Objectives

To determine if men with adverse pathology but undetectable ultrasensitive (<0.01 ng/mL) PSA are at high-risk for biochemical recurrence (BCR), or if there is a subset of patients at low-risk for whom the benefit of adjuvant radiation therapy might be limited.

Patients and methods

We evaluated 411 patients treated with RP from 2001 to 2013 without adjuvant radiation who had an undetectable (<0.01 ng/mL) PSA level after RP but with adverse pathology [positive surgical margins (PSMs), extraprostatic extension (EPE), and/or seminal vesicle invasion (SVI)]. Multivariable Cox regression analyses tested the relationship between pathological characteristics and BCR to identify groups of men at highest risk of early BCR.

Results

On multivariable analysis, only pathological Gleason 7 (4 + 3), Gleason ≥8, and SVI independently predicted BCR (P = 0.019, P < 0.001, and P = 0.001, respectively), although on two-way analysis men with Gleason 7 (4 + 3) did not have significantly higher rates of BCR compared with patients with Gleason ≤6 (log-rank, P = 0.074). Men with either Gleason ≥8 (with PSMs or EPE) or SVI (15% of the cohort) defined a high-risk group vs men without these characteristics (3-year BCR risk of 50.4% vs 11.9%, log-rank, P < 0.001).

Conclusions

Among men with adverse pathology but an undetectable (<0.01 ng/mL) PSA level after RP, the benefits of adjuvant radiation are probably limited except for men with Gleason 8-10 (with PSMs or EPE) or SVI who are at high-risk of early BCR.

Objective

To determine whether there are subsets of men with pathological high grade prostate cancer (Gleason score 8-10) with particularly high or low 2-year biochemical recurrence (BCR) risk after radical prostatectomy (RP) when stratified into groups based on combinations of pathological features, such as surgical margin status, extracapsular extension (ECE) and seminal vesicle invasion (SVI).

Materials and methods

We identified 459 men treated with RP with pathological Gleason score 8-10 prostate cancer in the SEARCH database. The men were stratified into five groups based on pathological characteristics: group 1, men with negative surgical margins (NSMs) and no ECE; group 2, men with positive surgical margin (PSMs) and no ECE; group 3, men with NSMs and ECE; group 4, men with PSMs and ECE; and group 5, men with SVI. Cox proportional hazards models and the log-rank test were used to compare BCR among the groups.

Results

At 2 years after RP, pathological group was significantly correlated with BCR (log-rank, P < 0.001) with patients in group 5 (+SVI) having the highest BCR risk (66%) and those in group 1 (NSMs and no ECE) having the lowest risk (14%). When we compared groups 2, 3, and 4, with each other, there was no significant difference in BCR among the groups (~50% 2-year BCR risk; log-rank P = 0.28). Results were similar when adjusting for prostate-specific antigen, age, pathological Gleason sum and clinical stage, or after excluding men who received adjuvant therapy.

Conclusions

In patients with high grade (Gleason score 8-10) prostate cancer after RP, the presence of either PSMs, ECE or SVI was associated with an increased risk of early BCR, with a 2-year BCR risk of ≥50%. Conversely, men with organ-confined margin-negative disease had a very low risk of early BCR despite Gleason score 8-10 disease.