Long-Term Results and Predictive Factors of Three-Dimensional Conformal Salvage Radiotherapy for Biochemical Relapse After Prostatectomy

Neuhoff. IJROBP 2007;67:1411

Radical prostatectomy (RP) is one of the first-line therapeutic options for patients with organ-confined prostate cancer. Biopsy Gleason score, pretreatment prostate-specific antigen (PSA) level, and clinical stage are established variables to predict outcomes after prostatectomy. However, even with use of these prognostic factors in patient selection, an increased level of PSA (0.2 ng/mL or greater) occurs in 15% of patients undergoing RP. Considering only the patients with positive margin, the risk of PSA recurrence increases to about 40%.

It is unclear whether PSA elevation after surgery indicates only local disease recurrence or distant metastatic disease. As a result, the optimal management for recurrent prostate cancer in patients with no clinical evidence of disease remains uncertain. In patients who experience PSA recurrence after RP, it is essential to distinguish local recurrence from distant metastases to determine which treatment may be the most appropriate. Slow PSA progression may indicate local recurrence, and a short PSA doubling time suggests metastatic disease. It has been suggested that the combination of pathologic stage, Gleason score, and PSA progression best identifies those patients at high risk developing distant metastases.

Salvage radiotherapy (RT) to the prostate bed has been used to treat patients who have a presumed or biopsy-proven local recurrence with no clinical evidence of metastatic disease. The aim of salvage RT is to eradicate presumed local recurrence in the prostatic bed and to reduce the risk of consequent progression to distant metastasis and death from prostate cancer. Salvage RT for patients with rising PSA level after RP is among the rare therapies in radiation oncology that may be administered without biopsy proof of cancer manifestation. Although retrospective series have demonstrated that salvage RT will result in biochemical response in approximately 75% of patients, long-term durable response is much lower and noted in approximately 20–40% of patients. Therefore, there is a need to identify prognostic factors that may predict those patients who will respond best to salvage RT.

Several retrospective studies have previously described the outcome of patients treated with salvage RT for biochemical relapse after radical prostatectomy for prostate cancer. Data from a prospective randomized trial are available for adjuvant RT after prostatectomy (EORTC trial 22911), but not for salvage RT after postoperative biochemical failure. To our knowledge, the study of Pisansky with 166 patients involved the largest patient cohort of a single institution. The present retrospective study analyzed the outcome of 171 patients also treated with salvage RT at a single institution.

After salvage RT, serum PSA decreased in 82.5% of the patients, which is comparable to the approximate 75% (range, 51–93%) response rate in previously published studies. These findings suggest that locally persistent disease was at least partially responsible for the serum PSA level in most patients. However, a substantial proportion of patients with an initially favorable biochemical response to RT subsequently had an increasing serum PSA profile. In the present study, 35.1% of the patients were expected to remain biochemical relapse-free 5 years after RT. In previous studies, the 5-year recurrence-free rate achieved by RT in patients with rising PSA after surgery varied between 10% and 46%

The predominant pattern of clinical relapse resulted from hematogenous dissemination: 83.3% (20 patients) of the 24 patients with clinical relapse showed distant disease relapse. Only one patient had clinical evidence of local disease relapse. In the present study, salvage RT effectively reduced the risk of local failure. However, there seemed to be patients who had occult systemic disease and did not benefit substantially from salvage RT alone.

To identify patients who may benefit most from salvage RT, the correlation between PSA progression after salvage RT and the clinical, biochemical, and pathologic parameters of the patients was analyzed. On univariate analysis, five pretherapy factors were predictive of PSA recurrence after RT: preoperative PSA level, pathologic tumor classification, Gleason score, tumor grading, and pre-RT PSA level. To our knowledge, preoperative PSA level was a statistically significant prognostic factor for biochemical relapse-free survival only in one other study. In our study, the biochemical relapse-free survival at 4 years was 46.3% for patients with preoperative PSA <10 ng/mL and 14.8% for patients with preoperative PSA between 20 and 50 ng/mL.

Other studies have also identified pathologic tumor stage as predictor of biochemical relapse after salvage RT. In the study of Pisansky, 5-year biochemical relapse-free survival rates by tumor stage T2, T3a, and T3b were 54%, 51%, and 26%, respectively. In our study, the 5-year biochemical relapse-free survival rate was 46.9%, 33.4%, and 14.2% (T3b-4), respectively.

The predictive value of the Gleason score has been pointed out in most previous reports. In the present study, the Gleason score was a statistically significant predictor in both univariate and multivariate analysis. Our study suggests that patients with Gleason scores of ≥7 are likely to have systemic disease and may not, therefore, benefit from salvage RT alone. Tumor grade (G3–4 vs. G2) also was identified as predictor of biochemical relapse after salvage RT, as shown in one other retrospective study. According to the Gleason score, this may be due to the higher risk of distant metastasis associated with high-grade prostate cancer. Our findings suggest that patients with high-grade lesions should be considered for additional treatment (e.g., hormonal therapy). In light of the results of RTOG 94-13 , which was a study of men treated primarily with RT, whole pelvis treatment in the postoperative setting is also receiving attention as additional treatment. In RTOG 94-13, the combination of whole-pelvis RT plus short-term androgen deprivation (STAD) resulted in reduced progression compared with the other three arms (prostate-only RT alone, prostate-only RT + STAD, and whole-pelvis RT alone). Because there is a risk of increased complications with whole-pelvis RT in the postoperative setting and there is as yet no proven benefit to it, such treatment is not recommended outside of a formal trial.

Preradiotherapy PSA level has been the most consistent prognostic factor in univariate and multivariate analyses . Various cutoff points have been selected, for example, 1.1 ng/mL, 1.7 ng/mL , 2 ng/mL, and 2.7 ng/mL. In the present study, the biochemical relapse-free survival at 5 years was 50.3% for patients with pre-RT PSA <1 ng/mL and 25.8% for patients with pre-RT PSA ≥1 ng/mL. The variability of the cutoff points may be related to the heterogeneity of patient cohorts. Unlike pathologic stage and grade, the risk of relapse increased in a continuous, and not stepwise, manner with serum PSA, which created difficulty in selecting an optimal cutoff point for this factor. A strict cutoff point has not been defined. In the multivariate analysis of the present study, pre-RT PSA as a continuous variable was a statistically significant independent predictor of biochemical relapse-free status after salvage RT (Hazards ratio, 1.538). Our results suggest that patients with low pre-RT PSA level may benefit most from salvage RT. Salvage RT should be given early when PSA is at a low level.

In previous studies, positive margin was predictive of better outcome after salvage RT. A positive margin indicates residual disease in the prostate bed for which salvage RT would be effective. Contrary to these studies, margin status in our study was not a statistically significant predictor for biochemical relapse-free survival. Patients with positive margin even showed an inferior biochemical relapse-free survival rate at 5 years compared with patients with negative margin (24.7% vs. 38.8%).

The aim of this study was not to analyze a possible benefit of adding hormonal therapy (HT) to salvage RT. Retrospective studies have found that using androgen suppression may improve the PSA outcome of postprostatectomy RT, especially for patients with high-risk features (i.e., pre-RT PSA >1 ng/mL, pathologic tumor stage T3b, and Gleason score ≥7). A retrospective study by King of 122 patients revealed even a statistically significant improvement in overall survival rates (100% vs. 87%) at 5 years in 53 patients who received RT and short-term androgen deprivation compared with patients who received RT alone. However, until now insufficient numbers of patients have been studied. Randomized studies of salvage RT and androgen deprivation therapy are needed. The RTOG recently completed accrual to a phase III trial (RTOG 96-01) comparing salvage RT alone vs. salvage RT plus 2 years of HT. In our study, patients who had received HT before RT surprisingly had a statistically significant lower 5-year clinical relapse-free survival rate than did patients without HT (67% vs. 87.7%). One could speculate that patients receiving HT before RT had worse clinical, biochemical, and pathologic parameters, but there was no difference in patient and tumor characteristics between patients with and without HT before RT.

Salvage RT is generally associated with a low rate of chronic morbidity. Other studies have indicated that urinary incontinence in 0–6% of cases, mild to moderate proctopathy in 0–17%, and chronic cystitis in 0–13% may result from RT in this setting. In our study, salvage RT was also tolerated without severe adverse effects: 8.2% of the patients developed mild or moderate rectal symptoms, and 7% of the patients had cystitis with transient hematuria. However, we determined toxicity solely on the basis of retrospective review of physician-reported genitourinary and gastrointestinal toxicities, which may have resulted in an underestimate of the actual complication rate.

This study represents one of the largest retrospective studies analyzing the outcome of patients treated with salvage RT for biochemical relapse after radical prostatectomy for prostate cancer. Our findings suggest that patients with Gleason score <7 (5-year biochemical relapse-free survival: 63.1%) and low pre-RT PSA level are the best candidates for salvage RT, whereas patients with high-grade lesions should be considered for additional treatment, such as hormonal therapy.