A Roadmap to an Effective Cleaning Program: Validation Considerations

Once products have been grouped and worst case(s) selected, the next logical question to be answered is "How much product can remain on equipment?" or "What limit should be established for a given cleaning situation?" Another form this question takes is "How clean is clean?" There are many bases for establishing limits for cleaning processes. Some of them are: Based on medical dose combined with a safety factor Based on toxicity Based on analytical detectability Based on the process capability of the cleaning process Limits Based on Medical Dose Combined with a Safety Factor Medical dosage level is probably the most common basis for limits calculations in the pharmaceutical industry. It is based on allowing a certain fraction of a daily dose to carry over to a daily dose of a following product. The fraction that reduces the dosage is referred to as a safety factor or a risk assessment factor and takes the form of a fraction such as 1/100th, 1/1000th, or 10,000th of the original daily dose. As an example, let us suppose that we first manufacture product A and then product B and we simply ask the question "What is an acceptable level of product A to carry over and be present in product B and yet not cause a medical effect in the patient consuming product B?" We could start with some basic mathematical relationships and build upon them. This math is very straightforward and logical. The maximum allowable carryover (MACO) could be expressed simply as: MACO = (allowable carryover into a single daily dose of next product) x (# of daily doses in complete batch of next product) Or, MACO = (allowable carryover into a single daily dose of next product) x (batch size of next product)/daily dose of next product) The allowable carryover product A into a single dose of product B would be: (Daily dose of A) x (Safety Factor) Thus, the total maximum allowable carryover depends on four variables: The daily dose of product A The safety factor selected The batch size of product B The daily dose of product B Let's suppose, for the present example, that we have five products (A, B, C, D, and E) in the current product grouping, and they are all products given by the oral route of administration. Let's further assume that the group of five all has different dosages and batch sizes as represented in Figure 15. Figure 15 Doses and Batch Size Information Product Daily Therapeutic Dose Formula Daily Dose Weight Batch Size Product A 10 mg 100 mg 50 kg Product B 30 mg 150 mg 100 kg Product C 100 mg 200 mg 100 kg Product D 200 mg 250 mg 10 kg Product E 500 mg 300 mg 150 kg Since the limits calculated will depend upon the specific sequence of products manufactured, it should be noted that there are 20 different combinations and permutations of possible manufacturing sequences for only a five-product group. Even if the initial product is specified, then there could be four other products manufactured subsequently, thus four different limits. For this reason, many companies use an equation which takes into account the "worst case" situations for all products in the same group. For purposes of an example, let's select Product A as our initial product. In order to calculate a limit for carryover of product A into any other product, we could use the following equation: The worst case number of doses in the following product would be determined by a combination of the largest daily dose (including excipients) of any of the other product in the group and the smallest batch size of any other product in the group by the relationship: Using the data in Figure 15 for the current group of five products would indicate that Product E has the largest Formula Daily Dose Weight (300 mg) and Product D has the Smallest Batch Size (10 Kg) of the product group. Thus the Worst Case Number of Doses would be: Note: The 10 Kg was converted to mg so that the units would agree. Also, please note that the Formula Daily Dose Weight includes the weight of the excipients since the excipient weight is also typically included in the batch weight and these two terms must represent the same components. This value would now be substituted back into the MACO calculation with the following results: We could now perform a 'worst case' calculation for each of the five products in the group and this would reduce the number of calculations from twenty down to five. The resulting calculations for the group of five products are shown in Figure 16. Figure 16 MACOs for All Products in Group Product MACO A 333 B 990 C 3330 D 6660 E 16650 Limits Based on Toxicity In many cases, the therapeutic dose of a material may not be known. This is the case, for example, for precursors and intermediates in API facilities as well as for cleaning agents. In this case, we often turn to toxicity as a measure of the potency in order to calculate limits. This method, known as the No Observed Effect Level (NOEL) method, makes use of toxicity data from animal studies and uses some of the same product parameters used for calculations based on therapeutic doses discussed in the previous section of this article. The method is based on the work of M.L. Layton, et. al.2 The method described by Layton was developed for the purpose of establishing acceptable levels of chemicals in drinking water and air. Hall has applied the same approach, which is based on toxicity data in the form of LD50 data, to pharmaceutical products and the detailed calculations can be found in Reference 1. At this point in time, this method appears to be the only method available to establish acceptable cleaning limits for certain materials. Even though the method is based on toxicity data derived from animal studies rather than data from man, it contains several safety factors and safety assumptions that allow its use for human and veterinary pharmaceutical products...