Drugs in Search of Diseases

The traditional approach to drug discovery and development is to start with a specific disease, identify a target based on knowledge of the biology, and screen a compound library for a chemical which will interact appropriately with the target, resulting in the expression of a desired effect. The chemical is then tested for efficacy in animal models of disease, for toxicology preclinically, and ultimately for toleration and pharmacokinetics in humans.

At least two thirds of compounds die at this stage before we can establish if our target selection was correct in treating the human disease. The attrition in this forward testing approach of unprecedented mechanisms can be as high as 49 in 50 compounds tested. In this article, we propose a reverse approach to drug discovery and development once chemical tractability and pharmacological tolerability of a particular mechanism have been established.

PHARMACOLOGICALLY TOLERANT GENE

Recent publications have analysed how the intersection of the proportion of the human genome that is tractable to chemical intervention, and the number of drug targets that are relevant to human disease may be as few as 600-1,500 (1). The final subset not considered in this analysis is those where pharmacological intervention would be tolerated – the pharmacologically tolerant genome. This is a difficult parameter to estimate, as our willingness to tolerate adverse effects of drugs varies considerably depending on the disease being treated and the seriousness of the adverse event. Even if we are optimistic, our ability to tolerate pharmacological intervention of these targets may reduce the total number of tractable targets by up to a half.

This ever diminishing number highlights the need to ensure that all potential diseases which could be treated by a new mechanism of action are identified once the safety of this mechanism has been established. This tactic of identifying new indications for proven drugs has been used with great success in the past, meeting significant areas of medical need and adding billions of dollars to the commercial value of products.

A survey conducted in 1998 demonstrated that 40 per cent of blockbuster drug sales were from alternative indications (2). Good examples include gabapentin, originally developed as an anti-epileptic medication, with subsequently proven efficacy in neuropathic pain, and sildenafil, developed for erectile dysfunction and subsequently demonstrated to offer a major advance in the treatment of primary pulmonary hypertension. Many of these alternative indications result from data generated in small-scale investigator-initiated studies postapproval, or from serendipitous case reports of unexpected benefits. The challenge is how to systematically identify and evaluate these opportunities earlier in the drug development and approval process.