Ten years after the appearance of the first tricyclic antidepressant, a survey showed that only about two-thirds of depressed inpatients benefit from somatic treatment. The second-generation drugs and the selective reuptake inhibitors do not appear clinically more efficacious than the traditional tricyclic antidepressants. Along with the development of new drugs with alternative profiles of action, there have been many attempts to identify biologic tests and markers of affective disorders which could be useful in clarifying diagnosis and planning its treatment. This paper has reviewed studies of 11 biochemical predictor's - that is, dexa-methasone suppression test, thyrotropin-releasing hormone test, intravenous L-tryptophan test, 3-methoxy-4-hydroxyphenyl glycol, 5-hydroxyindoleacetic acid, precursor amino acids, amino acid transport into erythrocytes, platelet serotonin, platelet imipramine binding, platelet monoamine oxidase, and tyramine sulfate excretion - of antidepressant response and their application for clinical practice. Although some biochemical markers may be associated with antidepressant response in patient samples a reliable prediction of antidepressant response in individual patients is not available at present. On the basis of the state of knowledge of the biology of depression and the availability of treatments with various profiles of action, it is unsatisfactory that depression is not better treated. Improved and standardized methods for clinical description and diagnosis and for creating meaningful biochemical measures are needed.