Investigators from the Critical Analytics for Manufacturing Personalized-Medicine (CAMP) interdisciplinary research group at the Singapore-MIT Alliance for Research and Technology (SMART), MIT’s research initiative in Singapore, have collaborated with MIT, A*STAR Skin Research Labs, and the National University of Singapore to create an innovative technique that can quickly and autonomously identify and monitor microbial contamination in cell therapy products (CTPs) early in the production process. By gauging ultraviolet light absorption of cell culture fluids and employing machine learning to discern light absorption patterns related to microbial contamination, this preliminary testing approach aims to lessen the overall duration of sterility testing and, in turn, reduce the waiting time for patients needing CTP doses. This timely delivery is particularly vital, as prompt administration of treatments can prove lifesaving for patients with terminal illnesses.
Cell therapy stands as a promising new frontier in medicine, particularly for addressing ailments such as cancers, inflammatory diseases, and chronic degenerative conditions through the manipulation or replacement of cells to restore functionality or combat disease. Nonetheless, a significant hurdle in CTP production is the swift and effective assurance that cells are uncontaminated before their administration to patients.
Current sterility testing techniques, relying on microbiological procedures, are labor-intensive and can take up to 14 days to identify contamination, which may negatively impact critically ill patients requiring urgent treatment. Although advanced techniques like rapid microbiological methods (RMMs) can shorten the testing duration to seven days, they still necessitate complicated processes such as cell extraction and growth enrichment mediums, and they heavily depend on skilled professionals for tasks including sample extraction, measurement, and analysis. This creates an urgent demand for novel methods that provide faster results without sacrificing the quality of CTPs, adhere to patient timelines, and utilize a straightforward workflow that doesn’t require extensive preparation.
In an article titled “Machine learning aided UV absorbance spectroscopy for microbial contamination in cell therapy products,” published in the journal Scientific Reports, researchers at SMART CAMP elucidated how they integrated UV absorbance spectroscopy to develop a machine learning-assisted approach for label-free, non-invasive, and real-time detection of cell contamination during the early phases of production.
This novel technique presents significant benefits over both conventional sterility tests and RMMs, as it does away with the need for cell staining to identify labeled organisms, circumvents the invasive procedure of cell extraction, and provides results in under thirty minutes. It offers an intuitive, rapid “yes/no” assessment for contamination, enabling automation of cell culture sampling through a straightforward workflow. Moreover, the established method does not call for specialized equipment, resulting in reduced costs.
“This swift, label-free technique is intended to serve as an initial step in the CTP manufacturing process as a form of continuous safety testing. It enables users to detect contamination early and take timely corrective measures, including employing RMMs only when potential contamination is indicated. This strategy reduces expenses, optimizes resource allocation, and ultimately hastens the total manufacturing timeline,” states Shruthi Pandi Chelvam, senior research engineer at SMART CAMP and the primary author of the study.
“Historically, cell therapy production is labor-intensive and prone to operator variability. By incorporating automation and machine learning, we aim to streamline the manufacturing of cell therapy and lower the risk of contamination. Specifically, our approach supports automated cell culture sampling at specified intervals to monitor for contamination, thereby decreasing manual tasks such as sample extraction, measurement, and analysis. This allows for continuous monitoring of cell cultures and early detection of contamination,” remarks Rajeev Ram, the Clarence J. LeBel Professor in Electrical Engineering and Computer Science at MIT, a principal investigator at SMART CAMP, and the corresponding author of the paper.
Looking ahead, future investigations will seek to expand the application of the method to cover a broader spectrum of microbial contaminants, particularly those representative of current good manufacturing practices environments and previously identified CTP contaminants. Additionally, the resilience of the model can be assessed across various cell types beyond MSCs. Furthermore, this technique can also be adapted for the food and beverage sector as an integral part of microbial quality control testing to ensure that food products comply with safety regulations.