Background: Recent Wilms' tumor (WT) trials and studies have tried to determine the minimal therapy needed for cure. The goal was survival without morbidity.
Patients and methods: From January 1989 to March 1994 the German Society of Pediatric Oncology and Hematology registered 440 patients (median age 2.9 years; 231 male, 209 female) with WTs (preoperative chemotherapy 362) for therapy according to the International Society of Pediatric Oncology Trial and Study 9. Therapy for relapse depended on site of relapse and therapy already received. Follow-up included inquiries for morbidity. Prognostic factors for relapse and death were evaluated.
Results: Five-year survival of WTs was 89.5%; 98.2% (385 of 392) of survivors had a follow-up of 5 years (range 0.8-12.6; median 8). In non-anaplastic WTs, young age (<2 years) was of significance (P = 0.026) for a better survival. Non-anaplastic WTs (407 patients) had a 5-year survival of 92.3%, versus 48.5% in anaplastic WTs (33 patients), and a 5-year relapse-free survival of 87.6% versus 42.4%. Survival after relapse was significantly worse for anaplastic than for non-anaplastic WTs (residual 3-year survival 11.8% versus 54.3%; P <0.0001). In preoperatively treated WTs, anaplasia was a strong prognostic factor for death [relative risk (RR) 4.7], followed by poor response to preoperative therapy (RR 3.6), stage IV (RR 3.2) and abdominal stage III (RR 2.2). Low abdominal stages (<III) dominated (280 versus 82). In the 334 unilateral stage I-IV WTs (median age 3.2 years), diffuse anaplasia (21 patients) had a 5-year relapse-free survival of 38.1%, versus 58.4% in blastemal WTs (25 patients); survival was 42.9% in diffuse anaplasia versus 84% in blastemal WTs. None of 46 patients (median age 1.9 years; 91.3% stages I or II) with differentiated WTs (nine epithelial, 37 stromal) relapsed despite their non-response; two died (one therapy related, one due to bilaterization). In the 25 non-anaplastic bilateral WTs, differentiated cases (one epithelial, eight stromal, 33.3% abdominal stage III) were more frequent (P = 0.048) than in unilateral WTs (one stromal, abdominal stage III relapsed). In all, 52.9% of the 5-year survivors had received adriamycin (250-400 mg/m(2)), 25.7% radiation, 6.4% ifosfamide (24-30 g/m(2)) and 6.7% carboplatin plus etoposide. Abnormal parameters according to the National Cancer Institute score were seen in 18.9% during follow-up, but only 6.4% were treated for morbidity at the end of follow-up. Three WTs developed renal failure due to Drash syndrome, but none due to tumor therapy. After adriamycin 1.9% of WTs (9% of those receiving 400 mg/m(2)) required therapy for cardiac toxicity.
Conclusions: Initial therapy should be more individualized, taking the above risk groups (age in non-anaplastic WTs, poor response, anaplasia, etc.) into account, as morbidity even after relapse therapy with ifosfamide, carboplatin and etoposide was not high. Milder therapy in low stages of differentiated and of well responding WTs should be tested.