PTEN loss in biopsy tissue predicts poor clinical outcomes in prostate cancer

Transcription

1 bs_bs_banner International Journal of Urology (214) 21, doi: /iju Original Article: Clinical Investigation PTEN loss in biopsy tissue predicts poor clinical outcomes in prostate cancer Prabhakar Mithal, 1,4 Emma Allott, 2,4 Leah Gerber, 2 Julia Reid, 3 William Welbourn, 3 Eliso Tikishvili, 3 Jimmy Park, 3 Adib Younus, 3 Zaina Sangale, 3 Jerry S Lanchbury, 3 Steven Stone 3 and Stephen J Freedland 2,4 1 University of Massachusetts Medical School, Worcester, Massachusetts, 2 Duke University, Durham, North Carolina, 3 Myriad Genetics, Salt Lake City, Utah, and 4 Surgery Department, Durham VA Medical Center, Durham, North Carolina, USA Abbreviations & Acronyms ADT = androgen deprivation therapy BCR = biochemical recurrence CRPC = castration-resistant prostate cancer PC = prostate cancer PCSM = prostate cancer specific mortality PSA = prostate-specific antigen RP = radical prostatectomy TURP = transurethral resection of the prostate Correspondence: Stephen J Freedland M.D., Departments of Surgery and Pathology, Duke University Medical Center, DUMC Box 2626, Suite 457, MSRB-I, 571 Research Drive, Durham, NC 2771, USA. steve.freedland@duke.edu Received 17 December 213; accepted 17 June 214. Online publication 5 August 214 Objectives: To determine whether PTEN status in prostate biopsy represents a predictor of intermediate and long-term oncological outcomes after radical prostatectomy, and whether PTEN status predicts response to androgen deprivation therapy. Methods: In a retrospective analysis of 77 men treated by radical prostatectomy who underwent diagnostic biopsy between , biopsy samples were stained for PTEN expression by the PREZEON assay with >1% staining reported as positive. Cox proportional hazards and log rank models were used to assess the correlation between PTEN loss and clinical outcomes. Results: During a median follow-up period after radical prostatectomy of 8.8 years, 39 men (51%) developed biochemical recurrence, four (5%) had castration-resistant prostate cancer, two (3%) had metastasis and two (3%) died from prostate cancer. PTEN loss was not significantly associated with biochemical recurrence (hazard ratio 2.1, 95% confidence interval.9 5.1, P =.1), but significantly predicted increased risk of castration-resistant prostate cancer, metastasis and prostate cancer-specific mortality (all log rank, P <.1), and time from androgen deprivation therapy to castration-resistant prostate cancer (log rank, P =.3). No patient without PTEN loss developed metastases or died from prostate cancer. Conclusions: PTEN loss at the time of biopsy seems to predict time to development of metastasis, prostate cancer-specific mortality and, for the first time, castration-resistant prostate cancer and response to androgen deprivation therapy after radical prostatectomy. If confirmed by larger studies, this would support the use of PTEN loss as an early marker of aggressive prostate cancer. Key words: biochemical recurrence, castration-resistant prostate cancer, prostate cancer, PTEN, prostate cancer-specific mortality. Introduction The need for better risk stratification in PC is well established. Despite the use of PSA, Gleason score and clinical staging, a substantial number of men with PC continue to be both under-and overtreated. 1 Thus, in order to better differentiate aggressive from indolent PC, clinicians are in need of more reliable molecular markers. A growing body of evidence suggests that the PTEN gene, a tumor suppressor that regulates the PI3K/p-AKT pathway, is one of the most commonly deleted genes in PC In animal models, complete PTEN loss recapitulates the major hallmarks of aggressive human PC local tumor invasion, metastases and castration resistance. 13 The role of PTEN in human PC progression is supported by multiple studies showing PTEN loss is more frequent in metastatic CRPC compared with localized PC. 6,11,12,14,15 However, it is unknown whether PTEN loss in localized tumor can help identify which men are at increased risk of future CRPC development. To understand the prognostic significance of PTEN, multiple studies have examined RP tissue and have shown that PTEN loss in the RP specimen is prognostic for BCR, an intermediate end-point. 3,5,1,11,16,17 However, given the marked tumor heterogeneity in PC, it is unclear whether using the more limited amount of tissue available in biopsy samples can yield similar prognostic information. The one study to date that examined this found no significant association between PTEN loss and BCR The Japanese Urological Association 129

2 P MITHAL ET AL. (a) (b) Fig. 1 Images taken at 2 magnification of biopsy samples stained by the PREZEON assay (Myriad Genetics). (a) PTEN-negative prostate biopsy tissue. (b) PTEN-positive prostate biopsy tissue. Furthermore, although BCR is an end-point linked with later outcomes, 19 not all men with BCR progress. As such, it is noteworthy that the two prior studies that examined the association of PTEN loss with clinical outcomes beyond BCR found that PTEN loss in the primary resected tumor predicted the development of metastatic disease among high-risk, but not PCSM in lower-risk men. 8,9 However, this has never been tested using biopsy tissue. Collectively, the prior studies leave open two key unanswered questions: (i) Can PTEN loss in the biopsy tissue predict clinical outcomes? and (ii) Can PTEN loss predict outcomes beyond BCR? To address these questions, we carried out a pilot study using an immunohistochemistry technique previously shown to be reliable, sensitive and specific for PTEN loss to assess whether PTEN status in diagnostic prostate biopsy tissue samples is associated with clinical outcomes in PC. 2 Furthermore, given that a recent study in mice with early-stage disease and PTEN loss suggested that ADT might actually accelerate disease progression, 21 we tested whether PTEN status predicted responses to ADT. Methods Study population and tissue samples A retrospective analysis was carried out in 77 men who underwent a diagnostic biopsy between 1992 and 26 and were subsequently treated by RP at the Veterans Affairs Medical Center in Durham, North Carolina, USA. The entry criteria for the present study included treatment between 1992 and 26, treatment <2 years after biopsy, no adjuvant radiation or hormonal therapy and 3 years of follow up or BCR within that time. Clinical and pathological variables PTEN protein expression levels were assessed in formalin-fixed paraffin embedded biopsy samples by immunohistochemistry assay, PREZEON (Myriad Genetics, Salt Lake City, UT, USA). This assay uses a rabbit monoclonal antibody, which detects endogenous levels of total PTEN protein. 2 PSA (ng/ml) values were obtained from the measurement closest to the date of, but preceding RP. BCR was defined as a single PSA value >.2 ng/ ml, two values of.2 ng/ml or secondary treatment for an elevated postoperative PSA. Distant metastases, defined as bone or visceral or distant (non-pelvic) adenopathy, were determined by review of radionuclide bone scans, magnetic resonance imaging scans, computed tomography scans, plain radiograph reports and clinical progress notes. The decision to carry out radiographic imaging was at the attending physician s discretion. CRPC was defined using the Prostate Cancer Working Group 2 criteria: a 25% PSA increase from the nadir and an increase of 2 ng/ml. PCSM was defined as death in any patient with metastasis showing progression after ADT. All information relevant to study variables was ascertained from medical records. Statistical analysis The tumor was considered PTEN negative ( ) if <1% of the tumor cells stained for PTEN expression and PTEN positive ( 1 ) otherwise (Fig. 1). This method was used because it reflected the distinctly bimodal staining pattern seen. Group characteristics were compared using the two-sample t-test for normally distributed continuous variables, Wilcoxon rank-sum test for non-normally distributed continuous variables and Fisher s exact test for categorical variables. Kaplan Meier plots and the log rank test were used to assess the correlation between PTEN loss and BCR (our primary outcome). Cox proportional hazards models were used to estimate hazard ratios and 95% confidence intervals of PTEN loss in predicting BCR. As our goal was to assess whether PTEN status in the biopsy could predict outcomes, we did not adjust for pathological characteristics in the RP specimen. In secondary analysis, we evaluated the association between PTEN loss and the development of CRPC, metastasis, PCSM, receipt of ADT after RP or response to ADT. To determine how well PTEN loss could predict future receipt of ADT receipt or response to ADT, we calculated the c-index. Because of the limited number of events for the studied outcomes, only univariable analyses were carried out. Two-sided P-values <.5 were considered statistically significant. Results Clinical characteristics of patient cohort The clinical characteristics of the study population are summarized in Table 1. Of the 77 men, nine (12%) had PTEN loss. PTEN status and clinical outcomes During a median follow-up period of 8.8 years, 39 men (51%) developed BCR, 2 men (26%) went on to receive ADT, four (5%) had CRPC, and two (3%) had progressed to metastatic disease and eventually died from PC. In our primary analysis, PTEN loss was suggestively, but not significantly, associated with higher risk of BCR (HR 2.1, 95% CI.9 5.1, P =.96; Fig. 2a). After adjusting for biopsy Gleason score, the associa The Japanese Urological Association

3 PTEN loss at biopsy predicts PC outcomes Table 1 Clinical and pathological characteristics of patients in radical prostatectomy cohort Characteristic PTEN-positive PTEN-negative P-value Total n 68 9 Mean age at surgery, years (SD) 62. (±5.5) 6.6 (±6.9).39 Race, n (%).44 Non-black 32 (47%) 3 (33%) Black 36 (53%) 6 (67%) Median PSA, ng/ml (IQR) 6.8 ( ) 7.1 (2.9 1.).4 Biopsy Gleason score, n (%) (54%) 2 (22%) 7 28 (41%) 4 (44%) (4%) 3 (33%) Median percent positive biopsy tissue (IQR) (n = 48) 1 (5 2) 3 (13 58).8 Median percent positive biopsy cores (IQR) (n = 64) 45 (29 5) 5 (29 57).6 Clinical stage (n = 66).82 T1 33 (55%) 3 (5%) T2/T3 27 (45%) 3 (5%) Median surgery year (IQR) 22 ( ) 1998 ( ).4 Pathological Gleason score (19%) 1 (11%) 7 44 (65%) 8 (89%) (16%) Extracapsular extension 2 (3%) 6 (67%).3 Positive margins 45 (66%) 7 (77%).49 Seminal vesicle invasion 1 (15%) 3 (38%).12 Median follow-up years (IQR) 8.7 ( ) 9.4 ( ).55 Unpaired t-test; χ 2 -test; rank-sum. (a) Proportion without BCR Log-rank P= PTEN positive PTEN loss (c) Proportion without metastasis Log-rank P< PTEN positive PTEN loss (b) Proportion without CRPC Log-rank P< PTEN positive PTEN loss (d) Proportion surviving Log-rank P< PTEN positive PTEN loss Fig. 2 (a) BCR-free survival in RP patients grouped by PTEN status at the time biopsy. (b) CRPC-free survival in RP patients grouped by PTEN status at biopsy. (c) Metastasis-free survival in RP patients grouped by PTEN status at biopsy. (d) Prostate cancer-specific survival in RP patients grouped by PTEN status at biopsy., PTEN-positive;, PTEN loss. tion with PTEN loss remained insignificant, and the hazard ratio was decreased only slightly (HR 1.87, 95% CI , P =.196). The slight attenuation in the hazard ratio was expected given PTEN correlated with biopsy Gleason scores (P =.2) and biopsy Gleason score correlated with BCR risk (HR 1.7, 95% CI , P =.12). In secondary analysis, PTEN loss was significantly associated with increased risk of several hard clinical end-points including the development of 214 The Japanese Urological Association 1211

4 P MITHAL ET AL. (a) Proportion without ADT Log-rank P=.4 (b) Proportion without CRPC Log-rank P= PTEN positive PTEN loss PTEN positive PTEN loss Fig. 3 (a) ADT-free survival in RP patients grouped by PTEN status at biopsy. (b) CRPC-free survival in RP patients receiving ADT grouped by PTEN status at biopsy., PTEN-positive;, PTEN loss. CRPC, development of metastasis and increased risk of death from PC (all log rank, P <.1; Fig. 2b d). No patient without PTEN loss developed metastatic disease or died from PC (Fig. 2b,c). Among six men with biopsy Gleason 8 disease, the three with PTEN loss all developed CRPC, whereas all three with intact PTEN did not (log rank, P =.6). PTEN status and response to hormonal therapy A recent study suggested that in mice with PTEN loss, ADT could actually accelerate disease progression. 22 Given that PTEN appeared to be more strongly correlated with post-adt end-points (i.e. CRPC, metastasis and PCSM) than pre-adt end-points (i.e. BCR), we evaluated this hypothesis. We found that while PTEN loss significantly predicted the receipt of ADT (log rank, P =.4; Fig. 3a), it was much more strongly linked with the time from ADT to development of CRPC (log rank, P =.3; Fig. 3b). The c-index for PTEN loss to predict time from surgery to receipt of ADT was.57 versus.76 for predicting time to CRPC from ADT initiation. Discussion Although PTEN loss in RP tissue has been consistently correlated with BCR, this has not been well tested in biopsy tissue. The sole study evaluating biopsy tissue and outcomes found that PTEN loss was not significantly prognostic for BCR. 18 Therefore, we examined biopsy tissue from men undergoing RP and found that PTEN loss occurred not infrequently (12%), only suggestively predicted BCR, but strongly predicted later outcomes including CRPC, metastases and PCSM after RP. As such, this is the first study to show that PTEN loss in the biopsy tissue is a significant prognostic variable in PC. Given the small sample size, these findings require validation. However, if confirmed by larger studies, these results suggest that assessing PTEN status by immunohistochemistry could greatly improve assessment of patient prognosis at the time of biopsy, while also offering guidance regarding initial treatment selection. PTEN is one of the most potent tumor suppressors known in human cancer. 22 First identified in human cancers as a frequent deletion on chromosome 1, preclinical studies have shown PTEN to be a negative regulator of the PI3K/AKT/mTOR pathway, the activation of which is implicated in 3 5% of human PC. PTEN deletions in rodents lead directly to the development of an aggressive PC with metastases and the development of castrate resistance after ADT. 13 As a result, PTEN has been investigated increasingly in human PC tissue with studies having repeatedly shown that PTEN loss in the RP occurs at rates of 16 44% in clinically localized PC, and is associated with markers of aggressive PC including higher Gleason score, higher clinical stage, larger tumor size, extracapsular extension and seminal vesicle invasion Rates of PTEN loss are much higher in CRPC and metastases at 5 86%, and several smaller studies have even shown significantly increased loss in CRPC and metastases relative to primary tumors within a given patient. 3,5,6,1,11,16,17 These findings have bolstered interest in PTEN as an early marker of aggressive PC. Of note, in the present study the rate of PTEN loss in biopsy samples was 12%, a rate lower than in prior studies. This could simply reflect sampling error. However, given that the risk of PCSM after RP is 15% at 15 years, any marker that is altered in greater or lower than 15% of men risks over- or underestimating the true prevalence of aggressive disease, respectively. 23 As such, a 12% rate of PTEN loss approximates the prevalence necessary for a clinically useful prognostic biomarker in early stage PC. While linking PTEN loss to pathological measures of aggressiveness in localized PC or showing that PTEN loss is increased in CRPC or metastases can provide insight into PC biology, few conclusions can be drawn for the purposes of risk assessment, for several reasons. First, this information is restricted to the select group of men who choose RP. Second, linking PTEN loss with pathological features is less than ideal given that pathological features themselves are not perfect predictors of clinical outcome. 1 Third, the increased rates of PTEN loss in CRPC and metastases hint at an association with PC progression, but cannot help us understand the prognostic value of PTEN loss. Fourth, showing that PTEN loss correlates with other features of aggressive disease within the RP specimen (i.e. stage and grade) does not mean that PTEN enhances these features in terms of predicting future events (i.e. PC progression). Therefore, PTEN status must be evaluated independently of grade and stage, at earlier time-points The Japanese Urological Association

5 PTEN loss at biopsy predicts PC outcomes and in relation to future clinical outcomes. In line with this, studies examining RP tissue have consistently shown that PTEN loss independently predicts increased future risk of BCR, and that the prognostic information is improved when combined with other biomarkers, such as AKT and ERG. 3,5,1,11,16,17,24 However, just two prior studies using BCR as the primary end-point examined PTEN status in non-rp tissue. Zafarana et al. studied 126 intermediate-risk PC patients undergoing radiotherapy to assess whether PTEN loss and/or c-myc gain were predictive of BCR risk using prostate biopsy tissue. 18 They found that having either a PTEN loss or c-myc gain resulted in significantly increased BCR risk, but importantly that PTEN loss alone did not predict BCR. Reid et al. showed that in TURP specimens from 38 patients receiving no initial treatment and treated conservatively for PC, PTEN loss alone did not predict significantly increased risk of 11-year PCSM. 8 Interestingly, and analogous to the only other study to use pretreatment tissue, Reid et al. noted that PTEN became prognostic only when it was combined with a lack of ERG/ETV1 rearrangement to identify a high-risk cohort. Thus, the prior studies using biopsy or TURP tissue suggest that evaluating PTEN loss might have value. Consistent with both of these studies, we did not find PTEN loss alone to significantly predict BCR risk. Although BCR is a key determinant of PC management, and precedes CRPC, metastasis and PCSM, a significant number of patients who recur do not progress. 19 Thus, directly identifying patients most likely to progress to CRPC, metastasis and PCSM, rather than relying on BCR, should provide more accurate risk stratification. Of the four prior studies investigating PTEN in localized tumor predicting these later clinical endpoints, three have done so in high-risk cohorts. Lotan et al. showed that in a cohort with known BCR, PTEN loss in RP tissue was associated with decreased time to metastasis. 9 In that cohort, 27% of patients had Gleason scores of 8 or higher, whereas 86% had extracapsular extension. Similarly, two studies reporting PTEN as an independent predictor of PCSM did so in cohorts of CRPC patients. 11,15 That PTEN was able to predict earlier metastasis and PCSM even in high-risk cohorts suggests that it could help identify patients with particularly poor outcomes. By contrast, the one prior study of later clinical outcomes in a relatively low-risk group showed that in TURP specimens from patients receiving no initial treatment and treated conservatively for PC, PTEN loss alone did not predict increased risk of 11-year PCSM. 8 The present study is the first to show that even within the limitations of biopsy tissue, PTEN loss can predict future development of CRPC, metastasis and PCSM. The identification of PTEN loss as a predictor of CRPC is of particular importance given the poor prognosis of CRPC patients. PTEN is believed to be the most frequently deleted gene in CRPC, with rates of PTEN loss between 52 86%. 11,12,14,15 Furthermore, multiple studies have implicated PTEN loss in molecular pathways leading to CRPC, such as decreased androgen receptor expression, as well as androgen-independent cell survival. 13,25,26 PTEN loss might promote the clonal proliferation of castration-resistant tumor foci while hindering responses to ADT. Remarkably, a recent preclinical study showed that in PTEN-null mice, high-grade prostatic intraepithelial neoplasia rapidly progressed to aggressive CRPC in response to castration, whereas no such transformation occurred in castration-naive animals. 21 The implication of castration and PTEN loss together in the development of CRPC led us to hypothesize that PTEN loss could both contribute to the development of CRPC and predict which patients among those receiving ADT developed CRPC. Indeed, PTEN loss at biopsy strongly predicted reduced time to CRPC development after receipt of ADT (c-index.76). Though the interaction between PTEN loss and ADT in the development of CRPC requires further investigation and confirmation in larger studies, if the strong prognostic value of PTEN loss for predicting the development of CRPC were confirmed, it would provide a compelling argument for PTEN testing when considering the initiation of ADT. The present study had several limitations. Given that PC is a genetically heterogeneous, multifocal disease any biopsy study of PC is limited by sampling errors. Also, the number of men included in the current pilot study was modest, resulting in limited statistical power and precluding multivariable analysis. Larger studies are required to validate these findings and to test whether the prognostic information provided by PTEN is independent of other clinical characteristics something we could not test because of the small sample size. Despite this, we still were able to detect differences in PC outcomes, which are consistent with prior studies. Of note, for BCR, which had the most events, adjusting for Gleason only slightly affected the hazard ratio, suggesting it is not a major confounder of PTEN loss for BCR risk. Though PTEN loss correlated with biopsy Gleason, it did not with pathological Gleason, which could reflect either small sample size or the well-known disparity between biopsy and pathological Gleason scores. Whether similar results would be seen for the later outcomes is unknown. Finally, our cohort included only men undergoing RP, and additional studies will be required to validate these findings in other patient populations including men treated with external beam radiation or brachytherapy, or those enrolled in active surveillance. In conclusion, we show that although PTEN loss is an uncommon event in early stage PC, it was not rare (12%). For the first time, we show that PTEN loss can predict CRPC and response to ADT after RP. Furthermore, PTEN loss in the biopsy predicted time to development of metastasis and PCSM, findings consistent with prior studies. If confirmed in larger studies, these findings would support the use of PTEN loss as an early marker of aggressive PC. Conflict of interest None declared. References 1 Epstein JI, Feng Z, Trock BJ, Pierorazio PM. Upgrading and downgrading of prostate cancer from biopsy to radical prostatectomy: incidence and predictive factors using the modified Gleason grading system and factoring in tertiary grades. Eur. Urol. 212; 61: McMenamin ME, Soung P, Perera S, Kaplan I, Loda M, Sellers WR. Loss of PTEN expression in paraffin-embedded primary prostate cancer correlates with high Gleason score and advanced stage. Cancer Res. 1999; 59: The Japanese Urological Association 1213

6 P MITHAL ET AL. 3 Halvorsen OJ, Haukaas SA, Akslen LA. Combined loss of PTEN and p27 expression is associated with tumor cell proliferation by Ki-67 and increased risk of recurrent disease in localized prostate cancer. Clin. Cancer Res. 23; 9: Dreher T, Zentgraf H, Abel U et al. Reduction of PTEN and p27kip1 expression correlates with tumor grade in prostate cancer. Analysis in radical prostatectomy specimens and needle biopsies. Virchows Arch. 24; 444: Yoshimoto M, Cunha IW, Coudry RA et al. FISH analysis of 17 prostate cancers shows that PTEN genomic deletion is associated with poor clinical outcome. Br. J. Cancer 27; 97: Schmitz M, Grignard G, Margue C et al. Complete loss of PTEN expression as a possible early prognostic marker for prostate cancer metastasis. Int. J. Cancer 27; 12: Bismar TA, Yoshimoto M, Vollmer RT et al. PTEN genomic deletion is an early event associated with ERG gene rearrangements in prostate cancer. BJU Int. 211; 17: Reid AH, Attard G, Ambroisine L et al.; Transatlantic Prostate Group. Molecular characterisation of ERG, ETV1 and PTEN gene loci identifies patients at low and high risk of death from prostate cancer. Br. J. Cancer 21; 12: Lotan TL, Gurel B, Sutcliffe S et al. PTEN protein loss by immunostaining: analytic validation and prognostic indicator for a high risk surgical cohort of prostate cancer patients. Clin. Cancer Res. 211; 17: Chaux A, Peskoe SB, Gonzalez-Roibon N et al. Loss of PTEN expression is associated with increased risk of recurrence after prostatectomy for clinically localized prostate cancer. Mod. Pathol. 212; 25: McCall P, Witton CJ, Grimsley S, Nielsen KV, Edwards J. Is PTEN loss associated with clinical outcome measures in human prostate cancer? Br. J. Cancer 28; 99: Han B, Mehra R, Lonigro RJ et al. Fluorescence in situ hybridization study shows association of PTEN deletion with ERG rearrangement during prostate cancer progression. Mod. Pathol. 29; 22: De Velasco MA, Uemura H. Preclinical remodeling of human prostate cancer through the PTEN/AKT pathway. Adv. Urol. 212; 212: Article ID pages. 14 Holcomb IN, Young JM, Coleman IM et al. Comparative analyses of chromosome alterations in soft-tissue metastases within and across patients with castration-resistant prostate cancer. Cancer Res. 29; 69: Sircar K, Yoshimoto M, Monzon FA et al. PTEN genomic deletion is associated with p-akt and AR signalling in poorer outcome, hormone refractory prostate cancer. J. Pathol. 29; 218: Yoshimoto M, Joshua AM, Cunha IW et al. Absence of TMPRSS2:ERG fusions and PTEN losses in prostate cancer is associated with a favorable outcome. Mod. Pathol. 28; 21: Choucair K, Ejdelman J, Brimo F, Aprikian A, Chevalier S, Lapointe J. PTEN genomic deletion predicts prostate cancer recurrence and is associated with low AR expression and transcriptional activity. BMC Cancer 212; 12: Zafarana G, Ishkanian AS, Malloff CA et al. Copy number alterations of c-myc and PTEN are prognostic factors for relapse after prostate cancer radiotherapy. Cancer 212; 118: Freedland SJ, Humphreys EB, Mangold LA et al. Risk of prostate cancer-specific mortality following biochemical recurrence after radical prostatectomy. JAMA 25; 294: Sangale Z, Prass C, Carlson A et al. A robust immunohistochemical assay for detecting PTEN expression in human tumors. Appl. Immunohistochem. Mol. Morphol. 211; 19: Jia S, Gao X, Lee SH et al. Opposing effects of androgen deprivation and targeted therapy on prostate cancer prevention. Cancer Discov. 213; 3: Song MS, Salmena L, Pandolfi PP. The functions and regulation of the PTEN tumour suppressor. Nat. Rev. Mol. Cell Biol. 212; 13: Wilt TJ, Brawer MK, Jones KM et al.; Prostate Cancer Intervention versus Observation Trial (PIVOT) Study Group. Radical prostatectomy versus observation for localized prostate cancer. N. Engl. J. Med. 212; 367: Bedolla R, Prihoda TJ, Kreisberg JI et al. Determining risk of biochemical recurrence in prostate cancer by immunohistochemical detection of PTEN expression and Akt activation. Clin. Cancer Res. 27; 13: Uzoh CC, Perks CM, Bahl A, Holly JM, Sugiono M, Persad RA. PTEN-mediated pathways and their association with treatment-resistant prostate cancer (Review). BJU Int. 29; 14: Dong JT, Li CL, Sipe TW, Frierson HF Jr. Mutations of PTEN/MMAC1 in primary prostate cancers from Chinese patients. Clin. Cancer Res. 21; 7: The Japanese Urological Association