Statin Use and Risk of Prostate Cancer in a Population of Men Who Underwent Biopsy

Transcription

1 Statin Use and Risk of Prostate Cancer in a Population of Men Who Underwent Biopsy Nelly Tan,* Eric A. Klein, Jianbo Li, Ayman S. Moussa and J. Stephen Jones From the Glickman Urological and Kidney Institute, Cleveland Clinic, Cleveland, Ohio Abbreviations and Acronyms BMI body mass index DRE digital rectal examination GS Gleason score PB prostate biopsy PC prostate cancer PSA prostate specific antigen Submitted for publication October 21, Study received institutional review board approval. * Correspondence: Department of Urology, Yale-New Haven Hospital, New Haven, Connecticut ( nelly.tan@gmail.com). Financial interest and/or other relationship with Pfizer, Cook, Abbott, Endocare and GTx. For another article on a related topic see page 302. Purpose: We determined the association between statin use and prostate cancer in men who underwent prostate biopsy. Materials and Methods: We performed a retrospective cohort study of men who underwent prostate biopsy from 2000 to 2007 at Cleveland Clinic. Statin use was determined using outpatient pharmacy records, and clinical and pathological outcomes were obtained. Multivariate logistic regression analysis to determine the effects of statins (and duration of use) was performed after adjusting for age, body mass index, African-American race, number of cores taken and prostate volume. Results: We analyzed data from 4,204 patients, and we identified 3,182 (75.7%) not on statins and 1,022 on statins. Men diagnosed with prostate cancer on statins compared to those not taking statins were less likely to have digital rectal examination positivity (5.3% vs 8.9%, OR 0.7, p 0.01), Gleason score 7 or greater prostate cancer (61.4% vs 72.4%, OR 0.78, p 0.02) and high volume prostate cancer (27.2 vs 31.4, p 0.01). Moreover statin users had lower prostate specific antigen compared to nonstatin users (5.13 vs 5.98, p 0.03). Multivariate analysis adjusted risk ratios for prostate cancer diagnosis, high grade prostate cancer (Gleason score 7 or greater) and 3 or more cores positive in statin users were 0.92 (95% CI ), 0.76 (95% CI ) and 0.86 (95% CI ) and only high grade prostate cancer persisted with length of use. Conclusions: Statin use was associated with a decreased risk of prostate cancer, less frequent high grade prostate cancer and lower volume of prostate cancer, suggesting that statin use has a protective effect against prostate cancer. Key Words: prostatic neoplasms, hydroxymethylglutaryl-coa reductase inhibitors PROSTATE cancer is the second most common cancer in the United States. Efforts to prevent this cancer remain important because screening has been shown to have only a modest effect on reducing mortality, 1,2 and while effective in curing early stage tumors, definitive therapy is associated with significant morbidity. Statins are a class of cholesterol lowering drugs that work by inhibiting 3-hydroxy-3-methylglutaryl coenzyme-a reductase and have recently demonstrated a chemopreventive effect in vitro on PC cell cultures. 3,4 There is controversy regarding the effects of statins on the overall risk reduction of PC, with some studies suggesting no effect while others suggest significant risk reduction, especially in advanced stage tumors. 5 7 However, there are minimal published data on a population of men being screened for prostate cancer and outcomes such as tumor volume have not been explored. We performed a retrospective cohort study to investi /11/ /0 Vol. 186, 86-90, July 2011 THE JOURNAL OF UROLOGY Printed in U.S.A by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH, INC. DOI: /j.juro

2 STATIN USE AND RISK OF PROSTATE CANCER 87 gate the association of statin use, and PC risk, high grade PC and PC volume in men who underwent prostate biopsy. MATERIALS AND METHODS Study Setting We performed a retrospective cohort study of patients who underwent prostate biopsy at a tertiary medical center, Cleveland Clinic. We used an institutional review board approved database of patients who underwent PB. All patients included in the study had a urologist perform the initial prostate biopsy from 2000 to Patients were excluded from study if we did not have complete pathological information. Study Subjects The population of patients evaluated at Cleveland Clinic includes self-referrals and provider-to-provider referrals. We identified 4,733 patients who underwent consecutive PB for increased PSA and/or abnormal DRE. Of the 4,733 men 529 (11.1%) were excluded from study because of incomplete pathological data (ie tumor grade information). We analyzed the remaining 4,204 patients, and we identified 3,182 (75.7%) patients not on statins and 1,022 who used statins. Exposure Statin use was evaluated in all patients who had electronic prescribing records, which were available for all patients in the study. Patients taking medications containing lovastatin, atorvastatin, cerivastatin, rosuvastatin, simvastatin, fluvastatin and/or pravastatin were considered statin users. Men were categorized as statin users if they were taking statins before undergoing PB. Those who did not have statin use information before biopsy were considered nonusers. The duration of statin use was determined by taking the difference between the start date of statin use and the date of PB. Statin use duration was categorized as 0 to 5 years and greater than 5-year use before PB was performed. This breakdown in years was used because a previous study suggested that a protective effect of statins was seen in men using statins for more than 5 years. 8 Clinical Characteristics Clinical variables collected were patient age, race (African-American vs all other), DRE status (positive or negative), BMI, serum PSA and prostate volume. Prostate volume was measured by transrectal ultrasound or by surgical prostatectomy specimen weight when transrectal ultrasound estimated volume was unavailable, which was the case in 20% of the patients. BMI was divided into 3 groups per National Institutes of Health obesity guidelines of less than 25, 25 to 30 and greater than 30 kg/m 2. 9 Prostate Cancer Ascertainment Pathological information collected from the pathology report of our electronic medical records included biopsy Gleason sum, number of cores obtained and number of cores positive. Biopsy specimens were reviewed by dedicated genitourinary pathologists according to a previously described protocol. 10 Indications for biopsy and prostate biopsy technique were published previously. 11 Tumor volume was assessed by the 2 measures of 3 or more cores positive for PC (considered to have high volume PC) and the percent of cores positive (number of cores positive/total cores taken). 12,13 Statistical Analysis Data were presented as means with standard deviations and medians with interquartile ranges. Group comparisons were done using the Wilcoxon rank sum test for continuous variables and the chi-square test for categorical variables. Multivariate logistic regression analysis was used to assess the effect of statin use and the duration of statin use on PC incidence, high grade PC and PC volume. Potential confounding factors such as age, BMI, African-American race, DRE positivity, prostate volume and number of cores surveyed were used as covariates in the modeling. The resulting odds ratios and CIs were converted to risk ratios for easy interpretation. A correlation test was used for the relationship between GS and duration of statin use. All analyses were done using the statistical software package R (R Development Core Team, and statistics were considered significant at RESULTS We analyzed 4,204 patients including 3,182 (75.7%) not on statins and 1,022 using statins. Patient clinical and pathological data are presented in table 1. Of the statin users 55.2% were diagnosed with PC, whereas 57.8% of patients not on statins had PC (p 0.15). The mean age of men with PC who were not on statin and who were on statin was 63.5 and 65.7 (p 0.01), respectively. Compared to patients not taking statins, men diagnosed with PC who were on statins were less likely to have DRE positivity (5.3 vs 8.9, p 0.01), GS 7 or greater PC (OR 0.78, p 0.02) and high volume prostate cancer (27.2 vs 31.4, p 0.01). Median PSA was lower in men being screened for PC who were on statins compared to those who were not on statins (5.13 vs 5.98, p 0.03). However, there was no difference in median PSA for men diagnosed with PC (6.44 vs 7.03, p 0.94). Multivariate logistic regression analysis demonstrated sustained protective associations between statin use and prostate cancer risk. After adjusting for age, BMI, African-American race, DRE, prostate volume and number of cores surveyed, patients on statins at any time were less likely to have prostate cancer (RR 0.92, 95% CI ) and high grade prostate cancer (RR 0.76, 95% CI ). Statin users also had a lower incidence of high volume (3 or more cores positive) PC (RR 0.86, 95% CI ). Moreover greater than 5 years of statin use predicted a lower rate of high grade PC (RR 0.75, 95% CI , table 2). The cumulative incidence of prostate cancer, high grade PC and high

3 88 STATIN USE AND RISK OF PROSTATE CANCER Table 1. Patient clinical and pathological data Statin Nonusers Statin Users Chi-Square p Value No. underwent PB (%) 3,182 (75.7) 1,022 (24.3) No. no PC (%) 1,340 (42.1) 457 (44.7) 0.15 No. GS 6 or greater (%) 1,842 (57.8) 565 (55.2) Mean age of men with PC (IQR) 63.5 (57 69) 65.7 (60 71) 0.01 No. African-American with PC (%) 370 (11.6) 157 (15.4) 0.01 No. DRE pos with PC (%) 275 (8.9) 54 (5.3) 0.01 No. kg/m 2 BMI with PC (%): Less than (23.6) 165 (18.3) ,026 (49.2) 432 (47.9) Greater than (27.2) 305 (33.8) Median PSA in all men (IQR) 5.98 ( ) 5.13 ( ) 0.01 Median PSA in men with PC (IQR) 6.44 ( ) 7.03 ( ) 0.94 No. pathological Gleason sum (%): (27.6) 218 (38.6) (42.5) 186 (32.9) (15.4) 77 (13.6) 8 or Higher 268 (14.5) 84 (14.9) No. cores obtained in PB (IQR) 11.6 (10 12) 12.6 (10 12) 0.01 No. pos cores (IQR) 2.0 (0 3) 1.9 (0 3) 0.36 No. 3 or more cores pos (%) 976 (31.4) 274 (27.2) 0.01 Mean cc prostate vol in men with PC (IQR) 48.1 (33 57) 46.0 ( ) 0.01 volume PC was lower in statin users as patient age increased (see figure). DISCUSSION In this cohort of patients the results indicate that statin users who underwent biopsy had a lower rate of DRE positivity, lower PSA, smaller prostate, as well as a lower rate of high grade and lower volume PC compared to nonusers who underwent biopsy. In addition, on multivariate analysis longer duration of statin therapy was associated with an overall lower rate of high grade PC. The lower risk of PC may be associated with antineoplastic effects of statins beyond the PSA suppression effects observed in published studies Since increased PSA is a common indication for prostate biopsy, suppression of PSA in statin users may complicate PC detection in men on statins. Consistent with other studies the median PSA in our patient population was 14% lower in statin users who underwent biopsy. However, in patients diagnosed with PC there was no difference in PSA, suggesting PSA lowering effects of statins in benign tissue but not in malignant tissues. Moreover since PSA was similar in the statin and control cohorts of men diagnosed with PC, one would presume a similar breakdown of low, moderate and high grade PC. Instead we found that statin users had a lower rate of GS 7 or greater PC, which suggests an antineoplastic effect of statin use. One mechanism by which statins could induce an antineoplastic effect is via apoptosis of prostate cancer cells, which was demonstrated in in vitro studies. 17,18 Some have suggested that the statin user population may inherently undergo increased screening, but published studies have shown the reduced risk of PC in statin users after adjusting for PSA testing. 6,15 Murtola et al compared statin users and nonusers undergoing screening for prostate cancer, and found that the overall incidence of PC was lower among statin users when bias due to differential PSA testing was eliminated. 15 Thus, more evidence is beginning to suggest that statins not only have PSA lowering effects, but they also may possess antineoplastic potential. Table 2. Risk ratio of overall incidence of PC, high grade PC and PC volume based on length of statin use Underwent PB No PC PC (GS 6 or greater) High Grade PC (GS 7 or greater) 3 or More Cores Pos No. no statin use (%) 3,182 1,340 (42.1) 1,842 (57.8) 1,334 (72.4) 976 (52.9) No. ever statin use (%): 1, (44.7) 565 (55.2) 347 (61.4) 274 (48.4) 0 5 Yrs (45.2) 495 (54.7) 305 (61.6) 241 (48.6) Greater than 5 yrs (40.6) 70 (59.3) 42 (60.0) 33 (47.1) RR (95% CI) ever statin use: Ref Ref 0.92 ( ) 0.76 ( ) 0.86 ( ) 0 5 Yrs Ref Ref 0.92 ( ) 0.78 ( ) 0.80 ( ) Greater than 5 yrs Ref Ref 0.95 ( ) 0.72 ( ) 0.81 ( )

4 STATIN USE AND RISK OF PROSTATE CANCER 89 Cumulative incidence of overall prostate cancer (A), high grade prostate cancer (B) and 3 or more cores positive (C) The effect of statins on the incidence of PC has been controversial in published studies. Reported study results have ranged from increased incidence of PC, 5 to no effect on PC incidence, 19 to a decrease in overall incidence, 20 to a decrease only in high grade/advanced PC 6,21,22 in statin users. Friis et al conducted a population based cohort study, and found no difference in PC between statin users and nonusers. 19 In contrast, Jacobs et al reported a reduced risk of only advanced PC in statin users (rate ratio 0.60, 95% CI ), 6 later supported by additional studies. 5,21,22 Other studies have demonstrated a reduced risk of PC 8,20,23 and high grade PC 6 in statin users. There is no consensus on the exact effect of statin use on PC in these population based cohort studies. All patients in our study underwent biopsy because of positive DRE or increased PSA and, thus, if statins have protective effects against PC as demonstrated in our study for overall PC and for high grade PC, those undergoing PB may be the group likely to show the association. Our results also demonstrate that statin use was associated with a lower volume of PC. Similar findings of lower prostate volume have been reported in men taking statins. Loeb et al studied prostatectomy specimens and found that men on statins had a lower percentage and lower volume of cancer. 24 Lower PC volume and lower rate of high grade cancers in men on statins may be explained by the cholesterol lowering effects of statins. Freeman and Solomon proposed a mechanism by which cholesterol could promote PC development, 25 in that high serum cholesterol contributes to the lipid raft component of the plasma membrane, which may stimulate signaling pathways that promote progression to hormone refractory disease. A nested case-control study demonstrated that a low plasma level was associated with a lower risk of high grade disease. 26 An additional study evaluated the association between serum cholesterol and prostate cancer, particularly high grade disease, in the placebo arm of the Prostate Cancer Prevention Trial. The study showed that men with normal serum cholesterol had a statistically significant lower risk of GS 8 to 10 PC (OR 0.41, 95% CI ), especially when the disease was also organ confined (OR 0.32, 95% CI ). 27 The protective effects of statins have been further suggested in several studies that report improved oncologic outcomes after treatment for PC. Gutt et al retrospectively analyzed 691 men treated with curative intent radiotherapy, and found that statin use was associated with improved freedom from biochemical failure and relapse-free survival. 28 Similar improved outcomes were also observed after prostatectomy. Hamilton et al examined 1,319 patients treated with radical prostatectomy and found that statin users had a dose dependent lower risk of PSA recurrence postoperatively. 29 In addition to potentially improved pretreatment effects of statins, these studies also purport additional benefits, specifically in posttreatment outcomes. There are a number of limitations to our study. We did not have information on certain potential confounders including comorbidities such as diabetes and coronary artery disease nor did we have information for medications considered to affect PC risk, such as 5 -reductase inhibitors, aspirin and antidiabetic use. In addition, there is selection bias. Since we have information for men who underwent biopsy, we cannot apply our results to the men who did not undergo biopsy. Our database is also from 1 tertiary care facility center. Thus, referral patterns and institutional practices may not be generalizable to the larger population. We do not have a number of PSA testing data for our patients and, thus, we cannot control for screening differences between the

5 90 STATIN USE AND RISK OF PROSTATE CANCER 2 populations. Finally, this is a retrospective study and so we cannot provide a causal relationship between statin use and risk of PC. Despite these limitations, this is the largest study published to our knowledge to show the effects of statins on PC in a population of men who underwent PB. This study is also the first to demonstrate statin effects on prostate cancer volume in a population of men before treatment. Furthermore, the lower prostate volume in statin users suggests the possibility that statins could prevent the development of benign prostatic hyperplasia. Additionally, the multivariate analysis of this study was rigorous, and controlled for more potential risk factors, confounders, than most published studies, including prostate volume, number of cores taken, BMI, DRE positivity, race and age. CONCLUSIONS This study provides evidence that patients screened for PC who use statins are less likely to have cancer detected at the time of initial biopsy, less likely to have moderate to high grade PC and more likely to have lower PC volume compared to nonstatin users. ACKNOWLEDGMENTS Kirsten I. Fishler and Zen Vuong provided critical review of the manuscript. REFERENCES 1. Andriole GL, Crawford ED, Grubb RL 3rd et al: Mortality results from a randomized prostate-cancer screening trial. N Engl J Med 2009; 360: Grubb RL 3rd, Pinsky PF, Greenlee RT et al: Prostate cancer screening in the Prostate, Lung, Colorectal and Ovarian cancer screening trial: update on findings from the initial four rounds of screening in a randomized trial. BJU Int 2008; 102: Wong WW, Dimitroulakos J, Minden MD et al: HMG-CoA reductase inhibitors and the malignant cell: the statin family of drugs as triggers of tumorspecific apoptosis. Leukemia 2002; 16: Hoque A, Chen H and Xu XC: Statin induces apoptosis and cell growth arrest in prostate cancer cells. Cancer Epidemiol Biomarkers Prev 2008; 17: Agalliu I, Salinas CA, Hansten PD et al: Statin use and risk of prostate cancer: results from a population-based epidemiologic study. Am J Epidemiol 2008; 168: 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: 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: 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: Clinical Guidelines on the Identification, Evaluation, and Treatment of Overweight and Obesity in Adults The Evidence Report. National Institutes of Health. Obes Res, suppl., 1998; 6: 51S. 10. Epstein JI, Allsbrook WC Jr, Amin MB et al: The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. Am J Surg Pathol 2005; 29: Tan N, Lane BR, Li J et al: Prostate cancers diagnosed at repeat biopsy are smaller and less likely to be high grade. J Urol 2008; 180: Freedland SJ, Csathy GS, Dorey F et al: Percent prostate needle biopsy tissue with cancer is more predictive of biochemical failure or adverse pathology after radical prostatectomy than prostate specific antigen or Gleason score. J Urol 2002; 167: Epstein JI, Walsh PC, Carmichael M et al: Pathologic and clinical findings to predict tumor extent of nonpalpable (stage T1c) prostate cancer. JAMA 1994; 271: Hamilton RJ, Goldberg KC, Platz EA et al: The influence of statin medications on prostate-specific antigen levels. J Natl Cancer Inst 2008; 100: Murtola TJ, Tammela TL, Määttänen L et al: Prostate cancer and PSA among statin users in the Finnish prostate cancer screening trial. Int J Cancer 2010; 127: Krane LS, Kaul SA, Stricker HJ et al: Men presenting for radical prostatectomy on preoperative statin therapy have reduced serum prostate specific antigen. J Urol 2010; 183: Lee SJ, Ha MJ, Lee J et al: Inhibition of the 3-hydroxy-3-methylglutaryl-coenzyme A reductase pathway induces p53-independent transcriptional regulation of p21(waf1/cip1) in human prostate carcinoma cells. J Biol Chem 1998; 273: Park C, Lee I and Kang WK: Lovastatin-induced E2F-1 modulation and its effect on prostate cancer cell death. Carcinogenesis 2001; 22: Friis S, Poulsen AH, Johnsen SP et al: Cancer risk among statin users: a population-based cohort study. Int J Cancer 2005; 114: Breau RH, Karnes RJ, Jacobson DJ et al: The association between statin use and the diagnosis of prostate cancer in a population based cohort. J Urol 2010; 184: Murtola TJ, Pennanen P, Syvala H et al: Effects of simvastatin, acetylsalicylic acid, and rosiglitazone on proliferation of normal and cancerous prostate epithelial cells at therapeutic concentrations. Prostate 2009; 69: Platz EA, Leitzmann MF, Visvanathan K et al: Statin drugs and risk of advanced prostate cancer. J Natl Cancer Inst 2006; 98: Shannon J, Tewoderos S, Garzotto M et al: Statins and prostate cancer risk: a case-control study. Am J Epidemiol 2005; 162: Loeb S, Kan D, Helfand BT et al: Is statin use associated with prostate cancer aggressiveness? BJU Int 2010; 105: Freeman MR and Solomon KR: Cholesterol and prostate cancer. J Cell Biochem 2004; 91: Platz EA, Clinton SK and Giovannucci E: Association between plasma cholesterol and prostate cancer in the PSA era. Int J Cancer 2008; 123: Platz EA, Till C, Goodman PJ et al: Men with low serum cholesterol have a lower risk of high-grade prostate cancer in the placebo arm of the prostate cancer prevention trial. Cancer Epidemiol Biomarkers Prev 2009; 18: Gutt R, Tonlaar N, Kunnavakkam R et al: Statin use and risk of prostate cancer recurrence in men treated with radiation therapy. J Clin Oncol 28: Hamilton RJ, Banez LL, Aronson WJ et al: Statin medication use and the risk of biochemical recurrence after radical prostatectomy: results from the Shared Equal Access Regional Cancer Hospital (SEARCH) Database. Cancer 116: 3389.