Levels of steroid hormones such as androgens and cortisol exhibit circadian variation, and their fluctuations are related to the sleep-wake cycle. Currently, the functional role of different stages of sleep in steroid hormone secretion remains unclear. The present study aims to explore the effect of slow-wave sleep (SWS) suppression on morning levels of cortisol and androgens. Twelve healthy male volunteers participated in two experimental sessions: a session with selective SWS suppression during night sleep and a session with regular night sleep (control). SWS suppression was achieved by stimulation using an acoustic tone. Salivary samples were collected in the morning immediately after awakening and again 40 min later. The samples were analysed by liquid chromatography-tandem mass spectrometry for testosterone, androstenedione (Ad), dehydroepiandrosterone (DHEA), 17α-hydroxyprogesterone (17-OHP), and cortisol. SWS suppression reduced overall SWS duration by 54.2% without significant changes in total sleep time and sleep efficiency. In the session with selective SWS suppression, the average level of morning testosterone was lower than in the control session (p = 0.017). Likewise, 17-OHP was lower in the SWS suppression condition (p = 0.011) whereas the ratio of DHEA/Ad was higher (p = 0.025). There were no significant differences between sessions in cortisol, Ad, or DHEA concentrations. The effect of selective SWS suppression on morning levels of testosterone and 17-OHP points to the importance of SWS for the synthesis and secretion of androgens. These results suggest that chronic sleep problems, which lead to reduced SWS, increase the risk for the development of androgen deficiency in the long term.