Investigations concerning a possible influence of thyroid disease and function on the risk for breast cancer have given contradictory results--although a slight hypothyroidism was repeatedly suggested and supported by increased thyrotropin concentrations and exaggerated TRH-responses. The selection of proper controls was stressed and in this study 179 consecutively diagnosed breast cancer patients were compared with 179 age-matched controls chosen from a computerized population register. The prevalence of thyroid disease and thyroxine treatment showed no differences between the groups. Slight but significant differences were found in the patient group--with a higher mean value for thyrotropin, reverse triodothyronine and T3-resin uptake and a lower mean value for triiodothyronine. No difference was found concerning thyroxine. This pattern of changes was inconsistent with a hypothyroidism, but has recently been found in many non-thyroidal diseases. It is suggested that it may represent a secondary and probably extrathyroidal metabolic change, most likely due to an altered peripheral conversion of the thyroxine.