Machine learning model enables detection of adventitious agents in cell cultures

Researchers have developed a rapid, label-free process analysis technology called the anomaly detection model to monitor microbial contamination in cell cultures in near real time.

Futuristic design of a glowing blue artificial intelligence/machine learning brain surrounded by circuit board

Using machine learning (ML), Singaporean researchers have developed a novel process analysis technology (PAT) for detecting adventitious microbial contamination in mesenchymal stromal cell cultures (MSC), enabling rapid and accurate testing of cell therapy products intended for use in patients .

The anomaly detection model uses ML to predict whether a culture is infected or not in a near real-time-like manner. According to the developers, the “breakthrough method” could be used during the cell manufacturing process to overcome some of the inefficiencies of endpoint testing.

More and more cell therapies are being developed or approved for a range of applications, with promising results in the treatment of cancer, autoimmune diseases, spinal cord injuries and neurological disorders, among other indications. As their utility increases, manufacturing methods and processes are continuously refined to ensure the safety, efficiency and sterility of these products for patient use.

The anomaly detection model, a fast, label-free PAT, was developed by the Critical Analytics for the Production of Personalized Medicine (CAMP) Interdisciplinary Research Group (IRG) at Singapore-MIT Alliance for Research and Technology (SMART), MIT’s research company in Singapore.

The ML model was developed by collecting sterile cell culture media samples from a series of MSC cultures with different culture conditions. Some of the collected samples were spiked with different bacterial strains in different colony forming units (CFUs). Using ultraviolet-visible (UV-Vis) spectrometry, the team obtained the absorption spectra of the sterile, non-spiked and bacteria-enriched samples, which were then used to train the model. Testing the ML model with a mixture of sterile and bacteria-enriched samples demonstrated the performance of the model in accurately predicting sterility.

“The practical application of this discovery is enormous: when combined with at-line technologies, the model can be used to continuously monitor cultures grown in bioreactors in good manufacturing practices (GMP) facilities. Consequently, GMP facilities can provide sterility testing for bacteria in spent culture media faster with less manpower in closed loop operations.Finally, patients receiving cell therapy as part of their treatment can rest assured that products have been thoroughly evaluated for safety and sterility,” said Shruthi Pandi Chelvam, lead author and research engineer at SMART CAMP, who collaborated with Derrick Yong and Stacy Springs, SMART CAMP Principal Investigators, on the development of this method.

According to the developers, their anomaly detection model can be used to detect the presence of adventitious microbial contamination in cell cultures within minutes, overcoming one of the key challenges in cell therapy production – the long (multi-day) incubation periods required for sterility testing. Other benefits of the in-process method are resource savings, as traditional methods usually test end products and if contamination is detected, cultures must be immediately discarded and reconstructed.

“Our increased adoption of machine learning in microbial anomaly detection has enabled us to develop a unique assay that performs rapid in-process contamination monitoring, marking a huge step forward in further streamlining the cell therapy manufacturing process. In addition to ensuring the safety and sterility of cell products prior to infusion in patients, this method also provides cost and resource effectiveness for manufacturers as it allows for a decisive batch restart and shutdown if the culture is contaminated,” added Yie Hou Lee. , scientific director of SLIM KAMP.

In the future, CAMP aims to develop an in-process monitoring pipeline in which this anomaly detection model can be integrated with some of the in-house at-line technologies being developed, which would enable periodic culture analysis using a bioreactor.

The anomaly detection model was explained in an oral summary (abstract 3), recently published in Cytotherapy

Leave a Comment

Your email address will not be published.