This paper summarizes our most recent results of steroid enzyme studies on cultured breast and endometrial cancer cells. It deals mainly with estrogen 17 beta-hydroxysteroid oxidoreductase (17 beta HSOR) activity, which presides over estradiol (E2) and estrone (E1) interconversion, a major metabolic pathway of estrogens. Assessment of either the oxidative or reductive component of 17 beta HSOR was carried out on intact cells by means of an original approach based on reverse phase-high performance liquid chromatography and radioactive detection on line. This system allows the continuous monitoring of both precursor degradation and formation of several radiometabolites to assess rates and direction of steroid metabolism. Overall, hormone-responsive, estrogen receptor (ER)-positive cells, regardless of whether they were derived from breast (MCF7) or endometrial (Ishikawa) tumor tissues, showed a prevalence for reductive metabolism (E1--E2), whilst oxidative pathways (E2--E1) were largely dominant in non-responsive, ER-poor mammary (MDA-MB231) and endometrial (HEC-1A) cells. The above estimates of 17 beta HSOR activity were at variance with those obtained using the classical enzymology approach, not only in quantitative terms (being markedly lower using intact cell analysis), but also because the prevalent direction of estrogen metabolism was often reversed. Although striking methodological differences may well account for this discrepancy, intact cell analysis is undoubtedly more similar to the in vivo state than the artificial requirements of classical enzymology procedures.