The Third National Wilms' Tumor Study sought to reduce treatment for low-risk patients and find better chemotherapy for those at high risk for relapse. Eligible patients (1439) were randomized according to stage (I-IV) and histology (favorable [FH] or unfavorable [UH]), and contributed data to survival and relapse-free survival (RFS) analyses. Four-year (postnephrectomy) survival percentages and randomized treatment regimens for low-risk patients were 96.5% for 607 Stage I/FH patients who received dactinomycin (Actinomycin D [AMD], Merck Sharp & Dohme, West Point, PA) and vincristine (VCR) for 10 weeks versus 6 months; 92.2% for 278 Stage II/FH patients; and 86.9% for 275 Stage III/FH patients who received AMD + VCR +/- Adriamycin (ADR, Adria Laboratories, Columbus, OH) for 15 months. Stage II/FH patients also had either zero or 2000 cGy irradiation (RT) postoperatively and Stage III/FH patients either 1000 or 2000 cGy. Four-year survival was 73.0% for 279 high-risk patients (any Stage IV, all UH) who received postoperative radiation therapy (RT) and AMD + VCR + ADR +/- cyclophosphamide (CPM). Statistical analysis of survival and RFS experience shows that the less intensive therapy does not worsen results for low-risk patients and CPM does not benefit those at high risk.

EFS and survival appear to be directly related to histologic response to neoadjuvant chemotherapy.

This paper presents an update from the Late Effects Study Group on 292 cases of second malignant neoplasms (SMN) occurring in individuals who were diagnosed with their first neoplasm in childhood. Data are presented regarding the types of first and second neoplasm, the therapy administered, and the predisposing factors. Of the 292 cases (308 SMN), the most common primary was retinoblastoma followed by Hodgkin's disease, soft-tissue sarcomas, and Wilms' tumor. This is not similar to the relative frequency of these cancers in children but rather reflects specific risk factors. Bone sarcomas were the most common SMN among the 208 SMN developing in previously irradiated sites while acute leukemia was the most common SMN unassociated with radiation. Known predisposing conditions to cancer were present in 73 cases; retinoblastoma was the most common of these, followed by neurofibromatosis. There were ten patients with three and three patients with four malignant neoplasms. In 14 patients, the cause of SMN was not suggested by known risk factors as these patients had negative family histories and received no radiation or chemotherapy. We note, therefore, that although most cases of SMN in survivors of childhood cancer can be attributed to radiation, genetic disease, chemotherapy, or combinations of these, unrecognized predisposition or chance may also play a role.

, Children's Cancer Study Group institutions entered 95 patients with recurrent brain tumors into a Phase I1 trial of carboplatin 560 mg/mz every 4 weeks. Complete or partial responses were observed for one of 19 evaluable children with brainstem glioma, two of 14 with ependymoma, six of 19 with medulloblastoma or central nervous system primitive neuroectodermal tumor (PNET), and none of 15 with high-grade astrocytoma. Of 33 children with medulloblastoma, ependymoma, or central nervous system PNET, five of 12 with no prior cisplatin exposure had responses, and two of 2 1 with prior cisplatin exposure had responses (P = 0.03). Thirty-four percent of patients had absolute neutrophil count nadirs less than 500/pl, and 37% had platelet count nadirs less than 25,000/ pl. Sixteen percent had moderate to severe otoxicity, 10% had nausea and vomiting, and none had nephrotoxicity. Cancer 662465-2469,1990.

Data from four pediatric hospitals concerning 20 children treated for renal cell carcinoma (RCC) from 1964-1978 were reviewed. Median age of the patients (pts) was 11.8 years (range, 14 months-19 years). Twelve were male and eight female; 17 were white and three black. Most patients presented with pain and hematuria with or without a palpable mass. An intrarenal tumor was detected at IV urography (17 pts), arteriography (2 pts), or at surgery (1 pt). Treatment consisted of nephrectomy in 15 pts, renal biopsy (4 pts), or no surgery (1 pt), followed by chemotherapy (5 pts), radiation therapy (1 pt), or both (7 pts). Ten pts died of distant metastases at a median of one year (range, 0.2 to two years) after diagnosis. The other 10 pts (50%) survive free of relapse at a median of 4 years (range, two to ten years) from diagnosis. Proportions surviving free of recurrent disease two or more years by National Wilms' Tumor Study (NWTS) Group were 5/5 in Group I, 3/7 in Group II, 1/3 in Group III, and 1/5 in Group IV; by age at diagnosis, 6/6 in those under 11 years old and 4/14 in those 11 or older; and by type of surgery, 10/15 who had nephrectomy and 0/5 with limited or no surgery. The data indicate that radiation and chemotherapy had only minor if any influences on relapse-free survival. We conclude that (1) RCC in children is similar to its counterpart in adults; (2) RCC has a worse prognosis than Wilms' tumor except for the earliest stage; (3) nephrectomy alone is adequate treatment for Group I RCC, and (4) young age (less than 11 years old) may be prognostically favorable.