Mithu
☆

India,
2020-07-09 09:18
(201 d 04:55 ago)

Posting: # 21659
Views: 2,064

## Ratio calculation of fully replicate study [RSABE / ABEL]

Dear All,

Hope all are doing well in such pandemic condition.

Today my question is, how to calculate ratio of fully replicate study in excel?

I have prepared one excel sheet in which i have put formula to calculate Partial AUC and AUC0-t.

I am getting correct PK for CMax & AUC's in line with SAS out put but when i am calculating Ratio, it is not matching with SAS out put.

Process followed to calculate ratio in excel is as follows:

1. from two T and two R taking average and then taking ratio of T /R.
2. Other case, i had also converted average of both T & R in LN and then taken ratio of LN T & LN R.

In both of the cases, i could not get ratio as per SAS. Does SAS has some other formula to calculate ratio?

Please help and educate me to calculate correct ratio of fully replicate study in line with SAS out put.

Regards,

Mithu

Edit: Category changed. [Helmut]
Helmut
★★★

Vienna, Austria,
2020-07-09 10:51
(201 d 03:23 ago)

@ Mithu
Posting: # 21660
Views: 1,683

## Ratio calculation of fully replicate study

Hi Mithu,

» […] how to calculate ratio of fully replicate study in excel?
» I have prepared one excel sheet in which i have put formula to calculate Partial AUC and AUC0-t.
» I am getting correct PK for CMax & AUC's in line with SAS out put but when i am calculating Ratio, it is not matching with SAS out put.

Without knowing your SAS code it would be reading tea leaves. Why on earth do you want to use Excel?

» Process followed to calculate ratio in excel is as follows:
» 1. from two T and two R taking average and then taking ratio of T /R.
» 2. Other case, i had also converted average of both T & R in LN and then taken ratio of LN T & LN R.
$$x/y=\exp (\log_{e}x-\log_{e}y)\;\left\{x,y \in \mathbb{R}^{+} \right\}\; {\color{red}{\neq}}\log_{e}x/\log_{e}y$$

Dif-tor heh smusma 🖖
Helmut Schütz

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

India,
2020-07-09 11:23
(201 d 02:50 ago)

@ Helmut
Posting: # 21661
Views: 1,678

## Ratio calculation of fully replicate study

Hello Sir,

» Without knowing your SAS code it would be reading tea leaves. Why on earth do you want to use Excel?

We have used Progesterone program.

» $$x/y=\exp (\log_{e}x-\log_{e}y)\;\left\{x,y \in \mathbb{R}^{+} \right\}\; {\color{red}{\neq}}\log_{e}x/\log_{e}y$$
Still I could not understand.

Regards,

Mithu
Helmut
★★★

Vienna, Austria,
2020-07-09 12:19
(201 d 01:54 ago)

@ Mithu
Posting: # 21662
Views: 1,668

## Ratio calculation of fully replicate study

Hi Mithu,

» » Without knowing your SAS code it would be reading tea leaves.
»
» We have used Progesterone program.

… which nowhere gives the subject’s T/R-ratios.

» » $$x/y=\exp (\log_{e}x-\log_{e}y)\;\left\{x,y \in \mathbb{R}^{+} \right\}\; {\color{red}{\neq}}\log_{e}x/\log_{e}y$$
» Still I could not understand.

Calculating the ratio of logs (2. of your OP) is nonsense.
The individual averaged log-differences $$I_{ij}$$ (for subject $$i$$ in sequence $$j$$) are computed in the FDA’s SAS macro by

ilat=0.5*(lat1t+lat2t-lat1r-lat2r);

Simply back-transform them.

Let’s have a look at the EMA’s full replicate reference data set:

SUBJECT   DATA  FORMULATION  PERIOD  SEQUENCE  logDATA     1    2285.96      R          1      BABA    7.734541    1    1955.82      T          2      BABA    7.578565    1    1345.94      R          3      BABA    7.204848    1    2856.24      T          4      BABA    7.957261

Then$$\overline{I_{11}}=0.5\times(7.578565+7.957261-7.734541-7.204848)=0.2982185$$and$$\overline{T_{11}/R_{11}}=\exp \overline{I_{11}}=\exp 0.2982185=1.347456$$Confirmed by Phoenix WinNonlin’s RSABE-template and my pocket calculator.

Which leaves my second question open:

» » Why on earth do you want to use Excel?

Dif-tor heh smusma 🖖
Helmut Schütz

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

India,
2020-07-09 12:46
(201 d 01:27 ago)

@ Helmut
Posting: # 21663
Views: 1,647

## Ratio calculation of fully replicate study

Hello Sir,

» Then$$\overline{I_{11}}=0.5\times(7.578565+7.957261-7.734541-7.204848)=0.2982185$$and$$\overline{T_{11}/R_{11}}=\exp \overline{I_{11}}=\exp 0.2982185=1.347456$$

Thank you Sir, now i understand.

» Which leaves my second question open:
»
» » » Why on earth do you want to use Excel?

Actually sir when we receive PK data from site (CRO), we check by using Excel. That's why i was using Excel for verification. Generally in average BE and partial replicate, we get it verified but for fully replicate study, i could not match the ratio.

This is the story.

thank you for your kind guidance and help.

regards,

Mithu

Helmut
★★★

Vienna, Austria,
2020-07-09 13:10
(201 d 01:03 ago)

@ Mithu
Posting: # 21664
Views: 1,627

## Purpose?

Hi Mithu,

» Actually sir when we receive PK data from site (CRO), we check by using Excel. That's why i was using Excel for verification. Generally in average BE and partial replicate, we get it verified but for fully replicate study, i could not match the ratio.
»
» This is the story.

I don’t understand the purpose of this ‘verification’. Calculate the study in SAS. The individual T/R-ratios don’t play any role in RSABE (log-differences are only an intermediate result).
Are you concerned about extreme values? They are common for HVD(P)s. 1 Contrary to ABEL (used in all jurisdictions for reference-scaling except in the US and China) assessing ‘outliers’ 2 is not acceptable for RSABE.

1. I have seen studies with subject ratios of 0.2–5 (‼) and still passing RSABE.
2. In ABEL of the reference treatment only.

Dif-tor heh smusma 🖖
Helmut Schütz

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

India,
2020-07-09 13:22
(201 d 00:51 ago)

@ Helmut
Posting: # 21665
Views: 1,628

## Purpose?

Dear Sir,

» I don’t understand the purpose of this ‘verification’. Calculate the study in SAS. The individual T/R-ratios don’t play any role in RSABE (log-differences are only an intermediate result).
» Are you concerned about extreme values? They are common for HVD(P)s. Contrary to ABEL (used in all jurisdictions for reference-scaling except in the US and China) assessing ‘outliers’ is not acceptable for RSABE.

Actually Sir, we do not have SAS at our end (sponsor) but now i got the point that such verification will not help in excel.

Actually at present i prepared excel for one of the our HVD and just learning how the data are interpreted.

Now i suggest my management to equipped us with proper tool to verify data correctly at our end.

CRO runs on SAS and providing PK sheet of data in excel to us and i was checking with excel. But today i got the correct way and understand how the software calculate ratio.

Regards,

Mithu.
Helmut
★★★

Vienna, Austria,
2020-07-09 13:56
(201 d 00:17 ago)

@ Mithu
Posting: # 21666
Views: 1,630

## Purpose?

Dear Mithu,

» Actually at present i prepared excel for one of the our HVD and just learning how the data are interpreted.

I have strong doubts that is possible to set up the FDA’s mixed-effects model for ABE in Excel. I would never ever rely on its Solver Add-in.

» CRO runs on SAS …
» Now i suggest my management to equipped us with proper tool to verify data correctly at our end.

Trust, but verify.    (Russian proverb)

You would need SAS, JMP, or Phoenix WinNonlin (maybe SPSS, Stata, STATISTICA). Don’t know which one is the least expensive. Only for SAS and Phoenix WinNonlin free code is available. For the others you are on your own.

BTW, why are you regularly ignoring some of my questions? Still open:

» » Are you concerned about extreme values? They are common for HVD(P)s. Contrary to ABEL (used in all jurisdictions for reference-scaling except in the US and China) assessing ‘outliers’ is not acceptable for RSABE.

What will you do (or have done in the past) with your ‘verification’? The T/R-ratios are not part of the RSABE SAS-output provided by the CRO.

Dif-tor heh smusma 🖖
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
jag009
★★★

NJ,
2020-07-09 18:01
(200 d 20:13 ago)

@ Helmut
Posting: # 21667
Views: 1,593

## Purpose?

Hi Helmut,

Sticking my nose in...

» I have strong doubts that is possible to set up the FDA’s mixed-effects model for ABE in Excel. I would never ever rely on its Solver Add-in.

No. Stay away from doing that... I've tried before (just to waste time in the office).

» What will you do (or have done in the past) with your ‘verification’? The T/R-ratios are not part of the RSABE SAS-output provided by the CRO.

I think he might be doing topline prelim analysis for his management team. You know, fresh off the oven? Unfortunately he doesn't have the necessary tools so he's stuck w Excel... I wouldn't verify what the CRO produces unless I have a glance at their SOPs and SAS codes. I mean I would verify but the CRO results are usually the final ones.

J
Mithu
☆

India,
2020-07-10 08:00
(200 d 06:13 ago)

@ Helmut
Posting: # 21668
Views: 1,559

## Purpose?

Dear Sir,

» » » Are you concerned about extreme values? They are common for HVD(P)s. Contrary to ABEL (used in all jurisdictions for reference-scaling except in the US and China) assessing ‘outliers’ is not acceptable for RSABE.
»
» What will you do (or have done in the past) with your ‘verification’? The T/R-ratios are not part of the RSABE SAS-output provided by the CRO.

First of all please accept my apology. My intention is not to ignoring your question. Actually i do not have any answer for it.

I do understand that ratio is not only part of SABE study passout. but this is the only tool which i could work out in excel. for 95% upper bound, i don't have any idea or knowledge to check in excel. Even i could also not able to calculate associate variability i.e. swr in tool like excel.

Learning form internet how to calculate PK and putting some efforts at my end.

I am learning statistics and when stuck rely on this forum for answer.

Again i am sorry if you still fell that i am ignoring your questions.

Regards,

Mithu.