To the Editor: The study of breast cancer gene expression signatures by Dr Acharya and colleagues1 provides the type of information needed to further the goals of personalized molecular medicine. However, the validation of such findings by other researchers is necessary before this type of information can be clinically useful. One of the major problems in validating such studies is the lack of a detailed description of how the biospecimens, which are the foundation of these molecular studies, were collected, transported, preserved, processed, and stored for the actual testing.2 This information was not provided in the article. An investigation of the references and Gene Expression Omnibus (GEO) records3 for the 5 data sets used in the study indicated that 2 did not supply any information about the tumor tissues from which the RNA was isolated for study, 2 gave basic information that the tissue was frozen, and 1 gave detailed information as to freezing and storage conditions.
Robb JA, Moore HM, Compton CC. Documenting Biospecimen Conditions in Reports of Studies. JAMA. 2008;300(6):650-651. doi:10.1001/jama.300.6.650-c