The value of in vitro data in metabolic disease: Editorial Board member Keith Suckling comments on a review by Clapham et al.

It is not unusual to read at the end of a paper presenting new biology that the novel mechanism or process being described may be a target for pharmacological intervention. The data may well have been obtained from studies in cell culture, perhaps of an immortalised human cell line, perhaps from cells genetically manipulated, and the basic picture (given the rigours of the reviewing process) may be very convincing. But there is always a strong doubt in the reader’s mind. This is because the pathway from a novel and potentially relevant piece of biology to a new drug with demonstrated clinical value is long and hazardous. At this early stage, the relevance to human disease may be little more than a hypothesis, and data of all kinds will be required as drug discovery and development takes place to validate the target.

It has been recognised for many years that one of the major problems to address is how to develop sufficient confidence in a target or agent to allow the major financial commitment that is required to be made. We need confidence that the cell culture models are relevant to an intact animal, and that the data we obtain in a pre-clinical species has relevance to the situation in the patient. Such confidence is hard to obtain, particularly for novel biology. In this review [1], Clapham and colleagues lay out the ground for metabolic disease. Apart from offering a very clear analysis of the issues involved, they make a very useful contribution by illustrating the value of data on drug activity obtained from cell culture studies and from parallel studies in laboratory animals, both in relation to the observed effects in patients. This kind of information is not readily available: it is mostly held in the experience of specialised scientists, an asset that is all too easily lost in the major reorganisation of drug companies.

A central point in the problems clearly illustrated by Clapham et al. is the proper understanding of what a cell culture model or an in vivo model can be expected to achieve. The question has been addressed for animal models in the closely related area of dyslipidaemia [1], but, probably because the relevant data are very widely scattered in the literature, the aligning of in vitro, in vivo and human data has not been widely done. The analysis of Clapham and colleagues is a welcome example of what could be done. But we need to go further. As I have noted recently [3,4], I believe we need to be much more critical of what our models can achieve and what their limitations are. Only then, as Clapham et al. clearly illustrate, will we be able to offer a rigorous scientific foundation for the difficult investment decisions that have to be made.


1. Clapham J.C., Hallén S. (2009), Cell based in vitro and ex vivo models in metabolic disease drug discovery: nice to have or critical path?, Expert Opinion on Drug Discovery, 4: 417-428.

2. Suckling K.E. (2006), The ideal model in atherosclerosis and dyslipidaemia - does it exist? Expert Opinion on Drug Discovery 1: 507-511.

3. Suckling KE (2007), Top down meets bottom-up. The critical interface for drug discovery. International Atherosclerosis Society Commentary.

4. Suckling KE (2008) Animal research: too much faith in models clouds judgement. Nature (letter) 2008, 455, 460.

By Keith Suckling, PhD, Editorial Board member of Expert Opinion on Therapeutic Patents