Log in
Register|
MiLi ☆ 2008-08-06 12:56 (6114 d 04:57 ago) (edited on 2008-08-06 15:14) Posting: # 2154 Views: 7,454 |
|
|
Hello, I am working on a old bioavailability study. there is only 7 sample times. The first analysis has been done with incremental troncated AUC. I have calculated AUCinf but the problem is that 1/3 of Tmax is at the last or the last but one sample time. The data even give some (1/6!!) negative Ke!!! It is not a problem in my program because you can verify it on graph. My question is : According to the study report, there has been no problem to perform the analysis on this data. But, I wonder about the (poor ?) credit that we can give to the results of this first analysis. Sorry for my poor english. -- Edit: Category changed. [Jaime] |
|
Helmut ★★★   Vienna, Austria, 2008-08-07 18:00 (6112 d 23:54 ago) @ MiLi Posting: # 2160 Views: 6,026 |
|
|
Dear miLi! ❝ I am working on a old bioavailability study. Sincere condolences.  ❝ there is only 7 sample times. The first analysis has been done with incremental troncated AUC. Interesting. I hope the protocol did state one metric as the main parameter - otherwise you will run into multiplicity issues. ❝ I have calculated AUCinf but the problem is that 1/3 of Tmax is at the last or the last but one sample time. I guess you mean ⅓ of Cmax? ❝ The data even give some (1/6!!) negative Ke!!! Do I get you right: you 'see' increasing values in 1/6 of subjects? ❝ It is not a problem in my program because you can verify it on graph. What program are you using? Kel by definition is positive (denoting a transfer from the central compartment); all programs specialized in NCA PK analysis would come up with an error message. ❝ My question is : According to the study report, there has been no problem to perform the analysis on this data. There's an American phrase for it: 'Lights on, nobody at home.' ❝ But, I wonder about the (poor ?) credit that we can give to the results of this first analysis. Forget it (sorry). ❝ Sorry for my poor english. No problem. If I misunderstood you, please correct me. — Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
MiLi ☆ 2008-08-07 19:28 (6112 d 22:26 ago) @ Helmut Posting: # 2161 Views: 5,934 |
|
|
Thanks for your answer. ❝ Sincere condolences.  ❝ Interesting. I hope the protocol did state one metric as the main parameter - otherwise you will run into multiplicity issues. ❝ I guess you mean ⅓ of Cmax? Yes, sorry ❝ Do I get you right: you 'see' increasing values in 1/6 of subjects? Yes!!! ❝ What program are you using? Kel by definition is positive (denoting a transfer from the central compartment); all programs specialized in NCA PK analysis would come up with an error message. I made the program myself with sas, maybe there are some errors in it but the results about elimination rate fit with the graphs. The data are really strange!!! ❝ There's an American phrase for it: 'Lights on, nobody at home.' I think I see the idea but what does that precisely mean?! Thanks to increase my culture!  ❝ Forget it (sorry). that's what i was afraid of. Thanks a lot |
|
Helmut ★★★ Vienna, Austria, 2008-08-07 19:54 (6112 d 21:59 ago) @ MiLi Posting: # 2162 Views: 6,074 |
|
|
Dear MiLi! ❝ I made the program myself with sas, maybe there are some errors in it but the results about elimination rate fit with the graphs. OK, if there are increasing values you may fit them, but without physiological meaning (assuming your formulation is not a transdermal patch, an i.m. suspension, etc.). Since in a linear PK model the slowest phase 'drives' the profile you would see the absorption in the later part of the profile and elimination in the former. That's the classical 'flip-flop' situation. How do your 'half lives' compare to reported ones from the literature? ❝ The data are really strange!!! Again; what's the 'half life'? If values are close to the LLOQ of the method and there's a lot of variability (old study = old method) it may also be possible that this increase is due to pure chance (with just 7 sampling points overall I would expect that quite often you had just 3 points available for estimating kel). ❝ ❝ There's an American phrase for it: 'Lights on, nobody at home.' ❝ ❝ I think I see the idea but what does that precisely mean?! Thanks to increase my culture! It describes a person who only seems to be present, but actually is absent-minded, or even worse has no mind at all.  I was referring to these guys running the first analysis without even noticing problems. — 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 |
|
d_labes ★★★ Berlin, Germany, 2008-08-08 10:42 (6112 d 07:11 ago) @ MiLi Posting: # 2163 Views: 6,177 |
|
|
Dear MiLi, ❝ [...] I made the program myself with sas, maybe there are some errors in it but [...] Have a look at Arturo Soto Matos-Pita, Bernardo de Miguel Lillo Noncompartmental Pharmacokinetics and Bioequivalence Analysis www.lexjansen.com/pharmasug/2005/statisticspharmacokinetics/sp07.pdf May be this helps to validate your own SAS program. But note the bioequivalence test (90% confidence interval) in this paper is not correct. 1.761 in the macro anova has to be replaced by the appropriate t-value. The sequence test is also not correct implemented. Regarding your task "re-analysis" I subscribe to the view of HS ❝ Sincere condolences. — Regards, Detlew |
Ing. Helmut Schütz