A systems biology approach to oncology drug development

Published on August 31, 2015   32 min
0:00
BIRGIT SCHOEBERL: Hello. My name is Birgit Schoeberl, and I work for Merrimack Pharmaceuticals. We are based in Cambridge, Massachusetts. The company is a biotech company based on systems biology. We are about 14 years old. And I'm happy to be here today to talk to you about the story of our anti-ErbB3 antibody MM-121 and use that as an example of how systems biology can be used in drug development.
0:29
So where has the state of oncology drug development been? In general, the focus has been on correlating how individual genes drive tumor growth, also called oncogenic drivers. And the single one-to-one correlation resulted in a series of very interesting targeted therapies, like Crizotinib to treat ALK mutations, Herceptin to treat tumors with Her2 amplifications. And they're, in general, very effective, but the problem with these types of therapies that are targeted at single mutations driving tumor growth, are that there's only very few patients that have tumors that are dependent on these mutations. So the prevalence ranges, in general, 10% to 15%. So why do we need to understand the networks and systems? Why do we need systems biology? Often there are no simple one-to-one correlations, and the tumors are dependent on multiple pathways, or the dependents are much more complex. And now, we are in the fortunate situation that most of the components of human cells are known, thanks to the Human Genome Project. So now it's really up to us to understand how these components play together in healthy as well as disease states. So we get to this holistic understanding of what is driving tumor growth, that will allow us to identify critical drivers of tumor growth, that are not necessarily mutated, but most likely, more abundant. And what I'd like to do today is show you an example of how we did that and identified ErbB3 as a target, and discovered 121 and anti-ErbB3 antibody, as well as predictive response biomarkers that were, later on, implemented in our clinical trials. So in the next slide, I will show you a schematic of how we use network biology, or systems biology, in practice.
2:46
So one of the first step in our discovery process is to identify networks, critical networks, that drive tumor growth, and use these networks to identify targets, and see what are the critical nodes in this networks that would make good targets to inhibit tumor growth. So diagnosing the network. As a second step, we use these computational models that we've built to understand or set therapeutic design criteria. So in silico simulate different mechanisms. A potential drug that targets this one target, or two targets, and what is the affinity? What should the mechanism of action be? So based on these design criteria, we then go about and either develop an antibody therapeutic, which could be, in the simplest case, a monoclonal antibody, bi-specific antibody, a mixture of antibodies, or it could be small molecules, nanoparticles. So there's no limit just thinking about the possibilities of drug design. And secondly, these computational models of pathways, so the understanding of what are the driver networks, can be used to identify predictive response biomarkers. So what should we measure in the tumor to see whether this tumor lesion is dependent on a particular network, or multiple networks. And then figure out what should be the single drug, or drug combination, which show the highest probability of success. These insights can also be used to then drive the clinical development, and identify indications where we believe these particular drug or drug combination will have the highest probability of success, meaning that the abundance of patients that are dependent on the critical networks identified is the highest.
4:59
So if we go to the next slide, it's just a cartoon of how networks usually look like, and the challenges they bring, as well as why it is important to understand the network as a whole, and not just a single gene. So often we see redundancy in networks. So there's more than one receptor, and if you inhibit one, the network is still driven by other receptors in the family, or outside the family. In addition, often we see feedback regulation to either amplify or dampen the signal, so that we convert a sustained signal into a transient signal. And very often we also see a very strong amplification from the receptor, down to the kinesis, or transcription factors. So in order to develop a very potent drug, it is important to understand if there is a strong amplification occurring from the receptor, down to the transcription factors. Because 90% inhibition, then, may just not be sufficient to potently inhibit a particular pathway. So let me now tell you, in more detail, the history of MM-121.
Hide

A systems biology approach to oncology drug development

Embed in course/own notes