Objective risk definition for endometrial lesion spectrum: A diagnostic algorithm

Abstract Objective. Investigations for risk definition in endometrial lesion spectrum still go on. In this study, molecular, morphometric, immunohistochemical techniques were combined with conventional morphology to realize whether an algorithm is definable for risk assessment to progress an invasive carcinoma in endometrial glandular lesion spectrum is possible. Methods. The study was carried out on 20 benign endometria, 35 hyperplasias, and 20 adenocarcinoma cases. Clonality of glandular cells, the volume percent of endometrial stroma (VPS), PTEN inactivation, and proliferative index (PI) were evaluated. Statistical analysis was evaluated to set an objective algorithm. Results. All benign tissues had polyclonal (PC), whereas all malignant tissues had monoclonal (MC) glandular epithelium. Of hyperplasias, 19 were MC, and 16 were PC. VPS value of 55% had 100% sensitivity, and 80% specificity ( n = 67) to distinguish MC from PC. Neither PTEN nor PI data augmented the specificity or the sensitivity of clonal distinction. Conclusion. Clonality and VPS values were found to be significant in differential of endometrial lesions. With this rationale, a diagnostic algorithm for endometrial risk lesions was set. This algorithm is based on HE morphology, VPS and clonality findings, and has 100% sensitivity and specificity to discriminate neoplastic endometrium from hyperplasia.