|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2000 American Society for Clinical Oncology Survival Advantage From Higher-Dose Radiation Therapy for Clinically Localized Prostate Cancer Treated on the Radiation Therapy Oncology Group TrialsFrom the Kimmel Cancer Center, Thomas Jefferson University; Radiation Therapy Oncology Group Statistical Headquarters; and Department of Radiation Oncology, Albert Einstein Medical Center, Philadelphia PA; Department of Radiation Oncology, McAuley Health Center, Ann Arbor; and Wayne State University, Detroit, MI. Address reprint requests to Richard Valicenti, MD, Bodine Center for Cancer Treatment, Thomas Jefferson University Hospital, 111 S 11th St, Philadelphia, PA 19107-5097; email richard.valicenti{at}mail.tju.edu
PURPOSE: We evaluated the effect of external-beam radiation therapy on disease-specific survival (death from causes related to prostate cancer) and overall survival in men with clinically localized prostate cancer. METHODS: From 1975 to 1992, 1,465 men with clinically localized prostate cancer received radiation therapy on four Radiation Therapy Oncology Group phase III randomized trials and were pooled for this analysis. No one received androgen-deprivation therapy with his initial treatment. All original histology had central pathologic review for grading using the Gleason classification system. Total delivered radiation dose ranged from 60 to 78 Gy (median, 68.4 Gy). The median follow-up time was 8 years. RESULTS: A Cox regression model revealed that Gleason score was an independent predictor of disease-specific survival and overall survival. The 10-year disease-specific survival rates by Gleason score were as follows: score of 2 through 5, 85%; score of 6, 79%; score of 7, 62%; and score of 8 through 10, 43%. Stratifying outcome by this important prognostic factor revealed that higher radiation dose was a significant predictor for improved disease-specific survival and overall survival only for those patients whose cancers had Gleason scores of 8 through 10 (P < .05). After adjusting for clinical T stage, nodal status, and age, treating with a higher radiation dose was associated with a 29% lower relative risk of death from prostate cancer and 27% reduced mortality rate (P < .05). CONCLUSION: These data demonstrate that higher-dose radiation therapy can significantly reduce the risk of dying from prostate cancer in men with clinically localized disease. This survival benefit is restricted to men with poorly differentiated cancers.
THE EFFICACY OF radiation therapy for clinically localized prostate cancer is controversial. Underlying the concept of an intrinsic effect for this localized therapy is the existence of a radiation dose-response relationship. Central to this concept is also the notion that improved local control can positively influence disease-specific survival and therefore reduce the risk of dying from prostate cancer. The most widely cited radiation dose-response data were obtained from the Patterns of Care studies and the patients treated at Washington University Medical Center in St. Louis.1,2 These studies demonstrated a possible relationship between the use of higher radiation doses and improved local tumor control for patients with stage T3 cancer. Recently, Hanks et al3 and several other investigators4-7 have described a dose response with conformal radiation therapy and prostate seed implantation to reduce future biochemical failure. In these series, the importance of radiation dose also seemed to apply to patients with stages T1/T2 with or without Gleason scores of 8 through 10. In an update of the Fox Chase Cancer Center experience, Hanks et al8 noted the ability of dose-escalated radiation therapy to improve survival in men with clinically localized prostate cancer. The Radiation Therapy Oncology Group (RTOG) has conducted four prospective randomized trials for clinically localized prostate cancer that included one radiation therapy–alone arm.9-12 In RTOG 77-06 and 75-06, all patients were treated with radiation therapy alone initially and were evaluated regarding the extent of the radiation treatment field. In RTOG 85-31 and 86-10, high-risk patients were randomized between radiation therapy alone initially or radiation therapy with hormonal therapy. On these trials, 1,560 patients were treated initially with external-beam radiation therapy alone and have been pooled together to create a centralized database. These patients had received radiation doses ranging from 60 to 78 Gy. In this database, we previously found that centrally reviewed Gleason score was the strongest predictor of overall and disease-specific survival.13 For these reasons and to allow for direct comparison with other series, we evaluated the relative effect of using higher radiation dose on clinical outcome by stratifying the results according to Gleason score.
Patient Population From 1975 to 1992, 1,560 men were treated initially with external-beam radiation therapy alone (no androgen-deprivation therapy) for clinically localized prostate cancer on one of four prospective phase III randomized trials. All these men were assessable and eligible for the trials. The patients received treatment on one of both arms of RTOG 77-06 and 75-06 or on the control arm of the randomized trials RTOG 86-10 or 85-31. If patients did not have central pathologic review of prostate biopsy specimens (n = 92) or radiation therapy dose was not available (n = 3), they were excluded from the analysis. This left a total of 1,465 (94%) patients. In addition to radiation dose, prognostic risk factors considered in this study included age, clinical T stage, lymph nodal status, and Gleason score. Positive nodal status (N1) had to be confirmed pathologically. Otherwise, nodal status was considered unknown (NX). Because prostate-specific antigen (PSA) testing was not used until 1988 and was thus available for relatively few patients (17%), it was not evaluated as a pretreatment variable or as a measure of treatment response.
End Points
Statistical Analysis To account for the difference in the follow-up times between early and later studies, we stratified the Cox models according to whether patients were entered in early or later studies. The studies 75-06 and 77-06 were one stratum and the later studies 85-31 and 86-10 were the other stratum. An advantage of a stratified Cox model is to estimate the risk factors separately within each level of stratifying variable and to pool the estimated hazards ratios over all strata. With this method we can account for the potential effect of studies without entering it in the model as a covariate.
The distribution of patient and tumor characteristics in the entire population and by study is listed in Table 1. The median patient age was 69 years, with a range of 45 to 87 years. The breakdown of palpable T stage was 92 (6%) in T1, 539 (35%) in T2, and 929 (59%) in T3. The distribution of Gleason score was as follows: Gleason score of 2 through 5, 208 patients (13%); Gleason score of 6, 554 (36%); Gleason score of 7, 272 (17%); Gleason score of 8 through 10, 434 (27%); and score unknown, 92 (6%). Ten percent of the patients had pathologically proven involvement of the lymph nodes, although given the pretreatment characteristics, the true incidence was probably higher.
Statistical Distribution of Radiation Doses The distribution of patients according to delivered radiation dose is given in Fig 1. In concordance with treatment protocol and the International Commission on Radiation Units and Measurement,16 dose was reported at the center of the prostate. The median radiation dose was 68.4 Gy, ranging from 10.8 to 77.7 Gy. The 25th and 75th percentile doses were at 66 Gy and 70.2 Gy, respectively.
Table 2 lists the results of the radiation therapy quality assurance review for the 1,465 patients. The variation in total radiation dose is attributed in part to the allowable prescribed dose on the four trials (65 to 72 Gy). In addition, a deviation of 5% or less in prescribed dose was considered as per treatment protocol. A total of 1,399 patients (95%) met these criteria. Larger variations were in minor (n = 35) or major (n = 15) protocol violation if dose deviated by 5% to 10% or 10% to 20%, respectively. Only 16 patients (1%) were potentially in noncompliance with treatment protocol.
Analysis of Radiation Dose Effect Patients who had cancers with Gleason scores of 2 to 5, 6, 7, and 8 to 10 had an 85%, 79%, 62%, and 43% chance, respectively, of surviving prostate cancer at 10 years from date of study entry (Fig 2). In a previous study, Gleason score was the single most important prognostic factor to predict for the risk of dying from prostate cancer.13 For this reason, an analysis of radiation dose response was based on stratifying outcome according to this factor. For this analysis, patients with a Gleason score of 2 to 5 and a Gleason score of 6 were combined, because the 10-year disease-specific survival for these groups did not differ significantly. We selected 25th, 50th, and 75th percentile doses of the dose distribution as the cutoff values for our analysis. Therefore, the hypotheses, based on the selection, were independent of the outcome of these groups of patients.
On univariate analysis, the 25th percentile radiation dose revealed a significant effect on disease-specific survival and overall survival (Figs 3 and 4) and a marginally significant effect on local progression (Fig 5). This benefit was restricted to men with cancers having Gleason scores of 8 to 10. No other dose cutoff for the other Gleason score groupings had a significant effect on any of the end points. At 10 years, the overall disease-specific survival rate in patients with cancers having Gleason scores of 8 to 10 receiving more than 66.0 Gy was 46%, compared with 31% for those receiving smaller doses (P = .041). Patients receiving larger radiation doses had a 10-year overall survival rate of 27%, compared with 16% for those receiving smaller doses (P = .012).
Using the Cox model, multivariate analysis of factors associated with local progression, disease-specific survival, and overall survival showed age, tumor stage, nodal status, and Gleason score to be independently significant (Table 3). The model stratified according to the length of follow-up time based on the study in which the patients were randomized. The model also identified statistical interactions between Gleason score and the radiation dose groups. The only significant interaction was the interaction between Gleason score of 8 to 10 and radiation dose  66 Gy versus more than 66 Gy. After adjustment for age and stage, treating with radiation dose more of than 66 Gy was associated with a 29% lower relative risk of death from prostate cancer (P = .043) for men with cancers having Gleason scores of 8 to 10. This corresponded with a 37% reduction in local progression (P = .018).
We evaluated the data for potential bias in favor of men receiving higher radiation doses. One potential source of bias was that the use of a lower radiation dose was associated with poor patient compliance or lower expected survival. To control for this factor, we carried out the same analysis for the 1,399 patients who received radiation therapy as per treatment protocol. For these patients, we found the same significant effect for using radiation doses greater than 66.0 Gy on all three clinical end points (Table 3).
We were able to determine the effect of high-dose external-beam radiation therapy on the risk of dying from prostate cancer by stratifying clinical outcome by centrally reviewed Gleason score. In men with clinically localized prostate cancer with Gleason scores of 8 to 10, the use of higher radiation doses significantly reduced the likelihood of local progression, disease-specific death, and overall mortality. This study represents the largest multi-institutional cohort of patients treated initially with external-beam radiation therapy alone for which long-term follow-up was available. Our data strongly suggest that both treatment and histologic factors were important determinants of a patient’s chance of survival at 10 to 15 years after radiation therapy. In 1999, the American Cancer Society estimates approximately 37,000 deaths resulting from prostate cancer in the United States.17 Until now, there have been few reliable data indicating that local therapy can positively influence the natural history of disease progression. We found an overall survival and disease-specific survival benefit from the use of higher doses of external-beam radiation therapy restricted to men with high-grade tumors. Although it has been previously suggested that this group was less likely to benefit because of the presence of occult micrometastatic disease, our results suggest improved outcome when comparisons were made with another series.18 Despite the majority of men having T3 cancers, we found a relatively high 10-year disease-specific survival of 43% for cancers with Gleason scores of 8 to 10, with a rate of 46% for doses greater than 66.0 Gy. Albertsen et al19 published a 15-year analysis of a population-based cohort of 767 men managed conservatively for localized prostate cancer. The original biopsy specimens were reviewed for grading using the Gleason classification system. Approximately 10% of the men had cancers with Gleason scores of 8 to 10. Depending on age, conservative management with no initial treatment or treatment with immediate or delayed hormonal therapy led to a 15-year disease-specific survival rate between 13% to 40% for these men. Aus et al20 published a retrospective analysis of 301 Swedish men who initially had localized prostate cancer. In their study there was also a strong influence of tumor grade on cause-specific survival. High-grade tumors were clearly the most malignant, with a high mortality rate of 60% and a low 10-year cause-specific survival rate of 19%. In a pooled analysis of 828 men treated conservatively, Chodak et al18 found a 34% 10-year disease-specific survival rate for men with a similar Gleason grouping, but with the majority having T1/T2 stage disease. The results of this study support two key hypotheses regarding the use of radiation therapy for the treatment of localized prostate cancer. First, increasing radiation dose can lead to a significant improvement in freedom from local progression. We found that doses greater than 66 Gy resulted in as much as a 37% reduction in local progression. In this study we defined local progression as either an increase of more than 50% in tumor size (cross-sectional area), recurrence of palpable tumor after initial clearance, or prostate biopsy showing adenocarcinoma of the prostate 2 years or more after study entry. Because the dose cutoff represented the 25th percentile dose, the higher-dose group had a median dose of 69 Gy as compared with 64 Gy for the lower-dose group. This represents an average of a 6.2% increase in dose between the two groups. This difference wouldn’t be expected to lead to an observable effect unless there was a relatively steep local control probability curve and may explain in part an observable effect restricted to poorly differentiated prostate cancers. These results are consistent with a subgroup analysis of the only completed phase III trial evaluating high versus conventional external-beam radiation therapy for T3/T4 prostate cancer.21 With an increase of 12.5% in dose, Shipley et al21 found significant effect on local control in men with poorly differentiated cancer. In their study, higher doses led to an 8-year local control rate of 84% as compared with 19% for lower doses, but neither a recurrence-free survival nor overall survival benefit was identified. This finding is in contrast with our study, in which we also found a significant improvement in both disease-specific survival and overall survival. Although high-dose radiation therapy was not administered with a proton boost, our result supports the second hypothesis that improvement in local tumor control may translate into a significant reduction in the relative risk of prostate cancer death. To date, we found no evidence of a dose effect for well-differentiated or moderately differentiated cancer. In patients with lower-grade tumors, there may be a shallower slope in the dose-response function, perhaps because of a greater clonal heterogeneity in radiosensitivity. Such tumors may be controlled with modest doses or may require follow-up periods of greater than 15 years for a dose response to manifest itself. In addition, stratification of outcome according to pretreatment PSA may be more important in this group than for poorly differentiated disease to optimally evaluate treatment efficacy.22 In this pooled analysis, there are several limitations. The patients were accrued to the studies over a 20-year time period. During this time, PSA testing became available for routine clinical use. In addition, these studies spanned the introduction of newer imaging modalities, such as transrectal ultrasound, computed tomography, and magnetic resonance imaging. As yet, none of these technologies have demonstrated benefit regarding long-term clinical outcome in men treated for localized prostate cancer. Nevertheless, many men with regional or metastatic prostate cancer were included in our analysis as indicated by 58% of the men requiring androgen deprivation at 10 years. Although the patients were followed-up prospectively, the indication for the use of varying radiation doses or the initiation of salvage androgen deprivation was occasionally not clearly ascertainable because of incomplete or absent records. To control for these uncertainties, we evaluated the subset of patients who received radiation therapy as per treatment protocol, and we found agreement with our initial results regarding the dose effect on all three clinical end points evaluated. Finally, as with any other pooled analysis, whether based on prospective or retrospective data, it is unclear how best to evaluate all selection biases that may influence the overall analysis. In summary, centrally reviewed biopsy Gleason score is a strong and independent predictor of the risk of dying from prostate cancer. This information can provide reliable estimates of outcome for men treated initially with external-beam radiation therapy that can be compared directly with estimates of other management approaches. Because we stratified outcome according to Gleason score, we were able to determine the effect of using higher radiation doses on reducing the risk of death from prostate cancer. Because external-beam radiation therapy seems to significantly lower rates of local progression, disease-specific survival, and mortality for men with cancers having Gleason scores of 8 to 10, a randomized trial of dose-escalated radiation therapy for groups of similar men is warranted.
Supported by Radiation Therapy Oncology Group grant nos. U10 CA21661, CCOP U10 CA37422, and Stat U10 CA32115 from the National Cancer Institute, Bethesda, MD.
1. Perez CA, Walz BJ, Zivnuska FR, et al: Irradiation of carcinoma of the prostate localized to the pelvis: Analysis of tumor response and prognosis. Int J Radiat Oncol Biol Phys 6:555-563, 1980[Medline] 2. Hanks GE, Martz KL, Diamond JJ: The effect of dose on local control of prostate cancer. Int J Radiat Oncol Biol Phys 15:1299-1305, 1988[Medline] 3. Hanks G, Hanlon A, Schultheiss T, et al: Dose escalation with 3D conformal treatment: Five year outcomes, treatment optimization, and future directions. Int J Radiat Oncol Biol Phys 41:501-510, 1998[Medline] 4. Stock RG, Stone NN, Tabert A, et al: A dose response study for I-125 prostate implants. Int J Radiat Oncol Biol Phys 41:101-108, 1998[Medline] 5. Pollack A, Zagars G: External beam radiotherapy dose response of prostate cancer. Int J Radiat Oncol Biol Phys 39:1101-1018, 1997 6. Zelefsky MJ, Leibel SA, Gaudin PB, et al: Dose escalation with three-dimensional conformal radiation therapy affects the outcome in prostate cancer. Int J Radiat Oncol Biol Phys 41:491-500, 1998[Medline] 7. Roach M, Meehan S, Kroll S, et al: Radiotherapy for high grade clinically localized adenocarcinoma of the prostate. J Urol 156:1719-1723, 1996[Medline] 8. Hanks GE, Hanlon AL, Pinover WH, et al: Survival advantage for prostate cancer patients treated with high-dose three-dimensional conformal radiotherapy. Cancer J Sci Am 5:152-158, 1999[Medline] 9. Asbell SO, Krall JM, Pilepich MV, et al: Elective pelvic irradiation in stage A2, B carcinoma of the prostate: Analysis of RTOG 77-06. Int J Radiat Oncol Biol Phys 15:1307-1316, 1988[Medline] 10. Pilepich MV, Krall JM Sause WT, et al: Prognostic factors in carcinoma of the prostate: Analysis of RTOG study 75-06. Int J Radiat Oncol Biol Phys 13:339-349, 1987[Medline] 11. Pilepich MV, Krall JM, Al-Sarraf M, et al: Androgen deprivation with radiation therapy compared with radiation therapy alone for locally advanced prostatic carcinoma: A randomized comparative trial of the Radiation Therapy Oncology Group. Urology 45:616-623, 1995[Medline]
12.
Pilepich MV, Caplan R, Byhardt RW, et al: Phase III trial of androgen suppression using goserelin in unfavorable-prognosis carcinoma of the prostate treated with definitive radiotherapy: Report of Radiation Oncology Group Protocol 85-31. J Clin Oncol 15:1013-1021, 1997 13. Roach M, Lu J, Pilepich M, et al: Long-term survival after radiotherapy alone: Radiation therapy oncology group prostate cancer trials. J Urol 161:864-868, 1999[Medline] 14. Cox DR: Regression models and life tables. J R Stat Soc B 34:187-220, 1972 15. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958 16. International Commission on Radiation Units and Measurements:ICRU Report No. 50: Prescribing, Recording, and Reporting Photon Beam Therapy. Washington, DC,International Commission on Radiation Units and Measurements, 1993
17.
Landis SH, Murray T, Bolden S, et al: Cancer Statistics, 1999. CA Cancer J Clin 49:8-31, 1999
18.
Chodak GW, Thisted RA, Gerber GS, et al: Results of conservation management of clinically localized prostate cancer. N Engl J Med 330:242-248, 1994
19.
Albertsen PC, Hanley JA, Gleason DF, et al: Competing risk analysis of men aged 55 to 74 years to diagnosis managed conservatively for clinically localized prostate cancer. JAMA 280:975-980, 1998 20. Aus G, Hugosson J, Norlen L: Long-term survival and mortality in prostate cancer treated with noncurative intent. J Urol 154:460-465, 1995[Medline] 21. Shipley WU, Verhey LJ, Munzenrider JE, et al: Advanced prostate cancer: The results of a randomized comparative trial of high dose irradiation boosting with conformal protons compared with conventional dose irradiation using photons alone. Int J Radiat Oncol Biol Phys 32:3-12, 1995[Medline]
22.
Lee WR, Hanks GE, Hanlon An increasing prostate-specific antigen profile following definitive radiation therapy for localized prostate cancer: Clinical observations. J Clin Oncol 15:230-238, 1996 Submitted October 6, 1999; accepted March 16, 2000.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2000 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
