|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2002 American Society for Clinical Oncology Determinants of Prostate Cancer–Specific Survival After Radiation Therapy for Patients With Clinically Localized Prostate CancerByFrom the Departments of Radiation Oncology and Pathology, Brigham and Women’s Hospital and Dana-Farber Cancer Institute, Boston, MA, and Department of Mathematics, Millersville University, Millersville, PA. Address reprint requests to Anthony V. D’Amico, MD, PhD, Department of Radiation Oncology, Brigham and Women’s Hospital, 75 Francis St, L-2 Level, Boston, MA 02215; email: adamico{at}lroc.harvard.edu
PURPOSE: Identifying pretreatment and posttreatment predictors of time to prostate cancer–specific death (PCSD) after external-beam radiation therapy (RT) was the subject of this study. PATIENTS AND METHODS: A Cox regression analysis was used to evaluate the ability of the pretreatment risk group to predict time to PCSD for 381 patients who underwent RT for clinically localized prostate cancer. Posttreatment factors analyzed for the 94 patients who experienced prostate-specific antigen (PSA) failure included the time to PSA failure, the posttreatment PSA doubling time (DT), and the timing of salvage hormonal therapy.
RESULTS: Despite the median age of 73 years at diagnosis, 45% of patients with high-risk disease were estimated to die from prostate cancer within 10 years after RT compared with 0% (P = .004) and 6% (P = .05) for patients with low- or intermediate-risk disease, respectively. Predictors of time to PCSD after PSA failure included PSA DT (P = .01) and delayed use of hormonal therapy (P CONCLUSION: Prostate cancer was a major cause of death during the first decade after RT for patients with clinically localized but high-risk disease, and the cause of death for patients with a short PSA DT after RT was nearly always prostate cancer. These data provide evidence to propose the hypothesis that a short posttreatment PSA DT may serve as a possible surrogate for PCSD. Prospective validation is needed.
PROSTATE-SPECIFIC antigen (PSA) failure after either radical prostatectomy (RP)1 or external-beam radiation therapy (RT)2 for patients with clinically localized prostate cancer occurs in approximately 30% to 50% of cases within 10 years after treatment and is a great source of anxiety for both patients and physicians. Therefore, in an attempt to identify patients at high risk of failure after RP or RT, investigators have developed pretreatment risk groups3 and nomograms4 on the basis of the relative values of time to posttreatment PSA failure after RP or RT. Yet, for whom PSA failure predicts death from prostate cancer remains unanswered. The inability to answer this question has been attributed to inadequate power in databases because of relatively short follow-up and the protracted clinical course of prostate cancer and the competing causes of mortality in this patient population.5 A second issue related to PSA failure after RT that has had significant ramifications on the patient’s quality of life and the overall cost to the health care system is whether survival is prolonged when salvage hormonal therapy is initiated at or after PSA failure at a time when the bone scan is negative as compared with positive. Unfortunately, because of patient and physician views on this matter, a randomized trial of early versus delayed hormonal therapy after PSA failure has not been successfully completed. Therefore, this study had two goals. The first goal was to determine whether pretreatment risk groups3 that have been shown to stratify patients by time to posttreatment PSA failure could also stratify patients by time to posttreatment prostate cancer–specific death (PCSD). The second goal was to evaluate whether posttreatment factors could predict for time to PCSD after PSA failure. Realizing that previous investigators have shown that both the time interval to PSA failure after RP6 and the PSA DT after RP6 or RT7 were predictors of time to distant failure, these factors in addition to the timing of salvage hormonal therapy were included in our analysis.
Patient Selection Three hundred eighty-one patients with a diagnosis of clinically localized prostate cancer and treated with external-beam RT by a single physician group at a Harvard-associated community outreach facility (St Anne’s Hospital, Fall River, MA) from 1987 to 2000 constituted the study cohort. The median age of the patient population at the time of initial therapy was 73 years (range, 49 to 86 years). The pretreatment clinical characteristics of the entire study cohort and the 94 patients (25%) who sustained PSA failure are listed in Table 1.
Staging, Treatment, and Follow-Up In all cases, staging evaluation involved a history and physical examination including a digital rectal examination (DRE), serum PSA, and a transrectal ultrasound-guided needle biopsy of the prostate with Gleason score histologic grading.8 The prostate biopsy was performed using an 18-gauge Tru-Cut needle (Travenol Laboratories, Deerfield, IL) through a transrectal approach. All biopsy material was reviewed and assigned a primary and secondary Gleason grade by a single genitourinary pathologist (A.A.R.). Before 1996, all patients had a computed tomographic scan of the pelvis and bone scan. After 1996, patients with both a pretreatment PSA level less than 10 ng/mL and a biopsy Gleason score of  6 did not undergo radiologic staging because of the less than 1% chance that these studies would reveal metastatic disease.9 The clinical stage was obtained from the DRE findings using the 1992 American Joint Committee on Cancer staging system.10 Radiologic and biopsy information were not used to determine clinical stage. The PSA measurement was obtained on an ambulatory basis within 6 weeks of the start of RT and before radiologic studies and biopsy. All PSA measurements were made using the Hybritech (San Diego, CA), Tosoh (Foster City, CA), or Abbott (Chicago, IL) assays. PSA values before 1989 were obtained as part of an American Cancer Society–sponsored screening program.
The treatment performed was conformal RT starting in 1994. Conventional RT was performed before 1994. However, a randomized trial of conventional versus conformal RT11 has not shown a difference in cancer control outcomes. The total median dose delivered to the prostate was 70.4 Gy (range, 69.3 to 70.4 Gy) after using a 95% normalization. A first course of RT to the prostate and seminal vesicles was prescribed for patients with either a PSA level more than 10 ng/mL or a biopsy Gleason score The median follow-up for the entire study cohort of 381 patients was 4 years (range, 0.5 to 13 years) using the first day of RT as time 0. The median time interval from diagnosis to start of RT was 2 months (range, 1.5 to 3.5 months). Before PSA failure, which was defined using the American Society for Therapeutic Radiology and Oncology consensus criteria,12 all patients generally had a serum PSA measurement and DRE performed every 3 months after RT for 2 years, then every 6 months for 3 additional years, and then annually thereafter. After PSA failure, patients were followed in rotation among radiation, medical oncology, and urology on an every-3- to 6-month basis until death. The median follow-up defining the date of PSA failure as time 0 for the 94 patients who have experienced PSA failure was 2.9 years (range, 0.5 to 9.8 years). No patient was lost to follow-up and there have been 54 deaths, 20 of which were from prostate cancer.
Determination of the Cause of Death
Salvage Hormonal Therapy
Statistical Methods
The pretreatment risk group and the timing of salvage hormonal therapy were treated as categorical variables, whereas the time interval to PSA failure and the PSA DT were treated first as continuous and then as categorical variables in separate Cox regression analyses. The categories selected for the time interval to PSA failure and the PSA DT were 2 years and 12 months, respectively. These times were selected for the purpose of illustration and because, on the basis of the results of previous studies,6,7,14 these values were suggested to be clinically useful breakpoints for predicting time to distant failure. The PSA DT was calculated assuming first-order kinetics and using a minimum of three PSA values each separated in time by a minimum of 3 months. Timing of salvage hormonal therapy was categorized as being initiated at a PSA level of The relative risk of PCSD and all-cause death were calculated for patients on the basis of the coefficients from the Cox regression model and reported with 95% confidence intervals (CIs). For the purpose of illustration, estimates of prostate cancer–specific survival (PCSS) and overall survival (OS) were determined using the actuarial method of Kaplan and Meier15 and were graphically displayed. Comparisons of survivorship between groups were made using the log-rank test and an adjustment for multiple comparisons was made using the methodology of Bonferonni.13 In order to avoid the potential for overestimating cause-specific death using the method of Kaplan and Meier15 given the competing causes of mortality in this patient cohort,16 the cumulative incidence method was also applied to calculate this end point in cases where PCSD and all-cause death were compared.
Pretreatment Prognostic Factors The median age and follow-up for patients in the low-, intermediate-, and high-risk groups were 73, 73, and 73 years and 3.9, 3.8, and 4.2 years, respectively. The results of the Cox regression time to PSA failure analysis evaluating the pretreatment risk groups are listed in Table 2 and 10-year estimates of PCSS (100% v 94% v 55%; P = .005) and OS (89% v 79% v 39%; P = .03) are illustrated in Figs 1 and 2. The relative risk (RR) of death caused by prostate cancer was 6 (95% CI, 2.5 to 10) for high-risk compared with low- or intermediate-risk patients. There was no significant difference (all pairwise values of P > .54) in the estimates of non-PCSS, being 12% versus 17% versus 27% at 10 years when stratified by the low, intermediate, and high pretreatment risk groups, respectively, as noted in Fig 3.
Posttreatment Prognostic Factors The median age and follow-up beyond PSA failure for patients in the PSA DT 12 versus more than 12 months groups were 73 and 72 years and 2.9 and 2.9 years, respectively. Similarly, the median age and follow-up beyond PSA failure for patients who received salvage hormonal therapy at the time of a negative bone scan and PSA of 10 ng/mL or greater than 10 ng/mL versus a positive bone scan were 73, 72, and 72 years and 2.9, 3.2, and 2.9 years, respectively. PSA DT as a continuous variable and the initiation of salvage hormonal therapy at the time of a positive bone scan were significant independent predictors of both time to death from prostate cancer and any cause as listed in Table 2 and illustrated in Figs 4 through 7. The lack of significance for the time interval to PSA failure on multivariable analysis can be explained by the high degree of concordance between a short time interval to PSA failure ( 2 years) and a short PSA DT ( 12 months).
Twenty patients (17 high-risk and three intermediate-risk) have died of prostate cancer and five from other causes among the 94 patients who experienced PSA failure. Of the 57 patients with a PSA DT 12 months, 18 have died, 17 (94%) from prostate cancer (15 high-risk and two intermediate-risk); whereas for the 37 patients with a PSA DT more than 12 months, seven have died, three (44%) from prostate cancer (two high-risk and one intermediate-risk). For men with a short PSA DT (ie, 12 months), estimates of PCSD and all-cause death after PSA failure were nearly identical, as shown in Figs 4 and 5, respectively. This similarity in estimates of PCSD and all-cause death remained unchanged when the cumulative incidence method16 was used to estimate PCSS in patients with a short PSA DT (ie, 12 months). Specifically, the 5-year estimate of PCSD after PSA failure was 52% and 49% using the Kaplan-Meier15 and cumulative incidence16 methods, respectively, both of which closely approximated the 53% 5-year estimate for all-cause death. The RR of PCSD and all-cause death was 5.1 (95% CI, 2 to 8.9; P = .03) and 2.2 (95% CI, 1.2 to 4; P = .05) for patients, respectively, with a PSA DT 12 compared with more than 12 months.
Patients who began salvage hormonal therapy when the bone scan was positive versus negative had an RR of 12 (95% CI, 6.2 to 18; P = .0006) and 9.1 (95% CI, 4 to 14.3; P = .0001) for PCSD and all-cause death, respectively. After correcting for multiple comparisons, there was a near significant increase in PCSS (P = .03) but not in OS (P = .51) if salvage hormonal therapy was initiated when the bone scan was negative with a PSA
To evaluate whether an imbalance in prognostic factors could have contributed to the survival differences displayed in Figs 6 through 9, the distribution of the known prognostic factors were compared for patients who initiated salvage hormonal therapy when the bone scan was positive as compared with negative. As shown in Table 3, there were no imbalances noted in the pretreatment risk group (P = .97) or the posttreatment PSA DT (P = .73) distributions among patients who received salvage hormonal therapy when the bone scan was positive versus negative.
The widespread use of monitoring serum PSA after treatment for patients with clinically localized prostate cancer has been the basis for the generation of pretreatment risk groups3 and nomograms4 that provide the probability of being free from PSA failure after RP or RT. These tools have enabled the identification of men at high risk for PSA recurrence on the basis of pretreatment parameters,3,4 posttreatment17 parameters, or both.18 However, it remains unclear as to whether patients who sustain PSA failure after primary therapy die of prostate cancer compared with other causes. Therefore, this study was performed to identify the determinants of PCSD on the basis of pretreatment and posttreatment predictors for men who underwent RT for clinically localized prostate cancer diagnosed during the PSA era. The results of the study disclosed that 45% of patients with high-risk disease were estimated to have died of prostate cancer within 10 years after RT compared with 0% (P = .004) and 6% (P = .05) for patients with low- or intermediate-risk disease, respectively. Evaluating this finding in conjunction with Fig 3, where the 10-year estimate of non-PCSD was 27% for patients with high-risk disease, revealed that high-risk prostate cancer was the leading cause of mortality. This observation is particularly important, given the competing causes of mortality expected in men whose diagnosis was made at a median age of 73. Whether the addition of concurrent and adjuvant androgen suppression therapy to RT for patients with high-risk but clinically localized disease will prolong survival, as has been shown for men with locally advanced prostate cancer,19 awaits further follow-up of completed randomized trials.
The results of the multivariable analysis of posttreatment factors showed that the PSA DT when evaluated as a continuous variable was a significant predictor of both time to PCSD and all-cause death. Specifically, the study revealed that patients with a short PSA DT ( To further support the potential prognostic significance of the PSA DT after primary local therapy, there are now several reports from patients managed both surgically6,20,21 and with radiation7,14,22 suggesting that a rapid posttreatment PSA DT (6 to 12 months) is a significant predictor of time to distant failure after PSA failure. In addition, one of the radiation studies also found the PSA DT to be predictive of time to PCSD.14 Specifically, they estimated the 5-year PCSD to be 52% versus 10% (P = .007) for patients with a posttreatment PSA DT of approximately 1 year or less compared with greater than 1 year, respectively, similar to the results in the current study. Further follow-up of the published studies associating the posttreatment PSA DT with time to distant failure6,7,14,20-22 will enhance our understanding of the potential prognostic significance of the posttreatment PSA DT. To date, there has been no randomized study that has evaluated whether a difference in survival exists for the use of early as opposed to delayed salvage hormonal therapy after PSA failure after primary RT. There is evidence, however, from randomized studies that support a survival benefit to adjuvant as opposed to salvage hormonal therapy in patients with locally advanced19 and metastatic23 prostate cancer treated using RT and node-positive24 prostate cancer managed with RP. Although these adjuvant therapy trials did not specifically address the question of salvage hormonal therapy, they provided the basis for the hypothesis that a prolongation in survival may be possible for early as opposed to delayed initiation of salvage hormonal therapy for patients who have experienced PSA failure after primary local therapy. Therefore, in evaluating the determinants of PCSD and all-cause death after PSA failure in this study, the question of whether the timing of salvage hormonal therapy impacted on PCSS and OS was also addressed. The final finding in this study was a prolongation in both PCSS and OS for early (any PSA level and a negative bone scan) as opposed to delayed (any PSA level and a positive bone scan) initiation of salvage hormonal therapy for patients who had sustained PSA failure after RT. Although these data are retrospective, there were no imbalances measured in the proportion of patients within each pretreatment risk group and posttreatment PSA DT cohort for patients who began salvage hormonal therapy when the bone scan was positive versus negative, as listed in Table 3. However, a retrospective study cannot control for unknown prognostic factors, and the sample size of patients with a positive bone scan is small (n = 10). Therefore, this result awaits validation from a prospective randomized trial. Beyond the timing of the initiation of salvage hormonal therapy, other unanswered issues remain regarding salvage hormonal therapy that are not addressed in this study and include type (luteinizing hormone-releasing hormone antagonist or orchiectomy with or without a nonsteroidal antiandrogen) and duration (intermittent v continuous).
In summary, despite a median age of 73 at diagnosis, prostate cancer was a major cause of death during the first decade after RT for patients with clinically localized but high-risk disease. Moreover, the cause of death in patients with a short PSA DT (
We thank Sidney Feldman, MD, an outstanding urologist, whose love for his family, patients, and others provided the inspiration for this study. Although he was taken early from this life, the love he shared with those fortunate enough to have known him is endless.
1. Han M, Lai S, Partin AW, et al: Biochemical (PSA): Recurrence probability following radical retropubic prostatectomy for clinically localized prostate cancer. J Urol 165: 149, 2001 (abstr 610)[CrossRef][Medline]
2. Shipley WU, Thames HD, Sandler HM, et al: Radiation therapy for clinically localized prostate cancer: A multi-institutional pooled analysis. JAMA 281: 1598-1604, 1999
3. D’Amico AV, Whittington R, Malkowicz SB, et al: Biochemical outcome after radical prostatectomy, external beam radiation therapy, or interstitial radiation therapy for clinically localized prostate cancer. JAMA 280: 969-974, 1998
4. Kattan MW, Easthan JA, Stapleton AMF, et al: A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst 90: 766-771, 1998
5. Albertsen PC, Hanley JA, Gleason DF, et al: Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer. JAMA 280: 975-980, 1998
6. Pound CR, Partin AW, Eisenberger MA, et al: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 281: 1591-1597, 1999
7. Lee WR, Hanks GE, Hanlon A: Increasing prostate-specific antigen profile following definitive radiation therapy for localized prostate cancer: Clinical observations. J Clin Oncol 15: 230-238, 1997 8. Gleason DF and the Veterans Administration Cooperative Urological Research Group: Histologic grading and staging of prostatic carcinoma, in Tannenbaum M (ed): Urologic Pathology. Philadelphia, PA, Lea & Febiger, 1977, pp 171-187 9. Lee CT, Oesterling JE: Using prostate-specific antigen to eliminate the staging radionuclide bone scan. Urol Clin North Am 24: 389-394, 1997[CrossRef][Medline] 10. Beahrs OH, Henson DE, Hutter RVP: American Joint Committee on Cancer, Manual for Staging of Cancer, ed 4 . Philadelphia, PA, Lippincott, 1992 11. Dearnaley D, Khoo V, Norman A, et al: Comparison of radiation side effects of conformal and conventional radiotherapy in prostate cancer: A randomized trial. Lancet 353: 267-272, 1999[CrossRef][Medline] 12. Cox JD, for the American Society for Therapeutic Radiology and Oncology Consensus Panel: Consensus statement: Guidelines for PSA following radiation therapy. Int J Radiat Oncol Biol Phys 37: 1035-1041, 1997[CrossRef][Medline] 13. Neter J, Wasserman W, Kutner M (eds): Simultaneous inferences and other topic in regression analysis-1, in Applied Linear Regression Models. Homewood, IL, Richard D. Irwin, Inc, 1983, pp 150-153 14. Sandler HM, Dunn RL, McLaughlin PW, et al: Overall survival after prostate-specific-antigen-detected recurrence following conformal radiation therapy. Int J Radiat Oncol Biol Phys 48: 629-633, 2000[CrossRef][Medline] 15. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53: 457-481, 1958[CrossRef] 16. Gaynor JJ, Feur EJ, Tan CC, et al: On the use of cause-specific failure and conditional failure probabilities: Examples from clinical oncology data. J Am Stat Assoc 88: 400-409, 1993[CrossRef]
17. Kattan MW, Wheeler TM, Scardino PT: Postoperative nomogram for disease recurrence after radical prostatectomy for prostate cancer. J Clin Oncol 17: 1499-1507, 1999
18. D’Amico AV, Whittington R, Malkowicz SB, et al: Utilizing predictions of early prostate-specific antigen failure to optimize patient selection for adjuvant therapy trials. J Clin Oncol 18: 3240-3246, 2000
19. Bolla M, Gonsalez D, Warde P, et al: Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. N Engl J Med 337: 295-300, 1997 20. Patel A, Dorey F, Franklin J, et al: Recurrence patterns after radical retropubic prostatectomy: Clinical usefulness of prostate specific antigen doubling times and log slope prostate specific antigen. J Urol 158: 1441-1445, 1997[CrossRef][Medline] 21. Roberts SG, Blute ML, Bergstralh EJ, et al: PSA doubling time as a predictor of clinical progression after biochemical failure following radical prostatectomy for prostate cancer. Mayo Clin Proc 76: 576-581, 2001[Medline] 22. Sartor CI, Strawderman MH, Lin XH, et al: Rate of PSA rise predicts metastatic versus local recurrence after definitive radiotherapy. Int J Radiat Oncol Biol Phys 38: 941-947, 1997[CrossRef][Medline] 23. Immediate versus deferred treatment for advanced prostatic cancer: Initial results of the Medical Research Council Trial—The Medical Research Council Prostate Cancer Working Party Investigators Group. Br J Urol 79:235-246, 1997
24. Messing EM, Manola J, Sarosdy M, et al: Immediate hormonal therapy compared with observation after radical prostatectomy and pelvic lymphadenectomy in men with node-positive prostate cancer. N Engl J Med 341: 1781-1788, 1999 25. Prentice RL: Surrogate endpoints in clinical trials: Definition and operational criteria. Stat Med 8: 431-440, 1989[Medline] Submitted March 12, 2002; accepted August 7, 2002. This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
7, and the median dose was 45.0 Gy (range, 45.0 to 50.4 Gy). No patient received neoadjuvant, concurrent, or adjuvant hormonal therapy. 
