A firm's justification for the frequency of media fills in relation to shifts should be risk based, depending on the type of operations and the media fill study design. For closed, highly automated systems run on multiple shifts, a firm with a rigorous media fill design may be justified to conduct a lower number of total media fill runs. Such a program can be appropriate provided that it still ensures performance of media fills for each aseptic processing line at least semiannually. The 2004 guidance for industry on Sterile Drug Products Produced by Aseptic Processing states that "[A]ctivities and interventions representative of each shift, and shift changeover, should be incorporated into the design of the semi-annual qualification program." In addition, the EU Annex 1, Manufacture of Sterile Medicinal Products, states that "Normally, process simulation tests should be repeated twice a year per shift and process."
Certain modern manufacturing designs (isolators and closed vial filling) afford isolation of the aseptic process microbiological contamination risks (e.g., operators and surrounding room environment) throughout processing. For such closed systems,1 if the design of the processing equipment is robust and the extent of manual manipulation in the manufacturing process is minimized, a firm can consider this information in determining its media fill validation approach. For example, it is expected that a conventional aseptic processing line that operates on two shifts be evaluated twice per year per shift and culminate in four media fills. However, for aseptic filling conducted in an isolator over two shifts, it may be justified to perform fewer than four media fill runs per year, while still evaluating the line semiannually to ensure a continued state of aseptic process control. This lower total number of media fill runs would be based on sound risk rationale and would be subject to reevaluation if contamination issues (e.g., product nonsterility, media fill failure, any problematic environmental trends) occur.
