Karin Conde-Knape, Senior Vice President of Early Development at Novo Nordisk, gave an insightful talk on how to use different tools and methodologies to advance drug discovery. Conde-Knape called for a better understanding of the diseases that scientists are trying to treat and the importance of meeting a high standard of care, and when possible, surpassing this.
Furthermore, it is paramount that there is a strong focus on understanding disease mechanisms across interconnected organs like the heart, liver, and kidney rather than treating them in isolation. By combining and analysing large-scale multi-omics data and real-world evidence, scientists can identify residual risks and therapeutics.
At Novo Nordisk, the main therapeutic focus is on cardiovascular and rare diseases. Haemophilia is a rare disease with significantly high unmet need despite the existing standard of care. The company is currently developing a treatment for haemophilia patients using Mim8, a coagulation factor 8 bispecific antibody that assembles with activated factor 9a and 10 to initiate a coagulation cascade and prevent bleeding episodes. The team is also leveraging generative AI and patient-centric data to tweak treatment strategies and redefine heart failure subtypes.
However, over the last 7 years, Novo Nordisk has been on a journey to diversify its pipeline. The company is renowned for its foundation in insulin biology and its transition into hormone therapies with GLP-1 analogues.
Conde-Knape partially attributed Novo’s success with GLP-1 analogues to clever and thoughtful use of AI/ML. These tools enable scientists to move from target validation to human testing and market approval in a more efficient, speedy, and cost-effective manner. Yet AI/ML is not at the stage where it can answer all of our problems. Conde-Knape expressed that in the context of target identification, AI/ML is not quite where it needs to be.
Alongside AI/ML, Novo uses high-throughput technologies, robotics, digital activities, and collaborations with partners like Microsoft and Valo to accelerate target discovery, molecule generation, and validation, supported by access to powerful computing and diverse data sets. Novo Nordisk even has access to computing services out of GEFION, Denmark’s AI supercomputer.
Novo’s efforts over the last seven years have enabled them to increase their scalability by over 80%. Data availability is not an issue in biopharma, but there are respective challenges when it comes to making data usable. Novo Nordisk has access to the SELECT data set, the largest cardiovascular outcome trial conducted on patients living with obesity and cardiovascular disease.
Conde-Knape outlined the significance of the trial: “This is the first time in the space connecting weight loss through an intervention, a pharmaceutical intervention, where it really demonstrates that lowering body weight plus additional mechanisms leads to cardiovascular protection.” From the 17,000 individuals from whom samples were collected, proteomics, metabolomics, epigenetics, and genetic analysis of these individuals have been conducted.
Finally, Conde-Knapper concluded by reiterating that the main challenge in drug discovery is not the lack of tools or data, but the need to ask the right scientific questions and define the right strategies for target identification and validation. She also stressed the importance of early clinical translation and collaborative efforts.
