The immune system, our body’s natural and elite guardian, is connected to almost every type of disease the body encounters and can even build up its own memory, so that—even years later—it can recognize unwanted threats from previous attacks and remember how to respond. Scientists at Sanofi, a global healthcare company, are inspired by the wonder of the natural world and are working to figure out how humans can develop something as elegant as what organically exists. With the help of technology like artificial intelligence (AI), its teams are pioneering the field of immunoscience, which studies the role the immune system plays in how diseases work, and how to treat and prevent them.

Today, there’s an immunoscience renaissance happening, and Sanofi is pushing medical research in this field forward at a much faster pace than was ever previously possible. With the use of new technology, it is helping to make connections between seemingly unrelated diseases, unlock treatment options, and usher in a new era of drug discovery.

Houman Ashrafian, Sanofi’s executive vice president, head of research and development, looks at this from a panoramic view of what’s possible. “Can we apply immunoscience to areas beyond the classic diseases? Can we address the non-canonical aspects of disease—frustration, fatigue, depression? Can we impact the great health challenges of our time—obesity, mental health, infectious disease, cancer—through our palette of tools for the immune response?” said Ashrafian. “The answer is going to be yes.”

By using AI to supercharge research, scientists can find new ways to effectively harness the immune system, which is crucial in accelerating treatments. This innovative, tech-powered approach to looking at human health holds the potential to change the future of care across multiple disease areas and help get more effective treatments to patients sooner.

In a recent clinical trial, Sanofi researchers were able to fast track clinical development as a result of their confidence. “We proved clinically that our in silico predictions were accurate, and then we were able to pivot into a later phase study with more patients. That saved us a phase and years in development time,” said Helen Merianos, global head of R&D strategy and portfolio management at Sanofi.

The Possibilities of Immunoscience

The therapies that have had the biggest impact on health and disease today are a result of manipulating the immune response. From the 1950s to the 1980s, there was a surge in immune system research, resulting in important discoveries: like the concept of the body learning to recognize its own cells and tissues, which prevents it from launching an immune response against itself. Over the last few decades, researchers have figured out how to apply the immune response therapeutically, with great results. “The single biggest impact in public health of our time is vaccines—it’s immune response,” said Ashrafian.

According to the World Health Organization, 41 million people die per year from non-communicable diseases such as cancer, MS, asthma, and diabetes—all of which immunoscience research and breakthroughs can continue to have considerable impacts on. The potential of immunoscience—of understanding and exploiting a sophisticated process that regulates the majority of biological processes in the body—is vast.

“We’ve got the toolbox,” said Ashrafian, “It’s the first time I can say, ‘I can make an impact on a patient's disease,’ and AI is a critical part of that.”

Emmanuel Frenehard, chief digital officer at Sanofi, thinks that this next generation of immunoscience will be able to better treat the more common diseases, and also deliver targeted, precision treatment to people suffering from lesser-known conditions. Statistics estimate that 300 million people around the world are living with a rare disease, and only 5% of known rare diseases have an available treatment. In this new era, fewer people may fall through the cracks and be able to get the care they deserve. “The future in this space is limitless,” said Frenehard.

How AI is Enhancing Research and Development at Sanofi

AI is being used across the value chain at Sanofi to get treatments to patients sooner. It’s driving research and development forward, accelerating discoveries and clinical trials, creating better understanding of disease, and helping to ensure manufacturing quality.

Some 22,000 employees have access to a platform where they can see AI-powered insights from all segments of the company, with teams leveraging more than one billion data points that help with decision making. While AI can require massive amounts of computing power, Sanofi is committed to Eco-Responsibility—serving as one of five pillars in its Responsible AI policy—with their teams assessing and monitoring the environmental risks of AI systems before deployment and even using AI to reduce the impact of their processes.

According to Merianos, AI is a tool that’s used for the sole benefit of innovating on behalf of patients to find the optimal path to treatment. One example of this is the use of AI to conduct in silico trials, or virtual clinical trials that use modeling and simulations to evalsuate new medicines for efficacy and safety before they’re tested on animals or humans. “AI can help make us work faster so we can get transformational solutions to patients more quickly,” she said. “It’s also making us smarter, by ensuring that we’re identifying and matching the right medicine to the right patient.”

As of August 2024, Sanofi has 78 clinical-stage projects, 33 of which are in phase 3 or have been submitted to regulators for approval. These include projects that use new molecular entities and those using existing products in new indications or different combinations.

Once products reach the manufacturing stage, AI is used to validate product batches, and detect potential quality control issues or defects. If advanced models detect deviations, it reports them to a human for review. “We always make sure there's a human in the loop; AI is here to be a sidekick, not to take over the decision from the human,” said Frenehard, emphasizing how crucial it is to ethically and safely integrate this new technology. “We are very public about adopting a responsible approach to AI.”

The Next Generation of Medicine

AI-powered immunoscience can push the boundaries of discovery, increase the probability of success, and expose fewer patients to unnecessary interventions. Sanofi recently released a large language model that’s fed by 10,000 mRNA sequences from a diverse set of organisms. This large, open-source data set helps scientists overcome a common barrier in mRNA vaccines: correctly predicting sequences. Frenehard thinks it will cut the prediction time in half. “It’s a perfect example of biology and technology working together,” he said. “We made it open source so it’s available to everyone because we want to give back to the scientific community.

Another way AI is helping overcome the challenge of long drug development cycles is by digitizing decades of experiments so scientists can use existing knowledge to find optimal treatment paths. By analyzing large quantities of data, scientists can uncover the immunological links, like common disease mechanisms or pathways, between seemingly unrelated conditions.

While clinical development has a 90% failure rate for new drug candidates, AI and technology are helping scientists improve this. Beyond learning from what’s been done in the past through digitized research, researchers are also using digital twins, or digital representations of things like livers or hearts, to simulate treatment to see the toxicity and efficacy of a certain therapy in the body during in silico trials with virtual patients.

This can ultimately lessen the potential negative impact on humans. “We’ll do everything to not go into animal and human trials until we’re as sure as we can be that it will work,” Frenehard said.

In the future, deep learning will help anticipate the unknowns in complex diseases like cancer, minimizing potential negative experiences or outcomes for patients who participate in clinical trials. Eventually, the goal is to reach a point where a phase 3 clinical trial will be a formality because everything—from molecular design to efficacy and safety testing—will have been simulated already with success. It is only by pushing the frontiers of science and pioneering new approaches like AI and immunoscience that we will ignite the discovery of novel therapeutic options for patients across the spectrum of human disease.