Log in
Register|
mathews ● 2007-09-17 15:04 (6056 d 03:30 ago) Posting: # 1086 Views: 11,975 |
|
|
I want help to clear some doubts in sample size calculation for a 2 x 2 crossover design. I know only the Cmax value of the drug from the literature. It is around 42.34 ± 13.27. How i can calculate the intra subject CV from this Cmax value? (is it required for Sample Size calculation?) which value of "delta" we commonly used for calculating Sample Size? there are different sample size formulas for additive model and multiplicative model? which model we commonly used? why? Mathews |
|
Helmut ★★★   Vienna, Austria, 2007-09-17 17:59 (6056 d 00:35 ago) @ mathews Posting: # 1087 Views: 10,562 |
|
|
Dear Mathews! ❝ I know only the Cmax value of the drug from the literature. It is around 42.34 ± 13.27. ❝ How i can calculate the intra subject CV from this Cmax value? (is it required for Sample Size calculation?) Unfortunatelly there's no way to calculate CVintra from CVtotal (CVtotal = CVintra + CVinter). Actually you don't even have CVtotal (in log scale), because CV = SD/mean (what you have) comes from untransformed data. Also see this post. ❝ which value of "delta" we commonly used for calculating Sample Size? With exception of Canada, where delta should be set to zero (ratio=1) and a correction for actual content should be made in the assessment of BE, delta should be set to the expected difference (e.g., from a reasonably sized pilot study). The commonly applied approach is to set the expected ratio to 0.95 (or -5%). Rationale behind is that release specifications of a batch generally are ±5%; since this applies to both test and reference - and we assume an average deviation from declared content of ±2.5% we end up with ±5% for the study. ❝ there are different sample size formulas for additive model and multiplicative model? Sure. Different model --> different error distribution / power function / sample sizes. ❝ which model we commonly used? why? It's a generally accepted assumption that many biological variables (including PK metrics like AUC, Cmax) follow a lognormal distribution. Furthermore serial dilutions in analytics also lead to multiplicative errors. Since additivity of effects and homoscedasticity (homogeneity of variance) is a prerequisite for ANOVA, we apply a multiplicative model (i.e., work with log-transformed data instead with raw data). By this the distribution skewed to the right becomes symetrical and values below zero are avoided. For tmax an additive model is most suitable; although a nonparametric analysis is also a must (ANOVA is incorrect for discrete data). — 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 |
|
mathews ● 2007-09-18 13:17 (6055 d 05:17 ago) @ Helmut Posting: # 1091 Views: 10,344 |
|
|
Dear HS, Thanks for your reply.  I have some more doubts. I want the intra CV value of some drugs for sample size calculation. In case if no pilot study is conducted what can i do? (I think in most of the BE studies CV is calculated fom literature). I am a fresher in this field. How can I get the intra CV value from the literature? Is there any web site giving information about drugs CV and all? From one of your post I understood that we can easily calculate the CV from the MSE of the ANOVA table. But this is possible if we conduct the pilot study (right?). From the paper "On Sample Size Calculation in Bioequivalence Trials" by Chow and Wang, they give two type formulas for calculation of S.S., one formula for raw data and one for log transformed data. what is the difference between the value of "sigma" in both case? Usually we use intra CV for the calculation of sample size. Is there any formula for calculating sample size using inter subject CV? please give some reference for sample size calculation in BE studies thanks Mathews |
|
Helmut ★★★ Vienna, Austria, 2007-09-18 15:51 (6055 d 02:43 ago) @ mathews Posting: # 1093 Views: 10,411 |
|
|
Dear Mathews! ❝ I want the intra CV value of some drugs for sample size calculation. You are not alone…  The only collection of CVs I know is rather outdated: Steinijans VW, Sauter R, Hauschke D, Diletti E, Schall R, Luus HG, Elze E, Blume H, Hoffmann C, Franke G, and W Siegmund ❝ In case if no pilot study is conducted what can i do? ❝ (I think in most of the BE studies CV is calculated fom literature). You are right! ❝ Is there any web site giving information about drugs CV and all? I don’t think so. ❝ From one of your post I understood that we can easily calculate the CV from the MSE of the ANOVA table. Sometimes (though rarely!) the entire ANOVA-table is published; in such a case the calculation is straightforward: \(CV=\sqrt{e^{MSE}-1}\); also see this post. ❝ From the paper "On Sample Size Calculation in Bioequivalence Trials" by Chow and Wang,… Caution: This paper contains a lot of typographic errors. Please get the corrections also: S-C Chow and H Wang ❝ … they give two type formulas for calculation of S.S., one formula for raw data and one for log transformed data. what is the difference between the value of "sigma" in both case? There’s a lot of confusion in terminology. Some authors use the term CVpooled instead of CVtotal.
Midha KK, Ormsby ED, Hubbard JW, McKay G, Hawes EM, Gavalas L, and IJ McGilveray For the multiplicative model (cross-over) \(CV=\sqrt{e^{MSE}-1}\) For the additive model (cross-over) \(CV=\sqrt{MSE}/\bar{x}_R\) ❝ Usually we use intra CV for the calculation of sample size. If you mean by “usually” non-replicate cross-over designs, yes. ❝ Is there any formula for calculating sample size using inter subject CV? This does not make sense. You get CVinter only from a cross-over study (as said above a parallel design will give you CVtotal). If you have CVinter, most likely you also will have CVintra; if there are no PK/clinical reasons to perform a parallel study, I would always go with a cross-over study, which is more powerful. ❝ please give some reference for sample size calculation in BE studies OK, you know Chow and Wang’s paper already. Please see two responses:
Approximations for the multiplicative cross-over model: Hauschke D, Steinijans VW, Diletti E and M Burke Sample Size Determination for Bioequivalence Assessment Using a Multiplicative Model J Pharmacokin Biopharm 20(5), 557–61 (1992) Exact sample size tables for the multiplicative cross-over model: Diletti E, Hauschke D and VW Steinijans Sample size determination for bioequivalence assessment by means of confidence intervals Int J Clin Pharm Ther Toxicol. 29(1), 1–8 (1991) Diletti E, Hauschke D and VW Steinijans Sample size determination: Extended tables for the multiplicative model and bioequivalence ranges of 0.9 to 1.11 and 0.7 to 1.43 Int J Clin Pharm Ther Toxicol 30/Suppl.1, S59–62 (1992) Exact sample size tables for the additive cross-over model: KF Phillips Pharmacometrics. Power of the Two One-Sides Tests J Pharmacokin Biopharm 18(2), 137–44 (1990) — 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 |
|
mathews ● 2007-09-24 13:11 (6049 d 05:22 ago) @ Helmut Posting: # 1120 Views: 10,288 |
|
|
Dear HS, Thanks for ur reply post. do u have any idea about the intra CV of Lansoprazole? from one published paper i got different CVintra for AUC and Cmax value of Lansoprazole. look this link http://www.mhra.gov.uk/home/groups/l-unit1/documents/websiteresources/con2023110.pdf and also from another paper i got a another CVintra for Lansoprazole.look this http://www.mhra.gov.uk/home/groups/l-unit1/documents/websiteresources/con2023104.pdf I want to find the Sample Size for a single dose 2 x 2 crossover design under fasting condition. So which CVintra is best for finding the Sample Size?. If we use the CVintra for Cmax value, the number of subjects required is almost double than that of CVintra of AUC value? So please help me for finding the intra CV. Thanks mathews |
|
Helmut ★★★ Vienna, Austria, 2007-09-25 03:12 (6048 d 15:21 ago) @ mathews Posting: # 1123 Views: 10,282 |
|
|
Dear Mathews, ❝ do u have any idea about the intra CV of Lansoprazole? Regrettably, yes...  ❝ I want to find the Sample Size for a single dose 2 x 2 crossover design under fasting condition. ❝ So which CVintra is best for finding the Sample Size?. ❝ If we use the CVintra for Cmax value, the number of subjects required is almost double than that of CVintra of AUC value? That’s a pitty, but you have to demonstrate BE for both AUC and Cmax, although you may widen the acceptance range for Cmax to 0.75–1.33 based on clinical grounds and the high variability. If the study is intended for submission in the EU, some countries will accept literature data showing high variability (e.g., UK), whereas others (e.g., NL), will ask for a replicate design study to show that the reference is a HVDP (see the Q&A document): A high CV as estimated from the ANOVA model is thus an indicator for high within-subject variability. However, a replicate design is needed to assess within-subject variability. There have been already a couple of threads concerning *prazoles and problems with high variability, gastric resistance, and outliers. I’ve seen studies succeeding with sample sizes of 32, and others failing in 90+…  — 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 |
|
mathews ● 2007-09-25 10:22 (6048 d 08:12 ago) @ Helmut Posting: # 1128 Views: 10,052 |
|
|
Dear HS, Thank you so much for ur reply. regards, Mathews |
Ing. Helmut Schütz