Dehydroepiandrosterone (DHEA) has been shown to affect the growth of mammary carcinomas both in vitro and in vivo. In humans, very high levels of DHEA and/or dehydroepiandrosterone sulfate (DHEAS) have been found in breast tissues and secretions, and epidemiological studies suggest a role of these steroids in the modulation of breast cancer growth. An uptake from plasma and a transformation from precursors can be both postulated, but the main source of the adrenal C19 steroids found within the breast is debated. Attempting to clarify this point, in ten patients undergoing surgery for breast cancer we studied: a) DHEAS and DHEA concentrations in tumor tissue; b) the differences between DHEAS (or DHEA) concentration in peripheral venous plasma and that draining the affected breast, that we assume to reflect the arteriovenous gradient of these steroids; c) DHEA sulfatase activity in tumor tissue. Results show that DHEA sulfatase activity is not related to DHEAS or DHEA concentrations in breast cancer tissue. A negative DHEA plasma gradient across the breast is unveiled, whereas DHEAS levels are not different in blood supplying and draining the breast with cancer. The DHEA plasma gradient across the breast is positively related to DHEA concentration in tumor tissue. Data are consistent with the hypothesis that the plasma source contributes remarkably to DHEA found within breast cancer tissue.