Résumé : Clinical-grade preparations of human chorionic gonadotropin (hCG) have been shown to be toxic to Kaposi's sarcoma (KS) cells. However, the results of clinical studies using commercial hCG preparations KS remain highly contradictory. More particularly, some hCG preparations could have a paradoxical growth effect on KS. Such discrepant results may be explained by the fact that the anti-KS activity is not associated with hCG itself but with one or more factors that are co-purified with the hormone. We found here that crude urine from first trimester pregnant women, the current source for commercial hCG, had a growth stimulatory effect on KS cells. By contrast, urine from last trimester pregnant women, from non-pregnant young women, from menopausal women and from men exhibited neither a growth stimulatory nor a growth inhibitory effect on KS cells. The amplitude of this pregnancy urine-associated pro-KS activity/hCG unit was higher than that achieved with clinical-grade hCG preparations. Partial co-purification of pregnancy-associated factors during the extraction procedure of commercial hCG from urine may explain the pro-KS activity achieved with some hCG preparations. We, therefore, suggest a cautious use of hCG purified from pregnancy urine for the treatment of KS.