This study updates technical principles and results of 3-dimensional conformal radiation therapy (3D-CRT) in localized carcinoma of the prostate. Between January 1992 and December 1999, 312 patients were treated with 3D-CRT and 135 patients were treated with bilateral arcs standard radiation therapy (SRT) alone for clinical stage T1b-c or T2 histologically confirmed prostate cancer. None of these patients received hormonal therapy. Mean follow-up for patients in the 3D-CRT group was 3.2 years (range, 2-5.9 years) and for SRT patients, 4.7 years (range, 4-7 years). For 3D-CRT, 7 intersecting fields were used (cerrobend blocking or multileaf collimation) to deliver 68-74 Gy to the prostate. Standard radiation therapy consisted of bilateral 120 degree rotational arcs, with portals using 2-cm margins around the prostate to deliver 68-70 Gy to the prostate. The criterion for chemical disease-free survival was a postirradiation prostate-specific antigen (PSA) value following the American Society for Therapeutic Radiology and Oncology guidelines. Symptoms during treatment were quantitated weekly, and late effects were assessed every 4-6 months. Dose-volume histograms showed a two-thirds reduction with 3D-CRT in normal bladder or rectum receiving > or = 70 Gy with 3D-CRT. Higher 5-year chemical disease-free survival was observed with 3D-CRT (75%; for T1b-c and 79%; for T2 tumors) compared with SRT (61% and 65%, P = 0.01 and P = 0.12, respectively). There was no statistically significant difference in chemical disease-free survival in patients with Gleason score of < or = 4 (P = 0.85), but, with Gleason score of 5-7, the 5-year survival rates were 83% with 3D-CRT and 59% with SRT (P < or = 0.01). In 245 patients with pretreatment PSA of < or = 10 ng/mL treated with 3D-CRT, the chemical disease-free rate was 80% versus 72% in 98 patients treated with SRT (P = 0.21). In patients with PSA of 10.1-20 ng/mL, the chemical disease-free survival rate for 50 patients treated with 3D-CRT was 71% compared with 43% for 20 patients treated with SRT (P = 0.02). The corresponding values were 59% and 16%, respectively, for patients with PSA levels > 20 ng/mL (P = 0.09). On multivariate analysis, the most important prognostic factors for chemical failure were pretreatment PSA (P = 0.004), nadir PSA (P = 0.001), and 3D-CRT technique (P = 0.012). Moderate dysuria was reported by 2%-5% of patients treated with 3D-CRT in contrast to 6%-9% of patients treated with SRT. The incidence of moderate loose stools or diarrhea, usually after the fourth week of treatment, was 3%-5% in the 3D-CRT patients and 8%-19% in the SRT group. Late intestinal grade 2 morbidity (proctitis or rectal bleeding) was 1% in the 3D-CRT group in contrast to 7% in SRT patients. The 3D-CRT spares more normal tissues, yields higher chemical disease-free survival, and results in less treatment morbidity than SRT in treatment of stage T1-T2 prostate cancer. Follow-up at > or = 10 years is needed to confirm these observations.