- Email: [email protected]
The association of statin therapy with clinicopathologic outcomes and survival among patients with localized renal cell carcinoma undergoing nephrectomy
Urologic Oncology: Seminars and Original Investigations ] (2015) ∎∎∎–∎∎∎
Original article
The association of statin therapy with clinicopathologic outcomes and survival among patients with localized renal cell carcinoma undergoing nephrectomy Boyd R. Viers, M.D.a, R. Houston Thompson, M.D.a, Sarah P. Psutka, M.D.a, Christine M. Lohse, M.S.b, John C. Cheville, M.D.c, Bradley C. Leibovich, M.D.a, Matthew K. Tollefson, M.D.a, Stephen A. Boorjian, M.D.a,* b
a Department of Urology, Mayo Clinic, Rochester, MN Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN c Department of Anatomic Pathology, Mayo Clinic, Rochester, MN
Received 25 November 2014; received in revised form 14 January 2015; accepted 14 January 2015
Abstract Objectives: Although statins have been found to induce apoptosis and demonstrate antimetastases activity both in vitro and in vivo for renal cell carcinoma (RCC), clinical evidence of a role for these medications is limited. We evaluated the association of statin therapy with outcomes among patients with surgically treated localized RCC. Methods and materials: We reviewed 2,357 patients who underwent nephrectomy between 1995 and 2009 for pNx/0, M0 RCC. Of these, 630 (27%) were taking statins within 3 months of surgery. Progression-free survival, cancer-specific survival, and overall survival were estimated using the Kaplan-Meier method. The associations of statin use with clinicopathologic outcomes were evaluated with multivariable logistic and proportional hazards regression models. Results: Statin therapy at the time of nephrectomy was not significantly associated with the risks of locally advanced (pT3–4) pathologic tumor stage (odds ratio ¼ 0.96; P ¼ 0.80) or high (3–4) tumor grade (odds ratio ¼ 1.11; P ¼ 0.30). Median postoperative follow-up was 7.8 years. Compared with patients not on statin therapy, patients taking statins at surgery had similar 10-year progression-free survival (80% vs. 79%; P ¼ 0.56), cancer-specific survival (85% vs. 84%; P ¼ 0.71), and overall survival (59% vs. 64%; P ¼ 0.11). On multivariable analysis, statin use was not significantly associated with the risks of disease progression (hazard ratio [HR] ¼ 1.22; P ¼ 0.10), death from RCC (HR ¼ 1.02; P ¼ 0.90), or all-cause mortality (HR ¼ 0.84; P ¼ 0.05). Conclusions: We found no independent association between preoperative statin therapy and oncologic outcomes among patients with surgically treated localized RCC. Our data thus do not support an anticancer role for statin therapy in this setting. r 2015 Elsevier Inc. All rights reserved.
Keywords: Statin; HGM-CoA reductase inhibitor; Renal cell carcinoma; localized; Nephrectomy
1. Introduction Statins, which inhibit 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, have been used extensively Take home message: Despite the mechanistic data supporting antitumor activity of HMG-CoA reductase inhibitors in renal cell carcinoma (RCC), we found no independent association between statin use and cancer progression, RCC-specific, or all-cause mortality among patients with localized RCC undergoing nephrectomy. * Corresponding author. Tel.: þ1-507-284-4015; fax: þ1-507-284-4951. E-mail address: [email protected] (S.A. Boorjian). http://dx.doi.org/10.1016/j.urolonc.2015.01.009 1078-1439/r 2015 Elsevier Inc. All rights reserved.
for the treatment of hypercholesterolemia and prevention of cardiovascular events since their approval in 1987 [1]. Evidence suggests that statins contain antineoplastic properties as well, through the inhibition of cellular proliferation, induction of apoptosis, and reduction in angiogenesis [2,3]. Specifically, HMG-CoA reductase inhibitors halt the conversion of HMG-CoA to mevalonate, a critical component in the maintenance of cellular membrane integrity, signaling, protein synthesis, and cell cycle progression [2,4]. As mutant p53 transcriptionally activated mevalonate pathway genes play a vital role in tumorigenesis [5], a disruption of
2
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
these uncontrolled cellular processes may thereby inhibit cancer growth and progression [6]. Indeed, anticancer properties of statins have been demonstrated among multiple solid malignancies, including prostate [7–10], colorectal [11,12], and breast cancers [13,14]. For renal cell carcinoma (RCC), in vitro and in vivo evidence suggests antitumor effects of statins through the modulation of multiple cellular processes, including the mTOR and JAK2/STAT3 pathways [15–17]. However, despite such mechanistic data supporting activity of HMG-CoA reductase inhibitors in RCC, the clinical utility of such therapy remains unclear. Epidemiological studies assessing primary prevention have demonstrated variable results [18–20]. Meanwhile, there remains paucity of data regarding the potential therapeutic implications for statin use among patients undergoing treatment for RCC [21–23]. Herein, then, we evaluated the association of statin use with clinicopathologic outcomes and survival among patients with localized RCC undergoing nephrectomy.
2. Materials and methods Following Institutional Review Board approval, we reviewed the Mayo Clinic Renal Tumor Registry to identify 2,357 patients treated with radical or partial nephrectomy for sporadic, unilateral, and localized (pNx/0, M0) RCC at our institution between 1995 and 2009. Clinical variables recorded included age, sex, year of surgery, type of surgery, symptoms at presentation, smoking status, Eastern Cooperative Oncology Group (ECOG) performance status, Charlson comorbidity index, and body mass index (BMI). The medical record was reviewed to identify patients with documented statin use within 3 months before surgery. Patients with a palpable flank or abdominal mass, discomfort, gross hematuria, acute-onset varicocele, or constitutional symptoms including rash, sweats, weight loss, fatigue, early satiety, and anorexia were considered symptomatic. Pathologic features studied included histologic subtype, tumor size, 2010 primary tumor classification, nuclear grade, coagulative tumor necrosis, and sarcomatoid differentiation. A genitourinary pathologist (J.C.C.) reviewed the microscopic slides from all specimens without the knowledge of patient outcome. Follow-up after nephrectomy was generally quarterly for the first 2 years, semiannually for the next 2 years, and annually thereafter for patients without evidence of recurrent disease. For survival end points, vital status was identified from death certificates or physician correspondence. For patients followed elsewhere, the nephrectomy registry monitors outcomes annually by correspondence with the patient and the local treating physician. Comparisons of clinicopathologic features between patients with and without statin use were performed using Wilcoxon rank sum and chi-square tests, as appropriate. Associations with advanced stage (pT3 or pT4) and
high-grade (3 or 4) tumors at nephrectomy were evaluated using logistic regression models. Progression-free survival, cancer-specific survival (CSS), and overall survival (OS) were estimated as the time from nephrectomy to event or last follow-up using the Kaplan-Meier method. Disease progression was defined as distant metastases or death due to RCC. Cox proportional hazards regression models were used to evaluate the association of statin use with outcomes. Moreover, associations with time to disease progression and death due to RCC after accounting for the competing risk of death without disease progression or death due to non–RCC-related causes were evaluated using a proportional subdistribution model [24]. In all logistic and proportional hazards regression models, year of surgery was used as a stratification effect to account for changes in clinicopathologic features and patient outcome over time. Finally, to evaluate the findings of Hamilton et al. [21] in our own data set, an identical multivariable model to the one from that study was tested, including age at surgery, sex, race (African American vs. other), type of surgery, Charlson comorbidity index, primary tumor classification, preoperative estimated glomerular filtration rate (calculated using the CKD-EPI equation), symptoms, and year of surgery (analyzed as a stratification effect). P o 0.05 was considered statistically significant. Statistical analyses were performed using the SAS software package (SAS Institute, Cary, NC). 3. Results Of the 2,357 patients who underwent radical or partial nephrectomy at our institution between 1995 and 2009, 630 (27%) had documentation of statin use within 3 months of surgery (Table 1). Clinicopathologic features for patients with and without statin use are provided in Table 2. As can be seen, patients treated with statin therapy were older (median age 66 vs. 61; P o 0.001), more likely to be male (71% vs. 65%; P ¼ 0.01), have a greater Charlson index (median ¼ 1 vs. 0; P o 0.001), worse ECOG performance score (Z1, 15% vs. 12%; P ¼ 0.05), local or constitutional symptoms, greater BMI (median ¼ 30 vs. 28; P o 0.001), incidence of obesity (BMI Z 30, 49% vs. 39%; Table 1 Statin medication use at time of nephrectomy Medication
n (%)a
Atorvastatin Simvastatin Lovastatin Pravastatin Rosuvastatin Fluvastatin Cerivastatin
329 259 38 37 24 17 2
(52) (41) (6) (6) (4) (3) (o1)
a n ¼ 558 On 1 statin within 3 months of surgery, n ¼ 68 on 2 statins, and n ¼ 4 on 3 statins.
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
3
Table 2 Clinicopathologic characteristics of cohort Feature
No statin (n ¼ 1,727)
Statin (n ¼ 630)
Total (n ¼ 2,357)
P value
Median (IQR) Age at surgery, y Charlson score (n ¼ 2,355) BMI (n ¼ 2,177) Tumor size, (n ¼ 2,351)
61 0 28 4.5
(52–70) (0–2) (25–33) (2.7–7.5)
66 1 30 4.1
(59–73) (0–3) (27–34) (2.7–6.5)
63 1 29 4.5
(54–71) (0–2) (26–33) (2.7–7.0)
o0.001 o0.001 o0.001 0.01
n (%) Male sex
1,130 (65)
448 (71)
1,578 (67)
Type of surgery Partial Radical
800 (46) 927 (54)
340 (54) 290 (46)
1,140 (48) 1,217 (52)
Symptoms (n ¼ 2,354)
815 (47)
214 (34)
1,029 (44)
o0.001
Constitutional symptoms (n ¼ 2,354)
250 (14)
60 (10)
310 (13)
0.002
Smoking status (n ¼ 2,320) Current Former
298 (18) 662 (39)
94 (15) 294 (47)
392 (17) 956 (41)
1,523 (88) 203 (12)
535 (85) 94 (15)
2,058 (87) 297 (13)
622 (39)
292 (49)
914 (42)
ECOG performance status (n ¼ 2,355) 0 Z1 BMI (n ¼ 2,177) Z30 RCC histologic subtype Clear cell Papillary Chromophobe Clear cell papillary Mucinous and spindle cell Translocation-associated Collecting duct NOS
0.001
0.001
0.05
o0.001 0.20 1,304 282 111 8 2 2 2 16
(76) (16) (6) (o1) (o1) (o1) (o1) (1)
464 (74) 123 (20) 30 (5) 7 (1) 1 (o1) 1 (o1) 0 4 (1)
1,768 405 141 15 3 3 2 20
(75) (17) (6) (1) (o1) (o1) (o1) (1)
2010 Primary tumor classification (n ¼ 2,343) pT1a pT1b pT2a pT2b pT3a pT3b pT3c pT4
758 408 146 84 228 71 10 9
(44) (24) (9) (5) (13) (4) (1) (1)
304 152 40 23 91 12 2 5
(48) (24) (6) (4) (15) (2) (o1) (1)
1,062 560 186 107 319 83 12 14
(45) (24) (8) (5) (914) (4) (1) (1)
Nuclear grade 1 2 3 4 Coagulative tumor necrosis (n ¼ 2,356)
148 868 611 100 405
(9) (50) (35) (6) (23)
34 312 257 27 118
(5) (50) (41) (4) (19)
182 1,180 868 127 523
(8) (50) (37) (5) (22)
Sarcomatoid differentiation (n ¼ 2,356)
0.01
0.04
0.06
28 (2)
6 (1)
34 (1)
0.01 0.20
NOS ¼ not otherwise specified.
P o 0.001), history of smoking (63% vs. 56%; P ¼ 0.001), and undergo partial nephrectomy (54% vs. 46%; P o 0.001). Pathologically, patients with statin use had a smaller tumor size (median ¼ 4.1 vs. 4.5 cm; P ¼ 0.01) and were less likely to have coagulative tumor necrosis (19% vs. 23%; P ¼ 0.01). In addition, on multivariable
analysis, controlling for other preoperative variables associated with pathologic stage (Table 3), we found that statin therapy at the time of nephrectomy was not significantly associated with the risk of locally advanced (pT3–4) pathologic tumor stage (odds ratio ¼ 0.96; P ¼ 0.80) or high tumor grade (odds ratio ¼ 1.11; P ¼ 0.30).
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
4
Table 3 Multivariable analysis of factors associated with advanced pathologic stage and greater tumor grade at nephrectomy Advanced pathologic stagea
Feature
OR (95% CI) Age at surgery Sex (ref. female) Constitutional symptoms Smoking status (ref. never) Current Former ECOG performance status (ref. ¼ 0) Charlson score BMI (ref. o30) Statins
High tumor gradeb P value
OR (95% CI)
P value
c
c
1.38 (1.24–1.55) 1.43 (1.10–1.87) 5.35 (4.03–7.10)
o0.001 0.008 o0.001
1.28 (1.18–1.40) 1.78 (1.46–2.17) 3.01 (2.29–3.95)
o0.001 o0.001 o0.001
1.27 1.18 1.91 0.85 1.55 0.96
(0.90–1.80) (0.91–1.54) (1.36–2.69) (0.78–0.92)d (1.21–1.97) (0.73–1.27)
0.20 0.20 o0.001 o0.001 o0.001 0.80
1.31 1.19 1.41 0.91 0.98 1.11
(1.00–1.70) (0.97–1.46) (1.06–1.88) (0.86–0.97)d (0.81–1.18) (0.90–1.37)
0.05 0.09 0.02 0.002 0.80 0.30
Pathologic tumor stage ZpT3. Tumor grade Z3. c Odds ratio and CI represent a 10-year increase in age. d Odds ratio and CI represent a 1-unit increase in Charlson score. a
b
Median duration of follow-up after surgery was 7.8 years (interquartile range [IQR]: 5.3–11.2) among patients still alive. During that time, 433 patients experienced disease progression, at a median 1.7 years following nephrectomy (IQR: 0.5–4.0). Furthermore, 811 died (median ¼ 4.5 y; IQR: 2.1–8.0), of whom 298 died of RCC (median ¼ 2.9 y; IQR: 1.5–6.1). We found that patients taking statins at surgery had a similar 10-year progression-free (80% vs. 79%; P ¼ 0.56) (Fig. 1), CSS (85% vs. 84%; P ¼ 0.71) (Fig. 2), and OS (59% vs. 64%; P ¼ 0.11) (Fig. 3) compared with patients not taking statins at nephrectomy. Next, we assessed the association between statin exposure and disease progression, cancer-specific mortality, non–RCCrelated mortality, and all-cause mortality on multivariable analysis, controlling for clinical variables including age, sex, smoking, ECOG performance status, Charlson comorbidity index, BMI, and pathologic features identified at the time of
nephrectomy (Table 4). We found here that statin use was not significantly associated with the risks of disease progression (hazard ratio [HR] ¼ 1.22; P ¼ 0.10), death due to RCC (HR ¼ 1.02; P ¼ 0.90), or all-cause mortality (HR ¼ 0.84; P ¼ 0.05). Further, even after accounting for the competing risks of death without disease progression and death from non–RCC-related causes, statin use was not independently associated with progression (HR ¼ 1.23, 95% CI: 0.96–1.58; P ¼ 0.10) or RCC-specific mortality (HR ¼ 0.98, 95% CI: 0.71–1.36; P ¼ 0.90). However, and importantly, the intended effect of statin usage remained protective against death from non–RCC-related causes (HR ¼ 0.70, 95% CI: 0.55-0.89; P ¼ 0.004). Finally, in an effort to validate the findings of Hamilton et al. [21], we performed an additional multivariable analysis controlling for age, sex, race, type of surgery, Charlson comorbidity index, pathologic tumor stage, preoperative
Fig. 1. Progression-free survival following nephrectomy for pNx/0, M0 RCC, stratified by statin exposure.
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
5
Fig. 2. Cancer-specific survival following nephrectomy for pNx/0, M0 RCC, stratified by statin exposure.
estimated glomerular filtration rate, patient symptomology, and year of surgery. We again found that statin exposure was not independently associated with cancer progression (HR ¼ 1.23, 95% CI: 0.97–1.56; P ¼ 0.09), death due to RCC (HR ¼ 1.06, 95% CI: 0.78–1.43; P ¼ 0.70), or all-cause mortality (HR ¼ 0.88, 95% CI: 0.74–1.05; P ¼ 0.20).
4. Discussion We found here, in a large cohort of patients with surgically treated RCC with long-term follow-up, that statin use at the time of nephrectomy was not associated with the risks of subsequent cancer progression, RCC-specific mortality, or allcause mortality. Likewise, we found no independent
association of this medication with advanced pathologic stage or greater tumor grade. These findings remained after adjusting for competing risks of progression or mortality. To our knowledge, only 3 prior series to date have assessed the therapeutic role of statin medications in patients with RCC [21–23]. Among these, only the study by Hamilton et al. [21] focused specifically on patients with localized (pNx/0, M0) disease, and therein reported a favorable correlation between statin exposure and outcomes in a population undergoing nephrectomy. On the contrary, as the data presented here do not demonstrate a significant association between preoperative statin use and oncologic outcomes among patients with surgically treated pNx/0, M0 RCC, our findings do not support a current anticancer role for statin therapy among patients with localized RCC undergoing nephrectomy.
Fig. 3. Overall survival following nephrectomy for pNx/0, M0 RCC, stratified by statin exposure.
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
6
Table 4 Multivariable analysis of factors associated with disease progression, death due to RCC, and all-cause mortality following nephrectomy Feature
Disease progression HR (95% CI)
Age at surgery Sex (ref. female) Type of surgery (ref. partial) Constitutional symptoms Smoking status (ref. never) Current Former ECOG performance status (ref. ¼ 0) Charlson score BMI (ref. BMI o30) Tumor size RCC histologic subtype (ref. other) Clear cell, collecting duct, NOS 2010 Primary tumor classification (ref. ¼ pT1a) pT1b pT2a pT2b pT3a pT3b pT3c/ pT4 Nuclear grade (ref. ¼ grade 1 or 2) 3 4 Coagulative tumor necrosis Sarcomatoid differentiation Statins
Death due to RCC P value
HR (95% CI)
All-cause mortality P value
o0.001 0.80 0.10 0.02
0.80 0.10 o0.001 0.03 0.60 0.001
1.19 1.16 2.30 1.25 1.04 1.08
(0.94–1.50) (0.98–1.37) (1.90–2.79) (1.20–1.31)b (0.89–1.22) (1.05–1.12)c
0.10 0.09 o0.001 o0.001 0.60 o0.001
o0.001
1.42 (1.16–1.75)
0.40 0.40 0.40 0.05
1.08 0.72 1.04 1.41
(0.95–1.22) (0.53–0.97) (0.66–1.63) (1.05–1.90)
0.30 0.03 0.90 0.02
0.94 1.13 1.24 0.99 1.09 1.04
(0.69–1.28) (0.90–1.43) (0.91–1.69) (0.92–1.07)b (0.88–1.36) (1.00–1.07)c
0.70 0.20 0.20 0.90 0.40 0.04
1.06 1.26 1.85 1.10 1.08 1.07
(0.73–1.55) (0.95–1.68) (1.30–2.64) (1.01–1.21)b (0.83–1.41) (1.03–1.12)c
2.15 (1.38–3.36)
P value
(1.39–1.63) (0.86–1.22) (0.97–1.45) (1.05–1.59)
(0.87–1.06) (0.71–1.14) (0.83–1.69) (1.01–1.65)
o0.001
HR (95% CI) 1.50 1.03 1.18 1.29
0.96 0.90 1.19 1.29
2.32 (1.61–3.35)
a
a
a
o0.001
3.88 5.35 7.75 10.30 16.47 19.14
(2.38–6.32) (3.03–9.44) (4.03–14.92) (6.05–17.54) (8.97–30.24) (9.66–37.92)
o0.001 o0.001 o0.001 o0.001 o0.001 o0.001
3.19 2.94 4.68 7.87 12.25 11.67
(1.72–5.92) (1.41–6.10) (2.04–10.71) (4.05–15.28) (5.89–25.46) (4.99–27.29)
o0.001 0.004 o0.001 o0.001 o0.001 o0.001
1.02 0.61 0.92 1.28 1.73 1.59
(0.82–1.28) (0.42–0.88) (0.58–1.48) (0.94–1.75) (1.14–2.61) (0.91–2.79)
0.90 0.009 0.70 0.10 0.010 0.10
2.09 3.05 1.69 2.54 1.22
(1.56–2.81) (1.97–4.74) (1.32–2.17) (1.46–4.42) (0.95–1.57)
o0.001 o0.001 o0.001 0.001 0.10
2.72 4.30 1.92 1.68 1.02
(1.82–4.05) (2.48–7.45) (1.43–2.59) (0.88–3.22) (0.74–1.39)
o0.001 o0.001 o0.001 0.10 0.90
1.26 2.28 1.45 1.08 0.84
(1.05–1.51) (1.62–3.21) (1.19–1.75) (0.63–1.85) (0.69–1.00)
0.01 o0.001 o0.001 0.80 0.05
a
Hazard ratio and CI represent a 10-year increase in age. Hazard ratio and CI represent a 1-unit increase in Charlson score. c Hazard ratio and CI represent a 1-cm increase in tumor size. b
The association between statin use and cancer-specific outcomes is complex and yet to be fully elucidated. Statins inhibit cholesterol biosynthesis, which remains a vital component of cellular membrane integrity and multiple downstream stream cellular processes [2]. In particular, the mevalonate pathway, a byproduct of cholesterol synthesis directly inhibited by HMG-CoA reductase inhibitors, plays a critical role in the development and progression of multiple malignancies through the promotion of invasion, 3-dimensional growth, and cell survival [5]. In fact, Freed-Pastor et al. [14] demonstrated that a mutant p53 suppressor gene up-regulates the expression of multiple mevalonate pathway genes, leading to the maintenance and progression of breast cancer cell lines. Likewise, the in vitro inhibition of mevalonate production within prostate cancer cells by lipophilic HMG-CoA reductase inhibitors led to a downstream reduction in isoprenoid GGPP, an activator of RhoA activity, culminating in attenuated migration/invasion of prostate cancer cell lines [25]. Accordingly, large epidemiological series have validated these laboratory findings with statin therapy demonstrating a 17% to 33% reduction in recurrence among breast cancer survivors [13] and a 45% reduction in prostate cancer mortality [7]. The role of statins in RCC pathophysiology, meanwhile, remains to be established. That is, research by Fang et al.
[15] demonstrated proapoptotic and metastasis-inhibitory effects of statins on RCC cell lines both in vitro and in vivo through the attenuation of the AKT/mTOR, ERK, and JAK2/STAT3 pathways. Likewise, in a dose-dependent manner, fluvastatin has been shown to induce apoptosis and reduce proliferation of RCC cell lines in vitro via direct suppression of Akt phosphorylation/activation, resulting in inhibition of the mTOR cascade [16]. Moreover, in vitro evidence suggest that statins may potentiate the cytotoxic activity of sorafenib through G1 phase cell cycle arrest [17], further supporting the potential for a therapeutic role of HMG-CoA reductase inhibitors among patients with RCC. However, epidemiological studies assessing the influence of statin exposure on the incidence of RCC have met with conflicting results. In a case-control analysis of 500,000 veterans, statin use was associated with a 48% risk reduction in the incidence of RCC [18], whereas in a prospective analysis of the general population, statins only demonstrated a risk reduction among women [19]. At the same time, others have found no significant association between statin usage and RCC incidence [20,26,27]. Regarding a role for statins among patients with surgically treated RCC, to our knowledge, the largest prior series reported was by Hamilton et al. [21]. In that study, the investigators retrospectively reviewed 2,608 surgically
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
treated patients with localized RCC [21] and noted a 33% risk reduction in cancer progression and improved OS when modeling for statin use as a time-dependent covariate [21]. However, limitations of that study include a short duration of follow-up and a lack of adjustment for histologic features noted at the time of nephrectomy such as tumor grade, sarcomatoid differentiation, and necrosis. We sought here to investigate the potential association of statin therapy with cancer outcomes after nephrectomy in a large cohort of patients with extended follow-up. Interestingly, unlike Hamilton et al. [21], we found no independent association between statin exposure and disease progression, CSS, or OS. Indeed, even when accounting for the competing risks of mortality, statin use was not independently associated with progression or RCC-specific death. Discrepancies in these findings may be the result of the extended follow-up in our series, as well as our incorporation of additional adverse pathologic features that have been associated with adverse oncologic outcomes [28]. We recognize that our study is limited by its retrospective, nonrandomized nature. Specifically, there was significant heterogeneity between cohorts, as those receiving statin therapy were older, with a worse functional status, greater obesity, and more likely to have a smoking history. In addition, we must acknowledge that we were not able here to accurately define the duration of statin exposure, dosage, or patient compliance, which may underlie the noted in vitro [16] relationship between statin use and outcomes, nor could we evaluate the effect of statin use after surgery on subsequent survival. Importantly, however, our results should not undermine the intended effect of statin therapy; which was independently protective against death owing to non–RCC-related causes. As such, in light of ongoing controversy, future prospective analyses are warranted to elucidate the anticancer role of statin use among patients with surgically treated RCC.
5. Conclusions We found no significant association between statin use and outcomes among patients with localized RCC undergoing nephrectomy. Specifically, we found no independent association with cancer progression, RCC-specific mortality, or all-cause mortality. These data suggest a limited anticancer role for statins in this setting. References [1] Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7–22. [2] Chan KK, Oza AM, Siu LL. The statins as anticancer agents. Clin Cancer Res 2003;9:10–9.
7
[3] Gauthaman K, Fong CY, Statins Bongso A. stem cells, and cancer. J Cell Biochem 2009;106:975–83. [4] Thurnher M, Nussbaumer O, Gruenbacher G. Novel aspects of mevalonate pathway inhibitors as antitumor agents. Clin Cancer Res 2012;18:3524–31. [5] Gruenbacher G, Thurnher M. Mevalonate metabolism in cancer. Cancer Lett 2015;356:192–6. [6] Demierre MF, Higgins PD, Gruber SB, Hawk E, Lippman SM. Statins and cancer prevention. Nat Rev Cancer 2005;5:930–42. [7] Yu O, Eberg M, Benayoun S, Aprikian A, Batist G, Suissa S, et al. Use of statins and the risk of death in patients with prostate cancer. J Clin Oncol 2014;32:5–11. [8] Banez LL, Klink JC, Jayachandran J, Lark AL, Gerber L, Hamilton RJ, et al. Association between statins and prostate tumor inflammatory infiltrate in men undergoing radical prostatectomy. Cancer Epidemiol Biomarkers Prev 2010;19:722–8. [9] Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancer-related mortality. N Engl J Med 2012;367:1792–802. [10] Wang C, Tao W, Wang Y, Bikow J, Lu B, Keating A, et al. Rosuvastatin, identified from a zebrafish chemical genetic screen for antiangiogenic compounds, suppresses the growth of prostate cancer. Eur Urol 2010;58:418–26. [11] Cardwell CR, Hicks BM, Hughes C, Murray LJ. Statin use after colorectal cancer diagnosis and survival: a population-based cohort study. J Clin Oncol 2014;32:3177–83. [12] Cho SJ, Kim JS, Kim JM, Lee JY, Jung HC, Song IS. Simvastatin induces apoptosis in human colon cancer cells and in tumor xenografts, and attenuates colitis-associated colon cancer in mice. Int J Cancer 2008;123:951–7. [13] Ahern TP, Lash TL, Damkier P, Christiansen PM, Cronin-Fenton DP. Statins and breast cancer prognosis: evidence and opportunities. Lancet Oncol 2014;15:e461–8. [14] Freed-Pastor WA, Mizuno H, Zhao X, Langerod A, Moon SH, Rodriguez-Barrueco R, et al. Mutant p53 disrupts mammary tissue architecture via the mevalonate pathway. Cell 2012;148:244–58. [15] Fang Z, Tang Y, Fang J, Zhou Z, Xing Z, Guo Z, et al. Simvastatin inhibits renal cancer cell growth and metastasis via AKT/mTOR, ERK and JAK2/STAT3 pathway. PLoS One 2013;8:e62823. [16] Woodard J, Sassano A, Hay N, Platanias LC. Statin-dependent suppression of the Akt/mammalian target of rapamycin signaling cascade and programmed cell death 4 up-regulation in renal cell carcinoma. Clin Cancer Res 2008;14:4640–9. [17] Bil J, Zapala L, Nowis D, Jakobisiak M, Golab J. Statins potentiate cytostatic/cytotoxic activity of sorafenib but not sunitinib against tumor cell lines in vitro. Cancer Lett 2010;288:57–67. [18] Khurana V, Caldito G, Ankem M. Statins might reduce risk of renal cell carcinoma in humans: case-control study of 500,000 veterans. Urology 2008;71:118–22. [19] Liu W, Choueiri TK, Cho E. Statin use and the risk of renal cell carcinoma in 2 prospective US cohorts. Cancer 2012;118:797–803. [20] Chiu HF, Kuo CC, Kuo HW, Lee IM, Lee CT, Yang CY. Statin use and the risk of kidney cancer: a population-based case-control study. Expert Opin Drug Saf 2012;11:543–9. [21] Hamilton RJ, Morilla D, Cabrera F, Leapman M, Chen LY, Bernstein M, et al. The association between statin medication and progression after surgery for localized renal cell carcinoma. J Urol 2014;191: 914–9. [22] Choi SK, Min GE, Jeon SH, Lee HL, Chang SG, Yoo KH. Effects of statins on the prognosis of local and locally advanced renal cell carcinoma following nephrectomy. Mol Clin Oncol 2013;1: 365–8. [23] Manoukian GE, Tannir NM, Jonasch E, Qiao W, Haygood TM, Tu SM. Pilot trial of bone-targeted therapy combining zoledronate with fluvastatin or atorvastatin for patients with metastatic renal cell carcinoma. Clin Genitourin Cancer 2011;9:81–8.
8
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
[24] Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 1999;94:494–509. [25] Brown M, Hart C, Tawadros T, Ramani V, Sangar V, Lau M, et al. The differential effects of statins on the metastatic behaviour of prostate cancer. Br J Cancer 2012;106:1689–96. [26] Coogan PF, Rosenberg L, Strom BL. Statin use and the risk of 10 cancers. Epidemiology 2007;18:213–9.
[27] Kuoppala J, Lamminpaa A, Pukkala E. Statins and cancer: a systematic review and meta-analysis. Eur J Cancer 2008;44:2122–32. [28] Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL, Zincke H. An outcome prediction model for patients with clear cell renal cell carcinoma treated with radical nephrectomy based on tumor stage, size, grade and necrosis: the SSIGN score. J Urol 2002;168: 2395–400.
Original article
The association of statin therapy with clinicopathologic outcomes and survival among patients with localized renal cell carcinoma undergoing nephrectomy Boyd R. Viers, M.D.a, R. Houston Thompson, M.D.a, Sarah P. Psutka, M.D.a, Christine M. Lohse, M.S.b, John C. Cheville, M.D.c, Bradley C. Leibovich, M.D.a, Matthew K. Tollefson, M.D.a, Stephen A. Boorjian, M.D.a,* b
a Department of Urology, Mayo Clinic, Rochester, MN Division of Biomedical Statistics and Informatics, Mayo Clinic, Rochester, MN c Department of Anatomic Pathology, Mayo Clinic, Rochester, MN
Received 25 November 2014; received in revised form 14 January 2015; accepted 14 January 2015
Abstract Objectives: Although statins have been found to induce apoptosis and demonstrate antimetastases activity both in vitro and in vivo for renal cell carcinoma (RCC), clinical evidence of a role for these medications is limited. We evaluated the association of statin therapy with outcomes among patients with surgically treated localized RCC. Methods and materials: We reviewed 2,357 patients who underwent nephrectomy between 1995 and 2009 for pNx/0, M0 RCC. Of these, 630 (27%) were taking statins within 3 months of surgery. Progression-free survival, cancer-specific survival, and overall survival were estimated using the Kaplan-Meier method. The associations of statin use with clinicopathologic outcomes were evaluated with multivariable logistic and proportional hazards regression models. Results: Statin therapy at the time of nephrectomy was not significantly associated with the risks of locally advanced (pT3–4) pathologic tumor stage (odds ratio ¼ 0.96; P ¼ 0.80) or high (3–4) tumor grade (odds ratio ¼ 1.11; P ¼ 0.30). Median postoperative follow-up was 7.8 years. Compared with patients not on statin therapy, patients taking statins at surgery had similar 10-year progression-free survival (80% vs. 79%; P ¼ 0.56), cancer-specific survival (85% vs. 84%; P ¼ 0.71), and overall survival (59% vs. 64%; P ¼ 0.11). On multivariable analysis, statin use was not significantly associated with the risks of disease progression (hazard ratio [HR] ¼ 1.22; P ¼ 0.10), death from RCC (HR ¼ 1.02; P ¼ 0.90), or all-cause mortality (HR ¼ 0.84; P ¼ 0.05). Conclusions: We found no independent association between preoperative statin therapy and oncologic outcomes among patients with surgically treated localized RCC. Our data thus do not support an anticancer role for statin therapy in this setting. r 2015 Elsevier Inc. All rights reserved.
Keywords: Statin; HGM-CoA reductase inhibitor; Renal cell carcinoma; localized; Nephrectomy
1. Introduction Statins, which inhibit 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) reductase, have been used extensively Take home message: Despite the mechanistic data supporting antitumor activity of HMG-CoA reductase inhibitors in renal cell carcinoma (RCC), we found no independent association between statin use and cancer progression, RCC-specific, or all-cause mortality among patients with localized RCC undergoing nephrectomy. * Corresponding author. Tel.: þ1-507-284-4015; fax: þ1-507-284-4951. E-mail address: [email protected] (S.A. Boorjian). http://dx.doi.org/10.1016/j.urolonc.2015.01.009 1078-1439/r 2015 Elsevier Inc. All rights reserved.
for the treatment of hypercholesterolemia and prevention of cardiovascular events since their approval in 1987 [1]. Evidence suggests that statins contain antineoplastic properties as well, through the inhibition of cellular proliferation, induction of apoptosis, and reduction in angiogenesis [2,3]. Specifically, HMG-CoA reductase inhibitors halt the conversion of HMG-CoA to mevalonate, a critical component in the maintenance of cellular membrane integrity, signaling, protein synthesis, and cell cycle progression [2,4]. As mutant p53 transcriptionally activated mevalonate pathway genes play a vital role in tumorigenesis [5], a disruption of
2
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
these uncontrolled cellular processes may thereby inhibit cancer growth and progression [6]. Indeed, anticancer properties of statins have been demonstrated among multiple solid malignancies, including prostate [7–10], colorectal [11,12], and breast cancers [13,14]. For renal cell carcinoma (RCC), in vitro and in vivo evidence suggests antitumor effects of statins through the modulation of multiple cellular processes, including the mTOR and JAK2/STAT3 pathways [15–17]. However, despite such mechanistic data supporting activity of HMG-CoA reductase inhibitors in RCC, the clinical utility of such therapy remains unclear. Epidemiological studies assessing primary prevention have demonstrated variable results [18–20]. Meanwhile, there remains paucity of data regarding the potential therapeutic implications for statin use among patients undergoing treatment for RCC [21–23]. Herein, then, we evaluated the association of statin use with clinicopathologic outcomes and survival among patients with localized RCC undergoing nephrectomy.
2. Materials and methods Following Institutional Review Board approval, we reviewed the Mayo Clinic Renal Tumor Registry to identify 2,357 patients treated with radical or partial nephrectomy for sporadic, unilateral, and localized (pNx/0, M0) RCC at our institution between 1995 and 2009. Clinical variables recorded included age, sex, year of surgery, type of surgery, symptoms at presentation, smoking status, Eastern Cooperative Oncology Group (ECOG) performance status, Charlson comorbidity index, and body mass index (BMI). The medical record was reviewed to identify patients with documented statin use within 3 months before surgery. Patients with a palpable flank or abdominal mass, discomfort, gross hematuria, acute-onset varicocele, or constitutional symptoms including rash, sweats, weight loss, fatigue, early satiety, and anorexia were considered symptomatic. Pathologic features studied included histologic subtype, tumor size, 2010 primary tumor classification, nuclear grade, coagulative tumor necrosis, and sarcomatoid differentiation. A genitourinary pathologist (J.C.C.) reviewed the microscopic slides from all specimens without the knowledge of patient outcome. Follow-up after nephrectomy was generally quarterly for the first 2 years, semiannually for the next 2 years, and annually thereafter for patients without evidence of recurrent disease. For survival end points, vital status was identified from death certificates or physician correspondence. For patients followed elsewhere, the nephrectomy registry monitors outcomes annually by correspondence with the patient and the local treating physician. Comparisons of clinicopathologic features between patients with and without statin use were performed using Wilcoxon rank sum and chi-square tests, as appropriate. Associations with advanced stage (pT3 or pT4) and
high-grade (3 or 4) tumors at nephrectomy were evaluated using logistic regression models. Progression-free survival, cancer-specific survival (CSS), and overall survival (OS) were estimated as the time from nephrectomy to event or last follow-up using the Kaplan-Meier method. Disease progression was defined as distant metastases or death due to RCC. Cox proportional hazards regression models were used to evaluate the association of statin use with outcomes. Moreover, associations with time to disease progression and death due to RCC after accounting for the competing risk of death without disease progression or death due to non–RCC-related causes were evaluated using a proportional subdistribution model [24]. In all logistic and proportional hazards regression models, year of surgery was used as a stratification effect to account for changes in clinicopathologic features and patient outcome over time. Finally, to evaluate the findings of Hamilton et al. [21] in our own data set, an identical multivariable model to the one from that study was tested, including age at surgery, sex, race (African American vs. other), type of surgery, Charlson comorbidity index, primary tumor classification, preoperative estimated glomerular filtration rate (calculated using the CKD-EPI equation), symptoms, and year of surgery (analyzed as a stratification effect). P o 0.05 was considered statistically significant. Statistical analyses were performed using the SAS software package (SAS Institute, Cary, NC). 3. Results Of the 2,357 patients who underwent radical or partial nephrectomy at our institution between 1995 and 2009, 630 (27%) had documentation of statin use within 3 months of surgery (Table 1). Clinicopathologic features for patients with and without statin use are provided in Table 2. As can be seen, patients treated with statin therapy were older (median age 66 vs. 61; P o 0.001), more likely to be male (71% vs. 65%; P ¼ 0.01), have a greater Charlson index (median ¼ 1 vs. 0; P o 0.001), worse ECOG performance score (Z1, 15% vs. 12%; P ¼ 0.05), local or constitutional symptoms, greater BMI (median ¼ 30 vs. 28; P o 0.001), incidence of obesity (BMI Z 30, 49% vs. 39%; Table 1 Statin medication use at time of nephrectomy Medication
n (%)a
Atorvastatin Simvastatin Lovastatin Pravastatin Rosuvastatin Fluvastatin Cerivastatin
329 259 38 37 24 17 2
(52) (41) (6) (6) (4) (3) (o1)
a n ¼ 558 On 1 statin within 3 months of surgery, n ¼ 68 on 2 statins, and n ¼ 4 on 3 statins.
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
3
Table 2 Clinicopathologic characteristics of cohort Feature
No statin (n ¼ 1,727)
Statin (n ¼ 630)
Total (n ¼ 2,357)
P value
Median (IQR) Age at surgery, y Charlson score (n ¼ 2,355) BMI (n ¼ 2,177) Tumor size, (n ¼ 2,351)
61 0 28 4.5
(52–70) (0–2) (25–33) (2.7–7.5)
66 1 30 4.1
(59–73) (0–3) (27–34) (2.7–6.5)
63 1 29 4.5
(54–71) (0–2) (26–33) (2.7–7.0)
o0.001 o0.001 o0.001 0.01
n (%) Male sex
1,130 (65)
448 (71)
1,578 (67)
Type of surgery Partial Radical
800 (46) 927 (54)
340 (54) 290 (46)
1,140 (48) 1,217 (52)
Symptoms (n ¼ 2,354)
815 (47)
214 (34)
1,029 (44)
o0.001
Constitutional symptoms (n ¼ 2,354)
250 (14)
60 (10)
310 (13)
0.002
Smoking status (n ¼ 2,320) Current Former
298 (18) 662 (39)
94 (15) 294 (47)
392 (17) 956 (41)
1,523 (88) 203 (12)
535 (85) 94 (15)
2,058 (87) 297 (13)
622 (39)
292 (49)
914 (42)
ECOG performance status (n ¼ 2,355) 0 Z1 BMI (n ¼ 2,177) Z30 RCC histologic subtype Clear cell Papillary Chromophobe Clear cell papillary Mucinous and spindle cell Translocation-associated Collecting duct NOS
0.001
0.001
0.05
o0.001 0.20 1,304 282 111 8 2 2 2 16
(76) (16) (6) (o1) (o1) (o1) (o1) (1)
464 (74) 123 (20) 30 (5) 7 (1) 1 (o1) 1 (o1) 0 4 (1)
1,768 405 141 15 3 3 2 20
(75) (17) (6) (1) (o1) (o1) (o1) (1)
2010 Primary tumor classification (n ¼ 2,343) pT1a pT1b pT2a pT2b pT3a pT3b pT3c pT4
758 408 146 84 228 71 10 9
(44) (24) (9) (5) (13) (4) (1) (1)
304 152 40 23 91 12 2 5
(48) (24) (6) (4) (15) (2) (o1) (1)
1,062 560 186 107 319 83 12 14
(45) (24) (8) (5) (914) (4) (1) (1)
Nuclear grade 1 2 3 4 Coagulative tumor necrosis (n ¼ 2,356)
148 868 611 100 405
(9) (50) (35) (6) (23)
34 312 257 27 118
(5) (50) (41) (4) (19)
182 1,180 868 127 523
(8) (50) (37) (5) (22)
Sarcomatoid differentiation (n ¼ 2,356)
0.01
0.04
0.06
28 (2)
6 (1)
34 (1)
0.01 0.20
NOS ¼ not otherwise specified.
P o 0.001), history of smoking (63% vs. 56%; P ¼ 0.001), and undergo partial nephrectomy (54% vs. 46%; P o 0.001). Pathologically, patients with statin use had a smaller tumor size (median ¼ 4.1 vs. 4.5 cm; P ¼ 0.01) and were less likely to have coagulative tumor necrosis (19% vs. 23%; P ¼ 0.01). In addition, on multivariable
analysis, controlling for other preoperative variables associated with pathologic stage (Table 3), we found that statin therapy at the time of nephrectomy was not significantly associated with the risk of locally advanced (pT3–4) pathologic tumor stage (odds ratio ¼ 0.96; P ¼ 0.80) or high tumor grade (odds ratio ¼ 1.11; P ¼ 0.30).
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
4
Table 3 Multivariable analysis of factors associated with advanced pathologic stage and greater tumor grade at nephrectomy Advanced pathologic stagea
Feature
OR (95% CI) Age at surgery Sex (ref. female) Constitutional symptoms Smoking status (ref. never) Current Former ECOG performance status (ref. ¼ 0) Charlson score BMI (ref. o30) Statins
High tumor gradeb P value
OR (95% CI)
P value
c
c
1.38 (1.24–1.55) 1.43 (1.10–1.87) 5.35 (4.03–7.10)
o0.001 0.008 o0.001
1.28 (1.18–1.40) 1.78 (1.46–2.17) 3.01 (2.29–3.95)
o0.001 o0.001 o0.001
1.27 1.18 1.91 0.85 1.55 0.96
(0.90–1.80) (0.91–1.54) (1.36–2.69) (0.78–0.92)d (1.21–1.97) (0.73–1.27)
0.20 0.20 o0.001 o0.001 o0.001 0.80
1.31 1.19 1.41 0.91 0.98 1.11
(1.00–1.70) (0.97–1.46) (1.06–1.88) (0.86–0.97)d (0.81–1.18) (0.90–1.37)
0.05 0.09 0.02 0.002 0.80 0.30
Pathologic tumor stage ZpT3. Tumor grade Z3. c Odds ratio and CI represent a 10-year increase in age. d Odds ratio and CI represent a 1-unit increase in Charlson score. a
b
Median duration of follow-up after surgery was 7.8 years (interquartile range [IQR]: 5.3–11.2) among patients still alive. During that time, 433 patients experienced disease progression, at a median 1.7 years following nephrectomy (IQR: 0.5–4.0). Furthermore, 811 died (median ¼ 4.5 y; IQR: 2.1–8.0), of whom 298 died of RCC (median ¼ 2.9 y; IQR: 1.5–6.1). We found that patients taking statins at surgery had a similar 10-year progression-free (80% vs. 79%; P ¼ 0.56) (Fig. 1), CSS (85% vs. 84%; P ¼ 0.71) (Fig. 2), and OS (59% vs. 64%; P ¼ 0.11) (Fig. 3) compared with patients not taking statins at nephrectomy. Next, we assessed the association between statin exposure and disease progression, cancer-specific mortality, non–RCCrelated mortality, and all-cause mortality on multivariable analysis, controlling for clinical variables including age, sex, smoking, ECOG performance status, Charlson comorbidity index, BMI, and pathologic features identified at the time of
nephrectomy (Table 4). We found here that statin use was not significantly associated with the risks of disease progression (hazard ratio [HR] ¼ 1.22; P ¼ 0.10), death due to RCC (HR ¼ 1.02; P ¼ 0.90), or all-cause mortality (HR ¼ 0.84; P ¼ 0.05). Further, even after accounting for the competing risks of death without disease progression and death from non–RCC-related causes, statin use was not independently associated with progression (HR ¼ 1.23, 95% CI: 0.96–1.58; P ¼ 0.10) or RCC-specific mortality (HR ¼ 0.98, 95% CI: 0.71–1.36; P ¼ 0.90). However, and importantly, the intended effect of statin usage remained protective against death from non–RCC-related causes (HR ¼ 0.70, 95% CI: 0.55-0.89; P ¼ 0.004). Finally, in an effort to validate the findings of Hamilton et al. [21], we performed an additional multivariable analysis controlling for age, sex, race, type of surgery, Charlson comorbidity index, pathologic tumor stage, preoperative
Fig. 1. Progression-free survival following nephrectomy for pNx/0, M0 RCC, stratified by statin exposure.
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
5
Fig. 2. Cancer-specific survival following nephrectomy for pNx/0, M0 RCC, stratified by statin exposure.
estimated glomerular filtration rate, patient symptomology, and year of surgery. We again found that statin exposure was not independently associated with cancer progression (HR ¼ 1.23, 95% CI: 0.97–1.56; P ¼ 0.09), death due to RCC (HR ¼ 1.06, 95% CI: 0.78–1.43; P ¼ 0.70), or all-cause mortality (HR ¼ 0.88, 95% CI: 0.74–1.05; P ¼ 0.20).
4. Discussion We found here, in a large cohort of patients with surgically treated RCC with long-term follow-up, that statin use at the time of nephrectomy was not associated with the risks of subsequent cancer progression, RCC-specific mortality, or allcause mortality. Likewise, we found no independent
association of this medication with advanced pathologic stage or greater tumor grade. These findings remained after adjusting for competing risks of progression or mortality. To our knowledge, only 3 prior series to date have assessed the therapeutic role of statin medications in patients with RCC [21–23]. Among these, only the study by Hamilton et al. [21] focused specifically on patients with localized (pNx/0, M0) disease, and therein reported a favorable correlation between statin exposure and outcomes in a population undergoing nephrectomy. On the contrary, as the data presented here do not demonstrate a significant association between preoperative statin use and oncologic outcomes among patients with surgically treated pNx/0, M0 RCC, our findings do not support a current anticancer role for statin therapy among patients with localized RCC undergoing nephrectomy.
Fig. 3. Overall survival following nephrectomy for pNx/0, M0 RCC, stratified by statin exposure.
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
6
Table 4 Multivariable analysis of factors associated with disease progression, death due to RCC, and all-cause mortality following nephrectomy Feature
Disease progression HR (95% CI)
Age at surgery Sex (ref. female) Type of surgery (ref. partial) Constitutional symptoms Smoking status (ref. never) Current Former ECOG performance status (ref. ¼ 0) Charlson score BMI (ref. BMI o30) Tumor size RCC histologic subtype (ref. other) Clear cell, collecting duct, NOS 2010 Primary tumor classification (ref. ¼ pT1a) pT1b pT2a pT2b pT3a pT3b pT3c/ pT4 Nuclear grade (ref. ¼ grade 1 or 2) 3 4 Coagulative tumor necrosis Sarcomatoid differentiation Statins
Death due to RCC P value
HR (95% CI)
All-cause mortality P value
o0.001 0.80 0.10 0.02
0.80 0.10 o0.001 0.03 0.60 0.001
1.19 1.16 2.30 1.25 1.04 1.08
(0.94–1.50) (0.98–1.37) (1.90–2.79) (1.20–1.31)b (0.89–1.22) (1.05–1.12)c
0.10 0.09 o0.001 o0.001 0.60 o0.001
o0.001
1.42 (1.16–1.75)
0.40 0.40 0.40 0.05
1.08 0.72 1.04 1.41
(0.95–1.22) (0.53–0.97) (0.66–1.63) (1.05–1.90)
0.30 0.03 0.90 0.02
0.94 1.13 1.24 0.99 1.09 1.04
(0.69–1.28) (0.90–1.43) (0.91–1.69) (0.92–1.07)b (0.88–1.36) (1.00–1.07)c
0.70 0.20 0.20 0.90 0.40 0.04
1.06 1.26 1.85 1.10 1.08 1.07
(0.73–1.55) (0.95–1.68) (1.30–2.64) (1.01–1.21)b (0.83–1.41) (1.03–1.12)c
2.15 (1.38–3.36)
P value
(1.39–1.63) (0.86–1.22) (0.97–1.45) (1.05–1.59)
(0.87–1.06) (0.71–1.14) (0.83–1.69) (1.01–1.65)
o0.001
HR (95% CI) 1.50 1.03 1.18 1.29
0.96 0.90 1.19 1.29
2.32 (1.61–3.35)
a
a
a
o0.001
3.88 5.35 7.75 10.30 16.47 19.14
(2.38–6.32) (3.03–9.44) (4.03–14.92) (6.05–17.54) (8.97–30.24) (9.66–37.92)
o0.001 o0.001 o0.001 o0.001 o0.001 o0.001
3.19 2.94 4.68 7.87 12.25 11.67
(1.72–5.92) (1.41–6.10) (2.04–10.71) (4.05–15.28) (5.89–25.46) (4.99–27.29)
o0.001 0.004 o0.001 o0.001 o0.001 o0.001
1.02 0.61 0.92 1.28 1.73 1.59
(0.82–1.28) (0.42–0.88) (0.58–1.48) (0.94–1.75) (1.14–2.61) (0.91–2.79)
0.90 0.009 0.70 0.10 0.010 0.10
2.09 3.05 1.69 2.54 1.22
(1.56–2.81) (1.97–4.74) (1.32–2.17) (1.46–4.42) (0.95–1.57)
o0.001 o0.001 o0.001 0.001 0.10
2.72 4.30 1.92 1.68 1.02
(1.82–4.05) (2.48–7.45) (1.43–2.59) (0.88–3.22) (0.74–1.39)
o0.001 o0.001 o0.001 0.10 0.90
1.26 2.28 1.45 1.08 0.84
(1.05–1.51) (1.62–3.21) (1.19–1.75) (0.63–1.85) (0.69–1.00)
0.01 o0.001 o0.001 0.80 0.05
a
Hazard ratio and CI represent a 10-year increase in age. Hazard ratio and CI represent a 1-unit increase in Charlson score. c Hazard ratio and CI represent a 1-cm increase in tumor size. b
The association between statin use and cancer-specific outcomes is complex and yet to be fully elucidated. Statins inhibit cholesterol biosynthesis, which remains a vital component of cellular membrane integrity and multiple downstream stream cellular processes [2]. In particular, the mevalonate pathway, a byproduct of cholesterol synthesis directly inhibited by HMG-CoA reductase inhibitors, plays a critical role in the development and progression of multiple malignancies through the promotion of invasion, 3-dimensional growth, and cell survival [5]. In fact, Freed-Pastor et al. [14] demonstrated that a mutant p53 suppressor gene up-regulates the expression of multiple mevalonate pathway genes, leading to the maintenance and progression of breast cancer cell lines. Likewise, the in vitro inhibition of mevalonate production within prostate cancer cells by lipophilic HMG-CoA reductase inhibitors led to a downstream reduction in isoprenoid GGPP, an activator of RhoA activity, culminating in attenuated migration/invasion of prostate cancer cell lines [25]. Accordingly, large epidemiological series have validated these laboratory findings with statin therapy demonstrating a 17% to 33% reduction in recurrence among breast cancer survivors [13] and a 45% reduction in prostate cancer mortality [7]. The role of statins in RCC pathophysiology, meanwhile, remains to be established. That is, research by Fang et al.
[15] demonstrated proapoptotic and metastasis-inhibitory effects of statins on RCC cell lines both in vitro and in vivo through the attenuation of the AKT/mTOR, ERK, and JAK2/STAT3 pathways. Likewise, in a dose-dependent manner, fluvastatin has been shown to induce apoptosis and reduce proliferation of RCC cell lines in vitro via direct suppression of Akt phosphorylation/activation, resulting in inhibition of the mTOR cascade [16]. Moreover, in vitro evidence suggest that statins may potentiate the cytotoxic activity of sorafenib through G1 phase cell cycle arrest [17], further supporting the potential for a therapeutic role of HMG-CoA reductase inhibitors among patients with RCC. However, epidemiological studies assessing the influence of statin exposure on the incidence of RCC have met with conflicting results. In a case-control analysis of 500,000 veterans, statin use was associated with a 48% risk reduction in the incidence of RCC [18], whereas in a prospective analysis of the general population, statins only demonstrated a risk reduction among women [19]. At the same time, others have found no significant association between statin usage and RCC incidence [20,26,27]. Regarding a role for statins among patients with surgically treated RCC, to our knowledge, the largest prior series reported was by Hamilton et al. [21]. In that study, the investigators retrospectively reviewed 2,608 surgically
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
treated patients with localized RCC [21] and noted a 33% risk reduction in cancer progression and improved OS when modeling for statin use as a time-dependent covariate [21]. However, limitations of that study include a short duration of follow-up and a lack of adjustment for histologic features noted at the time of nephrectomy such as tumor grade, sarcomatoid differentiation, and necrosis. We sought here to investigate the potential association of statin therapy with cancer outcomes after nephrectomy in a large cohort of patients with extended follow-up. Interestingly, unlike Hamilton et al. [21], we found no independent association between statin exposure and disease progression, CSS, or OS. Indeed, even when accounting for the competing risks of mortality, statin use was not independently associated with progression or RCC-specific death. Discrepancies in these findings may be the result of the extended follow-up in our series, as well as our incorporation of additional adverse pathologic features that have been associated with adverse oncologic outcomes [28]. We recognize that our study is limited by its retrospective, nonrandomized nature. Specifically, there was significant heterogeneity between cohorts, as those receiving statin therapy were older, with a worse functional status, greater obesity, and more likely to have a smoking history. In addition, we must acknowledge that we were not able here to accurately define the duration of statin exposure, dosage, or patient compliance, which may underlie the noted in vitro [16] relationship between statin use and outcomes, nor could we evaluate the effect of statin use after surgery on subsequent survival. Importantly, however, our results should not undermine the intended effect of statin therapy; which was independently protective against death owing to non–RCC-related causes. As such, in light of ongoing controversy, future prospective analyses are warranted to elucidate the anticancer role of statin use among patients with surgically treated RCC.
5. Conclusions We found no significant association between statin use and outcomes among patients with localized RCC undergoing nephrectomy. Specifically, we found no independent association with cancer progression, RCC-specific mortality, or all-cause mortality. These data suggest a limited anticancer role for statins in this setting. References [1] Heart Protection Study Collaborative Group. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360:7–22. [2] Chan KK, Oza AM, Siu LL. The statins as anticancer agents. Clin Cancer Res 2003;9:10–9.
7
[3] Gauthaman K, Fong CY, Statins Bongso A. stem cells, and cancer. J Cell Biochem 2009;106:975–83. [4] Thurnher M, Nussbaumer O, Gruenbacher G. Novel aspects of mevalonate pathway inhibitors as antitumor agents. Clin Cancer Res 2012;18:3524–31. [5] Gruenbacher G, Thurnher M. Mevalonate metabolism in cancer. Cancer Lett 2015;356:192–6. [6] Demierre MF, Higgins PD, Gruber SB, Hawk E, Lippman SM. Statins and cancer prevention. Nat Rev Cancer 2005;5:930–42. [7] Yu O, Eberg M, Benayoun S, Aprikian A, Batist G, Suissa S, et al. Use of statins and the risk of death in patients with prostate cancer. J Clin Oncol 2014;32:5–11. [8] Banez LL, Klink JC, Jayachandran J, Lark AL, Gerber L, Hamilton RJ, et al. Association between statins and prostate tumor inflammatory infiltrate in men undergoing radical prostatectomy. Cancer Epidemiol Biomarkers Prev 2010;19:722–8. [9] Nielsen SF, Nordestgaard BG, Bojesen SE. Statin use and reduced cancer-related mortality. N Engl J Med 2012;367:1792–802. [10] Wang C, Tao W, Wang Y, Bikow J, Lu B, Keating A, et al. Rosuvastatin, identified from a zebrafish chemical genetic screen for antiangiogenic compounds, suppresses the growth of prostate cancer. Eur Urol 2010;58:418–26. [11] Cardwell CR, Hicks BM, Hughes C, Murray LJ. Statin use after colorectal cancer diagnosis and survival: a population-based cohort study. J Clin Oncol 2014;32:3177–83. [12] Cho SJ, Kim JS, Kim JM, Lee JY, Jung HC, Song IS. Simvastatin induces apoptosis in human colon cancer cells and in tumor xenografts, and attenuates colitis-associated colon cancer in mice. Int J Cancer 2008;123:951–7. [13] Ahern TP, Lash TL, Damkier P, Christiansen PM, Cronin-Fenton DP. Statins and breast cancer prognosis: evidence and opportunities. Lancet Oncol 2014;15:e461–8. [14] Freed-Pastor WA, Mizuno H, Zhao X, Langerod A, Moon SH, Rodriguez-Barrueco R, et al. Mutant p53 disrupts mammary tissue architecture via the mevalonate pathway. Cell 2012;148:244–58. [15] Fang Z, Tang Y, Fang J, Zhou Z, Xing Z, Guo Z, et al. Simvastatin inhibits renal cancer cell growth and metastasis via AKT/mTOR, ERK and JAK2/STAT3 pathway. PLoS One 2013;8:e62823. [16] Woodard J, Sassano A, Hay N, Platanias LC. Statin-dependent suppression of the Akt/mammalian target of rapamycin signaling cascade and programmed cell death 4 up-regulation in renal cell carcinoma. Clin Cancer Res 2008;14:4640–9. [17] Bil J, Zapala L, Nowis D, Jakobisiak M, Golab J. Statins potentiate cytostatic/cytotoxic activity of sorafenib but not sunitinib against tumor cell lines in vitro. Cancer Lett 2010;288:57–67. [18] Khurana V, Caldito G, Ankem M. Statins might reduce risk of renal cell carcinoma in humans: case-control study of 500,000 veterans. Urology 2008;71:118–22. [19] Liu W, Choueiri TK, Cho E. Statin use and the risk of renal cell carcinoma in 2 prospective US cohorts. Cancer 2012;118:797–803. [20] Chiu HF, Kuo CC, Kuo HW, Lee IM, Lee CT, Yang CY. Statin use and the risk of kidney cancer: a population-based case-control study. Expert Opin Drug Saf 2012;11:543–9. [21] Hamilton RJ, Morilla D, Cabrera F, Leapman M, Chen LY, Bernstein M, et al. The association between statin medication and progression after surgery for localized renal cell carcinoma. J Urol 2014;191: 914–9. [22] Choi SK, Min GE, Jeon SH, Lee HL, Chang SG, Yoo KH. Effects of statins on the prognosis of local and locally advanced renal cell carcinoma following nephrectomy. Mol Clin Oncol 2013;1: 365–8. [23] Manoukian GE, Tannir NM, Jonasch E, Qiao W, Haygood TM, Tu SM. Pilot trial of bone-targeted therapy combining zoledronate with fluvastatin or atorvastatin for patients with metastatic renal cell carcinoma. Clin Genitourin Cancer 2011;9:81–8.
8
B.R. Viers et al. / Urologic Oncology: Seminars and Original Investigations ] (2015) 1–8
[24] Fine JP, Gray RJ. A proportional hazards model for the subdistribution of a competing risk. J Am Stat Assoc 1999;94:494–509. [25] Brown M, Hart C, Tawadros T, Ramani V, Sangar V, Lau M, et al. The differential effects of statins on the metastatic behaviour of prostate cancer. Br J Cancer 2012;106:1689–96. [26] Coogan PF, Rosenberg L, Strom BL. Statin use and the risk of 10 cancers. Epidemiology 2007;18:213–9.
[27] Kuoppala J, Lamminpaa A, Pukkala E. Statins and cancer: a systematic review and meta-analysis. Eur J Cancer 2008;44:2122–32. [28] Frank I, Blute ML, Cheville JC, Lohse CM, Weaver AL, Zincke H. An outcome prediction model for patients with clear cell renal cell carcinoma treated with radical nephrectomy based on tumor stage, size, grade and necrosis: the SSIGN score. J Urol 2002;168: 2395–400.