## Group effects FDA/EMA [Two-Stage / GS Designs]

Hi zizou,

sorry for reviving this old thread. See also this lengthy one.

» […] there is nothing stated from EMA on testing of such groups (or did I miss something?).

I don’t think so. I know only one case (biosimilar in 2016) where the statistician (“borrowed” from the MHRA) required the FDA’s model 2.

» I have opinion that group effect is not important in 2x2 crossover with proper planning/realization.

Agree – especially with this one.

» Moreover I have never seen group testing for EMA and I like the sentence "With one week between groups I would never ever thought a millisecond of setting up a group model." from this post.

Agree.

» I prepared

THX! I could reproduce your results with my R-code.

» Because the data are written by me randomly and then changing and changing to get the nice and not nice results at the same. The groups separately provide us with next BE results:

»

»

»

» CVintra = 30.9%

» PE (LL,UL) =

»

»

»

» CVintra = 19.4%

» PE (LL,UL) =

Nice example. My current thinking is: The decision scheme recommended by the FDA leads to nowhere. If there is a significant G×T interaction we are not allowed to pool. So far so good. If groups are not of the same size, everybody would present the results of the largest group evaluated by the conventional model. But: What if groups have equal sizes (like in your example)? Cherry-pick and present the “better” one? I bet the assessor would ask for the other one as well. If I would be a regulator I would require that

Furthermore, if there is a true G×T interaction the treatment effect might be biased. To which extent is unknown (cannot be estimated from the data). It might well be that a smaller group is closer to the true treatment effect than a larger one. We simply don’t know. IMHO, the assumption that “size matters” is false.

Your example is even more telling in another respect. Not only the PEs are different but also the variances. Should we blindly pool them? I know, the common model assumes equal variances anyhow, but…

I have just limited data (always tried to avoid equal group sizes). These are the results of my data sets which show a significant G×T interaction in model 1 and I evaluated the equally sized largest groups by model 3:

» If group testing was required by EMA I would prefer test for group*treatment interaction.

Disagree. What if you find one? Expected in 10% of studies by pure chance. Then you end up with the story from above.

» But I am still convinced that group effect is not required to test in standard 2x2 crossover study. For me it seems similar to have 2x2 crossover study on 24 subjects in 1 group and test for Period*Treatment interaction (e.g. in model Period*Treatment, Sequence, Subjects(Sequence) - if Period*Treatment significant, use data from one period as in parallel design.

Agree. Grizzle’s flawed testing for a sequence- (or better: unequal carryover-) effect… Cannot be handled statistically and only avoided

My current thinking for the EMA (maybe I’m wrong): Avoid a potential G×T interaction

sorry for reviving this old thread. See also this lengthy one.

» […] there is nothing stated from EMA on testing of such groups (or did I miss something?).

I don’t think so. I know only one case (biosimilar in 2016) where the statistician (“borrowed” from the MHRA) required the FDA’s model 2.

» I have opinion that group effect is not important in 2x2 crossover with proper planning/realization.

Agree – especially with this one.

» Moreover I have never seen group testing for EMA and I like the sentence "With one week between groups I would never ever thought a millisecond of setting up a group model." from this post.

Agree.

» I prepared

**example**(no real data ... to get "unlucky" results).THX! I could reproduce your results with my R-code.

» Because the data are written by me randomly and then changing and changing to get the nice and not nice results at the same. The groups separately provide us with next BE results:

»

»

**Group 1:**»

`data1=subset(data,data[,"Groups"]==1)`

» CVintra = 30.9%

» PE (LL,UL) =

**0.8774 (0.7554,1.0191)**»

»

**Group 2:**»

`data2=subset(data,data[,"Groups"]==2)`

» CVintra = 19.4%

» PE (LL,UL) =

**1.1689 (1.0626,1.2858)**Nice example. My current thinking is: The decision scheme recommended by the FDA leads to nowhere. If there is a significant G×T interaction we are not allowed to pool. So far so good. If groups are not of the same size, everybody would present the results of the largest group evaluated by the conventional model. But: What if groups have equal sizes (like in your example)? Cherry-pick and present the “better” one? I bet the assessor would ask for the other one as well. If I would be a regulator I would require that

*both*pass or – if not – make a conservative decision: fail.Furthermore, if there is a true G×T interaction the treatment effect might be biased. To which extent is unknown (cannot be estimated from the data). It might well be that a smaller group is closer to the true treatment effect than a larger one. We simply don’t know. IMHO, the assumption that “size matters” is false.

Your example is even more telling in another respect. Not only the PEs are different but also the variances. Should we blindly pool them? I know, the common model assumes equal variances anyhow, but…

I have just limited data (always tried to avoid equal group sizes). These are the results of my data sets which show a significant G×T interaction in model 1 and I evaluated the equally sized largest groups by model 3:

`AUC`

drug group n df mse CV(%)

a 1 9 7 0.006026 7.77

a 2 9 7 0.005799 7.63

b 1 12 10 0.014851 12.23

b 2 12 10 0.006714 8.21

c 1 9 7 0.001881 4.34

c 2 9 7 0.001462 3.82

Cmax

drug group n df mse CV(%)

b 1 12 10 0.036411 19.26

b 2 12 10 0.026247 16.31

*p*0.6232. Drugs a & c (with extremely low CVs) were NTIDs and the PI wanted groups for safety reasons.» If group testing was required by EMA I would prefer test for group*treatment interaction.

Disagree. What if you find one? Expected in 10% of studies by pure chance. Then you end up with the story from above.

» But I am still convinced that group effect is not required to test in standard 2x2 crossover study. For me it seems similar to have 2x2 crossover study on 24 subjects in 1 group and test for Period*Treatment interaction (e.g. in model Period*Treatment, Sequence, Subjects(Sequence) - if Period*Treatment significant, use data from one period as in parallel design.

Agree. Grizzle’s flawed testing for a sequence- (or better: unequal carryover-) effect… Cannot be handled statistically and only avoided

*by design*.My current thinking for the EMA (maybe I’m wrong): Avoid a potential G×T interaction

*by design*(*i.e.*, comply with the FDA’s conditions for pooling, keep the interval between groups short, if possible use the staggered approach).- State in the SAP that
*“it is a source of variation that cannot be reasonably assumed to have an effect on the response variable”*. Pool the data and evaluate the study by the common 2×2 model.

- Essentially the FDA is correct that the multigroup nature of the study should be taken into account – if (!) groups are separated by a long interval (months). If you want to go this way (remember: I have seen only a
*single*case where the EMA asked for it), state it in the SAP.*No pre-test*of the G×T interaction by model 1! Pool the data and evaluate the study by the FDA’s model 2. The loss in power compared to the common 2×2 model is small.

—

Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮

Science Quotes

*Dif-tor heh smusma*🖖Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮

Science Quotes

### Complete thread:

- Potvin C in the EU Helmut 2013-04-16 17:40
- Potvin C in the EU ElMaestro 2013-04-17 02:31
- Potvin C in the EU Helmut 2013-04-17 12:23
- 2 Groups model FDA d_labes 2013-04-17 13:10
- 2 Groups model FDA Helmut 2013-04-17 16:37
- Group effects obsolete? d_labes 2013-04-18 09:55
- Group effects FDA/EMA Helmut 2013-04-19 14:34
- Group effects FDA/EMA mittyri 2014-08-21 15:34
- Group effects FDA/EMA ElMaestro 2014-08-21 17:31
- Group effects EMA mittyri 2014-10-01 11:25
- Group effects EMA ElMaestro 2014-10-01 11:31
- Group effects EMA Helmut 2014-10-01 13:24
- Group effects EMA VStus 2016-10-06 22:06
- Group effects EMA ElMaestro 2016-10-07 12:45
- Group effects EMA mittyri 2016-10-07 15:06
- Fixed effects model with Group term mittyri 2016-10-09 13:27
- Fixed effects model: changing the F, p values mittyri 2016-10-10 18:33
- Fixed effects model: changing the F, p values ElMaestro 2016-10-10 20:10
- Fixed effects model: changing the F, p values Helmut 2016-10-11 00:17
- Holy War of type III d_labes 2016-10-11 13:17
- Significance of Group effect in Russia: why the type III is so 'important' mittyri 2016-10-11 15:07

- Fixed effects model: changing the F, p values zizou 2016-10-11 01:06
- FDA group model in R mittyri 2016-10-11 13:43
- FDA group model in R ElMaestro 2016-10-12 01:07
- FDA group model in R VStus 2016-10-17 15:32
- FDA group model in R ElMaestro 2016-10-17 18:58
- FDA group model in R VStus 2016-10-20 13:57

- FDA group model in R ElMaestro 2016-10-17 18:58

- FDA group model in R mittyri 2016-10-11 13:43

- Fixed effects model: changing the F, p values ElMaestro 2016-10-10 20:10
- Fixed effects model with Group term VStus 2016-10-20 13:54

- Fixed effects model: changing the F, p values mittyri 2016-10-10 18:33

- Fixed effects model with Group term mittyri 2016-10-09 13:27

- Group effects EMA mittyri 2016-10-07 15:06

- Group effects EMA ElMaestro 2016-10-07 12:45

- Group effects EMA VStus 2016-10-06 22:06

- Group effects FDA/EMA zizou 2016-12-31 02:54
- Group effects FDA/EMAHelmut 2017-06-03 14:57

- Group effects EMA mittyri 2014-10-01 11:25

- Group effects FDA/EMA ElMaestro 2014-08-21 17:31

- Group effects FDA/EMA mittyri 2014-08-21 15:34

- Group effects FDA/EMA Helmut 2013-04-19 14:34
- 2 Groups model FDA hiren379 2013-08-22 14:32
- Homework Helmut 2013-08-23 11:53

- Group effects obsolete? d_labes 2013-04-18 09:55

- 2 Groups model FDA Helmut 2013-04-17 16:37

- 2 Groups model FDA d_labes 2013-04-17 13:10

- Potvin C in the EU Helmut 2013-04-17 12:23

- Potvin C in the EU ElMaestro 2013-04-17 02:31