Therapeutic postprostatectomy irradiation

The purpose of this study was to determine the outcome of patients receiving external beam radiation for an elevated postprostatectomy prostate-specific antigen (PSA) level. Between December 1991 and September 1998, 108 patients received definitive radiation therapy for elevated postprostatectomy PSA levels. The median dose of irradiation was 68 Gy (range, 48-74 Gy). During treatment, the PSA levels were checked an average of 5 times (range, 3-7 times). Prostate-specific antigen values were judged to decline or increase during treatment if they changed by more than 0.2 ng/mL. After treatment, biochemical failure was defined as a measurable or rising PSA > 0.2 ng/mL. Median follow-up was 51 months (range, 3-112 months). Fifty-eight patients (54%) had evidence of biochemical failure. The 3- and 5-year actuarial biochemical relapse-free (bNED) survivals for all patients were 55% and 39%, respectively. Upon univariate analysis, intratreatment PSA and preradiation PSA were significant predictors of bNED survival. Patients with a PSA level that decreased during treatment had a 5-year bNED survival of 43% compared to 10% in patients with an increasing PSA level (P = 0.0002). Using the preradiation therapy PSA value as a continuous variable, higher preradiation therapy PSA levels were associated with an increased risk of failure (P = 0.004). Cut points of pretreatment PSA were derived at 0.9 ng/mL and 4.2 ng/mL using the Michael Leblanc recursive partitioning algorithm. The 5-year bNED rate for patients with a preradiation therapy PSA < 0.9 ng/mL was 45% versus 42% for patients with preradiation therapy PSA between 0.9 and 4.2 ng/mL and 21% for patients > or = 4.2 ng/mL (P = 0.0003). Patients with a Gleason score of < or = 7 had a 5-year bNED rate of 38% compared to 37% for patients with a Gleason score > 7 (P = 0.27). Other factors examined individually that did not reach statistical significance included time from surgery to radiation therapy, race, seminal vesicle involvement, pathological stage, surgical margin, and perineural invasion. Upon multivariate analysis, only preradiation therapy PSA (P < 0.001) and the PSA trend during radiation therapy (P < 0.001) were significant factors. The results of therapeutic radiation for patients with elevated postprostatectomy PSA levels are sufficiently poor; other strategies should be explored as alternatives, including early adjuvant postprostatectomy irradiation or the use of combined hormonal and radiation therapy in the salvage situation.