomarker where things can go wrong, such as problems with the study design or variations in the sample collection.

Even if you do everything right, more often than not, it turns out that your biomarker is simply not robust, sensitive or specific enough to cut it in the real life clinical setting. But, while the odds may seem stacked against us, the potential payoff for the researcher, the healthcare system and, most importantly, the patients, are huge. It then becomes a question of reducing risk and improving the chances of getting into the clinic. This is where a rigorous and systematic validation of the biomarker can make all the difference. The closer you get to the final clinical application of a biomarker the more thorough you need to be. Legislators are also starting to recognise this and introduce stricter recommendations and regulations, such as the new EU legislation on in vitro diagnostic devices.

SW: What is the correct way to validate a biomarker?

As it is so often the case, there simply is no magic pill that fits all. It largely depends on the type and application of the biomarker. Of course, there are some common steps that need to be included in any good biomarker validation plan. But, since there are so many potential pitfalls and many academic researchers simply do not know where to start, it is best to get a professional on board who is experienced in validation and approaches it from a more industrial point of view.

This is why IBBL has decided to share its expertise and carries out all of these steps of early validation for its clients. Pre-analytical validation assesses the robustness of the biomarker to common pre-analytical variables. Analytical validation assesses the precision, the matrix effect, the analytical range, the specificity, the sensitivity and the stability of the method. We then also perform a clinical verification, which is essentially a pilot study on a small sample size to evaluate the performance of the biomarker in a clinical setting, as well as a comparison to other markers and gold standards.

Once this early validation phase is completed, the client will be able to use the validation report and recommendations that we provide to help him/her decide how to proceed before embarking on the final large-scale clinical validation. He/she may use this report in business pitches to investors, in patent applications, in regulatory submissions or for publications. By approaching the validation this way and evaluating the biomarker after each step, we are able to save resources, reduce risk and ultimately bridge this gap between the discovery and the clinic application.

SW: Thank you, Monica for taking the time and answering my questions.  


Photo © IBBL