In CMC process transfer, method transfer, and scaled-down model building scenarios, it is necessary to compare production data from different batches to prove that the process before and after the transfer is consistent, or to prove that the scaled-down model is consistent with the large-scale one. We can choose different statistical methods to prove this, such as TOST equivalence test, Mean±3sd, tolerance interval, principal component analysis, etc. This presentation will introduce different scenarios corresponding to these different methods to reasonably illustrate the consistency between different batches.
