Résumé : The hypothesis that growth hormone (GH) can affect immune responses in man has been evaluated by monitoring cytokine expression in cultures from peripheral blood mononuclear cells, by enzyme-linked immunosorbent assay (ELISA) and ribonuclease protection assay, and in tonsillar cells by ELISA. In addition to pituitary GH (GH-N), the placental form (GH-V), differing from pituitary GH by 13 amino acids has also been tested. Only few effects reached statistical significance and were in no case greater than 15%. Pituitary GH slightly reduced IL-5 production and stimulated IFN-γ production. The latter effect was also observed with prolactin and could thus be induced through the prolactin receptor. It is proposed that GH has no strong effects on the parameters investigated, possibly as a result of redundancy in the cytokine network. Alternatively, effects on leukocytes are mediated by other tissues such as the liver or are clear only in response to stronger challenges.