Hutchy_7 ☆ UK, 2019-11-28 16:17 (1838 d 11:23 ago) Posting: # 20887 Views: 6,149 |
|
I am analyzing a 'simple' 2 way cross bioequivalence study of a generic oral capsule. Our standard approach for BE studies is to replace BLQs with zero. However, one subject has a profile for test product where all values are BLQ; their reference product profile is relatively in line with the rest of the dataset. If this was the other way around (reference profile all BLQ) the EMA BE guidelines would allow exclusion from the analysis; alas we are not in that position. There are no other clinical/medical reasons to exclude this subject from the PK analysis set, however the zero values for Cmax and AUC cannot be log transformed and therefore this subject's data cannot included in the statistical analysis. My initial thought would be to replace BLQ with LLOQ/2 for this subject only. Does that seem like a logical solution? Very grateful for anyone's thoughts/experiences from a regulatory acceptability perspective. Simon Edit: Category changed; see also this post #1. [Helmut] |
Ibrahim Komeil ☆ Egypt, 2019-11-28 16:55 (1838 d 10:45 ago) @ Hutchy_7 Posting: # 20889 Views: 5,040 |
|
Hi Simon I think that we will face a problem in optimization the linearity for this volunteer and weighing the calibration curve, i think that you can't modify the bioanalytical validation for single volunteer. But you can do it for all volunteers not for one person only by changing in LLOQ Regards, Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut] |
Helmut ★★★ Vienna, Austria, 2019-11-28 17:26 (1838 d 10:14 ago) @ Hutchy_7 Posting: # 20890 Views: 5,201 |
|
Hi Simon, ❝ Our standard approach for BE studies is to replace BLQs with zero. Not ideal. I would define the concentration-column as character instead of numeric (Phoenix/WinNonlin) and keep them coded BQL . If you are using an R-package (bear , PKNCA ) keep the column numeric but replace all BQLs by NA . In any case use the linear-up/logarithmic-down trapezoidal method for AUC. The linear trapezoidal should go to the waste bin.❝ However, one subject has a profile for test product where all values are BLQ; their reference product profile is relatively in line with the rest of the dataset. If this was the other way around (reference profile all BLQ)… Exactly, bad luck. The EMA’s logic behind: If the reference product shows low/irregular profiles sometimes (not unusual for highly variable drugs / drug products), it does not matter. There were no safety/efficacy issues in phase III and IV – despite occasional “bad” product performance. Hence, not an issue. On the other hand, there are no safety/efficacy data available for the test product. Therefore, exclusion in BE (which is a surrogate for TE) is considered not acceptable. ❝ There are no other clinical/medical reasons to exclude this subject from the PK analysis set, … That’s bad. Otherwise, you would have a justification for exclusion. ❝ … however the zero values for Cmax and AUC cannot be log transformed and therefore this subject's data cannot included in the statistical analysis. Correct in principle. Nevertheless, exclusion “on the basis of statistical analysis or for pharmacokinetic reasons alone” is not acceptable (BE-GL p. 14). ❝ My initial thought would be to replace BLQ with LLOQ/2 for this subject only. Why only for this subject? BE-GL p. 14: “The data from all treated subjects should be treated equally.” What does your protocol say? ❝ Does that seem like a logical solution? No – what’s so special about dividing by 2? Why not \(\hat{C}=LLOQ/\textrm{e}\) or \(\hat{C}=LLOQ/\pi\)? — Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
Hutchy_7 ☆ UK, 2019-11-28 17:47 (1838 d 09:53 ago) @ Hutchy_7 Posting: # 20891 Views: 5,042 |
|
Thank you both for the swift reply... will obviously need to give this more thought! |
ElMaestro ★★★ Denmark, 2019-11-28 18:44 (1838 d 08:56 ago) @ Hutchy_7 Posting: # 20892 Views: 5,040 |
|
Hello Hutchy, ❝ Thank you both for the swift reply... will obviously need to give this more thought! Thinking won't solve it, you still need to find some practical ways to handle this case.
Audit the whole thing, and do it well. For example pt. a. above is not about looking at the CoA and just thinking "well, the thingy was at 101.3% of labeled content so we are all good". Did the samples of this subject's run sit with other subject samples in the same run? Did you see anything out of the ordinary there? Did you investigate? Did the lab technician pipette the right matrix for extraction? Did the same technician pipette samples for another trial that day, if yes did the other trial also have a zero profile? Did you see irregularities when doing dissolution and F2? Was any test for potency/content/etc repeated for any reason? Did Watson LIMS grab an output from Analyst? Did it do so correctly? Did you use a Hamilton? Did the Hamilton pressure profiles for the samples in question look normal (note: Hamiltons generate myriads of data which noone ever looks at but it is available). If you sample the content in XYZ tablets (individually), would one of them show 0% ? Can you do a lab investigation and look at the primary and/or secondary samples again (regardless of how the guideline is worded, this is part of an investigation, not a classical repeat)? ISR's for that subject informative or not done yet? And so forth. This could all be a bit of an ordeal, but the burden associated with doing it right now is much, much smaller than the burden of an inspection later on if you get flagged, and you may get flagged if you are not very proactively pursuing explanations. So I would do these things right away. — Pass or fail! ElMaestro |