Postop PSA to Stratify Risk After Radical Prostatectomy

Background

Although radical prostatectomy provides excellent control for localized prostate cancer, pathologic examination of approximately one-third of specimens will reveal positive surgical margins (PSM).^[1,2] Numerous studies report that the presence of PSM adversely affects cancer specific outcomes and considerably increases the risk of biochemical recurrence (BCR).^[3] However, the optimum management of patients with PSM remains controversial.

Evidence from randomized trials suggests that immediate radiotherapy after the surgery, rather than watchful waiting, is more appropriate for the patient with pathologically advanced disease because it can improve cancer-specific and overall survival.^[4–6] Despite the fact that the most contemporary guidelines do not uniformly recommend adjuvant therapy for patients with adverse pathologic characteristics at radical prostatectomy, it has been shown that in daily clinical practice the patients with PSM were significantly more likely to receive immediate adjuvant treatment.^[7]

But not all of the patients with PSM develop BCR and the policy of adjuvant radiotherapy could result in considerable over-treatment. Therefore correct identification of those patients most likely to benefit from adjuvant management is of paramount importance. However, attempts to improve early staging after the surgery already hint at several difficulties. The ability of imaging modalities remains limited.^[8] The impact of PSM-associated variables (location, focality, length and Gleason score at the margin) on clinical decision making was not clearly established yet.^[2] Even routine use of frozen section on all cases has not fulfilled its expectation to provide effective control of surgical margin status.^[9,10]

Postoperative PSA measurements are generally performed 3 months after the surgery, although a significant decline in PSA may be detectable much earlier. Results from several studies indicate that intensive monitoring of PSA changes early after radical prostatectomy may provide clinically useful information, which facilitates identification of surgical failure.^[11–14] Moreover, recently introduced ultrasensitive PSA detection techniques are offering a new insight into the changes in serum PSA at very low concentration. It has been demonstrated that after a properly performed radical prostatectomy, measurable PSA is most likely attributed to the presence of active prostate cancer cells rather than to retained benign prostatic tissue.^[15,16] It is therefore conceivable, that real candidates for immediate adjuvant therapy, who have active prostate cancer cells remaining in the body after the surgery, should present with higher postoperative serum PSA in a comparison to individuals with incorrect diagnosis of PSM.

This hypothesis prompted us to evaluate the ability of early postoperative ultrasensitive PSA levels to reduce the overtreatment rate by further stratification of potential candidates for immediate adjuvant radiotherapy.

Table 1. Differences in potential predictive parameters according to the appearance of BCR
Parameters	BCR	BCR-free	p value
n	55	61
Pre-operative PSA (ng/ml, range)	9.20 (2.85–38.20)	7.60 (3.08–59.7)	0.172
Pathologic Gleason sum ≥ 7	27 (49.1%)	32 (52.5%)	0.717
Pathologic extraprostatic extension	25 (45.5%)	26 (42.6%)	0.393
Seminal vesicle invasion	8 (14.5%)	3 (5%)	0.077
PSA – day 14 (ng/ml, range)	0.298 (0.071–1.940)	0.238 (0.045–3.020)	0.264
PSA – day 30 (ng/ml, range)	0.085 (0.008–0.747)	0.022 (0.003–0.692)	<0.0001
PSA – day 60 (ng/ml, range)	0.041 (0.003–0.882)	0.010 (0.003–0.618)	<0.0001
PSA – day 90 (ng/ml, range)	0.042 (0.003–1.010)	0.008 (0.003–0.853)	<0.0001
PSA – day 180 (ng/ml, range)	0.013 (0.015–1.167)	0.011 (0.003–0.931)	<0.0001
PSA nadir (ng/ml, range)	0.022 (0.003–0.137)	0.007 (0.003–0.039)	<0.0001
Time to PSA nadir (months, range)	2 (0.5–3)	2 (1–12)	0.005

BCR, biochemical recurrence; PSA, prostate specific antigen;
Data presented as n (%) or median (range).

Table 2. Cox regression analyses of PSA cut-offs at particular postoperative measurement day as a predictor of BCR
Parameters	Hazard ratio	p value
PSA – day 14 (0.707 ng/ml)	2.60 (1.24–5.56)	0.133
PSA – day 30 (0.073 ng/ml)	4.35 (2.39–8.06)	<0.001
PSA – day 60 (0.041 ng/ml)	9.59 (5.15–17.95)	<0.001
PSA – day 90 (0.012 ng/ml)	12.34 (4.98–41.09)	<0.001
PSA – day 180 (0.021 ng/ml)	12.85 (6.29–29.82)	<0.001

BCR, biochemical recurrence; PSA, prostate specific antigen; PSM, positive surgical margins.
Data presented as PSA-measurement day (cut-off value) and hazard ratio (95% confidence interval).