This is the third of several Cleaning Memos dealing with selection of a worst-case product in a grouping approach for cleaning validation. In August we considered use of the pharmacopeia solubility categories, and also presented what may be a more useful system based on how “solubility bands” are defined. In September we considered issues in the use of typical evaluation systems involving calculation of a risk score that include a solubility parameter in addition to other parameters such as toxicity, potency, and operator cleaning difficulty. If you haven’t done so, please review those before continuing.
The focus this month will be selecting the worst case (the “most difficult to clean” product) in a grouping approach focused on Oral Solid Dosage formulations These considerations are applicable for solid tablets and the filling materials inside a solid capsule, (but may have to be modified for “gummy-type” drug products because they are in some sense semi-solids). In addition, this discussion will include the assumption that the worst-case product is the “most difficult to clean”, and that a protocol will be performed on that product using the lowest limit of any product in the group.
Overall Approach
Using the assumption that excipient ingredients and active ingredients both affect the difficulty of cleaning, a rating and also a ranking can be established for each product based on the a combination of the solubility rating of each ingredient and the percentage of each ingredient in a formulation. It is easiest to start off with an example (Example A) of a simple situation. Note clearly that in the discussion below, the solubility rating is not identical to the USP solubility categories (even though the solubility rating is based on the USP solubility categories.
Example A
Below are three formulations along with a solubility rating and the weight percent of each component in the drug product. For purposes of this evaluation, any component with a solubility of 100 ppm or less in water is considered “Insoluble”. This would include all components in the USP “Practically insoluble” or “Insoluble” categories (the lowest USP solubility categories). For purposes of this discussion, all other USP solubility categories would be considered “Not Insoluble”. Note that some may prefer to establish the dividing line between “Insoluble” and “Not Insoluble” at 1000 ppm. That could be acceptable as long as those companies apply it consistently. However, it is likely that any substance with a solubility of greater than 100 ppm in water is not likely to be left behind in equipment in significant amounts with the use of multiple water rinses.
Here are three different formulations made on the same equipment and cleaned by the same cleaning procedure. For each product formulation, components along with whether the component is considered “Insoluble” as defined above (solubility at 100 ppm or less) and the weight percent of the component in the formulation.
| Product A1 | Product A2 | Product A3 | ||||||
| Component | Insoluble? | Wt. % | Insoluble? | Wt. % | Insoluble? | Wt. % | ||
| API-1 | No | 5 | ||||||
| API-2 | Yes | 5 | ||||||
| API-3 | No | 5 | ||||||
| Excip-1 | No | 10 | No | 10 | No | 40 | ||
| Excip-2 | Yes | 40 | Yes | 10 | ||||
| Excip-3 | Yes | 45 | Yes | 40 | Yes | 45 | ||
| Excip-4 | No | 45 | ||||||
| Total % Insolubles | 85 | 45 | 55 | |||||
In this situation using the conventional approach of focusing mainly on the solubility of the active, it is likely that Product A2 would be considered the most difficult to clean. In the newer approach presented in this Cleaning Memo, Product A1 would be considered the “most difficult to clean” because of the Total % Insolubles” of 85% even though the active in Product A1 was not considered ‘Not Insoluble”.
I fully realize that these three formulations were selected to illustrate a point. So the next three examples (B, C, and D) further demonstrate how this approach can be modified in cases where different products share the same “Total % Insolubles”.
Example B
Here are two formulations with Product B1 being identical to Product A1. Both B1 and B2 are made on the same equipment and cleaned by the same cleaning procedure. The data in the table is organized just like in Example A.
| Product B1 | Product B2 | ||||
| Component | Insoluble? | Wt. % | Insoluble? | Wt. % | |
| API-1 | No | 5 | |||
| API-2 | No | 10 | |||
| Excip-1 | No | 10 | No | 5 | |
| Excip-2 | Yes | 40 | Yes | 45 | |
| Excip-3 | Yes | 45 | Yes | 40 | |
| Total % Insolubles | 85 | 85 | |||
In the newer approach presented in this Cleaning Memo, both products appear to be equally difficult to clean because each has total insolubles of 85%. One way to “break this tie” would be to consider the weight percent of actives. In this situation, Product B2 has a significantly greater weight percent of API; so as a result of being cleaned the same, it would be more likely to leave behind higher levels of residue of the active. And, if the limits approach is the lowest limit of any product in the group, it would be the one product most likely to fail if cleaning were inadequate.
Example C
Here are two more formulations with Product C1 being identical to Product A1. Both C1 and C2 are made on the same equipment and cleaned by the same cleaning procedure. The data in the table is organized just like for previous examples.
| Product C1 | Product C2 | ||||
| Component | Insoluble? | Wt. % | Insoluble? | Wt. % | |
| API-1 | No | 5 | |||
| API-2 | Yes | 5 | |||
| Excip-1 | No | 10 | No | 15 | |
| Excip-2 | Yes | 40 | Yes | 40 | |
| Excip-3 | Yes | 45 | Yes | 40 | |
| Total % Insolubles | 85 | 85 | |||
As in Example B, both products in Example C appear to be equally difficult to clean because each has “Total % Insolubles” of 85%. One way to “break this tie” would be to consider that Product C2 has an “Insoluble” API, whereas Product C1 has an active that is “Not Insoluble”. Other things being equal, Product C2 is likely to leave behind higher levels of residue of the active. And, if the limits approach is the lowest limit of any product in the group, C2 would be the one product most likely to fail if cleaning were inadequate.
Example D
Here are two more formulations with Product D2 being identical to Product C2. Both D1 and D2 are made on the same equipment and cleaned by the same cleaning procedure. The data in the table is organized just like for previous examples.
| Product D1 | Product D2 | ||||
| Component | Insoluble? | Wt. % | Insoluble? | Wt. % | |
| API-1 | No | 10 | |||
| API-2 | Yes | 5 | |||
| Excip-1 | No | 5 | No | 15 | |
| Excip-2 | Yes | 40 | Yes | 40 | |
| Excip-3 | Yes | 45 | Yes | 40 | |
| Total % Insolubles | 85 | 85 | |||
Again, both products appear to be equally difficult to clean because each has “Total % Insolubles” of 85%. But breaking the “tie” here is more problematic. In this situation, Product D1 has the highest concentration of active, but Product D2 has the active with the “Insoluble” rating. So, which is more critical? This may be a matter of judgment. In this case it might be helpful to evaluate which solubility categories in Table 3 in the September Cleaning Memo these two actives fit into. For example, if API-2 fits into Band V (solubility NMT 10 ppm) and API-1 fits into Band III (solubility from 1000 ppm down to >100 ppm), we might be more likely to judge the solubility rating to be the factor to determine the most “difficult to clean”. An alternative to break the tie in this case is perform some kind of lab test, such as a modified dissolution test comparison to see the rates or amounts of dissolution of the actives in a simulated cleaning situation (see the Cleaning Memo of June 2020).
Example E
The example here is not on how to “break a tie”, but may be considered a way to “establish a tie” in a situation where the “Total % Insolubles” are different, but very close to each other. This is illustrated with two formulations for Product E1 and Product E2, each with different actives but with similar levels of the same excipients. The data in the table is organized just like for previous examples.
| Product E1 | Product E2 | ||||
| Component | Insoluble? | Wt. % | Insoluble? | Wt. % | |
| API-1 | Yes | 5 | |||
| API-2 | Yes | 5 | |||
| Excip-1 | No | 15 | No | 13 | |
| Excip-2 | Yes | 40 | Yes | 41 | |
| Excip-3 | Yes | 40 | Yes | 41 | |
| Total % Insolubles | 85 | 87 | |||
For this example, if we follow the principles established for the highest “Total % Insolubles” rating, we would clearly choose Product E2 as the “most difficult to clean”. However, we may take a look at the difference between 87% and 85% and say there likely is little significant difference in terms of difficulty of cleaning (which would likely be the case if we had designed a robust cleaning process). We furthermore might like a have a rationale for using Product E1 as the representative product for use in our grouping strategy (for example, it is made more frequently than Product E2 and therefore would be easier to obtain the requisite number of cleaning validation runs).
One way to do this is to specify putting the different products into specified “difficulty of cleaning” bands based on “Total % Insolubles”. For example, we might establish bands as given below.
| Total % Insolubles | Band Designation |
| 90 – 100% | I |
| 80 – 89% | II |
| 70 – 79% | III |
| 60 – 69% | IV |
| < 60% | V |
For simplicity, values between 89% and 90% were not included. Based on the judgment that a difference between 89.8% and 89.5% is practically insignificant for determining the difficult of cleaning, it may be appropriate to specify “rounding” of values so any percentage between 89% and 90% is rounded up or down.
The band ranges given above may be suitable for solid oral dose formulations. However, the most suitable ranges for liquid products may be as given below.
| Total % Insolubles | Band Designation |
| 40 – 100% | I |
| 30 – 39% | II |
| 20 – 29% | III |
| 10 – 19% | IV |
| < 10% | V |
In both of these examples of bands, the breakdown of the band ranges are strictly given as examples. In establishing the most relevant ranges for you facility, it may be practical to see whether there appear to be “natural” ranges for the products in your facility.
Note further that even though bands do allow products with closely related “Total % Insolubles” to be considered equivalent, it may be possible to “break a tie” by considering the principles covered in Examples B, C and D. I know this may seem to be inconsistent after we have used “bands” to allow the establishment of “ties”; however, I am reminded Emerson said that “a foolish consistency is the hobgoblin of little minds”.
Some Caveats
In considering the use of the approach given in this Cleaning Memo, particularly pay attention to the reasoning and logic given for the approach. Second, this approach assumes an overall strategy of determining the “most difficult to clean” product and using that product as the group representative product in a cleaning validation protocol where the limit is set at the lowest limit (such as a L3 limit in µg/cm2) of any product in the group. Third, as with any significant change in a validation approach, do some serious exploration as to the feasibility of using this approach as well as the impact of adopting this newer approach on the validity of past cleaning validation that has been done.
Next month we will continue with applying these concepts to liquid products and to API manufacturing.
Copyright © 2023 by Cleaning Validation Technologies