The SOSA (selective optimization of side activities) approach represents a validated alternative to HTS. It consists of testing "old" drugs on new pharmacological targets, subjecting a limited set of structurally and therapeutically diverse drug molecules with known safety and bioavailability in humans to pharmacological screening to shorten hit identification time and cost. The approach proceeds in two steps: (1) Screening a smart library of about 1000 compounds, with hits being "druglike" due to existing bioavailability and toxicity data; (2) Optimizing hits via chemistry to increase affinity for the new target and decrease affinity for other targets, transforming side activity into the main effect. The concept is based on drugs acting on multiple receptors and a limited chemical universe safely administered to humans, covered by available drugs. Rationale includes most drugs interacting with multiple targets (e.g., diazepam inhibiting phosphodiesterases, gabazine being a monoamine-oxidase type A inhibitor, spiperone having affinity for 5-HT2a receptors, clozapine binding to 14 receptors), justifying testing old drugs on new targets. Successful examples of the SOSA approach are reviewed, including sulfonamides as leads to endothelin ETA receptor antagonists and myocardiac sodium/hydrogen exchange (NHE) inhibitors, dihydropyridines as leads to R1A-adrenergic antagonists and multidrug-resistance modulators, cyclic analogues of β-blockers as leads to potassium channel blockers, aminopyridazines as leads to muscarinic M1 receptor ligands, reversible acetylcholinesterase inhibitors, and CRF antagonists, neuroleptic benzamides as leads to D3 and D4 selective ligands, and thalidomide, diclofenac, and captopril as leads to compounds with different activity profiles.