Prostate Size and Risk of High-Grade, Advanced Prostate Cancer and Biochemical Progression After Radical Prostatectomy: A Search Database St
http://www.100md.com
《临床肿瘤学》
the Department of Urology, Johns Hopkins School of Medicine
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health
Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins Medical Institutions, Baltimore, MD
Department of Surgery, Veterans Administration Medical Center
Section of Urology, Medical College of Georgia, Augusta, GA
Urology Section, Department of Surgery, Veterans Administration, Greater Los Angeles Healthcare System
Department of Urology, University of California, Los Angeles, School of Medicine, Los Angeles
Department of Urology, San Diego Naval Hospital, San Diego
Department of Urology, Stanford University School of Medicine
Urology Section, Department of Surgery, Veterans Administration Medical Center, Palo Alto
Urology Section, Department of Surgery, Veterans Administration Medical Center
Department of Urology, University of California, San Francisco School of Medicine, San Francisco, CA
ABSTRACT
PURPOSE: Prostate growth and differentiation are under androgenic control, and prior studies suggested that tumors that develop in hypogonadal men are more aggressive. We examined whether prostate weight was associated with tumor grade, advanced disease, or risk of biochemical progression after radical prostatectomy (RP).
PATIENTS AND METHODS: We evaluated the association of prostate weight with pathologic tumor grade, positive surgical margins, extracapsular disease, and seminal vesicle invasion using logistic regression and with biochemical progression using Cox proportional hazards regression among 1,602 men treated with RP between 1988 and 2003 at five equal-access medical centers, which composed the Shared Equal Access Regional Cancer Hospital (SEARCH) Database.
RESULTS: In outcome prediction models including multiple predictor variables, it was found that the predictor variable of prostate weight was significantly inversely associated with the outcomes of high-grade disease, positive surgical margins, extracapsular extension (all P .004), and biochemical progression (comparing prostate weight < 20 v 100 g: relative risk = 8.43; 95% CI, 2.9 to 24.0; P < .001). Similar associations were seen between preoperative transrectal ultrasound–measured prostate volume and high-grade disease, positive surgical margins, extracapsular extension (all P .005), seminal vesicle invasion (P = .07), and biochemical progression (P = .06).
CONCLUSION: Men with smaller prostates had more high-grade cancers and more advanced disease and were at greater risk of progression after RP. These results suggest that prostate size may be an important prognostic variable that should be evaluated for use pre- and postoperatively to predict biochemical progression.
INTRODUCTION
Prostate growth and differentiation are highly dependent on sex steroid hormones, particularly dihydrotestosterone (DHT).1 In dogs, DHT results in marked prostate enlargement, especially when combined with estrogens.2 In humans, inhibition of 5-alpha reductase type II, which catalyzes the conversion of testosterone to the more potent androgen, DHT reduces prostate volume by approximately 25%,3 whereas chemical castration reduces volume by more than 50%,4 suggesting that at least maintenance, if not growth, of the adult human prostate is dependent on DHT.
Among men with metastatic disease, lower pretreatment serum testosterone concentrations have consistently been associated with worse treatment response and worse survival.5-9 Among men with clinically localized disease, some studies found that low serum testosterone concentrations were associated with reduced recurrence rates after primary therapy,10 whereas other studies found higher recurrence rates, more advanced disease, or higher grade disease,11-14 and yet other studies found no association with pathologic stage.15 However, examination of serum testosterone concentrations at the time of diagnosis may not reflect intraprostatic androgenicity, androgenicity during the time of tumor development, or life-long cumulative androgenicity. Thus, simply comparing outcomes between men with low and high serum androgen concentrations may not lend insight into the association between low androgenicity and prostate cancer aggressiveness. We hypothesized that prostate size, which is largely under androgenic control, may be a better surrogate of long-term in vivo androgenicity than serum androgen concentrations at the time of diagnosis.
To indirectly test whether low androgenicity was associated with more aggressive prostate cancer, we examined the association of prostate weight with pathologic grade and stage and biochemical progression after radical prostatectomy (RP). We further hypothesized that if smaller prostate size was associated with high-grade and advanced disease, then prostate size might be a valuable prognostic marker among men with prostate cancer. To accomplish this, we used data from the Shared Equal Access Regional Cancer Hospital (SEARCH) Database of men treated with RP.16
PATIENTS AND METHODS
Study Population and Assessment of Prostate Weight and Other Clinicopathologic Variables
After obtaining institutional review board approval from each institution for data abstraction, data from patients treated with RP from 1988 to 2003 at the Veterans Affairs Medical Centers in West Los Angeles, Palo Alto, and San Francisco, CA, and Augusta, GA, and the San Diego Naval Hospital were combined into the SEARCH Database.16 This database includes information on patient age at the time of surgery, race, height, weight, clinical stage, biopsy Gleason grade, preoperative prostate-specific antigen (PSA), surgical specimen pathology (specimen weight, tumor grade, stage, and surgical margin status), and follow-up PSA data for a mean and median of 46 and 34 months, respectively (range, 1 to 187 months). Patients treated with preoperative androgen deprivation or radiation therapy were excluded. Of 2,028 patients in the SEARCH Database, 400 with missing RP specimen weight data and 26 men diagnosed from a transurethral resection specimen (clinical stage T1a/T1b) were excluded, resulting in a study population of 1,602 men.
The prostatectomy specimens were sectioned per each institution's protocol.16 All institutions determined prostate weight by measurement of the gross RP specimen weight, including seminal vesicles and tips of the vasa. Patients were observed per each institution's protocol to determine biochemical progression, which was defined as a single PSA level of more than 0.2 ng/mL, two PSA concentrations at 0.2 ng/mL, or secondary treatment for an elevated postoperative PSA.17 Patients with no follow-up data (n = 80) were included for evaluating differences in preoperative and pathologic characteristics but not for biochemical progression.
Statistical Analysis
Prostate weight was entered into the models as a series of indicator variables for 20-g intervals. We also explored the association between prostate size and outcomes by entering prostate weight as a continuous variable, after logarithmic transformation, into all multivariable models. Odds ratios (OR) of pathologic variables were estimated for weight groups using logistic regression. The primary pathologic outcome of pathologic Gleason sum was defined as 7 (high grade) versus less than 7 (low grade). We also considered the following binary pathologic outcomes: positive surgical margins, extracapsular extension, and seminal vesicle invasion; few men had lymph node metastases (n = 26). We adjusted for age at RP (5-year intervals), race (black or other v white), body mass index (BMI; 25 to 29.9, 30 to 34.9, and 35.0 v < 25 kg/m2), height (tertiles: < 69, 69 to 72, and > 72 inches), and year of surgery (3-year intervals). In the model predicting pathologic Gleason sum, we also adjusted for clinical stage (T2b/T2c/T3 v T1c/T2a)18 and the following binary pathologic outcomes: positive surgical margins, extracapsular extension, and seminal vesicle invasion. For predicting the binary pathologic outcomes, we also adjusted for pathologic Gleason sum (7 and 8 to 10 v 2 to 6). In separate models, we additionally adjusted for presurgery PSA concentration (10 to < 20 and 20 v < 10 ng/mL).19 We tested for trend by entering median prostate weight of each 20-g interval as a continuous term into the model and evaluating the coefficient by the Wald test.
Time to biochemical progression was compared across the prostate weight categories using Kaplan-Meier plots and the log-rank test. To estimate the relative risk (RR) of biochemical progression associated with prostate weight, indicator variables for 20-g intervals were entered into a Cox proportional hazards regression model adjusting for age, race, BMI, height, year of surgery, clinical stage, pathologic Gleason sum, positive surgical margins, extracapsular extension, seminal vesicle invasion, and lymph node metastasis.
We tested for interactions between prostate weight and age, race, BMI, PSA concentration, and height by including both main effect terms and an interaction term, which represented the cross product of the two main effect terms in the same model. For tests of interactions, we coded the covariates as follows: prostate weight (continuous variable after logarithmic transformation), age (continuous variable), race (black v nonblack), BMI (continuous variable), height (continuous variable), and PSA concentration (continuous variable after logarithmic transformation). We tested the coefficient of the cross-product term by the Wald test.
The associations between other clinical, demographic, and anthropometric factors and outcomes have previously been studied in the SEARCH Database.16,20,21 The distribution of all clinical and pathologic variables was similar among the centers contributing to the SEARCH Database. Therefore, data from all centers were combined for analyses. All statistical analyses were performed using STATA 8.0 (Stata Corp, College Station, TX).
RESULTS
Mean and median RP specimen weights were 44 g (standard deviation, 23 g) and 39 g, respectively (range, 5 to 255 g). Mean patient age was 62.5 years (standard deviation, 6.8 years; Table 1). Smaller prostate weight was associated with younger age (P < .001) and lower preoperative PSA (P < .001).
Pathologic Gleason Sum and Other Pathologic Characteristics
The odds of high-grade disease, positive surgical margins, and extracapsular extension increased with decreasing prostate weight after adjusting for age at RP; the associations were enhanced after multivariable adjustment (Table 2). After including in the model preoperative PSA, the expression of which is under androgenic regulation and is strongly correlated with prostate volume, smaller prostate weight was associated with statistically significantly increased odds of high-grade disease in the RP specimen (comparing < 20 v 100 g: OR = 7.61; per 1-unit increase in log-transformed prostate weight, OR = 0.46; P < .001; Table 2). Even after adjusting for the higher prevalence of high-grade disease, smaller prostate weight was associated with statistically significantly increased odds of positive surgical margins (comparing < 20 v 100 g: OR = 3.09; per 1-unit increase in log-transformed prostate weight, OR = 0.46; P < .001) and extracapsular extension (comparing < 20 v 100 g: OR = 1.73; per 1-unit increase in log-transformed prostate weight, OR = 0.60; P = .004; Table 2). There was no statistically significant association between prostate weight and seminal vesicle invasion (P = .32). Similar trends were noted at each of the centers contributing data to the SEARCH Database. Both biopsy Gleason sum (P = .10) and clinical stage (P = .13) were inversely associated with prostate weight, although the associations were not statistically significant.
Biochemical Progression
Mean and median follow-up times among men without progression were 46 months (standard deviation, 40 months) and 34 months, respectively. During this time, 404 patients (27%) experienced disease progression. Smaller prostate weight was statistically significantly associated with a greater risk of biochemical progression (P = .007, Fig 1). Adjusting for age at RP, smaller prostate weight was associated with a statistically significantly increased risk of biochemical progression (comparing < 20 v 100 g: RR = 3.94; per 1-unit increase in log-transformed prostate weight, RR = 0.60; P = .001), the association for which was even stronger after adjustment for patient, clinical, and pathologic characteristics (comparing < 20 v 100 g: RR = 5.89; per 1-unit increase in log-transformed prostate weight, RR = 0.56; P < .001; Table 3). After further adjusting for preoperative PSA, the RR of biochemical progression comparing less than 20 with 100 g increased to 8.43 (per 1-unit increase in log-transformed prostate weight, RR = 0.48; P < .001). Similar trends were noted at each of the centers contributing data to the SEARCH Database.
Influence of Extreme Prostate Weights on the Association of Prostate Weight With Pathologic Characteristics and Biochemical Progression
The association of prostate weight with adverse pathologic characteristics and biochemical progression seemed to be strongest when contrasting men with the largest ( 100 g) and the smallest prostates (< 20 g). To explore whether prostate weight was associated with adverse pathologic characteristics and biochemical progression for the majority of men who had prostate weights of 20 g or more and less than 100 g, we reanalyzed our data excluding men with prostate weights of less than 20 g or 100 g or more. When these men were excluded, on multivariable analysis, smaller prostate weight remained statistically significantly inversely associated with high-grade cancer (P < .001), positive surgical margins (P < .001), extraprostatic disease (P = .01), and biochemical progression (P = .01).
Influence of PSA on the Association of Prostate Weight With Pathologic Characteristics and Biochemical Progression
To explore whether these associations varied by preoperative PSA, we stratified by PSA concentration. There was no statistically significant interaction between prostate weight and PSA (all P > .1); within each PSA stratum (< 10, 10 to < 20, and 20 ng/mL), smaller prostate weight was associated with higher odds of high-grade disease, positive surgical margins, and extracapsular extension. There was no statistically significant interaction between preoperative PSA and prostate weight for predicting biochemical progression (P = .72); smaller prostate weight was statistically significantly associated with progression among men with low (n = 950, RR = 1.32 for 20-g decrease in prostate weight, P = .03), intermediate (n = 250, RR = 1.38, P = .03), and high PSA concentrations (n = 92, RR = 2.38, P = .003).
Larger prostate weight was associated with higher PSA concentration, which would potentially lead to earlier diagnosis because of elevated PSA as a biopsy indication. To exclude this potential for detection bias, we excluded all men who underwent biopsy solely because of an elevated PSA (clinical stage T1c) and reanalyzed the data examining only men who were biopsied because of an abnormal digital rectal examination. When only men with clinical stage T2/T3 tumors were examined, smaller prostate size remained statistically significantly associated with high-grade disease (comparing < 20 to 100 g: OR = 8.48; 95% CI, 1.74 to 41.42; P = .01), positive surgical margins (OR = 7.08; 95% CI, 1.23 to 40.80, P = .003), extracapsular extension (OR = 1.50; 95% CI, 0.24 to 9.22; P = .04; Table 4), and biochemical progression (RR = 11.75; 95% CI, 2.76 to 49.96; P = .001; Table 5).
Evaluation of Potential Interactions
Within each stratum of BMI, height, or race, the association of smaller prostate weights with high-grade disease, positive surgical margins, extracapsular extension, and biochemical progression was similar. Overall, there were no statistically significant interactions (all P .09) between prostate weight and BMI, height, age, or race for predicting adverse pathology or biochemical progression with one exception; there was a significant interaction (P = .04) between prostate size and age in that larger prostate weight was more strongly related to lower odds of seminal vesicle invasion among older men than among younger men.
Evaluation of PSA Density
PSA density (PSAD; PSA concentration divided by prostate weight) has been suggested to better predict biochemical progression after RP than PSA.22 Therefore, we examined whether prostate weight remained statistically significantly associated with biochemical progression after adjusting for PSAD. When PSAD (0.3 to 0.7 and > 0.7 v < 0.3 ng/mL/g)22 was substituted for PSA in the multivariable model, smaller prostate weight remained statistically significantly associated with biochemical progression (comparing < 20 v 100 g: RR = 3.22; 95% CI, 1.16 to 8.90; P = .01).
Transrectal Ultrasound–Measured Prostate Volume As a Preoperative Prognostic Factor
Data were available for 753 men on preoperative transrectal ultrasound (TRUS) prostate volume. There was a strong correlation between TRUS volume and RP specimen weight (Spearman, r = 0.71; P < .001). We evaluated associations between 20-g intervals of TRUS-measured prostate volume and pathologic findings and biochemical progression after RP. Because we sought to determine whether TRUS volume could be used preoperatively for risk assessment, postoperative variables (pathologic Gleason sum, positive surgical margin, extracapsular extension, seminal vesicle invasion, or lymph node positivity) were not included in the analyses. On multivariable analysis (adjusting for biopsy Gleason sum, clinical stage, age, year of surgery, preoperative PSA, race, BMI, and height), smaller preoperative TRUS volume was associated with high-grade disease in the RP specimen (P < .001), positive surgical margins (P < .001), extracapsular extension (P = .005), seminal vesicle invasion (P = .07), and biochemical progression (comparing < 20 to 100 cc: RR = 5.99; 95% CI, 0.77 to 46.68; P = .06).
DISCUSSION
In an equal-access population, we found that men with smaller prostates had more high-grade cancers and more advanced disease at the time of RP and were at significantly greater risk of biochemical progression. Similar findings were observed whether we examined pathologic RP specimen weight or preoperative TRUS-measured prostate volume. These findings suggest that prostate size may be an important prognostic variable in prostate cancer.
On the basis of several observations, we hypothesized that small prostate size, after adjusting for age, may be a surrogate of low in vivo androgenicity. First, prostate growth and differentiation are highly dependent on sex steroid hormones, particularly DHT.1 In patients with 5-alpha reductase type II deficiency and, subsequently, low to absent DHT levels, the prostate develops rudimentarily.23 In animal models, only androgen withdrawal therapy is capable of pharmacologically reducing prostate size.24 In humans, 5-alpha reductase type II inhibition reduces prostate size by approximately 25%, whereas medical castration results in a more than 50% reduction.3,4 Prostate size in animal models is most closely correlated with serum androgen concentrations and advancing age.25 In human cross-sectional studies, the only serum hormone concentrations that have been associated with prostate volume are free testosterone and estrogens.26 Moreover, shorter CAG repeats within the androgen receptor gene, which result in a more active androgen receptor, have been correlated with larger prostates.27 However, other factors, namely estrogen, insulin, and insulin-like growth factor 1, all of which have been associated with prostate cancer development and progression in some epidemiologic studies, also influence prostate size.26,28,29 Therefore, for a given age, a small prostate is likely associated with both lower androgenicity and lower overall intraprostatic growth factor concentrations. Whether this simply results in a hostile environment in which only aggressive cancers can grow or the lower growth factor/androgenic concentrations actually induce more aggressive cancers is unclear.
Although conflicting data exist, several studies found that men with genetic polymorphisms favoring increased androgenicity were at higher risk for developing prostate cancer.30,31 Among men with metastatic disease, lower pretreatment total testosterone concentrations have consistently been associated with worse response to treatment and worse survival.5-8 Among men with clinically localized disease undergoing radiation therapy, Zagars et al10 found that low testosterone concentrations were associated with decreased risk of developing metastatic disease. For patients undergoing RP, some studies found no association between serum testosterone concentration and pathologic stage,15 whereas others found that low testosterone concentrations were associated with a higher grade, advanced stage, and higher progression rates.11-14 However, serum testosterone concentrations at the time of diagnosis may not reflect either intraprostatic androgenicity or the cumulative life-long androgen exposure.
Smaller prostate weight was associated with higher grade disease and more advanced stage, and even after controlling for these findings, smaller prostate weight independently predicted biochemical progression. Similar findings were observed whether prostate size was measured by RP specimen weight or preoperative TRUS-measured prostate volume. Moreover, prostate weight remained a significant predictor of biochemical progression after adjustment for PSAD. The higher prevalence of high-grade and advanced disease among men with smaller prostates would be expected to result in worse outcomes after other forms of therapy, although further studies are needed to confirm this. Prostate size is often used to help guide decisions regarding therapy (discouragement of men with larger prostates from brachytherapy32 and encouragement of men with large prostates and associated urinary symptoms to undergo surgery). These practice patterns may theoretically impact treatment outcomes and should be taken into account by adjusting for prostate size when comparing results across treatments. Additionally, studies found that Veterans Affairs patients had smaller prostates than men at tertiary-care referral centers.33,34 These prostate size differences between centers may, in part, explain disparate results between centers, although this needs to be specifically addressed in future studies. Consequently, inclusion of prostate size into pre- and post-treatment nomograms would likely result in both improved risk assessment and improved generalizability of these nomograms to men treated in different practice settings.
It is possible that men with larger prostates had their tumors detected earlier because of PSA-driven biopsies resulting from PSA elevation from an enlarged gland. This lead-time bias would be expected to result in better outcomes. Several studies found that prostate volumes more than 75 cc were associated with more favorable outcomes after RP35-37; all of the studies concluded that lead-time bias accounted for most or all of the results. Several findings suggest that lead-time bias may not explain all of the findings in the current study. First, after excluding men with PSA-detected tumors (clinical stage T1c) and examining only men with abnormal digital rectal examinations, the associations between smaller prostate size and higher grade advanced disease and biochemical progression remained statistically significant. Second, men with larger prostates were older. If one proposes that men with larger prostates had their cancers detected earlier, one would expect these patients to be younger and not older. Third, men with smaller prostates had more high-grade cancers. Whether high-grade cancers result from dedifferentiation of low-grade cancers over time or develop de novo is unknown. However, recent data suggest that, on rebiopsy of men in watchful waiting programs, tumor grade remains stable over several years.38,39 Although this does not exclude grade progression over the long term, it seems unlikely that grade progression could explain the 7.5-fold increased odds of high-grade disease among men with prostates less than 20 g v 100 g. Finally, even after controlling for and stratifying by preoperative PSA and eliminating men with PSA-detected cancers, men with smaller prostates had higher grade and more advanced disease. It is important to note that the validity of prostate size as a prognostic marker is independent of the reason for the worse outcomes among men with smaller prostates. Whether the worse outcomes were a result of detection bias, more aggressive tumors developing in a low growth factor/androgenic environment, or some alternate explanation, the end result remains the same; men with smaller prostates had more high-grade advanced disease and higher progression rates.
It has been argued that a marked reduction in androgenicity may interfere with prostate cancer grading, resulting in artifactual upgrading that does not reflect tumor biology.40 In the current study, it seems unlikely that the higher grade tumors among men with smaller prostates were a result of grading artifacts because these high-grade cancers were associated with other signs of aggressive behaviors, such as advanced pathologic stage and higher progression rates.
Serum and tissue androgen concentrations, androgen receptor concentrations and activity levels, and the androgen receptor gene CAG repeat length were unavailable in the SEARCH Database. Thus, we are unable to comment on the true in vivo androgenicity of our patients, and thus, it remains a hypothesis that smaller prostate size is a surrogate of decreased in vivo androgenicity. Seminal vesicles and tips of the vasa were included in the RP specimen weight. Given that we examined prostate weight as a categoric variable, even if seminal vesicle weight had not been included, most patients would be in the same or, at most, one lower weight category. Thus, the progressive overall trends observed for decreasing weight and worse outcomes cannot be explained by inclusion of seminal vesicle weight in the total RP specimen weight. Moreover, seminal vesicle weight is also under androgenic control and is often used as an in vivo bioassay for androgenicity.41,42 Thus, the total RP specimen weight, which includes prostate and seminal vesicle weight, may be a better surrogate of in vivo androgenicity than prostate weight alone. In the current study, all patients had prostate cancer and underwent surgery for their disease. Therefore, we are unable to comment on any possible relationship between androgenicity and the risk of developing prostate cancer. In addition, mean follow-up time was relatively short. Finally, we examined biochemical progression as our end point. Several studies found that, among men treated with RP, a shorter time to biochemical progression is associated with increased risk for developing metastatic disease.43,44 Although recent studies suggested that postoperative PSA doubling time may be a better surrogate than biochemical progression for prostate cancer–specific mortality,45,46 PSA doubling times were not available in the SEARCH Database.
In conclusion, we found that men with smaller prostate sizes had higher grade and more advanced disease at the time of RP and a greater risk of biochemical progression. Because similar results were found whether prostate size was measured as RP specimen weight or preoperative TRUS-measured prostate volume, the current data support the evaluation of prostate size as a pre- and postoperative prognostic marker in prostate cancer.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by the Department of Veterans Affairs, National Institutes of Health Grant No. R01CA100938 (W.J.A.), National Institutes of Health Specialized Programs of Research Excellence Grant No. P50 CA92131-01A1 (W.J.A.), the Georgia Cancer Coalition (M.K.T.), Center for Prostate Disease Research grant from the United States Army Medical Research and Materiel Command (C.L.A.), the Department of Defense, Prostate Cancer Research Program Grant No. PC030666 (S.J.F.), and the American Foundation for Urological Disease/American Urological Association Education and Research Scholarship Award (S.J.F.).
The views and opinions of and endorsements by the authors do not reflect those of the US Army or the Department of Defense.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Marker PC, Donjacour AA, Dahiya R, et al: Hormonal, cellular, and molecular control of prostatic development. Dev Biol 253:165-174, 2003
Isaacs JT, Coffey DS: Changes in dihydrotestosterone metabolism associated with the development of canine benign prostatic hyperplasia. Endocrinology 108:445-453, 1981
Thompson IM, Goodman PJ, Tangen CM, et al: The influence of finasteride on the development of prostate cancer. N Engl J Med 349:215-224, 2003
Noldus J, Ferrari M, Prestigiacomo A, et al: Effect of flutamide and flutamide plus castration on prostate size in patients with previously untreated prostate cancer. Urology 47:713-718, 1996
Chodak GW, Vogelzang NJ, Caplan RJ, et al: Independent prognostic factors in patients with metastatic (stage D2) prostate cancer: The Zoladex Study Group. JAMA 265:618-621, 1991
Ribeiro M, Ruff P, Falkson G: Low serum testosterone and a younger age predict for a poor outcome in metastatic prostate cancer. Am J Clin Oncol 20:605-608, 1997
Chen SS, Chen KK, Lin AT, et al: The correlation between pretreatment serum hormone levels and treatment outcome for patients with prostatic cancer and bony metastasis. BJU Int 89:710-713, 2002
Iversen P, Rasmussen F, Christensen IJ: Serum testosterone as a prognostic factor in patients with advanced prostatic carcinoma. Scand J Urol Nephrol Suppl 157:41-47, 1994
Oefelein MG, Agarwal PK, Resnick MI: Survival of patients with hormone refractory prostate cancer in the prostate specific antigen era. J Urol 171:1525-1528, 2004
Zagars GK, Pollack A, von Eschenbach AC: Serum testosterone: A significant determinant of metastatic relapse for irradiated localized prostate cancer. Urology 49:327-334, 1997
D'Amico AV, Chen MH, Malkowicz SB, et al: Lower prostate specific antigen outcome than expected following radical prostatectomy in patients with high grade prostate and a prostatic specific antigen level of 4 ng/ml or less. J Urol 167:2025-2030, 2002
Massengill JC, Sun L, Moul JW, et al: Pretreatment total testosterone level predicts pathological stage in patients with localized prostate cancer treated with radical prostatectomy. J Urol 169:1670-1675, 2003
Hoffman MA, DeWolf WC, Morgentaler A: Is low serum free testosterone a marker for high grade prostate cancer? J Urol 163:824-827, 2000
Schatzl G, Madersbacher S, Thurridl T, et al: High-grade prostate cancer is associated with low serum testosterone levels. Prostate 47:52-58, 2001
Monda JM, Myers RP, Bostwick DG, et al: The correlation between serum prostate-specific antigen and prostate cancer is not influenced by the serum testosterone concentration. Urology 46:62-64, 1995
Freedland SJ, Amling CL, Dorey F, et al: Race as an outcome predictor after radical prostatectomy: Results from the Shared Equal Access Regional Cancer Hospital (SEARCH) database. Urology 60:670-674, 2002
Freedland SJ, Sutter ME, Dorey F, et al: Defining the ideal cutpoint for determining PSA recurrence after radical prostatectomy. Urology 61:365-369, 2003
Freedland SJ, Presti JC Jr, Terris MK, et al: Improved clinical staging system combining biopsy laterality and TNM stage for men with T1c and T2 prostate cancer: Results from the SEARCH database. J Urol 169:2129-2135, 2003
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
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 22:446-453, 2004
Freedland SJ, Presti JC Jr, Kane CJ, et al: Do younger men have better biochemical outcomes after radical prostatectomy? Urology 63:518-522, 2004
Freedland SJ, Wieder JA, Jack GS, et al: Improved risk stratification for biochemical recurrence after radical prostatectomy using a novel risk group system based on prostate specific antigen density and biopsy Gleason score. J Urol 168:110-115, 2002
Walsh PC, Madden JD, Harrod MJ, et al: Familial incomplete male pseudohermaphroditism, type 2: Decreased dihydrotestosterone formation in pseudovaginal perineoscrotal hypospadias. N Engl J Med 291:944-949, 1974
Coffey DS, Walsh PC: Clinical and experimental studies of benign prostatic hyperplasia. Urol Clin North Am 17:461-475, 1990
Berry SJ, Coffey DS, Ewing LL: Effects of aging on prostate growth in beagles. Am J Physiol 250:R1039-R1046, 1986
Partin AW, Oesterling JE, Epstein JI, et al: Influence of age and endocrine factors on the volume of benign prostatic hyperplasia. J Urol 145:405-409, 1991
Roberts RO, Bergstralh EJ, Cunningham JM, et al: Androgen receptor gene polymorphisms and increased risk of urologic measures of benign prostatic hyperplasia. Am J Epidemiol 159:269-276, 2004
Roberts RO, Jacobson DJ, Girman CJ, et al: Insulin-like growth factor I, insulin-like growth factor binding protein 3, and urologic measures of benign prostatic hyperplasia. Am J Epidemiol 157:784-791, 2003
Hammarsten J, Hogstedt B: Hyperinsulinaemia as a risk factor for developing benign prostatic hyperplasia. Eur Urol 39:151-158, 2001
Makridakis NM, Reichardt JK: Molecular epidemiology of androgen-metabolic loci in prostate cancer: Predisposition and progression. J Urol 171:S25-S28, 2004
Giovannucci E, Stampfer MJ, Krithivas K, et al: The CAG repeat within the androgen receptor gene and its relationship to prostate cancer. Proc Natl Acad Sci U S A 94:3320-3323, 1997
Merrick GS, Wallner KE, Butler WM: Permanent interstitial brachytherapy for the management of carcinoma of the prostate gland. J Urol 169:1643-1652, 2003
Nixon RG, Meyer GE, Brawer MK: Differences in prostate size between patients from University and Veterans Affairs Medical Center populations. Prostate 38:144-150, 1999
Terris MK, Prestigiacomo AF, Stamey TA: Comparison of prostate size in university and Veterans Affairs Health Care System patients with negative prostate biopsies. Urology 51:412-414, 1998
D'Amico AV, Whittington R, Malkowicz SB, et al: A prostate gland volume of more than 75 cm3 predicts for a favorable outcome after radical prostatectomy for localized prostate cancer. Urology 52:631-636, 1998
Foley CL, Bott SR, Thomas K, et al: A large prostate at radical retropubic prostatectomy does not adversely affect cancer control, continence or potency rates. BJU Int 92:370-374, 2003
Stamey TA, Yemoto CM, McNeal JE, et al: Prostate cancer is highly predictable: A prognostic equation based on all morphological variables in radical prostatectomy specimens. J Urol 163:1155-1160, 2000
Carter CA, Donahue T, Sun L, et al: Temporarily deferred therapy (watchful waiting) for men younger than 70 years and with low-risk localized prostate cancer in the prostate-specific antigen era. J Clin Oncol 21:4001-4008, 2003
Epstein JI, Walsh PC, Carter HB: Dedifferentiation of prostate cancer grade with time in men followed expectantly for stage T1c disease. J Urol 166:1688-1691, 2001
Thompson IM, Klein EA, Lippman SM, et al: Prevention of prostate cancer with finasteride: US/European perspective. Eur Urol 44:650-655, 2003
Skarda J: Bioassay of steroid hormone agonist and antagonist activities of anti-androgens on mammary gland, seminal vesicles and spleen of male mice. J Vet Med A Physiol Pathol Clin Med 50:204-212, 2003
Kousteni S, Chen JR, Bellido T, et al: Reversal of bone loss in mice by nongenotropic signaling of sex steroids. Science 298:843-846, 2002
Pound CR, Partin AW, Eisenberger MA, et al: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 281:1591-1597, 1999
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
D'Amico AV, Moul JW, Carroll PR, et al: Surrogate end point for prostate cancer-specific mortality after radical prostatectomy or radiation therapy. J Natl Cancer Inst 95:1376-1383, 2003
Ward JF, Blute ML, Slezak J, et al: The long-term clinical impact of biochemical recurrence of prostate cancer 5 or more years after radical prostatectomy. J Urol 170:1872-1876, 2003(Stephen J. Freedland, Wil)
Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health
Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins Medical Institutions, Baltimore, MD
Department of Surgery, Veterans Administration Medical Center
Section of Urology, Medical College of Georgia, Augusta, GA
Urology Section, Department of Surgery, Veterans Administration, Greater Los Angeles Healthcare System
Department of Urology, University of California, Los Angeles, School of Medicine, Los Angeles
Department of Urology, San Diego Naval Hospital, San Diego
Department of Urology, Stanford University School of Medicine
Urology Section, Department of Surgery, Veterans Administration Medical Center, Palo Alto
Urology Section, Department of Surgery, Veterans Administration Medical Center
Department of Urology, University of California, San Francisco School of Medicine, San Francisco, CA
ABSTRACT
PURPOSE: Prostate growth and differentiation are under androgenic control, and prior studies suggested that tumors that develop in hypogonadal men are more aggressive. We examined whether prostate weight was associated with tumor grade, advanced disease, or risk of biochemical progression after radical prostatectomy (RP).
PATIENTS AND METHODS: We evaluated the association of prostate weight with pathologic tumor grade, positive surgical margins, extracapsular disease, and seminal vesicle invasion using logistic regression and with biochemical progression using Cox proportional hazards regression among 1,602 men treated with RP between 1988 and 2003 at five equal-access medical centers, which composed the Shared Equal Access Regional Cancer Hospital (SEARCH) Database.
RESULTS: In outcome prediction models including multiple predictor variables, it was found that the predictor variable of prostate weight was significantly inversely associated with the outcomes of high-grade disease, positive surgical margins, extracapsular extension (all P .004), and biochemical progression (comparing prostate weight < 20 v 100 g: relative risk = 8.43; 95% CI, 2.9 to 24.0; P < .001). Similar associations were seen between preoperative transrectal ultrasound–measured prostate volume and high-grade disease, positive surgical margins, extracapsular extension (all P .005), seminal vesicle invasion (P = .07), and biochemical progression (P = .06).
CONCLUSION: Men with smaller prostates had more high-grade cancers and more advanced disease and were at greater risk of progression after RP. These results suggest that prostate size may be an important prognostic variable that should be evaluated for use pre- and postoperatively to predict biochemical progression.
INTRODUCTION
Prostate growth and differentiation are highly dependent on sex steroid hormones, particularly dihydrotestosterone (DHT).1 In dogs, DHT results in marked prostate enlargement, especially when combined with estrogens.2 In humans, inhibition of 5-alpha reductase type II, which catalyzes the conversion of testosterone to the more potent androgen, DHT reduces prostate volume by approximately 25%,3 whereas chemical castration reduces volume by more than 50%,4 suggesting that at least maintenance, if not growth, of the adult human prostate is dependent on DHT.
Among men with metastatic disease, lower pretreatment serum testosterone concentrations have consistently been associated with worse treatment response and worse survival.5-9 Among men with clinically localized disease, some studies found that low serum testosterone concentrations were associated with reduced recurrence rates after primary therapy,10 whereas other studies found higher recurrence rates, more advanced disease, or higher grade disease,11-14 and yet other studies found no association with pathologic stage.15 However, examination of serum testosterone concentrations at the time of diagnosis may not reflect intraprostatic androgenicity, androgenicity during the time of tumor development, or life-long cumulative androgenicity. Thus, simply comparing outcomes between men with low and high serum androgen concentrations may not lend insight into the association between low androgenicity and prostate cancer aggressiveness. We hypothesized that prostate size, which is largely under androgenic control, may be a better surrogate of long-term in vivo androgenicity than serum androgen concentrations at the time of diagnosis.
To indirectly test whether low androgenicity was associated with more aggressive prostate cancer, we examined the association of prostate weight with pathologic grade and stage and biochemical progression after radical prostatectomy (RP). We further hypothesized that if smaller prostate size was associated with high-grade and advanced disease, then prostate size might be a valuable prognostic marker among men with prostate cancer. To accomplish this, we used data from the Shared Equal Access Regional Cancer Hospital (SEARCH) Database of men treated with RP.16
PATIENTS AND METHODS
Study Population and Assessment of Prostate Weight and Other Clinicopathologic Variables
After obtaining institutional review board approval from each institution for data abstraction, data from patients treated with RP from 1988 to 2003 at the Veterans Affairs Medical Centers in West Los Angeles, Palo Alto, and San Francisco, CA, and Augusta, GA, and the San Diego Naval Hospital were combined into the SEARCH Database.16 This database includes information on patient age at the time of surgery, race, height, weight, clinical stage, biopsy Gleason grade, preoperative prostate-specific antigen (PSA), surgical specimen pathology (specimen weight, tumor grade, stage, and surgical margin status), and follow-up PSA data for a mean and median of 46 and 34 months, respectively (range, 1 to 187 months). Patients treated with preoperative androgen deprivation or radiation therapy were excluded. Of 2,028 patients in the SEARCH Database, 400 with missing RP specimen weight data and 26 men diagnosed from a transurethral resection specimen (clinical stage T1a/T1b) were excluded, resulting in a study population of 1,602 men.
The prostatectomy specimens were sectioned per each institution's protocol.16 All institutions determined prostate weight by measurement of the gross RP specimen weight, including seminal vesicles and tips of the vasa. Patients were observed per each institution's protocol to determine biochemical progression, which was defined as a single PSA level of more than 0.2 ng/mL, two PSA concentrations at 0.2 ng/mL, or secondary treatment for an elevated postoperative PSA.17 Patients with no follow-up data (n = 80) were included for evaluating differences in preoperative and pathologic characteristics but not for biochemical progression.
Statistical Analysis
Prostate weight was entered into the models as a series of indicator variables for 20-g intervals. We also explored the association between prostate size and outcomes by entering prostate weight as a continuous variable, after logarithmic transformation, into all multivariable models. Odds ratios (OR) of pathologic variables were estimated for weight groups using logistic regression. The primary pathologic outcome of pathologic Gleason sum was defined as 7 (high grade) versus less than 7 (low grade). We also considered the following binary pathologic outcomes: positive surgical margins, extracapsular extension, and seminal vesicle invasion; few men had lymph node metastases (n = 26). We adjusted for age at RP (5-year intervals), race (black or other v white), body mass index (BMI; 25 to 29.9, 30 to 34.9, and 35.0 v < 25 kg/m2), height (tertiles: < 69, 69 to 72, and > 72 inches), and year of surgery (3-year intervals). In the model predicting pathologic Gleason sum, we also adjusted for clinical stage (T2b/T2c/T3 v T1c/T2a)18 and the following binary pathologic outcomes: positive surgical margins, extracapsular extension, and seminal vesicle invasion. For predicting the binary pathologic outcomes, we also adjusted for pathologic Gleason sum (7 and 8 to 10 v 2 to 6). In separate models, we additionally adjusted for presurgery PSA concentration (10 to < 20 and 20 v < 10 ng/mL).19 We tested for trend by entering median prostate weight of each 20-g interval as a continuous term into the model and evaluating the coefficient by the Wald test.
Time to biochemical progression was compared across the prostate weight categories using Kaplan-Meier plots and the log-rank test. To estimate the relative risk (RR) of biochemical progression associated with prostate weight, indicator variables for 20-g intervals were entered into a Cox proportional hazards regression model adjusting for age, race, BMI, height, year of surgery, clinical stage, pathologic Gleason sum, positive surgical margins, extracapsular extension, seminal vesicle invasion, and lymph node metastasis.
We tested for interactions between prostate weight and age, race, BMI, PSA concentration, and height by including both main effect terms and an interaction term, which represented the cross product of the two main effect terms in the same model. For tests of interactions, we coded the covariates as follows: prostate weight (continuous variable after logarithmic transformation), age (continuous variable), race (black v nonblack), BMI (continuous variable), height (continuous variable), and PSA concentration (continuous variable after logarithmic transformation). We tested the coefficient of the cross-product term by the Wald test.
The associations between other clinical, demographic, and anthropometric factors and outcomes have previously been studied in the SEARCH Database.16,20,21 The distribution of all clinical and pathologic variables was similar among the centers contributing to the SEARCH Database. Therefore, data from all centers were combined for analyses. All statistical analyses were performed using STATA 8.0 (Stata Corp, College Station, TX).
RESULTS
Mean and median RP specimen weights were 44 g (standard deviation, 23 g) and 39 g, respectively (range, 5 to 255 g). Mean patient age was 62.5 years (standard deviation, 6.8 years; Table 1). Smaller prostate weight was associated with younger age (P < .001) and lower preoperative PSA (P < .001).
Pathologic Gleason Sum and Other Pathologic Characteristics
The odds of high-grade disease, positive surgical margins, and extracapsular extension increased with decreasing prostate weight after adjusting for age at RP; the associations were enhanced after multivariable adjustment (Table 2). After including in the model preoperative PSA, the expression of which is under androgenic regulation and is strongly correlated with prostate volume, smaller prostate weight was associated with statistically significantly increased odds of high-grade disease in the RP specimen (comparing < 20 v 100 g: OR = 7.61; per 1-unit increase in log-transformed prostate weight, OR = 0.46; P < .001; Table 2). Even after adjusting for the higher prevalence of high-grade disease, smaller prostate weight was associated with statistically significantly increased odds of positive surgical margins (comparing < 20 v 100 g: OR = 3.09; per 1-unit increase in log-transformed prostate weight, OR = 0.46; P < .001) and extracapsular extension (comparing < 20 v 100 g: OR = 1.73; per 1-unit increase in log-transformed prostate weight, OR = 0.60; P = .004; Table 2). There was no statistically significant association between prostate weight and seminal vesicle invasion (P = .32). Similar trends were noted at each of the centers contributing data to the SEARCH Database. Both biopsy Gleason sum (P = .10) and clinical stage (P = .13) were inversely associated with prostate weight, although the associations were not statistically significant.
Biochemical Progression
Mean and median follow-up times among men without progression were 46 months (standard deviation, 40 months) and 34 months, respectively. During this time, 404 patients (27%) experienced disease progression. Smaller prostate weight was statistically significantly associated with a greater risk of biochemical progression (P = .007, Fig 1). Adjusting for age at RP, smaller prostate weight was associated with a statistically significantly increased risk of biochemical progression (comparing < 20 v 100 g: RR = 3.94; per 1-unit increase in log-transformed prostate weight, RR = 0.60; P = .001), the association for which was even stronger after adjustment for patient, clinical, and pathologic characteristics (comparing < 20 v 100 g: RR = 5.89; per 1-unit increase in log-transformed prostate weight, RR = 0.56; P < .001; Table 3). After further adjusting for preoperative PSA, the RR of biochemical progression comparing less than 20 with 100 g increased to 8.43 (per 1-unit increase in log-transformed prostate weight, RR = 0.48; P < .001). Similar trends were noted at each of the centers contributing data to the SEARCH Database.
Influence of Extreme Prostate Weights on the Association of Prostate Weight With Pathologic Characteristics and Biochemical Progression
The association of prostate weight with adverse pathologic characteristics and biochemical progression seemed to be strongest when contrasting men with the largest ( 100 g) and the smallest prostates (< 20 g). To explore whether prostate weight was associated with adverse pathologic characteristics and biochemical progression for the majority of men who had prostate weights of 20 g or more and less than 100 g, we reanalyzed our data excluding men with prostate weights of less than 20 g or 100 g or more. When these men were excluded, on multivariable analysis, smaller prostate weight remained statistically significantly inversely associated with high-grade cancer (P < .001), positive surgical margins (P < .001), extraprostatic disease (P = .01), and biochemical progression (P = .01).
Influence of PSA on the Association of Prostate Weight With Pathologic Characteristics and Biochemical Progression
To explore whether these associations varied by preoperative PSA, we stratified by PSA concentration. There was no statistically significant interaction between prostate weight and PSA (all P > .1); within each PSA stratum (< 10, 10 to < 20, and 20 ng/mL), smaller prostate weight was associated with higher odds of high-grade disease, positive surgical margins, and extracapsular extension. There was no statistically significant interaction between preoperative PSA and prostate weight for predicting biochemical progression (P = .72); smaller prostate weight was statistically significantly associated with progression among men with low (n = 950, RR = 1.32 for 20-g decrease in prostate weight, P = .03), intermediate (n = 250, RR = 1.38, P = .03), and high PSA concentrations (n = 92, RR = 2.38, P = .003).
Larger prostate weight was associated with higher PSA concentration, which would potentially lead to earlier diagnosis because of elevated PSA as a biopsy indication. To exclude this potential for detection bias, we excluded all men who underwent biopsy solely because of an elevated PSA (clinical stage T1c) and reanalyzed the data examining only men who were biopsied because of an abnormal digital rectal examination. When only men with clinical stage T2/T3 tumors were examined, smaller prostate size remained statistically significantly associated with high-grade disease (comparing < 20 to 100 g: OR = 8.48; 95% CI, 1.74 to 41.42; P = .01), positive surgical margins (OR = 7.08; 95% CI, 1.23 to 40.80, P = .003), extracapsular extension (OR = 1.50; 95% CI, 0.24 to 9.22; P = .04; Table 4), and biochemical progression (RR = 11.75; 95% CI, 2.76 to 49.96; P = .001; Table 5).
Evaluation of Potential Interactions
Within each stratum of BMI, height, or race, the association of smaller prostate weights with high-grade disease, positive surgical margins, extracapsular extension, and biochemical progression was similar. Overall, there were no statistically significant interactions (all P .09) between prostate weight and BMI, height, age, or race for predicting adverse pathology or biochemical progression with one exception; there was a significant interaction (P = .04) between prostate size and age in that larger prostate weight was more strongly related to lower odds of seminal vesicle invasion among older men than among younger men.
Evaluation of PSA Density
PSA density (PSAD; PSA concentration divided by prostate weight) has been suggested to better predict biochemical progression after RP than PSA.22 Therefore, we examined whether prostate weight remained statistically significantly associated with biochemical progression after adjusting for PSAD. When PSAD (0.3 to 0.7 and > 0.7 v < 0.3 ng/mL/g)22 was substituted for PSA in the multivariable model, smaller prostate weight remained statistically significantly associated with biochemical progression (comparing < 20 v 100 g: RR = 3.22; 95% CI, 1.16 to 8.90; P = .01).
Transrectal Ultrasound–Measured Prostate Volume As a Preoperative Prognostic Factor
Data were available for 753 men on preoperative transrectal ultrasound (TRUS) prostate volume. There was a strong correlation between TRUS volume and RP specimen weight (Spearman, r = 0.71; P < .001). We evaluated associations between 20-g intervals of TRUS-measured prostate volume and pathologic findings and biochemical progression after RP. Because we sought to determine whether TRUS volume could be used preoperatively for risk assessment, postoperative variables (pathologic Gleason sum, positive surgical margin, extracapsular extension, seminal vesicle invasion, or lymph node positivity) were not included in the analyses. On multivariable analysis (adjusting for biopsy Gleason sum, clinical stage, age, year of surgery, preoperative PSA, race, BMI, and height), smaller preoperative TRUS volume was associated with high-grade disease in the RP specimen (P < .001), positive surgical margins (P < .001), extracapsular extension (P = .005), seminal vesicle invasion (P = .07), and biochemical progression (comparing < 20 to 100 cc: RR = 5.99; 95% CI, 0.77 to 46.68; P = .06).
DISCUSSION
In an equal-access population, we found that men with smaller prostates had more high-grade cancers and more advanced disease at the time of RP and were at significantly greater risk of biochemical progression. Similar findings were observed whether we examined pathologic RP specimen weight or preoperative TRUS-measured prostate volume. These findings suggest that prostate size may be an important prognostic variable in prostate cancer.
On the basis of several observations, we hypothesized that small prostate size, after adjusting for age, may be a surrogate of low in vivo androgenicity. First, prostate growth and differentiation are highly dependent on sex steroid hormones, particularly DHT.1 In patients with 5-alpha reductase type II deficiency and, subsequently, low to absent DHT levels, the prostate develops rudimentarily.23 In animal models, only androgen withdrawal therapy is capable of pharmacologically reducing prostate size.24 In humans, 5-alpha reductase type II inhibition reduces prostate size by approximately 25%, whereas medical castration results in a more than 50% reduction.3,4 Prostate size in animal models is most closely correlated with serum androgen concentrations and advancing age.25 In human cross-sectional studies, the only serum hormone concentrations that have been associated with prostate volume are free testosterone and estrogens.26 Moreover, shorter CAG repeats within the androgen receptor gene, which result in a more active androgen receptor, have been correlated with larger prostates.27 However, other factors, namely estrogen, insulin, and insulin-like growth factor 1, all of which have been associated with prostate cancer development and progression in some epidemiologic studies, also influence prostate size.26,28,29 Therefore, for a given age, a small prostate is likely associated with both lower androgenicity and lower overall intraprostatic growth factor concentrations. Whether this simply results in a hostile environment in which only aggressive cancers can grow or the lower growth factor/androgenic concentrations actually induce more aggressive cancers is unclear.
Although conflicting data exist, several studies found that men with genetic polymorphisms favoring increased androgenicity were at higher risk for developing prostate cancer.30,31 Among men with metastatic disease, lower pretreatment total testosterone concentrations have consistently been associated with worse response to treatment and worse survival.5-8 Among men with clinically localized disease undergoing radiation therapy, Zagars et al10 found that low testosterone concentrations were associated with decreased risk of developing metastatic disease. For patients undergoing RP, some studies found no association between serum testosterone concentration and pathologic stage,15 whereas others found that low testosterone concentrations were associated with a higher grade, advanced stage, and higher progression rates.11-14 However, serum testosterone concentrations at the time of diagnosis may not reflect either intraprostatic androgenicity or the cumulative life-long androgen exposure.
Smaller prostate weight was associated with higher grade disease and more advanced stage, and even after controlling for these findings, smaller prostate weight independently predicted biochemical progression. Similar findings were observed whether prostate size was measured by RP specimen weight or preoperative TRUS-measured prostate volume. Moreover, prostate weight remained a significant predictor of biochemical progression after adjustment for PSAD. The higher prevalence of high-grade and advanced disease among men with smaller prostates would be expected to result in worse outcomes after other forms of therapy, although further studies are needed to confirm this. Prostate size is often used to help guide decisions regarding therapy (discouragement of men with larger prostates from brachytherapy32 and encouragement of men with large prostates and associated urinary symptoms to undergo surgery). These practice patterns may theoretically impact treatment outcomes and should be taken into account by adjusting for prostate size when comparing results across treatments. Additionally, studies found that Veterans Affairs patients had smaller prostates than men at tertiary-care referral centers.33,34 These prostate size differences between centers may, in part, explain disparate results between centers, although this needs to be specifically addressed in future studies. Consequently, inclusion of prostate size into pre- and post-treatment nomograms would likely result in both improved risk assessment and improved generalizability of these nomograms to men treated in different practice settings.
It is possible that men with larger prostates had their tumors detected earlier because of PSA-driven biopsies resulting from PSA elevation from an enlarged gland. This lead-time bias would be expected to result in better outcomes. Several studies found that prostate volumes more than 75 cc were associated with more favorable outcomes after RP35-37; all of the studies concluded that lead-time bias accounted for most or all of the results. Several findings suggest that lead-time bias may not explain all of the findings in the current study. First, after excluding men with PSA-detected tumors (clinical stage T1c) and examining only men with abnormal digital rectal examinations, the associations between smaller prostate size and higher grade advanced disease and biochemical progression remained statistically significant. Second, men with larger prostates were older. If one proposes that men with larger prostates had their cancers detected earlier, one would expect these patients to be younger and not older. Third, men with smaller prostates had more high-grade cancers. Whether high-grade cancers result from dedifferentiation of low-grade cancers over time or develop de novo is unknown. However, recent data suggest that, on rebiopsy of men in watchful waiting programs, tumor grade remains stable over several years.38,39 Although this does not exclude grade progression over the long term, it seems unlikely that grade progression could explain the 7.5-fold increased odds of high-grade disease among men with prostates less than 20 g v 100 g. Finally, even after controlling for and stratifying by preoperative PSA and eliminating men with PSA-detected cancers, men with smaller prostates had higher grade and more advanced disease. It is important to note that the validity of prostate size as a prognostic marker is independent of the reason for the worse outcomes among men with smaller prostates. Whether the worse outcomes were a result of detection bias, more aggressive tumors developing in a low growth factor/androgenic environment, or some alternate explanation, the end result remains the same; men with smaller prostates had more high-grade advanced disease and higher progression rates.
It has been argued that a marked reduction in androgenicity may interfere with prostate cancer grading, resulting in artifactual upgrading that does not reflect tumor biology.40 In the current study, it seems unlikely that the higher grade tumors among men with smaller prostates were a result of grading artifacts because these high-grade cancers were associated with other signs of aggressive behaviors, such as advanced pathologic stage and higher progression rates.
Serum and tissue androgen concentrations, androgen receptor concentrations and activity levels, and the androgen receptor gene CAG repeat length were unavailable in the SEARCH Database. Thus, we are unable to comment on the true in vivo androgenicity of our patients, and thus, it remains a hypothesis that smaller prostate size is a surrogate of decreased in vivo androgenicity. Seminal vesicles and tips of the vasa were included in the RP specimen weight. Given that we examined prostate weight as a categoric variable, even if seminal vesicle weight had not been included, most patients would be in the same or, at most, one lower weight category. Thus, the progressive overall trends observed for decreasing weight and worse outcomes cannot be explained by inclusion of seminal vesicle weight in the total RP specimen weight. Moreover, seminal vesicle weight is also under androgenic control and is often used as an in vivo bioassay for androgenicity.41,42 Thus, the total RP specimen weight, which includes prostate and seminal vesicle weight, may be a better surrogate of in vivo androgenicity than prostate weight alone. In the current study, all patients had prostate cancer and underwent surgery for their disease. Therefore, we are unable to comment on any possible relationship between androgenicity and the risk of developing prostate cancer. In addition, mean follow-up time was relatively short. Finally, we examined biochemical progression as our end point. Several studies found that, among men treated with RP, a shorter time to biochemical progression is associated with increased risk for developing metastatic disease.43,44 Although recent studies suggested that postoperative PSA doubling time may be a better surrogate than biochemical progression for prostate cancer–specific mortality,45,46 PSA doubling times were not available in the SEARCH Database.
In conclusion, we found that men with smaller prostate sizes had higher grade and more advanced disease at the time of RP and a greater risk of biochemical progression. Because similar results were found whether prostate size was measured as RP specimen weight or preoperative TRUS-measured prostate volume, the current data support the evaluation of prostate size as a pre- and postoperative prognostic marker in prostate cancer.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by the Department of Veterans Affairs, National Institutes of Health Grant No. R01CA100938 (W.J.A.), National Institutes of Health Specialized Programs of Research Excellence Grant No. P50 CA92131-01A1 (W.J.A.), the Georgia Cancer Coalition (M.K.T.), Center for Prostate Disease Research grant from the United States Army Medical Research and Materiel Command (C.L.A.), the Department of Defense, Prostate Cancer Research Program Grant No. PC030666 (S.J.F.), and the American Foundation for Urological Disease/American Urological Association Education and Research Scholarship Award (S.J.F.).
The views and opinions of and endorsements by the authors do not reflect those of the US Army or the Department of Defense.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Marker PC, Donjacour AA, Dahiya R, et al: Hormonal, cellular, and molecular control of prostatic development. Dev Biol 253:165-174, 2003
Isaacs JT, Coffey DS: Changes in dihydrotestosterone metabolism associated with the development of canine benign prostatic hyperplasia. Endocrinology 108:445-453, 1981
Thompson IM, Goodman PJ, Tangen CM, et al: The influence of finasteride on the development of prostate cancer. N Engl J Med 349:215-224, 2003
Noldus J, Ferrari M, Prestigiacomo A, et al: Effect of flutamide and flutamide plus castration on prostate size in patients with previously untreated prostate cancer. Urology 47:713-718, 1996
Chodak GW, Vogelzang NJ, Caplan RJ, et al: Independent prognostic factors in patients with metastatic (stage D2) prostate cancer: The Zoladex Study Group. JAMA 265:618-621, 1991
Ribeiro M, Ruff P, Falkson G: Low serum testosterone and a younger age predict for a poor outcome in metastatic prostate cancer. Am J Clin Oncol 20:605-608, 1997
Chen SS, Chen KK, Lin AT, et al: The correlation between pretreatment serum hormone levels and treatment outcome for patients with prostatic cancer and bony metastasis. BJU Int 89:710-713, 2002
Iversen P, Rasmussen F, Christensen IJ: Serum testosterone as a prognostic factor in patients with advanced prostatic carcinoma. Scand J Urol Nephrol Suppl 157:41-47, 1994
Oefelein MG, Agarwal PK, Resnick MI: Survival of patients with hormone refractory prostate cancer in the prostate specific antigen era. J Urol 171:1525-1528, 2004
Zagars GK, Pollack A, von Eschenbach AC: Serum testosterone: A significant determinant of metastatic relapse for irradiated localized prostate cancer. Urology 49:327-334, 1997
D'Amico AV, Chen MH, Malkowicz SB, et al: Lower prostate specific antigen outcome than expected following radical prostatectomy in patients with high grade prostate and a prostatic specific antigen level of 4 ng/ml or less. J Urol 167:2025-2030, 2002
Massengill JC, Sun L, Moul JW, et al: Pretreatment total testosterone level predicts pathological stage in patients with localized prostate cancer treated with radical prostatectomy. J Urol 169:1670-1675, 2003
Hoffman MA, DeWolf WC, Morgentaler A: Is low serum free testosterone a marker for high grade prostate cancer? J Urol 163:824-827, 2000
Schatzl G, Madersbacher S, Thurridl T, et al: High-grade prostate cancer is associated with low serum testosterone levels. Prostate 47:52-58, 2001
Monda JM, Myers RP, Bostwick DG, et al: The correlation between serum prostate-specific antigen and prostate cancer is not influenced by the serum testosterone concentration. Urology 46:62-64, 1995
Freedland SJ, Amling CL, Dorey F, et al: Race as an outcome predictor after radical prostatectomy: Results from the Shared Equal Access Regional Cancer Hospital (SEARCH) database. Urology 60:670-674, 2002
Freedland SJ, Sutter ME, Dorey F, et al: Defining the ideal cutpoint for determining PSA recurrence after radical prostatectomy. Urology 61:365-369, 2003
Freedland SJ, Presti JC Jr, Terris MK, et al: Improved clinical staging system combining biopsy laterality and TNM stage for men with T1c and T2 prostate cancer: Results from the SEARCH database. J Urol 169:2129-2135, 2003
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
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 22:446-453, 2004
Freedland SJ, Presti JC Jr, Kane CJ, et al: Do younger men have better biochemical outcomes after radical prostatectomy? Urology 63:518-522, 2004
Freedland SJ, Wieder JA, Jack GS, et al: Improved risk stratification for biochemical recurrence after radical prostatectomy using a novel risk group system based on prostate specific antigen density and biopsy Gleason score. J Urol 168:110-115, 2002
Walsh PC, Madden JD, Harrod MJ, et al: Familial incomplete male pseudohermaphroditism, type 2: Decreased dihydrotestosterone formation in pseudovaginal perineoscrotal hypospadias. N Engl J Med 291:944-949, 1974
Coffey DS, Walsh PC: Clinical and experimental studies of benign prostatic hyperplasia. Urol Clin North Am 17:461-475, 1990
Berry SJ, Coffey DS, Ewing LL: Effects of aging on prostate growth in beagles. Am J Physiol 250:R1039-R1046, 1986
Partin AW, Oesterling JE, Epstein JI, et al: Influence of age and endocrine factors on the volume of benign prostatic hyperplasia. J Urol 145:405-409, 1991
Roberts RO, Bergstralh EJ, Cunningham JM, et al: Androgen receptor gene polymorphisms and increased risk of urologic measures of benign prostatic hyperplasia. Am J Epidemiol 159:269-276, 2004
Roberts RO, Jacobson DJ, Girman CJ, et al: Insulin-like growth factor I, insulin-like growth factor binding protein 3, and urologic measures of benign prostatic hyperplasia. Am J Epidemiol 157:784-791, 2003
Hammarsten J, Hogstedt B: Hyperinsulinaemia as a risk factor for developing benign prostatic hyperplasia. Eur Urol 39:151-158, 2001
Makridakis NM, Reichardt JK: Molecular epidemiology of androgen-metabolic loci in prostate cancer: Predisposition and progression. J Urol 171:S25-S28, 2004
Giovannucci E, Stampfer MJ, Krithivas K, et al: The CAG repeat within the androgen receptor gene and its relationship to prostate cancer. Proc Natl Acad Sci U S A 94:3320-3323, 1997
Merrick GS, Wallner KE, Butler WM: Permanent interstitial brachytherapy for the management of carcinoma of the prostate gland. J Urol 169:1643-1652, 2003
Nixon RG, Meyer GE, Brawer MK: Differences in prostate size between patients from University and Veterans Affairs Medical Center populations. Prostate 38:144-150, 1999
Terris MK, Prestigiacomo AF, Stamey TA: Comparison of prostate size in university and Veterans Affairs Health Care System patients with negative prostate biopsies. Urology 51:412-414, 1998
D'Amico AV, Whittington R, Malkowicz SB, et al: A prostate gland volume of more than 75 cm3 predicts for a favorable outcome after radical prostatectomy for localized prostate cancer. Urology 52:631-636, 1998
Foley CL, Bott SR, Thomas K, et al: A large prostate at radical retropubic prostatectomy does not adversely affect cancer control, continence or potency rates. BJU Int 92:370-374, 2003
Stamey TA, Yemoto CM, McNeal JE, et al: Prostate cancer is highly predictable: A prognostic equation based on all morphological variables in radical prostatectomy specimens. J Urol 163:1155-1160, 2000
Carter CA, Donahue T, Sun L, et al: Temporarily deferred therapy (watchful waiting) for men younger than 70 years and with low-risk localized prostate cancer in the prostate-specific antigen era. J Clin Oncol 21:4001-4008, 2003
Epstein JI, Walsh PC, Carter HB: Dedifferentiation of prostate cancer grade with time in men followed expectantly for stage T1c disease. J Urol 166:1688-1691, 2001
Thompson IM, Klein EA, Lippman SM, et al: Prevention of prostate cancer with finasteride: US/European perspective. Eur Urol 44:650-655, 2003
Skarda J: Bioassay of steroid hormone agonist and antagonist activities of anti-androgens on mammary gland, seminal vesicles and spleen of male mice. J Vet Med A Physiol Pathol Clin Med 50:204-212, 2003
Kousteni S, Chen JR, Bellido T, et al: Reversal of bone loss in mice by nongenotropic signaling of sex steroids. Science 298:843-846, 2002
Pound CR, Partin AW, Eisenberger MA, et al: Natural history of progression after PSA elevation following radical prostatectomy. JAMA 281:1591-1597, 1999
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
D'Amico AV, Moul JW, Carroll PR, et al: Surrogate end point for prostate cancer-specific mortality after radical prostatectomy or radiation therapy. J Natl Cancer Inst 95:1376-1383, 2003
Ward JF, Blute ML, Slezak J, et al: The long-term clinical impact of biochemical recurrence of prostate cancer 5 or more years after radical prostatectomy. J Urol 170:1872-1876, 2003(Stephen J. Freedland, Wil)