Mahmoud
☆

Jordan,
2020-07-03 14:09
(473 d 03:57 ago)

Posting: # 21642
Views: 2,533

## SAS code for ABE [Software]

Dear All

in the following SAS code

proc mixed data=d1; class treat sequence subject period; model AUCT=sequence period treat/DDFM=kr; random treat/type=CSH subject=subject G; repeated/group=treat subject=subject; estimate 'T-R' treat 1 -1 /cl alpha=0.10; run;

In the Random statement TYPE=CSH could possibly be replaced by TYPE=FA(1)

FA(1) is not the same as FA0(2)

Thank You

M.Youseed

Helmut
★★★

Vienna, Austria,
2020-07-03 14:46
(473 d 03:19 ago)

@ Mahmoud
Posting: # 21644
Views: 2,100

## Origin of SAS code?

Hi Mahmoud,

» in the following SAS code

where does this code come from? It’s not the standard one recommended by the FDA here and there.

» model AUCT=sequence period treat/DDFM=kr;

I would suggest to use DDFM=SATTER (as the FDA recommends). The default of DDFM=KR uses the observed information matrix (SCORING=0) as does jmp. You may run into troubles if your study is re-evaluated in other software which uses the expected information matrix (e.g., Stata, R-package replicateBE). In SAS you can get the expected information matrix by setting SCORING=1.

» random treat/type=CSH subject=subject G;
»
» In the Random statement TYPE=CSH could possibly be replaced by TYPE=FA(1)

Acc. to the FDA’s guidance, Appendix E:

In the Random statement, TYPE=FA0(2) could possibly be replaced by TYPE=CSH.

So why compound symmetry in the first place?

» FA(1) is not the same as FA0(2)

Of course. FA(q) = factor analytic and FA0(q) = no diagonal factor analytic.
If this is a full replicate design, I would follow the FDA’s recommendation and use FA0(2).
However, in the stupid partial replicate designs (TRR|RTR|RRT or TRR|RTR) the optimizer may fail to converge since the model is over-specified (T not repeated). Then you’ve performed a study and don’t get a result cause SAS shows you the finger.* I suggest to specify FA0(1) instead. State that already in the SAP.

• If you are lucky SAS throws only a warning.
Convergence criteria met but final hessian is not positive definite.
But if Murphy’s law hits:
WARNING: Did not converge. WARNING: Output 'Estimates' was not created.

Dif-tor heh smusma 🖖
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Mahmoud
☆

Jordan,
2020-07-03 16:49
(473 d 01:17 ago)

@ Helmut
Posting: # 21646
Views: 2,090

## SAS code for ABE

Thank you very much

The BE study is (TRRT,RTTR) two sequence four period

                                                         GMR        Lower      Upper when i used FDA code the result for  Cmax                85.734    79.363     92.617 when i used code FDA but type=FA(1) the result is Cmax   85.925    80.087     92.190

Helmut
★★★

Vienna, Austria,
2020-07-03 17:00
(473 d 01:05 ago)

@ Mahmoud
Posting: # 21647
Views: 2,052

## SAP?

Hi Mahmoud,

»                                                          GMR        Lower      Upper
» when i used FDA code the result for  Cmax                85.734    79.363     92.617
» when i used code FDA but type=FA(1) the result is Cmax   85.925    80.087     92.190

Are you trying to “rescue” a failed study? What is stated in the SAP?

Dif-tor heh smusma 🖖
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Mahmoud
☆

Jordan,
2020-07-03 17:17
(473 d 00:48 ago)

@ Helmut
Posting: # 21648
Views: 2,035

## SAP?

Dear Dr

Thank you very much for your information

in This study I found that and for Cmax Pk that

CVintra =27.3217 and CVinter=20.7271 this is not true in BE studies

From most studies CVinter is more more than CVintra

I think that the two types=CSH or FA0(1) in the FDA code are not suitable to obtain these results

» Are you trying to “rescue” a failed study?

No

Helmut
★★★

Vienna, Austria,
2020-07-03 17:31
(473 d 00:34 ago)

@ Mahmoud
Posting: # 21649
Views: 2,033

## SAP?

Hi Mahmoud,

» Dear Dr

Still not a Dr. Told you before.

» in This study I found that and for Cmax Pk that
» CVintra =27.3217 and CVinter=20.7271 this is not true in BE studies

Truth does not belong to the realm of science.

» From most studies CVinter is more more than CVintra

Correct; sometimes CVintra > CVinter.

» I think that the two types=CSH or FA0(1) in the FDA code are not suitable to obtain these results

Why, can you elaborate?

» » Are you trying to “safe” a failed study?
» No

I see. Please answer my other question (also in the subject line)

» » What is stated in the SAP?

as well.

Dif-tor heh smusma 🖖
Helmut Schütz

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