The term soft tissue sarcoma refers to a large variety of malignant tumors arising in extraskeletal connective tissues that connect, support, and surround discrete anatomic structures. All visceral organs also contain a connective stroma that can undergo malignant transformation. Because of the histological similarities of this group of tumors and their relative rarity, treatment prescriptions for patients that have disseminated disease are most often uniform. In this study, we asked the question whether adding a third drug (cyclophosphamide or actinomycin D) to Adriamycin (Adr [Adria Laboratories, Columbus, OH])-(3,3-dimethyl-1-triazeno)- imidazole-4-carboxamide (DTIC) would improve the response rate and/or survival. A unique feature of this cooperative group clinical trial was the mandatory pathology review of the histological material. All patients of the Southwest Oncology Group between June 1, 1976, and November 17, 1979, who had a biopsy-confirmed diagnosis of a soft tissue sarcoma with convincing clinical or biopsy-documented evidence of metastatic disease were eligible for the study. Patients were randomized to receive (1) Adr, 60 mg/m2 intravenously, day 1, and DTIC, 250 mg/m2 every 3 weeks (104 patients); (2) Adr and DTIC as in (1) and cyclophosphamide, 500 mg/m2, day 1 (112 patients); or (3) Adr and DTIC as in (1) and actinomycin D, 1.2 mg/m2, day 1, (119 patients). There was no statistically significant difference in response rates (33%, 34%, and 24%) (P = .25). Median durations of response were 31 weeks in the Adr-DTIC arm, 26 weeks in the cyclophosphamide-DTIC-Adr arm, and 23 weeks in the Adr-DTIC-Actinomycin D arm (P = .78). Median durations of survival were 37, 42, and 50 weeks, respectively. Again, no statistically significant differences were observed (P = .59). Toxicities from each of these treatment arms were formidable and were equivalent. Prognostic factor analysis showed a prognosis based on bone marrow reserve, sex, and pathology subtype favorable to patients.