- Email: [email protected]
Expression Profile of Autophagy-related Markers in Localized Prostate Cancer: Correlation With Biochemical Recurrence After Radical Prostatectomy
Oncology Expression Profile of Autophagy-related Markers in Localized Prostate Cancer: Correlation With Biochemical Recurrence After Radical Prostatectomy Bing Liu, Hideaki Miyake, Masatomo Nishikawa, Hiromoto Tei, and Masato Fujisawa OBJECTIVE
METHODS
RESULTS
CONCLUSION
To evaluate the expression of multiple molecular markers involved in autophagy, a cellular degradation pathway for the clearance of damaged or superfluous proteins and organelles, in localized prostate cancer (PC) to clarify the prognostic significance of these markers in patients undergoing radical prostatectomy (RP). Expression levels of 5 autophagy markers, including autophagy-related gene 5, autophagy-related gene 9, Beclin1, microtubule-associated protein light chain 3, and UNC-51elike kinase 1 (ULK1), in RP specimens from 160 consecutive patients with clinically localized PC were measured by immunohistochemical staining. Of these 5 markers, ULK1 expression was significantly correlated with the incidence of biochemical recurrence (BR). On univariate analysis, ULK1 expression, serum prostate-specific antigen level, pathologic stage, Gleason score, seminal vesicle invasion, and surgical margin status were identified as significant predictors of BR. All these significant factors except for seminal vesicle invasion were independently associated with BR on multivariate analysis. Furthermore, significant differences in BR-free survival according to the positive numbers of these 5 independent risk factors were noted, that is, BR occurred in 2 of 33 patients negative for risk factors (6.1%), 20 of 76 patients positive for 1 or 2 risk factors (26.3%), and 38 of 51 patients positive for 3 risk factors (74.5%). Collectively, these findings suggest that measurement of expression levels of potential autophagy markers, particularly ULK1, in RP specimens, in addition to conventional parameters, may contribute to the accurate prediction of BR after RP for localized PC. UROLOGY 85: 1424e1430, 2015. 2015 Elsevier Inc.
R
adical prostatectomy (RP) has been the mainstay of treatment for localized prostate cancer (PC); however, biochemical recurrence (BR), defined as a persistent elevation of serum prostate-specific antigen (PSA), develops in approximately 30% of men undergoing RP.1 To date, there have been a number of studies performed to identify parameters closely associated with prognostic outcomes after RP, which would help determine additional postoperative therapy, as well as the follow-up schedule for an individual patient. As a result, various factors, including molecular biomarkers, are currently regarded as being useful for predicting the prognostic outcomes of RP, such as serum PSA level, Gleason score,
Financial Disclosure: The authors declare that they have no relevant financial interests. From the Division of Urology, Kobe University Graduate School of Medicine, Kobe, Japan Address correspondence to: Hideaki Miyake, M.D., Ph.D., Division of Urology, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan. E-mail: [email protected] Submitted: November 11, 2014, accepted (with revisions): March 6, 2015
1424
ª 2015 Elsevier Inc. All Rights Reserved
pathologic stage, lymph node metastasis, and surgical margin status (SMS),2,3 and several recent studies further reported the value of nomograms considering multiple clinicopathologic parameters in predicting survival after definitive therapy for localized PC.4,5 Nevertheless, PC has been shown to exhibit unique biological features and markedly heterogeneous genetic backgrounds,6 indicating the limitations for predicting postoperative prognostic outcomes in men with localized PC solely using conventional clinicopathologic parameters. Autophagy, a lysosome-dependent pathway for protein degradation, is characterized by the formation of doublemembraneebound organelles known as autophagosomes, and this process is activated in response to a wide variety of metabolic stresses, such as starvation and hypoxia, to maintain cellular homeostasis and viability.7 The dysregulation of autophagy has been demonstrated to be involved in various pathologic conditions, including cardiovascular disorders, microbial infection, and neurodegeneration. However, the functional role of autophagy in carcinogenesis http://dx.doi.org/10.1016/j.urology.2015.03.006 0090-4295/15
is still unclear, that is, the role of autophagy, as a prosurvival mechanism to protect cancer cells from various forms of cellular stress, is supported by most accumulated evidence, whereas, paradoxically, the disruption of autophagyassociated genes has been shown to contribute to accelerated tumorigenesis under some conditions.8,9 In recent years, there have been several studies reporting that the expression profiles of molecular markers consisting of the autophagy pathway could be used as useful predictors of clinical courses in several types of human malignant tumor10-14; however, limited information is available on the prognostic significance of autophagy-related proteins in patients with PC.15-17 Considering these findings, we evaluated the expression patterns of multiple autophagy-related proteins, including autophagy-related gene 5 (Atg5), autophagyrelated gene 9 (Atg9), Beclin1, microtubule-associated protein light chain 3 (LC3), and UNC-51elike kinase 1 (ULK1), in RP specimens from 160 patients with localized PC to analyze the prognostic significance of these markers in this cohort of patients.
METHODS This study included a total of 160 consecutive patients who were diagnosed with clinically organ-confined PC and then underwent RP and bilateral pelvic lymphadenectomy without any neoadjuvant therapies between 2007 and 2010 at our institution. Clinical staging was determined according to the results of digital rectal examination, transrectal ultrasonography, measurement of serum PSA level, computed tomography, magnetic resonance imaging, and/or bone scan. In this series, RP was performed based on the procedure described by Walsh18 in combination with that reported by others.19 The median duration of postoperative follow-up of this study was 51 months (range, 6-76 months). Informed consent for performing immunohistochemical staining of RP specimens was obtained from all patients, and the study design was approved by the Research Ethics Committee of our institution. Pathologic examinations were generally performed under the guidance of a single pathologist according to the 2002 TNM classification system. The surface of the resected RP specimen was inked and fixed, and whole-mount step sections were cut transversely at 3-mm intervals from the apex of the prostate to the tips of the seminal vesicles. A positive surgical margin was defined as the presence of cancer cells at the inked margin. Gleason grading was evaluated according to the modified system of the International Society of Urological Pathology Consensus Conference.20 After RP, follow-up of patients was conducted by periodic measurement of the serum PSA level at least every 3 months for the first 2 years, and every 6 months thereafter. BR was defined as a PSA level of 0.2 ng/mL on 2 consecutive measurements. Irrespective of pathologic findings suggesting a poor prognosis, no patient received any adjuvant therapies until their serum PSA levels reached 0.4 ng/mL. Immunohistochemical staining of RP specimens was performed as described previously.21 Briefly, sections from formaldehyde-fixed paraffin-embedded tissue from 160 RP specimens were deparaffinized and rehydrated. After blocking endogenous peroxidase with hydrogen peroxide, the sections were incubated with 5% normal blocking serum for 20 minutes. UROLOGY 85 (6), 2015
The sections were incubated with the following antihuman antibodies overnight at 4 C: Atg5 (Cell Signaling Technology, Beverly, MA), Atg9 (Santa Cruz Biotechnology, Santa Cruz, CA), Beclin1, LC3 (Cell Signaling Technology), ULK1 (Abcam, Cambridge, United Kingdom), and mammalian target of rapamycin (mTOR; Cell Signaling Technology) and were subsequently incubated with their corresponding secondary antibodies (Vector Laboratories, Burlingame, CA) for 30 minutes at room temperature. After incubation in an avidin-biotin peroxidase complex for 30 minutes, the sections were exposed to diaminobenzidine tetrahydrochloride solution and counterstained with methyl green. All findings on immunohistochemical staining were evaluated by 2 independent investigators (B.L. and M.N.) who were blinded to the clinicopathologic data. If inconsistent interpretations were obtained, differences were resolved by a joint review or consultation with a third observer familiar with immunohistochemical pathology. According to a previous study,12 the expression levels of Atg5, Atg9, and ULK1 were scored by the multiplication of the proportion of immunoreactive cells (1: 0%-5%, 2: 6%-25%, 3: 26%-75%, or 4: 75%100%) by the staining intensity (1: weak, 2: medium, or 3: strong), and a score >6 was considered as strong expression. Beclin1 and LC3 expressions were scored by the summation of the proportion of immunoreactive cells (0: 0%-9%, 1: 10%49%, or 2: 50%-100%) and the staining intensity (0: no staining, 1: weak, 2: medium, or 3: strong), and a score of 4 or 5 was considered to represent strong expression, as previously described.10 As previously described,22 tumors with <10% cells with weak staining of mTOR were scored as 0, those with >10% cells with weak staining or <20% cells with intermediate-tostrong staining of mTOR were scored as 1, and those with >20% cells with intermediate-to-strong staining of mTOR were scored as 2, and a staining score of either 1 or 2 was considered to represent strong mTOR expression. All statistical analyses were performed using StatView 5.0 software (Abacus Concepts, Inc., Berkeley, CA), and P values <.05 were considered significant. A chi-square test was used to analyze the association between several factors and the incidence of BR. BR-free survival rates were calculated using the Kaplan-Meier method, and differences were determined by the log-rank test. The prognostic significance of certain factors was assessed using the Cox proportional hazards regression model.
RESULTS During the observation period of this study (median, 51 months), 60 of the 160 patients (37.5%) experienced BR, and 1-, 3-, and 5-year BR-free survival rates were 88.1%, 66.9%, and 63.2%, respectively (Fig. 1A). Table 1 presents the characteristics of the 160 patients with clinically localized PC according to the presence or absence of BR. Of the several clinicopathologic parameters, serum PSA level, pathologic stage, Gleason score, seminal vesicle invasion (SVI), and SMS were significantly correlated with the incidence of BR. Of the 5 autophagy markers investigated in this study, the expression level of ULK1, but not the remaining 4 markers, was shown to be significantly correlated with the probability of BR. In addition, there was a significant difference in serum PSA level (9.2 vs 12.1 ng/mL; P ¼ .0031), but not Gleason score (7.1 vs 7.0; P ¼ .73), 1425
Figure 1. (A) Biochemical recurrence (BR)efree survival of 160 patients with clinically localized prostate cancer (PC) who underwent radical prostatectomy (RP). (B) BR-free survival of patients with clinically localized PC who underwent RP according to the expression level of UNC-51elike kinase 1. (C) BR-free survival of patients with clinically localized PC who underwent RP according to the number of independent risk factors for postoperative BR, including the serum level of prostate-specific antigen, pathologic stage, Gleason score, surgical margin status, and expression level of UNC-51elike kinase 1. (Color version available online.)
between patients with a weak ULK1 expression and those with a strong expression. The expression level of mTOR, one of the potential inhibitors of autophagic pathway,23 was also examined, and the strong expression of mTOR was observed in 122 of the 160 RP specimens (76.3%); however, the expression levels of all 5 autophagy markers were shown to lack the significant association with the mTOR expression (data not shown). The significance of several clinicopathologic parameters, as well as multiple autophagy-related markers, as predictors of BR was analyzed by univariate and multivariate analyses using the Cox proportional hazards regression model. As shown in Table 2, univariate analysis identified a strong ULK1 expression in addition to elevated serum PSA level, advanced pathologic stage, high Gleason score, SVI, and positive SMS as significant predictors of BR. Of these 6 factors, all except for SVI were independently associated with BR on multivariate analysis. Representative findings on immunohistochemical staining to detect the expression of ULK1 are 1426
presented in Figure 2, and BR-free survival curves according to the expression level of ULK1 are shown in Figure 1B. To more precisely predict the biochemical outcomes in the 160 patients after RP, we categorized them into 3 groups according to positive numbers of the 5 independent risk factors for BR identified by multivariate analysis. BR occurred in 2 of the 33 patients without any risk factors (6.1%), 20 of the 76 patients with 1 or 2 risk factors (26.3%), and 38 of the 51 patients with 3 risk factors (74.5%). As shown in Figure 1C, there were significant differences in BR-free survival among these 3 groups.
COMMENT Autophagy is regarded as an important cellular process, through which waste intracellular materials, such as aged or damaged proteins and organelles, are digested and recycled for renewed energy production.7 This complex process is UROLOGY 85 (6), 2015
Table 1. Patient characteristics according to biochemical recurrence Biochemical Recurrence Variables Age (y) <70 70 Serum PSA level (ng/mL) <10 10 Pathologic stage pT2 pT3 Gleason score 6 or 7 8 Seminal vesicle invasion Negative Positive Surgical margin status Negative Positive LC3B Weak expression Strong expression Beclin1 Weak expression Strong expression Atg5 Weak expression Strong expression Atg9 Weak expression Strong expression ULK1 Weak expression Strong expression
Negative (n ¼ 99)
Positive (n ¼ 61)
58 42
44 16
72 28
23 37
71 29
23 37
85 15
40 20
96 4
59 11
72 28
20 40
64 36
36 24
43 57
28 32
15 85
13 47
57 43
39 21
76 24
16 44
P Value .051 <.001 <.001 .007 .003 <.001 .613 .651 .283 .317 <.001
ATG5, autophagy-related gene 5; ATG9, autophagy-related gene 9; LC3B, microtubule-associated- protein-light-chain-3B; PSA, prostate-specific antigen; ULK1, UNC-51elike kinase 1.
conducted by a group of evolutionarily conserved proteins, including the 4 following functional groups: Atg proteins mediating the induction of autophagy (Beclin1, ULK1, Atg13, and Atg17), vesicle nucleation (ULK1, Atg13, and Atg17), autophagosome formation (Atg5, Atg12, and LC3), and retrieval of autophagy-related proteins (Atg2, Atg9, and Atg18).7-9,24 Despite intensive efforts to clarify the molecular mechanism of autophagy, its functional role during the progression of human malignancies remains controversial.8,9 Furthermore, inconsistent findings have been reported with respect to whether the expression profile of autophagyrelated markers is associated with the prognosis of patients with malignant tumors.10-14 In the field of PC research as well, a few studies assessed the significance of molecular markers involved in autophagy.15-17 For example, Giatromanolaki et al17 reported that the strong and extensive expression of LC3 in PC specimens was closely associated with the Gleason score, whereas Jiang et al16 showed the significant impact of the elevated expression of a mitochondrion-associated autophagy inhibitor, leucine-rich pentatricopeptide repeat motifecontaining protein, on the UROLOGY 85 (6), 2015
prognosis of PC patients at a late stage with a poor prognosis. To our knowledge, however, there have not been any studies investigating the prognostic impact of molecular markers involved in the autophagic pathway for men with early stage PC. Considering these findings, in this study, a total of 160 consecutive patients with clinically localized PC were examined for the expression levels of 5 proteins playing major roles in autophagy, including Atg5, Atg9, Beclin1, LC3, and ULK1, in RP specimens to elucidate the significance of these markers regarding biochemical outcomes after RP. In this series, BR occurred in 37.5% of the included patients, and the 5-year BR-free survival rate was 63.2%, which is consistent with those reported in previous studies.1 Furthermore, immunohistochemical studies revealed that various expression levels of the 5 autophagy-related proteins were detectable, even at very low levels, in the majority of PC tissues. Based on precise assessment of the expression level of each marker in RP specimens considering both the staining area and its intensity, the expression level of ULK1, but not the remaining 4 markers, was shown to be significantly correlated with the incidence of BR. In addition, ULK1 expression appeared to have a significant impact on serum PSA level, but not on Gleason score. These findings suggest the crucial role of ULK1 in regulation of the disease extension of PC, rather than that of biological aggressiveness, at an early stage. The precise prediction of the postoperative clinical course of patients undergoing surgical resection of malignant tumors is important for determining the appropriate postoperative follow-up schedule, as well as additional therapeutic options. In patients with PC as well, a number of studies have been carried out to identify factors, including molecular markers, predicting the biochemical outcome after RP,2,3 and several types of nomograms for calculating the probability of BR after definitive therapy targeting localized PC have been reported.4,5 Although some of these systems were shown to be useful, various limitations concerning their abilities to predict the clinical course after RP using conventional parameters alone have also been suggested.25 Accordingly, we subsequently examined the impacts of potential markers on the autophagy pathway, in addition to those of several clinicopathologic parameters, on the BR-free survival after RP. Univariate analysis showed that postoperative BR was significantly correlated with the serum PSA level, pathologic stage, Gleason score, SVI, SMS, and expression levels of ULK1. Of these factors, the serum PSA level, pathologic stage, Gleason score, SMS, and ULK1 expression were identified as independent predictors of BR-free survival on multivariate analysis. The prognostic significance of the 4 independent conventional parameters in patients with PC undergoing RP has been well recognized2-4; however, this may be the first study to demonstrate the potential impact of ULK1 expression on the postoperative prognosis of patients with PC. Considering these findings, it is necessary to compare the present finding with other molecular markers and nomograms and to conduct additional analysis of PC 1427
Table 2. Association of various parameters with biochemical recurrenceefree survival in 160 patients by univariate and multivariate analyses Univariate Analysis Variables Age (<70 vs 70 y) Serum PSA level (<10 vs 10 ng/mL) Pathologic stage (pT2 vs pT3) Gleason score (6 or 7 vs 8) Seminal vesicle invasion (negative vs positive) Surgical margin status (negative vs positive) LC3B (weak expression vs strong expression) Beclin1 (weak expression vs strong expression) Atg5 (weak expression vs strong expression) Atg9 (weak expression vs strong expression) ULK1 (weak expression vs strong expression)
Hazard Ratio (95% CI) 0.65 2.83 3.04 2.62 3.83 4.22 1.08 0.80 0.70 0.76 4.84
(0.37-1.12) (1.68-4.76) (1.80-5.14) (1.53-4.50) (1.97-7.45) (2.45-7.26) (0.65-1.82) (0.48-1.34) (0.38-1.29) (0.45-1.30) (2.72-8.59)
Multivariate Analysis
P Value
Hazard Ratio (95% CI)
P Value
.142 <.001 <.001 <.001 <.001 <.001 .76 .40 .26 .32 <.001
— (1.49-4.40) (1.00-3.29) (1.90-6.27) (0.66-2.91) (1.73-5.46) — — — — 5.18 (2.78-9.67)
— .001 .048 <.001 .395 <.001 — — — — <.001
2.56 1.82 3.45 1.38 3.07
CI, confidence interval; LC3B, microtubule-associated- protein-light-chain-3B; other abbreviations as in Table 1.
Figure 2. Typical immunohistochemical staining of clinically localized prostate cancer with an antibody against UNC-51elike kinase 1 (ULK1). (A) Prostate cancer (PC) specimen (pT2; Gleason score, 3 þ 4) with weak expression of ULK1 from a 70year-old patient with a serum prostate-specific antigen (PSA) level of 5.9 ng/mL. (B) PC specimen (pT2; Gleason score, 4 þ 3) with weak expression of ULK1 from a 71-year-old patient with a serum PSA level of 9.7 ng/mL. (C) PC specimen (pT3; Gleason score, 4 þ 3) with strong expression of ULK1 from a 68-year-old patient with a serum PSA level of 11.1 ng/mL. (D) PC specimen (pT2; Gleason score, 4 þ 3) with strong expression of ULK1 from a 71-year-old patient with a serum PSA level of 23.4 ng/mL. (Color version available online.)
specimens, including genomic assessment to develop a more precise system to predict the prognosis of PC patients after RP. It is of interest to consider the theoretical background behind the unfavorable impact of ULK1 overexpression on the postoperative prognosis of patients with PC. ULK1, one of the major human autophagy-related genes with a serine-threonine kinase activity, forms a stable complex with other autophagy-associated proteins. The high-performance mechanisms to initiate autophagy are very important to organisms when cells undergo stress 1428
under normal conditions, and the ULK1 complex is the core of the autophagy induction machinery.26 Taken together, in cancer cells, particularly those in early-stage disease, a high level of ULK1 expression may contribute to activate the initiation of autophagy and protect them from apoptosis, which could explain the present findings showing unfavorable biochemical outcomes after RP in patients with a strong expression of ULK1. Furthermore, considering the direct involvement of mTORC1, but not mTORC2, in the regulation of autophagy via ULK1,23 the lack of significant association between the ULK1 and UROLOGY 85 (6), 2015
mTOR expression may suggest the important role of the crosstalk between ULK1 and mTORC2 in the progression of early-stage PC. In fact, we previously reported the high expression levels of basal autophagy-associated proteins, including ULK1, in human renal cell carcinoma tissues, despite the simultaneous presence of an active mTOR signaling.22,27 However, there have been several studies presenting conflicting outcomes regarding the expression level of ULK1 as a prognostic marker in cancer patients.12,28 For example, Jiang et al12 reported that ULK1 expression was an independent prognostic indicator for the survival of patients with esophageal squamous cell carcinoma and that the suppression of ULK1 expression in esophageal squamous cell carcinoma cell lines by specific small interfering RNA resulted in the inhibition of cell proliferation and induction of apoptosis under starvation conditions, whereas Tang et al28 showed that the decreased expression of ULK1 was associated with breast cancer progression and closely related to decreased autophagic capacity. Collectively, these findings suggest that ULK1 may have different roles in different types of cancer; thus, it is necessary to intensively investigate the impact of ULK1 expression on the prognosis of patients with a wide variety of malignant tumors. Another point of interest is to develop a system to more precisely predict the biochemical outcome after RP by combining potential prognostic indicators. Based on the outcomes achieved by multivariate analysis, the 160 patients were divided into the following 3 groups according to the positive number of independent risk factors for BR: negative for any risk factor, positive for 1 or 2 risk factors, and positive for 3 risk factors, and there appeared to be significant differences in BR-free survivals among these 3 groups. It is necessary to strictly evaluate whether the improved prediction of BR by additional investigation of autophagy-related markers by immunohistochemical staining justifies the effort and cost when compared with the use of conventional clinicopathologic factors alone; these findings suggest that a simultaneous consideration of the 5 major risk factors identified in this series may contribute to develop a novel system that can more accurately predict BR after RP. Here, we would like to mention the limitations of this study. Initially, a sample size of 160 men in such a common disease like localized PC is not large enough to draw definitive conclusions on prognostic issues. In addition, there have been several studies showing significant differences between Japanese and Western populations in the characteristics of PC29; therefore, it might be difficult to apply the present findings to general populations with PC. Finally, this study focused on 5 representative components involved in the autophagy pathway as potential molecular markers for localized PC; however, there may be some other molecules mediating autophagy that are more closely associated with postoperative prognostic outcomes in PC patients.
UROLOGY 85 (6), 2015
CONCLUSION We assessed the significance of the baseline expression levels of multiple molecular markers associated with the autophagy pathway in RP specimens in addition to several conventional parameters as prognostic predictors in patients with clinically localized PC undergoing RP. Of the 5 major markers examined in this study, only ULK1 expression appeared to be independently associated with BR-free survival in these patients. Furthermore, it was suggested that the combined assessment of independent risk factors for BR, including the serum PSA level, pathologic stage, Gleason score, SMS, and ULK1 expression, could be useful for further refinement of prognostic assessment for this cohort of patients. However, prospective studies are required to confirm these findings and provide a better understanding of the role of autophagy in the progression of localized PC after RP. References 1. Amling CL. Biochemical recurrence after localized treatment. Urol Clin North Am. 2006;33:147-159. 2. Quinn DI, Henshall SM, Haynes AM, et al. Prognostic significance of pathologic features in localized prostate cancer treated with radical prostatectomy: implications for staging systems and predictive models. J Clin Oncol. 2001;19:3692-3705. 3. Spencer ES, Johnston RB, Gordon RR, et al. Prognostic value of ERG oncoprotein in prostate cancer recurrence and cause-specific mortality. Prostate. 2013;73:905-912. 4. Kattan MW, Easthan JA, Stapleton AM, et al. A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst. 1998;90:766-771. 5. Punnen S, Freedland SJ, Presti JC Jr, et al. Multi-institutional validation of the CAPRA-S score to predict disease recurrence and mortality after radical prostatectomy. Eur Urol. 2014;65:1171-1177. 6. Mackinnon AC, Yan BC, Joseph LJ, et al. Molecular biology underlying the clinical heterogeneity of prostate cancer: an update. Arch Pathol Lab Med. 2009;133:1033-1040. 7. He C, Klionsky DJ. Regulation mechanisms and signaling pathways of autophagy. Annu Rev Genet. 2009;43:67-93. 8. Mathew R, Karantza-Wadsworth V, White E. Role of autophagy in cancer. Nat Rev Cancer. 2007;12:961-967. 9. Eisenberg-Lerner A, Kimchi A. The paradox of autophagy and its implication in cancer etiology and therapy. Apoptosis. 2009;14:376-391. 10. Kim KM, Yu TK, Chu HH, et al. Expression of ER stress and autophagy-related molecules in human non-small cell lung cancer and premalignant lesions. Int J Cancer. 2012;131:362-370. 11. Shen Y, Li DD, Wang LL, et al. Decreased expression of autophagyrelated proteins in malignant epithelial ovarian cancer. Autophagy. 2008;4:1067-1068. 12. Jiang S, Li Y, Zhu YH, et al. Intensive expression of UNC-51-like kinase 1 is a novel biomarker of poor prognosis in patients with esophageal squamous cell carcinoma. Cancer Sci. 2011;102: 1568-1575. 13. Giatromanolaki A, Koukourakis MI, Koutsopoulos A, et al. High Beclin 1 expression defines a poor prognosis in endometrial adenocarcinomas. Gynecol Oncol. 2011;123:147-151. 14. Lee YJ, Ha YJ, Kang YN, et al. The autophagy-related marker LC3 can predict prognosis in human hepatocellular carcinoma. PLoS One. 2013;8:e81540. 15. Jiang X, Zhong W, Huang H, et al. Autophagy defects suggested by low levels of autophagy activator MAP1S and high levels of autophagy inhibitor LRPPRC predict poor prognosis of prostate cancer patients. Mol Carcinog. 2014. http://dx.doi.org/10.1002/mc.22193.
1429
16. Jiang X, Li X, Huang H, et al. Elevated levels of mitochondrionassociated autophagy inhibitor LRPPRC are associated with poor prognosis in patients with prostate cancer. Cancer. 2014;120:1228-1236. 17. Giatromanolaki A, Sivridis E, Mendrinos S, et al. Autophagy proteins in prostate cancer: relation with anaerobic metabolism and Gleason score. Urol Oncol. 2014;32:39.e11-39.e18. 18. Walsh PC. Radical prostatectomy: a procedure in evolution. Semin Oncol. 1994;21:662-671. 19. Koch MO. Management of the dorsal vein complex during radical retropubic prostatectomy. Semin Urol Oncol. 2000;18:33-37. 20. 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;29:1228-1242. 21. Kurahashi T, Miyake H, Hara I, et al. Expression of major heat shock proteins in prostate cancer: correlation with clinicopathological outcomes in patients undergoing radical prostatectomy. J Urol. 2007;177:757-761. 22. Nishikawa M, Miyake H, Harada K, et al. Expression of molecular markers associated with the mammalian target of rapamycin pathway in nonmetastatic renal cell carcinoma: effect on prognostic outcomes following radical nephrectomy. Urol Oncol. 2014;32:49. e15-49.e21.
1430
23. Alers S, L€offler AS, Wesselborg S, et al. Role of AMPK-mTORUlk1/2 in the regulation of autophagy: cross talk, shortcuts, and feedbacks. Mol Cell Biol. 2012;32:2-11. 24. Mari M, Reggiori F. Shaping membranes into autophagosomes. Nat Cell Biol. 2007;9:1125-1127. 25. Stephenson AJ, Scardino PT, Eastham JA, et al. Preoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Natl Cancer Inst. 2006;98: 715-717. 26. Hara T, Takamura A, Kishi C, et al. FIP200, a ULK-interacting protein, is required for autophagosome formation in mammalian cells. J Cell Biol. 2008;181:497-510. 27. Nishikawa M, Miyake H, Liu B, et al. Expression pattern of autophagy-related markers in non-metastatic clear cell renal cell carcinoma: Association with disease recurrence following radical nephrectomy. J Cancer Res Clin Oncol. in press. 28. Tang J, Deng R, Luo RZ, et al. Low expression of ULK1 is associated with operable breast cancer progression and is an adverse prognostic marker of survival for patients. Breast Cancer Res Treat. 2012;134:549-560. 29. Takechi H, Ito K, Yamamoto T, et al. Prostate-specific antigen kinetics in screen-detected prostate cancer in Japan. Urology. 2008; 72:1111-1115.
UROLOGY 85 (6), 2015
METHODS
RESULTS
CONCLUSION
To evaluate the expression of multiple molecular markers involved in autophagy, a cellular degradation pathway for the clearance of damaged or superfluous proteins and organelles, in localized prostate cancer (PC) to clarify the prognostic significance of these markers in patients undergoing radical prostatectomy (RP). Expression levels of 5 autophagy markers, including autophagy-related gene 5, autophagy-related gene 9, Beclin1, microtubule-associated protein light chain 3, and UNC-51elike kinase 1 (ULK1), in RP specimens from 160 consecutive patients with clinically localized PC were measured by immunohistochemical staining. Of these 5 markers, ULK1 expression was significantly correlated with the incidence of biochemical recurrence (BR). On univariate analysis, ULK1 expression, serum prostate-specific antigen level, pathologic stage, Gleason score, seminal vesicle invasion, and surgical margin status were identified as significant predictors of BR. All these significant factors except for seminal vesicle invasion were independently associated with BR on multivariate analysis. Furthermore, significant differences in BR-free survival according to the positive numbers of these 5 independent risk factors were noted, that is, BR occurred in 2 of 33 patients negative for risk factors (6.1%), 20 of 76 patients positive for 1 or 2 risk factors (26.3%), and 38 of 51 patients positive for 3 risk factors (74.5%). Collectively, these findings suggest that measurement of expression levels of potential autophagy markers, particularly ULK1, in RP specimens, in addition to conventional parameters, may contribute to the accurate prediction of BR after RP for localized PC. UROLOGY 85: 1424e1430, 2015. 2015 Elsevier Inc.
R
adical prostatectomy (RP) has been the mainstay of treatment for localized prostate cancer (PC); however, biochemical recurrence (BR), defined as a persistent elevation of serum prostate-specific antigen (PSA), develops in approximately 30% of men undergoing RP.1 To date, there have been a number of studies performed to identify parameters closely associated with prognostic outcomes after RP, which would help determine additional postoperative therapy, as well as the follow-up schedule for an individual patient. As a result, various factors, including molecular biomarkers, are currently regarded as being useful for predicting the prognostic outcomes of RP, such as serum PSA level, Gleason score,
Financial Disclosure: The authors declare that they have no relevant financial interests. From the Division of Urology, Kobe University Graduate School of Medicine, Kobe, Japan Address correspondence to: Hideaki Miyake, M.D., Ph.D., Division of Urology, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan. E-mail: [email protected] Submitted: November 11, 2014, accepted (with revisions): March 6, 2015
1424
ª 2015 Elsevier Inc. All Rights Reserved
pathologic stage, lymph node metastasis, and surgical margin status (SMS),2,3 and several recent studies further reported the value of nomograms considering multiple clinicopathologic parameters in predicting survival after definitive therapy for localized PC.4,5 Nevertheless, PC has been shown to exhibit unique biological features and markedly heterogeneous genetic backgrounds,6 indicating the limitations for predicting postoperative prognostic outcomes in men with localized PC solely using conventional clinicopathologic parameters. Autophagy, a lysosome-dependent pathway for protein degradation, is characterized by the formation of doublemembraneebound organelles known as autophagosomes, and this process is activated in response to a wide variety of metabolic stresses, such as starvation and hypoxia, to maintain cellular homeostasis and viability.7 The dysregulation of autophagy has been demonstrated to be involved in various pathologic conditions, including cardiovascular disorders, microbial infection, and neurodegeneration. However, the functional role of autophagy in carcinogenesis http://dx.doi.org/10.1016/j.urology.2015.03.006 0090-4295/15
is still unclear, that is, the role of autophagy, as a prosurvival mechanism to protect cancer cells from various forms of cellular stress, is supported by most accumulated evidence, whereas, paradoxically, the disruption of autophagyassociated genes has been shown to contribute to accelerated tumorigenesis under some conditions.8,9 In recent years, there have been several studies reporting that the expression profiles of molecular markers consisting of the autophagy pathway could be used as useful predictors of clinical courses in several types of human malignant tumor10-14; however, limited information is available on the prognostic significance of autophagy-related proteins in patients with PC.15-17 Considering these findings, we evaluated the expression patterns of multiple autophagy-related proteins, including autophagy-related gene 5 (Atg5), autophagyrelated gene 9 (Atg9), Beclin1, microtubule-associated protein light chain 3 (LC3), and UNC-51elike kinase 1 (ULK1), in RP specimens from 160 patients with localized PC to analyze the prognostic significance of these markers in this cohort of patients.
METHODS This study included a total of 160 consecutive patients who were diagnosed with clinically organ-confined PC and then underwent RP and bilateral pelvic lymphadenectomy without any neoadjuvant therapies between 2007 and 2010 at our institution. Clinical staging was determined according to the results of digital rectal examination, transrectal ultrasonography, measurement of serum PSA level, computed tomography, magnetic resonance imaging, and/or bone scan. In this series, RP was performed based on the procedure described by Walsh18 in combination with that reported by others.19 The median duration of postoperative follow-up of this study was 51 months (range, 6-76 months). Informed consent for performing immunohistochemical staining of RP specimens was obtained from all patients, and the study design was approved by the Research Ethics Committee of our institution. Pathologic examinations were generally performed under the guidance of a single pathologist according to the 2002 TNM classification system. The surface of the resected RP specimen was inked and fixed, and whole-mount step sections were cut transversely at 3-mm intervals from the apex of the prostate to the tips of the seminal vesicles. A positive surgical margin was defined as the presence of cancer cells at the inked margin. Gleason grading was evaluated according to the modified system of the International Society of Urological Pathology Consensus Conference.20 After RP, follow-up of patients was conducted by periodic measurement of the serum PSA level at least every 3 months for the first 2 years, and every 6 months thereafter. BR was defined as a PSA level of 0.2 ng/mL on 2 consecutive measurements. Irrespective of pathologic findings suggesting a poor prognosis, no patient received any adjuvant therapies until their serum PSA levels reached 0.4 ng/mL. Immunohistochemical staining of RP specimens was performed as described previously.21 Briefly, sections from formaldehyde-fixed paraffin-embedded tissue from 160 RP specimens were deparaffinized and rehydrated. After blocking endogenous peroxidase with hydrogen peroxide, the sections were incubated with 5% normal blocking serum for 20 minutes. UROLOGY 85 (6), 2015
The sections were incubated with the following antihuman antibodies overnight at 4 C: Atg5 (Cell Signaling Technology, Beverly, MA), Atg9 (Santa Cruz Biotechnology, Santa Cruz, CA), Beclin1, LC3 (Cell Signaling Technology), ULK1 (Abcam, Cambridge, United Kingdom), and mammalian target of rapamycin (mTOR; Cell Signaling Technology) and were subsequently incubated with their corresponding secondary antibodies (Vector Laboratories, Burlingame, CA) for 30 minutes at room temperature. After incubation in an avidin-biotin peroxidase complex for 30 minutes, the sections were exposed to diaminobenzidine tetrahydrochloride solution and counterstained with methyl green. All findings on immunohistochemical staining were evaluated by 2 independent investigators (B.L. and M.N.) who were blinded to the clinicopathologic data. If inconsistent interpretations were obtained, differences were resolved by a joint review or consultation with a third observer familiar with immunohistochemical pathology. According to a previous study,12 the expression levels of Atg5, Atg9, and ULK1 were scored by the multiplication of the proportion of immunoreactive cells (1: 0%-5%, 2: 6%-25%, 3: 26%-75%, or 4: 75%100%) by the staining intensity (1: weak, 2: medium, or 3: strong), and a score >6 was considered as strong expression. Beclin1 and LC3 expressions were scored by the summation of the proportion of immunoreactive cells (0: 0%-9%, 1: 10%49%, or 2: 50%-100%) and the staining intensity (0: no staining, 1: weak, 2: medium, or 3: strong), and a score of 4 or 5 was considered to represent strong expression, as previously described.10 As previously described,22 tumors with <10% cells with weak staining of mTOR were scored as 0, those with >10% cells with weak staining or <20% cells with intermediate-tostrong staining of mTOR were scored as 1, and those with >20% cells with intermediate-to-strong staining of mTOR were scored as 2, and a staining score of either 1 or 2 was considered to represent strong mTOR expression. All statistical analyses were performed using StatView 5.0 software (Abacus Concepts, Inc., Berkeley, CA), and P values <.05 were considered significant. A chi-square test was used to analyze the association between several factors and the incidence of BR. BR-free survival rates were calculated using the Kaplan-Meier method, and differences were determined by the log-rank test. The prognostic significance of certain factors was assessed using the Cox proportional hazards regression model.
RESULTS During the observation period of this study (median, 51 months), 60 of the 160 patients (37.5%) experienced BR, and 1-, 3-, and 5-year BR-free survival rates were 88.1%, 66.9%, and 63.2%, respectively (Fig. 1A). Table 1 presents the characteristics of the 160 patients with clinically localized PC according to the presence or absence of BR. Of the several clinicopathologic parameters, serum PSA level, pathologic stage, Gleason score, seminal vesicle invasion (SVI), and SMS were significantly correlated with the incidence of BR. Of the 5 autophagy markers investigated in this study, the expression level of ULK1, but not the remaining 4 markers, was shown to be significantly correlated with the probability of BR. In addition, there was a significant difference in serum PSA level (9.2 vs 12.1 ng/mL; P ¼ .0031), but not Gleason score (7.1 vs 7.0; P ¼ .73), 1425
Figure 1. (A) Biochemical recurrence (BR)efree survival of 160 patients with clinically localized prostate cancer (PC) who underwent radical prostatectomy (RP). (B) BR-free survival of patients with clinically localized PC who underwent RP according to the expression level of UNC-51elike kinase 1. (C) BR-free survival of patients with clinically localized PC who underwent RP according to the number of independent risk factors for postoperative BR, including the serum level of prostate-specific antigen, pathologic stage, Gleason score, surgical margin status, and expression level of UNC-51elike kinase 1. (Color version available online.)
between patients with a weak ULK1 expression and those with a strong expression. The expression level of mTOR, one of the potential inhibitors of autophagic pathway,23 was also examined, and the strong expression of mTOR was observed in 122 of the 160 RP specimens (76.3%); however, the expression levels of all 5 autophagy markers were shown to lack the significant association with the mTOR expression (data not shown). The significance of several clinicopathologic parameters, as well as multiple autophagy-related markers, as predictors of BR was analyzed by univariate and multivariate analyses using the Cox proportional hazards regression model. As shown in Table 2, univariate analysis identified a strong ULK1 expression in addition to elevated serum PSA level, advanced pathologic stage, high Gleason score, SVI, and positive SMS as significant predictors of BR. Of these 6 factors, all except for SVI were independently associated with BR on multivariate analysis. Representative findings on immunohistochemical staining to detect the expression of ULK1 are 1426
presented in Figure 2, and BR-free survival curves according to the expression level of ULK1 are shown in Figure 1B. To more precisely predict the biochemical outcomes in the 160 patients after RP, we categorized them into 3 groups according to positive numbers of the 5 independent risk factors for BR identified by multivariate analysis. BR occurred in 2 of the 33 patients without any risk factors (6.1%), 20 of the 76 patients with 1 or 2 risk factors (26.3%), and 38 of the 51 patients with 3 risk factors (74.5%). As shown in Figure 1C, there were significant differences in BR-free survival among these 3 groups.
COMMENT Autophagy is regarded as an important cellular process, through which waste intracellular materials, such as aged or damaged proteins and organelles, are digested and recycled for renewed energy production.7 This complex process is UROLOGY 85 (6), 2015
Table 1. Patient characteristics according to biochemical recurrence Biochemical Recurrence Variables Age (y) <70 70 Serum PSA level (ng/mL) <10 10 Pathologic stage pT2 pT3 Gleason score 6 or 7 8 Seminal vesicle invasion Negative Positive Surgical margin status Negative Positive LC3B Weak expression Strong expression Beclin1 Weak expression Strong expression Atg5 Weak expression Strong expression Atg9 Weak expression Strong expression ULK1 Weak expression Strong expression
Negative (n ¼ 99)
Positive (n ¼ 61)
58 42
44 16
72 28
23 37
71 29
23 37
85 15
40 20
96 4
59 11
72 28
20 40
64 36
36 24
43 57
28 32
15 85
13 47
57 43
39 21
76 24
16 44
P Value .051 <.001 <.001 .007 .003 <.001 .613 .651 .283 .317 <.001
ATG5, autophagy-related gene 5; ATG9, autophagy-related gene 9; LC3B, microtubule-associated- protein-light-chain-3B; PSA, prostate-specific antigen; ULK1, UNC-51elike kinase 1.
conducted by a group of evolutionarily conserved proteins, including the 4 following functional groups: Atg proteins mediating the induction of autophagy (Beclin1, ULK1, Atg13, and Atg17), vesicle nucleation (ULK1, Atg13, and Atg17), autophagosome formation (Atg5, Atg12, and LC3), and retrieval of autophagy-related proteins (Atg2, Atg9, and Atg18).7-9,24 Despite intensive efforts to clarify the molecular mechanism of autophagy, its functional role during the progression of human malignancies remains controversial.8,9 Furthermore, inconsistent findings have been reported with respect to whether the expression profile of autophagyrelated markers is associated with the prognosis of patients with malignant tumors.10-14 In the field of PC research as well, a few studies assessed the significance of molecular markers involved in autophagy.15-17 For example, Giatromanolaki et al17 reported that the strong and extensive expression of LC3 in PC specimens was closely associated with the Gleason score, whereas Jiang et al16 showed the significant impact of the elevated expression of a mitochondrion-associated autophagy inhibitor, leucine-rich pentatricopeptide repeat motifecontaining protein, on the UROLOGY 85 (6), 2015
prognosis of PC patients at a late stage with a poor prognosis. To our knowledge, however, there have not been any studies investigating the prognostic impact of molecular markers involved in the autophagic pathway for men with early stage PC. Considering these findings, in this study, a total of 160 consecutive patients with clinically localized PC were examined for the expression levels of 5 proteins playing major roles in autophagy, including Atg5, Atg9, Beclin1, LC3, and ULK1, in RP specimens to elucidate the significance of these markers regarding biochemical outcomes after RP. In this series, BR occurred in 37.5% of the included patients, and the 5-year BR-free survival rate was 63.2%, which is consistent with those reported in previous studies.1 Furthermore, immunohistochemical studies revealed that various expression levels of the 5 autophagy-related proteins were detectable, even at very low levels, in the majority of PC tissues. Based on precise assessment of the expression level of each marker in RP specimens considering both the staining area and its intensity, the expression level of ULK1, but not the remaining 4 markers, was shown to be significantly correlated with the incidence of BR. In addition, ULK1 expression appeared to have a significant impact on serum PSA level, but not on Gleason score. These findings suggest the crucial role of ULK1 in regulation of the disease extension of PC, rather than that of biological aggressiveness, at an early stage. The precise prediction of the postoperative clinical course of patients undergoing surgical resection of malignant tumors is important for determining the appropriate postoperative follow-up schedule, as well as additional therapeutic options. In patients with PC as well, a number of studies have been carried out to identify factors, including molecular markers, predicting the biochemical outcome after RP,2,3 and several types of nomograms for calculating the probability of BR after definitive therapy targeting localized PC have been reported.4,5 Although some of these systems were shown to be useful, various limitations concerning their abilities to predict the clinical course after RP using conventional parameters alone have also been suggested.25 Accordingly, we subsequently examined the impacts of potential markers on the autophagy pathway, in addition to those of several clinicopathologic parameters, on the BR-free survival after RP. Univariate analysis showed that postoperative BR was significantly correlated with the serum PSA level, pathologic stage, Gleason score, SVI, SMS, and expression levels of ULK1. Of these factors, the serum PSA level, pathologic stage, Gleason score, SMS, and ULK1 expression were identified as independent predictors of BR-free survival on multivariate analysis. The prognostic significance of the 4 independent conventional parameters in patients with PC undergoing RP has been well recognized2-4; however, this may be the first study to demonstrate the potential impact of ULK1 expression on the postoperative prognosis of patients with PC. Considering these findings, it is necessary to compare the present finding with other molecular markers and nomograms and to conduct additional analysis of PC 1427
Table 2. Association of various parameters with biochemical recurrenceefree survival in 160 patients by univariate and multivariate analyses Univariate Analysis Variables Age (<70 vs 70 y) Serum PSA level (<10 vs 10 ng/mL) Pathologic stage (pT2 vs pT3) Gleason score (6 or 7 vs 8) Seminal vesicle invasion (negative vs positive) Surgical margin status (negative vs positive) LC3B (weak expression vs strong expression) Beclin1 (weak expression vs strong expression) Atg5 (weak expression vs strong expression) Atg9 (weak expression vs strong expression) ULK1 (weak expression vs strong expression)
Hazard Ratio (95% CI) 0.65 2.83 3.04 2.62 3.83 4.22 1.08 0.80 0.70 0.76 4.84
(0.37-1.12) (1.68-4.76) (1.80-5.14) (1.53-4.50) (1.97-7.45) (2.45-7.26) (0.65-1.82) (0.48-1.34) (0.38-1.29) (0.45-1.30) (2.72-8.59)
Multivariate Analysis
P Value
Hazard Ratio (95% CI)
P Value
.142 <.001 <.001 <.001 <.001 <.001 .76 .40 .26 .32 <.001
— (1.49-4.40) (1.00-3.29) (1.90-6.27) (0.66-2.91) (1.73-5.46) — — — — 5.18 (2.78-9.67)
— .001 .048 <.001 .395 <.001 — — — — <.001
2.56 1.82 3.45 1.38 3.07
CI, confidence interval; LC3B, microtubule-associated- protein-light-chain-3B; other abbreviations as in Table 1.
Figure 2. Typical immunohistochemical staining of clinically localized prostate cancer with an antibody against UNC-51elike kinase 1 (ULK1). (A) Prostate cancer (PC) specimen (pT2; Gleason score, 3 þ 4) with weak expression of ULK1 from a 70year-old patient with a serum prostate-specific antigen (PSA) level of 5.9 ng/mL. (B) PC specimen (pT2; Gleason score, 4 þ 3) with weak expression of ULK1 from a 71-year-old patient with a serum PSA level of 9.7 ng/mL. (C) PC specimen (pT3; Gleason score, 4 þ 3) with strong expression of ULK1 from a 68-year-old patient with a serum PSA level of 11.1 ng/mL. (D) PC specimen (pT2; Gleason score, 4 þ 3) with strong expression of ULK1 from a 71-year-old patient with a serum PSA level of 23.4 ng/mL. (Color version available online.)
specimens, including genomic assessment to develop a more precise system to predict the prognosis of PC patients after RP. It is of interest to consider the theoretical background behind the unfavorable impact of ULK1 overexpression on the postoperative prognosis of patients with PC. ULK1, one of the major human autophagy-related genes with a serine-threonine kinase activity, forms a stable complex with other autophagy-associated proteins. The high-performance mechanisms to initiate autophagy are very important to organisms when cells undergo stress 1428
under normal conditions, and the ULK1 complex is the core of the autophagy induction machinery.26 Taken together, in cancer cells, particularly those in early-stage disease, a high level of ULK1 expression may contribute to activate the initiation of autophagy and protect them from apoptosis, which could explain the present findings showing unfavorable biochemical outcomes after RP in patients with a strong expression of ULK1. Furthermore, considering the direct involvement of mTORC1, but not mTORC2, in the regulation of autophagy via ULK1,23 the lack of significant association between the ULK1 and UROLOGY 85 (6), 2015
mTOR expression may suggest the important role of the crosstalk between ULK1 and mTORC2 in the progression of early-stage PC. In fact, we previously reported the high expression levels of basal autophagy-associated proteins, including ULK1, in human renal cell carcinoma tissues, despite the simultaneous presence of an active mTOR signaling.22,27 However, there have been several studies presenting conflicting outcomes regarding the expression level of ULK1 as a prognostic marker in cancer patients.12,28 For example, Jiang et al12 reported that ULK1 expression was an independent prognostic indicator for the survival of patients with esophageal squamous cell carcinoma and that the suppression of ULK1 expression in esophageal squamous cell carcinoma cell lines by specific small interfering RNA resulted in the inhibition of cell proliferation and induction of apoptosis under starvation conditions, whereas Tang et al28 showed that the decreased expression of ULK1 was associated with breast cancer progression and closely related to decreased autophagic capacity. Collectively, these findings suggest that ULK1 may have different roles in different types of cancer; thus, it is necessary to intensively investigate the impact of ULK1 expression on the prognosis of patients with a wide variety of malignant tumors. Another point of interest is to develop a system to more precisely predict the biochemical outcome after RP by combining potential prognostic indicators. Based on the outcomes achieved by multivariate analysis, the 160 patients were divided into the following 3 groups according to the positive number of independent risk factors for BR: negative for any risk factor, positive for 1 or 2 risk factors, and positive for 3 risk factors, and there appeared to be significant differences in BR-free survivals among these 3 groups. It is necessary to strictly evaluate whether the improved prediction of BR by additional investigation of autophagy-related markers by immunohistochemical staining justifies the effort and cost when compared with the use of conventional clinicopathologic factors alone; these findings suggest that a simultaneous consideration of the 5 major risk factors identified in this series may contribute to develop a novel system that can more accurately predict BR after RP. Here, we would like to mention the limitations of this study. Initially, a sample size of 160 men in such a common disease like localized PC is not large enough to draw definitive conclusions on prognostic issues. In addition, there have been several studies showing significant differences between Japanese and Western populations in the characteristics of PC29; therefore, it might be difficult to apply the present findings to general populations with PC. Finally, this study focused on 5 representative components involved in the autophagy pathway as potential molecular markers for localized PC; however, there may be some other molecules mediating autophagy that are more closely associated with postoperative prognostic outcomes in PC patients.
UROLOGY 85 (6), 2015
CONCLUSION We assessed the significance of the baseline expression levels of multiple molecular markers associated with the autophagy pathway in RP specimens in addition to several conventional parameters as prognostic predictors in patients with clinically localized PC undergoing RP. Of the 5 major markers examined in this study, only ULK1 expression appeared to be independently associated with BR-free survival in these patients. Furthermore, it was suggested that the combined assessment of independent risk factors for BR, including the serum PSA level, pathologic stage, Gleason score, SMS, and ULK1 expression, could be useful for further refinement of prognostic assessment for this cohort of patients. However, prospective studies are required to confirm these findings and provide a better understanding of the role of autophagy in the progression of localized PC after RP. References 1. Amling CL. Biochemical recurrence after localized treatment. Urol Clin North Am. 2006;33:147-159. 2. Quinn DI, Henshall SM, Haynes AM, et al. Prognostic significance of pathologic features in localized prostate cancer treated with radical prostatectomy: implications for staging systems and predictive models. J Clin Oncol. 2001;19:3692-3705. 3. Spencer ES, Johnston RB, Gordon RR, et al. Prognostic value of ERG oncoprotein in prostate cancer recurrence and cause-specific mortality. Prostate. 2013;73:905-912. 4. Kattan MW, Easthan JA, Stapleton AM, et al. A preoperative nomogram for disease recurrence following radical prostatectomy for prostate cancer. J Natl Cancer Inst. 1998;90:766-771. 5. Punnen S, Freedland SJ, Presti JC Jr, et al. Multi-institutional validation of the CAPRA-S score to predict disease recurrence and mortality after radical prostatectomy. Eur Urol. 2014;65:1171-1177. 6. Mackinnon AC, Yan BC, Joseph LJ, et al. Molecular biology underlying the clinical heterogeneity of prostate cancer: an update. Arch Pathol Lab Med. 2009;133:1033-1040. 7. He C, Klionsky DJ. Regulation mechanisms and signaling pathways of autophagy. Annu Rev Genet. 2009;43:67-93. 8. Mathew R, Karantza-Wadsworth V, White E. Role of autophagy in cancer. Nat Rev Cancer. 2007;12:961-967. 9. Eisenberg-Lerner A, Kimchi A. The paradox of autophagy and its implication in cancer etiology and therapy. Apoptosis. 2009;14:376-391. 10. Kim KM, Yu TK, Chu HH, et al. Expression of ER stress and autophagy-related molecules in human non-small cell lung cancer and premalignant lesions. Int J Cancer. 2012;131:362-370. 11. Shen Y, Li DD, Wang LL, et al. Decreased expression of autophagyrelated proteins in malignant epithelial ovarian cancer. Autophagy. 2008;4:1067-1068. 12. Jiang S, Li Y, Zhu YH, et al. Intensive expression of UNC-51-like kinase 1 is a novel biomarker of poor prognosis in patients with esophageal squamous cell carcinoma. Cancer Sci. 2011;102: 1568-1575. 13. Giatromanolaki A, Koukourakis MI, Koutsopoulos A, et al. High Beclin 1 expression defines a poor prognosis in endometrial adenocarcinomas. Gynecol Oncol. 2011;123:147-151. 14. Lee YJ, Ha YJ, Kang YN, et al. The autophagy-related marker LC3 can predict prognosis in human hepatocellular carcinoma. PLoS One. 2013;8:e81540. 15. Jiang X, Zhong W, Huang H, et al. Autophagy defects suggested by low levels of autophagy activator MAP1S and high levels of autophagy inhibitor LRPPRC predict poor prognosis of prostate cancer patients. Mol Carcinog. 2014. http://dx.doi.org/10.1002/mc.22193.
1429
16. Jiang X, Li X, Huang H, et al. Elevated levels of mitochondrionassociated autophagy inhibitor LRPPRC are associated with poor prognosis in patients with prostate cancer. Cancer. 2014;120:1228-1236. 17. Giatromanolaki A, Sivridis E, Mendrinos S, et al. Autophagy proteins in prostate cancer: relation with anaerobic metabolism and Gleason score. Urol Oncol. 2014;32:39.e11-39.e18. 18. Walsh PC. Radical prostatectomy: a procedure in evolution. Semin Oncol. 1994;21:662-671. 19. Koch MO. Management of the dorsal vein complex during radical retropubic prostatectomy. Semin Urol Oncol. 2000;18:33-37. 20. 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;29:1228-1242. 21. Kurahashi T, Miyake H, Hara I, et al. Expression of major heat shock proteins in prostate cancer: correlation with clinicopathological outcomes in patients undergoing radical prostatectomy. J Urol. 2007;177:757-761. 22. Nishikawa M, Miyake H, Harada K, et al. Expression of molecular markers associated with the mammalian target of rapamycin pathway in nonmetastatic renal cell carcinoma: effect on prognostic outcomes following radical nephrectomy. Urol Oncol. 2014;32:49. e15-49.e21.
1430
23. Alers S, L€offler AS, Wesselborg S, et al. Role of AMPK-mTORUlk1/2 in the regulation of autophagy: cross talk, shortcuts, and feedbacks. Mol Cell Biol. 2012;32:2-11. 24. Mari M, Reggiori F. Shaping membranes into autophagosomes. Nat Cell Biol. 2007;9:1125-1127. 25. Stephenson AJ, Scardino PT, Eastham JA, et al. Preoperative nomogram predicting the 10-year probability of prostate cancer recurrence after radical prostatectomy. J Natl Cancer Inst. 2006;98: 715-717. 26. Hara T, Takamura A, Kishi C, et al. FIP200, a ULK-interacting protein, is required for autophagosome formation in mammalian cells. J Cell Biol. 2008;181:497-510. 27. Nishikawa M, Miyake H, Liu B, et al. Expression pattern of autophagy-related markers in non-metastatic clear cell renal cell carcinoma: Association with disease recurrence following radical nephrectomy. J Cancer Res Clin Oncol. in press. 28. Tang J, Deng R, Luo RZ, et al. Low expression of ULK1 is associated with operable breast cancer progression and is an adverse prognostic marker of survival for patients. Breast Cancer Res Treat. 2012;134:549-560. 29. Takechi H, Ito K, Yamamoto T, et al. Prostate-specific antigen kinetics in screen-detected prostate cancer in Japan. Urology. 2008; 72:1111-1115.
UROLOGY 85 (6), 2015