Drug discovery can be a long and gruelling process, but this effort is vital to advance human health. As is the case in many fields, the increasing sophistication of genomics and other omics, quantified at single-cell and spatial resolution, have, and will continue to, improve the process.
However, genomic data has in fact played a role in this process for a long time now, and it persists as a staple tool in the drug discovery journey. In this feature, we take a look at the state of genomics in the drug discovery field, and dive into the past, present and future to give you a comprehensive overview of the research landscape.
The drug discovery and development process is arduous, and it is not unusual for it to take more than 15 years. Furthermore, with 90% of drugs in the pipeline ultimately failing, it is clear that we need as many avenues as possible to successfully bring novel drugs to the market. Genetic evidence is one piece of the puzzle that can be used to validate potential targets and help to get new treatments out of the lab and into the clinic. In fact, drug targets supported by genetic data are more than twice as likely to be approved.
There are two stages to getting a drug into the clinic: the discovery phase and the development phase. Discovery can take up to 15 years and involves finding relevant biological targets (e.g., genes and proteins) that are associated with disease pathology. This phase also involves the identification of molecules that can be used to modulate the target, alongside pre-clinical and first-in-human studies.
The second stage, development, can take up to 10 years on top of this and involves late-stage clinical trials. Safety, efficacy and optimal dose are all assessed within this second phase. But with each disease having different parameters, it can take a long time to observe relevant clinical outcomes following drug administration.
The impact of genomics in drug discovery has been well documented since the late 20th century. In fact, this article from 2000 detailed the progress in the field as a result of ever-increasing new tech. Plus, the creation of the human reference genome in the wake of the Human Genome Project exponentially increased the opportunities to find disease-associated biomarkers.
There are a number of examples of drugs approved off the back of significant genetic evidence over the years. Perhaps the most famous is PCSK9. Both gain- and loss-of-function mutations in the PCSK9 gene are linked to altered levels of low-density lipoprotein cholesterol, the first link being discovered in 2003 and the second in 2005. Lowered cholesterol levels associated with the loss-of-function of the PCSK9 protein were seen to lead to a decreased risk of coronary heart disease.
In light of this evidence, the FDA approved PCSK9 inhibitors for those at risk of coronary heart disease, as a means to lower cholesterol levels. The first drug was approved in 2015, 12 years after the publication of the first genetic evidence. The genetic data has been credited as an integral part of the development of PCSK9 inhibitors. Today, there are three such drugs on the market, and more are in the pipeline. Additionally, there are similar inhibitors in the clinic to treat common diseases including HIV and osteoporosis.
With an increase in the use of other omics, a significant amount of work has gone into figuring out exactly how this data drives the transition of drugs from lab to clinic. A 2019 article posited that, with genetic evidence, a drug is two times more likely to be approved. And a 2022 investigation revealed that over two thirds of approved drugs from the previous year had some kind of genetic evidence backing them. These drugs are primarily used to treat cancers, but other diseases such as chronic kidney disease also feature.
The amount of time it takes for a drug to be developed, and the gap from discovery of genetic evidence to approval, is steadily decreasing thanks to constantly improving technology and methodologies. In addition to genomic evidence, artificial intelligence and machine learning are also playing a role in driving this journey forward. However, the process is still far too long, contributing to ongoing unmet clinical need and highlighting a real need for system reform.
The profiling of genomics, transcriptomics, proteomics, metabolomics and a variety of other techniques have all played a pivotal role in drug discovery over the last year. Let’s take a look at some recent papers that summarise the current omics drug discovery landscape.
Future prospects for human genetics and genomics in drug discovery – This article explores the impact of genomic data on the drug discovery process and the value of genetic evidence in pushing drugs through the clinical trial process, by leveraging large-scale population data (Ghoussani, Nelson and Dunham, 2023).
Harnessing the Power of Electronic Health Records and Genomics for Drug Discovery – This review from January 2023 highlights the benefits of leveraging genetic data for drug discovery and also drug repurposing. It explores the use of biobanks and electronic health records to inform the process (Krebs and Milani, 2023).
Bioinformatics Paradigms in Drug Discovery and Drug Development – This publication explores the use of bioinformatics analyses alongside the use of various omics in order to expediate the drug discovery process (Sharma et al., 2023).
Deep learning in image-based phenotypic drug discovery – This study from July 2023 details the use of artificial intelligence deep learning models to sort through thousands of images from phenotypic-based drug discovery processes, as a means to identify cellular changes (Krentzel, Shorte and Zimmer, 2023).
AI-powered therapeutic target discovery – This opinion piece assesses the use of artificial intelligence to sort through genetic data as a means to find new drug targets, a crucial early stage of the drug discovery process (Pun, Ozerov and Zhavaronkov, 2023).
Modern drug discovery for inflammatory bowel disease: The role of computational methods – This 2023 review discusses the use of bioinformatics methods to improve the drug discovery process for inflammatory bowel diseases such as Crohn’s disease, and how this has led to the emergence of a number of small molecule candidates (Johnson et al., 2023).
AlphaFold2 protein structure prediction: Implications for drug discovery – Another opinion piece, this article explores the use of protein structure prediction in the journey towards identifying and designing effective small molecule drugs, specifically using AlphaFold2 technology (Borkakoti and Thornton, 2023).
Metabolomics- and systems-biology-guided discovery of metabolite lead compounds and druggable targets – This review from early 2023 discusses the use of metabolomics and systems-level interpretation of other omics data to derive conclusions about metabolites in biological systems, which can assist in target identification and lead compound discovery (Palermo, 2023).
Using genetic association data to guide drug discovery and development: Review of methods and applications – Another review, this paper describes statistical methods to assess and apply genetic data in drug discovery and the opportunities to use Mendelian randomization in the process (Burgess et al., 2023).
In the spotlight
From target discovery to clinical drug development with human genetics – Trajanoska et al., 2023.
Genetic data is clearly a valuable aspect of the drug discovery toolkit, but just how instrumental has it been over time? In this 2023 study, the team set out to answer this question, searching for drugs that have made it to the clinic off the back of genetic data.
How crucial was this evidence, and how many examples have there been over the decades? The team focused on non-cancer drugs, and searched for those that would likely not have been approved without this supporting genetic data. The process they followed, and the number of drugs identified at each stage, are detailed in the figure below. Examples where direct genetic evidence was found after the drug was approved were excluded between steps 4 and 5.
Figure 2: Flowchart showing the process to identify drugs where genetic data was crucial in its approval. Screenshot taken directly from Front Line Genomics webinar, where lead author Katerina Trajanoska presented the work.
Ultimately, the team found 48 first-in-class drugs for 40 targets that heavily relied on genetic data for approval. Among these, most were for metabolic diseases, and 36 of the targets corresponded to rare diseases. The median gap between establishing this genetic evidence and approving the drug was 25 years, but since the completion of the Human Genome Project this has decreased significantly. This also coincides with the development of new technologies, and so the gap should, in theory, continue to come down.
Most of the drugs were biologics. Additionally, they were primarily used to compensate for loss-of-function mutations. It is important to remember that this is a subset of non-cancer drugs for which genetic data was integral to approval; genetic data has also played a smaller role in the approval of many other drugs over the years.
Last month, we had the pleasure of hosting a webinar series on Genomics in Drug Discovery. Our expert speakers across the series each gave their own insightful views and unique perspectives on the topic. Talks included:
From Target Discovery to Clinical Drug Development with Human Genetics – Katerina Trajanoska, Postdoctoral Research Fellow, McGill University
Leveraging Public Genomic Data to Identify Therapeutic Targets – Stefanie Morgan, Head of Science, Watershed Bio
Discovery Forum: National Genomic Data Driving Pharmaceutical and Biotech R&D – James Duboff, Strategic Partnerships Director, Genomics England
Integrating Omics into Drug Discovery: from Target ID to Phase 3 and Beyond – Tom Lanz, Senior Director of Multi-Omics and Biomarkers, Pfizer
Advancing Drug Discovery Through Genetics and Genomics – Nikolina Nakic, Senior Director, Head of V2G2F Computational Biology, GSK
Multi-Omics Approaches to Inform Disease Mechanism and Drug Target Identification – Andrew Jarnuczak, Associate Principal Scientist, AstraZeneca
Building AI Models: Lessons Learned – Richard Lewis, Director, Data Science, Computer-Aided Drug Design, Novartis
Proteomics and Single-Cell: Applications to the Drug Discovery Process – Yann Abraham, Associate Scientific Director, Janssen
There is still a huge amount of unmet clinical need in today’s world, and the vast majority of illnesses do not have suitable drugs on the market. It is clear that genomics and other omics are going to play a key role in changing this.
But more drugs becoming simply available over time is not the entire story; the timeframe from establishing evidence to ultimately gaining approval must be dramatically decreased. This would mean ensuring that, as the potential of our technology exponentially improves, the emergence of new drugs can keep up with the incoming evidence. Ensuring that the frameworks exist to implement genetic data into every relevant step of the drug discovery process will be key, as will improving the clinical trials process, for example through better tracking of side effects.
But there are a number of key challenges that still need to be addressed before we can harness the full potential of new tools; one of these challenges is the integration of multi-omics data, something that a number of institutes are currently working on.
Another aspect that is sure to further improve the process is the use of artificial intelligence. With excellent pattern recognition and the ability to sort through vast amounts of data quicker than a human could, AI can be used to identify drug targets much more efficiently than traditional means. With the wealth of new drug targets that are produced from vast scale multi-omics data, AI can drastically reduce a list of drug targets down to those that are most likely-to-succeed.
Plus, the technology also has the potential to transform the latter stages of the drug discovery process, such as predicting side effects and integrating personalised approaches in the clinic and in trials. However, questions remain to ensure that this is done properly, and in his recent talk at a Front Line Genomics webinar, Richard Lewis (Novartis) stated that we must build models that people can trust if we are to move forward in this space.
Additionally, new omics tech can be used not only in the drug discovery and development process but in clinical trials and beyond, as a means to assess, for example, genotoxicity. An example of this was given by Tom Lanz (Pzifer) in our recent webinar.
As AI becomes more sophisticated and we solve many of the challenges related to multi-omics integration and around lesser used omics such as metabolomics, we will certainly see the drug discovery process speed up exponentially.
References and further reading
With thanks to all our speakers in the Genomics in Drug Discovery ONLINE webinar series, whose talks influenced this feature.
Chandler, R.J., LaFave, M.C., Varshney, G.K., Trivedi, N.S., Carrillo-Carrasco, N., Senac, J.S., Wu, W., Hoffmann, V., Elkahloun, A.G., Burgess,S.M., Venditti,C.P. 2015. Vector design influences hepatic genotoxicity after adeno-associated virus gene therapy. The Journal of Clinical Investigation, 125(2), pp.870-880.
Isert, C., Kromann, J.C., Stiefl, N., Schneider, G. and Lewis, R.A., 2023. Machine Learning for Fast, Quantum Mechanics-Based Approximation of Drug Lipophilicity. ACS Omega, 8(2), pp.2046-2056.
King, E.A., Davis, J.W. and Degner, J.F., 2019. Are drug targets with genetic support twice as likely to be approved? Revised estimates of the impact of genetic support for drug mechanisms on the probability of drug approval. PLoS Genetics, 15(12), p.e1008489.
Ochoa, D., Karim, M., Ghoussaini, M., Hulcoop, D.G., McDonagh, E.M. and Dunham, I., 2022. Human genetics evidence supports two-thirds of the 2021 FDA-approved drugs. Nature Reviews Drug Discovery, 21(8), p.551.
Trajanoska, K., Bhérer, C., Taliun, D., Zhou, S., Richards, J.B. and Mooser, V., 2023. From target discovery to clinical drug development with human genetics. Nature, 620(7975), pp.737-745.