Client success stories

Increasing competition and growing compliance requirements are pressuring biotech companies. Every step in the manufacturing process needs to be streamlined and made as efficient as possible in order to extract every last drop out of every batch in every process across the entire line. But batch optimization requires the constant analysis of huge volumes of data—a complex and costly operation.

A leading biotech company reached out to Aizon to identify how artificial intelligence (AI) could tackle these challenges and be used to further optimize yield. Using Aizon’s Fast Forward and Blueprint services, the integrated team was able to identify real opportunities to leverage AI in their specific environment and build a roadmap to achieve those goals.

After the development of a pharmaceutical product in an R&D environment, the regulatory approval procedures that follow result in industrial-scale manufacturing production recipes. For the biotechnology industry, in particular, such process operating conditions specifications cause considerable difficulties due to the high volatility that is natural in biological processes like upstream manufacturing.

Although manufacturers can use in- line and on-line sensors to directly measure Critical Process Parameter (CPP) variability, this will only help to identify an out-of-control process that risks entire batch rejection upon reaching out-of-specification (OOS) conditions. A subsequent major challenge is how to leverage such information to directly interact with a pre-approved recipe, and maintain a state of control while driving process deviations within established conditions (ECs) for successful drug manufacturing according to GxP guidelines.

In this context, Aizon is leading the “CPV of the Future” project to answer questions such as how to apply advanced analytics and artificial intelligence (AI) to support decision-making and maintain regulatory compliance.

A site for a large global biotechnology company was looking to explore predictive models to improve the yield of their downstream bioprocess using historical data. The site produces a rare disease medicine where the value per gram can be hundreds of thousands of dollars, making it important to optimize for every gram per batch.

Their goal was to figure out how to choose the best selection of available bags in inventory that, across 6 runs, would provide the optimal total yield of protein, without risk of sacrificing the yield of later runs by using the ‘best bags’ in the early runs of a campaign. How could they pool selection of bags in each run, and 1) fulfill the ‘hard requirements’ per the control strategy (e.g. total start protein concentration, harvest day distribution, etc.) and 2) optimize to predicted yield per run and in aggregate for the campaign?

A multinational biopharmaceutical company was looking to reduce their operational inefficiencies and reduce costs by avoiding numerous recirculations at their ultrafiltration process step. Even with a lot of effort and analysis, they struggled to get the targeted concentration of the drug product (polarimetry) correct or “Right First Time.”

Leveraging an accurate prediction model, this biopharmaceutical company has now proven the ability to go from multiple recirculations to getting the ultrafiltration process step “Right First Time.” Recirculations are now no longer necessary and they estimated a 61% reduction in the total number of runs and a 53% boost in process effectiveness. Moving forward, they now have all of the data they need for a good AI approach. The data is no longer lost and it is all captured in a GxP-compliant manner making any future audits easier, too.