[Somatic and biological factors predicting a response to antidepressive agents].

In order to assess the predictive value of somatic and biological factors in antidepressant trials, non specific parameters, i.e. natural course of illness, life events, placebo effect ... have to be controlled by means of studies vs placebo. Among somatic factors, retardation seems to predict a positive response to antidepressants. The predictive value of other endogenous signs--like insomnia or weight loss--is still questioned. Few biochemical parameters appear relevant when metabolites of central monoamines, their precursors and the enzymatic processes involved are considered. The serotoninergic system is the focus of many studies. Among the neuroendocrine indices, the DST proved too poorly specific of depression. Among the physiological parameters, some characteristics of sleep EEG, like a shortening of REM latency, seem promising. Pharmacological challenges, for instance response to stimulant drugs, gave inconsistent results and should be discussed on ethical grounds. Many studies have been undertaken but presently no routine reliable biological index is available to predict a response to antidepressants.