Men Presenting for Radical Prostatectomy on Preoperative Statin Therapy Have Reduced Serum Prostate Specific Antigen

Men Presenting for Radical Prostatectomy on Preoperative Statin Therapy Have Reduced Serum Prostate Specific Antigen L. Spencer Krane,* Sanjeev A. Kaul, Hans J. Stricker, James O. Peabody, Mani Menon and Piyush K. Agarwal From the Department of Urology, Wake Forest University, Winston-Salem, North Carolina (LSK), and Vattikuti Urology Institute, Henry Ford Health Systems, Detroit, Michigan

Abbreviations and Acronyms BMI ⫽ body mass index CoA ⫽ coenzyme A hMG ⫽ 3-hydroxy-3-methylglutaryl LDL ⫽ low-density lipoprotein PSA ⫽ prostate specific antigen RP ⫽ radical prostatectomy Submitted for publication May 14, 2009. * Correspondence: Vattikuti Urology Institute, Henry Ford Health Systems, 2799 W. Grand Blvd., Detroit, Michigan 48202 (e-mail: spencerkrane@ gmail.com).

For another article on a related topic see page 345.

Purpose: Studies have suggested that statin (3-hydroxy-3-methyl-glutaryl-coenzyme A reductase inhibitors) medication use may decrease prostate specific antigen in healthy men. We determined the effect of preoperative statin use on total preoperative prostate specific antigen and the risk of biochemical recurrence in patients with prostate cancer presenting for radical prostatectomy. Materials and Methods: A retrospective review of 3,828 patients undergoing radical prostatectomy from January 2001 to July 2008 at our institution identified 1,031 on statin medications. We compared these 1,031 patients to the remaining 2,797 not on statins preoperatively. We evaluated differences in prostate specific antigen overall, and when patients were stratified by age specific groups, body mass index and Gleason grades on final pathology. We also investigated differences in biochemical recurrence rates. Results: Overall median serum prostate specific antigen was lower in patients on preoperative statins (5.0 vs 5.2 ng/ml, p ⫽ 0.002). Median prostate specific antigen was lower in men on statins with Gleason grades 7 or 8/9 disease (p ⬍0.05). Using a multivariate logistic regression model statin therapy was associated with a 4.7% decrease in prostate specific antigen (p ⬍0.001). Statin therapy was not associated with an overall decreased risk of biochemical recurrence (p ⫽ 0.73) at a mean followup of 26 months. Conclusions: In this cohort of men presenting for radical prostatectomy serum prostate specific antigen is significantly lower in patients with prostate cancer on preoperative statins compared to those not taking these medications. Prospective studies are required to evaluate if this decrease in prostate specific antigen leads to later detection of prostate cancer or variations in oncological outcomes. Key Words: prostatic neoplasms, hydroxymethylglutaryl-CoA reductase inhibitors, prostatectomy

118

www.jurology.com

STATIN medications are among the most widely prescribed medications in America today.1 They have demonstrated effects in reducing cardiovascular morbidity in patients with low high-density lipoprotein or increased LDL.2 They are also indicated for primary treatment of hyperlipidemia in patients with diabetes3 and following

cardiovascular events or acute myocardial infarction.4 They function by inhibiting the conversion of hMG-CoA to mevalonate, a precursor of cholesterol.5 The indications of these medications continue to increase and epidemiological trends indicate that current prescribing patterns will continue.1 As the medications become ge-

0022-5347/10/1831-0118/0 THE JOURNAL OF UROLOGY® Copyright © 2010 by AMERICAN UROLOGICAL ASSOCIATION

Vol. 183, 118-125, January 2010 Printed in U.S.A. DOI:10.1016/j.juro.2009.08.151

STATINS DECREASE PROSTATE SPECIFIC ANTIGEN IN MEN WITH PROSTATE CANCER

neric during the next decade and prices become more affordable, the overall use and compliance will certainly increase. These medications have demonstrated a complex interaction with the prostate, PSA and the development of prostate cancer. In 2 independent studies PSA decreased in patients prescribed statin medications.6,7 These studies involved screening populations of men and prostate cancer was not evaluated with biopsies. In vitro, statins have demonstrated antitumor effects by inhibiting endothelial cell migration, inducing apoptosis of prostate cancer cell lines and inhibiting angiogenesis.8 Patients with prostate cancer presenting for radiation therapy on statin therapy may have more favorable characteristics when presenting for radiation therapy, although this has not been confirmed by multiple centers.9 We present the preoperative characteristics of men on statin therapy seeking surgical therapy for clinically localized prostate cancer between 2001 and 2008, and compare pathological features with men not on statin medication. Additionally, we evaluate the role of statin medications in the risk of biochemical recurrence following radical prostatectomy.

MATERIALS AND METHODS We identified 3,828 men with biopsy proven prostate cancer who underwent computer assisted (robotic) RP between January 2001 and August 2008 at the Vattikuti Urology Institute. Preoperative characteristics including age, BMI (kg/m2), PSA (ng/ml), current medication use (specific type but not dose, duration or indication), pathological Gleason sum, percentage tumor involvement, pathological stage and prostate size were entered prospectively into an institutional review board approved database. All hMG-CoA reductase inhibitors (including combination therapies such as ezetimibe/simvastatin) were included in the statin group. We did not have institutional review board approval to collect indication for statin therapy, dose of medication or high-density lipoprotein/LDL levels. Race data were not complete for all patients and, therefore, were not included in the analysis. Patients with a reported diabetes diagnosis or preoperative use of hyperglycemic medication were included in further diabetes analysis due to the association of PSA with diabetic status.10 Patients were considered overweight if they had a BMI between 25 and 29.9 kg/m2, and obese if BMI was greater than 30 kg/m2. The 1997 UICC-American Joint Committee on Cancer TNM system was used for staging and the Gleason system was used for grading. Fellowship trained genitourinary pathologists centrally reviewed all pathology data. Biochemical recurrence was defined as the American Urological Association consensus definition of a single PSA of 0.2 ng/ml or greater with another increasing value.11 Student’s t test, chi-square, Fisher’s exact test and nonparametric tests were used as appropriate. The Kaplan-Meier method was used to evaluate the risk of biochemical recurrence. Multivariable linear regression

119

analysis was performed to evaluate the effect of statin therapy after assuring the data satisfied assumptions of linear regression. Variables in the multivariable model included age (continuous), BMI (less than 25, 25.0 to 29.9, greater than 30), finasteride therapy, timing of surgery (2001 to 2005, 2006 to 2008), Gleason sum, history of diabetes, prostate weight (log transformed) and pathological stage. Multivariate analysis was performed using a Cox proportional hazards model predicting the time to recurrence. Statistical analysis was performed using JMP® 7. All comparisons were made using 2-sided tests with p ⬍0.05 considered statistically significant.

RESULTS We identified 1,031 patients who reported preoperative statin use at RP. These men were slightly older and had a higher BMI than those not on statins preoperatively (table 1). The overall median PSA was significantly lower in the patients on statins (5.0 vs 5.2 ng/ml, p ⫽ 0.0015). Overall men on statin therapy presented with a more aggressive distribution of Gleason sums vs those not on statins. The patients on statin therapy had a higher proportion of Gleason 7 disease detected on final pathology (58% vs 54%, p ⫽ 0.02 for trend). The men on statin therapy were more likely to be overweight, have a diagnosis of diabetes and have surgery later in the cohort (p ⬍0.05). Statin use was lowest in patients younger than 50 years (14%) and was highest in men 60 to 69 years old (33%) (table 2). Statin prevalence in BMI groups demonstrated increasing use with increasing BMI as there was 20% use in men with a BMI less than 25 kg/m2, 28% in men with a BMI between 25 and 29.9 and 31% in men with a BMI greater than 30. At all ages and BMI classes PSA was lower in patients on statin therapy, although this finding was not statistically significant for all groups. Stratified by pathological Gleason sums, median PSA was statistically significantly lower in men on statin therapy with Gleason 7 or Gleason 8-10 disease (p ⬍0.05) (table 3). When evaluating PSA density in patients with similar Gleason sums this difference was statistically significant in all cohorts. In the multivariable linear regression model, controlling for confounders such as age, BMI, finasteride therapy and history of diabetes, we find that statin therapy was associated with an approximately 4% overall decrease in PSA (table 4). Patients having surgery later in the cohort also had a slightly decreased PSA. Statins did not change the rate of biochemical recurrence. Overall biochemical recurrence for the entire cohort was not significantly different between the statin and nonstatin groups (see figure). In a subgroup analysis based on Gleason sum we did not find that statin therapy decreased the rate of recurrence. Stratification of BMI categories did not induce statistical significance of statin therapy on bio-

120

STATINS DECREASE PROSTATE SPECIFIC ANTIGEN IN MEN WITH PROSTATE CANCER

Table 1. Preoperative and pathological patient characteristics Statin No. men (%) Mean age (SD) Median BMI (IQR) No. BMI (%): Less than 25 25.0–29.9 Greater than 30 Median PSA (IQR) Median prostate wt (IQR) Median PSA density (IQR) No. preop finasteride therapy (%) No. yr of prostatectomy (%): 2001–2005 2006–2008 No. history of diabetes mellitus (%): No diabetes mellitus Diabetes mellitus No. clinical stage (%): pT1c pT2/3 No. pathological Gleason sum (%): 6 7 8/9 Median % Ca involvement of prostate (IQR) No. pathological stage (%): pT2 pT3a pT3b/T4

1,031 61.4 28

No Statin

(27) (6.6) (26–30)

2,807 59.4 27

p Value

(73) (7.5) (25–30)

154 (15) 544 (53) 323 (32) 5.0 (4.1–6.5) 46 (37–57) 0.11 (0.08–0.15) 13 (1.3)

630 (23) 1,395 (51) 718 (26) 5.2 (4.1–7.2) 44 (36–56) 0.12 (0.08–0.17) 45 (1.6)

483 548

(47) (53)

1,547 1,260

(55) (45)

903 128

(88) (12)

2,655 152

(95) (5)

750 281

(73) (27)

2,044 763

(73) (27)

323 591 113 10

(31) (58) (11) (5–20)

1,011 1,510 266 15

(36) (54) (10) (5–20)

712 251 66

(70) (24) (6)

2,019 615 166

(72) (22) (6)

Not available 0.001* 0.001†

0.0015† 0.0008† 0.0001† 0.43‡ 0.0001‡

0.0001‡

0.96‡

0.02§

0.13† 0.20§

* Student’s t test. † Wilcoxon rank sum test. ‡ Chi-square test. § Fisher’s exact test.

chemical recurrence. On multivariate analysis including pathological stage, Gleason grade, PSA, margin status and statin use only statin use was not significant in predicting the risk of biochemical recurrence (table 5). Margin status was similar between the 2 groups (p ⫽ 0.43). Mean followup in these patients was 26 months.

DISCUSSION The interactions between statins and prostate cancer is multifaceted, and as of yet has not been fully Table 2. PSA and statin use

Pt age: 40–49 50–59 60–69 70–79 BMI: Less than 25 25–29.9 30⫹

No. Pts on Statins (%)

Median PSA With Statins (IQR)

Median PSA Without Statins (IQR)

p Value*

43 (14) 341 (22) 534 (33) 113 (30)

4.5 (3.5–6.3) 4.6 (3.9–5.9) 5.1 (4.2–7.0) 5.2 (4.3–7.8)

4.7 (3.5–6.4) 5.1 (4.1–6.9) 5.4 (4.3–7.4) 5.4 (4.5–6.9)

0.90 0.0001 0.056 0.297

154 (20) 544 (28) 323 (31)

4.7 (3.6–6.0) 5.0 (4.1–6.7) 5.1 (4.2–6.5)

5.0 (4–6.7) 5.3 (4.2–7.4) 5.2 (4.2–7.2)

0.09 0.005 0.16

* Kruskal-Wallis rank sum test.

elucidated. Despite early excitement that these medications may prevent prostate cancer, metaanalyses have demonstrated that the overall risk of prostate cancer is not reduced with statin use.12 Three meta-analyses, evaluating as many as 930,000 patients, have shown that the use of statin medication does not decrease overall risk.2,13,14 This lack of association was also confirmed in a study of northern Californian men in an integrated health care program.15 However, in a Health Professionals Follow Up Study Platz et al reported that while the overall risk of prostate cancer is not decreased, there is a substantial decrease in advanced or metastatic prostate cancer in statin users.16 This phenomenon was dose and duration dependent. Another study by Jacobs et al confirmed the decreased risk of advanced or metastatic prostate cancer in patients using cholesterol lowering agents.17 Murtola et al found this association between statin therapy and decreased risk of advanced prostate cancer only with the use of atorvastatin, lovastatin and simvastatin.18 Flick et al noted a decreased risk of prostate cancer only in men routinely using nonsteroidal anti-inflammatory drugs.19 In evaluating patients from the Surveillance, Epidemiology, and End Results database Boudreau et al saw a decreased risk

STATINS DECREASE PROSTATE SPECIFIC ANTIGEN IN MEN WITH PROSTATE CANCER

121

Table 3. PSA and Gleason sum

Statin median PSA (IQR) No statin median PSA (IQR) p Value* Statin median PSA density (IQR)† No statin median PSA density (IQR)† p Value*

Gleason Sum 6

Gleason Sum 7

4.7 4.7

5.0 5.4

(3.6–5.8) (3.7–6.1) 0.70 0.09 (0.07–0.13) 0.10 (0.07–0.14) 0.045

(4.1–6.6) (4.3–7.3) 0.0001 0.11 (0.08–0.16) 0.13 (0.09–0.18) 0.0001

Gleason Sum 8-10 6.1 6.6

(4.4–8.2) (4.9–11.0) 0.045 0.13 (0.08–0.20) 0.15 (0.12–0.24) 0.004

* Kruskal-Wallis rank sum test. † PSA density is PSA/pathological prostate weight.

of prostate cancer in men on statin therapy. However, this association was not statistically significant nor was a dose dependent relationship observed.20 Multiple theories for the role of statins in preventing metastatic or advanced prostate cancer have been proposed. In vitro the presence of statins has been known to cause apoptosis of prostate cancer cell lines directly.8 Isoprenoid production is also inhibited by statin therapy and these proteins are known to induce progression through the cell cycle. Inflammation has been shown to be a contributing factor in the development of prostate cancer and statin therapy is associated with a decrease in overall inflammatory state. Increased intraprostatic cholesterol deposits are seen with aggressive prostate cancer and, therefore, decreasing production of cholesterol may provide a protective effect.21 While statins are not believed to affect systemic androgen levels, their effect on intraprostatic testosterone or dihydrotestosterone levels has not been evaluated.22 Our findings have been hinted at in previous studies but not fully investigated. In an evaluation of patients presenting for brachytherapy for clinically localized prostate cancer Moyad et al noted that patients on statins had a statistically lower mean PSA of 5.7 vs 6.9 ng/ml in nonusers.9 Soto et al Table 4. Multivariate linear regression evaluating change in PSA

2

BMI for each increased 1 kg/m Pathological Gleason sum: 6 7 8–10 Surgery yr: 2001–2005 2006–2008 Statin therapy: No Yes

% Change in PSA (95% CI)

p Value*

⫺0.7 (⫺0.4–⫺1.2)

0.001

3.2 18

Ref (0.2–6.2) (14–21)

0.03 0.0001

Ref ⫺4.0 (⫺2.0–⫺6.0)

0.001

Ref ⫺4.7 (⫺2.6–⫺7.0)

0.001

Model adjusted for age, BMI, prostate weight (log transformed), finasteride therapy, diabetes diagnosis and pathological stage. * Entered in the model as natural log transformation of preoperative PSA and back transformed for ease of interpretation.

also found that patients on preoperative statin therapy presenting for radiation therapy had a statistically lower PSA than nonusers (3.1 vs 4.6 ng/ml, p ⬍0.0001).23 Our findings confirm these previous findings, and expand on our current understanding of the interaction between serum PSA and statin therapy. Even when including potential confounders such as year of surgery, finasteride therapy and history of diabetes the interaction between PSA and statins still exists. Other authors have demonstrated that PSA decreases when statin therapy is initiated. This was independently reported in 2 studies.6,7 In 1 study 15 airplane pilots were started on statin therapy and compared to 85 pilots not on statins.6 The men on statins had a 41% PSA decrease 5 years after the initiation of therapy, whereas those not on statin therapy had a 38% increase in PSA during the same period. In a study of veterans who were initiated on statin therapy Hamilton et al reported an overall 4% decrease in PSA after initiating statin therapy.7 The overall decrease in PSA associated with statin therapy in our cohort was similar at 4.7% when accounting for confounding variables. More clinically relevant is that in patients with higher baseline PSA the decrease was more pronounced. Hamilton et al found that if a PSA of 4 ng/ml was used as a trigger point for biopsy, 40% of men initiating statin therapy would no longer qualify for a biopsy, which could potentially delay the diagnosis of prostate cancer. We found that in patients with higher Gleason grades there was a larger decrease in overall PSA and PSA density. This may be due to the higher PSA in these patients and, therefore, a larger effect on serum PSA by the statin medication as reported previously.7 Mills et al found that statin therapy did not decrease PSA in 174 patients treated with atorvastatin for 6 months, but the brief duration of therapy may help explain why they did not find a significant difference in PSA.24 The results from this study in collaboration with others suggest that patients on statin medications may require a lower threshold for PSA screening. However, these conclusions should be taken with caution. We and others have not provided any evidence that this decrease in PSA actually delays

122

STATINS DECREASE PROSTATE SPECIFIC ANTIGEN IN MEN WITH PROSTATE CANCER

Biochemical recurrence based on preoperative statin therapy

prostate cancer diagnosis or that it produces an increased risk of prostate cancer specific mortality. We did not identify a decreased rate of locally advanced disease in this cohort. However, this is not a screening population, but rather patients already diagnosed with prostate cancer and referred for cancer treatment. The increased rate of Gleason 7 disease in the statin cohort could also be explained by patient referral bias rather than an effect of the medication. Patients on statins with other comorbidities and less aggressive or lower volume disease may be more inclined to undergo radiation therapy or pursue active surveillance for prostate cancer. Patients on statin therapy were more likely to have diabetes or other medical comorbidities and these require routine evaluation by a primary care physician.

Table 5. Multivariate Cox proportional hazard ratios for biochemical recurrence

PSA* Pathological grade: 6 7 8–10 Pathological stage: pT2 pT3a pT3b/pT4 Pos margins Statin use * Log transformed.

HR (95% CI)

p Value

1.65 (1.33–1.98)

⬍0.0001

1—Reference 1.02 (0.77–1.32) 2.78 (1.78–4.54)

⬍0.0001

1—Reference 0.99 (0.80–1.22) 2.31 (1.76–3.06) 1.38 (1.17–1.64) 0.99 (0.83–1.18)

⬍0.0001

0.0002 0.93

Therefore, these patients are more likely to undergo routine digital rectal examination and PSA testing for prostate cancer screening. As a result prostate cancer is likely detected earlier and, therefore, at a more clinically localized stage with a lower serum PSA. This detection bias may explain why other studies have shown a decrease in locally advanced prostate cancer in patients on statin medications. The finding that the patients on statins are actually slightly older would argue against this in our patient population but these criticisms cannot be entirely ignored. Patients with increased BMI have been shown to have a lower PSA due to hemodilution or other processes.25 In this cohort of patients PSA remains relatively stable as BMI increases. However, in the multivariate regression model we see increasing BMI is associated with a PSA decrease of 0.7% for each increase in kg/m2. We note several limitations to this study, all of which indicate the need for additional study of this subject. We did not have dose, duration or indication for statin therapy available for the entire cohort. As increased duration of therapy may provide a protective benefit, this could further delineate the risk for biochemical recurrence.16 The end point for this study is not cancer specific death and, therefore, we cannot comment if statin therapy has had a role in preventing prostate cancer death as some have theorized.26 Additionally, mean followup in this cohort is 26 months, which might not be sufficiently long for the antineoplastic effects of statin therapy to be seen. Due to the low rate of biochemical recurrence

STATINS DECREASE PROSTATE SPECIFIC ANTIGEN IN MEN WITH PROSTATE CANCER

in this cohort, analyses of recurrence may be underpowered. Moreover this study does not represent a screening population of men diagnosed with prostate cancer, but rather there is selection bias because the data were retrieved from a tertiary referral center specializing in RP. Due to the heterogeneity of this patient population no standardized screening protocol or surrogate for routine medical evaluations could be established. Therefore, further longitudinal population based studies may be required to validate the effect of statin therapy on PSA in men diagnosed with prostate cancer. The change in PSA associated with statin therapy has been linked with changes in LDL levels, but as this

123

is not included in our institutional review board we cannot comment on this possible confounder.

CONCLUSIONS The role that statins have in the development of prostate cancer is still being elucidated. In this cohort of men presenting for radical prostatectomy pretreatment PSA was significantly lower in patients on statin therapy. Further study on the effect of statin therapy on PSA is warranted to ensure that patients on statin medications are being appropriately screened for prostate cancer.

REFERENCES 1. Mann D, Reynolds K, Smith D et al: Trends in statin use and low-density lipoprotein cholesterol levels among US adults: impact of the 2001 National Cholesterol Education Program guidelines. Ann Pharmacother 2008; 42: 1208. 2. Baigent C, Keech A, Kearney PM et al: Efficacy and safety of cholesterol-lowering treatment: prospective meta-analysis of data from 90,056 participants in 14 randomised trials of statins. Lancet 2005; 366: 1267. 3. Kearney PM, Blackwell L, Collins R et al: Efficacy of cholesterol-lowering therapy in 18,686 people with diabetes in 14 randomised trials of statins: a meta-analysis. Lancet 2008; 371: 117. 4. Cannon CP, Braunwald E, McCabe CH et al: Intensive versus moderate lipid lowering with statins after acute coronary syndromes. N Engl J Med 2004; 350: 1495. 5. Knopp RH: Drug treatment of lipid disorders. N Engl J Med 1999; 341: 498. 6. Cyrus-David MS, Weinberg A, Thompson T et al: The effect of statins on serum prostate specific antigen levels in a cohort of airline pilots: a preliminary report. J Urol 2005; 173: 1923.

progression-free survival after brachytherapy for clinically localized prostate cancer. Urology 2005; 66: 1150. 10. Werny DM, Saraiya M and Gregg EW: Prostatespecific antigen values in diabetic and nondiabetic US men, 2001–2002. Am J Epidemiol 2006; 164: 978. 11. Cookson MS, Aus G, Burnett AL et al: Variation in the definition of biochemical recurrence in patients treated for localized prostate cancer: the American Urological Association Prostate Guidelines for Localized Prostate Cancer Update Panel report and recommendations for a standard in the reporting of surgical outcomes. J Urol 2007; 177: 540. 12. Shannon J, Tewoderos S, Garzotto M et al: Statins and prostate cancer risk: a case-control study. Am J Epidemiol 2005; 162: 318. 13. Browning DR and Martin RM: Statins and risk of cancer: a systematic review and metaanalysis. Int J Cancer 2007; 120: 833. 14. Dale KM, Coleman CI, Henyan NN et al: Statins and cancer risk: a meta-analysis. JAMA 2006; 295: 74.

7. Hamilton RJ, Goldberg KC, Platz EA et al: The influence of statin medications on prostate-specific antigen levels. J Natl Cancer Inst 2008; 100: 1511.

15. Friedman GD, Flick ED, Udaltsova N et al: Screening statins for possible carcinogenic risk: up to 9 years of follow-up of 361,859 recipients. Pharmacoepidemiol Drug Saf 2008; 17: 27.

8. Hamilton RJ and Freedland SJ: Review of recent evidence in support of a role for statins in the prevention of prostate cancer. Curr Opin Urol 2008; 18: 333.

16. Platz EA, Leitzmann MF, Visvanathan K et al: Statin drugs and risk of advanced prostate cancer. J Natl Cancer Inst 2006; 98: 1819.

9. Moyad MA, Merrick GS, Butler WM et al: Statins, especially atorvastatin, may favorably influence clinical presentation and biochemical

17. Jacobs EJ, Rodriguez C, Bain EB et al: Cholesterol-lowering drugs and advanced prostate cancer incidence in a large U.S. cohort. Cancer Epidemiol Biomarkers Prev 2007; 16: 2213.

18. Murtola TJ, Tammela TL, Lahtela J et al: Cholesterol-lowering drugs and prostate cancer risk: a population-based case-control study. Cancer Epidemiol Biomarkers Prev 2007; 16: 2226. 19. Flick ED, Habel LA, Chan KA et al: Statin use and risk of prostate cancer in the California Men’s Health Study cohort. Cancer Epidemiol Biomarkers Prev 2007; 16: 2218. 20. Boudreau DM, Yu O, Buist DS et al: Statin use and prostate cancer risk in a large populationbased setting. Cancer Causes Control 2008; 19: 767. 21. Freeman MR and Solomon KR: Cholesterol and prostate cancer. J Cell Biochem 2004; 91: 54. 22. Hall SA, Page ST, Travison TG et al: Do statins affect androgen levels in men? Results from the Boston area community health survey. Cancer Epidemiol Biomarkers Prev 2007; 16: 1587. 23. Soto DE, Daignault S, Sandler HM et al: No effect of statins on biochemical outcomes after radiotherapy for localized prostate cancer. Urology 2009; 73: 158. 24. Mills IW, Crossland A, Patel A et al: Atorvastatin treatment for men with lower urinary tract symptoms and benign prostatic enlargement. Eur Urol 2007; 52: 503. 25. Banez LL, Hamilton RJ, Partin AW et al: Obesityrelated plasma hemodilution and PSA concentration among men with prostate cancer. JAMA 2007; 298: 2275. 26. Colli JL and Amling CL: High cholesterol levels are associated with reduced prostate cancer mortality rates during periods of high but not low statin use in the United States. Urol Oncol 2009; 27: 170.

EDITORIAL COMMENTS Several recent studies have identified the statin class of drugs as potential agents for the prevention of prostate cancer (references 12 and 16 in article).

Krane et al studied the PSA in men undergoing prostatectomy who were classified as users vs nonusers of statins. In a prostatectomy cohort they

124

STATINS DECREASE PROSTATE SPECIFIC ANTIGEN IN MEN WITH PROSTATE CANCER

found that statin users had lower PSA values than the nonuser group. Although it is tempting to conclude from this study that statins decreased PSA in the statin group, there are certain limitations that preclude this assumption. This result could have been a screening effect in that men who were screened for lipid abnormalities may have concurrently been screened for PSA values as well. In addition, men on statins could have had greater levels of obesity and, thus, had decreased PSA values due to hemodilution. This limitation is further borne out in the finding that not only did the low BMI group have a low rate of statin use, but also had

no difference in PSA between users and nonusers. It is well-known that statin drugs have significant health benefits due to cardiovascular and other potential benefits. However, direct data for the use of these drugs as chemopreventive agents are lacking. Future trials should focus on the study of these drugs in prospective trials with primary end points focused on prostate cancer outcomes.

This is the first study to demonstrate lower PSA among statin users compared with nonusers in men diagnosed with prostate cancer while considering the potential confounding effect of diabetes mellitus, a common condition among statin users and also associated with decreased PSA (reference 10 in article). These findings are in concordance with previous observations (reference 7 in article). The topic is of general public health interest because in theory lower PSA in statin users could lead to delayed prostate cancer diagnostics and possibly to more advanced tumors at diagnosis. The authors’ finding of more high grade tumors in statin users would be consistent with this notion. However, as the authors

demonstrate this does not translate into a higher short-term recurrence rate after prostatectomy. Furthermore, several population based studies have conversely reported a lower incidence of advanced and high grade tumors in statin users.1 Thus, whether the PSA decreasing effect of statins bears clinical significance and really delays prostate cancer diagnostics remains to be determined.

Mark Garzotto Department of Urology and Radiation Medicine Portland VA Medical Center Portland, Oregon

Teemu J. Murtola Department of Surgery Central Finland Central Hospital, Jyväskylä and School of Public Health, University of Tampere, Tampere, Finland

REFERENCE 1. Murtola TJ, Visakorpi T, Lahtela J et al: Statins and prostate cancer prevention: where we are now, and future directions. Nat Clin Pract Urol 2008; 5: 376.

This study adds to the building literature supporting a relationship between statins and prostate biology. It is encouraging that our finding of a 4.1% PSA reduction associated with statin use in men without prostate cancer is nearly identical to this population of selected men with prostate cancer undergoing robotic prostatectomy (4.7%) (reference 7 in article). Because statins appear to decrease PSA, it is tempting to think statins could prevent prostate cancer. However, the urological community learned a valuable lesson from the SELECT (Selenium and Vitamin E Cancer Prevention Trial): to curb enthusiasm about potential preventive agents until a definitive randomized controlled trial is performed.1

Until we have a sufficient rationale to launch randomized controlled trials to answer this question definitively, we must continue to examine the relationship between statin use and prostate biology in different clinical scenarios (eg before cancer diagnosis [prevention], during active surveillance, radiation and surgical therapy, and in hormone refractory and metastatic disease states) through epidemiological and cohort studies. Future studies addressing these issues should explore the influence of statin dose and changes in cholesterol to better understand the mechanisms through which statins may act on the prostate. Robert J. Hamilton University of Toronto Toronto, Ontario, Canada

REFERENCE 1. Lippman SM, Klein EA, Goodman PJ et al: Effect of selenium and vitamin E on risk of prostate cancer and other cancers: the Selenium and Vitamin E Cancer Prevention Trial (SELECT). JAMA 2009; 301: 39.

STATINS DECREASE PROSTATE SPECIFIC ANTIGEN IN MEN WITH PROSTATE CANCER

125

REPLY BY AUTHORS Certainly there are limitations in assuming that statins routinely decrease serum PSA. PSA has been associated with obesity and hemodilution (reference 25 in article). We could not evaluate medical compliance, duration of therapy or screening bias, and the clinical relevance of a 4.7% decrease in serum PSA is debatable. However, the decrease in PSA with statin therapy has been demonstrated in large screening (reference 7 in article) and prostate cancer populations. This association does not appear to be accidental and warrants further investigation.

We agree that decreased PSA in men presenting for prostate cancer can be a concern and could potentially cause a delay in diagnosis. Fortunately we did not see any difference in the biochemical recurrence rate between men taking and not taking statins preoperatively. This article adds to the mounting literature suggesting that there is a complex interaction between statin medications and prostate cancer. While it is tempting to initiate a trial to see if statin use can prevent prostate cancer, a mechanism of the biological interaction must be elucidated. Otherwise such a trial would be premature.