Three-Dimensional Conformal Radiation Therapy in Early Stage Prostate Cancer

AbstRact & Commentary

Source: Hanks GE, et al. Cancer J Sci Am 1999;5:152-158.

Prostate cancer is often detected on the basis of an abnormal serum prostate-specific antigen (PSA) level. Recent studies have provided some guidance regarding which of these asymptomatic cancers may represent a threat to life. Albertsen and colleagues1 have shown that the histology of the tumor, graded by the Gleason score, can identify the level of risk of dying from prostate cancer. The 15-year probability of dying from prostate cancer was 4-7% for an untreated patient with Gleason score 2-4, 18-30% for an untreated patient with Gleason score of 6, and 60-87% for an untreated patient with Gleason score 8-10.

In patients who opt for treatment, the majority of patients today undergo radical prostatectomy. While this is a highly effective management approach, the majority of individuals who undergo radical prostatectomy become impotent, and a significant fraction develop incontinence. Radiation therapy is an alternative approach to treatment. Standard external-beam radiation therapy can be effective in experienced hands; however, like surgery, radiation therapy may be associated with unpleasant treatment side effects such as radiation proctitis (diarrhea) and radiation cystitis (dysuria and frequency). The argument continues regarding the efficacy of radiation therapy; most radiation therapists believe radiation therapy is equal to surgery in the control of the cancer, while urologists usually claim that surgery is superior to radiation therapy.

The efficacy of local therapy depends on the extent of disease. In patients with locally advanced cancers, radiation therapy is often used as primary therapy. A recent clinical trial suggested that the use of hormonal therapy (goserelin), used concomitantly with the external-beam radiation therapy, resulted in improved survival compared to external-beam radiation therapy alone in patients with T2 and T3 tumors without node involvement.2

With standard treatment planning, it is difficult to deliver more than 70 Gy to the prostate bed without an unacceptable level of toxicity to the bladder and rectum. In recent years, several new techniques have been developed to deliver radiation therapy more effectively into the prostate with higher doses to the tumor bed and lower doses to the adjacent normal structures. These methods include ultrasound-guided transperineal prostate implant (IMP), intensity-modulated external-beam radiation therapy (IMRT), and three-dimensional conformal external-beam radiation therapy (3D-CRT). Data relating outcome to treatment dose in prostate cancer are limited and no prospective randomized studies with dose as the main variable have been reported. Furthermore, the new techniques for delivering higher doses of radiation therapy have not been compared to one another or to standard techniques in prospective randomized studies. However, a considerable body of data on 3D-CRT has been collected and Hanks and colleagues at Fox Chase Cancer Center have endeavored to conduct a retrospective analysis on the question of dose effects.

Hanks et al compared outcomes in two groups of patients. Group I consisted of 296 patients treated by 3D-CRT technique with doses greater than 74 Gy, and 296 patients treated by conventional or 3D-CRT techniques with doses less than 74 Gy, patients matched for stage (based on palpation), histologic grade, and PSA level (< 10 ng/mL, 10-19.9 ng/mL, ³ 20 ng/mL). Group II consisted of 357 patients treated with doses greater than 74 Gy, and 357 treated with doses less than 74 Gy matched on stage and grade, but not on PSA levels. Four outcome variables were assessed—freedom from PSA relapse, freedom from distant metastasis, cause-specific survival, and overall survival.

In group I, the high-dose group had significantly improved freedom from PSA relapse at five years (73% vs 55%), freedom from distant metastases at five years (97% vs 91%), and cause-specific survival at five years (100% vs 97%) than the low-dose group, but overall survival was not different. For group II patients, all four outcome variables were superior in the high dose group; freedom from PSA relapse at five years was 71% for high dose and 56% for low dose, freedom from distant metastases at five years was 97% for high dose and 88% for low dose, cause-specific survival at five years was 99% for high dose and 94% for low dose, and overall survival at five years was 88% for high dose and 79% for low dose. When subset analysis was performed, patients with stage T2c and T3 appeared to benefit from the higher dose radiation therapy as did those with low Gleason grade (2-6). However, patients with T1/T2ab lesions or Gleason grade 7-10 did not seem to have a better outcome with higher doses of radiation therapy.

Hanks et al interpret the data as supporting the existence of a dose-response curve to radiation therapy in prostate cancer and showing that doses greater than 74 Gy, which cannot be safely given by standard approaches, are associated with a superior outcome, including better survival, when delivered by 3D-CRT.


3D-CRT appears to permit the safe delivery of higher and more effective doses of radiation therapy to patients with prostate cancer. Although the differences seen between groups of patients, relatively well matched retrospectively who received different doses of radiation therapy, appear to support this idea, it is important to note that the study was not a prospective randomized trial. The fact that the numbers of patients in each group were identical indicates that some sort of selection of cases from a larger pool was undertaken. The basis for deciding which patients would get higher doses and which would get lower doses is not discussed and the basis for omitting patients from groups, if it happened, is not mentioned. However, other experiences with 3D-CRT are consistent with this report.

Zelefsky and colleagues from Memorial Sloan-Kettering conducted a phase I study in 743 patients with prostate cancer in which the tumor target dose was increased from 64.8 Gy to 81 Gy in 5.4 Gy increments.3 The induction of a clinical response (defined as a PSA nadir £ 1 ng/mL) was dose-dependent; 90% in those receiving 75.6-81 Gy, 76% for those treated with 70.2 Gy and, 56% for those treated with 64.8 Gy. Five-year freedom from PSA relapse was influenced by the presence or absence of poor prognostic factors. Favorable factors were stage T1-2, PSA levels less than 10 ng/mL, and Gleason score 6 or lower. Patients with all favorable factors were considered good prognosis, those with two of these features were considered intermediate prognosis, and those with one or none of these features were considered unfavorable. Patients with intermediate or unfavorable prognosis had a significantly better freedom from PSA relapse if they received 75.6 Gy or more radiation therapy. Patients underwent surveillance sextant prostate biopsies at least 2.5 years after treatment. A positive biopsy was observed in only 1 of 15 (7%) patients who received 81 Gy. Positive biopsies were obtained from 48% of those receiving 75.6 Gy, 45% of those receiving 70.2 Gy, and 57% of those receiving 64.8 Gy. Thus, these data would argue that 81 Gy is the most effective dose of radiation therapy delivered by 3D-CRT.

How does 3D-CRT compare to the other novel methods of delivering radiation therapy to the prostate? Unfortunately, we can only make rough estimates because appropriate studies have not yet been conducted. Peschel and colleagues at Yale have studied IMP in 120 patients with prostate cancer.4 Four-year freedom from PSA relapse was 80% for the favorable subset of patients, 66% for the intermediate subset, and 57% for the unfavorable subset. Those numbers are similar to the results obtained at Memorial Sloan Kettering with 3D-CRT (85% for the favorable group, 79% for the intermediate group, and 55% for the unfavorable group). However, these comparisons are crude. Insufficient numbers of patients were treated at the most effective doses of 3D-CRT to make valid comparisons to optimally delivered implant radiation therapy.

Even if the data permitted us to declare 3D-CRT the winner, it would be difficult to obtain the benefits of this approach for our patients. Hanks et al cite a personal communication with Jean Owen, who participated in a national Patterns of Care Study surveying the practice of radiation therapy in 1500 radiation oncology facilities in the United States. Data from that study dealt with patients treated in 1994. Only 19% of the facilities used 3D-CRT and among those facilities, only 27% of treated patients received more than 70 Gy and only 1.2% received as much as the 74 Gy that the Hanks data suggested was superior. Thus, despite the apparent dose-related advantages of 3D-CRT delivery, only a very small number of centers have translated these results to practice.

It would appear that improvements in techniques have resulted in the ability to safely deliver larger and more effective doses of radiation therapy to the prostate. Doses greater than 74 Gy and perhaps as high as 80 Gy are associated with excellent local control, improved freedom from PSA relapse, lower rates of metastasis, and a modest, but significant overall survival advantage. Critics might suggest that these findings require replication in the setting of a prospective randomized trial. Such a trial should be undertaken. The challenge now is to find a place where these benefits might be obtained for our patients who may not be eligible for a study due to intercurrent illness or other problems. Fox Chase Cancer Center and Memorial Sloan-Kettering Cancer Center appear to be two places where state-of-the-art practice is ongoing. Do your radiation oncology colleagues use 3D-CRT to deliver 74-80 Gy to the prostate? If they don’t, it might be instructive to learn why not. v


1. Albertsen PC, et al. JAMA 1998;280:975-980.

2. Bolla M, et al. N Engl J Med 1997;337:295-300.

3. Zelefsky MJ, et al. Int J Radiat Oncol Biol Phys 1998;41:491-500.

4. Peschel RE. Cancer J Sci Am 1999;5:145-146.

Which of the following statements best escribes the results of using three-dimensional conformal radiation therapy (3D-CRT) in patients with early stage prostate cancer?

a. Dose-response effects have been noted in which doses of more than 74 Gy appear more effective than those less than 74 Gy.

b. No dose-response effects have been noted in the range of doses that can be safely administered by this technique.

c. Prospective randomized studies have documented that 3D-CRT is more effective than radiation therapy delivered by implants.

d. Prospective randomized studies have documented that 3D-CRT is more effective than surgery.

e. 3D-CRT does not permit the delivery of higher doses of radiation therapy than well-designed external beam radiation therapy using four opposing fields.