This study evaluates the degree of kidney damage during cisplatin/ifosfamide-based combination chemotherapy and its possible prevention by amifostine. Thirty-one patients with solid tumors stratified according to pretreatment were randomized to receive cisplatin/ifosfamide-based chemotherapy with or without amifostine (1000 mg absolute) given as a short infusion prior to cisplatin. Chemotherapy consisted of cisplatin (50 mg/m2), ifosfamide (4 g/m2) and either etoposide (500 mg/m2) (VIP regimen) or paclitaxel (175 mg/m2) (TIP regimen) repeated at 3 weekly intervals. For all patients the glomerular filtration rate (GFR) measured by creatinine clearance, serum creatinine, electrolytes and differential urinary protein excretion were determined prior to, during and after each treatment cycle. A total of 62 cycles of chemotherapy were evaluable. In the amifostine arm the GFR was almost completely maintained after application of two cycles of chemotherapy (121 to 108 ml/min), whereas in the control group a 30% reduction of the GFR (105 to 80 ml/min) was observed. In both groups marked increases of glomerular and tubular marker profiles peaking at day 3 after chemotherapy were found with a nearly complete reversibility of these changes prior to the next chemotherapy cycle. Patients receiving amifostine had a lower degree of hypomagnesemia, as well as a lower urinary excretion ofN-acetyl-glucosaminidase and albumin, indicating less tubular damage compared to the control patients. Treatment with 1000 mg amifostine resulted in an almost complete preservation of GFR. This corresponded to a slightly reduced excretion of tubular marker proteins and a lower incidence of hypomagnesemia during chemotherapy in amifostine patients compared to controls. This dose of amifostine may be sufficient for nephroprotection in patients without pre-existing risk factors for renal damage who undergo a restricted number of chemotherapy cycles.