Breast-duct endoscopy to study stages of cancerous breast disease

Background In breast-cancer research, we lack reproducible non-invasive access to breast tissue. Breast cancer is thought to start in the lining of the milk duct or lobule. We have had no direct access to this area other than in tissue removed surgically or by fine-needle aspiration. Our objective was to explore an intraductal approach to studying breast cancer and precancerous changes by duct cannulation and endoscopy. Methods In this pilot study, women in a university hospital were asked to participate. Nine patients with previously diagnosed ductal carcinoma-in-situ (DCIS) or invasive breast cancer who were about to undergo mastectomy under general anaesthesia agreed. After the patient was asleep but before the operation, we spent 15 minutes cannulating their ducts, obtaining washings, and attempting endoscopy. The ducts that had been cannulated were marked by instillation of dye or other materials, the nipple was sutured shut, and the mastectomy proceeded as planned. The mastectomy specimen underwent extensive histopathological examination. Findings We were successful in intraductal cannulation and endoscopy in seven of the nine patients, In five of the nine, we obtained epithelial cells in the washings. In one the cells were consistent with proliferative disease, in three there was atypical epithelium, and in one there was frank DCIS. With several different contrast materials injected into separate duct orifices, DCIS was confined to a single duct system. Interpretation We found that the intraductal approach is feasible for the study of the early changes of breast cancer. Technical difficulties include identification of the breast-duct orifices, determining the distensibility of the duct, and developing a reliable technique to obtain washings.