E-CADHERIN EXPRESSION AS A MARKER OF TUMOR AGGRESSIVENESS IN ROUTINELY PROCESSED RADICAL PROSTATECTOMY SPECIMENS ANGELO M. DE MARZO, BEATRICE KNUDSEN, KIRK CHAN-TACK, AND JONATHAN I. EPSTEIN ABSTRACT Objectives. Approximately 30% of clinically localized prostate adenocarcinomas treated by radical prosta- tectomy (RP) will recur within 10 years. To prevent recurrence, new adjuvant therapies are in development that seek to treat high-risk patients after surgery. To identify patients as candidates for these treatments, improved biomarkers for predicting prognosis are needed. Reduced expression of E-cadherin has been proposed as a new marker for predicting prognosis in prostate adenocarcinoma. Since few studies have examined the relation between risk factors for disease progression and E-cadherin expression using rou- tinely processed RP specimens, we used RP specimens to correlate downregulation of E-cadherin and pathologic stage at RP. Methods. Primary adenocarcinomas (n = 76) from formalin-fixed and paraffin-embedded RP specimens were evaluated by immunohistochemistry against E-cadherin (HECD-1) using heat-induced epitope retrieval and automated staining (BioTek Solutions). Normal appearing prostate epithelium was used as an internal control for each specimen. Staining was scored as an estimate of the percentage of tumor cells in each specimen that showed strong plasma membrane staining. Results. Specimens were divided into three categories with respect to Gleason score: intermediate (score 5 to 6, n = 31), intermediate to high (score 7, n = 25), and high (score 8 to 9, n = 20). For pathologic stage, specimens were divided into three categories: low stage/organ confined (pT2, n = 30), intermediate stage/ limited extraprostatic extension (pT3a, n = 25), and high stage/seminal vesicle-pelvic lymph node metas- tases (pT3b-any pTN1, n = 21). In univariate analysis, reduced levels of E-cadherin correlated with advanced Gleason score (P = 0.003) and advanced pathologic stage (P = 0.008). In multivariate analysis, E-cadherin, preoperative prostate-specific antigen, and Gleason score all contributed independently to the prediction of high-stage disease (P 0.0001). Ten pelvic lymph node metastases from this same patient cohort were stained for E-cadherin. All were positive and 9 of 10 were moderately to strongly positive. Conclusions. Since essentially all patients in the high-stage category have a very high likelihood of disease recurrence, we conclude that the study of E-cadherin in a prospective manner as a potential biomarker of disease progression in patients with clinically organ-confined prostate cancer who undergo RP is warranted. Additionally, our finding that most metastatic tumor cells in pelvic lymph nodes express E-cadherin supports the notion that the establishment of the distant colonization and growth of metastatic tumor cells may be facilitated by expression or re-expression of previously downregulated E-cadherin. This would strongly suggest that irreversible genetic inactivation through mutation or allelic loss at 16q2.3 is probably not the mechanism of E-cadherin downregulation in most abnormally expressing primary prostate carcinomas. UROLOGY 53: 707–713, 1999. © 1999, Elsevier Science Inc. All rights reserved. This study was funded in part by an intradepartmental grant to Dr. De Marzo and by Public Health Services Specialized Program in Research Excellence (SPORE) in Prostate Cancer, grant P50CA58236. From the Department of Pathology, Johns Hopkins University Medical Institutions, Baltimore, Maryland; Department of Pa- thology, New York Hospital-Cornell Medical Center, New York, New York; and Department of Medicine, University of Pennsyl- vania School of Medicine, Philadelphia, Pennsylvania Reprint requests: Angelo M. De Marzo, M.D., Ph.D., Depart- ment of Pathology, Ross 512B, 720 Rutland Avenue, Baltimore, MD 21205 Submitted: August 3, 1998, accepted (with revisions): October 6, 1998 ADULT UROLOGY © 1999, ELSEVIER SCIENCE INC. 0090-4295/99/$20.00 ALL RIGHTS RESERVED PII S0090-4295(98)00577-9 707