New hurdles for translational research

Characterization of the molecular constituents of cancer and of the relationship of genotypic and phenotypic features to the clinical behaviour of the disease(s) are of enormous contemporary interest. This activity is fuelled by the exponential rate of discovery of new genes, an increasing understanding of the biological consequences of their expression, and a recognition that application of this knowledge in the clinic should provide the opportunity for individually optimized therapeutic approaches.Additionally, tumours may be explored as a primary research material in their own right, which is probably best epitomized by the recently announced Cancer Genome Project [1], in which the aim is to establish all of the genotypic defects of a selection of human carcinomas. The rapidly approaching completion of the Human Genome Project may be expected to extend the number of candidate genes for clinical evaluation by several orders of magnitude. Some of the new analytical tools required are already in place (although they need substantial refinement to keep pace) to handle this vast amount of new information in a miniaturized manner (eg DNA arrays and tissue arrays enable the simultaneous analysis of the expression of thousands of genes in a single sample, and of the expression of a small number of genes in several hundred tumours, respectively).Excitement about the possible advances resulting from molecular pathology has never been greater, and therefore the imposition of restrictions on this work that unnecessarily impede progress is of substantial concern.Over the past few years there has been a perceived need for official guidelines to ensure that access to and analysis of tissues from patients can occur within a framework that will not damage the individuals from whom the tissues were collected and will free the clinical scientist from concerns of litigation. It is therefore welcome that a number of bodies have considered this matter and that their views are availabl