Indications for excluding the seminal vesicles when treating clinically localized prostatic adenocarcinoma with radiotherapy alone

Abstract

The indications for treating the seminal vesicles (SV) in patients with clinically localized carcinoma of the prostate are controversial. We sought to define subgroups of patients in whom coverage could be avoided, using pretreatment prostate specific antigen (PSA) values and the Gleason score. Because the rectum is the major dose-limiting structure, we also measured the extent of rectal sparing achieved by excluding the SV from external beam treatment fields. We retrospectively studied lateral x-ray simulation films of 43 patients treated with standard four-field radiotherapy and dose-volume histograms of eight patients treated with conformal radiotherapy. The rectal surface areas were measured and the volumes were calculated including and excluding the SV. The pathology reports of patients treated with radical prostatectomy alone between 1987 and 1993 were reviewed. Patients without preoperative PSA levels or biopsy Gleason scores, or who received neoadjuvant hormonal therapy were excluded. Of the 368 remaining patients, 66 (18%) had preoperative PSA levels < or = 4, 172 (47%) had PSA levels 4-10, and 130 (35%) had PSA levels > 10. The Gleason score was < or = 6 in 269 (73%), and 99 (27%) had a score > or = 7. The reduction in the total irradiated rectal areas to full doses when the SV were excluded ranged from 5 to 67% in individual patients (median, 44%). The median reduction in the irradiated rectal volumes to 50% of the prescribed dose, as determined by dose-volume histograms, was 51% (range: 37-76%). The median reduction in bladder volumes was 9% (range: 6%-15%). The incidence of SV involvement was 19% (70 out of 368). Patients with normal PSA levels (< or = 4 ng/mL) had a 3% (2 out of 66) rate of SV involvement. Patients with PSA levels between 4-10 ng/mL had a 8% (10 out of 126) rate of SV involvement. All other cases had at least 20% rate of SV involvement, with a combined rate of 33%. Using a cutoff of 15%, two low risk groups were identified: all patients with PSA levels < or = 4, and patients with PSA 4-10 and Gleason score < or = 6. These constituted 52% (192 out of 368) of all patients. Overall, the low-risk patients had a 6% (12 out of 192) incidence of SV involvement vs. 33% (58 out of 176) for the high risk patients (p < or = .001). Excluding the SV from the treatment field can significantly reduce (40%-50%) the volume of irradiated rectum. Our data confirm that pretreatment PSA levels and Gleason scores can be effectively used to define subgroups of patients in whom SV irradiation can be avoided. We propose excluding the SV in all patients with PSA levels < or = 4, and patients with PSA levels 4-10 and a Gleason score < or = 6.