In Reply: Dr Lash raises the issue of the independent trial assumption in determining our study power. He also requests the presentation of outcomes limited only to the person-time that had not been included in the earlier reports, to support meta-analysis. Our patient cohort did include some cases previously reported (22% of the overall cohort). However, as noted in the article, their use was based on a longer clinical follow-up and expansion of the genotyping to include multiple alleles *3, *4, *10, and *41. We agree that these cases cannot be viewed as completely independent, yet, because of the extended follow-up and new data with newly assigned CYP2D6 alleles, any distinction between previous and current person-time is not feasible.
Schroth W, Schwab M, Brauch H. Association Between CYP2D6 Polymorphisms and Breast Cancer Outcomes—Reply. JAMA. 2010;303(6):516-517. doi:10.1001/jama.2010.94