Over the past few months, several authors have written in the pages of this journal and elsewhere about the critical role that genomics must play in the development of new pharmaceutical agents. But as Nature Biotechnology correctly stated in its editorial" The genome rush of 1996,"'many significant problems must be solved before genomic efforts revolutionize drug discovery. Perhaps the most worrisome of these problems is that genomics will provide a great many potential targets, most of which are not targets at all. Indeed, genomics initially increases the probability, not of finding new drugs, but of finding new blind alleys; that is, potential targets whose inactivation or activation has no impact on disease progression.

Jurgen Drews2 has calculated the number of new targets that will emerge from genomics: 3,000 to 10,000 interesting targets for the 100 or so diseases whose treatment would alter health care substantially. Even at only 3,000 to 10, 0000 targets of interest, drug discovery will be slow; as Drews notes, only 417 identified molecular targets are affected by drugs now available through the enormous efforts of the pharmaceutical industry. One must note, sadly, that the US Food and Drug Administration (FDA, Rockville, MD) scorecard for new chemical entities (NCEs) approved for any indication averages about thirty each year, though 68 were approved in 1996; some paradigm shift is needed to make genoroics a driver for drug discovery.