feature Omic profiling for drug safety assessment: current trends and public–private partnerships William M. Gallagher, David Tweats and Jochen Koenig, jochen.koenig@genedata.com The drug development process is currently being hindered by non-optimal prediction of toxicity. Advances in molecular profiling approaches, such as transcriptomics, proteomics and metabolomics, offer the potential to provide a more comprehensive insight into toxicological effects than hitherto possible. These new technologies present their own challenges, however, particularly in relation to standardization and assessment. The focus of this article is on describing the current trends concerning the application of omic approaches in drug safety assessment, with specific emphasis on the role of public–private partnerships in advancing this emerging arena. Introduction Despite huge investment in target and lead compound discovery processes, the number of new drugs entering the market has stagnated. Of those compounds that enter into preclinical development, about 90% currently fail to reach the market [1]. Problems regarding lack of efficacy and concerns about safety are of paramount importance. Since the removal of a compound due to toxicity may take place at a late stage in drug development, failure costs can be very high. It has been estimated that it costs in the region of $1400 million, spent over 10 years on average, to bring a new drug to the market [2]. It is, therefore, crucial that compounds that ultimately will be unsuccessful fail at the earliest stage possible, in order to save long-term costs, free up develop- ment resources for more promising compounds and avoid adverse side effects in patients. The hurdles for a new compound to become a registered drug are already very high. As a result, new drugs can generally be regarded safe and effective in treating the condition for which they were developed. More and more drugs, however, are introduced to target either chronic diseases like diabetes and Alzheimer’s or widespread conditions like obesity. Long-term therapy and the sheer number of patients being treated both potentially increase the risk that rare but serious side effects, which are unlikely to be observed during drug development, may become appar- ent only after market introduction [3]. At the same time, Society has become less prepared to accept therapy-related risks. As a consequence, it is proving increasingly difficult to weigh the benefits of a drug designed to benefit the vast majority of patients against the possible risk of serious adverse effects for a small minority. The lack of knowledge about the exact (molecular) mechanism of action of many drugs and the mechanism underlying potential side effects only compounds this problem. It has become clear that discrete genetic alterations between individuals can result in significantly different efficacy and safety prop- erties of drugs [4]. Typically, this is due to dif- ferences in the affinity of a drug for off-target receptors or alternative metabolic modification of the drug or underlying, unrecognized enzyme deficiencies, resulting in increased susceptibility of individual patients. Effects like these may have been the cause of the low frequency of adverse events that led to the withdrawal of products like Merck’s Vioxx and Bayer’s Lipobay (Baycol in the US) [5,6]. Therefore, the challenge for pharma is to improve the prediction of rare safety risks and increase the understanding of on-target and off- target modes of action of drug candidates. Large-scale application of the new, molecular assays (i.e. ‘omics’ technologies) is seen as the big hope to drive this generation of additional knowledge. The focus of this article is on describing the current trends concerning the application of omic approaches in drug safety assessment, with Perspective  FEATURE Drug Discovery Today  Volume 14, Numbers 7/8  April 2009 PERSPECTIVE 1359-6446/06/$ - see front matter ß 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.drudis.2009.02.001 www.drugdiscoverytoday.com 337