OBJECTIVE To obtain information about preorchiectomy gonadal function in patients with testicular germ cell cancer to improve the clinical management of fertility and other andrologic aspects in these men. METHODS In group 1, a group of 83 consecutive patients with testicular germ cell cancer (TGCC) investigated before orchiectomy, semen analysis was carried out in 63 patients and hormonal investigations, including measurement of follicle-stimulating hormone, luteinizing hormone (LH), testosterone, estradiol, sex hormone-binding globulin (SHBG), inhibin B, and human chorionic gonadotropin (hCG), in 71 patients. Hormone levels in patients with elevated hCG (n = 41) were analyzed separately. To discriminate between general cancer effects and specific effects associated with TGCC, the same analyses were carried out in a group of 45 consecutive male patients with malignant lymphoma (group 2). Group 3 comprised 141 men employed in a Danish company who served as controls in the comparison of semen parameters. As a control group in hormone investigations, 193 men were selected randomly from the Danish National Personal Register to make up group 4. RESULTS We found significantly lower sperm concentration (median, 15 x 10(6)/mL; range, 0 to 128 x 10(6)/mL) and total sperm count (median, 29 x 10(6)/mL; range, 0 to 589 x 10(6)) in patients with testicular cancer than in patients with malignant lymphomas (sperm concentration: median, 48 x 10(6)/mL; range, 0.04 to 250 x 10(6)/mL; sperm count: median, 146 x 10(6); range, 0.05 to 418 x 10(6)) (P < .001 and P < .001) and healthy men (sperm concentration: median, 48 x 10(6)/mL; range, 0 to 402 x 10(6)/mL; sperm count: median, 162 x 10(6); range, 0 to 1253 x 10(6)) (P < .001 and P < .001). FSH levels were increased in men with testicular cancer (median, 5.7 IU/L; range, 2.0 to 27 IU/L) compared with both men with malignant lymphomas (median, 3.3 IU/L; range, 1.01 to 12.0 IU/L) and healthy controls (median, 4.1 IU/L; range, 1.04 to 21 IU/L)(P = .001 and P = .007, respectively). Surprisingly, we found significantly lower LH in the group of men with TGCC (median, 3.6 IU/L; range, 1.12 to 11.9 IU/L) than in healthy men (median, 4.7 IU/L; range, 1.3 to 11.9 IU/L) (P = .01). We could not detect any differences between men with testicular cancer and men with malignant lymphomas and healthy men with regard to serum levels of testosterone, SHBG, and estradiol. Men with testicular cancer who had increased hCG levels had significantly lower LH and significantly higher testosterone and estradiol than those without detectable hCG levels. CONCLUSIONS Spermatogenesis is already impaired in men with testicular cancer before orchiectomy. Neither local suppression of spermatogenesis by tumor pressure nor a general cancer effect seems to fully explain this impairment. The most likely explanation is preexisting impairment of spermatogenesis in the contralateral testis in men with testicular cancer. The question of whether also a pre-existing Leydig cell dysfunction is present in men with testicular cancer could not be answered in this study because the tumor seems to have a direct effect on the Leydig cells. Men with testicular cancer had low LH values as compared with controls. We speculate that increased intratesticular level of hCG also in men without measurable serum hCG may play a role by exerting LH-like effects on the Leydig cells, causing increased testosterone and estrogen levels and low LH values in the blood.