Experimental models have demonstrated the Adriamycin (doxorubicin; Adria Laboratories, Columbus, OH)-potentiating activity of lonidamine. Phase II clinical trials on advanced breast cancer have shown that this drug induced a 16% objective response rate. Present multicentric randomized trial was conducted to verify whether lonidamine can potentiate the antineoplastic effects of conventional fluorouacil, Adriamycin, cyclophosphamide (FAC) chemotherapy in advanced breast cancer. From January 1987 to December 1989, 265 patients were enrolled in this study, and 231 are evaluable for response. After stratification according to institution and ECOG performance status (PS), the patients were randomly allocated to receive either standard FAC therapy (group A) or FAC plus lonidamine (600 mg orally daily three times a day) (group B). After three FAC courses, the patients with no progressive disease were further randomized to either receive continuous treatment up to the time of tumor progression (maximum: 10 courses) or to discontinue therapy when a response "plateau" was reached. In this latter group, the same therapy was restarted at relapse or disease progression. Objective response (complete response plus partial response) was significantly higher in group B (66.3%) compared to group A (42.3%). The actuarial median times to disease progression was also significantly longer (P less than 0.0001) in group B (median 9 months) than in group A (median 6 months). Other than myalgia and gastric pain, no increased toxicity was observed in the lonidamine. The analysis of second randomization are yet available because of the longer follow-up time required. Present findings suggest an interesting additive effect of lonidamine when combined with FAC chemotherapy and warrant further investigation in other therapeutic regimens and in other neoplastic diseases.