Bioequivalence and Bioavailability Forum • Lansoprazole

shankar
★

India,
2007-06-21 11:39
(6945 d 16:29 ago)

Posting: # 825
Views: 9,806

Lansoprazole [Design Issues]

Hi All,

We have to plan bioequivalence study on Lansoprazole under fasting and fed conditions. Now, as Lansoprazole is highly variable in fed condition (intrasubject CV (%) is 70-82%) based on available literature. The sample size coming around 450 subjects.

In my opinion we can plan replicate study, but with how many no. of subjects??

Please suggest how to plan the study design.

Shankar NJ

—
Shankaranarayanan

drshiv ★ India, 2007-06-27 12:22 (6939 d 15:46 ago) @ shankar Posting: # 837 Views: 7,218	Lansoprazole Post reply
	Hi Shankar, See the FDA Industry Guidance on Statistical Issues in Bioavailability and Bioequivalence Studies. Download from FDA site. There are tables given to find out the number of subjects required in replicate design based on %CV. drshiv -- Edit: Full quote removed. Please see this post! [HS]

Helmut ★★★ Vienna, Austria, 2007-06-30 16:02 (6936 d 12:06 ago) @ shankar Posting: # 850 Views: 7,234	HVDs Post reply
	Dear Shankar! ❝ We have to plan bioequivalence study on Lansoprazole under fasting and fed conditions. For which country? Answer depends on the regulation… — 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

shankar ★ India, 2007-07-04 13:17 (6932 d 14:51 ago) @ Helmut Posting: # 863 Views: 7,134	HVDs Post reply
	Dear Hs, This study for EU Submission. Shankar NJ

Helmut
★★★

Vienna, Austria,
2007-07-06 18:34
(6930 d 09:34 ago)

(edited on 2007-12-13 17:36)
@ shankar
Posting: # 869
Views: 7,366

HVDs in the EU :-((

Post reply

Dear shankar!

❝ This study for EU Submission.

See this post on *prazoles and my experiences in the EU.

Where did you get your CV% from? Even for lansoprazole >50% is rather high!
There are studies published with a CV% of 30–40 for C_max. :-D

According to the 'Q&A Document to the NfG' and the 'Concept Paper on HVDs/HVDPs', I would guess that:

you will have to demonstrate high variability in a replicate design study (this is not the pivotal BE).
if you were able to do so, you may widen the acceptance range (for C_max only), if clinically justifiable.
you may perform the pivotal BE either as a conventional 2×2 (sample size N), a replicate design (3×2 of sample size 2N/3), or a replicate design (4×2 of sample size N/2). If your 2×2 sample size is 48, you essentially get the same power in a 3×2 replicate in 36 subjects and in a 4×2 replicate in 24 subjects.
since in the Recommendation for an Update of the NfG Reference Scaled Average Bioequivalence (RSABE) is not mentioned (some European members states raised major objects to the Concept Paper on HVDs), I would not dare suggesting this approach any more (at least not before a positive opinion in a scientific advisory meeting).
Although some members states may accept RSABE, in my opinion a centralized application or a MRP procedure would be almost futile...

--
Edit: The Concept Paper on HVDs/HVDPs was deleted from the EMEA's website in October 2007; you may download a copy here.

—
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

drks
☆

2007-07-07 17:53
(6929 d 10:15 ago)

@ Helmut
Posting: # 870
Views: 7,231

HVDs in the EU :-((

Post reply

Dear Helmut,

❝ Where did you get your CV% from? Even for lansoprazole >50% is rather high!

❝ There are studies published with a CV% of 30-40 for Cmax. :-D

I would agree with Shankar in this regard. There are studies with such high variability (70-80%) observed with lansoprazole. We have observed similar variability for lansoprazole, particularly with fed studies. In fact, for various studies submitted to MHRA, the confidence interval has been widened on the basis of high variability. One can find different approvals with studies showing such high variability at the MHRA site upon searching for lansoprazole.

Helmut
★★★

Vienna, Austria,
2007-07-08 15:01
(6928 d 13:07 ago)

@ drks
Posting: # 871
Views: 7,496

Lansoprazole in the UK

Post reply

Dear Kshitij,

thanks for your hint!

❝ I would agree with Shankar in this regard. There are studies with such high variability (70-80%) observed with lansoprazole. We have observed similar variability for lansoprazole, particularly with fed studies. In fact, for various studies submitted to MHRA, the confidence interval has been widened on the basis of high variability.

I found two applications (authorisation granted). Studies were performed on the 30 mg capsules (and 15 mg authorisation was waived based on in vitro data).
In the first application the acceptance range for Cmax was set to 70%-143% (23 subjects), and in the second on to 75%-133% (52 subjects).
For the first product the lower limit was exceeded for Cmax in fed state (69.8%), whereas for the second product upper limits in fed stated (multiple dose) were exceeded, both for Cmax and AUC. In this study even point estimates were outside the acceptance limits (Cmax 144.92%, AUCtau 157.87%). The study wa not only underpowered to demonstrate BE, but actually bioinequivalence was proven!
CVs calculated from CIs were 27.5%-38.6% in studies of the first product, and 17.7%-30.1% in studies of the second product with the exeption of the fed/MD study, were CV was 81.7% for AUCtau and 90.7% for Cmax (!).

Most interesting is the Assessor's Comment for the second product:

The results of the fasted study indicate unequivocal equivalence. The 90% confidence intervals (CI) for the AUC and Cmax ratios fall within the conventional 80-125% bioequivalence range in the fasting study, but outside this range in the fed one.
The outside conventional bioequivalence range of CI shown in the fed study can be accepted in the case of lansoprazole particularly with its high intrasubject variability; for this implies a large range of blood drug levels, even when the same product is administered to the same subject on different ocassions.
The study under the fed conditions shows a delay in absorption for both reference and test products. The fed study also shows both the reference and test products have their bioavailability reduced, although the reference product is more affected. However, this reduction is not clinically relevant and is consistent with literature data.

We all know that guidelines are not carved out of stone, but I would not risk designing such a study without a scientific advisory meeting!

—
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