A major objective of the second Intergroup Rhabdomyosarcoma Study (IRS-II) (1978 to 1984) was to preserve the bladder without compromising the survival of children with localized genitourinary sarcomas arising in or near the bladder. After incisional biopsy, 109 patients with localized, gross residual sarcoma of the prostate (43 patients), bladder (43 patients), vagina (20 patients), or cervix/uterine corpus (3 patients) were treated with vincristine, dactinomycin, and cyclophosphamide (VAC). After two to four drug courses, radiation therapy and/or surgery were used to treat patients with residual or recurrent tumor. The 3-year survival rate of patients treated on this primary chemotherapy regimen (70%) was similar to that of the primary surgery regimens of IRS-I (78%; P = 0.46), but the 3-year disease-free survival rate was significantly inferior (52% versus 70%; P = 0.02). Since the IRS-II encouraged bladder preservation at the onset of therapy, the percentage of patients with bladder and prostate tumors who retained the bladder was initially substantially higher in IRS-II (97%) than in IRS-I (58%). However, the percentages of 95 patients with bladder-prostate (BP) tumors in IRS-II who retained the bladder and were alive at 2 and 3 years after starting treatment were only 33% and 22%, respectively, compared with 26% and 23%, respectively, in the 66 patients with BP tumors in IRS-I. Thus, sequential treatment with primary chemotherapy, followed by radiation therapy and/or surgery as given in IRS-II, failed to improve the eventual bladder salvage rate.