jag009
★★★

NJ,
2019-02-25 22:03

Posting: # 19972
Views: 3,693

## History lessons - FDA BE 90% CI [Regulatives / Guidelines]

Hi,

Can someone tell me what the BE requirements were 1990/before, FDA? I mean the 90% CI requirement. Yeah I was still in school....

Thanks
J
nobody
nothing

2019-02-25 22:39

@ jag009
Posting: # 19973
Views: 3,463

## History lessons - FDA BE 90% CI

75/25, 90%CI was rocket science

Kindest regards, nobody
Helmut
★★★

Vienna, Austria,
2019-02-26 00:44

@ nobody
Posting: # 19974
Views: 3,432

## History lessons - FDA BE 90% CI

Γεια σου, λακωνικό Οδυσσέα!

» 75/25, 90%CI was rocket science

Don’t forget #3 and #4 mentioned in this post.

The 75/75 rule was proposed in 1978 for tricyclic antidepressants.1

(2) The test drug product shall be deemed to meet the bioequivalence requirement for in vivo testing in humans if the following conditions are met:
(ii) In at least 75 percent of the subjects administered the drug, the test drug product has a bioavailability of greater than 75 percent relative to that of the administered reference material utilizing each subject as his own comparison;

It was defended by the FDA2 but immediately criticized3 because already earlier4,5 it was shown to reject >50% of studies if the variability is large even for T/R=1.

I don’t know when the 80/20 rule (aka the “power approach”) was introduced (mid-1980s?). Schuirmann showed why it is crap.6 Unfortunately (for historic reasons) it is still part of Phoenix/WinNonlin’s output and happily reported by some (see my rant, follow-up).

@John: The 90% CI was recommended in 1992.7

1. Fed. Reg. 6913 (Feb. 17, 1978); §320.107:6968–9. free resource.
2. Cabana BE. Assessment of 75/75 Rule: FDA Viewpoint. J Pharm Sci. 1983;72(1):98–99. doi:10.1002/jps.2600720127.
3. Haynes JD. FDA 75/75 Rule: A Response. J Pharm Sci. 1983;72(1):99–100. doi:10.1002/jps.2600720128.
4. Haynes JD. Statistical Simulation Study of New Proposed Uniformity Requirement for Bioequivalency Studies. J Pharm Sci. 1981;70(6):673–5. doi:10.1002/jps.2600700625.
5. Metzler CM, Huang DC. 1983. Statistical methods for bioavailability and bioequivalence. Clin Res Pract Drug Reg Aff. 1983;1:109–132.
6. Schuirmann DJ. A comparison of the Two One-Sided Tests Procedure and the Power Approach for Assessing the Equivalence of Average Bioavailability. J Pharmacokin Biopharm. 1987;15(6):657–80. doi:10.1007/BF01068419.
7. FDA/CDER. Guidance for Industry. Statistical Procedures for Bioequivalence Studies Using a Standard Two-Treatment Crossover Design. Jul 1992.

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,
2019-02-26 15:19

@ Helmut
Posting: # 19976
Views: 3,330

## History lessons - FDA BE 90% CI

Thanks guys!

» @John: The 90% CI was recommended in 1992.7

Thanks Helmut. The reason I asked is because I read some report (dated 1990; the synopsis only) and the one sided 90% CI was reported as lower = -18.33%, upper = -6.36%; T/R Ratio = 93.86, n=24. I found it strange to report CI in such a way so I was wondering if the reporting was based on some old rules.

Thanks
J
nobody
nothing

2019-02-26 16:07

@ jag009
Posting: # 19978
Views: 3,374

## History lessons - FDA BE 90% CI

...maybe guys at that time where looking if the CI included 0 (Danish style, so to say...) or trying to hide that the lower bound is below 0.8... ? Just kidding...

Kindest regards, nobody
Helmut
★★★

Vienna, Austria,
2019-02-28 12:03

@ jag009
Posting: # 19979
Views: 3,244

## Strange result

Hi John,

» […] I read some report […] and the one sided 90% CI was reported as lower = -18.33%, upper = -6.36%; T/R Ratio = 93.86, n=24. I found it strange to report CI in such a way so I was wondering if the reporting was based on some old rules.

Strange!
As the name tells, TOST are interval (significance) tests giving two p-values – one for the 1st (left) test where

H01: PE < ln(1–0.2)
H11: PE ≥ ln(1–0.2)

and one for the 2nd (right) test where

H02: PE > ln(1+0.2)
H12: PE ≤ ln(1+0.2).

Only if both H01 and H02 are rejected at the level α, BE is concluded. In my entire career I haven’t seen a single report where this was done1 – though most claim that TOST was applied. The confidence interval inclusion test is two-sided (though OTST is difficult to pronounce) and “operationally equivalent” to TOST. Some call that just “an algebraic coincidence”.2 There is no CI in TOST.3 The FDA’s 1992 guidance4 mixes both approaches up:

The two one-sided hypotheses at the α = 0.05 level of significance should be tested for AUC and Cmax by con- structing the 90% confidence interval for the ratio between the test and reference averages.

However, conclusions of TOST and the confidence interval inclusion method are the same.

Coming back to your case. I’ve never seen results reported in such a way.
If we would report the CI as usual we would give (93.86 – 18.33)% = 75.53% and (93.86 – 6.36)% = 87.53% and the study failed (like nobody assumed). But then we have another problem with the reported T/R-ratio which transforms to (93.86 – 100)% = –6.14%. Then –18.33% < –6.14% < –6.36%, or what? Based on 100√0.7553 × 0.8753 we get 81.29% ≠ 93.86%.
Or are the boundaries given relative to 100%? Bizarre. However, then the study would pass since (100 – 18.33)% = 81.67% and (100 – 6.36)% = 93.64%. Note that the reported PE is outside the CI…

» […] (dated 1990; the synopsis only)

Was the study performed for Health Canada? In the 1989 draft 80–120% (untransformed data) were recommended and changed to 80–125% (log-transformed) in 1991.
Then the study would have passed again cause –18.33% > –20% and –6.36% < +20%. However, the problem with the PE persists cause 100(–0.1833 + (–0.0636)) / 2 = –12.35% ≠ –6.14%. I don’t get it.

1. I did it myself in a few studies upon sponsor’s wish. In all cases assessors asked for the 90% CI later…
For ages I do it the other way ’round. Describe the confidence interval inclusion approach according to the guidelines and – to make newbies happy – state that it is “operationally equivalent” to TOST.
2. Brown LD, Casella G, Hwang JTG. Optimal Confidence Sets, Bioequivalence, and the Limaçon of Pascal. J Amer Statist Assoc. 1995;90(431):880–9. doi:10.1080/01621459.1995.10476587. free resource.
3. Schuirmann DJ. A comparison of the Two One-Sided Tests Procedure and the Power Approach for Assessing the Equivalence of Average Bioavailability. J Pharmacokin Biopharm. 1987;15(6):657–80. doi:10.1007/BF01068419.
4. FDA/CDER. Guidance for Industry. Statistical Procedures for Bioequivalence Studies Using a Standard Two-Treatment Crossover Design. Jul 1992.

Dif-tor heh smusma 🖖
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
nobody
nothing

2019-03-01 08:50

@ Helmut
Posting: # 19981
Views: 3,157

## Strange result

imho it's the CI for the difference in AUC(?) between T and R, although this is really a strange way to report the results. Might be to obfuscate somefink. Anyway I would be careful with data that old, if you don't have access to methods (including bioanalytics) and raw data.

Kindest regards, nobody
jag009
★★★

NJ,
2019-03-01 17:40

@ nobody
Posting: # 19982
Views: 3,209

## Pandora's box?

Hi Guys (and Gals!),

It's a 2-way crossover US study (definitely not anywhere else, clinical done in 1990 so was report). Here is the exact wordings from the table (if that makes any difference to what I posted). Note that the summary table also presented +/- Westlake 95% CI..

Cmax(n=24):
Detectable differences(%)=13.28
T/R Ratio=93.86%
+/- Westlake 95% CI=20.12%
One sided t-test 90% CI:Lower limit= -18.33%, Upper Limit= -6.36%

Thanks
J
Helmut
★★★

Vienna, Austria,
2019-03-02 15:37

@ jag009
Posting: # 19985
Views: 3,047

## Confusing

Hi John,

» Here is the exact wordings from the table (if that makes any difference to what I posted). Note that the summary table also presented +/- Westlake 95% CI..
»
» Cmax(n=24):
» Detectable differences(%)=13.28

“Detectable difference” stinks of the FDA’s 80/20 rule (at least 80% post hoc power to detect a 20% difference). Was applied till 1992. IMHO, since 13.28% < 20%, Cmax failed.

» T/R Ratio=93.86%
» +/- Westlake 95% CI=20.12%

Why 95%? However, Westlake’s CI is an information sink (example). By fiddling around with the t-values his CI is always symmetric around 100%. He suspected that clinicians are not comfortable with a CI which is asymmetric (see this post). As a side effect the T/R-ratio should not be given.

» One sided t-test 90% CI:Lower limit= -18.33%, Upper Limit= -6.36%

Here’s the other way ’round. Should be 95%.

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,
2019-03-01 17:44

@ Helmut
Posting: # 19983
Views: 3,117

## Strange result

Hi Helmut!

» Was the study performed for Health Canada? In the 1989 draft 80–120% (untransformed data) were recommended and changed to 80–125% (log-transformed) in 1991.
» Then the study would have passed again cause –18.33% > –20% and –6.36% < +20%. However, the problem with the PE persists cause 100(–0.1833 + (–0.0636)) / 2 = –12.35% ≠ –6.14%. I don’t get it.

That I do not know (if it's for Canada) but it was a us study w US products. But your suggestion about Canada using non-transformed make sense(?) Can you tell me (or pt to me) about the Canadian guidance 89?

Thanks
J
Helmut
★★★

Vienna, Austria,
2019-03-02 15:09

@ jag009
Posting: # 19984
Views: 3,075

## Transformation, acceptance range

Hi John,

» » Was the study performed for Health Canada? In the 1989 draft 80–120% (untransformed data) were recommended and changed to 80–125% (log-transformed) in 1991.
»

Another goody: At the “International Open Conference on Dissolution, Bioavailability, Bioequivalence” (Toronto, June 15–17, 1992) the current draft was provided (without a date), which stated

95% CI of relative mean AUC 80%–125%.

John Ruedy (Chairman, Expert Advisory Committee of Biovailability, HC’s HPB) pointed out in his presentation that the final guidance will require a 90% CI.
Couldn’t find the pre-1992 guidance so far. Maybe my memory mixed the transformation-business up with this story (see the end of this post).

Dif-tor heh smusma 🖖
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
nobody
nothing

2019-03-02 16:42

@ Helmut
Posting: # 19986
Views: 3,071

## Transformation, acceptance range

Take the report and make little pieces out of it, makes perfect notepads. Otherwise no use of such data, except if you dig out some raw data

Kindest regards, nobody