Login to Access Video or Poster Abstract: MP57-20
Sources of Funding: This work was supported by Roswell Park Cancer Institute and National Cancer Institute (NCI) grant P30CA016056.

Introduction

Our group reported on low detectable prostate-specific antigen (PSA) during the first 3 years after radical prostatectomy and proposed criteria to predict subsequent biochemical recurrence (BCR). The purpose of this study is to extend follow-up, add new patients, and compare specific predictors of low detectable PSA that progress to BCR.

Methods

An institutional database was queried to identify men with ultrasensitive PSA monitoring who underwent open or robot-assisted radical prostatectomy (RARP) between Jan 1993 to Dec 2011. Serum PSA and its pattern during the first 3 years of follow-up divided 663 men into 3 groups: 1)undetectable PSA (≤detectable PSA in lab where measured), 2) low detectable-stable PSA (≥detectable PSA in lab where measured and <0.2 ng/ml, no 2 subsequent increases and/or PSA velocity <0.05 ng/yr), 3)low detectable-unstable PSA (≥detectable PSA in lab where measured and <0.2 ng/ml, 2 subsequent increases and/or PSAV ≥0.05 ng/yr). The primary endpoint was BCR, defined as PSA of 0.2 ng/ml or greater, or receipt of radiation therapy beyond 3 years of follow-up. Time to BCR was summarized using standard Kaplan-Meier and log-rank tests.

Results

The five and ten-year BCR- free survival for all groups was 93% (95% CI: 0.9-0.95) and 73% (95% CI: 0.67-0.79), respectively. Five-year BCR-free survival differed among the 3 groups (p<0.001): 99% (95% CI 0.97-0.99) for undetectable; 89% (95% CI: 0.82-0.94) for low detectable-stable group; and 62% (95% CI: 0.48-0.74) low detectable-unstable group. Ten-year BCR-free survival was 80% (95% CI 0.72-0.86) for undetectable, 67% (95% CI: 0.53-0.78) for low detectable-stable group, and 46% (95% CI: 0.29-0.61) low detectable-unstable group (p<0.001). Low detectable-stable patients were more likely to have lower NCCN risk category (57.3% vs. 39.6%, p=0.04), low pre-operative PSA (91.2% vs. 88.7%, p=0.03), and an open procedure (30.1% vs. 15.1%, p=0.04) than low detectable-unstable patients. No difference in ≥pT2 stage (31.9% vs. 34%, p=0.79), pathologic Gleason score >7 (8% vs. 9.4%, p=0.38), positive margin status (22.1% vs. 24.5%,p=0.73), or extent of focal margin positivity (44% vs. 61.5%, p=0.38) was found.

Conclusions

Low detectable PSA patients have risk of BCR more similar to undetectable PSA patients and may warrant monitoring instead of salvage radiation.

Funding

This work was supported by Roswell Park Cancer Institute and National Cancer Institute (NCI) grant P30CA016056.