Bioequivalence and Bioavailability Forum

Dear Ramesh - or Debbie in PharmPK

❝ … all the standards (STD 1 to STD 11) are with in the specifications for accuracy, but, both blank and blank IS are not meeting the acceptance criteria (less 20 % of the LOQ(STD 1)) against the STD 1.

Where does this criterion (<20% of LLOQ) come from?
And how do you calculate it actually - since you don't know the concentration, do you compare the measured response of the blank with the calculated response at zero (=intercept of the CC)?

❝ so we have removed the STD 1 from the regression and calculated the interference in the STD balnk and STD Zero samples (blank sample with IS), against the STD 2, making STD 2 as the new LOQ, …

… in trusting the interference more than your calibrators?

Most people would allow removal of data points from the calibration curve only (for <25% of data points) if the back-calculated concentration deviates more than 15% (or 20% at the LLOQ) from the nominal concentration in the following cases:

• not more than 25% of data points are affected, and
  
• the validated regression model would not change (http://www.fda.gov/cder/guidance/4252fnl.pdf FDA).
  
• Excluded points should not be consecutive (ANVISA).

Removal is possible only for CC data points back-calculated from the entire regression model. IMHO you are trying to remove data points, because you have some problems with contamination in both CC-blanks.
I assume, you don’t have contaminations at the other levels of the CC, because in this case you would get an increased (compared to other CCs) intercept, which should give a back-calculated value of ≪LLOQ for the blanks.

Your method of defining a new LLOQ higher than than the lowest QC-level simply leads to a rejected run.

---

Edit: Link corrected for FDA’s new site. [Helmut]