Hi Mariana, I’ll try to work through your questions.
Firstly, I think your assumption
- In WNL, the input values for <LLOQ concentrations must be different depending if they happen in absorption (before Tmax) or in elimination phase (after Tmax)?
is incorrect. Depending what your analysis method is modelling and/or NCA you can quite possibly leave them as a non-numeric value i.e. missing and the program will manage the missing information. For instance in NCA there are various rules used to insert a concentration for AUC calculation when it is missing at time zero i.e. dosing, dependent on the route (and regimen e.g. if first dose or steady-state). Look in the WNL user guide (page 27 in v6.4) for the section “Data checking and pre-treatment” and I think you will find many of the assumptions you are trying to incorporate are already handled for you.
Going on to your specific questions I am going to assume initially you are looking at NCA,and you meant sampling, not dosing times…
1.1) Suppose that they appear before Tmax and we have the first three dosing times with <LLOQ concentrations:
In WNL, we must define the concentration values as “0” for these three dosing times (conWNL) or we must define only the first dosing time (t=0) as “0”, and the remaining two as “missing” (conWNL2)?
There is any difference in using linear method or lin/log down method if the <LLOQ concentrations happen in dosing times before Tmax?
- the difference you will see is whether a Tlag is output, and whether it calcualtes the first ‘triangle’ oing back to t0 or t3
1.2.) Suppose that they appear after Tmax and we have the last two dosing times with <LLOQ concentrations:
-personally leave as non-numeric, you don’t know the value with confidence, it’s not ZERO, so leave it missing, i.e. <LLOQ or indeed <5 if that was the LoQ. I prefer LinUp/LogDown since it covers possibilites of reabsorption later inthe profile
1.3.) Suppose that they appear after Tmax and we have at least one dosing time with <LLOQ concentrations before the last dosing time:
If I understand your question you have 2 or more BQL vlaues and then your profile has quantifiable values again; you may want to define the subesequent concentration values as missing (or indeed 0), this is dependent on your SOP/protocol. If you want to do this then the BQL wizard can be instructed to do so. However in the data you quoted above I would be more inclined to query the two BQLs embedded in the profile as it looks like you are losing valid data and the apparent elimination phase by censoring the concentrations measured after these points.
I will attach a simple project file illustrating the choice of using only some of the rules discussed above and how the NCA might look different.
Simon.
BQL_forum.phxproj (482 KB)BQL.doc (59 KB)