Bioequivalence and Bioavailability Forum

Dear Niraj,

I agree with Kumar. Note that his reference deals with superiority testing, parallel groups, normal dis­tributed data, and common – known! – variance. Not what we have in BE. According to guidelines only two stages are acceptable. For more stages see Gould.¹ I tried Gould’s two-stage method in the late 1990s and neither Germany’s BfArM nor France’s AFSSAPS accepted the protocols… Hence, I gave up.

❝ […] group sequential design and adaptive design. Is there any difference?

See the review by Schwartz & Denne.² Adaptive designs are a subset of group sequential designs, where not only the sample size is re-estimated based on the observed variance, but based on the effect size (PE in the BE setting) as well.

❝ I guess that in adaptive design alpha level can not be the same (Method C, Potvin literature).

Not necessarily. The method of Fuglsang³ is based on Methods B/C, whereas methods of Karalis & Macheras⁴ and Karalis⁵ are based on Methods C/D. Fuglsang adapts α₂ based on the outcome of the first stage (requiring simulations in the interim analysis), whereas fixed values are used by Karalis & Macheras (α₁ 0.05 or 0.0294, α₂ 0.0294) and Karalis (α₁ 0.05 or 0.0280, α₂ 0.0280).
Personally I would be cautious applying adaptive methods in BE. Unlike in phase II, where due to the large sample sizes the PE generally is sufficiently precise, in BE this might not be the case. Although these methods control the error type I, power might be low – especially for combinations of small sample sizes in stage 1 and large CVs.⁶

---

References:
1. Gould LA. Group Sequential Extension of a Standard Bioequivalence Testing Procedure. J Pharmacokinet Biopharm. 1995;23(1):57–86. doi 10.1007/BF02353786.
   
2. Schwartz TA, JS Denne JS. Common threads between sample size recalculation and group sequential procedures. Pharmaceut Statist. 2013;2:263–71. doi 10.1002/pst.68.
   
3. Fuglsang A. Controlling type I errors for two-stage bioequivalence study designs.
   Clin Res Reg Aff. 2011;28(4):100–5. doi 10.3109/10601333.2011.631547.
   
4. Karalis V, Macheras P. An Insight into the Properties of a Two-Stage Design in Bioequivalence Studies. Pharm Res. 2013;30(7):1824–35. doi 10.1007/s11095-013-1026-3.
   
5. Karalis V. The role of the upper sample size limit in two-stage bioequivalence designs. Int J Pharmaceut. 2013:456(1):87–94. doi 10.1016/j.ijpharm.2013.08.013
   
6. Fuglsang A. Futility Rules in Bioequivalence Trials with Sequential Designs. AAPS J (pre-print November 2013). doi 10.1208/s12248-013-9540-0.