We report a comparative study of CA 15.3 and EMCA (epithelial mucin core antigen) in 77 consecutive women with newly diagnosed UICC assessable metastatic breast cancer, 59 patients received hormones and 18 chemotherapy. Assessments of response were made prior to commencing therapy and repeated 2 monthly. Sites of metastatic disease included bone (34), pulmonary (8), bone and pulmonary (14) and visceral (21). Using a cut-off of 33 U ml-1 changes in EMCA at 2, 4 and 6 months showed a highly significant correlation (P < 0.001) with UICC assessed response at 6 months; selectivity 70%, sensitivity 80%, specificity 91%, positive predictive value 84%; negative predictive value 89% at 2 months. Corresponding values for CA 15.3: selectivity 89%, sensitivity 85%, specificity 91%, PPV 92% and NPV 91%. Four of eight patients unassessable by CA 15.3 were assessable by EMCA; four patients expressed neither marker. EMCA appears to reflect tumour bulk and may be useful in monitoring therapy in patients with advanced breast cancer. With an easier and more robust assay format than CA 15.3, EMCA is potentially a more useful marker.