Clinical value of immunohistochemically detected lymphovascular invasion in endometrioid endometrial cancer

Introduction. Lymphovascular Invasion (LVSI) of tumour cells is marked as an important step in the process of tumour metastases and is an important prognostic factor in Endometrial Cancer (EC). Currently, the standard method for assessing LVSI is light microscopic examination of H&E stained sections. Tumour cells within lymphovascular spaces can evade detection on H&E staining if they are present in very small numbers or surrounded by a greater number of circulating cells. Dual immunostaining for epithelial and endothelial cell markers cell markers has been shown to increase detection rate of LVSI significantly. Objectives. To investigate the clinical significance of LVSI as detected by H&E (LVSI-H&E) and itnimmohistochemically (LVSI-IHC) in clinically stage I endometrioid EC patients. Methods. Single representative section of 90 patients with stage I endometriod EC were immumostained in accordance with established streptavidin-biotin peroxidase method using a mouse monoclonal pancytokeratin (PCK), clone AE1/AE3 and CD31 endothelial cell marker. The H&E sections and their corresponding immunostained sections were re-examined to identify LVSI. Clinical records were available on 72 patients. The following data were collected: age, race, parity, presentation, associated medical disorders (obesity, diabetes and hypertension), use of Tamoxifen or HRT, menopausal state, recurrence and survival. Results. Overall, LVSI was present in 45 (50%) cases and absent in 45 (50%) cases on IHC, as compared with 17 (19%) and 73 (81%) cases, respectively. on H&E. Statistical analysis revealed significant association between LVSI-H&E and depth of myometrial invasion (P < 0.0001). The median follow-up period was 161 months (range 5-207 months). During the follow-up period, six of 14 cases with evidence of LVSI-H&E presented with recurrence as opposed to six of 58 patients with no evidence (OR = 6.26, 95%: CI = 1.3-30.6). There was a significant association between tumour recurrence rate and LVSI-H&E (P = 0.01). The 5-year recurrence-free survival was 54% for the group with H&E evidence of LVSI (95%: CI = 44-64%) compared with 89% for the group without (95%: CI = 82-97%). There was a significant difference in the recurrence-free survival between the two groups (Chi-square = 6.96, P = 0.008). In contrast, LVSI-IHC was found to be significantly associated only with high-grade tumours (P = 0.01) and survival analysis revealed no statistically significant association with recurrence or survival. Conclusions. LVSI-H&E in stage I EC remains an important predictive factor of recurrent disease and reduced disease-free interval. Immunohistochemically detected LVSI is a common event, associated with tumour grade and appears to be of no statistically significant clinical value. (C) 2003 Elsevier Inc. All rights reserved.