Perineural invasion is not predictive of biochemical outcome following prostate brachytherapy

The purpose of this article is to evaluate whether the presence of perineural invasion (PNI) in the biopsy specimen is predictive of 5-year biochemical disease-free outcome after prostate brachytherapy.

Four hundred twenty-five patients underwent transperineal ultrasound-guided prostate brachytherapy using either 103Pd or 125I for clinical T1b/T3a NXMO (1997 American Joint Committee on Cancer) adenocarcinoma of the prostate gland from April 1995 to October 1999. No patient was lost to follow-up, and no patient underwent pathological lymph node staging. Two hundred twenty-one patients were implanted with 103Pd, and 204 patients were implanted with 125I. Of this cohort, 190 patients were implanted without supplemental beam radiation, and 235 received moderate-dose external-beam radiation therapy followed by a prostate brachytherapy boost. One hundred sixty-three patients received hormonal manipulation in conjunction with the brachytherapy implant (86 of these patients received moderate-dose external-beam radiation therapy before brachytherapy), and 262 patients were hormone naïve. Perineural invasion, defined as carcinoma tracking along or around a nerve within the perineural space, was identified in 105 patients (24.7% of the population). The median patient age was 68 years (range, 48-81 years). The mean follow-up was 37.1 +/- 15.2 months, and the median follow-up was 35.4 months (range, 6-74 months). Follow-up was calculated from the date of implantation. Biochemical disease-free survival was defined by the American Society of Therapeutic Radiation and Oncology (ASTRO) consensus definition.

When Kaplan-Meier survival analysis was performed, the presence of PNI did not predict failure. When PNI was entered into a Cox regression analysis with evaluated clinical predictors of failure (age, clinical T stage, pretreatment prostate-specific antigen level, and Gleason score) or treatment parameters (use of neoadjuvant hormonal therapy, supplemental external-beam radiation therapy, and choice of isotope), PNI did not add to the predictive strength of these variables. The median disease-free prostate-specific antigen level was 0.1 ng/mL for the entire cohort.

Our results indicate that actuarial 5-year biochemical outcomes with a prostate brachytherapy approach that utilizes generous periprostatic margins is not dependent on the presence of PNI. In addition, PNI should not be used as an independent prognosticator in determining the need for combined-modality therapy in patients undergoing prostate brachytherapy.