I wrote several Cleaning Memos last year about the importance of using terminology more carefully. While it may seem to be more of a problem now (perhaps because of the widespread dissemination of information on the internet), it is not something new in the arena of cleaning validation. In this Cleaning Memo I will focus on two examples from the year 1993, the year sometimes considered the “birth” of cleaning validation (although clearly cleaning validation was “conceived” much earlier).
The first is the publication “Determining Cleaning Validation Acceptance Limits for Pharmaceutical Manufacturing Operations” by two scientists at Lilly (Gary Fourman and Michael Mullen). It was published in the April 1993 issue of Pharmaceutical Technology. It provided the fundamental basis for calculating limits for pharmaceutical cleaning validation by proposing limits on a carryover basis of the more stringent of 0.001 of a dose and 10 ppm, as well as requiring that the equipment be visually clean. The dose criterion is given as “No more than 0.001 dose of any product will appear in the maximum daily dose of another product”. The 10 ppm criterion is given as “No more than 10 ppm of any product will appear in another product”. You might ask, “What’s the problem with that?”
The problem is that some people have read (or interpreted) that as meaning a criteria referring to limits of a drug product in another drug product. So when they take a look at the calculation, they base the dose criteria on the dose of the cleaned drug product in another drug product, or base the 10 ppm criterion on no more than 10 ppm of any drug product in another drug product. It should have clear from the examples the Lilly scientists used that the residue of concern was the active (API) in the cleaned drug product. In one example given, the factor “I” related to the dose of the cleaned product is based on “milligrams of the active ingredient”. That is, the calculation was not a “0.001 of a minimum dose of one drug product in another drug product”, but rather was “0.001 of a minimum dose of the active of one drug product in the maximum dose of another drug product”. That is, rather than using the term “product” loosely as meaning either a “drug product” or an “active ingredient”, it would have clearer to use more precise terms.
This also carries over to the use of the 10 ppm criterion. It is not 10 ppm drug product in another drug product. Again, in the example given for the 10 ppm criterion, the term “R” is “10 mg of the active ingredient in product A/kg product B” emphasis added. So the calculation is not “10 ppm of one product in any other product” but rather 10 ppm of the active of one drug product in any other drug product.
Don’t get me wrong here and think I’m complaining about that publication. That paper was critical in the early days of cleaning validation for providing a basis for how limits could be set. A careful reading of the paper should allow a reader to clearly know when the “product” referred to was the “active ingredient” and when it was actually the “drug product”. I am merely using this as an example to encourage us to be more careful in the use of terms.
The second example I will cover is also from 1993. It is the FDA’s cleaning validation guidance document. And this example also covers the issue of limits. In that document the FDA state that “limits that have been mentioned by industry representatives … include analytical detection levels such as 10 PPM, biological activity levels such as 1/1000 of the normal therapeutic dose, and organoleptic levels such as no visible residue”. It is likely that those three “industry examples” are from the Fourman and Mullen paper since that publication is listed in the “references” at the end of the FDA guidance. Yes, the FDA presentation is slightly different in that it refers to the “normal therapeutic dose” as opposed to the minimum therapeutic dose, as well as expanding visually clean to other organoleptic evaluations.
However, what I want to focus on here is the FDA’s reference to 10 ppm as the analytical detection level. As an analytical detection level, this sounds like it means that as long as the analytical sample test result is below 10 ppm, any measured residue is acceptable. We all know this can’t be the case; if it were we would just extract our swab samples with a much larger volume of diluent to make sure the measured value was below 10 ppm. But that is not what stated in the Fourman/Mullen paper. The 10 ppm is 10 ppm as the concentration in the next drug product. I suspect that this is part of the reason that I still see people wanting to set limits just based on a 10 ppm value in an analytical sample. For clarification, doing so may be possible, as long as 10 ppm in an analytical sample was less than a limit calculated on a dose criterion and on a 10 ppm in the next product criterion.
So, is there a practical point to giving these examples? Of course! The first is to never use the word “product” in a generic sense unless you clearly state you are doing so. For example, sometimes I will write “As used in this report, “product” may mean drug product, active, excipient or cleaning agent”. But, if I am specifically referring to the formulated product, I try to refer to it as the “drug product” (or sometimes as formulated drug product, to be somewhat redundant). And I refer to the active as drug substance, API or drug active. It may seem wordy, but it avoids potential confusion. Second, when you refer to a limit make sure you state what that limit is a limit in (or what it is a limit for). There is a clear difference between 10 ppm in the next product and10 ppm in the extracted swab sample. This is one of the reasons that I try to use the “L0, L1, L2, L3, L4a, L4b, and L4c” terminology (see the September 2012 Cleaning Memo) when I discuss limits.
Some of you may tire of me talking about the importance of clear expression of ideas. However, if you continue reading my Cleaning Memos in the future, I’m sure I will occasionally come back to another example of the need to be more careful in the language we use.