Bioequivalence and Bioavailability Forum

Hi Beholder,

❝ You are right regarding the meaning. Yes, it is rather "should not". But looks like I was not clear enough.

Oh, you were.

❝ My question was caused by an article […]

The link is not working. Can you check?

❝ According to the literature, the minimum arithmetic average Cmax is 1,479.61 ng / ml.

Arithmetic means of concentrations are stupid. Reporting concentrations to six significant digits even more. I assume that the “minimum arithmetic average C_max” was calculated from different studies; each with a dose of 300 mg.

❝ Thus, the LLOQ should not exceed 74 ng / ml.

OK, 5% of a doubtful value (likely the geometric mean is lower). Setting the LLOQ based on (any) average is stupid as well.

❝ However, it should be taken into account that, in accordance with the requirements of the EAEU, the LLOQ is calculated based on the minimum value of Cmax from the entire sample of subjects.

Makes sense. (Lack of) carry-over has to be assessed on a subject-level. See what I wrote at the end of my previous post. We cannot get an unbiased treatment-effect in the presence of true carry­over. Hence, we have to avoid CO by design. One method is to assess pre-dose concentrations in periods >1. And yes, make sure that we are able to measure 5% of every subject’s C_max after each treatment.

❝ In the analyzed studies, the minimum value of Cmax was 60.8 ng / ml,

Wow, that’s extreme! Not clear to me whether the authors are referring to the same dose of 300 mg in all studies.

❝ while the calculated value of LLOQ is 3 ng / ml."

5% again. If the same dose was used in all studies (I doubt it) it would be a nightmare:

• LLOQ 3 ng/mL makes sense.
  
• Given such a high between-subject variability, what will be the upper limit of individual C_max-values? 3,000 ng/mL, 5,000 ng/mL, or even higher? Would drive most detectors to the edge.
  Yes, one could use a quadratic calibration but that would decrease the accuracy at high concentrations (the curve is getting flatter). Set the ULOQ not that high, but dilute and reanalyze samples >ULOQ?

---

PS: It can be even worse. Sometimes in studies in patients (mainly phase III) different LLOQs are used. Might be caused by limited sample volumes or interactions with co-medications. The world is not ideal.