The clonal nature of neoplastic lesions such as invasive breast cancer and ductal carcinoma in situ (DCIS) has been widely proven by several proliferative, genetic or other malignancy-associated markers. The aim of this study is to clarify whether benign hyperplastic lesions such as ductal hyperplasia of usual type (DH) and papilloma can be distinguished from neoplastic lesions such as DCIS by X-chromosome inactivation analysis. Clonal analysis was performed using a polymerase chain reaction-based assay for non-random X-chromosome inactivation of the human androgen receptor gene (HUMARA). Formalin-fixed and paraffin-embedded archival tissue of ten DCIS, sixteen DH, nine papillomas, and seven normal terminal ductal lobular units (TDLUs) was laser-microdissected to avoid contamination with surrounding tissue. All of the cases analysed revealed a monoclonal origin. Furthermore, in one of these cases, opposite X chromosomes were inactivated within the same breast. X-linked inactivation analysis clearly demonstrates that, at least in the breast, monoclonality is not restricted to neoplastic processes. The data support the hypothesis that the mammary gland is organized into distinct stem cell-derived monoclonal patches and that TDLUs are monoclonal in origin. Any proliferative lesion arising within such a pre-existing clonal patch should therefore be clonal, irrespective of whether it originates from one or more patch cells. Thus, X-chromosome inactivation analysis cannot be considered a valid method for distinguishing between neoplastic and hyperplastic breast lesions.