Camille Bouget (M17): AI at the service of precision immunology

Reflets Magazine: What was your background before co-founding Scienta Lab?
Camille Bouget: I originally followed a scientific path, studying pharmacy in Bordeaux with the aim of becoming a hospital pharmacist. But during my studies I discovered an interest in the business side of the pharmaceutical industry, because it allowed me to gain varied experience in terms of activities, company types, and geographic scope. As part of my sixth year of industrial pharmacy, I chose the ESSEC Master’s in Marketing Management and Digital, because at that point I was already convinced that marketing—at the heart of a pharmaceutical company’s activity—would expose me to the rest of the functions and give me an overall view of how the industry works. So I joined Cergy in 2016 and, at the end of 2017, landed my first role as Product Manager for a range of biosimilar medicines in immunology and inflammation at Sandoz France, a subsidiary of the Swiss group Novartis. But I had a strong desire to live abroad—I wanted to see something different. The opportunity arose in 2018, when Sanofi was looking for a Product Manager for a V.I.E at its global headquarters in Boston to work on the marketing of a treatment for rheumatoid arthritis. The role matched exactly the skills I had developed at Sandoz, and in August 2018 I flew to Boston with my suitcases. The American adventure lasted a year and a half and continued at Sanofi’s European headquarters in Amsterdam. I believe that my time at ESSEC, followed by this experience in the United States, truly transformed my career path, making me realise I was capable of building an entrepreneurial project. That was when the idea of Scienta Lab started to take shape with my brother Vincent, an engineer from Centrale and one of the company’s three co-founders.

RM: What sparked the creation of Scienta Lab—how did the idea come about?
C. Bouget: It was 2020, in the middle of Covid. I must have been a bit restless because I was juggling several projects! More seriously, at Sanofi I was working on a rheumatoid arthritis drug for which researchers were beginning to develop AI approaches to predict efficacy in certain patients. I found that fascinating. However, it seemed a pity to apply this to just one drug when dozens exist for the disease. The idea of an algorithm capable of discriminating—among all these options—the one that would work best for each patient profile struck me as far more relevant. I spoke to Vincent, who found it interesting. Without my knowing, I had just planted a seed. A few months later he told me he’d thought deeply about it: he loved the scientific approach and wanted to make this tool a reality, but felt we needed reinforcement on the technology side. He introduced me to Julien, one of his classmates from CentraleSupélec—a true maths whizz, passionate about fencing and chess—who became the third co-founder. The project quickly gathered momentum, leading to a paper in a scientific journal and a presentation to a room of rheumatologists at the French Rheumatology Congress. Then, in 2021, I made a decision that changed everything: rather than letting the idea slip away, I decided to turn this scientific project into a genuine entrepreneurial venture. Together with Vincent and Julien, we were convinced it was time to found the company and that we shouldn’t wait. That’s how Scienta Lab was officially born, in June 2021.                                      

RM: What data do you compare?
C. Bouget: We use a large set of anonymised patient data to model autoimmune diseases, moving from the most global to the most granular levels: clinical, biological, proteomic, and even transcriptomic data. The goal is to recreate an in silico “diseased” system that faithfully reflects the complexity of the pathology. Once this model is rebuilt, we subject it to various perturbations. In practice, we use data that allows us to reproduce the mechanism of action of the candidate drug we aim to assess. This gives us a preliminary view of how the candidate might act in different pathologies—long before it is administered to a patient. This approach provides very early, valuable information on a treatment’s potential efficacy and thus guides pharmaceutical companies’ strategic choices more effectively: which target to prioritise, which patient profile to select, and in which disease area to launch a clinical trial to maximise its chances of success. By making this technology accessible to pharmaceutical and biotech companies, we enable them both to accelerate and to increase the probability of success in their R&D efforts for new medicines.

RM: You operate at the crossroads of science, medicine, and tech; how do you juggle these three worlds?
C. Bouget: I very quickly realised that our success would come from bridging two often siloed worlds—mathematical science and medical science—coupled with the right commercial positioning. Given my training, I naturally have this multidimensional outlook, with the scientific aspect on the one hand and a business layer on the other. On a daily basis, my role is to juggle these hats effectively to move the company’s projects forward. But a large part of the expertise also comes from my co-founders and my team, particularly the biomedical and technological know-how.

RM: How do you handle recruitment?
C. Bouget: Recruitment is a major challenge for Scienta Lab because we know that team quality is the key to our success. First, all recruitment is conducted collectively by all the co-founders. We have precise processes that allow us to evaluate both the skills, values, and expertise required, to ensure complementarity among the profiles we bring in. Next, we look for talent that combines two dimensions: strong domain expertise—in immunology, artificial intelligence, or bioinformatics—together with robust soft skills. With us, it’s not enough to excel in your discipline: you must also build bridges between sciences, engage in dialogue with other specialities, and contribute to collective projects. It’s this ability to connect immunology and AI that makes all the difference. Finally, we uphold a strong culture of excellence. Our hiring processes are deliberately long and demanding because we insist on selecting the right people. This high bar means that once new hires join, we can quickly give them autonomy and responsibilities, to immediately verify the match between expectations and capabilities.

RM: You’re in an extremely competitive space; what most sets you apart from competitors?
C. Bouget: The space we operate in is highly competitive, but our approach stands out clearly from most players in the market. Many of our competitors work with medico-economic approaches. They model factors external to the medicine—such as care pathways, costs, or certain demographic criteria—to try to predict its efficacy. At Scienta Lab, we take a radically different approach. We develop true scientific models of biological systems. In concrete terms, this means We integrate very precise preclinical data—such as results observed in cells or in animal models—to predict the expected effect in humans. This approach enables us to obtain a much finer, candidate-specific prediction, even when a drug is entirely new and no retrospective data yet exist. 
It’s this ability to transform raw experimental data into reliable clinical projections that sets us apart. It positions us not only as a unique player in immuno-inflammation, but also as a partner capable of concretely increasing the success rates of development programmes.

RM: Working in this way and along this axis, won’t AI also open doors to oncology, HIV, or other diseases?
C. Bouget: Oncology is indeed a good example of a possible field of application. In immuno-oncology, some treatments aim to reactivate the immune system so it destroys the tumour, whereas in autoimmunity the challenge is the opposite: to calm an overactive immune system. These two approaches speak to each other, so there are interesting bridges between the fields. However, from a technological and scientific standpoint, the issues differ considerably. Immunity is systemic: it runs through the whole body, circulates in the blood, affects different organs, and remains very hard to characterise. We don’t always know how to measure it, nor what triggers an inflammatory flare. This is exactly where our added value lies: being able, thanks to AI, to identify the right information, at the right time, in the right tissue, in order to maximise the probability of success in therapeutic development. For now, our priority is not to scatter across all diseases, but to become undisputed experts in autoimmunity. That’s already a major challenge, as these pathologies affect nearly 8% of the Western population and represent a medical need that remains largely unmet. Succeeding here would already be a considerable step forward.

RM: How do you see the future for Scien-ta Lab and for yourself?
C. Bouget: Scienta Lab’s objective is to put its predictive tools at the service of therapeutic innovation, improving care for patients with immunological diseases. I am convinced that the future of pharmaceutical research rests on our ability to make the right decisions as early as possible in a drug’s development. Scienta Lab’s true success will therefore be to remain faithful to its mission: harnessing the power of AI to predict, from the translational phases onwards, the efficacy of candidate medicines. By refining these predictions upstream, we significantly increase the probability of clinical success. That means not only reducing wasted time, resources, and investment on molecules doomed to fail, but also speeding up the arrival of new therapies on the market. In short, it offers the pharmaceutical industry a strategic tool to innovate faster and more effectively. This will be the focus of our next fundraising round: strengthening our international presence—particularly in the United States, our main market—through ambitious communication efforts and the structuring of dedicated local teams.
But beyond scientific and economic performance, the challenge remains deeply human: the more we improve the precision of these predictions, the more we help bring about a new generation of truly effective treatments, capable of meeting the needs of patients who, even today, do not always have satisfactory solutions.

Photos: Arnaud Calais