BACKGROUND: Prepubertal patients receiving chemotherapy are relatively resistant to cyclophosphamide-induced germinal cell alterations. OBJECTIVE: To study the possible protective effect of testosterone used to inhibit germinal cell activity in men who are receiving cyclophosphamide. DESIGN: Randomized, clinical trial. SETTING: University medical center. PATIENTS: 15 patients with the nephrotic syndrome who were treated with cyclophosphamide for 6 to 8 months. INTERVENTION: Five patients received daily oral cyclophosphamide, five received cyclophosphamide in monthly bolus injections, and five received monthly intravenous boluses of cyclophosphamide plus testosterone (100 mg intramuscularly every 15 days). MEASUREMENTS: Sperm counts, serum follicle-stimulating hormone levels, and serum luteinizing hormone levels were measured before, during, and after treatment with cyclophosphamide alone or cyclophosphamide plus testosterone. RESULTS: The 10 patients who did not receive testosterone became azoospermic during cyclophosphamide therapy. In only 1 of the 10 patients did the sperm count return to normal 6 months after discontinuation of therapy. Follicle-stimulating hormone levels were elevated in these patients (mean +/- SE, 19.20 +/- 1.28 IU/L in patients receiving oral cyclophosphamide and 16.04 +/- 2.22 IU/L in patients receiving intravenous cyclophosphamide alone). All 5 patients who received testosterone became azoospermic or severely oligospermic during treatment but had a normal sperm count 6 months after the discontinuation of therapy. In these patients, the mean sperm count was 45.78 +/- 3.89 x 10(6)/mL and follicle-stimulating hormone levels were normal (5.08 +/- 0.56 IU/L). CONCLUSION: Testosterone given to men before and during an 8-month cycle of cyclophosphamide therapy for the nephrotic syndrome may preserve fertility.