Camille Bouget discusses how artificial intelligence is influencing innovations in the treatment of immune-mediated diseases.
“Immuno-inflammatory diseases are often described as a niche. They are not,” explains Camille Bouget, CEO and founder of the health care startup Scienta Lab.
Since as many as 10 people in Western countries may be affected by the immune system, he noted, symptoms tend to worsen, with treatment options available frequently, they are seriously inadequate for the majority of patients.
That is why in 2021 Bouget co-founded Scienta Lab, a biotechnology company that aims to advance research within immunology with modern technologies such as artificial intelligence and EVA, a multimodal AI organizational model designed for translational research in immunology and inflammation.
He said, “It was designed to answer the pressing questions that R&D teams face during drug development: which therapeutic targets should be pursued, which biological markers are strong enough to advance to clinical trials, and which patients are most likely to respond to a given drug?”
Used in many stages of the pipeline, he explains, early EVA can measure the effectiveness of treatment before the patient begins treatment. As the program progresses, EVA can assess whether the molecular signals observed in animal studies are likely to translate to humans and in the clinical phase, it will support the identification of patient subgroups for more precisely designed trials.
“The main beneficiaries are biopharmaceutical and biotech companies that work in immunology and inflammation. In the end, however, the bottom beneficiary is the patient: better designed tests, fewer failed programs and faster access to treatments that address unmet needs in diseases such as rheumatoid arthritis, lupus and inflammatory bowel disease,” he said.
Why AI?
The constant challenge of Bouget and the industry in which he works, has been to communicate properly the complexity of what they do, in a way that is accessible to all major stakeholders, be it investors, partners or the general public.
“Diseases affecting the immune system are known to be very difficult to identify”, he noted, since often even experts do not know how to measure them, what may cause an outbreak or why treatment works for one patient, but fails to work for another.
He said, “Convincing people that AI can navigate that complexity without over-promising takes a lot of effort”, especially when you’re young. deep-tech in the industry that is currently dominated by those who call the main important players.
He is of the opinion that what is said to patients is important, not least because the immunology drug development pipeline has historically suffered from delays and programs that fail in phase II or III after years of investment and work.
“Each of these failures represents not only financial costs but delayed or denied access to potentially effective treatments. AI that proactively improves the accuracy of preclinical decision-making can meaningfully shorten that timeline and shift more resources to more likely candidates.”
Strong foundations
But it’s not just a matter of access to advanced technology. For Bouget, multidisciplinary teams of scientists and engineers are critical to the overall success of any organization attempting to revolutionize immunology research and development.
He said, “Multidisciplinary teams are the perfect foundation for doing this well. The failure mode we see most often in the use of AI in drug development is the disconnect between the computational capabilities of the model and its biological relevance.
“A model trained without a deep immunological understanding may be ready for a negative signal. Conversely, a team with outstanding biological knowledge but limited machine learning skills will have difficulty extracting meaningful structures from the scale of data generated by today’s multiomics.”
At Scienta Lab, the founding team consists of a chemist and former industrial strategist, a biomedical engineer and a mathematician with deep AI expertise.
He explained, “Day by day, our team combines immunology, bioinformatics, machine learning and clinical pharmacology. The ability to build bridges between those fields, to have a conversation where the wet-lab immunologist and the transformer builder learn from each other sincerely, is what allows us to build technically and biologically robust models.”
Bouget added, “Organizations that try to solve this problem through pure data science or biological science will reach the top. The translation gap in drug development is fundamentally not a data problem or a computer problem alone, but an understanding that requires truly integrated teams.”
Don’t miss out on the information you need to succeed. Sign up for Daily BriefSilicon Republic’s digest of must-know sci-tech news.
