Bioequivalence and Bioavailability Forum

Dear Pravin!

❝ [...] drug like Fluoxetine and its active metabolite Norfluoxetine half life is 90 hrs so in such cases what we do.

OK, this is really a quite nasty compund.
Just as an example have a look at this paper:

Pan RN, Chen TH, Huang CSH, and CH Hsiong
Pharmacokinetics and Bioequivalent Study of Generic Fluoxetine Capsules Preparation
Journal of Food and Drug Analysis 10(1), 13–7 (2002)

20mg were administered to 8 subjects b.i.d. for 13 days. According to the reported half life for the reference of 87.2±37.2 hours one might calculate an accumulation factor of 11.0±15.5, which is not reflected in the data (Figure 5); I would estimate the accumulation in the study as about 4 (200ng/mL instead of 50ng/mL).
Since for normal distributed data P(mu+sigma) = 0.15866, one may assume that at least one subject (0.15866×8) showed a half life of >87.2+37.2 or more than 124.4 hours, therefore steady state conditions were not obtained in all subjects…

BTW, Martindale (2002) reports a half life of 4-16 days for norfluoexetine, which also makes steady-state conditions in healthy subjects infeasible.
At least you may opt for a parallel design, OAD dosing.

Quoting

A Marzo
Bioequivalence Behind the Scenes
Pharmaceutical Development and Regulation 1(3), 179–89 (2003)

“A long elimination half-life […] calls for decision making about study design, namely crossover or parallel group design, the latter design being more adequate for drugs cleared with a half-life >10 days. Ethical problems concerning drug administration in healthy individuals (e.g. […] in a repeated-dose regimen) is another issue. Drugs that produce unacceptable side effects in a repeated-dose regimen should be studied in healthy volunteers only in single dose.”

In such a case IMHO it is imperative to go for a ‘Scientific Advisory Meeting’ with the regulators in the planning phase of the study.