The Prostate 75:211^217 (2015)
Statin Use after Radical Prostatectomy Reduces Biochemical Recurrence in MenWith Prostate Cancer Cheryn Song,1 Sejun Park,2 Jinsung Park,3 Myungsun Shim,1 Aram Kim,1 In Gab Jeong,1 Jun Hyuk Hong,1 Choung-Soo Kim,1 and Hanjong Ahn1* 1
Department of Urology, Asan Medical Center,University of Ulsan College of Medicine, Seoul, Korea 2 Department of Urology,Ulsan University Hospital,Ulsan, Korea 3 Department of Urology, Eulji University Hospital, Daejeon, Korea
BACKGROUND. The impact of statin use on biochemical recurrence (BCR) of prostate cancer after radical prostatectomy (RP) is controversial. METHODS. In 2,137 consecutive RP patients between 1998 and 2011 at Asan Medical Center, we aimed to assess the presence and impact of statin use according to types (hydrophilic vs. hydrophobic), dose equivalents (DEs), and postoperative duration of usage (36 months). Between non-users and preoperative or postoperative users, clinicopathological characteristics, and impact of statin use on BCR were analyzed using Cox proportional hazards model. Mean (range) follow-up was 39.4 (8–183) months. RESULTS. Compared to non-users, preoperative users had lower PSA (5.8 vs. 7 ng/ml), but the rates of organ confined disease, pathologic Gleason score (GS) or positive surgical margin (PSM) were not different. After adjusting for pathologic stage, postoperative statin use was associated with a higher BCR-free survival. In multivariate analysis, 36 months’ statin use independently decreased the risk of BCR along with PSA, pathologic GS, pathologic stage, and PSM. Risk reduction was observed especially in patients with pathologic GS 7 (HR 0.27, 95% CI 0.13–0.59, P ¼ 0.001), NSM disease (HR 0.18, 95% CI 0.05–0.63, P ¼ 0.007), or PSA 10.0 ng/ml (HR 0.30, 95% CI 0.11–0.81, P ¼ 0.018). Increasing duration of use nullified the effect. Preoperative statin use did not demonstrate significant risk reduction for BCR in any of the stratified multivariate models. CONCLUSION. In Korean men undergoing RP, preoperative statin use was not associated with different pathologic outcome. However, postoperative statin use until 36 months decreased the risk of BCR independently especially in patients with high-risk disease. Prostate 75:211–217, 2015. # 2014 Wiley Periodicals, Inc. KEY WORDS: prostate cancer; radical prostatectomy; statin; HMG-CoA reductase inhibitor; biochemical recurrence; prognosis
INTRODUCTION Statins (3-hydroxy-3-methylglutaryl coenzyme A [HMG-CoA] reductase inhibitors) are widely prescribed for the treatment of hypercholesterolemia and have been shown to reduce cardiovascular events and mortality in several randomised clinical trials [1–3]. Statins have been shown to arrest cell cycle progression [4], induce apoptosis [5], suppress angiogenesis, and alter the adhesion and migration of tumor cells [6] in vitro and in vivo. With respect to prostate cancer, the mechanisms through which statins inhibit prostate carcinogenesis or progression processes can be classified into cholesterol-mediated and non-cholesterolmediated. Statin decreases serum cholesterol level by ß 2014 Wiley Periodicals, Inc.
inhibiting HMG-CoA reductase. As cholesterol is a major component in lipid rafts, which are involved in signaling pathways regulating prostate cancer cell survival and proliferation, reduction in cholesterol and changes in cell wall lipid raft integrity adversely Conflict of interest: None. Correspondence to: Hanjong Ahn, Department of Urology, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap 2 Dong, Songpa-gu, Seoul 138-736, Korea. E-mail: [email protected] Received 14 May 2014; Accepted 27 August 2014 DOI 10.1002/pros.22907 Published online 18 October 2014 in Wiley Online Library (wileyonlinelibrary.com).
212
Song et al.
influences these signaling pathways [7,8]. Additionally, statins may influence the rate of prostate cancer progression by inhibiting the synthesis of isoprenoids. Apart from cholesterol-mediated effects, statins were shown to induce anti-angiogenic as well as antiinflammatory effects although these molecular effects were considered to be of less clinical significance. Several observational studies evaluated the relationship between statin use and prostate cancer risk [8–11]. With variability across the studies, several studies uniformly reported that the use of statin was inversely related to the overall risk of prostate cancer especially the aggressive prostate cancer with long term statin use. Based on these finding, statin were suggested to be potentially chemopreventive. Furthermore, the association between statin use and biochemical recurrence (BCR) of prostate cancer after radical prostatectomy (RP) has been examined evaluating the statin’s potential for recurrence prevention. However, results were mostly inconsistent across the studies, and two recent meta-analyses [12,13] concluded that statin use was not associated with BCR after RP. As pointed out in the meta-analyses, study populations were heterogeneous with most studies not adjusting for the type, dose or duration of statin use. In the current study, in a homogeneous group of Korean patients who underwent RP we aimed to identify the impact of perioperative statin use on BCR after RP. MATERIALS AND METHODS Study Population After obtaining institutional review board approval, data from 2,666 consecutive Korean patients with prostate cancer who had undergone RP between 1998 and 2011 at Asan Medical Center were reviewed retrospectively. RP was performed by three surgeons via either open retropubically or with robot assistance. Following surgery, patients were followed for serum PSA every 3–6 months depending on pathologic stage for the initial 2 years, every 6 months for 3 years, and annually thereafter. After excluding 217 patients who received neoadjuvant therapy or adjuvant therapy and 312 patients without sufficient data, 2,137 patients remained for analysis in the study to assess BCR, defined as PSA greater than 0.2 ng/ml.
prescription. We obtained the detailed data on preoperative statin use from each patient’s medical record. Postoperative statin use was evaluated through telephone survey, which included questions on continued or discontinued use (for preoperative users), new use after surgery (for non-users), types and dose of statins. When unclear, we contacted prescribing physicians. Of the 2,137 patients, 241 patients (11.3%) were preoperative users. After surgery, 211 additional patients had begun using statin. Postoperative users were further categorised as 36 months (n ¼ 176, 38.9%) according to the postoperative duration of use. There were seven types of statins used: atorvastatin (57.2%), simvastatin (19.2%), rosuvastatin (11.4%), pitavastatin (6.2%), pravastatin (3.1%), lovastatin (2.0%), and fluvastatin (0.9%). Statin doses were translated into dose equivalents (DEs) based on published guideline [14], with simvastatin 20 mg being assigned a value of 1. Pitavastatin users (n ¼ 28) were excluded for analysis of DEs because of the efficacy of pitavastatin was not clear. DEs 1 was 38 (8.9%), 280 (66.0%) and 106 (25.0%) patients, respectively. Median duration of statin use was 27.7 months (IQR 16.6, 47.8). Statistical Analysis Clinicopathological characteristics were compared between preoperative statin users and non-users using chi-square and Student T-test. Kaplan–Meier method was used to evaluate the risk of BCR according to postoperative statin use excluding preoperative users (n ¼ 241), and multivariate analysis was performed to evaluate the impact of statin use including type, dose and duration of statins on BCR. According to the duration of usage, statin users were categorised as 36 months users. Type of statin was categorised as hydrophilic (pravastatin, rosuvastatin) or hydrophobic (atorvastatin, simvastatin, lovastatin, and fluvastatin) according to the hydrophilicity. Also according to the calculated DE, statins were categorised as DE ¼ 1 and DE >1 or 0.05). Postoperative statin users demonstrated a significantly different pathologic stage distribution com-
213
pared to the non-users (Table I). While all statin users appeared to have better pathologic stage than the nonusers, users with
Statin Use after Radical Prostatectomy Reduces Biochemical Recurrence in MenWith Prostate Cancer Cheryn Song,1 Sejun Park,2 Jinsung Park,3 Myungsun Shim,1 Aram Kim,1 In Gab Jeong,1 Jun Hyuk Hong,1 Choung-Soo Kim,1 and Hanjong Ahn1* 1
Department of Urology, Asan Medical Center,University of Ulsan College of Medicine, Seoul, Korea 2 Department of Urology,Ulsan University Hospital,Ulsan, Korea 3 Department of Urology, Eulji University Hospital, Daejeon, Korea
BACKGROUND. The impact of statin use on biochemical recurrence (BCR) of prostate cancer after radical prostatectomy (RP) is controversial. METHODS. In 2,137 consecutive RP patients between 1998 and 2011 at Asan Medical Center, we aimed to assess the presence and impact of statin use according to types (hydrophilic vs. hydrophobic), dose equivalents (DEs), and postoperative duration of usage (36 months). Between non-users and preoperative or postoperative users, clinicopathological characteristics, and impact of statin use on BCR were analyzed using Cox proportional hazards model. Mean (range) follow-up was 39.4 (8–183) months. RESULTS. Compared to non-users, preoperative users had lower PSA (5.8 vs. 7 ng/ml), but the rates of organ confined disease, pathologic Gleason score (GS) or positive surgical margin (PSM) were not different. After adjusting for pathologic stage, postoperative statin use was associated with a higher BCR-free survival. In multivariate analysis, 36 months’ statin use independently decreased the risk of BCR along with PSA, pathologic GS, pathologic stage, and PSM. Risk reduction was observed especially in patients with pathologic GS 7 (HR 0.27, 95% CI 0.13–0.59, P ¼ 0.001), NSM disease (HR 0.18, 95% CI 0.05–0.63, P ¼ 0.007), or PSA 10.0 ng/ml (HR 0.30, 95% CI 0.11–0.81, P ¼ 0.018). Increasing duration of use nullified the effect. Preoperative statin use did not demonstrate significant risk reduction for BCR in any of the stratified multivariate models. CONCLUSION. In Korean men undergoing RP, preoperative statin use was not associated with different pathologic outcome. However, postoperative statin use until 36 months decreased the risk of BCR independently especially in patients with high-risk disease. Prostate 75:211–217, 2015. # 2014 Wiley Periodicals, Inc. KEY WORDS: prostate cancer; radical prostatectomy; statin; HMG-CoA reductase inhibitor; biochemical recurrence; prognosis
INTRODUCTION Statins (3-hydroxy-3-methylglutaryl coenzyme A [HMG-CoA] reductase inhibitors) are widely prescribed for the treatment of hypercholesterolemia and have been shown to reduce cardiovascular events and mortality in several randomised clinical trials [1–3]. Statins have been shown to arrest cell cycle progression [4], induce apoptosis [5], suppress angiogenesis, and alter the adhesion and migration of tumor cells [6] in vitro and in vivo. With respect to prostate cancer, the mechanisms through which statins inhibit prostate carcinogenesis or progression processes can be classified into cholesterol-mediated and non-cholesterolmediated. Statin decreases serum cholesterol level by ß 2014 Wiley Periodicals, Inc.
inhibiting HMG-CoA reductase. As cholesterol is a major component in lipid rafts, which are involved in signaling pathways regulating prostate cancer cell survival and proliferation, reduction in cholesterol and changes in cell wall lipid raft integrity adversely Conflict of interest: None. Correspondence to: Hanjong Ahn, Department of Urology, Asan Medical Center, University of Ulsan College of Medicine, 388-1 Pungnap 2 Dong, Songpa-gu, Seoul 138-736, Korea. E-mail: [email protected] Received 14 May 2014; Accepted 27 August 2014 DOI 10.1002/pros.22907 Published online 18 October 2014 in Wiley Online Library (wileyonlinelibrary.com).
212
Song et al.
influences these signaling pathways [7,8]. Additionally, statins may influence the rate of prostate cancer progression by inhibiting the synthesis of isoprenoids. Apart from cholesterol-mediated effects, statins were shown to induce anti-angiogenic as well as antiinflammatory effects although these molecular effects were considered to be of less clinical significance. Several observational studies evaluated the relationship between statin use and prostate cancer risk [8–11]. With variability across the studies, several studies uniformly reported that the use of statin was inversely related to the overall risk of prostate cancer especially the aggressive prostate cancer with long term statin use. Based on these finding, statin were suggested to be potentially chemopreventive. Furthermore, the association between statin use and biochemical recurrence (BCR) of prostate cancer after radical prostatectomy (RP) has been examined evaluating the statin’s potential for recurrence prevention. However, results were mostly inconsistent across the studies, and two recent meta-analyses [12,13] concluded that statin use was not associated with BCR after RP. As pointed out in the meta-analyses, study populations were heterogeneous with most studies not adjusting for the type, dose or duration of statin use. In the current study, in a homogeneous group of Korean patients who underwent RP we aimed to identify the impact of perioperative statin use on BCR after RP. MATERIALS AND METHODS Study Population After obtaining institutional review board approval, data from 2,666 consecutive Korean patients with prostate cancer who had undergone RP between 1998 and 2011 at Asan Medical Center were reviewed retrospectively. RP was performed by three surgeons via either open retropubically or with robot assistance. Following surgery, patients were followed for serum PSA every 3–6 months depending on pathologic stage for the initial 2 years, every 6 months for 3 years, and annually thereafter. After excluding 217 patients who received neoadjuvant therapy or adjuvant therapy and 312 patients without sufficient data, 2,137 patients remained for analysis in the study to assess BCR, defined as PSA greater than 0.2 ng/ml.
prescription. We obtained the detailed data on preoperative statin use from each patient’s medical record. Postoperative statin use was evaluated through telephone survey, which included questions on continued or discontinued use (for preoperative users), new use after surgery (for non-users), types and dose of statins. When unclear, we contacted prescribing physicians. Of the 2,137 patients, 241 patients (11.3%) were preoperative users. After surgery, 211 additional patients had begun using statin. Postoperative users were further categorised as 36 months (n ¼ 176, 38.9%) according to the postoperative duration of use. There were seven types of statins used: atorvastatin (57.2%), simvastatin (19.2%), rosuvastatin (11.4%), pitavastatin (6.2%), pravastatin (3.1%), lovastatin (2.0%), and fluvastatin (0.9%). Statin doses were translated into dose equivalents (DEs) based on published guideline [14], with simvastatin 20 mg being assigned a value of 1. Pitavastatin users (n ¼ 28) were excluded for analysis of DEs because of the efficacy of pitavastatin was not clear. DEs 1 was 38 (8.9%), 280 (66.0%) and 106 (25.0%) patients, respectively. Median duration of statin use was 27.7 months (IQR 16.6, 47.8). Statistical Analysis Clinicopathological characteristics were compared between preoperative statin users and non-users using chi-square and Student T-test. Kaplan–Meier method was used to evaluate the risk of BCR according to postoperative statin use excluding preoperative users (n ¼ 241), and multivariate analysis was performed to evaluate the impact of statin use including type, dose and duration of statins on BCR. According to the duration of usage, statin users were categorised as 36 months users. Type of statin was categorised as hydrophilic (pravastatin, rosuvastatin) or hydrophobic (atorvastatin, simvastatin, lovastatin, and fluvastatin) according to the hydrophilicity. Also according to the calculated DE, statins were categorised as DE ¼ 1 and DE >1 or 0.05). Postoperative statin users demonstrated a significantly different pathologic stage distribution com-
213
pared to the non-users (Table I). While all statin users appeared to have better pathologic stage than the nonusers, users with
