ASTM E3418-23

4.1.1 The origins for the calculation of cleaning validation limits for pharmaceuticals date back to the 1980's with the publication of an article in 1984, that stated that "limits must be safe and acceptable and in line with residual limits set for various substances in foods" (7). A second article in 1989 expanded upon these ideas adding that an "effect threshold" should be established in collaboration with toxicology and medical authorities (or alternatively, an appropriate safety factor. For example, 10× or 100× could be superimposed) and finally that limits for surface residue levels could then be calculated based on a smallest batch size/maximum dose combination. This article further mentioned that this calculation leads to many limits that could be verified through visual inspection (8). A third article in 1993, proposed the use of a combination of limits suggesting that carryover of product residues needed to meet these three criteria:

4.1.2 In 1993, United States Food and Drug Administration (USFDA) issued a guide for its inspectors requiring that "the basis for any limits must be scientifically justifiable" (6). In 1996, USFDA proposed that, in addition to penicillin, certain "classes" of compounds would also need to be manufactured in dedicated facilities and would expect manufacturers to identify any drugs that present the risk of cross-contamination and to implement measures necessary to eliminate that risk (10). Otherwise, nothing short of dedicated facilities or equipment would be sufficient. In 2005, the European Medicines Agency (EMA) similarly announced that it would require dedicated facilities for certain medicines in addition to potent sensitizers (11).

4.1.3 In response to these pending regulatory requirements, a guideline was published in 2010 by ISPE which introduced the concept of a health-based limit for calculating cleaning limits known as the Acceptable Daily Exposure (ADE) (1). The demonstration of adequate cleaning and control against ADE derived limits could avoid facility or equipment dedication. Several articles were published discussing why the dose-based and median lethal dose (LD₅₀) should no longer be used and should be replaced with the limits based on the ADE (12-15).

4.1.4 In 2015, EMA issued a guidance requiring the use of health based exposure limits (HBELs) for use in calculating cleaning limits (2). This requirement has now been incorporated into the European and Pharmaceutical Inspection Co-operation Scheme Good Manufacturing Practices (5 and 16) and has been adopted by Health Canada (17) and the World Health Organization (18).

FIG. 1 ASTM E3106 Risk (Hazard) Identification and Risk Analysis Steps (modified from FIG. 3 in E3106)

All medicinal products must have HBEls determined (2). Other chemical compounds identified as hazards to patients in the risk (hazard) identification step that cannot be eliminated or replaced should have HBELs determined if acceptable safety assessments or risk assessments are not available (see Guide E3219). After HBELs have been determined, manufacturing parameters such as batch sizes, maximum daily doses, total share surface areas, etc. are documented in the risk analysis step and used to calculate safe limits for swab and rinse samples.

4.2.1 The medical device industry is very broad and includes many diverse devices that have been handled differently than pharmaceuticals.

4.2.2 For example, cleaning acceptance limits for implantable medical devices have been based historically on testing after the cleaning process is completed by doing biological safety assessments that show that the final packaged product is safe and effective (ISO 19227, ISO 10993-1). Initial cleaning limits might be derived from historical data on the same types of devices using the same manufacturing processes and materials as a starting point and coupling that with clinical history that shows the devices produced using this methodology are safe and effective. Then biological safety testing, including extractables, is performed on devices exposed to the validated, controlled cleaning process.

4.2.3 The extractables testing also shows that the limits that were established for other manufacturing residuals that carry through on the part after cleaning are also at levels low enough to mitigate any local or systemic adverse reaction (this may be toxicological risk assessment of the extractables test data as well as biological testing).

4.2.4 While many medical devices use the approach described above, some medical devices can benefit from using an HBEL approach similar to pharmaceuticals. If an HBEL approach is used with a medical device, a risk assessment (this could be the HBEL monograph (see Guide E3219)) should address other potential risks (for example, patient exposure at the tissue level) from residue levels on the device beyond the general toxicological risk assessments typically performed for pharmaceuticals.

4.2.5 Application of the approach described within this guide applies science-based and risk-based concepts and principles for calculation of cleaning validation safe limits and performance-based limits (for example, statistical process control) introduced in Guide E3106.

4.2.6 Application of the approach described within this guide applies the science-based and risk-based concepts and principles for derivation of health based exposure limits introduced in Guide E3219.

4.2.7 Application of the approach described within this guide applies the science-based and risk-based concepts and principles for derivation of visual residue limits introduced in Practice E3263.

4.2.8 Key Concepts—This guide applies the following key concepts: (1) health based exposure limits, (2) quality risk management, (3) science-based approach, (4) statistics-based approach, (5) visual residue limits, and (6) statistical process control (SPC) limits.