The vast majority (>80%) of clinically non-functioning pituitary adenomas (NFPAs) are gonadotroph-cell adenomas, as demonstrated by immunocytochemistry. However, they are rarely associated with increased levels of dimeric LH or FSH. Increased levels of uncombined subunits (free alpha-subunit mainly, LH-beta subunit more rarely) are more frequently encountered, but are generally modest. The main problems raised by NFPA are mass effects problems, responsible for optic chiasm compression or deficient hormone secretion resulting from compression of normal anterior pituitary cells. The therapeutic management of NFPA may require combination of different options. The strategy of observation only for patients with incidentally discovered pituitary adenomas may be appropriate, provided that the tumor is well-delimited, small, has no extension with risk of neurological or visual chiasm compression, and that a meticulous hormonal work-up has ruled out the possibility of a minimal hormonal hypersecretion. Transsphenoidal surgery allows improvement in visual disturbances due to chiasmal syndrome in most patients, and sometimes, in pituitary function. After surgery alone, nearly 30% (between 10 and 69%, according to the series) of patients relapse within 5 to 10 yr. Radiotherapy is proposed either as a systematic adjunct or only if a significant remnant persists. Systematic radiation therapy is supported by the low relapse rate (mean, 11%; range, 6-21%) observed when radiation therapy is systematically associated with surgery. However, irradiation is almost always followed by hypopituitarism which might be associated with a reduction in life expectancy, despite appropriate replacement therapy. Results of medical treatment are disappointing. Dopamine agonist bromocriptine decreases gonadotropin and alpha-subunit in vitro and in vivo, but, in clinical studies, was poorly effective in reducing supranormal gonadotropins and free subunits levels, and rarely produced a minimal tumoral shrinkage. Cabergoline may be more efficacious. Somatostatin analogs are able to improve minimally visual problems in 20-40% of cases, but reduction in tumoral volume is anecdotic. Whether new somatostatin analogs (e.g. SOM230 which is a multiligand agonist) will improve these results is presently unknown. Administration of GnRH agonists is generally ineffective and may be hazardous. Prolonged administration of GnRH antagonist in a small number of patients with a secreting gonadotroph cell adenoma has been reported to reduce supranormal gonadotropins levels but not to produce any change in tumoral size.