Evaluation of the prognostic significance of MSMB and CRISP3 in prostate cancer using automated image analysis.

Despite prostate cancer being the most frequent cancer in men in the Western world, tissue biomarkers for predicting disease recurrence after surgery have not been incorporated into clinical practice. Our group has previously identified β-microseminoprotein (MSMB) and cysteine-rich secretory protein-3 (CRISP3) as independent predictors of biochemical recurrence after radical prostatectomy. The purpose of the present study was to use automated image analysis, enabling quantitative determination of MSMB and CRISP3 expressions in a large cohort and to validate the previous findings. MSMB and CRISP3 protein expressions were assessed on tissue microarrays constructed from 3268 radical prostatectomy specimens. Whole-slide digital images were captured, and a novel cytoplasmic algorithm was used to develop a quantitative scoring model for cytoplasmic staining. Classification regression tree analysis was used to group patients, with different risk for biochemical recurrence, depending on level of protein expression. Patients with tumors expressing high levels of MSMB had a significantly reduced risk for biochemical recurrence after radical prostatectomy (HR=0.468; 95% CI 0.394-0.556; P<0.001). Multivariate analysis adjusted for clinicopathological parameters revealed that MSMB expression was an independent predictor of decreased risk of recurrence (HR=0.710; 95% CI 0.578-0.872; P<0.001). We found no correlation between CRISP3 expression and biochemical recurrence. In this current study, we applied a novel image analysis on a large independent cohort and successfully verified that MSMB is a strong independent factor, predicting favorable outcome after radical prostatectomy for localized prostate cancer.