Initiation, progression, and completion of the cell cycle are regulated by various cyclin-dependent kinases (CDKs), which are thus critical for cell growth. Tumour development is closely associated with genetic alteration and deregulation of CDKs and their regulators, suggesting that inhibitors of CDKs may be useful anti-cancer therapeutics. Indeed, early results suggest that transformed and normal cells differ in their requirement for e.g. cyclin/CDK2 and that it may be possible to develop novel antineoplastic agents devoid of the general host toxicity observed with conventional cystostatic drugs. Numerous active-site inhibitors of CDKs have been studied; the main limitation with these ATP antagonists is kinase specificity for CDKs. However, screening of compound collections, as well as rational design based on enzyme-ligand complex crystal structures, are now yielding pre-clinical candidates, particularly certain purine and flavonoid analogues, with impressive potency and selectivity. Natural CDK inhibitors (CKIs), e.g. the tumour suppressor gene products p16(INK4), p21(WAF1), and p27(KIP1), form the starting point for the design of mechanism-based CDK inhibitors. A number of these small proteins have been dissected and inhibitory lead peptides amenable to peptidomimetic development have been identified. Conversion of these peptides into pharmaceutically useful molecules is greatly aided by the recent elucidation of CKI/CDK crystal and solution structures. Additional interaction sites on CDKs being exploited for the purposes of inhibitor design include: phosphorylation/dephosphorylation sites, macromolecular substrate binding site, CKS regulatory subunit binding sites, cyclin-binding site, cellular localisation domain, and destruction box. Finally, progress has recently been made in the application of antisense technology in order to target CDK activity.