| Sumario: | Objective: To study relationships of ERG, SPINK1 and PTEN status in prostate cancer initiation and progression. Methods: Study included 272 low and intermediate risk patients who underwent radical prostatectomy (RP) from 2000 to 2012 at Kingston General Hospital. All cases were represented on a tissue microarray. When possible, analysis included both high- and low-grade cancer components from index and secondary tumour lesions. Data and results: Consistent with previous studies, PTEN loss (both partial and/or complete) was significantly associated with biochemical recurrence (BCR), pathological stage and Gleason score (p<0.0001). Interestingly, very strong ERG expression on any core was associated with increased risk of BCR (p=0.002) and increased Gleason score (p<0.0001), whereas average ERG levels per case was not significantly associated with any of these variables. Patients with both PTEN loss and strong ERG positivity had the shortest survival compared to all other groups (p=0.007). Also, ERG and PTEN expression were different between index and spatially distant tumour foci (p=0.048; Pearson Chi-Square test). Lastly, we observed an extremely low rate of SPINK1 expressing early prostate cancers (n=10, <4%) and, contrary to previous studies, ERG and SPINK1 expression was not mutually exclusive (n=3). Conclusions: Combined analysis of ERG and PTEN is prognostically significant for radical prostatectomy patients. Focal/core-based expression of ERG is found to be more resistant to sampling error and is significantly correlated with Gleason score. These observations have important implications for the use of these biomarkers in biopsy tissue and indicates that their utility depends on sampling the index tumour focus.
|
|---|