Helmut ★★★ Vienna, Austria, 20121011 19:29 (4070 d 06:27 ago) Posting: # 9393 Views: 7,704 

Dear all, I’m wondering what we can say about the precision of estimated CV_{WR} in different replicate designs. One full replicate (TRTRRTRT) and the partial replicate (TRRRTRRRT) are mentioned by both the FDA and the EMA. An alternative full replicate, but requiring only three periods was used by the Lászlós in 2009 and is also mentioned in the new Russian draft GL, namely TRTRTR. On the contrary to the partial replicate we could also derive CV_{WT}. For CV 30%, T/R 95%, 80% target power I get n=20 for the 4period design and n=30 for both 3period designs with power of 82.02% in all designs. So far so good. What about the precision of the estimated CV_{WR}? Let’s look from how many subjects the value is estimated: TRTRRTRT: 20 (100%) Heretic question: From TRTRTR the estimate will be less precise (only half of the subjects used), but we get additional information about the test. Less chance of outliers? Duno. — Diftor heh smusma 🖖🏼 Довге життя Україна! _{} Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes 
jag009 ★★★ NJ, 20121012 18:52 (4069 d 07:04 ago) @ Helmut Posting: # 9407 Views: 6,501 

Hi Helmut, ❝ An alternative full replicate, but requiring only three periods was used by the Lászlós in 2009 and is also mentioned in the new Russian draft GL, namely TRTRTR. Could you provide the reference for the Laszlos paper? Thanks John 
Helmut ★★★ Vienna, Austria, 20121012 19:37 (4069 d 06:19 ago) @ jag009 Posting: # 9408 Views: 6,815 

Hi John! ❝ Could you provide the reference for the Laszlos paper? Tóthfalusi L, Endrényi L, García Arieta A. Evaluation of Bioequivalence for Highly Variable Drugs with Scaled Average Bioequivalence. Clin Pharmacokinet. 2009;48(11):725–43. doi:10.2165/1131804000000000000000 These simulations (GMR restriction 0.8–1.25, CV 35, 45, 55%; each data point representing 10 000 studies) are based on a TRTRTR design in 36 subjects. Circles: unscaled, squares: EMA’s method (without the 50% cap – paper published before the BEGL), triangles: FDA’s method. Studies passing = empiric power. — Diftor 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, 20121017 12:40 (4064 d 13:16 ago) @ Helmut Posting: # 9427 Views: 6,783 

Dear Helmut, ❝ I’m wondering what we can say about the precision of estimated CV_{WR} in different replicate designs. [...] Let’s look from how many subjects the value is estimated: ❝ ❝ ❝ ❝ Heretic question: From TRTRTR the estimate will be less precise (only half of the subjects used) ... The 50% is only valid if you talk EMA . Otherwise all the data are used for the fit of a (mixed) model. Why not use the confidence intervals of the covariance parameter estimates from the fit of a mixed model as precision? Ok, this leaves out the partial replicate design because we are not aware if the σ^{2}_{WR} is valid from fitting the mixed model, at least in the form of FDA code. At least we can think in terms of intrasubject contrasts to estimate σ^{2}_{WR} via an ANOVA with sequence group as the solely effect (as implemented in the FDA progesterone guidance in the framework of scaled ABE). The df for this analysis are nseq where n is the number of intrasubject contrasts evaluable (=subject/2 in case of design TRTRTR i.e. the 50% above are again correct). Thus take formulas given by your own in the early days of this forum for obtaining a confidence interval for σ^{2}_{WR} and then transformed to CV to obtain a measure of the precision of estimated CV_{WR}. Lets go with the numbers of subjects given by you and assume that the estimated CV is obtained always as 20% (statistically only with vanishing probability ): # R function upper confidence limit of CV  as one liner gives design n df upper CL BTW: May it be that there is an error in this post in calculating SSintra from MSintra ? The final result is again correct . That remains me on my old school days: Bad school grade (5) if final result correct but intermediate steps with errors .— Regards, Detlew 
Helmut ★★★ Vienna, Austria, 20121017 17:28 (4064 d 08:28 ago) @ d_labes Posting: # 9428 Views: 6,365 

Dear Detlew, you made my day! ❝ Lets go with the numbers of subjects given by you and assume that the estimated CV is obtained always as 20% (statistically only with vanishing probability ): ❝ ❝ ❝ ❝ ❝ gives ❝ ❝ ❝ ❝ Interesting that the CV from the partial replicate is (although slightly) more precise than the 4period 2sequence full replicate. I didn’t expect that. Which design is the ‘best’? Pros & cons:
❝ BTW: May it be that there is an error in this post in calculating Blast! Will correct it. ❝ The final result is again correct . That remains me on my old school days: Bad school grade (5) if final result correct but intermediate steps with errors . Happened to me regularly. Especially if the final result came from my neighbor – who was much better in maths than I ever was. — Diftor 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, 20121017 18:39 (4064 d 07:17 ago) @ Helmut Posting: # 9429 Views: 6,410 

Dear Helmut, ❝ you made my day! My pleasure! Being pleased to do it again. ❝ Which design is the ‘best’? Pros & cons: [...] May be that the precision of CV_{wR} is different between these designs if powered to the same value of stating BE. But in no guidance I know the precision of CV_{wR} is an criterion. And we don't use it in any decision. So what? Nice to know? — Regards, Detlew 
Helmut ★★★ Vienna, Austria, 20121017 21:11 (4064 d 04:45 ago) @ d_labes Posting: # 9430 Views: 6,410 

Dear Detlew! ❝ ❝ Which design is the ‘best’? Pros & cons: [...] ❝ ❝ May be that the precision of CV_{wR} is different between these designs if powered to the same value of stating BE. ❝ But in no guidance I know the precision of CV_{wR} is an criterion. And we don't use it in any decision. Yessir! ❝ So what? Nice to know? Exactly – in the spirit of BioInternational ’94, where “Need to know vs. nice to know” was the unofficial slogan. Primarily I was interested whether it is worthwhile to go with TRTRTR instead of the partial replicate because you get the bonus of CV_{WT}. Some sponsors are interested how their formulation performs but the 4period replicate has its drawbacks. — Diftor heh smusma 🖖🏼 Довге життя Україна! _{} Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes 