Drilling in on Disease Hallmarks: New Tricks to Solve Age-Old Problems

Ann Beliën, PhD, founder and CEO, Rejuvenate Biomed

The pharmaceutical industry has seemingly fully embraced artificial intelligence. While there are many companies out there that are simply looking to incorporate existing AI systems into their operations, some of the most exciting developments have been coming from organizations that are building their own AI platforms. The technology is incredibly useful in a variety of ways, which means that AI platforms built from the ground up can be designed to perform specific tasks very well.

This is a particularly valuable channel for pharma and biotech companies that are focused on developing treatments for individual conditions or diseases.

Ann Beliën, PhD, is the founder and CEO of Rejuvenate Biomed, a biotech focused on developing treatments for age-related conditions. The company’s approach to this is developing combination medications of existing drugs so that they can be applied to complex diseases and conditions.

“I decided to establish Rejuvenate Biomed after a very personal experience dealing with my father who was aging really fast,” says Beliën. “Each time when we were trying to help him with one age-related disease, another would pop up. I didn’t understand what happened, why this cascade of health problems seemed to have been initiated. As a researcher, I wanted to understand this and learned about the hallmarks of aging that describe the biology of aging.”

According to Beliën, no medication to date had been developed with this information as a base.

“My parents taught me that if you want to make a difference, you don’t complain that other people are not doing it. You do it yourself,” she says. “So, I got all of my courage together and started Rejuvenate Biomed with the aim to make a difference in age-related diseases. I figured out how we could do that in a different way than I had been doing drug development in the past—by taking into account the knowledge of the biology of aging.”

BUILDING PLATFORMS

In order to achieve these goals, Rejuvenate has developed two integrated AI-driven and in vivo platforms. This technology is used to explore data and find ways to treat multiple conditions by identifying unique combinations of medications that impact complex dependent processes, something which is impossible for the human brain to do, explains Beliën.

“AI was absolutely instrumental to allow us to do what we do today,” she says. “We only develop combination drugs, and this is because of the complexity of the diseases that we deal with. The biology of aging has very neatly laid out the hallmarks of aging. Typically, there are 12 general processes (called hallmarks) where you see that damage is occurring while aging. Within these processes, there are a lot of pathways and genes that are being dysregulated. What we want to do is have combination drugs that can have an impact on multiple pathways and genes simultaneously.”

Beliën adds that it is important to ensure these combinations are safe for different patient populations; therefore, there is the need to work with compounds that have human safety data already available.

“You only have human safety data for compounds that have been used a lot in humans,” she says. “As a result, we only use products that have been around for quite a while, for which we know that they are safe for a chronic use in the older patients. The biological knowledge that you have about these compounds is very broad, so to map that knowledge to all the biological knowledge that we have on the hallmarks of aging, creates a multi-dimensional space that we cannot decipher with our normal brain.”

According to Beliën, the process of developing the AI platforms initially focused mostly on data collection, digestion, and linking. As the amount of data in their library grew, Rejuvenate started to incorporate machine learning into the platform to study the interconnection at the next level of interaction. Eventually, AI was added so that it could identify multi-dimensional connections between data, analyze them, detect previously unrevealed patterns, and then provide users with an output that they can comprehend.

“It’s a lot of different technologies that we bring together into the system so that it can make connections that we would typically not see by traditional means of looking at the information,” she says. “The system can now give us a deep understanding of the information and predict which compounds can have an impact on multiple hallmarks simultaneously. It can also help us to predict for which indication certain combinations are best positioned, and that’s how we identified, amongst others, our lead compounds.”

SIMPLIFYING DATA

Beliën sees these platforms becoming much common in the coming years. The reason why is simple: the pharma industry is built on data and these platforms can make utilizing that data much easier.

“Everybody needs to have AI tools,” she says. “There’s so much information out there and the complexity of the information will never allow us to identify new patterns like machine learning and AI can do. Of course, as human beings, we will need to do the validation of the output and the interpretation of all the information that comes out of the AI. So, we must understand where we can use AI-tools and, just as importantly, where we cannot use them.”

To distinguish between the two, Beliën stresses the importance of leveraging people that have strong biological knowledge and the scientific background to interpret and validate these tools when used in the pharmaceutical setting. Those individuals can include data computational biologists, data engineers, and machine learning experts.

“We also need people to make any additional adjustments to the algorithms that are being used in that setting, because the system will basically do what you ask it to do,” says Beliën. “You can make it smarter to learn from its own analysis by telling the system what is correct and what is incorrect. There’s always the human factor that plays a role, but having this ability—it’s tremendous. Now you can search rooms full of data in seconds, where before you could only do smaller things like look at publications to find information.”

THE HUMAN ELEMENT

Platforms like these can be used to identify new potential targets within diseases, to expand insights in diseases, and to discover new indications for existing molecules. According to Beliën, the technology is equally capable across all these processes. This is due to the fact that while the technology can make predictions, it’s ultimately on the people using it to make the actual determinations on which questions need to be answered.

“The way we have set up our system is to help us identify new combinations for age-related diseases, but we also use it to identify new targets, pathways within disease, and unidentified interactions between compounds, which can lead to identification of unique characteristics,” she explains.

That includes using AI to prioritize an organization’s own individual portfolio. “Imagine you have five different compounds for one target,” says Beliën. “You create data in the lab, you put that back in your system, and now your system can tell you which of these five has the biggest potential to impact the target. It’s really a multi-applicable tool: depending on the need, a human will interrogate broadly the system and refine the questions to get to a validated answer. So it’s not a standalone tool where you just ask one question and it gives you the definitive answer. It’s like an interaction between human beings and AI, enabling technologies and validation in labs.”

By using these platforms, Rejuvenate is able to look at information and find patterns in ways that, Beliën says, the human brain would never be able to do on its own. By combining these new insights with knowledge that researchers already have, the solutions are starting to become predictive, she adds.

“What I envision is that in the future, you will have a personalized approach. This will allow you to understand what is happening in your body when you grow older,” says Beliën. “It is important to understand your personal health data and make observations of what goes wrong in your case, so that you can preemptively get on the right medication or prescription drugs that will prevent you or your organs from deteriorating.”