Working on new drugs is a complex, lengthy undertaking: it takes around twelve years to bring them to market – with an average total cost of around 2.8 billion US dollars. The reasons for this are increasingly complex products and study designs, increasing requirements for documentation and safety, and the laborious recruitment of participants for clinical studies. Pharmaceutical companies shy away from designing new active ingredients such as antibiotics if they are no longer considered profitable. However, systems with artificial intelligence (AI) offer a means of counteracting this for the industry and the healthcare system as a whole, according to a current study by the Learning Systems platform.
Advertisement
The process of drug development can be made significantly more efficient with the help of AI “from the first idea to approval,” writes the team led by Klemens Budde from the Charité in the network's white paper, which was published by the German Academy of Engineering Sciences (Acatech). is hung and is funded by the Federal Ministry of Research. This makes it possible to “save years of work and costly investments.” What is crucial for this is “systematic analyzes in data processing”, for example to recognize relevant patterns from big data.
With the help of AI, huge amounts of data could be systematically analyzed and extensive knowledge could be quickly evaluated, the authors explain. This makes it possible to find suitable drug targets and candidates in a short time, make better predictions about drug side effects and optimize chemical synthesis, i.e. the production of the drug. The key technology could also help with the selection and monitoring of test subjects for clinical studies and approval. AI-based data analysis also enables the development of personalized therapies, for example for the treatment of cancer. These can be better tailored to the patient's individual clinical picture.
Challenge: Fill gaps in the database
The AlphaFold software developed by Google DeepMind allows the AI-based prediction of crucial protein structures “within a few hours with a high level of accuracy,” the members of the platform’s Health, Medical Technology and Care working group give an example. In order to achieve comparable accuracy and resolution, such molecular strands would have had to be researched experimentally, sometimes over months of work. The US biotechnology company Insilico Medicine was also able to develop an active ingredient candidate against fibrosis through to the preclinical phase using AI support for less than $850,000. Traditionally, this would have cost around $664 million.
According to the analysis, the South Korean pharmaceutical tech company Standigm has also developed an AI-based platform for identifying drugs with new mechanisms of action, which allows these structures to be identified within an average of seven months compared to typically 30 months. The authors also address the opportunities of generative AI. Med-PalM, for example, is a language model developed by Google specifically for medical questions. It supports “the intuitive, text-based query of relevant genes for specific diseases based on information organized in knowledge graphs.” As a counterpart, Exscientia has published a chatbot for producing knowledge graphs. Generative AI can also be used to create new molecules or proteins. However, hallucinations of such language models are challenging.
The experts describe general hurdles as the lack of legal requirements as well as data quality and availability. What is important is the willingness of research companies to share information. There are gaps in particular in the database on human biology, such as disease mechanisms and the effects of drugs. These could be closed with high-quality measurements of the population, which would ideally be provided via the electronic patient record (ePA) or the health insurance companies. Politicians must set the right course, for example with the European Health Data Space (EHDS) and standards such as the Health Data Usage Act. However, AI-supported research should not be torpedoed by barriers such as those demanded by civil rights activists.
(bme)