Learnaboutgmp Community

Cleaning Validation query

Dear All,

We have a product with 12.5 mcg smallest therapeutic dose and according to cleaning validation calculation (Considering PDE approach) we are getting the final acceptance limit is 137.75 mcg/ swab which is too higher than the therapeutic dose.

Detail of last product (A) and next product (B) as defined below-

PDE of previous product= 0.022 mg/day
Batch Size of next product (B)= 60.0 Kg (600,000 tablets)
SRDD of previous product (A)= 12.5 mcg
LRDD of next product (B)= 300 mcg (or 300/12.5 = 24 tablets)
Swab Surface Area= 4 sq. inch
Total Surface Area of shared equipments= 15971 sq. inch
LOQ of previous product (A)= 0.014 mcg/mL
LOD of previous product (A)= 0.007 mcg/mL

Inference: Based on Health based Criteria (PDE) Calculation, The Final swab limit is higher than the SRDD (12.5 mcg).

We have also done the cleaning validation calculation considering other approaches but not getting the appropriate acceptance criteria.

What approach we have to follow and what carryover limit is acceptable in this situation?


That’s the problem with the ‘old’ cleaning validation criteria. Because it is so potent (it’s probably a hormone) the ADE or PDE will have an extremely low concentration based on the NOAEL (no observed adverse effect level, a measure of toxicity). Based on the current regulations in chapters 3 and 5 of the Eudralex the new equation would be:

MSC = ADE (mcg/kg) x BW (kg) x Lot A (kg)/lot B 9kg)

MSC is the 'Maximum Safe Concentration
ADE is the ‘acceptable daily exposure’ in mcg/Kg
BW is the average ‘body weight’ of the human adult which is 70 Kg
Lot A is the active lot in Kg
Lot B is the smallest production lot in this manufacturing train

Note, this equation does not contain the swab or rinse area yet (you may have to increase the area). You will have to go back and evaluate your current method to ensure it LOQ can even detect the MSC and even develop a new method. If you wish to continue to manufacture hormones you may have to dedicate the facility.

I think you can use this: