The right drug targets for the right patient: Using biomarkers to solve challenges in drug development

Protein biomarkers provide a dynamic and powerful approach to identify and understand novel drug targets in early drug development. The pharmaceutical industry utilizes biomarkers to help uncover disease progression markers, responder signatures, diagnostic biomarkers, and potential new indications.

In this virtual roundtable discussion, now available to watch on demand, we are joined by leading experts, Dr. Jürgen Braunger of Boehringer Ingelheim Pharma, Dr. Timothy Radstake of AbbVie, and Dr. Christopher D. Whelan of Biogen. These experts share their opinions on how biomarkers can help drive early clinical phase trials, discuss their work with biomarkers at present and predict how the field will evolve to meet future challenges.

As well as advancements, panelists also highlight potential challenges and issues that the pharmaceutical industry may encounter during the use of protein markers in discovery, and what needs to improve.

Read on for highlights from the live Q&A session or register to watch the webinar at a time that suits you.

Is it better to conduct pharmacogenomics or genetic studies for a population or an individual? And do you think precision medicine for an individual is cost-effective?

JB: In our experience, when conducting proteomic analysis with a big variety of markers, it’s used in a clinical trial setting, in a larger set of patients rather than an individualized view. And this is because you need good statistics with many patients, so you can differentiate the nonsignificant effects and focus on those patterns that are showing statistical and meaningful significance.

Coming to the question of whether precision medicine is cost-effective, I'm a real believer that this is true. Once you can focus your therapies on those patients who are in need and who have the most chances to benefit from the drugs, then this also limits the size of the population who gets treated, and in the end, focuses the medicine to be more cost-effective.

With the difficulty to establish normal ranges in biomarker studies due to the lack of exhaustive understanding of underlying biological processes, how can the confirmation and validation of biomarkers be done efficiently?

TR: If you look at the sole purpose of the biomarker, you want to try to find that protein or marker and whether it's associated with the disease or not. Ideally, you can use that biomarker to predict the outcome of a disease. I don’t think you have to understand how that biomarker is implicated in biological relevance or disease mechanism.

I think that we are moving into a field where we use the multi-omics and analysis technologies with machine learning and artificial intelligence, to really use that multitude of biomarkers to go to that next stage and to say, "Okay, if you look at that network of proteins, can we understand disease?" And in that sense, it's going to be important.

I think if you just look at the disease association, it's important to replicate because if you find a biomarker in one operation, which usually is pretty small, that doesn't tell you anything. You would see the same effect if you take a bigger population, maybe even a population across the globe. I think to the first question, you have to do at least two replication studies to make sure that your biomarker is actually telling you something.

CW: In genetics, we try to combine protein biomarker data with the genetic data on techniques like Mendelian randomization to determine causality. It helps us pass out biomarkers that might be simply associated with the disease versus biomarkers that might be driving the disease pathophysiology. By combining genetics and protein biomarkers, we can get at that question.

What are the sources for these biomarkers?

CW: We're looking at our proteins as typically the end products of genes. We think of genetics as the blueprint and the proteins are the end products of that blueprint. That's probably the more simplistic way of doing it.

JB: It need not necessarily be proteins. It can be any modification of proteins or disease expanded by modification of proteins. It could also be cleavage products of proteins that could be interesting as biomarkers. Any kind of protein products that you could detect from bodily fluids or other matrixes that can be detected by antibody technology, I would say.

How stable are protein biomarkers?

TR: I think there are three different answers to that question. First, what is meant by stability? You can talk about stability in a patient over time because as time evolves, patient disease changes. And then you have the stability issue in storing a sample, thawing and freezing, and all those things you do with a sample over time. And then I think the third one is, if you measure the same sample 10 times during the same day, what's that stability?

In rheumatology, for example, where you see a biomarker like anti-citrullinated peptide, which is associated with a risk to develop rheumatoid arthritis in patients that have a genetic risk and joint pain, that protein itself changes over time; it becomes a part of the disease driver because it activates P cells. I think some biomarkers you can use to look at phases before a disease critically affects the patient.

To answer the second question about stability over time. I think you have to be careful with that because some proteins are really unstable, and I don't know if they're affected by three storing cycles. Some proteins will completely disappear. It’s super important to make sure that you know what happened with your sample over time.

JB: In some instances, we observe proteins that are so unstable that without any stabilization agent like protease inhibitors, you won't see any of the anilides in your assay later on. So, from the moment you're taking the blood, the biological processes in the blood tube need to be stopped, otherwise, you don't detect anything.

SelectScience hosts 10+ webinars every month, discover more of our upcoming webinars>>