TUESDAY, Jan. 23 (HealthDay News) -- As many as half of microarray-based studies in oncology may contain flawed analysis or conclusions about associations between gene expression profiles and clinical outcome, according to a report published in the Jan. 17 issue of the Journal of the National Cancer Institute.
Alain Dupuy, M.D., and Richard M. Simon, D.Sc., of the National Cancer Institute in Bethesda, Md., reviewed 90 studies published through the end of 2004 that fell into three general categories of statistical analysis: outcome-related gene finding, class discovery, and supervised prediction. The researchers focused their attention on 42 studies published in 2004.
The researchers found that 21 (50 percent) of the 2004 studies contained at least one flaw, such as unclear or inadequate control for false-positive findings; a claim of correlation between expression clusters and clinical outcome that was spurious because there was inadequate analyses to support the conclusion; and an incorrect cross-validation procedure resulting in a biased estimation of the prediction accuracy for binary outcomes.
"Microarray studies are a fast-growing area for both basic and clinical research with an exponentially growing number of publications," the authors write. "As demonstrated by our results, common mistakes and misunderstandings are pervasive in studies published in good-quality, peer-reviewed journals." The authors provide guidelines in the form of a list of "Do's and Don'ts" for researchers.
Abstract
Full Text (subscription or payment may be required)
