Artificial intelligence ushers in revolution in drug development
Artificial intelligence (AI) is on the rise. Its potential is huge, including in drug development. Launching a new drug takes the biopharmaceuticals sector more than ten years and costs an average of 2 billion euros. AI makes a more efficient and therefore faster development process, reduces the price tag and marks a turning point in the development of new, innovative drugs for hitherto untreatable conditions.
It’s already an accepted fact that AI will eventually revolutions the drug development process.
Liesbeth Van Rompay, Equity Analyst at KBC Asset Management
Recent developments add momentum
Pharmaceuticals companies have been using computer simulations and computing power in their drugs research for many years. There are three factors which mean that AI, rather than disrupting the traditional process, will actually provide an additional boost.
- Breakthrough in self-learning systems
Progress is the driving force behind the wide deployment of AI. A good example is Natural Language Processing (NLP): the ability to process unstructured data such as text and extract patterns from it. ‘And that’s useful in drug development’, says Liesbeth Van Rompay, Equity Analyst at KBC Asset Management. ‘NLP is used to quickly and accurately analyse scientific publications or medical data and extract pertinent information.’
- More, more comprehensive and more accurate databases
AI is totally dependent on the quantity and quality of the data used to train it. Scientific literature, clinical trial reports and medical records contain a wealth of data, but it is often difficult to access, for example because of privacy concerns, data not being clearly defined by biological variation or data being subject to sampling bias. Progress is being made, however, as disparate datasets are becoming better aligned. In addition, biopharmaceutical companies, academic research groups, government and hospital research groups are using new tools to generate huge datasets. These include things such as remote health monitoring using wearables such as smartwatches, or automatic insulin pumps. The flow of new data being generated is huge.’
- Increased computing power
Increased computing power has enabled AI researchers to analyse and process the vast amounts of data faster, both in the training and application phases.
Pioneering protein research: a foretaste of what may come
Proteins are like the workhorses of biology. Once their shape and structure is this creates a better understanding of what they do. AlphaFold, an AI programme developed by DeepMind, the research arm of Alphabet (Google's parent company), presented groundbreaking research in 2022 predicting the 3D structure of a protein based on its genetic code. ‘Proteins are large, complex molecules that play a role in just about every function in our bodies, for example muscle contraction, or converting food into energy. The role of a protein is determined by its unique 3D structure. The ‘recipe’ for building proteins is contained within genes, which are encoded in our DNA. An error in the genetic recipe of a protein can result in a malformed protein and thus lead to disease. The genetic recipe for a protein does not automatically dictate its shape. And although AlphaFold cannot predict a protein's function, it does help scientists develop drugs based on a protein's unique shape’, says Van Rompay.
The database created by AlphaFold gives scientists a wealth of information. We are now finally beginning to understand the biological machinery of some 200 million proteins. That opens the door to fighting diseases like cancer, Alzheimer's and Parkinson's.
Liesbeth Van Rompay, Equity Analyst at KBC Asset Management
Immense added value, including economically
As well as its potential in scientific research, AI also adds value in the rest of the chain. ‘An expensive process that would normally take several years can be significantly shortened by using AI. This makes speeds up the process of determining whether or not a particular medicine is effective. And, of course, that is also good news for patients’, says Van Rompay.
AI currently supports 16% of drug research, and that percentage could double over the next three to five years.
The economic value of speeding up time-to-market is immense and also creates scope to reduce the development costs for drugs to treat rare diseases, which affect fewer patients and therefore have smaller market potential.
Liesbeth Van Rompay, Equity Analyst at KBC Asset Management
A new partnership is born
AI is and will always be software; it is humans who hold the reins. AI alone will never be able to develop a new drug from A to Z. However, new expertise is needed, in addition scientific expertise: engineers, AI and data specialists. Most pharma companies do not develop that expertise internally; the advances in AI and machine learning are moving too fast for that. That means that companies working on their own risk missing the boat. A better alternative is to work with qualified partners who can build the necessary software solutions.
The biotech industry is complex and requires specialised knowledge. Building a bridge between biological sciences and insights from IT stimulates innovation and takes the drug development process to the next level.
Liesbeth Van Rompay, Equity Analyst at KBC Asset Management
‘We are seeing more partnerships now that companies have been convinced of the added value of AI’, says Van Rompay. ‘Pfizer, Bristol Myers Squibb and AstraZeneca, to name just a few, are partnering with start-ups in their search for new drugs. Over the past two years, the potential value of partnerships was around 32 billion USD. Right now, there are no fewer than 25 AI-driven drugs in the pipeline.’
Growth in the AI-driven drug industry is assured thanks to the many new market entrants, significant capital investment and a continuous technological innovation tsunami.
Liesbeth Van Rompay, Equity Analyst at KBC Asset Management
‘Thanks to AI, the world of drug development faces unprecedented opportunities. We are still in the early stages, but it’s already clear that AI will lead to a revolution in drug development. Investors wanting to get a piece of the action will find that we at KBC Asset Management will be delighted to guide them in their search’, concludes Van Rompay.
Want to read more about thematic investing?
This article is informational only and should not be considered investment advice.
