Effect of Body Mass Index on Histopathologic Parameters: Results of Large European Contemporary Consecutive Open Radical Prostatectomy Series

Oncology Effect of Body Mass Index on Histopathologic Parameters: Results of Large European Contemporary Consecutive Open Radical Prostatectomy Series Hendrik Isbarn, Claudio Jeldres, Lars Budäus, Georg Salomon, Thorsten Schlomm, Thomas Steuber, Felix K. H. Chun, Sascha Ahyai, Umberto Capitanio, Alexander Haese, Hans Heinzer, Hartwig Huland, Markus Graefen, and Pierre Karakiewicz OBJECTIVES

METHODS

RESULTS

CONCLUSIONS

To determine whether an increased body mass index (BMI) is a predictor of advanced pathologic findings in European men undergoing radical prostatectomy (RP). The relationship between obesity and prostate cancer is controversial. Studies, predominantly derived from the United States, have suggested that an increased BMI is a significant predictor of adverse pathologic findings in patients treated with open RP. From April 2005 to June 2008, 1538 consecutive patients were treated with open RP at a single tertiary referral center. We tested the effect of BMI on the rate of extracapsular extension, seminal vesicle invasion, lymph node invasion, and positive surgical margins in univariate and multivariate logistic regression models. The covariates consisted of clinical stage, prostate-specific antigen, biopsy Gleason score, age, prostate volume, and rate of nerve-sparing surgery. On multivariate analysis, both continuously coded and categorically coded BMI was unrelated to the rate of extracapsular extension (odds ratio [OR] 1.02, P ⫽ .5), seminal vesicle invasion (OR 1.03, P ⫽ .3), lymph node invasion (OR 0.98, P ⫽ .7), or positive surgical margins (OR 1.03, P ⫽ .3). Obese patients who are candidates for open RP should not expect to have worse pathologic findings after surgery than their nonobese counterparts. Differences in patients’ weight and height between North America and Europe might explain the lack of adverse effects of an elevated BMI in this European cohort. UROLOGY 73: 615– 619, 2009. © 2009 Elsevier Inc.

O

besity is a common and growing problem in the United States and in Western European countries.1,2 Obesity has been shown to be an independent risk factor for various types of cancer and increased cancer mortality.3,4 However, the relationship between obesity and various prostate cancer (PCa) control endpoints, such as pathologic stage and rate of biochemical recurrence after radical prostatectomy (RP), remains a matter of debate. Several North American studies have suggested that an increased body mass index (BMI) predisposes to more unfavorable PCa grade and stage and/or to a greater bioP. I. Karakiewicz partially supported by the University of Montreal Health Center Urology Associates, Fonds de la Recherche en Santé du Quebec, University of Montreal Department of Surgery, and University of Montreal Health Center (CHUM) Foundation. From the Martini-Clinic, Prostate Cancer Center Hamburg-Eppendorf, Hamburg, Germany; Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Center, Montreal, Quebec, Canada; Department of Urology, University Hospital Hamburg-Eppendorf, Hamburg, Germany; and Department of Urology, VitaSalute San Raffaele, Milan, Italy Reprint requests: Pierre I. Karakiewicz, M.D., F.R.C.S.C., Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Center, 1058 Rue StDenis, Montréal, QC H2X 3J4 Canada. E-mail: [email protected] Submitted: August 27, 2008, accepted (with revisions): September 16, 2008

© 2009 Elsevier Inc. All Rights Reserved

chemical recurrence rate after RP.5-12 Conversely, a large prospective cohort study from the Mayo Clinic (n ⫽ 5313) with a median follow-up of 10 years, found that a greater BMI was associated with adverse pathologic features at RP but was not an independent predictor of biochemical recurrence or cancer-specific survival.13 European studies addressing the effect of obesity on PCa incidence have also shown conflicting results. Some have indicated that obese patients are at an increased risk of developing PCa. However, others could not corroborate these findings.3,14,15 Few European studies have addressed the relationship between obesity and PCa characteristics. To date, 3 such studies, with a combined sample size of 4606 patients, examined the effect of BMI on biopsy and RP Gleason grade and on the biochemical recurrence rate after RP.16-18 Two studies (n ⫽ 4313) found that BMI did not impede the ability to predict the PCa grade or rate of biochemical recurrence after RP.16,18 One reported no effect of BMI on pathologic stage (n ⫽ 293).17 Differences in the prevalence and magnitude of obesity between North America and Europe19,20 could explain the discrepancy between the North American and European 0090-4295/09/$34.00 doi:10.1016/j.urology.2008.09.038

615

findings. Alternatively, the lack of an effect between BMI and PCa characteristics might be related to the relatively small number of European studies that has examined the effect of BMI. Because of the paucity of European data, we decided to examine the effect of BMI on pathologic PCa characteristics, as well as on the rate of positive surgical margins (PSMs), with the intent of testing the premise that obese patients may present with more unfavorable pathologic findings, which in turn could predispose them to greater rates of PSMs.

MATERIAL AND METHODS From April 2005 to June 2008, 1740 patients underwent open RP at the Martini-Clinic, Prostate Cancer Center Hamburg-Eppendorf.21 Exclusions consisted of patients treated with neoadjuvant hormonal therapy (n ⫽ 148), because this could lead to more favorable pathologic findings,22,23 previous radiotherapy (n ⫽ 4), and missing BMI data (n ⫽ 50). This resulted in 1538 assessable patients. Lymph node dissection was only performed in patients who had a moderate or high risk of lymph node invasion (LNI), according to a previously validated nomogram.24 The whole RP specimen was surface inked and processed using serial step sections at 3 mm according to the Stanford protocol.25 Grading was performed according to the Gleason system.26 Pathologic stage was defined according to the 1992/2002 American Joint Committee on Cancer staging classification.27 Surgical margins were defined as positive if ⱖ1 cancer cell extended to the inked surface. According to the World Health Organization classification, the BMI was recorded prospectively before RP and was calculated by dividing the weight (kg) by the square of the height (m2).28 The BMI was stratified as follows: BMI ⬍25 kg/m2 (normal weight) vs 25-29.9 kg/m2 (overweight) vs 30-34.9 kg/m2 (mildly obese) vs ⱖ35 kg/m2 (moderately or severely obese). The ␹2 test and independent t test were used to assess the statistical significance of the mean values and proportions, respectively. Subsequently, the effect of BMI on extracapsular extension (ECE), seminal vesicle invasion (SVI), LNI, and the rate of PSMs was examined in 2 ways. First, we used the World Health Organization categories. Second, we relied on continuously coded BMI, because categorization could have either amplified or hid the true relationship between BMI and the 4 examined outcome parameters. The 2 BMI coding schemes were examined in univariate and multivariate logistic regression models. In the multivariate models, the covariates consisted of patient age, clinical stage, prostatespecific antigen (PSA) level, biopsy Gleason score, prostate volume, and nerve-sparing status at RP. On multivariate analysis, BMI was first grouped as BMI ⬍25 kg/m2 vs 25-29.9 kg/m2 vs 30-34.9 kg/m2 vs ⱖ35 kg/m2. Subsequently, owing to few observations, the mildly obese and moderately/severely obese categories were collapsed into a single stratum. The Bonferroni method was used to adjust for multiple comparisons in the analysis of variance model. Power calculations addressed the effect size that would have been required to achieve 80% power in a 2-sided comparison with an ␣ of 0.05. All statistical tests were performed using the Statistical Package for Social Sciences, version 16.0 (SPSS, Chicago, IL). Moreover, all tests were 2-sided, with a significance level set at .05. Power analyses were performed with NQuery Advisor, version 4 (Statistical Solutions, Saugus, MA). 616

RESULTS Table 1 lists the clinical and pathologic characteristics of the study population. The patient age range was 37-77 years (mean 63, median 64). Overall, 579 patients (37.7%) had a BMI of ⬍25 kg/m2, 789 (51.3%) had a BMI of 25-29.9 kg/m2, 148 (9.6%) had a BMI of 30-34.9 kg/m2, and 22 (1.4%) had a BMI of ⱖ35 kg/m2. The preoperative PSA level was 0.7-108.6 ng/mL (mean 8.1, median 6.2). Most patients had a biopsy Gleason score of ⱕ6 (48.9%) vs 3⫹4 in 30.6%, and most had clinical Stage T1c (84.9%). The prostate volume was 12-220 cm3 (mean 46, median 41). The overall PSM rate was 14.1%. It was 7.9% for those with in organ-confined disease and 21.3% for those with extraprostatic disease. ECE was reported in 328 patients (21.3%), SVI in 123 (8%), and LNI in 56 patients (3.6%). Data stratified according to BMI category revealed that PSA was the only cancer-related variable that was significantly greater statistically in patients with a BMI ⱖ35 kg/m2 compared with patients with a lower BMI (analysis of variance, P ⫽ .04). Nonetheless, after adjusting for multiple comparisons, the recorded P value (P ⫽ .04) did not reach the required significance (P ⫽ .007). Categorically, coded BMI had no statistically significant effect on the rate of ECE, SVI, LNI, or PSMs. After adjusting for clinical stage, PSA level, biopsy Gleason score, patient age, prostate volume, and status of nervesparing surgery, categorically coded BMI was unrelated to ECE (P ⫽ .6), SVI (P ⫽ .3), LNI (P ⫽ .9), or the PSM rate (P ⫽ .8; Table 2). Similarly, no statistically significant effect was recorded when the obese BMI categories (mild and moderate/severe obesity) were collapsed into a single stratum. Finally, continuously coded BMI had no effect on either of the examined endpoints (Table 2). The power analyses revealed that the proportion of men with a BMI that exceeded normal weight or overweight was insufficient to achieve 80% power. For example, for ECE, the rate of ECE in the mildly and moderately/severely obese men would have needed to be 25% (instead of 20.6%) to achieve 80% power with an ␣ of 0.05.

COMMENT Several North American studies have addressed the effect of BMI on PCa characteristics. Most of these studies found that a greater BMI predisposes a patient to a more advanced pathologic stage.5,6,8,10 This finding is important and indicates that obese patients with PCa might warrant different treatment considerations than their nonobese counterparts. For example, greater rates of ECE in obese patients might represent a relative contraindication for performing nervesparing surgery. Similarly, a greater rate of ECE and SVI might represent a relative indication for external beam radiotherapy instead of RP. Finally, a greater rate of LNI in obese patients might be indicative of a greater need for extended pelvic lymph node dissection in this patient group. Taken together, obese patients might warrant different treatment considerations than their nonobese counterparts. UROLOGY 73 (3), 2009

Table 1. Clinical and pathologic characteristics of 1538 men undergoing radical prostatectomy stratified by body mass index

Parameter

Entire Cohort

Normal Weight (BMI ⬍25 kg/m2)

Patients (n) 1538 579 (37.7) Preoperative characteristics Age (y) Mean 63 63 Median 64 65 Range 37-77 43-77 Preoperative PSA (ng/mL) Mean 8.1 7.7 Median 6.2 6.1 Range 0.7-108.6 1.0-84.0 Biopsy Gleason score ⱕ6 752 (48.9) 279 (48.2) 3⫹4 470 (30.6) 185 (32) 4⫹3 175 (11.4) 58 (10) ⱖ8 141 (9.2) 57 (9.8) Clinical stage T1c 1306 (84.9) 489 (84.5) ⱖT2 215 (14) 86 (14.9) Missing 17 (1.1) 4 (0.7) Prostate volume (cm3) Mean 46 43 Median 41 40 Range 12-220 15-143 Pathologic outcome characteristics ECE 328 (21.3) 126 (21.8) SVI 123 (8) 42 (7.3) pN stage pN0 823 (53.5) 312 (53.9) pN1 56 (3.6) 21 (3.6) pNx 659 (42.8) 246 (42.5) Prostatectomy Gleason score ⱕ6 456 (29.6) 173 (29.9) 3⫹4 796 (51.8) 303 (52.3) 4⫹3 221 (14.4) 80 (13.8) ⱖ8 64 (4.2) 22 (3.8) Nerve sparing None 81 (5.3) 28 (4.8) Unilateral 409 (26.6) 155 (26.8) Bilateral 1048 (68.1) 396 (68.4) PSM rate Overall 217 (14.1) 82 (14.2) OCD 86/1085 (7.9) 34/410 (8.3) ECE 70/328 (21.3) 28/126 (22.2) SVI 59/122 (48.4) 19/42 (45.2)

Moderately Overweight Mildly or Severely (BMI Obese (BMI Obese P 25-29.9 kg/m2) 30-34.9 kg/m2) (BMI ⱖ35 kg/m2) Value 789 (51.3)

148 (9.6)

22 (1.4) .3ⴱ

63 64 37-77

63 64 49-73

61 62 48-70 .04ⴱ

8.2 6.4 0.7-74.6

8.0 6.0 1.0-108.6

12.2 7.2 3.4-94.2 .3†

382 (48.4) 243 (30.8) 98 (12.4) 66 (8.4)

79 (53.4) 39 (26.4) 17 (11.5) 13 (8.8)

12 (54.5) 3 (13.6) 2 (9.1) 5 (22.7)

671 (85) 110 (14.1) 8 (1)

129 (87.2) 15 (10.1) 4 (2.7)

17 (77.3) 4 (18.2) 1 (4.5)

.5†

1.0ⴱ 47 42 12-220

49 42 15-138

48 42 23-120

167 (21.2) 63 (8)

32 (21.6) 13 (8.8)

3 (13.6) 5 (22.7)

416 (52.7) 30 (3.8) 343 (43.5)

82 (55.4) 4 (2.7) 62 (41.9)

13 (59.1) 1 (4.5) 8 (36.4)

.8† .08† 1.0†

.8† 382 (48.4) 243 (30.8) 98 (12.4) 66 (8.4)

46 (31.3) 73 (49.3) 23 (15.5) 6 (4.1)

5 (22.7) 11 (50) 3 (13.6) 3 (13.6)

41 (5.2) 214 (27.1) 534 (67.7)

10 (6.8) 33 (22.3) 105 (70.9)

2 (9.1) 7 (31.8) 13 (59.1)

109 (13.8) 46 (8.2) 36 (21.6) 26 (41.9)

21 (14.2) 5 (4.9) 6 (18.8) 10 (76.9)

5 (22.7) 1 (7.1) 0 (0) 4 (80)

.8†

.8†

BMI, body mass index; PSA, prostate-specific antigen; ECE, extracapsular extension; SVI, seminal vesicle invasion; PSM, positive surgical margin; OCD, organ-confined disease. Data presented as number of patients with percentages in parentheses, unless otherwise noted.

Thus, it is crucial to know the effect of BMI on the different patient groups, because the strength of this effect appears to differ according to the continent of origin. For example, BMI exerted no effect on the pathologic stage or rate of biochemical recurrences after RP in a European cohort of 293 patients.17 However, the limited sample size might have undermined the validity of the study. Only 2 other European studies have examined the effect of BMI on patients UROLOGY 73 (3), 2009

treated with RP.16,18 One of these (n ⫽ 1814) failed to corroborate the effect of BMI on pathologic grade.16 The other also failed to show that the prediction of the rate of biochemical recurrence is affected by BMI.18 The findings of these studies are reassuring. They indicate that pathologic Gleason grade appears to be unaffected by BMI. Moreover, it appears that RP surgeons are capable of adjusting the type of prostatectomy to ensure that the biochemical recurrence 617

Table 2. Multivariate logistic regression analyses of effect of BMI on various pathologic characteristics after adjusting for preoperative and operative factors BMI (kg/m2) Continuously coded OR 95% CI P value Categorically coded 25-29.9 vs ⬍25 OR 95% CI P value 30-34.9 vs ⬍25 OR 95% CI P value ⱖ35 vs ⬍25 OR 95% CI P value Categorically coded 25-29.9 vs ⬍ 25 OR 95% CI P value ⱖ30 vs ⬍25 OR 95% CI P value

ECE

SVI

LNI

PSM

1.02 0.97-1.06 .5 .6

1.03 0.97-1.10 .3 .3

0.98 0.90-1.07 .7 .9

1.03 0.98-1.08 .3 .8

0.99 0.75-1.31 .9

1.04 0.67-1.63 .9

0.96 0.52-1.78 1.0

0.95 0.69-1.31 .8

1.16 0.72-1.87 .5

1.24 0.60-2.60 .6

0.77 0.24-2.41 .7

0.98 0.57-1.70 1.0

0.48 0.13-1.82 .3 1.0

3.52 0.94-13.26 .06 .5

0.44 0.04-4.50 .2 .8

1.79 0.57-5.61 .3 .9

0.99 0.75-1.31 .9

1.04 0.67-1.63 .9

0.97 0.52-1.79 .9

0.95 0.69-1.31 .8

1.05 0.67-1.66 .8

1.49 0.77-2.91 .2

0.69 0.24-1.99 .5

1.07 0.64-1.78 .8

LNI, lymph node invasion; OR, odds ratio; CI, confidence interval; other abbreviations as in Table 1.

rates are also unaffected by the BMI. However, this hypothesis cannot be fully validated without examining the effect of BMI on the PCa pathologic characteristics and the PSM rate at RP. To address this issue, we performed a formal analysis of the effect of BMI on the 3 pathologic stages (ECE, SVI, and LNI) and on the rate of PSMs. Because BMI can be coded as a continuous or categorical variable, we tested its effect in 2 separate analyses. Both analyses failed to identify BMI as either a statistically significant or an independent predictor of any of the 4 examined endpoints. Moreover, the effect of collapsing the mildly obese category and infrequent (n ⫽ 22) moderate/severe obese patient stratum also failed to result in either a statistically significant or an independent predictor status for BMI. These findings provide conclusive evidence that being overweight or obesity of any degree does not predispose to unfavorable pathologic characteristics or to an elevated PSM rate. Thus, overweight or obese individuals should be given the same treatment considerations as their nonobese counterparts. Our findings pertaining to the lack of effect of BMI on PCa pathologic characteristics and on the PSM rate corroborate previous European studies addressing cancer control endpoints at or after RP. Moreover, the lack of effect of BMI on PCa characteristics in general agrees with several other studies that focused on populations other than men from the United States. For example, obesity was found to be unrelated to the serum PSA level before PCa diagnosis,29 PCa incidence,3,15 or Gleason grade at biopsy.16 Our findings are in disagreement with the results of sev618

eral North American studies, which virtually uniformly suggested that obese patients treated with RP have a more unfavorable stage and grade compared with their nonobese counterparts.5,6,10 The reason for the observed discrepancy between the United States and Europe or other countries might be because of the differences in the prevalence and magnitude of obesity between the United States and other countries.1,19,20 However, no formal analyses have examined this hypothesis. It is also conceivable that obesity might be linked to another trait in U.S. men. This relationship might predominantly exist in the United States. Thus, its manifestations would not be apparent in analyses reported from other countries. Additional studies attempting to investigate the location-specific effect of obesity may be warranted. Our study had some limitations. For example, we had no data available regarding the ethnic background of the individual patients. This information could be of special interest, because in multiethnic populations, some subgroups might have more unfavorable PCa characteristics than others.8,9,30 However, although we did not explicitly documented race, the vast majority of the patients of our study cohort were white. Thus, the number of Asians and blacks was very small and surely did not exceed 1% of the entire cohort. The relatively small number of mildly obese patients (n ⫽ 148), and the even smaller number of moderately or severely obese patients (n ⫽ 22), representing an overall 11% of the entire study population, might be another limitation of our study. The power analyses revealed that the rates of ECE and SVI were not great enough in the obese UROLOGY 73 (3), 2009

patients to achieve 80% power in the study cohort. Most of the studies from the United States, from which strong positive findings were reported, had a greater proportion of obese men in their study cohorts (range 17%-34%).4,8,9,11,13 Therefore, we could not entirely exclude the possibility that a greater proportion of obese men in our study cohort might have altered our results. However, because the frequency and magnitude of obesity in most European countries is not as severe as in the United States, this problem can be hardly overcome. The single-institution nature of our report can be regarded as both a limitation and as a strength. One or more selection biases might be more operational at a single institution vs multiple centers. Conversely, standardized surgical technique and standardized specimen evaluation at a single institution represent strengths. Additional strengths correspond to the contemporary nature of the present report, which ensured that our findings are most applicable to future projects.

CONCLUSIONS Obese patients treated with open RP in Europe did not show advanced or adverse pathologic features compared with their nonobese counterparts. Therefore, RP provides the same potential of cancer control for obese patients as for nonobese patients and, accordingly, should not be denied to obese patients because of the fear of inadequate cancer control. References 1. Ogden CL, Carroll MD, Curtin LR, et al. Prevalence of overweight and obesity in the United States, 1999-2004. JAMA. 2006;295: 1549-1555. 2. Seidell JC. Obesity in Europe: scaling an epidemic. Int J Obes Relat Metab Disord. 1995;19(suppl 3):S1-S4. 3. Rapp K, Schroeder J, Klenk J, et al. Obesity and incidence of cancer: a large cohort study of over 145 000 adults in Austria. Br J Cancer. 2005;93:1062-1067. 4. Calle EE, Rodriguez C, Walker-Thurmond K, et al. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of US adults. N Engl J Med. 2003;348:1625-1638. 5. Boorjian SA, Crispen PL, Carlson RE, et al. Impact of obesity on clinicopathologic outcomes after robot-assisted laparoscopic prostatectomy. J Endourol. 2008;22:1471-1476. 6. Amling CL, Kane CJ, Riffenburgh RH, et al. Relationship between obesity and race in predicting adverse pathologic variables in patients undergoing radical prostatectomy. Urology. 2001;58:723-728. 7. Amling CL, Riffenburgh RH, Sun L, et al. Pathologic variables and recurrence rates as related to obesity and race in men with prostate cancer undergoing radical prostatectomy. J Clin Oncol. 2004;22: 439-445. 8. Freedland SJ, Aronson WJ, Kane CJ, et al. Impact of obesity on biochemical control after radical prostatectomy for clinically localized prostate cancer: a report by the Shared Equal Access Regional Cancer Hospital database study group. J Clin Oncol. 2004;22:446453. 9. Freedland SJ, Terris MK, Presti JC Jr, et al. Obesity and biochemical outcome following radical prostatectomy for organ confined disease with negative surgical margins. J Urol. 2004; 172:520-524.

UROLOGY 73 (3), 2009

10. Castle EP, Atug F, Woods M, et al. Impact of body mass index on outcomes after robot assisted radical prostatectomy. World J Urol. 2008;26:91-95. 11. Motamedinia P, Korets R, Spencer BA, et al. Body mass index trends and role of obesity in predicting outcome after radical prostatectomy. Urology. Epub 2008 Jul 2. 12. Rohrmann S, Roberts WW, Walsh PC, et al. Family history of prostate cancer and obesity in relation to high-grade disease and extraprostatic extension in young men with prostate cancer. Prostate. 2003;55:140-146. 13. Siddiqui SA, Inman BA, Sengupta S, et al. Obesity and survival after radical prostatectomy: a 10-year prospective cohort study. Cancer. 2006;107:521-529. 14. Engeland A, Tretli S, Bjorge T. Height, body mass index, and prostate cancer: a follow-up of 950000 Norwegian men. Br J Cancer. 2003;89:1237-1242. 15. Andersson SO, Wolk A, Bergstrom R, et al. Body size and prostate cancer: a 20-year follow-up study among 135,006 Swedish construction workers. J Natl Cancer Inst. 1997;89:385-389. 16. Gallina A, Karakiewicz PI, Hutterer GC, et al. Obesity does not predispose to more aggressive prostate cancer either at biopsy or radical prostatectomy in European men. Int J Cancer. 2007;121: 791-795. 17. Paaskesen CE, Borre M. Body mass index and prognostic markers at radical prostatectomy. Scand J Urol Nephrol. 2008;42:230236. 18. Chun FK, Briganti A, Graefen M, et al. Body mass index does not improve the ability to predict biochemical recurrence after radical prostatectomy. Eur J Cancer. 2007;43:375-382. 19. Berghofer A, Pischon T, Reinhold T, et al. Obesity prevalence from a European perspective: a systematic review. BMC Pub Health. 2008;8:200-209. 20. Buschemeyer WC III, Freedland SJ. Obesity and prostate cancer: epidemiology and clinical implications. Eur Urol. 2007;52:331343. 21. Graefen M, Walz J, Huland H. Open retropubic nerve-sparing radical prostatectomy. Eur Urol. 2006;49:38-48. 22. Bazinet M, Zheng W, Begin LR, et al. Morphologic changes induced by neoadjuvant androgen ablation may result in underdetection of positive surgical margins and capsular involvement by prostatic adenocarcinoma. Urology. 1997;49:721-725. 23. Heidenreich A, Aus G, Bolla M, et al. EAU guidelines on prostate cancer. Eur Urol. 2008;53:68-80. 24. Briganti A, Chun FK, Salonia A, et al. Validation of a nomogram predicting the probability of lymph node invasion among patients undergoing radical prostatectomy and an extended pelvic lymphadenectomy. Eur Urol. 2006;49:1019-1027. 25. McNeal JE, Redwine EA, Freiha FS, et al. Zonal distribution of prostatic adenocarcinoma: correlation with histologic pattern and direction of spread. Am J Surg Pathol. 1988;12:897-906. 26. Gleason DF, Mellinger GT. Prediction of prognosis for prostatic adenocarcinoma by combined histological grading and clinical staging. J Urol. 1974;111:58-64. 27. Greene FL, for the American Joint Committee on Cancer. Updating the strategies in cancer staging. Bull Am Coll Surg. 2002;87:13-15. 28. WHO Expert Committee. Physical status: the use and interpretation of anthropometry—report of a WHO Expert Committee. WHO Tech Rep Ser. 1995;854:1-452. 29. Hutterer G, Perrotte P, Gallina A, et al. Body mass index does not predict prostate-specific antigen or percent free prostate-specific antigen in men undergoing prostate cancer screening. Eur J Cancer. 2007;43:1180-1187. 30. Ravery V, Dominique S, Hupertan V, et al. Prostate cancer characteristics in a multiracial community. Eur Urol. 2008;53:533-538.

619