The concept of an intraductal approach to evaluate the breast microenvironment assumes direct access to the cancer-containing duct. Central duct access to the cancer-affected lobe is essential if cytology or cell markers are to be useful indicators of pre-malignant change. Access to the cancer-bearing lobe would be less important if field change effects of malignant change were predominantly supra-lobar. The aim of this study was to determine how often duct lavage fluid drains the breast cancer-affected segment. 58 patients undergoing mastectomy for breast cancer were recruited among which 47 had at least one fluid-yielding duct. Following duct lavage, fluid-yielding ducts were perfused ex vivo with Polyurethane Elastomer (PU4ii) resin. Specimens were sliced sagittally, and the extent of resin perfusion and anatomical relationship to the cancer-affected segment was recorded. Computed tomography (CT) scanning was performed on selected mastectomies before cut-up for a feasibility study of 3D duct reconstruction. The median number of fluid-yielding ducts cannulated per cancer-affected breast was 2 (range 1-4). 35/47 (74%) mastectomy specimens were successfully cannulated for resin perfusion. 29/35 (83%) showed tracing of the cancer-affected duct system, 6/35 resin perfusions traced duct systems unaffected by cancer and 12/35 perfusions extravasated. The proportion of sagittal breast slices perfused by resin was 13-68% (median 43%). Volume rendering CT showed it is feasible to produce a simulated image of the perfused ducts. Duct access to the cancer-containing segment is feasible in the majority of patients. Fluid-yielding ducts proportionately drain a significant volume of the breast. Large symptomatic cancers may cause obstruction with distal collapse. Further quantitative study of breast perfusion CT scans may be helpful for estimating the volume fraction of breast tissue perfused by fluid-yielding ducts. The intraductal approach is a valid concept for biomarker assessment of cancer-containing breast segments.