Association of percent positive prostate biopsies and perineural invasion with biochemical outcome after external beam radiotherapy for localized prostate cancer

Purpose: Few studies have evaluated the significance of the percentage of positive biopsies (PPB) and perineural invasion (PNI) for patients treated with external beam radiotherapy (EBRT) for localized prostate cancer. Our goal was to investigate the value of these factors in predicting biochemical control (bNED) after EBRT. Methods and Materials: The study cohort consisted of 331 patients who received EBRT between 1993 and 1999 for clinically localized prostate cancer. The median follow-up was 4.4 years (range, 3 months to 9.6 years). The distribution by clinical T stage was as follows: T1 in 55 (17%), T2a in 94 (28%), T2b in 76 (23%), T2c in 74 (22%), T3a in 27 (8%), and Tab in 5 (2%). The pretreatment prostate-specific antigen (iPSA) level was less than or equal to10 ng/mL in 224 patients, 10.1-20 ng/mL in 72 patients, and >20 ng/mL in 35 patients. The biopsy Gleason score was less than or equal to6 in 216 patients and greater than or equal to7 in 115 patients. On the basis of the pathology report, the PPB was calculated for 239 patients and was <33% in 109, 34-66% in 72, and greater than or equal to67% in 58 patients. PNI was present in 30 patients. The median dose of EBRT was 68.4 Gy (range, 64-71 Gy). Patients were categorized into three risk groups: 142 patients were low risk (T1-T2, iPSA less than or equal to10 ng/mL, and Gleason score less than or equal to6), 137 were intermediate risk (increase in the value of one of the risk factors); and 52 patients were high risk (increase in value of two or more of the risk factors). Biochemical failure was defined as three consecutive rises in the PSA level. Results: The 5-year bNED rate for the entire cohort was 62%. The 5-year bNED rate for the low-, intermediate, and high-risk group was 79%, 51%, and 47%, respectively (p < 0.0001). On univariate analysis (log-rank test), clinical stage (p = 0.0073), grade (p < 0.0001), iPSA (p = 0.0043), risk group (p < 0.0001), PPB (p = 0.0193), and presence of PNI (p = 0.0137) correlated with bNED. For T1-T2a, T2b-T2c, and T3 patients, the 5-year bNED rate was 71%, 59%, and 40%, respectively. The 5-year bNED rate was 68% for those with an iPSA level of less than or equal to10 ng/mL and 49% for those with an iPSA level of >10 ng/mL. For patients with PPB <33%, 34-66%, and greater than or equal to67%, the 5-year bNED rate was 75%, 67%, and 51%, respectively. Within the intermediate-risk group, the PPB was significantly associated with the bNED rate: 67%, 52%, and 30% for those with PPB <33%, 34-66%, and greater than or equal to67%, respectively (p = 0.0046). This association was not seen in the low- or high-risk group. The 5-year bNED rate was 64% for patients without PNI and 48% for those with PNI. On multivariate analysis (Cox proportional hazards model), the statistically significant predictive factors for bNED were risk group (p = 0.0032) and PPB (p = 0.044). Using the chi-square test, statistically significant associations between T stage, PSA level, Gleason score, and risk group with PPB were found; PNI was significantly associated with T stage and PSA level only. Conclusion: Our results showed that PPB and PNI have a statistically significant impact on the bNED rate in patients treated with conventional dose of EBRT (less than or equal to71 Gy). Within the intermediate-risk group, the PPB was predictive of bNED, suggesting that patients with less than or equal to33% PPB had a statistically significant better treatment outcome compared with those with a greater PPB. PNI was not significant for bNED in multivariate analysis. The effects of these two prognostic factors in patients who have been treated with higher doses of RT (>75.6 Gy) should be studied. (C) 2004 Elsevier Inc.