I-08

Clinical Trial Design Optimization: A Pharmacokinetic Sampling Plan for a Phase 3 Clinical Trial in Mild Alzheimer Disease Patients

Earvin Liang (1), Farkad Ezzet (2), Jonathan Wagg (2), Margaret Cooney (1), Susan Abushakra (1)

(1) Elan Pharmaceuticals, Inc. South San Francisco, CA, (2) Pharsight Corporation, St. Louis, MO

Introduction: One aspect of design optimization concerns finding an efficient pharmacokinetic (PK) sampling schedule to properly characterize PK of a given drug of interest in the target population.  ELND005 (Scyllo-inositol) is being investigated as a potentially disease modifying oral treatment for Alzheimer's Disease (AD). 

Objective: To identify an optimal PK sampling plan for a Phase 3 clinical trial in Mild AD patients receiving oral ELND005 250 mg or placebo BID over 78 weeks. 

Methods: The method underling optimization approach involves the computation of "Fisher information matrix", whose determinant ("criterion") represents the degree of information inherent in the design using WinPOPT (http://www.winpopt.com/) sampling optimization based on a previously developed PK model of ELND005.  PK samples are to be collected from 450 patients per arm receiving 250 mg ELND005 or placebo BID.  A base PK sample collection study design occurs at baseline visit and weeks 6, 12, 24, 48 and 78, of which 1 sample collected at pre-dose and 2 post-dose collected with at least 1 hour apart from each other during baseline visit, and 2 blood samples drawn at approximately 5 to 9 hours post-dose with at least 1 hour apart from each other during follow-up clinical visits.

Results and Discussion: The base study design is likely to yield a criterion of 86 with %CV for clearance at ~51%.  When the PK sampling schedule is modified to 2 post dose samples with one soon after dosing (0.5 hours) and the other at ~2 (Design A), 3 (Design B), or 4 (Design C) hours post-dose during baseline visit, and to 1 post dose sample during the rest of the visits, the criterion and %CV of clearance are likely to improve to 85 and ~34%, 94 and ~20%, and 111 and ~14%, respectively.  Either Design is invariant in terms of efficiency to changes in sampling times anywhere between 5-9 hours post morning dose and there is no meaningful gain in efficiency if two samples instead of one are collected during PK visits at weeks 6 to 78.  For Designs B and C, if sample size is maintained at 300 and 200 patients while other conditions remain the same, %CV of clearance is likely to increase slightly to ~24% and ~20%, respectively.

Conclusion: Taking into consideration patient compliance, comfort and safety, as well as data richness and consistency, Design B is considered an efficient and practical sampling plan and therefore proposed for the future Phase 3 trials.

Reference: PAGE 21 () Abstr 2309 [www.page-meeting.org/?abstract=2309]

Poster: Study Design