Computerized quantification and planimetry of prostatic capsular nerves in relation to adjacent prostate cancer foci

Perineural invasion is discussed as a significant route of extraprostatic extension in prostate cancer (PCa). Recent in vitro studies suggested a complex mechanism of neuroepithelial interaction.

The present study was intended to investigate whether the concept of neuroepithelial interaction can be supported by a quantitative analysis and planimetry of capsular nerves in relation to adjacent PCa foci.

Whole-mount sections of the prostate were created from patients undergoing non-nerve-sparing laparoscopic radical prostatectomy. For each prostate, adjacent sections were created and stained both to identify capsular nerves (S100) and to localize cancer foci (hematoxylin and eosin).

Computerized quantification and planimetry of capsular nerves (ImageJ software) were performed after applying a digital grid to define 12 capsular sectors. For statistical analyses, mixed linear models were calculated using the SAS 9.3 software package.

Specimens of 33 prostates were investigated. A total of 1957 capsular nerves and a total capsular nerve surface area of 26.44 mm(2) were measured. The major proportion was found in the dorsolateral (DL) region (p<0.001). Adjacent tumor was associated with a statistically significant higher capsular nerve count compared with the capsules of tumor-free sectors (p<0.005). Similar results were shown for capsular nerve surface area (p<0.006). Subsequent post hoc analyses at the sector level revealed that the effect of tumor on capsular nerve count or nerve surface area is most pronounced in the DL region.

The presence of PCa foci resulted in a significantly increased capsular nerve count and capsular nerve surface area compared with tumor-free sectors. The present study supports former in vitro findings suggesting that the presence of PCa lesions may lead to complex neuroepithelial interactions resulting in PCa-induced nerve growth.