WHEN IS PARAMETRIC RELEASE USING TERMINAL STERILIZATION NOT STERILIZATION
3. “Your firm failed to establish laboratory controls that include scientifically sound and appropriate specifications, standards, sampling plans, and test procedures designed to assure that components, drug product containers, closures, in-process materials, labeling, and drug products conform to appropriate standards of identity, strength, quality, and purity (21 CFR 211.160(b).
For example, there was no scientific justification for the sampling plans utilized for environmental monitoring in areas that your firm uses to manufacture terminally sterilized injectables. This included the frequency and locations of viable airborne particulate sampling activities, the locations of non-viable airborne particulate sampling activities, the frequency and locations of non-product contact surface monitoring, and the evaluation of microorganisms found through environmental monitoring activities. Your environmental monitoring program was insufficient to detect contamination of concern, including mold observed on the clean side of the air filters supplying air to the sterile filling areas (author’s emphasis). In addition, your environmental monitoring records do not identify the locations from which environmental monitoring samples were taken in each manufacturing area and room including, but not limited to: viable airborne particulate monitoring, non-viable airborne particulate monitoring, and large non-product contact surface area monitoring. Your incomplete records make evaluating the air quality of the filling area during the time prior to this inspection more difficult. In your response, you state that you use a formal risk assessment to justify the type, extent, frequency, and location of sampling and test procedures. Baxter also will create a procedure for periodic reassessment of that risk assessment to incorporate any relevant new or emerging information, and you will continue to conduct bioburden testing.”
COMMENT:
The FDA observed that Baxter had “no scientific justification for the sampling plans utilized for environmental monitoring…” which “included frequency and locations of viable airborne and non-viable particulate sampling activities”. Their audit of the Client also found that the “environmental monitoring records do not identify the locations from which environmental monitoring samples were taken in each manufacturing area and room…” In Baxter’s response, Baxter indicated that they would be using a formal risk assessment and create a procedure for periodic reassessment of that risk assessment.
Environmental monitoring for both viable airborne and non-viable airborne particulates has been a concern for decades. More recently ISO 14644-1/2 (1999) has discussed this requirement as has FDA 2004 Guidance for Industry re: Aseptic Processing (September 2004), revised Annex I (March 2009) and USP<1116> (revised May 2011). With all of these historic documents, it is difficult to understand why Baxter in 2012 was still not following these guidance documents. FDA has on more than one occasion cited firms for having personnel employed that were incapable of performing the function for which they were employed. Please see my Blog re: Triad Group (H&P Industries, 2010/2011) and Medimmune (U.K., approx 2005).
“Although we acknowledge your use of formal risk assessments in implementing these revised environmental monitoring procedures and bioburden testing, your response is still deficient. In your response, you downplay the product quality and safety impacts posed by the mold observed on the clean side of HEPA filters supplying air to your sterile filling areas on the grounds that products made in these areas are terminally sterilized (author’s emphasis). Please note that sterile products should be protected from microbiological contamination during processing, even when terminally sterilized, in order to minimize sterilization challenge and byproducts of excessive bioburden.”
COMMENT:
Baxter’s attempts to “downplay the product quality and safety impacts posed by the mold observed on the clean side of HEPA filters supplying air to your sterile filling areas on the grounds that products made in these areas are terminally sterilized.” The FDA further states that “sterile products should be protected from microbiological contamination during processing, even when terminally sterilized, in order to minimize sterilization challenge and byproducts of excessive bioburden.”
There exists many examples in the literature where products which are released using parametric methods to include steam, EtO, dry heat and irradiation must maintain certain ranges of microorganisms before dosing levels are modified. If these bioburdens are exceeded, then dose setting parameters must be reestablished (see AAMI and PDA). An excellent example may also be viewed on an earlier Blog wherein H&P Industries (Triad Group) had excessive Bacillus cereus within their sterile 2X2 gauze pads and the their irradiating dose failed to kill the microorganisms. Based upon the FDA 483s, the Company released the product without resetting the dose. Their data demonstrated the presence of these Gram positive spore-forming rods, but never took sufficient actions and the FDA seized their products.
While this may not be the situation in Baxter’s case, mold falling or potentially falling into LVP prior to terminal sterilization could create an excessive bioburden situation and challenge the parametric release parameters that they had historically developed.
“The products covered during the inspection of the Marion facility are made using a parametric release approach instead of sterility testing. FDA permits the use of a properly qualified and adequately maintained parametric release program to meet the intent of the CGMP regulation for product sterility testing (see 21 CFR 211.165(a) and 211.167(a)). Further, FDA considers a properly qualified and maintained parametric release program to encompass multiple, integrated CGMP systems that are in a state of control, including 1) sterilization process validation and control, 2) verification by suitable load monitor(s), 3) a validated container/closure system, and 4) an effective Quality System. Failure to meet any one of these criteria could disqualify the parametric release program. For your bioburden-based sterilization process, basic environmental control is integral to ensuring continuing sterilization process control by preventing an excessive challenge to the sterilization process (author’s emphasis). We acknowledge that your firm has been performing sterility testing on your terminally sterilized products since December 2012 as part of your corrective action plan. We also urge you to evaluate all Baxter terminal sterilization operations to verify that the overall operation of your sterility assurance program meets CGMP and drug application requirements, including, but not limited to, application commitments pertaining to the use of parametric release. Please update all of your drug applications containing parametric release provisions to incorporate your newly enhanced environmental monitoring program as application commitments. FDA will evaluate the adequacy of your environmental control and monitoring program in a future inspection.”
COMMENT:
To circumvent the parametric release Baxter has been performing sterility testing since December 2012 as part of their corrective action plan. To move from parametric release to sterility testing represents a significant step backwards. Those familiar with sterility testing recognizes that it represents a poor choice when parametric release is available. Standard Industry practices suggests that a failure rate of approximately 20% is required to detect a sterility test positive.
Since the FDA indicated that Baxter’s facility personnel first made this Observation in 2010, one must wonder how many LVP lots may have left the facility potentially contaminated. The FDA has requested that Baxter investigate this. Again, it is surprising that the FDA had to illustrate this issue to a multi-national Company that appears to be operating in multiple “silos” within their North Carolina facility.
Leave a Reply