Blog

Pioneering the use of machine learning
algorithms to turbo-charge drug development

March 1, 2022

Anja-Vanessa Peter

5 Min read

Novel therapy development is expensive, slow, and largely unsuccessful, making it critical to find new ways of bringing medicines to market. MTIP portfolio company, Intelligencia is helping pharmaceuticals and biotech to manage the risk of clinical development and optimize portfolio strategy.

For more than a decade the pharmaceutical and life sciences sector has been scratching its proverbial head around an existential question – how long could the industry’s business model continue in the face of exponentially rising costs and time in the development of novel therapies?

Recent research reports estimate the time to bring a molecule from discovery to market is about a decade while the average cost of the R&D process is between $1.3 billion and $2.1 billion, staggeringly expensive given that a Deloitte study outlined that the expected return on investment from drug development has declined to less than 2%, meaning that ways to reduce the cost of bringing new drugs to market is imperative for the entire industry (Measuring the return from pharmaceutical innovation 2018, Deloitte).

How did Intelligencia’s journey start?

In 2014, members of Intelligencia’s leadership team, including co-founders Dimitrios Skaltsas and Vangelis Vergetis, and COO, Rezzan Köse worked at management consultancy, McKinsey, with clients in the pharma, healthcare and life sciences industries.

“The problem statement was very clear: how do we tackle the insurmountable amounts of time and money that go into drug development? Up until that moment, it had been tackled with the bricks, mortar, and people model but we started to get the question from clients: how can we be more analytical in our approach? Are there other ways to solve this problem? Is there room for applying data science because all the clinical trial data legally needs to be registered and is publicly available,” said Rezzan Köse.

Enter Intelligencia, the brainchild of Skaltsas and Vergetis, with the aim to innovate the drug development process by projecting the likelihood of success using publicly available data, advanced analytics as well as machine learning to assess, quantify and reduce the risks associated with clinical development.

The company first focused on the world’s largest pharmaceutical therapeutic area, oncology, which consistently has large numbers of trials taking place and higher success rates being observed. Intelligencia built a team with expertise in immuno-oncology that could make the associations between a target and an indication. Their first task was to create data sets for the machine learning algorithms they had written to undertake predictive modeling.

“Their high-quality and expertly curated database and their intuitive user interface impressed us during our due diligence last year. Both were recently highlighted by four of their life sciences customers while explaining what they liked about Intelligencia’s solution.“ said Danchen Zhao, Investment Director at MTIP.

The power of Machine learning

While there is still skepticism in some quarters about how much machine learning will change the face of pharma and life sciences, the COVID-19 pandemic started to turn the dial with AI used to save scientists’ time and accelerate the slow discovery process. Recently, professor of medicinal chemistry at King’s College London, Miraz Rahman, told the Financial Times he believes that in the next five to ten years AI will be completely integrated into drug discovery with all the data suggesting it is here to stay.”

Köse believes that when she talks to people about what Intelligencia does, the best approach is to outline the company’s blind experiments and how their predictions align with the drugs that have been successful or failed. “Basically, we can close our eyes and run our algorithm which will tell us if a drug has been approved or not and when we compare the results more than 80% of the predictions match what happened. We’ve used various algorithms with test sets and done the blind tests to measure the overlap. We know that this works,” she explains.

Intelligencia’s platform allows customers to:

  • assess the probability of success of pipeline drugs, with supporting rationale;
  • unlock optimal trial design, with comprehensive benchmarks linked to clinical trial success;
  • identify and prioritize additional potential indications to explore for each asset;
  • guide business development decisions, and;
  • identify emerging technologies and scientific breakthroughs to gain competitive insights and enable the early identification of “the next big thing”.

One of the most important challenges, Köse says, is to clearly articulate Intelligencia’s value proposition for a product that is complex and requires an understanding of both biology and data science to appreciate its full power. “Our technology is a decision support tool that helps executives and stakeholders to make better decisions, ensuring that they don’t fall into certain biases that may not be founded in reality. What we are saying is that if company X has a roundtable with 12 stakeholders, add one more seat for our AI algorithm and use it as another data point when you are making your decision. Use its power and understand why it says 60% as opposed to 80% or why it says 80% as opposed to 30%.”

2021 and their “Moneyball-moment”

Forbes recently included Intelligencia on its list of America’s top 50 most promising AI companies with co-founder Vangelis Vergetis telling the magazine that “biotech needs to catch up to baseball, and its own Moneyball moment is here”, a reference to the 2011 film in which a small-budget baseball team used advanced analytics to outperform expectations.

In August 2021, MTIP announced MTIP Fund II’s fourth investment in Intelligencia, with others including Big Pi Ventures and Synetro Group. This transaction is part of the rapid acceleration of venture funding in the use of artificial intelligence and machine learning innovations in diagnostics and drug discovery which, according to the financial data and software company, Pitchbook, will be a $131 billion market by 2023.

“There is an increasing understanding of how machine and deep learning is emerging as a crucial tool for transforming the process of drug development and this has been accelerated by the pandemic. We need to bring this urgency to drug discovery for all patients and at MTIP this investment aligns with our philosophy of supporting innovation that enhances patients’ quality of life and life expectations across therapeutic areas,” said Dr. Christoph Kausch, MTIP Managing Partner.

How Intelligencia is leading the machine learning revolution

It’s a sentiment seconded by Köse as she reflected on her journey from her time at McKinsey to her Intelligencia role now, “There is an overwhelming amount of data out there with so much potential waiting to be uncovered and the faster we apply machine learning to the things that are not humanly possible, the faster we can change people’s lives. The potential of curing all diseases is out there, it’s a matter of being able to access that information and that knowledge and cracking it. We are so close and so far at the same time and that fascinates me.”

Looking ahead Köse sees Intelligencia as leading the pharma machine learning revolution, “I want us to be able to stop and look back and say, do you guys remember how drug development was happening before? How it was taking 20 years? How it was extremely risky, and how all of that has changed because of our product? I see us as a leader and trailblazer and I’m so incredibly proud of that.”

MTIP invests in Intelligencia

Learn more about the author:

Learn more

Download Article as PDF:

Download

Share the article:

LinkedIn Button

Media

Trends and insights
from our health experts.