Bioequivalence and Bioavailability Forum

Hi Achievwin,

❝ It is not about potency correction, it is about controlling errors wherever possible …

Agree.

❝ I believe I heard (in the 2000 BE workshop) rationale for this 5% difference is to accomodate aging differences (time of manufacture to time of dosing) between Test and RLD (usually RLD is older than Test …

Which workshop? Let’s compile what we have:

• FDA ANDA Draft, Appendix A (2013)
  We recommend that the assayed drug content of the test product batch not differ from the reference product by more than ± 5 percent.
  
• EMA BE GL (2010)
  Section 4.1.2 Reference product
  The selection of the reference product used in a bioequivalence study should be based on assay content and dissolution data and is the responsibility of the Applicant. Unless otherwise justified, the assayed content of the batch used as test product should not differ more than 5% from that of the batch used as reference product determined with the test procedure proposed for routine quality testing of the test product. The Applicant should document how a representative batch of the reference product with regards to dissolution and assay content has been selected. It is advisable to investigate more than one single batch of the reference product when selecting reference product batch for the bioequivalence study.
  Section 4.1.8 Evaluation
  In bioequivalence studies, the pharmacokinetic parameters should in general not be adjusted for differences in assayed content of the test and reference batch. However, in exceptional cases where a reference batch with an assay content differing less than 5% from test product cannot be found (see section 4.1.2) content correction could be accepted. If content correction is to be used, this should be pre-specified in the protocol and justified by inclusion of the results from the assay of the test and reference products in the protocol.
  
• WHO Section 7.3.2 Choice of comparator product (2017)
  Content of the API(s) of the comparator product should be close to the label claim and the difference between two products being compared should not be more than ± 5%. If, because of the lack of availability of different batches of the comparator product, it is not possible to study batches with potencies within ± 5%, potency correction may be required on the statistical results from the bioequivalence study.

❝ … and also assay variabilities (Damn: everyone blames poor analytical chemist ).

That’s the point. What are the batch release spec’s? Generally ±10% and for NTIDs ±5%. Of course, you don’t get a CoA from the originator. Analyzing the reference with the method validated for the test is not a problem for IR. Little bit more tricky for MR. A nightmare for creams and ointments with their fantastic emulsifiers. You never can’t be sure. Yes, analytical (in)accuracy and (im)precision hits … Say, you have a great routine method with 2% and you measure a potency 100% for T and R. What are the true values? Can be <100% for T and >100% for R. That’s why in sample size estimation one should never (ever!) assume a T/R-ratio of 1.

In :

set.seed(20220909)
T <- rnorm(n = 1e7, mean = 1, sd = 0.02)
R <- rnorm(n = 1e7, mean = 1, sd = 0.02)
round(quantile(T / R, probs = c(0.025, 0.5, 0.975)), 4)
 2.5%   50% 97.5%
0.946 1.000 1.057


That’s why in most sample size functions of PowerTOST 0.95 is the default.