Preface to Genomic Pathology - a New Frontier

Abstract

We are writing this preface in the San Francisco Bay area, surrounded by one of the largest critical masses of the biotech industry in the world – a biological/pathological Silicon Valley. The companies here demonstrate the cutting edge of this new biology and what can be achieved by the integration of a few key technologies. The rapid rate of growth in this area is exciting but embracing and exploiting it poses enormous challenges to the academic research community and pathologists in particular. It is probably not an exaggeration to say that in 2001 we are at a new frontier. We may be about to witness the change from the ‘old biology’ of morphology and the dissection of biological and pathological pathways by the use of information on a small number of key features, be they morphological, immunocytochemical or molecular, to the ‘new biology’, where the interactions of thousands of pieces of data can be simultaneously assessed, be they gene polymorphisms, gene expression or protein expression analysed by new proteomic techniques. What has made this possible? The convergence of the first drafts of the human genome1, 2, robotic microarrays, proteomic technologies, and the rapid development of computing and bio-informatics have provided unparalleled opportunities for the interrogation of human biology and pathology. The biggest surprise of the Human Genome Project was that we have a relatively ‘simple’ genetic code with only 30 000–40 000 genes1, 2. At first sight, this may be viewed as a good thing, but it highlights the fundamental importance of gene–gene interactions, the regulation of gene expression, and protein–protein interactions. The development of microarray technologies has unleashed the potential not only to interrogate large numbers of genes, but also to place a representation of the entire human genome on a single slide. Arrays with 40 000 sequences have already been printed and soon the whole genome will be assessed in a single experiment. This has already been done for yeast and now it is our turn. It has never been easier to identify targets and the problem now shifts from data generation to data-mining and deriving information from such material. The data generated from these experiments will start to challenge historical pathological concepts and classifications. Large numbers of ‘superior’ gene expression markers, classifications, and prognostic markers will be hypothesized but they need to be validated and shown to be truly superior to classical pathological methods or rejected. Unfortunately, this can take many years unless the hypotheses are testable in archival paraffin-embedded tissue. Thus, the challenge moves from the test tube and array to the tissue and the pathologist to verify such findings. The challenge that the new biology offers to the pathologist requires advances in pathological techniques and these are now being met by tissue arrays, quantitative in situ techniques, and the development ofquantitative immunocytochemical methods. These technologies will simplify large-scale studies and answer the question as to whether a microarray can provide as much valuable information as a haematoxylin and eosin section. Bio-informatics is in its infancy in its application to pathology. This area is crying out for simple programs that can be used to interrogate meaningfully the large amounts of information that the new biology generates. This need arises not only in identifying complex patterns of gene expression, but also in areas such as simple programs for handling and analysing tissue arrays. What role will pathologists take on in the era of the new biology? We have choices. We can choose to remain with our morphological roots, or we can extend our expertise to use the new tools to our and our patients' benefit. If we wish to adopt the latter, we need to do it quickly as the momentum in this area is building quickly. In this Annual Review Issue we have tried to demonstrate some of the key technologies,