Log in
Register|
Dr Andrew Leary ★ Ireland, 2009-08-12 14:33 (5749 d 15:23 ago) Posting: # 4037 Views: 6,434 |
|
|
Dear Helmut et al In an upcoming BE study, we are planning to use the two-stage design as recommended in the draft guideline. Our rationale is as follows: (a) variability of the drug substance is unknown & not published; (b) for formulation reasons there is a significant risk of bio-inequivalence (making the option to stop after the first group an attractive one). To our surprise, in assessing the protocol the local regulators (Irish) have questioned our approach on the grounds that they are not sure whether the two-stage approach will ultimately prove acceptable in the future. Informal advice from the MHRA is similar. [Effectively they are saying we use this approach at our own risk. So, nothing new there!] It is clear that the two-stage design is acceptable to some regulators and not to others; I presume that those who accept this approach got their way when the new guideline was drafted. That said, it looks as though this will remain in the final guideline. My questions are: Has anyone heard otherwise (will two-stage stay or go)? Is anyone else using this approach yet? Kind regards Andrew Edit: Category changed. [Helmut] |
|
Helmut ★★★   Vienna, Austria, 2009-08-12 16:32 (5749 d 13:24 ago) @ Dr Andrew Leary Posting: # 4038 Views: 5,499 |
|
|
Dear Andrew! ❝ Our rationale is as follows: (a) variability of the drug substance is unknown & not published; OK. ❝ (b) for formulation reasons there is a significant risk of bio-inequivalence (making the option to stop after the first group an attractive one). I don't know which method you are planning to use. Potvin et al. (2008) was not validated for a stop-criterion based on the actual (observed) point estimate (PE) in stage 1. You estimate the power based on CVintra and the expected PE applied in study planning. If power ≥80%, you evaluate the study with α 0.05 (pass|fail). If power <80% you evaluate the study with α 0.0294 (pass|fail). If BE fails, calculate sample size for stage 2 based on the expected (!) PE used in study planning. That's the tricky part: if you already expected a high chance of failure (let's say a PE of only 85%) and planned the study for a PE of 95%, there's no (statistical) way out to to stop the study, even when you see an actual PE in stage 1 of only 82%. That's clearly a drawback of the method, and I would not recommend it in any case when you expect a reasonable high chance of bioinequivalence. In such a case I would rather perform a classical pilot study to get an idea of the PE. Quoting the paper:
Another possible application of sample size reestimation is to adjust the sample size based on the observed effect size seen in an initial sample. For example, one might choose a sample size based on the assumption that the ratio of geometric means is between 0.95 and 1.053, but the estimated ratio seen in an initial sample may be, say, 0.92. One might be tempted, in this case, to recalculate the sample size assuming that the true ratio of population means was 0.92. However, simulations carried out by Cui et al. indicate that if this procedure is applied naïvely, the overall type I error rate may be inflated by 30% or more. More complicated statistical procedures have been proposed to allow sample size reestimation based on the observed effect size. […] While these more complicated procedures could certainly be considered for a BE study, they lack the feature of using the usual test statistic formulae (possibly with a simple adjustment of the nominal type I error rate), and they have not been validated for use with two one-sided tests and unknown variance. We will not consider them further here. For months I want to ask Walter Hauck whether the group is working on such a method, but it still stays on my to-do-list… ❝ My questions are: ❝ Has anyone heard otherwise (will two-stage stay or go)? My last informal information date back to May 2009: will be kept. ❝ Is anyone else using this approach yet? Yes, me. See the footnote in this post. I expect to start the interim analysis next week. Edit 2009-08-18: Power was >80%, the study was stopped after the first stage and demonstrated BE. So no experience with a second stage yet. — Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
Dr Andrew Leary ★ Ireland, 2009-08-13 15:26 (5748 d 14:30 ago) @ Helmut Posting: # 4040 Views: 5,365 |
|
|
Helmut Thank you very much for such a comprehensive reply. You will recall that my statistical knowledge amounts to less than a pimple on the backside of most users of this forum, so please excuse me if what I have written below seems garbled. Firstly, it seems from your response that the method of Potvin, although having the drawback you mention, has the advantage of allowing you to calculate the size of the second group by using the data from the first group. This is logical. I had assumed that the regulators would be illogical and that, for a two-stage design, they would expect us to pre-specify the size of both first and second groups. Thus, we have planned a group sequential design with a first group of 24 to be followed by a second group of 24 if the interim analysis indicates the possibility of passing with 48. The idea would be to adjust the confidence intervals according the method of Gould 1995 (ie. as we plan two groups of equal size, we would calculate 93.9% confidence intervals at interim analysis). I'll confess that I do not fully (or even slightly) understand the method then used to decide on stop or proceed; I was relying on my statistican to work this out from the Gould paper. In your view, is Gould's approach unlikely to be acceptable to the regulators? Kind regards Andrew |
|
Helmut ★★★ Vienna, Austria, 2009-08-13 15:53 (5748 d 14:03 ago) @ Dr Andrew Leary Posting: # 4042 Views: 5,565 |
|
|
Dear Andrew! ❝ Thank you very much for such a comprehensive reply. Oh, I'm always at little bit epic. ❝ You will recall that my statistical knowledge amounts to less than a pimple on the backside of most users of this forum […]
❝ […] I had assumed that the regulators would be illogical and that, or a two-stage design, they would expect us to pre-specify the size of ]both[/i] first and second groups. In the drafted quideline they didn't mention a specific method. Health Canada is planning to implement Potvin's method in their guideline (see here; but nothing published yet). ❝ […] The idea would be to adjust the confidence intervals according the method of Gould 1995 […] I like Gould's paper and tried it a couple of times (three times in Germany, once in France). Regulators did not accept the proposals. I gave up in 2002. ❝ In your view, is Gould's approach unlikely to be acceptable to the regulators? No. Not only from said experiences, but also two members of the PK-drafting group telling me "Gould's method is ridiculous" without giving further explanations. For interested readers: LA Gould — 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 |
|
Dr Andrew Leary ★ Ireland, 2009-08-13 18:56 (5748 d 11:00 ago) @ Helmut Posting: # 4043 Views: 5,327 |
|
|
Once again, many thanks Helmut! |
Ing. Helmut Schütz