KNect365 Life Sciences is part of the Informa Connect Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 3099067.

Informa

Building predictive models for bioprocesses with Joey Studts, Boehringer Ingelheim

Joey Studts, Global BioProcess and Pharmaceutical Development at Boehringer Ingelheim, took us through how the company is utilizing data to build predictive models to better understand bioprocesses and increase the speed of drug development.

As part of the downstream production track at BioProcess International Europe in April 2019, Studt's presentation explored how Boehringer Ingelheim are trying to better utilize their data to make better process decisions.

RELATED ARTICLE: Research Report: Cell Therapy Process Automation, Scalability and Closed Systems

“We’ve taken a multi-tier approach to that, and one of the most exciting topics around this digitalization process is the modeling,” Studts said. “We’ve collaborated with several companies to try to optimize how we’re using our data to build models, both for statistical and mechanistic models. We can one, understand our data better, and two, predict what might happen outside of our datasets.”

Studts explained that this will greatly enhance the speed of drug development, although the project is still in the early stages. “Right now, we focus more on process-understanding and being better prepared to bring a safe drug consistently to the markets, to improve our understanding of process ability and process robustness, and once we get a good handle on that, I think then we can greatly increase speed-to-market and enhance our efforts.”

 

 

Studts went on to say that he envisions this happening over the next few years: “Right now, I’m hoping with the next drug or the drug after that that we’ll bring to the market at the late stage, we’ll be able to have a much more broad understanding of our process robustness to present to our management and to the regulatory authorities.”

He explained that once Boehringer Ingelheim have proven that they’ve established the models as digital twins of their data, then they can reduce the efforts put in at the development stage. In using the models to reduce the development process, they would accelerate getting the product to the market.

 

The rise of predictive modeling in biomanufacturing

It’s not just Boehringer Ingelheim who are looking into predictive modeling programs to enhance developments. “We pride ourselves at being close to the front, but I think a lot of people are using these models, especially mechanistic models,” said Studts. “The mathematics behind the models have been around since the 50s and 60s, [so] there’s nothing dramatically new, but as the computer power advances, we can now use the models in real time and really impact [them].”

Studts highlighted that Boehringer Ingelheim have collaborated with companies such as GoSilico who develop software platforms for predictive modeling. He added that these platforms are designed for anyone in the new wave of lab technicians to apply them without the need of a computer scientist.

Studts has seen this as a prime example of how the industry has evolved from using big data as a buzzword to actually utilizing it in the biomanufacturing field: “I think it’s a small step in the right direction,” he said.

Over 2000 biopharm scientists, engineers and executives will be gathering in Boston for BioProcess International September 9-12, 2019. Explore the agenda or book your pass here

He pointed out that the industry suffers around not being able to apply big data ideas in line with academic theory and that this is because it doesn’t have data properly structured. “We have lots of data but it’s in Excel files; it’s in dispersed databases that aren’t aligned, and now I think the industry needs to really focus on getting this data in a proper structure. Once it’s structured, then we can really optimize these technologies being developed in the academics field to do big data, to really understand what this data is telling us in a thorough way.”

Studts said that he believes we’re not far off from doing that as an industry standard. “I think we’re just a couple of years away, and Boehringer, ourselves, we've been running fully automated and fully digitalized manufacturing processes for over 15 years, and so the data is there in the databases, it's just normal glitches and stuff like this need to be aligned, but we’re almost there and I think other players in this industry are not too far behind.”

This interview was filmed for BPI TV at BioProcess International Europe in April 2019.

RELATED ARTICLE: Next generation commercial data management systems for biopharmaceuticals

Get articles like this by email