Goodwin-Arctoris Drug Discovery Masterclass Recap
Professionals in drug discovery and development face difficult odds: on average, it takes 12 to 15 years to take a new drug from the idea stage to market approval; that is around one-third of a scientist’s professional life! The costs are eye-watering as well: the price tag of taking a single new drug to the market is in the range of $2bn, with attrition rates in excess of 90%. Drug development and discovery indeed is a long and winding road.
Dan Thomas, our Head of Discovery Biology, has seen this road before. Dan has more than 20 years of experience working in early-stage drug discovery in the pharmaceutical industry – and he shared his insights giving a webinar as part of the Arctoris-Goodwin Masterclass. In his seminar, Dan gave an overview of the early-stage drug discovery process and shared his views on how to improve it to reduce attrition rates. In this blog, we present a summary of his seminar.
The Attrition Cliff
A promising compound needs to overcome a series of tightly regulated hurdles designed to ensure that only safe and effective drugs find their way to patients. It all starts in the early discovery phase, resulting in a handful of potential compounds. These are then tested in animal studies to explore dosage levels and possible adverse effects, to protect human trial participants. These tests form the backbone of an Investigational New Drug Application (IND), requesting authorization from the Food and Drug Administration (FDA) in the US — or the relevant regulatory counterpart in other countries — to administer the substance in humans.
After successfully filing an IND application, the drug candidate enters clinical trials — from Phase I for safety assessment, up to Phase III for a head-to-head comparison with the standard of care. Despite all the effort placed in ensuring that only promising candidates start this process, attrition rates are dismal.
Less than 12% of the candidates resulting from drug discovery programmes make it all the way through the clinical development funnel.
Failures can be attributed to several reasons, including dose-limiting toxicity, low efficacy, or poor bioavailability. Notably, the longer it takes to identify these shortcomings, the higher the cost a program will already have absorbed. Thus, ensuring that the earliest steps of the process are done with the utmost care can have enormous implications on the process’s overall economics.
Drug Discovery in a Nutshell
The initial phases in the path to market are collectively known as drug discovery. A general blueprint for this process has emerged over the past decades: the first step consists of identifying a gene or protein — known as a target — suspected of playing a pivotal role in a particular disease or physiological process. It is crucial to eliminate less promising targets promptly as going after a red herring can prove extremely costly.
Then, suitable assays are designed to be able to test and profile molecules for their effects on the target. These assays are subsequently used to carry out large-scale screening campaigns, where often millions of molecules are tested for effects on the chosen target. Molecules displaying the desired effect in these high-throughput screening campaigns — known as hits — are then singled out and taken to the next step in the process.
Hits are carefully analyzed and optimized, resulting in a series of lead compounds — molecules that are deemed promising enough to warrant additional work. Leads are optimized via iterations of Design-Make-Test-Analyse (DMTA) cycles to improve critical properties such as affinity, selectivity, metabolic stability, and bioavailability. The optimized molecules are then characterized to understand their mode of action before moving them towards the next stage, commonly referred to as drug development.
Drug development begins with pre-clinical tests where absorption, distribution, metabolism and excretion (ADME), as well as toxicity/ safety, are assessed in animal studies. In this part of the process, experiments are designed to evaluate potential benefits for certain dosages. The results from the pre-clinical program are used to prepare an IND application to be able to launch clinical development.
After all this hard work, we finally have a molecule ready for testing in humans. Regretfully about nine in ten of these molecules will never make it to the market. Perhaps we have to raise the standards to ensure that fewer but better molecules make it to the clinical stages of development?
Strategic Shift — Getting the Early Steps Right
To illustrate this approach, let’s take a step back. Imagine you are tasked with determining a quantity experimentally. Now, suppose that the experiment involves several subjective intervention points (SIP) — a choice between reagents from different vendors, different stirring methods, etc. Even if each of these choices introduces only minute errors, uncertainties rapidly compound leading to a broad range of potential outcomes (figure below). Things complicate further once there are transfers of responsibility along the way. Can we take the reported results from the previous steps at face value? Should we verify them? Is it even possible to do so? These questions are at the core of the reproducibility crisis that we have written about previously.
At Arctoris, our objective is to reduce the uncertainty of outcomes by addressing these critical issues, i.e. narrowing the potential space of outcomes:
- First, we minimise experimental uncertainty due to — largely avoidable — human error by relying on automated processes.
- Second, we aim to reduce the number of SIPs by fostering unambiguous protocols, where the choices of reagents and equipment are no longer subjective.
- Third, we ease the transfer of responsibility by generating full audit trails with detailed and complete capture of data and metadata.
- Finally, we reduce cycle times so that we can be less hesitant to take a step back if something needs to be verified.
Case Study: Ultra-High Definition Profiling and Accelerated Onboarding
To highlight the capabilities of automated experimentation and data collection, Dan also shared details of an example showing the gains in the resolution of data as well as the acceleration of the experimental process.
The example related to supporting clinical research in response to the COVID-19 pandemic, when the crisis had just started in March 2020. The scientific team at Arctoris decided to establish and run comprehensive assays to reveal which of the 24 currently available JAK inhibitors was most effective against the family of Janus Kinases. Dan highlighted the onboarding process for adding these new assays: in mere 10 hours, our platform, Ulysses, was able to run full reagent verification. Less than a week later, we were able to create and validate standardised experimental execution procedures and accompanying documentation for the new assays including full process maps, experiment blueprints, and analytical pipelines. All in all, for the JAK inhibitors, we were able to collect a detailed data set with more than 1 million data points in under 100 h — which enabled us to create accurate differentiating profiles to establish full activity and kinetic profiles, an essential first step of any drug discovery process and vital data supplementing ongoing efforts to find treatment for patients suffering from COVID-19. This preparatory work now enables us to generate data for our clients and partners using these (and many other) assay in less than 24 hours between compound delivery and report generation without sacrificing data quality or speed. In other words, we have moved the drug discovery and development process from the long-winding country road to the highway.
Dan presented data highlighting the onboarding process for adding a new assay such as the one against the Janus Kinases to the Arctoris portfolio. In mere 10 hours, our platform was able to run full reagent verification. Then, in less than a week, we were able to create and validate standardised experimental execution procedures and accompanying documentation for the new assay including a full process map, experiment blueprint, and analytical pipeline. This preparatory work now enables us to generate data for our clients and partners using this assay in less than 24 hours between compound delivery and report generation without sacrificing data quality or speed.
In other words, we have moved the drug discovery and development process from the long-winding country road to the highway.
Dr Daniel Thomas is a seasoned drug discovery professional with experience of over 20 years in early-stage research and development in the pharmaceutical industry. He has extensive theoretical and practical knowledge of assay development, molecular profiling and enzyme kinetics, and is an accomplished leader with a track record of developing trans-national matrix research teams and leading early-stage programmes across a broad range of therapeutic areas during his time at GSK. As the Head of Discovery Biology at Arctoris, he is responsible for the development and implementation of a comprehensive scientific strategy designed to deliver the best in data quality.