Bioequivalence and Bioavailability Forum • WinNonlin: Data handling

Anu
★

India,
2014-02-26 14:42
(4072 d 19:52 ago)

Posting: # 12508
Views: 4,387

WinNonlin: Data handling [Software]

Dear All,

Greetings!

While performing BE analysis with the help of WinNonlin, I handle the data as follows:

2-way cross over: If data of a particular subject for 1 period is missing, I do not consider that particular subject in my analysis.
3-way crossover: If data of a particular subject for 2 periods is missing, I do not consider that particular subject in my analysis.
Fully Replicate: If data of a particular subject for 3 periods is missing, I do not consider that particular subject in my analysis.

Is this approach is correct? If not kindly let me know,how to handle such type of missing data? Is their any guideline for the same.

Thanks & Regards
Anu

Edit: Category changed. [Helmut]

Helmut
★★★

Vienna, Austria,
2014-02-26 15:26
(4072 d 19:08 ago)

@ Anu
Posting: # 12511
Views: 3,645

WinNonlin: Data handling

Post reply

Hi Anu,

first see EMA’s GL, Section 4.1.8, Subject accountability:

Ideally, all treated subjects should be included in the statistical analysis. However, subjects in a
crossover trial who do not provide evaluable data for both of the test and reference products […] should not be included.

Yes.
Generally yes. But there is a case where you would drop the subject even if only one period is missing: Two test available and the reference missing.
By fully replicate you talk about the four-period RTRT|TRTR. No doubt about three missing periods. It’s tricky if both administrations of the test are missing. Although IMHO it is a little bit strange why a subject missing a period should come back for others. But EMA’s Q&A-document contains such a wacky example data set and I have seen a case in the wild. Would you keep the subject in the analysis only to get a better estimate of CV_WR? I would not; this time agreeing with EMA’s GL above.
Homework: Fully replicate three-period design (RTR|RTR).

—
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

Shuanghe
★★

Spain,
2014-02-26 19:50
(4072 d 14:44 ago)

@ Helmut
Posting: # 12519
Views: 3,566

WinNonlin: Data handling

Post reply

Hi Helmut,

Agree with the first 2 points.

❝ 3. By fully replicate you talk about the four-period RTRT|TRTR. No doubt about three missing periods. It’s tricky if both administrations of the test are missing. Although IMHO it is a little bit strange why a subject missing a period should come back for others. But EMA’s Q&A-document contains such a wacky example data set and I have seen a case in the wild. Would you keep the subject in the analysis only to get a better estimate of CV_WR? I would not; this time agreeing with EMA’s GL above.

Well, for FDA you might need to do it differently.

I asked FDA (along with other questions) if subject who missed 2 periods of test but has data of 2 periods of reference should be included to have a better estimate of S_WR and after 1 year and 3 months I finally got answers from them:

For 4-period replicate BE study subject who has 2 periods of reference should be included for S_WR (even if they has no data of test product) but for estimate the difference between T and R only subjects who have no missing period should be included (so if subject has only 3 periods, exclude him).

The response is shockingly slow but at least it leaves no doubt on this topic.

I think the principle for estimation of S_WR should be the same for EMA study as well for which I usually clearly stated in the protocols that subject with 2 periods of reference (even if no test data) will be included to estimate the S_WR but for evaluation of BE subject will have crossed at least once (minimum 1T and 1R) as indicated in the guideline and in the sample dataset in Q&A document.

—
All the best,
Shuanghe

Anu
★

India,
2014-02-27 06:28
(4072 d 04:06 ago)

@ Helmut
Posting: # 12523
Views: 3,608

WinNonlin: Data handling

Post reply

Hi Helmut,

Thank you so much for the reply :-)

❝ 3. Generally yes. But there is a case where you would drop the subject even if only one period is missing: Two test available and the reference missing.

ok right, can't take BE in this case too.

❝ Homework: Fully replicate three-period design (RTR|RTR).

In this case i.e. Partially replicate design, I would consider that subject, In which their is 1(T) & 1 (R) available and would not consider the Subjects with both (R)or with just 1(T) or 1 (R).

Thanks & Regards