Bioequivalence and Bioavailability Forum

Dear all,
I´m involved in the development of an IR generic product for EU submission. There are 2 strengths of the reference product and the SmPC states that the exposure appears to be non-linear. But, I cannot find a dose study in the literature where both strengths are dosed in the same study. So, I did a 3-way pilot study with both strengths of the reference and 1 strength of the test product. ABE analysis on dose-adjusted AUCt of the reference was used to test dose proportionality. And the results showed that the upper CI is just outside the 80-125. On the other hand, if you compare the dose-adjusted means of the AUCt for the 2 strengths the difference is 8%.
The EU guideline for IR states: “Assessment of linearity will consider whether differences in dose-adjusted AUC meet a criterion of ± 25%.”
Can I use this data to argue that PK is linear and a biowaiver for the lower strengths should be accepted?
Any comments/suggestion is greatly appreciated.