1. ¹BioStat Solutions Inc., Mount Airy, MD, USA
2. ²Eli Lilly and Company, Indianapolis, IN, USA
3. ³Johnson and Johnson Pharmaceutical Research and Development, Raritan, NJ, USA
4. ⁴Schering-Plough, Kenilworth, NJ, USA
5. ⁵Novartis Institute of Biomedical Research, Cambridge, MA, USA
6. ⁶Orion Corporation Orion Pharma, Espoo, Finland
7. ⁷Teva Pharmaceutical Industries, Netanya, Israel

Correspondence: Dr CM Bromley, BioStat Solutions Inc., 5 Ridgeside Court No. 202, Mount Airy, MD 21771, USA. E-mail: enab@BioStatSolutions.com

Received 12 February 2008; Revised 17 June 2008; Accepted 8 August 2008; Published online 16 September 2008.

Abstract

Pharmacogenetic association studies have the potential to identify variations in DNA sequence which impact drug response. Identifying these DNA variants can help to explain interindividual variability in drug response; this is the first step in personalizing dosing and treatment regimes to a patient's needs. There are many intricacies in the design and analysis of pharmacogenetic association studies, including having adequate power, selecting proper endpoints, detecting and correcting the effects of population stratification, modeling genetic and nongenetic covariates accurately, and validating the results. At this point there are no formal guidelines on the design and analysis of pharmacogenetic studies. The Industry Pharmacogenomics Working Group has initiated discussions regarding potential guidelines for pharmacogenetic study design and analyses (http://i-pwg.org) and the results from these discussions are presented in this paper.