The synthetic steroid 7alpha-methyl-19-nortestosterone (MENT) is a potent androgen that is resistant to 5alpha-reductase. It thus has decreased activity at the prostate and may have advantages over testosterone-based regimens in long term treatment or as part of a male contraceptive. Administration to eugonadal men results in suppression of gonadotropins, but its ability to support androgen-dependent behavior has not been investigated. For sustained release administration, MENT acetate was used, because its diffusion characteristics were more suitable for use in implants. However, upon release the acetate is rapidly hydrolyzed, and MENT is the biologically active moiety in circulation. We studied the effects of MENT on sexual interest and activity, spontaneous erection, and mood states in comparison with testosterone enanthate (TE) in 20 Caucasian and Chinese hypogonadal men recruited in Edinburgh and Hong Kong (n = 10 in each center). Outcomes were measured using a combination of daily diaries, semistructured interviews, and questionnaires. Nocturnal penile tumescence (NPT) was also recorded in the Edinburgh group. After withdrawal of androgen replacement treatment (wash-out phase) for a minimum of 6 weeks, subjects were randomized to two groups in a cross-over design. Drug treatment regimens were of 6-week duration and consisted of two implants, each containing 115 mg MENT acetate, inserted s.c. into the upper arm and removed after 6 weeks and two injections of TE (200 mg, i.m.) 3 weeks apart. MENT treatment resulted in stable plasma MENT concentrations of 1.4 +/- 0.1 nmol/L after 3 weeks and 1.3 +/- 0.1 nmol/L after 6 weeks (mean +/- SEM; all men). Nadir testosterone concentrations were 3.6 +/- 0.6 nmol/L at the end of the wash-out phase and 9.4 +/- 0.6 nmol/L 3 weeks after each injection. There were no differences in hormone concentrations between centers. There were no adverse toxicological effects. There were only minor differences between the two treatments. Both MENT and TE treatment resulted in significant increases in sexual interest and activity, spontaneous erection (both by self-report and NPT measurement), and increases in positive moods, with decreases in negative moods in the Edinburgh group. In the Hong Kong group, both treatments increased waking erection, with a trend toward increased sexual interest and activity. Mood states appeared to be less affected during the wash-out phase than in Edinburgh men and showed no significant response to either treatment. These results demonstrate that MENT has similar effects on sexual activity and mood states as testosterone in hypogonadal men. As NPT is a physiological androgen-dependant outcome, these data provide further evidence for the androgenicity of MENT. The lack of detected effect of either androgen in Hong Kong men other than on waking erection illustrates the importance of the cultural context of symptomatology and its measurement. The appropriate dose of MENT remains to be determined, but these results support its development as a potential androgen replacement therapy.