University of Kentucky UKnowledge CRVAW Faculty Journal Articles Center for Research on Violence Against Women 6-1-2008 Long-Term Survival After Radical Prostatectomy Compared to Other Treatments in Older Men With Local or Regional Prostate Cancer Liqian Liu University of Texas Health Science Center at Houston Ann L. Coker University of Kentucky, ann.coker@uky.edu Xianglin L. Du University of Texas Health Science Center at Houston, xianglin.l.du@uth.tmc.edu Janice N. Cormier University of Texas M.D. Anderson Cancer Center, jcormier@mdanderson.org Charles E. Ford University of Texas Health Science Center at Houston, charles.e.ford@uth.tmc.edu See next page for additional authors Click here to let us know how access to this document benefits you. Follow this and additional works at: https://uknowledge.uky.edu/crvaw_facpub Part of the Male Urogenital Diseases Commons, and the Public Health Commons Repository Citation Liu, Liqian; Coker, Ann L.; Du, Xianglin L.; Cormier, Janice N.; Ford, Charles E.; and Fang, Shenying, "Long-Term Survival After Radical Prostatectomy Compared to Other Treatments in Older Men With Local or Regional Prostate Cancer" (2008). CRVAW Faculty Journal Articles. 102. https://uknowledge.uky.edu/crvaw_facpub/102 This Article is brought to you for free and open access by the Center for Research on Violence Against Women at UKnowledge. It has been accepted for inclusion in CRVAW Faculty Journal Articles by an authorized administrator of UKnowledge. For more information, please contact UKnowledge@lsv.uky.edu.
Authors Liqian Liu, Ann L. Coker, Xianglin L. Du, Janice N. Cormier, Charles E. Ford, and Shenying Fang Long-Term Survival After Radical Prostatectomy Compared to Other Treatments in Older Men With Local or Regional Prostate Cancer Notes/Citation Information Published in, v. 97, no. 7, p. 583-591. Digital Object Identifier (DOI) http://dx.doi.org/10.1002/jso.21028 This article is available at UKnowledge: https://uknowledge.uky.edu/crvaw_facpub/102
2008;97:583 591 Long-Term Survival After Radical Prostatectomy Compared to Other Treatments in Older Men With Local/Regional Prostate Cancer LIQIAN LIU, MD, MS, 1 ANN L. COKER, PhD, 1 XIANGLIN L. DU, MD, PhD, 1 * JANICE N. CORMIER, MD, MPH, 2 CHARLES E. FORD, PhD, 1 AND SHENYING FANG, MD, MS 1 1 School of Public Health, University of Texas Health Science Center, Houston, Texas 2 Department of Surgical Oncology, University of Texas, M. D. Anderson Cancer Center, Houston, Texas Background: This study aimed to address long-term survival in a large population-based cohort of men with prostate cancer receiving radical prostatectomy compared to other treatments. Methods: We studied 5,845 patients diagnosed with local/regional stage prostate cancer at age 65 74 in 1992 with comorbidity score <2, who were defined as potential candidates for radical prostatectomy and identified from the SEER (Surveillance, Epidemiology and End Results)- Medicare cohort with median follow-up of 11 years. Results: Of 5,845 patients, 10-year all-cause survival rates were the highest for patients receiving radical prostatectomy (81.0%; 95% CI: 79.4 82.4%), followed by radical prostatectomy in combination with radiotherapy (67.6%; 62.0 72.5%), radiotherapy (60.5%; 58.3 62.6%), and were the lowest for watchful-waiting (50.7%; 47.5 53.8%). A similar pattern was found for 10-year prostate cancer-specific survivals by treatments. After adjusting for age, ethnicity, region, Gleason Score, comorbidity, median annual household income, hormone therapy and chemotherapy, the hazard ratio of all-cause mortality was 0.31 (95% CI: 0.25 0.37) for radical prostatectomy and 0.38 (95% CI: 0.28 0.52) for radical prostatectomy plus radiation therapy compared to those with watchful-waiting. Conclusions: There was a significant long-term survival benefit in men receiving radical prostatectomy compared to those receiving watchfulwaiting or radiotherapy. J. Surg. Oncol. 2008;97:583 591. ß 2008 Wiley-Liss, Inc. KEY WORDS: prostate cancer; treatment; radical prostatectomy; survival; older men INTRODUCTION The efficacy and effectiveness of radical prostatectomy compared to other primary treatments for prostate cancer, including radiotherapy and watchful-waiting (observational management), have not been convincingly demonstrated to date [1 13]. There have been four randomized trials addressing this issue in patients with localized prostate cancer [11 14], of which only one demonstrated a survival advantage of radical prostatectomy over watchful-waiting [14]. Two of these trials compared radical prostatectomy with external-beam radiotherapy [12][13] and found a significant reduction in disease progression associated with radical prostatectomy, one of which also found that radical prostatectomy was associated with favorable disease-specific survival but not overall survival [13]. However, the interpretation of the results from these randomized trials has been hampered by small sample size [2,11 13], questionable outcome measures [2][11 13], and incomplete data [12]. In addition, although randomized controlled trials are the gold standard for determining efficacy of a therapy, participants in these trials often do not represent the general population with the disease. Older patients in particular are under-represented in cancer clinical trials [15 17]. Moreover, it is important to know whether the efficacy documented under controlled ideal world conditions can be translated into real world effectiveness in the community. To the best of our knowledge, there has been no large populationbased study assessing the effectiveness of radical prostatectomy compared to other treatments specifically for older men with prostate cancer. Therefore, our objective was to examine the association between various prostate cancer treatments and long-term survival in older men using the nationwide and population-based Surveillance, Epidemiology, and End Results (SEER) and Medicare linked data. These data include a large cohort of men diagnosed with prostate ß 2008 Wiley-Liss, Inc. cancer at age 65 or older with up to 11.8 years of follow-up. The data also provides comprehensive information on patient and tumor characteristics, such as Gleason score, comorbidities, socioeconomic status as well as adjuvant therapy. We compared all-cause and prostate cancer-specific mortality in a cohort of patients who were potential candidates for radical prostatectomy and other treatment modalities such as radiation therapy, and observational management. We hypothesized that long-term survival rates would be higher among men with localized or regional stage prostate cancer who received radical prostatectomy relative to the other treatments. PATIENTS AND METHODS Data Sources The SEER program of the National Cancer Institute (NCI) includes population-based cancer registries in the 11 selected geographic areas, covering approximately 14% of the U.S. population. These areas are: the metropolitan areas of San Francisco/Oakland, Detroit, Atlanta and Seattle; Los Angeles county, the San Jose-Monterey area; and the states of Connecticut, Iowa, New Mexico, Utah and Hawaii. The Grant sponsor: Agency for Healthcare Research and Quality; Grant number: R01-HS016743. *Correspondence to: Xianglin L. Du, MD, PhD, Division of Epidemiology, University of Texas School of Public Health, 1200 Herman Pressler Drive, RAS-E631, Houston, TX 77030; Fax: 713-500-9264. E-mail: xianglin.l.du@uth.tmc.edu Received 3 January 2008; Accepted 19 February 2008 DOI 10.1002/jso.21028 Published online 31 March 2008 in Wiley InterScience (www.interscience.wiley.com).
584 Liu et al. population covered by SEER is comparable to the general U.S. population with respect to measures of poverty and education [18 20]. The Medicare program covers more than 97% of U.S. population age 65 years or older for hospital, physician, and other medical services [19,20]. The linked SEER-Medicare databases were created by matching cases in SEER against the Medicare master enrollment files. The SEER-Medicare data, from which this study population was identified, included cases in the SEER data from 1992 to 1999 and their Medicare claims through 2002. In this linkage, 93% of persons aged 65 and older in the SEER were matched to the Medicare enrollment files. The method of linking these data has been described elsewhere [19,20]. The Committee for Protection of Human Subjects at the University of Texas Health Science Center in Houston approved this study. Study Population There were 23,711 prostate cancer cases identified using the International Classification of Disease (ICD)-9-CM code of 185 reported to SEER in 1992. We studied the cases diagnosed in 1992 only to ensure every case has been followed up for at least 10 years after diagnosis. We excluded patients who did not have full coverage of both Medicare Part A (inpatient care) and Part B (outpatient care) or who were members of HMOs because the claims may be incomplete [21]. A total of 13,933 cases remained and were initially included in the analysis (Fig. 1). The cohort was further restricted to patients who were potential candidates for radical prostatectomy according to the treatment guidelines of National Cancer Institute (NCI) and National Comprehensive Cancer Network (NCCN) [22,23]. These include men with localized (tumor confined within prostate) prostate cancer with at least 10 years of life expectancy and with no serious comorbidities. Only 5,845 patients diagnosed with local/regional stage prostate cancer at age 65 74 years with comorbidity scores of less than 2 were included in our final analysis. Outcomes The survival time in months was calculated from the date of diagnosis to the date of death or to the date of last follow-up (December 31, 2002). Since SEER reported only the month and year of diagnosis, Fig. 1. Flowchart of subjects included in the analysis.
Survival After Radical Prostatectomy 585 we arbitrarily defined the day of diagnosis as the 15th of the month. All-cause mortality was defined as death from any cause that was the underlying cause of death, which was identified by the SEER program through linking the SEER data with the National Death Index data from the National Center for Health Statistics. Patients still alive at the last follow-up were censored. Prostate cancer-specific mortality was defined as prostate cancer as the underlying cause of death. In this specific analysis, patients who died of causes other than prostate cancer or were still alive at the last follow-up were censored. Treatments Among the defined cohort of patients, four treatment options were available, including radical prostatectomy, radiation therapy, radical prostatectomy in combination with radiation therapy, and watchfulwaiting (or observational management). Radical prostatectomy was identified if it was indicated in SEER (procedure codes of 40 70) or if there was a Medicare claim for radical prostatectomy (ICD-9-CM code of 605, or Common Procedure Terminology (CPT) codes of 55810 55815 or 55840 55845) [24,25]. Patients were defined as receiving radiation therapy if it was stated in SEER data or in Medicare with the following claims codes: [26,27] ICD-9-CM procedure codes of 9221 9229 for a hospital inpatient or outpatient facility claim of therapeutic radiology; the CPT codes of 77401 77499 or 77750 77799 for a physician or outpatient claim of radiotherapy or clinical brachytherapy; revenue center codes of 0330 or 0333 for radiotherapy. Patients were defined as receiving observational management if they did not receive either prostatectomy or radiation therapy. Adjuvant hormone therapy was defined if any of the following Medicare procedure codes were found within 6 months of the diagnosis of prostate cancer: the Healthcare Common Procedure Coding System (National Level II) codes of J1950 or J9217 J9219 for leuprolide, J9202 for goserelin; CPT (National Level I) codes of 54520 54521, 54530 or 54535 for orchiectomy [25,28]. Chemotherapy was determined if any of the following Medicare procedure codes were identified within 6 months of the diagnosis of prostate cancer as described elsewhere [21,28 30]: ICD-9-CM procedure codes of 9925 and V codes of V58.1, V66.2, or V67.2; the procedure codes of 96400 96549 or J9000 J9999 (except the codes for hormonal therapy above), Q0083 Q0085; and revenue center codes of 0331, 0332, and 0335 [25,28,31,32]. Comorbidity Index Patient comorbidity was identified from the Medicare claims according to diagnoses made or procedures performed in the year before the diagnosis of prostate cancer. The details of creating a comorbidity score have been described elsewhere [33]. In brief, the comorbidity index was based on the Medicare inpatient, outpatient and physician claim files. For physician and outpatient claims, a patient s comorbid diagnoses must appear on at least two different claims that were more than 30 days apart. Conditions that did not appear on two different claims were considered rule out diagnoses, and were not counted as comorbidities. We used the SAS macro rule-out programs available on the NCI website for calculations [28,34]. The program utilizes claim records and reflects the Deyo adaptation of the Charlson comorbidity index, with several procedure codes from the Romano adaptation [35 37]. Other Potential Confounding Variables Patient age, race/ethnicity, geographic regions, Gleason Score, tumor stage were available in SEER-Medicare database. We categorized age into four groups: 65 69, 70 74, 75 79 and 80 or older; race into white and non-white (blacks, Hispanics and others); region into Pacific (Los Angeles, San Jose, Seattle, San Francisco and Hawaii), Mountain (New Mexico and Utah), Central (Iowa), and East (Connecticut, Atlanta, and Detroit). Gleason Score was classified into four categories by the SEER: 2 4 (well differentiated), 5 7 (moderately differentiated), 8 10 (poorly differentiated), and unknown grade. Because local and regional diseases were combined as a single category in SEER data, we were unable to further stratify the analysis by local versus regional stage. However, this combined category may account for the upstaging that would likely occur if all clinically determined local stage prostate cancers were surgically staged [38]. We also used the American Joint Committee on Cancer (AJCC) stage available in SEER to control for residual confounding, although a large proportion of cases (37%) had missing information on AJCC stage. We used median annual household income at the zip code level from the 2000 Census available in the SEER-Medicare linked files as a proxy indicator of socioeconomic status (SES), which was categorized into quartiles in the analysis. Statistical Analysis Mean survival time and mortality rates per 1,000 person months were calculated according to treatment modality. Survival rates of 5 and 10 years were calculated and tested by log-rank statistics for differences between radical prostatectomy and other treatments. Hazard ratios were calculated for other treatments relative to watchful-waiting using the Cox proportional hazard model. The proportional assumption for the Cox proportional hazard model was assessed graphically and satisfied when the Kaplan Meier curves for survival functions by treatments were parallel without intersection while adjusting for age, race, region, Gleason score, AJCC stage, comorbidity, median annual household income, hormone therapy and chemotherapy [39]. The analyses were conducted using STATA Version 8 (StataCorp LP, College Station, TX, 2005). RESULTS Table I presents characteristics of the total population of prostate cancer cases reported to SEER in 1992 and included in the SEER- Medicare linked data, as well as the characteristics of patients who were treated with radical prostatectomy. About 55% of the patients were younger than 75 years old; a majority (85.5%) were white; most of the patients (76.7%) had local/regional cancer; 88% of the patients had a comorbidity score of 0 or 1. Younger men, those with local/ regional disease, fewer comorbidities, and higher median annual household income were more likely to be treated with radical prostatectomy (trend test of proportion, P < 0.0001). Patients with local/regional cancer, age 65 74 and comorbidity score of 0 or 1 represented the greatest proportion of patients who received radical prostatectomy and were defined as radical prostatectomy candidates as described in Study Population. The subsequent analyses were restricted to the 5,845 patients who were candidates for radical prostatectomy according to the clinical guidelines. Of the 5,845 patients, 2,567 actually received radical prostatectomy, 2,006 received radiation therapy, 302 received radical prostatectomy plus radiation therapy, and 970 received observational management (watchful-waiting) (Table II). Generally, younger, white, men with moderately differentiated tumor, those with low comorbidity score and with the highest median annual household income were more likely to receive radical prostatectomy relative to other treatments. In addition, patients treated with radical prostatectomy were less likely to receive hormone therapy than those who received watchful-waiting or radical prostatectomy plus radiation therapy, and also less likely to receive chemotherapy than the watchful-waiting group (Chi-square P values 0.05). There was no significant difference between radical prostatectomy and radical prostatectomy plus radiation therapy groups
586 Liu et al. TABLE I. Characteristics of Men With Local/Regional Stage Prostate Cancer in 1992 and the Number and Percentage That Had Radical Prostatectomy Characteristics n(%) a of patients (N ¼ 13,933) n(%) a of patients treated with radical prostatectomy (N ¼ 3,597) P value, trend of proportion Age (years) 65 69 3,445 (24.7) 1,787 (51.9) <0.0001 70 74 4,211 (30.2) 1,426 (33.9) 75 79 3,343 (24.0) 350 (10.5) 80 2,934 (21.1) 34 (1.2) Race White 11,908 (85.5) 3,191 (26.8) NA Non-White b 2,025 (14.5) 406 (20.1) Comorbidity 0 9,077 (65.2) 2,778 (30.6) <0.0001 1 3,128 (22.5) 626 (20.0) 2 1,098 (7.9) 157 (14.3) 3 630 (4.5) 36 (5.7) Gleason score 2 4 (well-differentiated) 2,437 (17.5) 395 (16.2) <0.0001 5 7 (moderately differentiated) 6,978 (50.1) 2,357 (33.8) 8 10 (poorly differentiated) 3,056 (21.9) 778 (25.5) Unknown 1,462 (10.5) 67 (4.6) SEER-recoded AJCC c stage I 1,696 (12.2) 376 (22.2) <0.0001 II 1,414 (10.2) 545 (38.5) III 1,931 (13.7) 1,282 (66.4) IV 1,550 (11.1) 201 (13.0) Unknown 7,342 (52.7) 1,193 (16.3) Extent of disease Local/regional 10,640 (76.4) 3,379 (31.8) <0.0001 Distant 1,074 (7.7) 16 (1.5) Unstaged 2,219 (15.9) 202 (9.1) Region d Pacific 5,735 (41.2) 1,764 (30.8) <0.0001 Mountain 1,604 (11.5) 587 (36.6) Central 1,627 (11.7) 353 (21.7) East 4,967 (35.7) 893 (18.0) Median annual household income 1st quartile (<$36951) 3,230 (23.2) 716 (22.2) <0.0001 2nd quartile ($36951 46845) 3,357 (24.1) 825 (24.6) 3rd quartile ($46846 60088) 3,380 (24.3) 950 (28.1) 4th quartile (60089) 3,437 (24.7) 975 (28.4) Unknown 529 (3.8) 131 (24.8) a Percentage of cases received radical prostatectomy in the strata (row percentage). b Non-White: 1,255 blacks (9.0%), 157 Hispanic (1.1%), 613 others (4.4%). c American Joint Committee on Cancer. d Pacific region includes Los Angeles, San Jose, Seattle, San Francisco and Hawaii; mountain region includes New Mexico and Utah; central region includes Iowa; east region includes Connecticut, Atlanta, and Detroit. except in Gleason score, SEER-recoded AJCC stage and receiving hormone therapy, and also no significant difference in median annual household income between radical prostatectomy and radiation therapy groups (Table II). Table III presents the hazard ratio of mortality for the four treatment groups. Men who were treated with radical prostatectomy and radical prostatectomy combined with radiation therapy had significantly favorable survivals than those who were treated with radiation therapy alone or watchful-waiting. The mean overall and prostate cancerspecific survival time was the greatest for those treated with radical prostatectomy, followed by radical prostatectomy plus radiation therapy, radiation therapy alone, and was the lowest for watchfulwaiting. Compared to the watchful-waiting group and adjusting for potential confounders, men who underwent radical prostatectomy were significantly less likely to die of any causes (ahr ¼ 0.31; 95% CI ¼ 0.25, 0.37) or to die of prostate cancer specifically (ahr ¼ 0.17; 95% CI ¼ 0.10, 0.28). Similarly, men receiving radical prostatectomy plus radiation therapy were also less likely to die of any cause (ahr ¼ 0.38; 95% CI ¼ 0.28, 0.52) or to die of prostate cancer (ahr ¼ 0.23; 95% CI ¼ 0.13, 0.48). The magnitude of mortality reduction was greater for prostate cancer-specific mortality than allcause mortality in patients with radical prostatectomy and radical prostatectomy plus radiation therapy compared to those with watchfulwaiting. Radiation therapy was also associated with a reduced allcause and prostate cancer-specific mortality but magnitude was smaller. The all-cause and prostate cancer-specific 5- and 10-year survival rates are presented in Table IV, and the Kaplan Meier survival curves are illustrated in Figures 2 and 3. The median follow-up was 132.59 months. As shown in Figure 2, the Kaplan Meier overall survivals were markedly different for patients treated with radical prostatectomy, radical prostatectomy plus radiation therapy, radiation therapy and watchful-waiting (log-rank test P < 0.0001). Men receiving radical prostatectomy and radical prostatectomy plus radiation
TABLE II. Comparison of Potential Risk Factors for Survival Among Various Treatment Modalities Survival After Radical Prostatectomy 587 Risk factor Radical prostatectomy, n(%)(n¼ 2,567) Radical prostatectomy and radiotherapy, n (%) (N ¼ 302) Radiotherapy, n (%) (N ¼ 2,006) Watchful-waiting, n(%)(n¼ 970) Total, n (%) (N ¼ 5,845) Age (years) 65 69 1,432 (53.0) 164 (6.1) 728 (26.9) 381 (14.1) 2,705 70 74 1,135 (36.2) 138 (4.4) 1,278 (40.7) 589 (18.8) 3,140 P value REF 0.6 <0.0001 <0.0001 Race White 2,284 (45.0) 267 (5.3) 1,736 (34.2) 787 (15.5) 5,074 Non-White 283 (36.7) 35 (4.5) 270 (35.0) 183 (23.7) 771 P value REF 0.8 0.01 <0.0001 Region Pacific 1,255 (50.8) 154 (6.2) 684 (27.7) 377 (15.3) 2,470 Mountain 415 (56.5) 38 (5.2) 148 (20.1) 134 (18.2) 735 Central 255 (40.5) 39 (6.2) 240 (38.2) 95 (15.1) 629 East 642 (31.9) 71 (3.5) 934 (46.4) 364 (18.1) 2,011 P value REF 0.2 <0.0001 0.0 Gleason score 2 4 (well differentiated) 286 (26.6) 20 (1.9) 422 (39.2) 349 (32.4) 1,077 5 7 (moderately differentiated) 1,768 (51.8) 153 (4.5) 1,128 (33.1) 362 (10.6) 3,411 8 10 (poorly differentiated) 473 (41.5) 128 (11.2) 365 (32.1) 173 (15.2) 1,139 Unknown 40 (18.4) 1 (0.5) 91 (41.7) 86 (39.5) 218 P value REF <0.0001 <0.0001 <0.0001 AJCC stage I 301 (36.4) 21 (2.5) 373 (45.1) 133 (16.1) 828 II 460 (55.6) 21 (2.5) 273 (33.0) 73 (8.8) 827 III 914 (67.0) 171 (12.5) 217 (15.9) 63 (4.6) 1,365 IV 112 (34.9) 41 (12.8) 64 (19.9) 104 (32.4) 321 Unkown 780 (31.2) 48 (1.9) 1,079 (43.1) 579 (23.1) 2,504 P value REF <0.0001 <0.0001 <0.0001 Comorbidity 0 2,102 (45.9) 247 (5.4) 1,516 (33.1) 720 (15.7) 4,585 1 465 (36.9) 55 (4.4) 490 (38.9) 250 (19.8) 1,260 P value REF 1.0 <0.0001 <0.0001 Median household income 1st quartile (<$36951) 495 (40.7) 50 (4.1) 412 (33.9) 259 (21.3) 1,216 2nd quartile ($36951 $46845) 584 (42.9) 74 (5.4) 496 (36.5) 206 (15.2) 1,360 3rd quartile ($46846 $60088) 662 (44.6) 81 (5.5) 528 (35.5) 215 (14.5) 1,486 4th quartile (>$60089) 738 (46.9) 88 (5.6) 513 (32.6) 236 (15.0) 1,575 Unknown 88 (42.3) 9 (4.3) 57 (27.4) 54 (26.0) 208 P value REF 0.8 0.1 <0.0001 Hormone therapy Yes 271 (36.0) 44 (5.8) 170 (22.6) 268 (35.6) 753 No 2,296 (45.1) 258 (5.1) 1,836 (36.1) 702 (13.8) 5,092 P value REF 0.0 0.0 <0.0001 Chemotherapy Yes 213 (39.4) 27 (5.0) 146 (27.0) 155 (28.7) 541 No 2,354 (44.4) 275 (5.2) 1,860 (35.1) 815 (15.4) 5,304 P value REF 0.7 0.2 <0.0001 P value: Chi-square test for the distribution of the factors among the treatment groups. REF, reference; race, region and SEER-recoded AJJC Stage are same as in Table I. therapy had significantly longer survival than men receiving radiation therapy and watchful-waiting. The 10-year all-cause survival rates were the highest for patients treated with radical prostatectomy (81.0%; 95% CI ¼ 79.4, 82.4), followed by patients with radical prostatectomy plus radiation therapy (67.6%; 95% CI ¼ 62.0, 72.5), radiation therapy (60.5; 95% CI ¼ 58.3, 62.6), and were the lowest for watchful-waiting (50.7%; 95% CI ¼ 47.5, 53.8). A similar pattern was found for 10-year prostate cancer-specific survival according to treatment rendered. Five-year allcause and prostate cancer-specific survival rates were consistently higher for radical prostatectomy and radical prostatectomy combined with radiation therapy relative to the other treatment groups. Patients receiving radical prostatectomy or radical prostatectomy plus radiation therapy were not significantly different in 5- and 10-year all-cause and prostate cancer-specific survival rates (log-rank test P values 0.05). In addition, as shown in Figure 3, the difference in prostate cancerspecific survival distributions between radical prostatectomy plus radiation therapy and radiation therapy was not statistically significant (log-rank test P ¼ 0.70). DISCUSSION This study compared the long-term survival among men who received radical prostatectomy with those who received other primary treatments such as radiotherapy, watchful-waiting and radical prostatectomy in combination with radiotherapy. The comparison was restricted to men with prostate cancer who are candidates for radical prostatectomy according to the clinical guidelines. We found that men
588 Liu et al. TABLE III. Hazard Ratio of Mortality for Patients Receiving Radical Prostatectomy Compared to Patients With Other Treatments Treatment modality Mean survival time in month (SE) Mortality rate a (95% CI) HR b (95% CI) Overall survival Radical prostatectomy 127.8 (0.6) 2.0 (1.8, 2.2) 0.31 (0.25, 0.37) Radical prostatectomy and radiotherapy 120.0 (2.0) 3.3 (2.7, 4.0) 0.38 (0.28, 0.52) Radiotherapy 112.5 (0.9) 4.2 (4.0, 4.5) 0.68 (0.56, 0.81) Watchful-waiting 101.2 (1.5) 5.6 (5.2, 6.1) 1.00 (reference) Prostate cancer-specific survival Radical prostatectomy 139.7 (0.3) 0.2 (0.1, 0.2) 0.17 (0.10, 0.28) Radical prostatectomy and radiotherapy 135.2 (1.4) 0.6 (0.4, 0.9) 0.23 (0.13, 0.48) Radiotherapy 134.6 (0.6) 0.6 (0.5, 0.8) 0.56 (0.37, 0.85) Watchful-waiting 129.6 (1.1) 1.1 (0.9, 1.4) 1.00 (reference) CI, confidence interval; HR, hazard ratio; SE, standard error. a Deaths per 1,000 person-months. b Adjusted for age, race, region, Gleason score, AJCC stage, comorbidity, median annual household income, hormone therapy and chemotherapy. who were treated with radical prostatectomy had significantly longer survival than those who received other treatments after adjusting for significant prognostic factors. Our findings are consistent with those reported by Bill-Axelson et al. [14] in the randomized clinical trial which demonstrated that radical prostatectomy was superior to watchful-waiting for localized prostate cancer in men with mean age of 64.7. With a median of 8.2 years of follow-up in the trial, radical prostatectomy significantly reduced the prostate cancer-specific (HR ¼ 0.56; 95%CI ¼ 0.36, 0.88) and the overall (HR ¼ 0.74; 95%CI ¼ 0.56, 0.99) mortality. We observed even stronger positive association between radical prostatectomy and survival compared to watchful-waiting in older men. The difference in findings may be associated with our longer followup time, larger sample size, and relatively healthy patients with low comorbidity scores. These may also be reasons that the 10-year overall and prostate cancer-specific survival rates in the present study are higher than those reported in the population-based study by Lu-Yao and Yao [40]. The difference is also likely attributable to the restricted cohort that included patients potentially eligible for radical prostatectomy and younger patients (65 74 years of age only). The present research suggests that there may be a benefit over the 10-year follow-up for patients treated with radical prostatectomy over radiotherapy among those who were candidates for radical prostatectomy as defined by age, stage and comorbid conditions. This conclusion is shared by Paulson [12] and Akakura et al. [13] who found the advantage of radical prostatectomy in cancer progression and cancer specific survival rates compared with radiotherapy in their clinical trials. The difference in risk of dying between radical prostatectomy and watchful-waiting increases over time in this study. This observation is consistent with the findings in the study by Holmberg et al. [5] The same pattern appears between radical prostatectomy and radiotherapy. These findings may reflect the fact that prostate cancer is a slowly progressive malignancy, therefore, the effectiveness of particular treatment modalities will only become evident in studies with long term follow-up. There are a number of limitations in this study. First and the most important concern would be the effect of potential selection bias because all patients were not randomized into the different treatment groups but self selected the treatment modality they preferred. The watchful-waiting group may have included heterogeneous patients. If the survival outcomes were associated with other factors that led the patients into different therapy groups rather than the treatment modalities themselves, the finding could be seriously biased if those factors were not properly controlled for. For this reason, we carefully restricted our analysis to the population who were potential candidates for radical prostatectomy according to the treatment guidelines. For example, only those patients with tumor confined within prostate, with at least 10 years of life expectancy and with no serious comorbidities were selected in order to minimize the selection bias. Furthermore, there are some differences in the distributions of age, race and cancer stage in the present study population that composed of 13,933 prostate cancer cases in the linked SEER-Medicare in 1992 and in the unlinked (excluded) SEER population that comprised 9,778 prostate cancer cases in the same year [41]. Compared to the full SEER population with prostate cancer, the present study population of SEER linked with Medicare included 3.1% (P ¼ 0.0028) fewer cases aged 65 69; 2.4% (P < 0.0001) more white men; 1.5% (P ¼ 0.0098) more men with local/regional stage cancer. While these differences are statistically significant due to the large numbers with prostate cancer, these differences on potential selection bias are unlikely to explain the significant findings for observed association between radical prostatectomy treatment and survival given that the SEER-Medicare TABLE IV. All Cause and Prostate Cancer-Specific Survival Rates for Patients Receiving Radical Prostatectomy Compared to Patients With Other Treatments All cause survival % (95% CI) Prostate cancer-specific survival % (95% CI) Treatment 5-Year 10-Year 5-Year 10-Year Radical prostatectomy 93.6 (92.6, 94.5) 81.0 (79.4, 82.4) 98.7 (98.1, 99.1) 98.1 (97.4, 98.6) Radical prostatectomy 91.1 (87.3, 93.8), P ¼ 0.7 67.6 (62.0, 72.5), P ¼ 0.1 97.3 (94.3, 98.8), P ¼ 0.8 94.5 (90.8, 96.8), P ¼ 0.1 and radiotherapy Radiotherapy 85.0 (83.3, 86.4), P < 0.0001 60.5 (58.3, 62.6), P < 0.0001 95.7 (94.5, 96.7), P < 0.0001 93.8 (92.3, 95.0), P < 0.0001 Watchful-waiting 75.5 (72.6, 73.4), P ¼ 0.1 50.7 (47.5, 53.8), P ¼ 0.0 91.5 (88.1, 94.0), P ¼ 0.0 86.3 (82.1, 89.6), P ¼ 0.0 P value: Log-rank test for the difference in survival distribution between radical prostatectomy and other treatment groups after adjusted for age, race, region, Gleason score, SEER-recoded AJCC stage, comorbidities, median annual household income, hormone therapy and chemotherapy. CI ¼ confidence interval.
Survival After Radical Prostatectomy 589 Survival probability 0.00 0.25 0.50 0.75 1.00 p value<0.0001 12 24 36 48 60 72 84 96 108 120 132 144 Time (months) RP RP+RT RT Watchful waiting Fig. 2. Overall survival in men with local/regional prostate cancer, by treatment modalities. [Color figure can be viewed in the online issue, available at www.interscience.wiley.com.] linked population would have better survival potential (lower stage at diagnosis and a greater proportion of white race) than the complete SEER population. Second, we were unable to control for the level of prostatespecific antigen (PSA) and physician/hospital characteristics. PSA was demonstrated to be associated with prostate cancer survival and relapse by Roehl et al. [42] and Pisansky et al. [43], and physician/hospital characteristics were suggested to be associated with cancer outcomes by Begg et al. [44] and Bach et al. [45] Third, misclassification bias may be present particularly because local and regional stage prostate cancers were combined into one category by the SEER. However, the strong survival advantage of radical prostatectomy over the other treatment modalities in this large, population-based study is unlikely to be completely explained by variations in PSA or health care provider characteristics. Fourth, our study was limited to men aged 65 74 years. Therefore, the results might not be generalized to Survival probability 0.00 0.25 0.50 0.75 1.00 p value<0.0001 12 24 36 48 60 72 84 96 108 120 132 144 Time (months) RP RP+RT RT Watchful waiting Fig. 3. Prostate cancer-specific survival in men with local/regional prostate cancer, by treatment modalities. [Color figure can be viewed in the online issue, available at www.interscience.wiley.com.]
590 Liu et al. younger men with prostate cancer. However, since most of the prostate cancer cases are older men [46], our findings will be valuable in generalization to a larger elderly population. Finally, this study did not examine the morbidity and complications such as impotence and urinary incontinence, particularly since it was known that radical prostatectomy is associated with these complications more than other treatment modalities [47,48]. Therefore, the adverse impact of radical prostatectomy on health-related quality of life may be a more serious concern, but that cannot be addressed in our study. This study has a number of strengths. First, we restricted the analyses to a cohort of patients who were candidates for radical prostatectomy based on the clinical guideline, extent of disease and estimated life expectancy. Such restriction allows a fairer comparison among the various treatment strategies. In addition, the analyses controlled for patient comorbidity identified from Medicare claims, which is a known confounder for treatment choices and survival. Furthermore, this is a population-based study including incident prostate cancer cases in 1992 in the SEER regions with up to 11 years of follow-up, which enables us to observe the long-term outcomes and the large sample size (5,845 radical prostatectomy candidates) increases the power to detect differences in survival. In conclusion, our study demonstrated a significant long-term survival benefit in men receiving radical prostatectomy compared to those receiving watchful-waiting or radiotherapy. However, any benefit of radical prostatectomy on survival must be weighed against the potential morbidity of surgical treatment, such as impotence and urinary incontinence. Patients need to be aware of both benefits and adverse effects of radical prostatectomy compared to other treatments so that the better and informed treatment decisions can be made. In addition, further study evaluating the potential survival advantages of radical prostatectomy among younger men with prostate cancer would be of interest. ACKNOWLEDGMENTS We acknowledge the efforts of the National Cancer Institute; Center for Medicare and Medicaid Services; Information Management Services, Inc.; and the SEER Program tumor registries in the creation of this database. The interpretation and reporting of these data are the sole responsibilities of the authors. Dr. Du is supported in part by a grant from the Agency for Healthcare Research and Quality (R01- HS016743). REFERENCES 1. Presti J: Neoplasms of the prostate gland. In: Tanagho EA, McAninch JW, editors. Smith s general urology. 16th edition. New York: Lange Medical Books/McGraw Hill; 2004. pp 367 385. 2. Graversen PH, Nielsen KT, Gasser TC, et al.: Radical prostatectomy versus expectant primary treatment in stages I and II prostatic cancer. A fifteen-year follow-up. Urology 1990;36:493 498. 3. Potters L, Klein EA, Kattan MW, et al.: Monotherapy for stage T1-T2 prostate cancer: Radical prostatectomy, external beam radiotherapy, or permanent seed implantation. Radiother Oncol 2004;71:29 33. 4. Lee D: Radical prostatectomy improves cancer-specific survival but not overall survival in early-stage prostate cancer. Clin Prostate Cancer 2003;2:15 17. 5. Holmberg L, Bill-Axelson A, Helgesen F, et al.: A randomized trial comparing radical prostatectomy with watchful-waiting in early prostate cancer. N Engl J Med 2002;347:781 789. 6. Kupelian PA, Elshaikh M, Reddy CA, et al.: Comparison of the efficacy of local therapies for localized prostate cancer in the prostate-specific antigen era: A large single-institution experience with radical prostatectomy and external-beam radiotherapy. J Clin Oncol 2002;20:3376 3385. 7. Keyser D, Kupelian PA, Zippe CD, et al.: Stage T1-2 prostate cancer with pretreatment prostate-specific antigen level < or ¼ 10 ng/ml: Radiation therapy or surgery? Int J Radiat Oncol Biol Phys 1997;38:723 729. 8. D Amico AV, Whittington R, Kaplan I, et al.: Equivalent biochemical failure-free survival after external beam radiation therapy or radical prostatectomy in patients with a pretreatment prostate specific antigen of >4 20 ng/ml. Int J Radiat Oncol Biol Phys 1997;37:1053 1058. 9. D Amico AV, Whittington R, Kaplan I, et al.: Equivalent 5-year bned in select prostate cancer patients managed with surgery or radiation therapy despite exclusion of the seminal vesicles from the CTV. Int J Radiat Oncol Biol Phys 1997;39:335 340. 10. Martinez AA, Gonzalez JA, Chung AK, et al.: A comparison of external beam radiation therapy versus radical prostatectomy for patients with low risk prostate carcinoma diagnosed, staged, and treated at a single institution. Cancer 2000;88:425 432. 11. Byar DP, Corle DK: VACURG randomised trial of radical prostatectomy for stages I and II prostatic cancer. Veterans Administration Cooperative Urological Research Group. Urology 1981;17:7 11. 12. Paulson DF: Randomized series of treatment with surgery versus radiation for prostate adenocarcinoma. NCI Monographs 1988;7: 127 131. 13. Akakura K, Isaka S, Akimoto S, et al.: Long-term results of a randomized trial for the treatment of Stages B2 and C prostate cancer: Radical prostatectomy versus external beam radiation therapy with a common endocrine therapy in both modalities. Urology 1999;54:313 318. 14. Bill-Axelson A, Holmberg L, Ruutu M, et al.: Scandinavian Prostate Cancer Group Study No. 4. Radical prostatectomy versus watchful-waiting in early prostate cancer. N Engl J Med 2005; 352:1977 1984. 15. Hutchins LF, Unger JM, Crowley JJ, et al.: Underrepresentation of patients 65 years of age or older in cancer-treatment trials. N Engl J Med 1999;341:2061 2067. 16. Trimble EL, Carter CL, Cain D, et al.: Representation of older patients in cancer treatment trials. Cancer 1994;74:2208 2214. 17. Unger JM, Hutchins LF, Albain KS: Underrepresentation of elderly patients in cancer clinical trials: Causes and remedial strategies. In: Balducci L, Lyman GH, Ershler WB, et al.: Comprehensive Geriatric Oncology. 2nd edition. New York, NY: Taylor & Francis; 2004. pp 259 274. 18. Ries LAG, Eisner MP, Kosary CL, et al.: SEER Cancer Statistics Review, 1975-2002, National Cancer Institute. Bethesda, MD, http://seer.cancer.gov/csr/1975_2002/ (accessed August 22, 2005). 19. Potosky AL, Riley GF, Lubitz JD, et al.: Potential for cancer related health services research using a linked Medicare-tumor registry database. Med Care 1993;31:732 748. 20. Warren JL, Klabunde CN, Schrag D, et al.: Overview of the SEER-Medicare data: Content, research applications, and generalizability to the United States elderly population. Med Care 2002;40:3 18. 21. Du XL, Goodwin JS: Increase of chemotherapy use in older women with breast carcinoma from 1991 to 1996. Cancer 2001; 92:730 737. 22. National Cancer Institute. http://www.cancer.gov/cancertopics/ pdq/treatment/prostate/healthprofessional (accessed June 21 2006). 23. The National Comprehensive Cancer Network. Practice Guidelines in Oncology, Prostate Cancer, v.1. 2005. 24. National Cancer Institute, Bethesda, MD. The SEER program code manual. NIH Publication No. 94-1999. Revised Edition, 1994. 25. American Medical Association. Physicians Current Procedural Terminology CPT 2000. Chicago: American Medical Association; 2000. 26. Du XL, Freeman JL, Goodwin JS: Information on radiation treatment in patients with breast cancer: The advantages of the linked medicare and SEER data. J Clin Epidemiol 1999;52:463 470.
Survival After Radical Prostatectomy 591 27. Virnig BA, Warren JL, Cooper GS, et al.: Studying radiation therapy using SEER-Medicare-linked data. Med Care 2002;40: 49 54. 28. Du XL, Fang S, Coker AL, et al.: Racial disparity and socioeconomic status in association with survival in older men with local/regional stage prostate cancer: Findings from a large community-based cohort. Cancer 2006;106:1276 1285. 29. Warren JL, Harlan LC, Fahey A, et al.: Utility of the SEER- Medicare data to identify chemotherapy use. Med Care 2002;40: 55 61. 30. Du XL, Goodwin JS: Patterns of use of chemotherapy for breast cancer in older women: Findings from Medicare claims data. J Clin Oncol 2001;19:1455 1461. 31. U.S. Public Health Services. International Classification of Diseases, 9th Revision, Clinical Modification. 5th edition. Los Angeles, CA: PMIC; 1996. 32. Health Care Financing Administration. HCFA Data Dictionary: Revenue Center Codes. June 17, 1999. 33. Du XL, Chan W, Giordano S, et al.: Variation in modes of chemotherapy administration for breast cancer and association with hospitalization for chemotherapy-related toxicity. Cancer 2005;104:913 924. 34. National Cancer Institute. http://healthservices.cancer.gov/seermedicare/program/comorbidity.html (accessed on August 22, 2005). 35. Charlson ME, Pompei P, Ales KL, et al.: A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 1987;40:373 383. 36. Romano PS, Roos LL, Jollis JG: Adapting a clinical comorbidity index for use with ICD-9-CM administrative data: Differing perspectives. J Clin Epidemiol 1993;46:1075 1079. 37. Deyo RA, Cherkin DC, Ciol MA: Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol 1992;45:613 619. 38. Litwiller SE, Djavan B, Klopukh BV, et al.: Radical retropubic prostatectomy for localized carcinoma of the prostate in a large metropolitan hospital: Changing trends over a 10-year period (1984 1994). Dallas Outcomes Research Group for Urological Disorders. Urology 1995;45:813 822. 39. Kleinbaum DG: Survival analysis: A self-learning text. New York, NY; Springer-Verlag: 1996. 40. Lu-Yao GL, Yao SL: Population-based study of long-term survival in patients with clinically localised prostate cancer. Lancet 1997;349:906 910. 41. Surveillance, Epidemiology, and End Results (SEER) Program SEER*Stat Database: Incidence SEER 11 Regs Public-Use, Nov 2001 Sub (1992 1999), National Cancer Institute, DCCPS, Surveillance Research Program, Cancer Statistics Branch. 42. Roehl KA, Han M, Ramos CG, et al.: Cancer progression and survival rates following anatomical radical retropubic prostatectomy in 3,478 consecutive patients: Long-term results. J Urol 2004;172:910 914. 43. Pisansky TM, Kahn MJ, Rasp GM, et al.: A multiple prognostic index predictive of disease outcome after irradiation for clinically localized prostate carcinoma. Cancer 1997;79:337 344. 44. Begg CB, Riedel ER, Bach PB, et al.: Variations in morbidity after radical prostatectomy. N Engl J Med 2002;346:1138 1144. 45. Bach PB, Pham HH, Schrag D, et al.: Primary care physicians who treat blacks and whites. N Engl J Med 2004;351:575 584. 46. Ries LAG, Eisner MP, Kosary CL, et al.: SEER Cancer Statistics Review, 1975 2002, National Cancer Institute. Bethesda, MD, http://seer.cancer.gov/csr/1975_2002/ (accessed August 22, 2005). 47. Steineck G, Helgesen F, Adolfsson J, et al.: Quality of life after radical prostatectomy or watchful waiting. N Engl J Med 2002;347:790 796. 48. Potosky AL, Davis WW, Hoffman RM, et al.: Five-year outcomes after prostatectomy or radiotherapy for prostate cancer: The prostate cancer outcomes study. J Natl Cancer Inst 2004;96: 1358 1367.