Page 87 - Assurance of Sterility for Sensitive Combination Products and Materials
P. 87

Aseptic processing   73


                 EMA proposed revisions to Annex 1 (Manufacture of Sterile Medicinal
                 Products), prepared by an international PIC/S working group that in-
                 cluded representatives from Europe, United States, Australia, and Japan.
                 The 2107 draft revision contains numerous specific recommendations
                 for risk approaches, with many more implied. This signals the openness
                 by the regulatory bodies for the industry to propose new approaches.
              •  Manufacturers are recognizing that traditional approaches may no longer fit
                 what we need: The future will involve new and more complex delivery sys-
                 tems as well as sterile ATMP (Advanced Therapy Medicinal Products), Cell
                 and Gene Therapy products, and personalized medicine. These new thera-
                 pies require aseptic processing. Traditional aseptic processing approaches do
                 not adequately address the unique needs of the manufacturing of autolo-
                 gous sterile products on a very small scale. New (and improved) science and
                 risk-based approaches are needed. The tight release criteria and the timeline
                 for these products will provide an opportunity for rapid microbiological
                 testing, and in-process and real-time release of sterile products.
              •  Better technology is here and waiting: There is a significant opportunity
                 to acquire and link data at unprecedented levels. This information and
                 knowledge can be used to better model, predict, and control aseptic
                 processes. Simulations along with virtual and augmented reality can be
                 used to significantly reduce costs and timelines for building facilities and
                 developing the process by minimizing change requests and proactively
                 addressing design problems prior to the build-out. Augmented reality
                 can provide more frequent and effective training and expedite qualifi-
                 cations with higher levels of worker skill level and awareness. Artificial
                 intelligence and machine learning can utilize data to optimize processes
                 and process control minimizing variability and improving quality.
                    It is important to note that the utilization of technology improve-
                 ments will require extensive use of automation and knowledge man-
                 agement. As processes become more automated and continuous, process
                 validation will shift from a matter of process testing and replicate runs
                 to qualification of the automated control systems that control process
                 parameters. As manufacturing intelligence and shared data acquisition
                 and utilization systems are used more, the notion of computer system
                 validation based on the US FDA Part 11 and computer software guid-
                 ance may need to change. Methodologies that have been developed for
                 the software industry [19, 20] may need to be adopted.
              •  The technology workforce is changing: It is more difficult to attract,
                 train, and retain technical resources. The shortage of skilled, qualified
   82   83   84   85   86   87   88   89   90   91   92