Bioequivalence and Bioavailability Forum

wien-ui ☆ Oman, 2014-06-11 20:32 (4393 d 23:37 ago) Posting: # 13058 Views: 5,897	WHO Guidelines draft, April 2014 [Regulatives / Guidelines] Post reply
	Dear All, I would like to open the discussion for the new WHO BE Guidelines draft http://www.who.int/medicines/areas/quality_safety/quality_assurance/BE-guidelines-revised-QAS14-583_23052014.pdf. We still have about 20 days (till July 2, 2014) to give feedback comments. Kind regards, Osama

wien-ui
☆

Oman,
2014-06-14 11:44
(4391 d 08:25 ago)

(edited on 2014-06-14 12:49)
@ wien-ui
Posting: # 13065
Views: 4,518

WHO Guidelines draft, April 2014

Post reply

Dear All,

Maybe I should start to activate the discussion on this issue.
I'm not sure but I think this is the first time (with exception of the WHO workshop on assessment of Bioequivalence at Sep.2010) in which the WHO mentioned to the two-stage sequential design in it's BE guidelines (page 41, line 1225) but WHO described it generally and didn't mentioned to any certain method (potvin B or C), Montague (method D) and which one they recommend. does this mean they will accept all? hoping for comments.
Kind regards,
Osama

Helmut
★★★

Vienna, Austria,
2014-06-16 15:14
(4389 d 04:55 ago)

@ wien-ui
Posting: # 13066
Views: 4,352

Two-Stage Designs

Post reply

Hi Osama,

❝ I'm not sure but I think this is the first time […] in which the WHO mentioned to the two-stage sequential design in it's BE guidelines

Correct.

❝ […] WHO described it generally and didn't mentioned to any certain method (potvin B or C), Montague (method D) and which one they recommend.

Unfortunately WHO followed in the wording EMA’s approach. My interpretation of lines 1237–44

In order to employ a two-stage design, adjustments must be made to protect the overall Type I error rate and maintain it at 5%. In order to do this both the interim and final analyses must be conducted at adjusted levels of significance, with the confidence intervals calculated using the adjusted values.
It is recommended that the same alpha for both stages be employed which gives an alpha of 0.0294 for this case,

speaks for Method B (α 0.0294) because in the first stage of Method C α is only adjusted if power <80%. The text continues with

However, the amount of alpha to be spent at the time of the interim analysis can be set at the study designer’s discretion. For example, the first stage may be planned as an analysis where no alpha is spent in the interim analysis since the objective of the interim analysis is to obtain information on the point estimate difference and variability and where all the alpha is spent in the final analysis with the conventional 90% confidence interval. In this case no test against the acceptance criteria is made during the interim analysis and bioequivalence cannot be proven at that point.

I’m not aware of any publication exploring no (!) adjustment in the first stage and stopping for futility. O’Brien-Flemming with α (0.005, 0.048), expected GMR 0.95, target power 80%, CV 20%, stage 1 sample size 12, and a futility criterion of 0.85 ≤ PE ≤ 1/0.85 leads to a slight inflation of the Type I Error. Trying

library(Power2Stage) power.2stage.fC(alpha=c(0.005, 0.048), CV=0.2, n1=12, GMR=0.95, targetpower=0.8, fCrit="PE", fClower=0.85, theta0=1.25, nsims=1e6)

gives

TSD with 2x2 crossover Method B: alpha (s1/s2) = 0.005 0.048 Interim power monitoring step included Target power in power monitoring and sample size est. = 0.8 Power calculation via non-central t approx. CV1 and GMR = 0.95 in sample size est. used Futility criterion PE outside 0.85 ... 1.176471 BE acceptance range = 0.8 ... 1.25 CV = 0.2; n(stage 1) = 12; GMR= 0.95 1e+06 sims at theta0 = 1.25 (p(BE)='alpha'). p(BE) = 0.053862 p(BE) s1 = 0.021617 Studies in stage 2 = 18.31% Distribution of n(total) - mean (range) = 13.8 (12 ... 78) - percentiles 5% 50% 95% 12 12 24

❝ does this mean they will accept all?

WHO’s guideline are intended to support regulators of countries which don’t have their own (local) guidelines in place. I guess if one wants to use anything deviating from variants of Method B he/she has to show in simulations that the overall-α is maintained.

—
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

ElMaestro
★★★

Denmark,
2014-06-16 15:41
(4389 d 04:27 ago)

@ Helmut
Posting: # 13068
Views: 4,411

interim analysis, wording

Post reply

Hi all,

❝ Unfortunately WHO followed in the wording EMA’s approach.

According to my psychic tasseographer, WHO's assessors are those who write the WHO guideline and they are to a large extent the same as the guys who are writing the EMA's guideline.
I don't know how she does it. Reading coffee grounds sounds weird but she is often spot on.

❝ My interpretation of lines 1237–44

❝ In order to employ a two-stage design, adjustments must be made to protect the overall Type I error rate and maintain it at 5%. In order to do this both the interim and final analyses must be conducted at adjusted levels of significance, with the confidence intervals calculated using the adjusted values.

I think the quoted exerpt above in actuality means that an applicant must pick both alphas so as to have the type I error controlled at 5%. I agree that in the strictest sense one could read it in such a way that they require an adjusted alpha after stage one (i.e. not 0.05) but I do not think this is what they strictly mean and I do not think this is what they will endeavour to enforce. So I think there's still hope for Potvin C.
If you wish you can try Potvin C at alpha₁=0.049 or 0.000083 or whatever and find a good correspond level of alpha₂ for an overall alpha of 0.05.

—
Pass or fail!
ElMaestro

Helmut
★★★

Vienna, Austria,
2014-06-16 16:11
(4389 d 03:58 ago)

@ ElMaestro
Posting: # 13069
Views: 4,527

interim analysis, wording

Post reply

Hi ElMeaestro,

❝ According to my psychic tasseographer, WHO's assessors are those who write the WHO guideline and they are to a large extent the same as the guys who are writing the EMA's guideline.

Supported.

❝ ❝ In order to employ a two-stage design, adjustments must be made to protect the overall Type I error rate and maintain it at 5%. In order to do this both the interim and final analyses must be conducted at adjusted levels of significance, with the confidence intervals calculated using the adjusted values.

❝

❝ I think the quoted exerpt above in actuality means that an applicant must pick both alphas so as to have the type I error controlled at 5%.

Agree.

❝ I agree that in the strictest sense one could read it in such a way that they require an adjusted alpha after stage one (i.e. not 0.05) but I do not think this is what they strictly mean and I do not think this is what they will endeavour to enforce.

Do you remember the cases leading to deficiency letters by European regulators if studies were performed according to Method C? For a particularly nasty example see here. AUC passed in the first stage (Method C – according to the protocol; unadjusted α since power >80%) but failed in the post-hoc switch to Method B. If the sponsor would have known beforehand that Method C is not acceptable, they would have initiated a – rather small – second stage and likely would have passed with all glory. :-(

❝ So I think there's still hope for Potvin C.

❝ If you wish you can try Potvin C at alpha₁=0.049 or 0.000083 or whatever and find a good correspond level of alpha₂ for an overall alpha of 0.05.

Correct. Nevertheless not everybody has the guts to perform own simulations, though Power2Stage is such a nice tool.

BTW, below type I errors (blue=low, red=high) for Methods B/C and a conventional fixed sample design (abscissa n₁ 12–60, ordinate CV 10–100%).

[image]

Bonus question: Which one (B or C) mimics the fixed sample design better? :pirate:

—
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

wien-ui ☆ Oman, 2014-06-18 22:45 (4386 d 21:23 ago) @ wien-ui Posting: # 13088 Views: 4,091	WHO Guidelines draft, April 2014 Post reply
	Hi Helmut and ElMaestro, I can't find anything in this draft indicates removing of subjects add on Design, like the EMA (May 2012)? has the WHO decided to keep it also valid? I thought the reason of establishing the two-stage sequential designs is as they give better control of patient's risk than the add on design. Best regards, Osama

Helmut
★★★

Vienna, Austria,
2014-06-22 16:31
(4383 d 03:37 ago)

@ wien-ui
Posting: # 13117
Views: 3,847

No Add-on studies

Post reply

Hi Osama,

❝ I can't find anything in this draft indicates removing of subjects add on Design, like the EMA (May 2012)?

Are you referring to the Q&A-document?

❝ … has the WHO decided to keep it also valid?

I don’t think so (cannot find Add-ons anywhere in the draft). Annex 7 to WHO’s TRS № 937 (2006) mentioned Add-ons in Section 6.3.2. At the end of the section it was stated:

Add-on designs must be carried out strictly according to the study protocol and SOPs, and must be given appropriate statistical treatment.

(my emphasis)
In the past Australia’s TGA and more recently Health Canada accepted Add-ons – if a Bonferroni-correction (i.e., α 0.025 → 95% CI) was applied. In the meantime both agencies dropped Add-ons.

❝ I thought the reason of establishing the two-stage sequential designs is as they give better control of patient's risk than the add on design.

Exactly. See also this post and followings. Based on statistical theory I would say that a Bonferroni-correction should maintain the patient’s risk ≤0.05 in any TSD (and Add-ons as well). Note that Bonferroni’s adjusted α of 0.025 is smaller than any α of the published TSD-methods. Though this would lead to a conservative patient’s risk, the sample size penalty will be larger than necessary – which is not desirable both from ethical and economical perspectives.

—
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

WHO Guidelines draft, April 2014 [Regulatives / Guidelines]

WHO Guidelines draft, April 2014

Two-Stage Designs

interim analysis, wording

interim analysis, wording

WHO Guidelines draft, April 2014

No Add-on studies