OBJECTIVES: Reliable prognostic indicators are needed for a better pretherapeutic assessment of the agressiveness of organ-confined prostate cancer (PC) lesions. The 67-kD laminin receptor (67LR) is a cell-surface-associated protein involved in the acquisition of the invasive and metastatic phenotype of a variety of human cancer cell types. We have previously shown that 67LR detection in PC tissues from radical prostatectomy (RP) specimens is an independent predictor of biochemical (PSA) relapse in patients with clinically localized PC. In this study, we assessed 67LR detection in diagnostic PC biopsies as a predictor of biochemical relapse after RP. METHODS: Diagnostic biopsy and subsequent RP tissue specimens from 151 patients with clinically localized PC were immunohistochemically analyzed for 67LR expression. The level of 67LR expression was evaluated by both intensity and extent of the staining. Clinicopathological preoperative and postoperative parameters, including 67LR expression, were correlated with each other and tested as predictors of biochemical relapse. RESULTS: 67LR was detected in 67.5 and 68.2% of biopsies and RPs, respectively. 67LR detection in RP specimens was an independent predictor of relapse. The level of 67LR expression in the biopsy was significantly associated with the biopsy Gleason score (p<0.05) but failed to predict the pathological stage (p>0.1). Biochemical progression-free estimates for patients whose biopsy did or did not express the protein differed with only borderline statistical significance (p = 0.05). Multivariate analysis identified biopsy Gleason score as the only independent preoperative predictor of recurrence. Significant discrepancies in levels of 67LR expression were found between matched biopsy and RP specimens (p<0.05), with exact agreement rates <40%. CONCLUSIONS: 67LR detection in PC biopsies was not a significant preoperative predictor of outcome after RP. Heterogeneity of 67LR expression and biopsy sampling errors most likely represented the main reasons for discordant results between biopsy and RP specimens.