Christine Galustian and Angus G. Dalgleish

Department of Oncology, Division of Cellular and Molecular Medicine, St. Georges University of London, Cranmer Terrace, London SW17 0RE.

Abstract

The discovery of effective cancer treatments is a key goal for pharmaceutical companies. However, the current costs of bringing a cancer drug to the market in the USA is now estimated at $1 billion per FDA approved drug, with many months of research at the bench and costly clinical trials. A growing number of papers highlight the use of data mining tools to determine associations between drugs, genes or protein targets, and possible mechanism of actions or therapeutic efficacy which could be harnessed to provide information that can refine or direct new clinical cancer studies and lower costs. This report reviews the paper by R.J. Epstein, which illustrates the potential of text mining using Boolean parameters in cancer drug discovery, and other studies which use alternative data mining approaches to aid cancer research.