Source
Department of Radiation Oncology, Duke Comprehensive Cancer Center, Duke University Medical Center, Durham, NC, USA.
Abstract
PURPOSE:
To determine the efficacy of radiotherapy (RT) in patients with an isolated elevation of prostate specific antigen (PSA) after radical prostatectomy (RP).
METHODS AND MATERIALS:
Between November 1987 and May 1993, 53 patients with adenocarcinoma of the prostate were referred for pelvic RT for an elevated PSA after RP. No patient had clinically or radiographically apparent local or distant disease, nor had an undergone prior androgen ablation. Patients received a median dose of 61.2 Gy to the prostatic bed. An undetectable PSA was required to be considered disease free (NED). Univariate and multivariate analyses were performed to identify factors predictive of becoming disease free after RT.
RESULTS:
The median follow-up was 15 months. Of the 53 patients, 16 (30%) became NED after RT and 15 (28%) had a declining (n = 11) or stable (n = 4) PSA at last evaluation. The median time after RT to achieve an undetectable PSA was 9.3 months. At 12 and 24 months, the actuarial disease-free survival was 30 and 23%, respectively; actuarial progression-free survival was 71 and 26%, respectively. By univariate analysis, the last PSA level before RT (i.e., the pre-RT PSA) and an undetectable PSA after RP were significant predictors of becoming NED (p = 0.0001 and 0.04, respectively). However, on multivariate analysis, only the pre-RT PSA remained significant (p = 0.01). The mean pre-RT PSA differed significantly between patients who became NED after RT and those who did not (1.5 +/- 0.2 ng/ml vs. 7.6 +/- 1.6 ng/ml, respectively; p = 0.018). Fourteen of 27 (52%) patients with pre-RT PSA levels < or = 2.5 ng/ml became NED, vs. only 2 of 26 (8%) patients with higher levels. There were no severe acute or late sequelae of RT.
CONCLUSION:
Prostatic-bed RT for an elevated serum PSA after RP is most effective in patients with a pre-RT PSA < or = 2.5 ng/ml. Patients with significantly higher PSA values are unlikely to benefit from RT, possibly due to the presence of occult distant metastases. The optimal therapy for this latter group remains to be determined.