ABSTRACT

A n important consideration in cleaning validation is the determination of "how clean is clean enough?" The Food and Drug Administration (FDA) does not establish analy-hcal acceptance criteria for manufacturers [l]. Specific analytical acceptance criteria for target residues must be established by the manufacturer. It is important in setting acceptance criteria that the limits are scientifically justified. An arbitrary setting of limits is just that-"arbitrary9'-and may raise concerns in any regulatory investigation. This situation is somewhat complicated by the fact that sometimes the term Limit is used loosely, referring to the acceptance limit in the next product, of surface contamination, or of the analyzed sample. While all these are interrelated, they are not necessarily of the same units or magnitude. For example, in the contamination of the next product (the product

subsequently manufactured in the cleaned equipment), the units typically are ppm or pg/g; for surface contamination, the units are usually pg/cm2; for the analyzed sample, the units are typically pg or pg/g. It should be clear that limits per surface area are different units and cannot be compared directly without other pieces of information (such as batch size and equipment surface area). In addition, limits in the subsequent product may be the same units as limits in the analyzed sample, but they also are not comparable without other information such as area swabbed and the swab recovery factor. What this means is that an acceptance limit of 3.2 ppm in the subsequent product is not necessarily the same as 3.2 ppm in the analyzed sampled prepared by a swabbing procedure. It is important to make these proper distinctions when discussing residue limits. This will insure that analytical methods are properly validated considering the appropriate limits for the residue in the analyzed sample. In the subsequent discussion, the calculation of residue limits for the different parts of a validated system will be discussed. The limitations and applicability of such calculations as applied to finished drug products will then be explored.