The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters.

Transcription

1 An Exploration of Risk Stratification for Active Surveillance and Androgen Deprivation Therapy Side Effects for Prostate Cancer Utilizing Data From the Surveillance, Epidemiology, and End Results Database The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters. Citation Accessed Citable Link Terms of Use Dinh, Kathryn Tindell An Exploration of Risk Stratification for Active Surveillance and Androgen Deprivation Therapy Side Effects for Prostate Cancer Utilizing Data From the Surveillance, Epidemiology, and End Results Database. Doctoral dissertation, Harvard Medical School. February 4, :14:23 AM EST This article was downloaded from Harvard University's DASH repository, and is made available under the terms and conditions applicable to Other Posted Material, as set forth at (Article begins on next page)

2 Abstract Part 1: Occult high-risk disease in clinically low-risk prostate cancer patients: Incidence and clinical predictors from SEER data Objective: First, to determine the incidence of pathologic upgrading and upstaging for contemporary, clinically low-risk patients and identify predictors of having occult, advanced disease to inform selection of patients for active surveillance. We will further consider the differing risk of upgrading at prostatectomy between clinically low-risk patients with!50% biopsy cores positive and other prostate cancer patients. Methods: For the first portion of the study, we identified 10,273 patients in the Surveillance, Epidemiology, and End Results (SEER) database diagnosed with clinically low-risk disease (ct1c/t2a, Prostate specific antigen (PSA)<10ng/mL, Gleason 3+3=6) in who received a prostatectomy. The primary outcome was the incidence of upgrading to pathologic Gleason score 7-10 or upstaging to pathologic T3-T4/N1 disease. Multivariable logistic regression (MVA) of men with complete biopsy data (n=5,581) identified significant predictors of upgrading or upstaging. As a second analysis, we identified 14,902 patients with prostate cancer of any risk-group diagnosed with prostatectomy who also had percent positive biopsy cores (PBC) available. Patients were categorized by NCCN clinical risk-groups, separating low-risk patients by percent PBC. We measured incidence of pathologic high-risk disease, defined as pt3a-t4 or Gleason 8-10, and used MVA to consider differing risk of advanced disease in patients with clinical low-risk disease and!50% PBC. Results: Of the first cohort, 44% of patients were upgraded and 9.7% were upstaged at prostatectomy. MVA showed age, PSA, and percent positive cores (all p<0.001), but not race, were associated with occult, more aggressive disease. With these variables dichotomized at the median, age >60 (Adjusted Odds Ratio [AOR] 1.39), PSA>5.0 (AOR 1.28), and >25% positive cores (AOR 1.76) were significantly associated with upgrading (all p<0.001). Similarly, age>60 (AOR 1.42), PSA>5.0 (AOR 1.44), and >25% positive cores (AOR 2.26) were associated with upstaging (all p<0.001). In the second cohort, 9.2% of clinically low-risk and <50%PBC, 18.6% of clinically low-risk and! #!

3 !50%PBC, and 27.6% of clinically intermediate-risk patients had occult, high-risk disease at prostatectomy (p<0.001). On MVA, low-risk with!50%pbc were more likely than low-risk with <50%PBC to have pathologic high-risk disease (AOR 2.28, 95%CI , p<0.001), had similar risk to favorable-intermediate disease overall (AOR 1.09, , p=0.33), and had higher risk than favorable-intermediate among men over 60 (AOR 1.28, , p=0.04). Conclusion: Nearly half of clinically low-risk patients harbor Gleason!7 or!pt3 disease. Percent PBC demonstrates utility for identifying a subset of NCCN low-risk patients who should have further testing before deciding on active surveillance as nearly one in five clinically low-risk prostate cancer patients with!50% positive biopsy cores harbored occult pt3a-t4 or Gleason This suggests that such patients should not be classified by national guidelines as low-risk and these patients should be made aware of this excess risk if considering active surveillance. Part 2: Association of Androgen Deprivation Therapy with Depression in Localized Prostate Cancer Using SEER-Medicare Objective: Androgen deprivation therapy (ADT) may contribute to depression, yet several studies!have not demonstrated a link. We aimed to determine if receipt of any ADT or longer duration of ADT for prostate cancer is associated with increasing risk of depression. Methods: We identified 78,552 men over 65 with stage I-III prostate cancer using the Surveillance, Epidemiology, and End Results-Medicare linked database from , excluding patients with psychiatric diagnoses within the prior year. Our primary analysis was the association of pharmacologic ADT with the diagnosis of depression, or receipt of inpatient- or outpatient-psychiatric treatment, using Cox-proportional hazard regression. Drug-data for treatment of depression was not available. Our secondary analysis was association of duration of ADT with each endpoint. Results: Overall, 43% (33,882) of patients received ADT and had higher 3-year cumulative incidence of depression (7.1% vs. 5.2%), inpatient- (2.8% vs. 1.9%), and outpatient-psychiatric treatment (3.4% vs. 2.5%) than patients without ADT (all! $!

4 p<0.001). Adjusted cox-analyses demonstrated patients with ADT had a 23% increased risk of depression (adjusted hazard ratio [AHR]=1.23,95%CI=[ ]), 29% increased risk of inpatient-psychiatric treatment (AHR=1.29,95%CI=[ ]), and a non-significant 7% increased risk of outpatient-psychiatric treatment (AHR=1.07,95%CI=[ ]) compared to patients without ADT. The risk of depression increased with duration of ADT, from 12% with "6, 26% with 7-11, to 37% with!12 months (p-trend<0.0001). Similar duration-effect was seen for inpatient- (ptrend<0.0001) and outpatient-psychiatric treatment (p-trend<0.0001). Conclusion: Pharmacologic ADT increased the risk of depression and inpatientpsychiatric treatment in this large study of elderly men with localized prostate cancer. This risk increased with longer duration of ADT. The possible psychiatric effects of ADT should be recognized by physicians and discussed with patients prior to initiating treatment.! %!

5 Table of Contents Abstract 2 Table of Contents 5 Glossary 6 Introduction 7 Part Materials and Methods 11 Results 12 Discussion 14 Conclusion 17 Part Materials and Methods 18 Results 20 Discussion 22 Conclusion 25 Part 1 Suggestions for Future Work 26 Part 2 28 Materials and Methods 28 Results 31 Discussion 33 Conclusion 36 Part 2 Suggestions for Future Work 37 Summary 39 Acknowledgements 40 Personal Contribution to this Work 40 References 41 Tables & Figures 53!! &!

6 Glossary AA: ADT: ADT+: ADT-: AHR: AOR: AS: bgs: CCI: CI: ct: MAO-A: MRI: MVA: NCCN: PBC: PCa: PCSM: pgs: PSA: pn: PNI: pt: RCT: RP: RR: RT: SEER: US: African American Androgen deprivation therapy Cohort that received ADT Cohort that did not receive ADT Adjusted hazard ratio Adjusted odds ratio Active surveillance Biopsy Gleason score Charlson comorbidity index Confidence interval Clinical T category Monoamine oxidase-a Magnetic resonance imaging Multivariable logistic analysis National clinical cancer network Positive biopsy cores Prostate cancer Prostate cancer specific mortality Pathologic Gleason score Prostate specific antigen Pathology N category Perineural invasion Pathologic T category Randomized control trial Radical prostatectomy Relative risk Radiation therapy Surveillance, Epidemiology, and End Results United States! '!

7 Introduction In 2015, an estimated 220,000 men will be diagnosed with prostate cancer (PCa) in the United States (US) and over 2.7 million will be living with this disease.(1) While PCa is the most common cancer diagnosis in men, the 5-year overall survival rate is greater than 99%,(1) suggesting that many cases of PCa detected may not lead to clinically significant disease during a man s lifetime. Men are commonly diagnosed through the screening test approved by the FDA in 1994, involving both a blood test for prostatespecific antigen (PSA) in combination with a digital prostate exam.(2) In the early era of PSA screening, numerous studies compared definitive treatment to observation or conservative management and produced mixed results,(3-5) specifically raising questions around the utility of treating low-risk PCa or PCa in patients 65 years and older. These early studies contributed further to the notion that screening detects many insignificant cases of PCa. The consequences of diagnosing clinically insignificant PCa are numerous, including harmful treatment side effects, waste of medical resources, and patient and family distress around PCa diagnosis and prognosis. Due to the concern that many PCa patients may often be over-treated, there is increased interest in selecting appropriate patients for active surveillance (AS) to reduce unnecessary treatment.(3, 4, 6, 7) AS is a treatment plan in which qualifying patients are followed with routine PSA testing and repeat prostate biopsies with guidelines for reclassification and initiation of definitive treatment. AS differs significantly from watchful waiting, in which patients are watched and treated only with androgen deprivation therapy (ADT) when symptomatic from PCa. Reviews of AS selection criteria have demonstrated that current practice is highly varied by institution and study design and many criteria are not sensitive or specific for insignificant disease,(8, 9) defined as Gleason"6 and organ-confined disease (T1c-T2c). Criteria for AS could be improved by identifying certain diagnostic features that place clinically low-risk men, defined as PSA<10, T1c-T2a, and Gleason 6, at increased risk of harboring occult Gleason 7-10 or T3-T4 disease. Past studies have shown a disparity between biopsy Gleason score and clinical stage with prostatectomy Gleason score and pathologic stage;(10-18) however, many of these studies included all risk-groups, limiting their application to current AS-! (!

8 eligible patients. Further work is necessary to understand the incidence of upgrading to Gleason 7-10 and upstaging to T3-T4 disease at prostatectomy among contemporary, low-risk patients and how understanding this could help to identify those at risk of having occult, more advanced disease and inform AS selection criteria. The purpose of part one of this thesis is three-fold. First, we used the Surveillance, Epidemiology, and End Results (SEER) database,(19) including newly available data on biopsy and pathologic Gleason sum and score, clinical T stage, and pathologic T stage, to estimate the prevalence of upgrading and upstaging among contemporary low-risk PCa patients diagnosed in and treated with prostatectomy. Second, we used a multivariable logistic analysis (MVA) to identify features associated with increased risk of harboring more advanced disease in these patients. These factors may help identify low-risk PCa patients with concerning features who should have further evaluation prior to selection for AS, such as advanced prostate imaging or additional magnetic resonance imaging (MRI)-guided biopsies. This work was published in Journal Urology in 2015 as Incidence and Predictors of Upgrading and Upstaging Among 10,000 Contemporary Patients with Low-risk Prostate Cancer,(20) which will be included in part one of this thesis. This work demonstrated that increasing percent positive biopsy core (PBC) as significantly associated with higher risk of upgrading or upstaging among low-risk PCa. In fact, we found that patients with >50% PBC and clinical low-risk disease were upgrading in 55.2% of cases and upstaged in 18.7%,(20) which are significant numbers. This led us to investigate what is understood about the disparity between clinical riskstratification and pathological risk after prostatectomy and what easily measurable clinical-factors, such as percent positive biopsy cores (PBC), PSA density, percent of cancer present in a single biopsy core, and PSA density, may improve the accuracy of risk-stratifying newly diagnosed PCa patients. Current risk-stratification guidelines for PCa from the National Clinical Cancer Network (NCCN) do not include percent PBC;(21) therefore, patients with clinical T1c-T2a, biopsy Gleason "6, and initial PSA "10.0 ng/ml are considered NCCN low-risk, even with a high percent PBC. Prior studies have shown that an increasing percent PBC is significantly associated with! )!

9 worse outcomes among intermediate-risk patients,(22, 23) with studies often dichotomizing patients at PBC of 50%.(22-25) However, this work is not generalizable to low-risk patients eligible for AS as they were not included in the study. Yet, the results were significant enough to lead to a recent proposal for future NCCN guidelines to create subgroups of intermediate-risk patients, including stratification by 50% PBC.(23) Percent PBC has been shown to impact outcomes after prostatectomy in low-risk patients in single-institution studies, with the worst outcomes for low-risk patients with >50% compared to <34% or 34-50% PBC,(16) but the correlation of percent PBC with pathologic outcomes has not been thoroughly considered in a population-based dataset. Therefore, based on prior work done in intermediate risk patients on percent PBC and our own finding that increasing percent PBC is associated with increasing rates of upgrading and upstaging among low-risk PCa patients, we also aim to compare pathologic outcomes at radical prostatectomy of low-risk patients with!50% PBC to other low- or intermediate-risk patients using SEER. This work will help inform how percent PBC could be included in future risk-stratification guidelines for prostate cancer. This work was published in Urology in 2015 as Disease in Clinically Low-Risk Prostate Cancer with 50% Positive Biopsy Cores: Should National Guidelines Stop Calling Them Low-Risk?,(26) which will also be included in part one of this thesis. On the other end of the spectrum, 38 percent of newly diagnosed PCa patients have high-risk disease at diagnosis.(27) Following multiple randomized controlled trials demonstrating improved survival with ADT alongside radiation for high-risk and locally advanced PCa,(28-34) the proportion of PCa patients initiated on ADT within twelve months of diagnosis increased to nearly 50% by 2002.(35) Though ADT has significant survival benefits, mounting evidence has demonstrated an extensive side effect profile, including metabolic, cardiovascular, bone, and cognitive effects.(36) There is reason to believe that ADT may negatively impact mood, causing clinically significant depression. While several small studies have not identified any association between ADT and clinical depression,(37-42) possibly due to inadequate power, others have found a significant association;(43-45) yet there remains no consensus on the subject. We aim! *+

10 to investigate the relationship between ADT and depression further in part two of this thesis, using the SEER database linked to Medicare claims to measure the association between the receipt of pharmacologic ADT and depression diagnosis as well as increased psychiatric utilization in men with localized PCa. In addition, we will specifically consider the effect of ADT duration on these outcomes. This work is accepted at the Journal of Clinical Oncology as Association of Androgen Deprivation Therapy with Depression in Localized Prostate Cancer, which will be included in part two of this thesis.! **

11 Part 1.1 Incidence and Predictors of Upgrading and Up Staging among 10,000 Contemporary Patients with Low Risk Prostate Cancer. Dinh KT, Mahal BA, Ziehr DR, Muralidhar V, Chen YW, Viswanathan VB, Nezolosky MD, Beard CJ, Choueiri TK, Martin NE, Orio PF, Sweeney CJ, Trinh QD, Nguyen PL The Journal of Urology. 2015;194(2): doi: /j.juro PubMed PMID: Materials and Methods Study cohort SEER is a population-based cancer registry, sponsored by the US National Cancer Institute, that collects demographic characteristics and cancer incidence, treatment, and survival data for approximately 28% of the US population.(19) We identified men in SEER diagnosed in with low-risk, histologically confirmed prostate adenocarcinoma primarily treated with prostatectomy. Low-risk PCa was defined as pretreatment PSA<10.0ng/mL, biopsy Gleason 3+3, and clinical T1c-T2a. The study included only to utilize new, PCa-specific variables in SEER, including total number of biopsy cores and number of positive cores. We used SEER*Stat to access data. Our initial cohort was 10,478 men, excluding patients diagnosed at autopsy or by death certificate, whose pathologic data was from autopsy, and those who received neoadjuvant chemo-radiation. Patients with incomplete data for biopsy Gleason sum, clinical T stage, pathologic Gleason sum, or pathologic T stage were excluded (205). Our final cohort was 10,273, of which 5,581 patients had complete data for the number of biopsy and positive cores. Demographic data included age at diagnosis, race, and marital status. Clinical data included pre-treatment PSA, biopsy Gleason score and sum, clinical T stage, number of biopsy cores, number of cores positive for PCa. The percent positive biopsy cores! *"

12 (PBC) was defined as the number of positive cores divided by the total number of cores biopsied. Pathologic data included pathology Gleason score and sum (pgs), pathology T (pt) and N (pn) stage, and tumor size (mm). Tumor size was available for 3,615 patients. Statistical analysis Demographic, clinical and pathologic characteristics of our sample were described. Upgrading was defined as pgs 7-10 and upstaging as pt3-t4/n0-1. We further stratified pgs 3+4 vs!4+3 and pt3a vs pt3b-4/n0-1 to define aggressive, locally advanced disease. Our primary outcome was the proportion of patients upgraded or upstaged at prostatectomy. We also calculated the rate of adverse tumor size, defined as tumors over 20mm. Our second outcome was to identify demographic and clinical characteristics associated with upgrading or upstaging by MVA using the subset of patients with complete biopsy data (n=5,581). Age, PSA and percent PBC were treated as continuous variables in our initial models. Race was defined as black or non-black and marital status as married or not married. In a secondary MVA (n=5,581), all variables were binary; age and PSA were dichotomized near the median, age "60 or >60, PSA "5.0 or >5.0, and percent PBC was "25% and >25%. Finally, we used two factors associated with upgrading and upstaging to create a stratified risk-table. All P values reported were two-sided with significance at p<0.05. We used STATA (version 11.1; StataCorp LP, College Station, TX) for statistical analysis. The Institutional Review Board at the study institution approved this study; a waiver for informed consent was obtained. Results Baseline demographic and clinical characteristics Table 1 lists baseline demographic, clinical, and pathologic characteristics for the total cohort. The median age was 60 years (range 34-92). Most patients were married (77.7%) and white (83.8%), with 1,215 African American patients (AA, 11.8%). The! *#

13 median PSA was 5.0ng/mL (range ), and 92.7% of patients were ct1c vs 7.3% ct2a. The median tumor size was 13mm (interquartile range 7-18mm). Incidence of upgrading and upstaging Forty-four percent (4,467) of patients in our cohort were upgraded to pgs 7-10 (Figure 1). Of these, 86.2% (3,842) were pgs 3+4=7 disease, 10.6% (473) were pgs 4+3=7 disease, and 1.3% (137) were pgs 8-10 disease. 9.7% (992) of patients were upstaged to pt3-4/n0-1 (Figure 2). Of these, 88.4% (815) were pt3a, 14.4% (133) were pt3b, 1.4% (14) were pt4, and 3.0% (30) had N1 disease. Tumor size was >20mm in 16.4% (593/3615) of patients, of whom 66.4% were upgraded and/or upstaged. In total, 39.0% (4002) of patients had pgs 3+4 without upstaging or pt3a with pgs "3+4, 5.9% (611) of patients had pgs!4+3 disease, and 1.7% (177) of patients had pt3b-4/n0-1. Fifty-four percent of patients (5538) had low-risk disease with pgs 3+3 and pt2.(8) Factors associated with upgrading and upstaging MVA of patients with complete biopsy data (n=5,581) found age, PSA, and percent PBC were associated with upgrading (all p<0.001, Table 2) and upstaging (all p<0.001, Table 2). These variables remained statistically associated when upgrading and upstaging were restricted to pgs!4+3 and pt3b-4/n0-1, respectively. Black race was not significantly associated with upgrading (p= 0.58) or upstaging (p=0.65), nor was marital status. Only percent PBC was significantly associated with having tumor over 20mm at prostatectomy (adjusted odds ratio [AOR] 5.96, 95%CI ). A second MVA (n=5,581) using dichotomized variables found that age>60 (AOR 1.39, ), PSA>5.0 (AOR 1.28, ), and >25% positive cores (AOR 1.76, ) were significantly associated with upgrading (all p<0.001, Table 2). Similarly, age>60 (AOR 1.42, ), PSA>5.0 (AOR 1.44, ), and >25% positive cores (AOR 2.26, ) were significantly associated with upstaging (all p<0.001, Table 2). These variables remained statistically associated when the outcome was restricted to only pgs!4+3 or pt3b-4/n0-1.! *$

14 Impact of percent positive biopsy cores for cancer Since 2010, SEER has included data on the number of biopsy cores and positive cores, which was only available for 54.3% of our cohort (n=5,581), and we found percent PBC was significantly associated with both upgrading and upstaging in this study. When the patients were stratified by percent PBC, the incidence of upgrading and upstaging ranged from 32% and 5.4% when "12.5% positive cores to 55% and 19% when >50% positive cores (Figure 3), demonstrating a positive correlation between percent PBC and increased risk of upgrading or upstaging. The number of biopsy cores ranged from 1 to >99 cores in this cohort with 63.8% receiving a standard 8-12 core biopsy,(46, 47) and only 5.3% receiving a sextant biopsy. Yet, the rate of upgrading and upstaging between these two groups were similar at 43.2% versus 44.9% and 10.7% versus 10.5%, respectively (p=0.56, p=0.90). Using PSA and percent positive biopsy cores for an individual patient s risk stratification We identified age, PSA, and percent PBC as the strongest clinical variables associated with upgrading or upstaging. Of these variables, we selected PSA and percent PBC to create a risk stratification table for upgrading or upstaging in our sample with complete biopsy data (n=5,581, table 3 & 4). We found a patient with PSA"2.5 and "12.5% PBC had a 13.6% risk of upgrading and 1.69% of upstaging compared to 61.8% and 19.1%, respectively, when PSA and >50% PBC. This table is color-coded for low, intermediate, and high risk of either upgrading or upstaging. Discussion In a cohort of over 10,000 patients in SEER with clinically low-risk prostate cancer (ct1c-t2a, PSA<10, and Gleason 3+3) treated with prostatectomy, we found that nearly half harbored pgs!7 or!pt3 disease. Specifically, 44% (4468) of these patients were upgraded to pgs 7 to 10 (5.9% pgs!4+3) and 9.7% (992) were upstaged to pt3- T4/N0-1 disease (1.7% pt3b-4/n0-1). Older age, higher PSA, and higher percent PBC were significantly associated with upgrading and upstaging (n=5,581, all p<0.001). To our knowledge, this is the largest population-based study of upgrading and upstaging exclusively in clinically low-risk PCa patients.! *%

15 Concerns about overtreatment have prompted adoption and encouragement of active surveillance in recent years. Our findings demonstrate that a surprising proportion of clinically low-risk patients eligible for AS may have occult, more advanced disease. While NCCN guidelines for AS eligibility include all clinically low-risk and some intermediate risk patients, some have advocated for even stricter criteria, particularly for younger patients. Specifically, previous work has shown that a subgroup of patients with very low risk disease (n=153), defined as ct1c, fewer than two positive biopsy cores, PSA-density <0.15 ng/ml/gm, and <50% involvement of a single core positive, had better pathologic outcomes at prostatectomy than patients with traditionally-defined lowrisk disease (n=7333). (48) While limiting AS purely to patients with very low-risk disease might overly limit the number of patients eligible for AS, our work supports the need to further risk-stratify among low-risk patients. As debates continue on appropriate criteria for AS, this paper serves as an important reminder that nearly 40% of biopsy Gleason 3+3 and ct1c-t2a patients who would generally qualify for AS may have pgs 3+4 disease and an additional 7% percent may be harboring pgs!4+3 or pt3b-4/n0-1 disease. The challenge for clinicians is to identify these patients prior to the initiation of AS, which current criteria has proven neither sensitive nor specific enough to do.(8) We developed a modified riskstratification table using PSA and percent PBC (Table 5 & 6) that may help identify patients at risk for upgrading or upstaging who would benefit from further work-up prior to AS, such as advanced imaging or MRI-targeted biopsy. (9, 49-53) It would not be prudent to perform advanced imaging on all low-risk patients; however, our results propose clinical features that could justify further work-up. For example, a patient with PSA and >25% PBC may have a 60% chance of harboring higher grade disease. It is important to remember that some older, low-risk patients who display clinical characteristics suggestive of more advanced or aggressive disease may be appropriate for active surveillance without further work-up, when considering their life expectancy, comorbidities, and potential negative effects of treatment. Prospective single-arm trials of AS generally demonstrated low rates of PCa specific mortality (PCSM, 0-3%) with 5-10 years follow-up, and randomized trials have mixed! *&

16 results when comparing overall mortality for RP and watchful waiting (3, 4, 6, 7). Therefore, AS may be a reasonable treatment for most low-risk patients; however, 14-41% of patients in prospective AS studies progressed to active treatment with intermediate follow-up. In the Klotz phase II study, 50% of patients who progressed to treatment on AS eventually recurred, raising the questions of whether they harbored more aggressive disease at presentation and whether treating that disease earlier could have altered their outcome.(54) The risk tables presented here can be used to help identify patients at increased risk of advanced disease before initiating AS for further work-up and, when appropriate, to give them the choice of being treated earlier, rather than when they progress. This paper is limited in reporting oncologic outcomes due to follow-up times of less than two years, as well as the lack of biochemical and progression free survival data reported by SEER. However, a previous, smaller study showed patients upgraded at prostatectomy had a relative risk (RR) of 1.86 ( ) for biochemical progression at six years,(16) and another showed patients upgraded from Gleason 6 to pgs 3+4 or 4+3 at prostatectomy had progression free survival similar to patients with biopsy Gleason 3+4 and 4+3 disease, respectively, at five years.(13) These and other prior, smaller studies, with rates of upgrading from 18-56%(10-18) and upstaging from 24-42%, 13,14,(55) are less applicable to modern AS-eligible populations because they often included all risk groups, used pre-2005 modification Gleason scores(56), and relied upon sextant prostate biopsy schemes. All patients in this present study were evaluated after the 2005 modification and 83% had an eight to 18 core biopsy. It is possible that the varying experience of pathologists represented in SEER, as compared to single institution studies with mandatory uropathology review, could explain higher rates of upgrading in the present study. For example, Fine et al found that correlation of biopsy and pathologic Gleason scores was lower at outside institutions (69.7%) than at John s Hopkins (75.9%, p=0.0002). 6 Notably, race was not significantly associated with upgrading or upstaging in our analysis (0.54, 0.65, respectively), unlike the landmark paper of 1,801 men treated with prostatectomy.(57) The reason for this discrepancy is unclear, but it could reflect the! *'

17 difference between a single-center study versus a geographically and institutionally diverse sample. Additionally, this study included a larger fraction of AA patients (11.8% vs 9.1%) and nearly five-fold more (1,215 vs 256 patients) than in the study published by Sundi et al.(57) There are some potential limitations to this study. First, we cannot know what factors influenced selection of these patients for prostatectomy over other treatment options, including AS. Further imaging after the initial biopsy, high PSA velocity, or positive family history may have influenced treatment decisions, but these variables are not included in SEER. Therefore, the incidence of upgrading and upstaging presented here may overestimate the true value for all clinically low-risk patients. However, given that the majority of low-risk PCa patients received treatment rather than surveillance in 2010 and 2011, the sample studied is likely representative of current low-risk patients. (58) Second, SEER does not report the percent of PCa in a biopsy core, which is often used to consider a patient s risk and is part of the European Association of Urology guidelines for selecting patients for AS. (8, 9, 46) Similarly, SEER does not specify prostate biopsy technique, such as if ultrasound guided or MRI-targeted, nor if a patient received a repeat biopsy prior to selecting a treatment, limiting our inclusion of biopsy technique into our risk model. Finally, the SEER database does not include other factors previously associated with upgrading or upstaging in smaller studies, including maximum percent of cancer in a single core, PSA density or velocity, obesity, and delay between biopsy and prostatectomy.(14-18) Conclusions In this study of over 10,000 contemporary, clinically low-risk prostate cancer patients eligible for active surveillance, 44% (4468) of these patients were upgraded to pathologic Gleason 7 to 10 (5.9% pgs!4+3) and 9.7% (992) were upstaged to pt3- T4/N0-1 disease (1.7% pt3b-4/n0-1) at prostatectomy. Low-risk patients should be considered for further diagnostic imaging or guided biopsies based on PSA and percent positive cores prior to choosing active surveillance.! *(

18 Part 1.2 Occult High-Risk Disease in Clinically Low-Risk Prostate Cancer with!50% Positive Biopsy Cores: Should National Guidelines Stop Calling Them Low-Risk? Dinh KT, Muralidhar V, Mahal BA, Chen YW, Nezolosky MD, Beard CJ, Choueiri TK, Martin NE, Orio PF, Sweeney CJ, Trinh QD, Nguyen PL Urology doi: /j.urology PubMed PMID: Material and Methods Study cohort The US National Cancer Institute sponsors the SEER database, which includes 18 regions or 28% of the US population.(59) The SEER database includes demographics, incidence, treatment, and survival outcomes for patients with cancer. We identified men in SEER diagnosed in with histologically confirmed prostate adenocarcinoma treated primarily with prostatectomy. The study was restricted to diagnoses in , as total number of biopsy cores and number of positive cores became available in We used SEER*Stat to access data. Our initial cohort was 51,301 men, excluding patients diagnosed at autopsy or by death certificate, whose pathologic data were from autopsy, and those who received neoadjuvant chemo-radiation. Patients who could not be categorized by NCCN risk groups, including patients with clinical T4 disease, N1, or M1 disease, as well as patients with incomplete data for PSA (ng/ml), biopsy Gleason sum, or clinical T stage were excluded (23,885). In addition, patients were excluded if they had unclear pathologic T stage, Gleason sum, or Gleason score (574). Finally, patients were excluded if incomplete biopsy core data (11,940), including both the total number of biopsy cores taken and the number of cores positive for cancer, leaving 14,902 patients in the final cohort. Demographic data included age at diagnosis and race, including White, Black, Asian, Native American/Alaskan Native, and Hispanic. Clinical data included pre-treatment! *)

19 PSA, biopsy Gleason score and sum (bgs), clinical T category (ct), number of biopsy cores, and number of cores positive for PCa. Percent positive biopsy cores (PBC) was defined as the number of positive cores divided by the total number of cores taken at biopsy. Pathologic data included prostatectomy Gleason score (pgs) and sum and pathologic T and N category (pt, pn). Statistical analysis Patients were initially categorized by NCCN-defined risk-groups, low, intermediate, and high-risk, with low-risk further divided into low-risk with <50% PBC and low-risk with!50% PBC. Patients were then further categorized into favorable and unfavorable intermediate risk, as described by Zumsteg et al.(23) Favorable-intermediate-risk was defined as one intermediate-risk factor, including ct2b-c, PSA ng/ml, or cgs 3+4=7, and <50% PBC, while unfavorable-intermediate risk was defined as bgs 4+3=7 with otherwise low-risk features, more than one intermediate-risk factor as described above, or!50% PBC with no high-risk features (ct3, PSA>20.0, or bgs 8-10). Demographic, clinical and pathologic characteristics of our sample were described and compared between risk-groups, including favorable- and unfavorable-intermediate risk. Pathologic data were used to categorize patients into three categorical outcomes: pt2a and pgs"6, pt2b-c and pgs"7, or pt3a-4 and/or pgs Our primary outcome was the proportion of patients with each outcome, stratified by NCCN risk-group and low-risk with <50% PBC and low-risk with!50% PBC. The incidence of occult, pathologic highrisk disease, defined as pt3a-4 or pgs 8-10, was then further stratified by age. Tumor size in mm was compared among these clinical-risk groups as a second pathologic outcome. Finally, MVA was used to measure the association between percent PBC and occult, high-risk pathology among NCCN low and intermediate PCa patients and to compare the risk of having pathologic high-risk disease by NCCN risk group, adjusting for age and race. Some analyses were repeated with Zumsteg s adjusted-risk groups, including favorable- and unfavorable-intermediate risk. All P values reported were two-sided with significance at p<0.05. We used STATA (version 11.1; StataCorp LP, College Station, TX) for statistical analysis. The! "+

20 Institutional Review Board at the study institution approved this study; a waiver for informed consent was obtained. Results Baseline demographic and clinical characteristics Of the overall cohort, 5,607 were NCCN low-risk (37.6%) of whom 1,122 or 20% of lowrisk had!50% PBC. Table 1 lists baseline demographic, clinical and pathological characteristics for the overall cohort, stratified by NCCN low-risk with <50% PBC vs.!50% PBC, favorable-intermediate risk vs. unfavorable-intermediate risk, and NCCN high-risk. Low-risk patients with <50% vs.!50% PBC had similar median PSA (5.0 vs. 5.1 ng/ml, p=0.22) and age (60 vs. 59 years, p=0.11), but low-risk with <50% PBC had more ct2a disease (7.7% vs. 6.0%, p=0.048) and had a higher proportion of nonminority patients (74.2 % vs. 70.2%, p=0.030). In comparing all five risk-groups, increasing risk-group was associated with older age (p<0.001). Clinical T category, bgs, and percent PBC were different between the groups by definition of the clinicalrisk categories. Incidence of occult, pathologic T3-4 or GS 8-10 Figure 1 shows the incidence of each pathologic categorical outcome by NCCN risk group with division of low-risk by percent PBC. The incidence of pathologic high-risk disease (pt3a-t4, or Gleason 8-10) was 9.2% in low-risk with <50% PBC, 18.6% in low-risk with!50% PBC, 27.6% in intermediate-risk, and 65.8% in high-risk (p<0.001). With further stratification of NCCN intermediate-risk into favorable- and unfavorableintermediate risk, the incidence of pathologic high risk disease (pt3a-4 or pgs 8-10) was 9.2% in low-risk with <50% PBC, 18.6% in low-risk with!50% PBC, 18.2% in favorable-intermediate risk, and 34.2% in unfavorable-intermediate risk (p<0.001, Figure 1). The association between percent positive biopsy core and high-risk pathology! "*

21 The association between increasing percent PBC and high-risk pathology was analyzed using an MVA within NCCN low-risk and intermediate risk-groups, adjusting for age, race, initial PSA, and ct category. This analysis demonstrated that percent PBC was associated with high-risk pathology among low risk patients (AOR per 1% increase PBC, , p<0.001) and among intermediate-risk patients (AOR 1.016, , p<0.001). When repeating the analysis using percent PBC dichotomized at 50%, the AOR of high-risk pathology was 2.29 for low-risk and 2.09 for intermediate-risk disease (all p<0.001). Risk of occult, high-risk disease in low-risk patients with!50% PBC MVA were performed to measure the risk of having occult, pt3a-4 or pgs 8-10, by risk group (Table 2). Each analysis was adjusted for age and race. Low-risk with!50% PBC had an increased risk of occult, pathologic high-risk disease than low-risk with <50% PBC (AOR 2.28, 95% Confidence interval , p<0.001), similar risk to favorableintermediate risk (AOR 1.09, 95%CI , p=0.33), and lower risk than unfavorable-intermediate risk disease (AOR 0.47, 95%CI , p<0.001). We conducted a sensitivity analysis excluding patients who may have had an MRI target biopsy, which we defined as having <6 cores, or a saturation biopsy, which we defined as having >18 cores, and found that low-risk with!50% PBC had an increased risk of occult, pathologic high-risk disease than favorable-intermediate risk disease (AOR 1.25, 95%CI , p=0.026, n=13,478). Pathologic tumor size The median tumor size at prostatectomy was 13mm for low-risk with <50% PBC, 16mm for low-risk with!50% PBC, 15mm with favorable-intermediate risk, 17mm for intermediate-risk, 18.5mm with unfavorable-intermediate risk, and 20mm for high-risk disease (p<0.001). Mean pathologic tumor size (mm) was larger for low-risk with!50% PBC than low-risk with <50%PBC (21.3 vs mm, p<0.0001) and nearly significantly larger than NCCN intermediate-risk patients (21.3 vs. 19.7mm, p=0.083). Mean tumor size was also larger for low-risk with!50% PBC than favorable-intermediate risk (21.3! ""

22 vs. 17.8mm, p<0.0001) and similar to unfavorable-intermediate risk patients (21.3 vs mm, p=0.82). Interaction between age and pathologic risk Within each clinical risk group, the incidence of occult, pathologic high risk disease approximately doubled for men age 70 and older compared to men age less than 50, increasing from 6.3 to 14.9% in low-risk with <50% PBC, 11.0 to 29.0% in low-risk with!50% PBC, 14.2 to 23.7% in favorable-intermediate risk, 17.9 to 34.5% in all intermediate-risk and 20.70% to 40.9% in unfavorable-intermediate risk (all p<0.002, Figure 2). There was a significant interaction for age and the relationship between lowrisk with!50% PBC and favorable-intermediate risk when considering the risk of occult, pathologic high-risk (p=0.044), such that low-risk with!50%pbc disease appeared to harbor more occult disease than favorable-intermediate risk for men aged 60 and older (AOR 1.28, 95%CI , p=0.048) but not necessarily among men younger than 60 years (AOR 0.91, 95%CI , p=0.50). Finally, among men age 70 and over, the incidence of occult high-risk disease was not significantly different when comparing lowrisk with!50% PBC to all intermediate-risk patients (29.0% vs. 34.5%, p=0.36). Discussion Nearly one in five patients with!50% PBC and NCCN low-risk features harbored occult, high-risk (pathologic T3a-4 or GS 8-10) disease at prostatectomy in this large population-based study of nearly 15,000 patients with localized PCa in The risk of occult, high risk disease in patients with low-risk and!50% PBC was higher than low-risk with <50% PBC and similar to patients classified as favorable-intermediate risk, as defined by Zumsteg et al, though the risk appears higher than favorable-intermediate risk for men older than 60, who represent the majority of men with prostate cancer.(23) Additionally, patients with low-risk and!50% PBC had larger tumor size at prostatectomy than favorable-intermediate risk, and similar tumor size to patients with unfavorable-intermediate risk. These results suggest that national guidelines should not categorize patients with clinically low-risk features and!50% cores positive as lowrisk.! "#

23 Our results question whether patients with low-risk and!50% PBC are appropriately treated by NCCN low-risk guidelines for low-risk prostate cancer. Furthermore, this pathologic evidence suggests that these patients may benefit from more aggressive treatment options defined for intermediate-risk patients. For example, for patients with low-risk and!50% PBC, NCCN guidelines state that observation is mandatory for those with a <10 year life-expectancy and active surveillance tends to be encouraged for those with longer life-expectancies. However, given that one in five men with low-risk and!50% PBC harbor pathologic high-risk disease, many healthy men in this category may be appropriately treated with more definitive therapy, as typically recommended for intermediate-risk patients.(21) These results should also encourage physicians treating NCCN low-risk and!50% PBC to consider further diagnostic imaging, such as a MRI, (60) to search for evidence of extra-capsular extension, seminal vesicle invasion, or foci of high-grade disease prior to selecting active surveillance or observation according to current treatment guidelines. There are plausible reasons why percent PBC could be a useful factor for risk-stratifying prostate cancer patients. Higher percent PBC was independently associated with an increased risk of pt3a-4 or pgs 8-10 disease and correlated with pathologic tumor size presently and in prior literature.(61, 62) Smaller studies with longer follow-up have found percent PBC to correlate with worse biochemical recurrence(16, 25, 63, 64) and prostate cancer specific mortality.(22, 23, 65-67) Increased tumor burden is a risk factor for occult, high-risk disease in prostate cancer patients as a larger tumor has more opportunity to invade the capsule or develop foci of high-grade disease that may be missed on biopsy. Increasing tumor size has been correlated with worse prostate cancer specific outcomes in prior studies,(63) as large tumors require more aggressive, local control. Finally, higher percent PBC may also represent multiple foci of disease within the prostate. Despite standard biopsy sampling schema of six locations bilaterally, multifocal disease may not be adequately sampled, leaving high-grade disease possibly undetected.(68) Our results provide additional evidence to suggest that NCCN guidelines and active surveillance (AS) criteria should include percent PBC in prostate cancer risk-! "$

24 stratification schema.(16-18, 22, 23, 63, 65, 69) Patients classified as NCCN low-risk have heterogeneous pathologic findings at prostatectomy,(16, 25, 62, 69) which was also demonstrated in the present cohort. By stratifying low-risk patients by percent PBC in this cohort, one in five with!50% harbored pt3a-4 or pgs8-10 disease compared to only one in ten patients with low-risk and <50% (p<0.001). These two groups could not be distinguished by current NCCN risk-factors, as they had similar initial serum PSA and biopsy Gleason score, and patients with <50% PBC had more ct2a disease, which suggests that percent PBC may correlate better with pathologic outcomes than clinical T category(70) or initial PSA. Presently, only some AS criterion exclude patients with!50% PBC, while a few other criterion exclude patients with a maximum number of cores positive for cancer or maximal percentage of cancer in a single biopsy core.(9, 71) Therefore, patients with low-risk and!50% PBC may read current treatment guidelines on their own and be falsely reassured that their disease is unlikely to need treatment or cause future morbidity or even mortality. Interestingly, among men age 70 and older, patients with low-risk and!50% PBC had similar incidence of occult, pathologic high risk disease to NCCN intermediate-risk patients, and low-risk!50% PBC had an increased risk of occult, pathologic high risk disease than favorable-intermediate risk among men with age over 60. Other studies have suggested that there is an increase in Gleason 8-10 with age;(60) therefore, older patients may have a higher risk of occult, high-risk disease than a younger patient with similar percent PBC. When considering older patients, physicians must always consider risk factors in the context of a patient s life expectancy, comorbidities, and potential negative-effects of treatment when deciding how aggressively to treat PCa. Further work is necessary to determine if age can contribute to pretreatment risk-stratification of men with PCa. There are some possible limitations to our analysis. First, because biopsy core data is only available in , there was not sufficient follow-up time to accrue survival data for the patients in this study. We suggest that finding occult, high-risk disease at prostatectomy is the best proxy available in SEER for future outcomes, as occult disease has been shown in institutional studies with longer follow-up to correlate with! "%

25 increased risk of biochemical progression at six years (Relative risk [RR] 1.86, 95% CI ),(16) and that patient outcomes correspond more strongly to prostatectomy Gleason than to biopsy Gleason score.(13) Second, an association has been demonstrated between the presence of perineural invasion (PNI) on biopsy and an increased risk of T3 disease at prostatectomy; however, the presence of PNI is not available in SEER at this time. Further work should be done to characterize the use of percent PBC with PNI data for risk-stratification of patients. Third, the pathologic results in SEER, which represents many pathologists, likely produces more variable results than studies with mandatory uropathology review. This was shown in a study that demonstrated a higher concordance between biopsy and pathologic Gleason scores at John s Hopkins (75.9%) than at a number of outside institutions (69.7%, p<0.001);(10) Therefore, the rates of discordant biopsy and pathologic Gleason score may be higher in this study than in single-institutional studies, but these results are likely more representative. Fourth, other clinical factors that may be associated with increased risk of occult disease on pathology and be useful for future risk-stratification models, such as PSA density, PSA velocity, and maximal involvement of a single biopsy core with PCa, are not available in SEER. Finally, this study is nonrandomized and has the general limitations of a retrospective, observational study. Conclusion Nearly one in five clinically low-risk prostate cancer patients with!50% positive biopsy cores had pathologic T3a-4 or Gleason 8-10 disease at prostatectomy. These patients had similar tumor size to unfavorable-intermediate risk, and among men over 60, had more occult high-risk disease than favorable-intermediate risk. Together, these results suggest that national guidelines should not categorize patients with low-risk and!50% cores positive as low risk and patients should be made aware of this excess risk if considering active surveillance.! "&

26 Part 1: Suggestions for Future Work Efforts to understand which patients truly have indolent PCa and, therefore, can be safely enrolled in active surveillance (AS) has been an area of intense research for more than two decades. The high rates of upgrading and upstaging found in the first paper included here, available online in February 2015, have been confirmed by other single institution and database studies,(72-77) including data from SEER.(72) Importantly, such results have begun to influence AS selection criteria. In these studies, various clinical criteria, always including percent PBC, were found to correlate with upgrading or upstaging. Reviews of AS criteria published since this work demonstrates that the role of percent PBC has been incorporated into many criteria, often including either "2 out of 12-cores positive for PCa or <33% (3/12) cores positive.(8, 76, 78) Importantly, many of the updated outcomes of prospective, AS studies have shown that increasing number of PBC correlates with increased likelihood of grade reclassification and/or treatment, further supporting our findings.(79-82) Therefore, it has been largely accepted that increasing number of cores positive for PCa correlates with worse pathologic outcomes or increased likelihood of treatment after AS. However, there are additional implications of this finding that should be explored with prospective study. Villa et al considered whether 12 biopsy cores was sufficient for finding aggressive PCa among seemingly low-risk PCa in a relatively small study.(83) Patients with >12 biopsy cores at initial biopsy had lower rates of upgrading or upstaging at prostatectomy, suggesting that the traditional 12-core biopsy may undersample the prostate in patients being considered for AS. Further work could consider the role for saturation biopsy in patients eligible for AS and how to interpret percent PBC in such a sample. Others have considered the role of MRI-guided or MRI-ultrasound fusion biopsies for detecting aggressive disease in patients eligible for AS.(53, 77) Siddiqui et al published a prospective study conducted at the national cancer institute of more than 1000 patients, suggesting that MRI-fusion biopsies were better at identifying high-risk disease while diagnosing fewer low-risk cases.(53) However, the application of this study is limited because many of the patients included had prior negative biopsies and there! "'

27 were no clinical endpoints measured. Further work is necessary to clarify the role of MRI-guided or MRI-fusion biopsies in the selection of patients for AS or in the detection of low-risk PCa. The barriers to such a study or the generalization of MRI biopsies is the availability of the technology, the expertise required to perform the biopsies, and the increased cost of performing such a biopsy. These studies will need to consider including a number needed to avoid treatment in standard biopsy versus MRI-guided biopsy as well as conduct a cost-analysis. Finally, other clinical factors that may be associated with increased risk of occult disease on pathology and be useful for future risk-stratification models, such as PSA density, PSA velocity, maximal involvement of a single biopsy core with PCa, and obesity are not available in SEER. A large, well-designed, prospective study with clinical endpoints is necessary to help evaluate the role for these additional clinical values in risk-stratifying PCa patients or selecting appropriate patients for AS. A large study will also provide the clinical endpoints to understand if occult disease correlates with worse clinical outcomes. However, most studies considering PCa outcomes after AS or treatment are not likely to be randomized, due to patient preference for different treatment modalities or for immediate, definitive treatment. In addition, clinical endpoint related to survival or PCSM generally take years to accrue in low-risk PCa given then 99% survival rate at 10 years. There is further need for agreement upon a surrogate endpoint, such as metastasis free survival or biochemical free survival for clinical trials in low-risk PCa to be feasible and have impact on men diagnosed with low-risk PCa in the near future.! "(

28 Part 2 Association of Androgen Deprivation Therapy with Depression in Localized Prostate Cancer Dinh KT, Reznor G, Muralidhar V, Mahal BA, Nezolosky MD, Choueiri TK, Hoffman KE, Hu JC, Sweeney CJ, Trinh QD, Nguyen PL. Accepted at Journal of Clinical Oncology 1/27/2016 Materials and Methods Data source Data was abstracted from the SEER, Medicare-linked database. SEER is a populationbased registry sponsored by the US National Cancer Institute including 18 regions, or 28% of the US population.(59) Medicare provides federal health insurance for approximately 97% of individuals age 65 years and older, and Medicare to SEER linkage is complete for 93% of eligible patients.(84) Study cohort The initial cohort of 172,733 patients included men age 66 and older diagnosed with clinically-localized PCa between , had no other malignancy in SEER, had both Medicare Part A and B claims available from one year prior to through thirty-six months beyond diagnosis, and were not enrolled in a health maintenance organization from Table 1 details the inclusion and exclusion criteria. Patients diagnosed at autopsy or by death certificate (n=5,315) or died within six months of diagnosis (n=21) were excluded. Patients with orchiectomy were excluded (n=2,986) to avoid confounding from effects that physical and anatomical changes from castration may have on depression. Additionally, to reduce confounding of advanced or recurrent disease, men with stage 4 or unknown stage PCa (n=68,133) were excluded as well as men initiated on ADT beyond six months from PCa diagnosis, as this could be salvage ADT (n=10,255). Finally, patients with a depression diagnosis from 12 months prior to diagnosis through t0 were excluded (n=7,511). These patients were identified if ICD-9! ")

29 codes (296.2, 296.3, 296.5, 296.6, 296.7,298.0, , ,301.13, 309.0, 309.1,311 further described in Table 2)(85-87) appeared in their SEER-Medicare record in the 12 months prior to PCa diagnosis. The final cohort included 78,552 patients with localized PCa. Patient sociodemographic and clinical characteristics Each patient s age at diagnosis, race, marital-status, and county-level population density were extracted from SEER. Zip-code level median income and education (percent with college diploma) were identified from census-linked data. Comorbidity status was assigned using the Klabunde-modification(88) of the Charlson comorbidity index (CCI) as 0, 1, or!2 from Medicare-claims data one year prior to PCa diagnosis. Clinically, overall stage was recorded as Stage I, II, or III. Gleason score was recorded from SEER-grade where, prior to 2003, well-differentiated was defined as Gleason 2-4, moderately-differentiated was Gleason 5-6, and poorly-differentiated was Gleason After 2004, Gleason score was recorded, and we categorized patients similarly. Three risk groups were defined by modification of Lu-Yao s risk groups,(89) with risk group A as Stage I-II and Gleason 2-6; risk group B was Stage I-II and Gleason 7-10; risk group C was Stage III with any grade.(89) Primary treatment type Patients were categorized as receiving ADT (ADT+) if ADT was the primary therapy or if given alongside radical prostatectomy (RP) or radiation therapy (RT) before t0, defined as six months after diagnosis. Patients were categorized as not receiving ADT (ADT-) if ADT was not started from diagnosis through the follow-up period of 36 months beyond diagnosis. Primary treatment was defined as RP with or without ADT, RT with or without ADT, or ADT alone if initiated before t0 as identified through SEER Medicare claims files, as previously described.(89, 90) Primary-treatment was assigned as observation if there was an absence of active treatment codes. ADT+ was stratified by duration, defined as 1-6, 7-11, and more than 12 months. End points! #+

30 The primary endpoint of interest was a depression diagnosis, defined as the presence of depression codes, described above (Table 2), in Medicare-claims data, including physician, outpatient, and inpatient claims, from t0 through 36 months from PCa diagnosis. Secondary endpoints included any outpatient-psychiatric or inpatientpsychiatric utilization.(91-93) Psychiatric utilization was measured by depression diagnosis code or psychiatric procedure-code and categorized as inpatient- or outpatient-treatment (table 2). The ICD-9 and procedure codes utilized to identify the endpoints in this study have not been formally validated, but have been used in prior studies as noted above. Statistical analysis Baseline demographic, clinical, and treatment characteristics were described for the overall cohort, stratified by ADT+ and ADT-, and compared using chi-square test for categorical variables and Wilcoxon rank-sum test for ordinal variables. The incidence of and time to primary and secondary endpoints were measured and compared by chisquared analysis and Wilcoxon-test. A number needed to harm was calculated. Coxproportional hazard analyses were performed to compare the possible effect of any ADT versus no ADT, controlling for demographic and clinical factors, on primary and secondary endpoints. The analysis was repeated for an association between duration of ADT, by strata defined above. Finally, propensity matched sensitivity-analyses, with 1:1 matching without replacement, were conducted for each endpoint as a sensitivity analysis.(94) Demographic covariates included age at diagnosis, marital status, race, county-level population density, zip-code income and education quartiles, and year of diagnosis. Clinical variables included modified-cci, PCa risk-group, and receipt of primary RP or RT. An interaction term was measured between year of diagnosis, defined as versus , and receipt of ADT on depression diagnosis to consider homogeneity of the effect of ADT over the study duration. Finally, unadjusted cumulative incidence function graphs were created comparing time to depression and inpatient-psychiatric treatment by receipt of ADT. Statistical testing was two-sided with a level of significance set at p=0.05. Analyses were performed using SAS, version 9.3 (SAS Institute,Cary,NC,USA). The Dana-Farber! #*

31 Cancer Institute s Institutional Review Board provided a waiver of informed consent for this study. Results Baseline patient demographics and treatment The study cohort included 78,552 men with localized PCa, of whom 33,882 (43.1%) received ADT within six months of diagnosis (ADT+). Baseline demographic and clinical features compared between ADT+ and patients who did not receive ADT (ADT-) are presented in Table 3. To summarize, the ADT+ cohort was older at diagnosis, had more comorbidities, had shorter life expectancies, and had more high-grade disease (all p<0.001). Of the 44,670 ADT- patients, 25.3% had RP, 44.2% had RT, and 30.5% received no treatment. Of the 33,882 ADT+ patients, 6.0% had RP plus ADT, 64.9% had RT plus ADT, and 29.1% had ADT alone. In the two years after diagnosis, 45.3% of ADT+ cohort received six or fewer months, 22.1% received seven to 11 months, and 32.6% received!12 months of ADT. In the top strata, 40.8% (4,515) patients had!18 months of ADT. Risk of depression and psychiatric treatment associated with any ADT The cumulative-incidence of new depression from six to 36 months after PCa diagnosis was higher in the ADT+ than ADT- cohorts (7.1 vs. 5.2%, p<0.001, Table 3). The number needed to harm was 53. More ADT+ patients had inpatient- (2.8 vs. 1.9%) and outpatient-psychiatric treatment (3.4 vs. 2.5%) as well (all p<0.001, Table 3). An adjusted Cox-proportional hazard analysis showed that, compared to ADT-, there was a 23% increased risk of depression associated with ADT+ (adjusted hazard ratio [AHR]=1.23, 95%CI=[ ], p<0.0001). Similarly, ADT+ had a 29% increased risk compared to ADT- of inpatient-psychiatric treatment (AHR=1.29; 95%CI=[ ], p<0.0001) and 7% increased risk of outpatient-psychiatric treatment (AHR=1.07; 95%CI= , p=0.17) compared to ADT-. These associations were strengthened when using propensity matching in sensitivity analyses (Table 4). Older age, being! #"

32 unmarried, and having a more comorbidity (CCI>0) were also associated with increased risk of depression (all p<0.0001). An interaction term between the era of PCa diagnosis, vs , and the receipt of ADT on depression was non-significant (p=0.306), suggesting that the effect of receiving ADT on a diagnosis of depression was not significantly different before or after the year Figure 1 and 2 show the unadjusted cumulative incidence function of time to depression and inpatient-psychiatric treatment, respectively, demonstrating that the incidence was increased almost immediately after initiating ADT therapy. Risk of depression and psychiatric treatment associated with duration of ADT ADT+ was stratified by duration of therapy, defined as six or fewer, seven to 11, and!12 months. The cumulative-incidence of depression increased by strata, from 6.1% to 7.6% to 8.0% with six or fewer, seven to 11, and!12 months of ADT respectively, compared to 5.2% in ADT- (p<0.0001). The incidence of inpatient-psychiatric treatment increased by strata, from 2.4% to 3.0% to 3.3%, respectively, compared to 1.9% in ADT- (p<0.0001), as did the incidence of outpatient-psychiatric treatment, from 2.8% to 3.5% to 4.1%, respectively, compared to 2.5% in ADT- (p<0.0001). Table 5 shows the adjusted Cox-proportional hazard analysis of the association between ADT duration and each endpoint, controlling for demographic and clinical variables. The risk for depression among ADT+ compared to ADT- cohorts increased with ADT duration, from 12% higher risk with six or fewer months (AHR=1.12; 95%CI=[ ]) to 26% with seven to 11 (AHR=1.26; 95%CI=[ ]) and 37% with!12 months (AHR=1.37; 95%CI=[ ], p-trend<0.0001). Similarly, the risk of inpatient-psychiatric treatment increased with ADT duration, compared to no ADT, from 16% higher risk with six or fewer months (AHR=1.16; 95%CI=[ ]) to 28% with seven to 11 (AHR=1.28; 95%CI=[ ]) and 47% with!12 months (AHR=1.47; 95%CI=[ ], p-trend<0.0001), while the risk for outpatient-psychiatric treatment increased slightly with duration of ADT from 3% lower risk with six or fewer months (AHR=0.97; 95%CI=[ ]) to 3% higher risk with seven to 11 (AHR=1.03l 95%CI=[ ]) and 20% higher risk with!12 months of ADT (AHR=1.20; 95%CI=[ ], p-trend=0.04).! ##

33 Discussion Patients who received any ADT had a 23% increased risk of new depression compared to those without ADT when adjusting for demographic and clinical variables in this large, population-based study of 78,000 US men with localized PCa age 66 and older. We also observed a significant trend between increasing duration of ADT and increasing risk of depression, finding that patients receiving six or fewer months of ADT had a 12% increased risk compared to no ADT versus a 37% increased risk with twelve or more months of ADT. This evidence strongly suggests an association between receiving ADT and developing new depression in men age 66 and older with localized PCa. The impact of ADT on depression may plausibly occur via deregulation of neurochemicals, like serotonin, in addition to the well-described physical effects.(95, 96) Though a direct mechanism of ADT causing depression has yet to be identified, studies in mice have demonstrated that enzymes responsible for maintaining the neurochemical balance are hormonally responsive and that gonectamized mice have higher activity of monoamine oxidase A (MAO-A) in the prefrontal cortex compared to controls.(95) This increased activity of MAO-A could be responsible for deregulation of serotonin and thus increased rates of depression. Additionally, low-testosterone has been associated with increased depression in otherwise healthy men,(97-99) though the literature is somewhat inconsistent.(39, 100) Additionally, the many known side-effects of ADT have been shown to negatively impact quality of life,(96) including physical effects, such as erectile dysfunction, shortening of penile length, vasomotor flushing, weight gain, and gynecomastia, as well as cognitive effects, including insomnia and a decline in verbal and executive function.(36, 100) As physician-claims data have been shown to underestimate disease incidence(101) and can also be subject to coding errors, we chose two additional measures of clinically-significant psychiatric disease: inpatient- and outpatient-psychiatric treatment. A similar measurement was utilized in a population-based study of PCa in Sweden.(93) We again found the risk for inpatient-psychiatric treatment increased with ADT duration, from 16% for six or fewer (AHD=1.16; 95%CI=[ ]) to 47% with!12 months (AHR=1.47; 95%CI=[ ], p-trend<0.001) compared to no ADT. The p-trend for! #$

34 both inpatient- and outpatient-psychiatric treatment were significant, demonstrating that with increasing duration of ADT the relative-risk of these outcomes increased compared to no ADT. The association of ADT with psychiatric treatment provides further evidence that ADT may have a clinically significant effect on psychiatric health. There are multiple implications of our findings. First, we contribute to mounting evidence around the extensive side-effect profile,(36, 100) which should include risk of clinically significant psychiatric disease. Physicians should note psychiatric illness as a possible side effect when discussing ADT and consider including such effects in consent-forms. Second, judicious use of ADT is warranted to avoid unnecessary side effects. ADT should not be used in low-risk patients, as it does not provide benefit.(102) Among intermediate-risk PCa the potential benefits of ADT must be weighed against the potential harms, including increased depression risk. Third, further work should identify what patient subgroups are at greatest risk of psychiatric side-effects on ADT and what interventions may reduce this risk. In this study, older age, being unmarried, and having more comorbidity were associated with increased depression and psychiatric treatment. While patients with recent psychiatric disease were excluded from the present study, other work has identified increased rates of depression among similar patients.(39, 103) These populations may be appropriate to target for early depression screening or intervention following ADT initiation. Depression in PCa also impacts healthcare utilization,(85, 104, 105) including increased emergency room visits and hospitalizations,(104) and patient outcomes, including increased suicide and overall mortality.(106) Therefore, identification of and support for men at risk for depression may have broad effects on quality of life, healthcare utilization, and survival. Prior work identifying interventions to reduce the risk of depression with ADT has suggested a role for exercise,(36, 101, 105) though two reviews have concluded the current evidence is lacking.(105, 107) High-quality research on interventions to mitigate depressive effects of ADT is needed. Prevention may be especially important as little is known about whether ADT use modifies the efficacy of commonly used treatments for depression in these patients.(108, 109) Prior work has! #%

35 demonstrated that further characterization of depressive symptoms in men receiving ADT may also guide identification and treatment.(110) Published literature considering the association between ADT and depression has been highly inconsistent.(37-45, 104, 111) Many of these studies were likely underpowered, failing to find an association between ADT and depression, while previous populationbased studies have used a broad definition of depression(111) or considered only primary rather than adjuvant ADT,(93) limiting the generalizability of their results. Our findings show that any ADT increases the risk of depression using a more specific definition of clinical depression, demonstrate an association between ADT and increased psychiatric treatment, and show that ADT is associated with these outcomes in the adjuvant setting. Furthermore, we demonstrated a dose-response with ADT duration, which may reflect a direct biological effect of increased exposure to ADT, an indirect biological effect due to other symptoms (e.g. decreased libido), or a negative emotional response to being told that one needs longer-course ADT. ADT duration has only been previously considered in three small studies, none of which demonstrated different risk with duration or between long- and short-course ADT.(39, 96, 101) There are some potential limitations of this study. First, based on the eligibility for Medicare, the cohort was limited to patients aged 66 and older at diagnosis. The results may not, therefore, reflect the psychiatric effects of ADT given to younger men, who account for approximately one-third of PCa cases.(112) Secondly, the outcome of depression was based on diagnosis, rather than symptoms, which relied on physician recognition and reporting.(113) Older men report depressive symptoms less and often present with somatic rather than emotional symptoms;(39, 114, 115) therefore, it is possible that many symptomatic men were undiagnosed and therefore not measured in this study. This should have affected each cohort similarly. Additionally, patients receiving ADT likely had more appointments than the ADT- cohort and physicians may have been more sensitive to symptoms of depression in patients receiving ADT or with more aggressive disease, increasing their likelihood of being diagnosed with depression in ADT+ over ADT- patients.! #&

36 We attempted to account for this by also analyzing endpoints such as inpatient and outpatient psychiatric treatment, which may be less likely to be affected by this potential bias. The secondary endpoint in this study was an exploratory analysis inspired by prior studies conducted outside of SEER-Medicare considering psychotherapy. Many codes included for this secondary endpoint come from the psychotherapy literature as well as from identification of depression ICD-9 codes described above associated with hospitalization (Table 2). The codes were carefully and thoughtfully chosen to most accurately reflect psychiatric utilization by patients. Third, we could not control for the use of psychotropic medication prior to PCa diagnosis, but did exclude patients based on history of depressive diagnosis or psychiatric treatment. Data on prescription of psychiatric medication was not available in SEER-Medicare for use as an endpoint. Fourth, increased depression associated with ADT may be secondary to confounding, as these patients were older, had more comorbidity, had more advanced disease, and experienced physical effects of ADT. These additional symptoms and comorbidity could have led to an over-diagnosis of depression in the ADT+ cohort. However, all MVAs were adjusted for demographic and clinical variables, including age, comorbidity (CCI), and PCa risk-group, and found a consistent increased-risk with duration of ADT across all three endpoints. In addition, a greater proportion of the ADT+ cohort were of ethnic minorities (17.3 vs. 15.4%) and had low comorbidity (CCI=0, 66.2 vs. 64.5%), which were both variables independently associated with lower rates of depression in MVAs. Fifth, we cannot control for disease progression on ADT, which may contribute to increased incidence of depression. Finally, this study is non-randomized and shares the general limitations associated with retrospective, observational studies. Conclusion We observed a significantly increased risk of depression and inpatient-psychiatric treatment in men treated with ADT for PCa, as well as a duration-response effect such that more ADT was linked to increasing risk of depression and inpatient- and outpatientpsychiatric treatment. The possible psychiatric effects of ADT should be recognized by physicians and discussed with patients prior to initiating treatment.! #'

37 Part 2: Suggestions for future work This work adds to the growing body of research supporting the relationship between receipt of ADT and clinically significant depression. However, a recent review by Donovan et al considering the psychological effects of ADT noted the relative lack of prospective studies considering this important relationship.(116) Currently published prospective studies have small sample sizes, and many suffer from either the lack of a control group or lack of a diagnostic interview relying on patient-reported survey results. Therefore, the most important next step in understanding the relationship between ADT and depression is a large, well-designed prospective study to compare PCa patients receiving ADT to those treated with definitive therapy to men without cancer. Further, a prospective trial could compare PCa patients on ADT to men with metastatic cancer that is not prostate who are receiving systemic therapy. This study will need to careful consider the use of self-reported surveys versus diagnostic interviews to evaluate the rate of depression as well as track and control for physical and cognitive side-effects of ADT. A prospective study published In 2015 made an important step forward, comparing 61 men with ADT to 61 men with radical prostatectomy and 61 men without cancer.(45) This study found that patients receiving ADT had a significant increase in rates of depression at 12 months compared to initiation of treatment (28% to 39%) when compared to both control groups, respectively (p<0.01, p=0.01). While this is still a fairly small study with only 12 months of follow-up, the prospective nature supports the relationship between ADT and depression. The cause of depression in patients receiving ADT needs to be further explored. In this paper, I discuss a possible biological mechanism between ADT and depression; however, the extensive physical and cognitive side effect profile of ADT should not be dismissed. Defining the effect of any one of the numerous side effects on rates of depression could inform treatment options to reduce the burden of depression in patients. Additionally, the effectiveness of various treatment options, including antidepressive medications, psychotherapy, cognitive behavioral therapy, exercise, or psychosocial interventions, needs to be studied in a prospective fashion. Some! #(

38 challenges to such a study will include the feasibility of randomization of men to screening for depression versus no screening and randomization to different treatment modalities for depression. A meta-analysis of studies was recently published, but suffers from poor-quality randomized trials; however, the results suggest that peer-support and psychotherapy may show significant improvement in depressive symptoms.(117) Finally, the most important work following this study will be to raise awareness among physicians prescribing ADT and patients taking ADT of the risk of depression. A study of nearly 700 patients in Germany found that only 51% of patients felt that expected side effects were clearly discussed prior to initiating ADT and that development of depression was significantly associated with non-adherence to ADT treatment.(118) The risk of depression needs to be incorporated into every discussion considering the risks and benefits of ADT. Finally, patients on ADT must be appropriately screened for depressive symptoms with a clear mechanism of referral to appropriate psychological evaluation and treatment if necessary. Recent work by Sharpley et al suggests that men experience additional symptoms of depression beyond those traditionally considered, such as major increases in irritability or aggression.(119) Prospective study should consider the utility of additional symptom categories in identifying men on ADT with clinically significant depression.! #)

39 Summary The Surveillance, Epidemiology and End Results (SEER) database provides value information about prostate cancer (PCa) patients in the United States. Two important areas of research in this field include the identification of appropriate patients for active surveillance (AS) and what side effects of androgen deprivation therapy (ADT) are clinically significant for patients with high-risk or metastatic PCa. This thesis utilizes the SEER and SEER-Medicare linked databases to answer specific questions in each of these two areas. In part 1, I used information about clinical and pathologic staging of PCa patients to evaluate the accuracy of the NCCN classification of low-risk PCa and to inform the selection of appropriate patients for AS. I first found that 44% of clinically low-risk PCa are upgraded or upstaged at prostatectomy to Gleason 7 to 10 or pathologic T3-T4/N0- N1 disease. Higher PSA and percent positive biopsy cores (PBC) best correlated with risk of upgrading and upstaging. Based on these results, I then considered the utility of percent PBC in improving the risk stratification of newly diagnosed PCa. I found that nearly one in five clinically low-risk prostate cancer patients with!50% PBC had pathologic T3a-4 or Gleason 8-10 disease at prostatectomy. Together, the results of these two studies suggest that future national guidelines should not categorize patients with clinically low-risk disease by the current NCCN definition and!50% PBC as low risk. Additionally, these patients should be made aware of the excess risk associated with high percent PBC if considering active surveillance and perhaps have further screening with prostate MRI for extracapsular or high-grade disease. In part 2, I utilized the SEER-Medicare linked database to measure the incidence of new depression and inpatient-psychiatric treatment in PCa patients treated with ADT compared to those without ADT. We found a significant increase in both outcomes, as well as a duration-response effect such that increasing doses of ADT was linked to an increasing risk of depression and inpatient-psychiatric treatment. Physicians should discuss the possible psychiatric effects of ADT with patients prior to initiating treatment as well as screen patients on ADT for depression.! $+

40 Acknowledgements This work would not have been possible without the mentorship from Dr. Paul Nguyen. He laid a comprehensive foundation of both research considering prostate cancer and database-driven research, enabling me to complete the included projects. I would also like to thank Dr. Janet Mullington, for encouraging me to pursue a research year and to complete this honors thesis. Thanks to biostatistician Gally Reznor for her help with statistics for the ADT and depression paper. Finally, I must thank my husband, Khang Dinh, for his endless support of my endeavors and dreams. This work is supported financially by Fitz s Cancer Warriors, David and Cynthia Chapin, the Prostate Cancer Foundation, Hugh Simons in honor of Frank and Anne Simons, Scott Forbes and Gina Ventre, Campbell family in honor of Joan Campbell, and a grant from an anonymous family foundation. In addition, Kathryn received financial support from the Scholar in Medicine office of Harvard Medical School. Personal Contribution to this Work The three studies included above were the primary work that I did during a year of research with Dr. Nguyen. While the primary idea was provided by Dr. Nguyen for Incidence and Predictors of Upgrading and Up Staging among 10,000 Contemporary Patients with Low Risk Prostate Cancer, I primarily designed the study, pulled and coded the data, performed the statistical analysis, as well as wrote and submitted the manuscript. The second paper included, Occult High-Risk Disease in Clinically Low- Risk Prostate Cancer with!50% Positive Biopsy Cores: Should National Guidelines Stop Calling Them Low-Risk?, was an idea that I had following the completion of the first paper. For this manuscript, I primarily designed the study, pulled and coded the data, performed the statistical analysis, as well as wrote and submitted the manuscript. Finally, the third paper, Association of Androgen Deprivation Therapy with Depression in Localized Prostate Cancer, was developed from a hypothesis suggested by Dr. Nguyen. I did the primary study design, but worked with biostatistician Gally Reznor who pulled and coded the data as well as conducted the primary statistical analyses. We worked together to refine the data and analyses before I primarily wrote and submitted the manuscript. In all three instances, I was the first author on the publication.! $*

41 References 1. Siegel RL, Miller KD, Jemal A. Cancer statistics, CA: a cancer journal for clinicians. 2015;65(1):5-29. doi: /caac PubMed PMID: Hankey BF, Feuer EJ, Clegg LX, Hayes RB, Legler JM, Prorok PC, et al. Cancer surveillance series: interpreting trends in prostate cancer--part I: Evidence of the effects of screening in recent prostate cancer incidence, mortality, and survival rates. J Natl Cancer Inst. 1999;91(12): PubMed PMID: Bill-Axelson A, Holmberg L, Ruutu M, Garmo H, Stark JR, Busch C, et al. Radical prostatectomy versus watchful waiting in early prostate cancer. The New England journal of medicine. 2011;364(18): Epub 2011/05/06. doi: /NEJMoa PubMed PMID: Wilt TJ, Brawer MK, Jones KM, Barry MJ, Aronson WJ, Fox S, et al. Radical prostatectomy versus observation for localized prostate cancer. The New England journal of medicine. 2012;367(3): Epub 2012/07/20. doi: /NEJMoa PubMed PMID: ; PubMed Central PMCID: PMC Wilt TJ, Brawer MK, Barry MJ, Jones KM, Kwon Y, Gingrich JR, et al. The Prostate cancer Intervention Versus Observation Trial:VA/NCI/AHRQ Cooperative Studies Program #407 (PIVOT): design and baseline results of a randomized controlled trial comparing radical prostatectomy to watchful waiting for men with clinically localized prostate cancer. Contemp Clin Trials. 2009;30(1):81-7. doi: /j.cct PubMed PMID: Wilt TJ. Management of low risk and low PSA prostate cancer: long term results from the prostate cancer intervention versus observation trial. Recent results in cancer research Fortschritte der Krebsforschung Progres dans les recherches sur le cancer. 2014;202: Epub 2014/02/18. doi: / _18. PubMed PMID: Lane JA, Donovan JL, Davis M, Walsh E, Dedman D, Down L, et al. Active monitoring, radical prostatectomy, or radiotherapy for localised prostate cancer: study design and diagnostic and baseline results of the ProtecT randomised phase 3 trial. The Lancet Oncology. 2014;15(10): doi: /s (14) Iremashvili V, Pelaez L, Manoharan M, Jorda M, Rosenberg DL, Soloway MS. Pathologic prostate cancer characteristics in patients eligible for active surveillance: a head-to-head comparison of contemporary protocols. European urology. 2012;62(3): Epub 2012/03/27. doi: /j.eururo PubMed PMID: Thomsen FB, Brasso K, Klotz LH, Roder MA, Berg KD, Iversen P. Active surveillance for clinically localized prostate cancer--a systematic review. Journal of surgical oncology. 2014;109(8): Epub 2014/03/13. doi: /jso PubMed PMID: Fine SW, Epstein JI. A contemporary study correlating prostate needle biopsy and radical prostatectomy Gleason score. The Journal of urology. 2008;179(4):1335-8; discussion 8-9. Epub 2008/02/22. doi: /j.juro PubMed PMID: ! $"

42 11. Isariyawongse BK, Sun L, Banez LL, Robertson C, Polascik TJ, Maloney K, et al. Significant discrepancies between diagnostic and pathologic Gleason sums in prostate cancer: the predictive role of age and prostate-specific antigen. Urology. 2008;72(4): Epub 2008/04/04. doi: /j.urology PubMed PMID: Krane LS, Menon M, Kaul SA, Siddiqui SA, Wambi C, Peabody JO, et al. Role of PSA velocity in predicting pathologic upgrade for Gleason 6 prostate cancer. Urologic oncology. 2011;29(4): Epub 2009/07/07. doi: /j.urolonc PubMed PMID: Pinthus JH, Witkos M, Fleshner NE, Sweet J, Evans A, Jewett MA, et al. Prostate cancers scored as Gleason 6 on prostate biopsy are frequently Gleason 7 tumors at radical prostatectomy: implication on outcome. The Journal of urology. 2006;176(3):979-84; discussion 84. Epub 2006/08/08. doi: /j.juro PubMed PMID: Rapiti E, Schaffar R, Iselin C, Miralbell R, Pelte MF, Weber D, et al. Importance and determinants of Gleason score undergrading on biopsy sample of prostate cancer in a population-based study. BMC urology. 2013;13:19. Epub 2013/04/13. doi: / PubMed PMID: ; PubMed Central PMCID: PMC 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. European urology. 2012;61(5): Epub 2012/02/18. doi: /j.eururo PubMed PMID: Freedland SJ, Kane CJ, Amling CL, Aronson WJ, Terris MK, Presti JC, Jr. Upgrading and downgrading of prostate needle biopsy specimens: risk factors and clinical implications. Urology. 2007;69(3): Epub 2007/03/27. doi: /j.urology PubMed PMID: ; PubMed Central PMCID: PMC Grossfeld GD, Chang JJ, Broering JM, Li YP, Lubeck DP, Flanders SC, et al. Under staging and under grading in a contemporary series of patients undergoing radical prostatectomy: results from the Cancer of the Prostate Strategic Urologic Research Endeavor database. The Journal of urology. 2001;165(3): Epub 2001/02/15. PubMed PMID: Gancarczyk KJ, Wu H, McLeod DG, Kane C, Kusuda L, Lance R, et al. Using the percentage of biopsy cores positive for cancer, pretreatment PSA, and highest biopsy Gleason sum to predict pathologic stage after radical prostatectomy: the Center for Prostate Disease Research nomograms. Urology. 2003;61(3): Epub 2003/03/18. PubMed PMID: Surveillance, Epidemiology, and End Results (SEER) Program ( SEER*Stat Database: Incidence - SEER 9 Regs Research Data, Nov 2014 Sub ( ) <Katrina/Rita Population Adjustment> - Linked To County Attributes - Total U.S., Counties, National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2015, based on the November 2014 submission.! $#

43 20. Dinh KT, Mahal BA, Ziehr DR, Muralidhar V, Chen YW, Viswanathan VB, et al. Incidence and Predictors of Upgrading and Up Staging among 10,000 Contemporary Patients with Low Risk Prostate Cancer. The Journal of urology. 2015;194(2): doi: /j.juro PubMed PMID: James L Mohler AJA, Robert R Bahnson, et al. NCCN Clinical Practice Guidelines in Oncology: Prostate Cancer. 2015(1.2015). Epub 10/24/ D'Amico AV, Renshaw AA, Cote K, Hurwitz M, Beard C, Loffredo M, et al. Impact of the percentage of positive prostate cores on prostate cancer-specific mortality for patients with low or favorable intermediate-risk disease. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2004;22(18): doi: /JCO PubMed PMID: Zumsteg ZS, Spratt DE, Pei I, Zhang Z, Yamada Y, Kollmeier M, et al. A new risk classification system for therapeutic decision making with intermediate-risk prostate cancer patients undergoing dose-escalated external-beam radiation therapy. European urology. 2013;64(6): doi: /j.eururo PubMed PMID: Yoon JH, Chen MH, Renshaw AA, Richie JP, D'Amico AV. Predictive factor analysis as the basis for the clinical utility of percent positive prostate biopsies in patients with intermediate-risk prostate cancer. Urology. 2002;60(3): PubMed PMID: Lieberfarb ME, Schultz D, Whittington R, Malkowicz B, Tomaszewski JE, Weinstein M, et al. Using PSA, biopsy Gleason score, clinical stage, and the percentage of positive biopsies to identify optimal candidates for prostate-only radiation therapy. International journal of radiation oncology, biology, physics. 2002;53(4): PubMed PMID: Dinh KT, Muralidhar V, Mahal BA, Chen YW, Nezolosky MD, Beard CJ, et al. Occult High-Risk Disease in Clinically Low-Risk Prostate Cancer with >/=50% Positive Biopsy Cores: Should National Guidelines Stop Calling Them Low-Risk? Urology doi: /j.urology PubMed PMID: Mahmood U, Levy LB, Nguyen PL, Lee AK, Kuban DA, Hoffman KE. Current clinical presentation and treatment of localized prostate cancer in the United States. The Journal of urology. 2014;192(6): doi: /j.juro PubMed PMID: D'Amico AV, Manola J, Loffredo M, Renshaw AA, DellaCroce A, Kantoff PW. 6- month androgen suppression plus radiation therapy vs radiation therapy alone for patients with clinically localized prostate cancer: a randomized controlled trial. Jama. 2004;292(7): doi: /jama PubMed PMID: D'Amico AV, Chen MH, Renshaw AA, Loffredo M, Kantoff PW. Androgen suppression and radiation vs radiation alone for prostate cancer: a randomized trial. Jama. 2008;299(3): doi: /jama PubMed PMID: Bolla M, Gonzalez D, Warde P, Dubois JB, Mirimanoff RO, Storme G, et al. Improved survival in patients with locally advanced prostate cancer treated with radiotherapy and goserelin. The New England journal of medicine. 1997;337(5): doi: /NEJM PubMed PMID: Bolla M, Collette L, Blank L, Warde P, Dubois JB, Mirimanoff RO, et al. Longterm results with immediate androgen suppression and external irradiation in patients! $$

44 with locally advanced prostate cancer (an EORTC study): a phase III randomised trial. Lancet. 2002;360(9327): PubMed PMID: Horwitz EM, Bae K, Hanks GE, Porter A, Grignon DJ, Brereton HD, et al. Tenyear follow-up of radiation therapy oncology group protocol 92-02: a phase III trial of the duration of elective androgen deprivation in locally advanced prostate cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2008;26(15): doi: /JCO PubMed PMID: Roach M, 3rd, Bae K, Speight J, Wolkov HB, Rubin P, Lee RJ, et al. Short-term neoadjuvant androgen deprivation therapy and external-beam radiotherapy for locally advanced prostate cancer: long-term results of RTOG Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2008;26(4): doi: /JCO PubMed PMID: Denham JW, Steigler A, Lamb DS, Joseph D, Mameghan H, Turner S, et al. Short-term androgen deprivation and radiotherapy for locally advanced prostate cancer: results from the Trans-Tasman Radiation Oncology Group randomised controlled trial. The Lancet Oncology. 2005;6(11): doi: /S (05)70348-X. PubMed PMID: Gilbert SM, Kuo YF, Shahinian VB. Prevalent and incident use of androgen deprivation therapy among men with prostate cancer in the United States. Urologic oncology. 2011;29(6): doi: /j.urolonc PubMed PMID: ; PubMed Central PMCID: PMC Nguyen PL, Alibhai SM, Basaria S, D'Amico AV, Kantoff PW, Keating NL, et al. Adverse Effects of Androgen Deprivation Therapy and Strategies to Mitigate Them. European urology doi: /j.eururo PubMed PMID: Pirl WF, Greer JA, Goode M, Smith MR. Prospective study of depression and fatigue in men with advanced prostate cancer receiving hormone therapy. Psychooncology. 2008;17(2): doi: /pon PubMed PMID: ; PubMed Central PMCID: PMC Cherrier MM, Aubin S, Higano CS. Cognitive and mood changes in men undergoing intermittent combined androgen blockade for non-metastatic prostate cancer. Psycho-oncology. 2009;18(3): doi: /pon PubMed PMID: ; PubMed Central PMCID: PMC Hervouet S, Savard J, Ivers H, Savard MH. Depression and androgen deprivation therapy for prostate cancer: a prospective controlled study. Health psychology : official journal of the Division of Health Psychology, American Psychological Association. 2013;32(6): doi: /a PubMed PMID: Timilshina N, Breunis H, Alibhai S. Impact of androgen deprivation therapy on depressive symptoms in men with nonmetastatic prostate cancer. Cancer. 2012;118(7): doi: /cncr PubMed PMID: Cary KC, Singla N, Cowan JE, Carroll PR, Cooperberg MR. Impact of androgen deprivation therapy on mental and emotional well-being in men with prostate cancer: analysis from the CaPSURE registry. The Journal of urology. 2014;191(4): doi: /j.juro PubMed PMID: Wiechno PJ, Sadowska M, Kalinowski T, Michalski W, Demkow T. Does pharmacological castration as adjuvant therapy for prostate cancer after radiotherapy! $%

45 affect anxiety and depression levels, cognitive functions and quality of life? Psychooncology. 2013;22(2): doi: /pon PubMed PMID: van Tol-Geerdink JJ, Leer JW, van Lin EN, Schimmel EC, Stalmeier PF. Depression related to (neo)adjuvant hormonal therapy for prostate cancer. Radiotherapy and oncology : journal of the European Society for Therapeutic Radiology and Oncology. 2011;98(2): doi: /j.radonc PubMed PMID: Saini A, Berruti A, Cracco C, Sguazzotti E, Porpiglia F, Russo L, et al. Psychological distress in men with prostate cancer receiving adjuvant androgendeprivation therapy. Urologic oncology. 2013;31(3): doi: /j.urolonc PubMed PMID: Lee M, Jim HS, Fishman M, Zachariah B, Heysek R, Biagioli M, et al. Depressive symptomatology in men receiving androgen deprivation therapy for prostate cancer: a controlled comparison. Psycho-oncology. 2015;24(4): Epub 2014/06/14. doi: /pon PubMed PMID: ; PubMed Central PMCID: PMCPMC Heidenreich A, Bastian PJ, Bellmunt J, Bolla M, Joniau S, van der Kwast T, et al. EAU guidelines on prostate cancer. part 1: screening, diagnosis, and local treatment with curative intent-update European urology. 2014;65(1): Epub 2013/11/12. doi: /j.eururo PubMed PMID: Eichler K, Hempel S, Wilby J, Myers L, Bachmann LM, Kleijnen J. Diagnostic value of systematic biopsy methods in the investigation of prostate cancer: a systematic review. The Journal of urology. 2006;175(5): Epub 2006/04/08. doi: /S (05) PubMed PMID: Tosoian JJ, JohnBull E, Trock BJ, Landis P, Epstein JI, Partin AW, et al. Pathological outcomes in men with low risk and very low risk prostate cancer: implications on the practice of active surveillance. The Journal of urology. 2013;190(4): Epub 2013/05/07. doi: /j.juro PubMed PMID: ; PubMed Central PMCID: PMC Bjurlin MA, Meng X, Le Nobin J, Wysock JS, Lepor H, Rosenkrantz AB, et al. Optimization of prostate biopsy: the role of magnetic resonance imaging targeted biopsy in detection, localization and risk assessment. The Journal of urology. 2014;192(3): Epub 2014/04/29. doi: /j.juro PubMed PMID: Sonn GA, Natarajan S, Margolis DJ, MacAiran M, Lieu P, Huang J, et al. Targeted biopsy in the detection of prostate cancer using an office based magnetic resonance ultrasound fusion device. The Journal of urology. 2013;189(1): Epub 2012/11/20. doi: /j.juro PubMed PMID: ; PubMed Central PMCID: PMC Puech P, Rouviere O, Renard-Penna R, Villers A, Devos P, Colombel M, et al. Prostate cancer diagnosis: multiparametric MR-targeted biopsy with cognitive and transrectal US-MR fusion guidance versus systematic biopsy--prospective multicenter study. Radiology. 2013;268(2): Epub 2013/04/13. doi: /radiol PubMed PMID: Moore CM, Robertson NL, Arsanious N, Middleton T, Villers A, Klotz L, et al. Image-guided prostate biopsy using magnetic resonance imaging-derived targets: a! $&

46 systematic review. European urology. 2013;63(1): Epub 2012/06/30. doi: /j.eururo PubMed PMID: Siddiqui MM, Rais-Bahrami S, Turkbey B, George AK, Rothwax J, Shakir N, et al. Comparison of MR/ultrasound fusion-guided biopsy with ultrasound-guided biopsy for the diagnosis of prostate cancer. Jama. 2015;313(4): doi: /jama PubMed PMID: Cooperberg MR, Carroll PR, Klotz L. Active surveillance for prostate cancer: progress and promise. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2011;29(27): Epub 2011/08/10. doi: /JCO PubMed PMID: Cooke EW, Shrieve DC, Tward JD. Clinical versus pathologic staging for prostate adenocarcinoma: how do they correlate? American journal of clinical oncology. 2012;35(4): Epub 2011/03/26. doi: /COC.0b013e fc. PubMed PMID: Epstein JI, Allsbrook WC, Jr., Amin MB, Egevad LL. The 2005 International Society of Urological Pathology (ISUP) Consensus Conference on Gleason Grading of Prostatic Carcinoma. The American journal of surgical pathology. 2005;29(9): Epub 2005/08/13. PubMed PMID: Sundi D, Ross AE, Humphreys EB, Han M, Partin AW, Carter HB, et al. African American men with very low-risk prostate cancer exhibit adverse oncologic outcomes after radical prostatectomy: should active surveillance still be an option for them? Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2013;31(24): Epub 2013/06/19. doi: /JCO PubMed PMID: ; PubMed Central PMCID: PMC Weiner AB, Patel SG, Etzioni R, Eggener SE. National trends in the management of low- and intermediate-risk prostate cancer in the United States. The Journal of urology Epub 2014/08/12. doi: /j.juro PubMed PMID: Surveillance E, and End Results (SEER) Program (,--./ :;) SEER*Stat Database: Incidence - SEER 18 Regs Research Data + Hurricane Katrina Impacted Louisiana Cases, Nov 2013 Sub ( varying) - Linked To County Attributes - Total U.S., Counties, National Cancer Institute, DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2014 (updated 5/7/2014), based on the November 2013 submission. 60. Futterer JJ, Briganti A, De Visschere P, Emberton M, Giannarini G, Kirkham A, et al. Can Clinically Significant Prostate Cancer Be Detected with Multiparametric Magnetic Resonance Imaging? A Systematic Review of the Literature. European urology doi: /j.eururo PubMed PMID: Sebo TJ, Cheville JC, Riehle DL, Lohse CM, Pankratz VS, Myers RP, et al. Predicting prostate carcinoma volume and stage at radical prostatectomy by assessing needle biopsy specimens for percent surface area and cores positive for carcinoma, perineural invasion, Gleason score, DNA ploidy and proliferation, and preoperative serum prostate specific antigen: a report of 454 cases. Cancer. 2001;91(11): PubMed PMID: Dinh KT, Mahal BA, Ziehr DR, Muralidhar V, Chen YW, Viswanathan VB, et al. Incidence and Predictors of Upgrading and Upstaging Among 10,000 Contemporary! $'

47 Patients with Low-risk Prostate Cancer. The Journal of urology doi: /j.juro PubMed PMID: Liauw SL, Fricano J, Correa D, Weichselbaum RR, Jani AB. Dose-escalated radiation therapy for intermediate-risk prostate cancer: patient selection for androgen deprivation therapy using percentage of positive cores. Cancer. 2009;115(8): doi: /cncr PubMed PMID: Rossi PJ, Clark PE, Papagikos MA, McCullough DL, Lee WR. Percentage of positive biopsies associated with freedom from biochemical recurrence after low-doserate prostate brachytherapy alone for clinically localized prostate cancer. Urology. 2006;67(2): doi: /j.urology PubMed PMID: Huang J, Vicini FA, Williams SG, Ye H, McGrath S, Ghilezan M, et al. Percentage of positive biopsy cores: a better risk stratification model for prostate cancer? International journal of radiation oncology, biology, physics. 2012;83(4): doi: /j.ijrobp PubMed PMID: Qian Y, Feng FY, Halverson S, Blas K, Sandler HM, Hamstra DA. The percent of positive biopsy cores improves prediction of prostate cancer-specific death in patients treated with dose-escalated radiotherapy. International journal of radiation oncology, biology, physics. 2011;81(3):e doi: /j.ijrobp PubMed PMID: Zumsteg ZS, Spratt DE, Pei X, Yamada Y, Kalikstein A, Kuk D, et al. Short-term androgen-deprivation therapy improves prostate cancer-specific mortality in intermediate-risk prostate cancer patients undergoing dose-escalated external beam radiation therapy. International journal of radiation oncology, biology, physics. 2013;85(4): doi: /j.ijrobp PubMed PMID: Capitanio U, Karakiewicz PI, Valiquette L, Perrotte P, Jeldres C, Briganti A, et al. Biopsy core number represents one of foremost predictors of clinically significant gleason sum upgrading in patients with low-risk prostate cancer. Urology. 2009;73(5): Epub 2009/02/07. doi: /j.urology PubMed PMID: Kattan MW, Eastham JA, Wheeler TM, Maru N, Scardino PT, Erbersdobler A, et al. Counseling men with prostate cancer: a nomogram for predicting the presence of small, moderately differentiated, confined tumors. The Journal of urology. 2003;170(5): Epub 2003/10/09. doi: /01.ju PubMed PMID: Reese AC, Landis P, Han M, Epstein JI, Carter HB. Expanded criteria to identify men eligible for active surveillance of low risk prostate cancer at Johns Hopkins: a preliminary analysis. The Journal of urology. 2013;190(6): doi: /j.juro PubMed PMID: Dall'Era MA, Albertsen PC, Bangma C, Carroll PR, Carter HB, Cooperberg MR, et al. Active surveillance for prostate cancer: a systematic review of the literature. European urology. 2012;62(6): doi: /j.eururo PubMed PMID: Caster JM, Falchook AD, Hendrix LH, Chen RC. Risk of Pathologic Upgrading or Locally Advanced Disease in Early Prostate Cancer Patients Based on Biopsy Gleason Score and PSA: A Population-Based Study of Modern Patients. International journal of! $(

48 radiation oncology, biology, physics. 2015;92(2): doi: /j.ijrobp PubMed PMID: Maurice MJ, Zhu H, Kiechle JE, Kim SP, Abouassaly R. Comorbid Disease Burden is Independently Associated with Higher Risk Disease at Prostatectomy in Patients Eligible for Active Surveillance. The Journal of urology doi: /j.juro PubMed PMID: Russo GI, Cimino S, Castelli T, Favilla V, Urzi D, Veroux M, et al. Percentage of cancer involvement in positive cores can predict unfavorable disease in men with lowrisk prostate cancer but eligible for the prostate cancer international: active surveillance criteria. Urologic oncology. 2014;32(3): doi: /j.urolonc PubMed PMID: Schiffmann J, Wenzel P, Salomon G, Budaus L, Schlomm T, Minner S, et al. Heterogeneity in D'Amico classification-based low-risk prostate cancer: Differences in upgrading and upstaging according to active surveillance eligibility. Urologic oncology. 2015;33(7):329 e13-9. doi: /j.urolonc PubMed PMID: Truong M, Slezak JA, Lin CP, Iremashvili V, Sado M, Razmaria AA, et al. Development and multi-institutional validation of an upgrading risk tool for Gleason 6 prostate cancer. Cancer. 2013;119(22): doi: /cncr PubMed PMID: Song SH, Pak S, Park S, Song C, Jeong IG, Choi HJ, et al. Predictors of unfavorable disease after radical prostatectomy in patients at low risk by D'Amico criteria: role of multiparametric magnetic resonance imaging. The Journal of urology. 2014;192(2): doi: /j.juro PubMed PMID: Romero-Otero J, Garcia-Gomez B, Duarte-Ojeda JM, Rodriguez-Antolin A, Vilaseca A, Carlsson SV, et al. Active surveillance for prostate cancer. Int J Urol doi: /iju PubMed PMID: Bul M, Zhu X, Valdagni R, Pickles T, Kakehi Y, Rannikko A, et al. Active surveillance for low-risk prostate cancer worldwide: the PRIAS study. European urology. 2013;63(4): doi: /j.eururo PubMed PMID: Newcomb LF, Thompson IM, Jr., Boyer HD, Brooks JD, Carroll PR, Cooperberg MR, et al. Outcomes of Active Surveillance for Clinically Localized Prostate Cancer in the Prospective, Multi-Institutional Canary PASS Cohort. The Journal of urology doi: /j.juro PubMed PMID: Reese AC, Feng Z, Landis P, Trock BJ, Epstein JI, Carter HB. Predictors of Adverse Pathology in Men Undergoing Radical Prostatectomy Following Initial Active Surveillance. Urology. 2015;86(5): doi: /j.urology PubMed PMID: Tosoian JJ, Mamawala M, Epstein JI, Landis P, Wolf S, Trock BJ, et al. Intermediate and Longer-Term Outcomes From a Prospective Active-Surveillance Program for Favorable-Risk Prostate Cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2015;33(30): doi: /JCO PubMed PMID: Villa L, Salonia A, Capitanio U, Scattoni V, Abdollah F, Suardi N, et al. The number of cores at first biopsy may suggest the need for a confirmatory biopsy in patients eligible for active surveillance-implication for clinical decision making in the! $)

49 real-life setting. Urology. 2014;84(3): doi: /j.urology PubMed PMID: Warren JL, Klabunde CN, Schrag D, Bach PB, Riley GF. Overview of the SEER- Medicare data: content, research applications, and generalizability to the United States elderly population. Medical care. 2002;40(8 Suppl):IV doi: /01.MLR PubMed PMID: Prasad SM, Eggener SE, Lipsitz SR, Irwin MR, Ganz PA, Hu JC. Effect of depression on diagnosis, treatment, and mortality of men with clinically localized prostate cancer. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2014;32(23): doi: /JCO PubMed PMID: ; PubMed Central PMCID: PMC Boyd CA, Benarroch-Gampel J, Sheffield KM, Han Y, Kuo YF, Riall TS. The effect of depression on stage at diagnosis, treatment, and survival in pancreatic adenocarcinoma. Surgery. 2012;152(3): doi: /j.surg PubMed PMID: ; PubMed Central PMCID: PMC Goodwin JS, Zhang DD, Ostir GV. Effect of depression on diagnosis, treatment, and survival of older women with breast cancer. J Am Geriatr Soc. 2004;52(1): PubMed PMID: ; PubMed Central PMCID: PMC Klabunde CN, Potosky AL, Legler JM, Warren JL. Development of a comorbidity index using physician claims data. Journal of clinical epidemiology. 2000;53(12): PubMed PMID: Lu-Yao GL, Albertsen PC, Li H, Moore DF, Shih W, Lin Y, et al. Does primary androgen-deprivation therapy delay the receipt of secondary cancer therapy for localized prostate cancer? European urology. 2012;62(6): doi: /j.eururo PubMed PMID: ; PubMed Central PMCID: PMC Sammon JD, Abdollah F, Reznor G, Pucheril D, Choueiri TK, Hu JC, et al. Patterns of Declining Use and the Adverse Effect of Primary Androgen Deprivation on All-cause Mortality in Elderly Men with Prostate Cancer. European urology doi: /j.eururo PubMed PMID: Burnett-Zeigler IE, Pfeiffer P, Zivin K, Glass JE, Ilgen MA, Flynn HA, et al. Psychotherapy utilization for acute depression within the Veterans Affairs health care system. Psychol Serv. 2012;9(4): doi: /a PubMed PMID: Crystal S, Sambamoorthi U, Walkup JT, Akincigil A. Diagnosis and treatment of depression in the elderly medicare population: predictors, disparities, and trends. J Am Geriatr Soc. 2003;51(12): PubMed PMID: ; PubMed Central PMCID: PMC Bill-Axelson A, Garmo H, Nyberg U, Lambe M, Bratt O, Stattin P, et al. Psychiatric treatment in men with prostate cancer--results from a Nation-wide, population-based cohort study from PCBaSe Sweden. European journal of cancer. 2011;47(14): doi: /j.ejca PubMed PMID: Stuart EA. Matching methods for causal inference: A review and a look forward. Stat Sci. 2010;25(1):1-21. doi: /09-STS313. PubMed PMID: ; PubMed Central PMCID: PMC ! %+

50 95. Meyers B, D'Agostino A, Walker J, Kritzer MF. Gonadectomy and hormone replacement exert region- and enzyme isoform-specific effects on monoamine oxidase and catechol-o-methyltransferase activity in prefrontal cortex and neostriatum of adult male rats. Neuroscience. 2010;165(3): doi: /j.neuroscience PubMed PMID: ; PubMed Central PMCID: PMC Dacal K, Sereika SM, Greenspan SL. Quality of life in prostate cancer patients taking androgen deprivation therapy. J Am Geriatr Soc. 2006;54(1): doi: /j x. PubMed PMID: Barrett-Connor E, Von Muhlen DG, Kritz-Silverstein D. Bioavailable testosterone and depressed mood in older men: the Rancho Bernardo Study. The Journal of clinical endocrinology and metabolism. 1999;84(2): doi: /jcem PubMed PMID: McIntyre RS, Mancini D, Eisfeld BS, Soczynska JK, Grupp L, Konarski JZ, et al. Calculated bioavailable testosterone levels and depression in middle-aged men. Psychoneuroendocrinology. 2006;31(9): doi: /j.psyneuen PubMed PMID: Shores MM, Moceri VM, Sloan KL, Matsumoto AM, Kivlahan DR. Low testosterone levels predict incident depressive illness in older men: effects of age and medical morbidity. The Journal of clinical psychiatry. 2005;66(1):7-14. PubMed PMID: Kumar RJ, Barqawi A, Crawford ED. Adverse events associated with hormonal therapy for prostate cancer. Reviews in urology. 2005;7 Suppl 5:S PubMed PMID: ; PubMed Central PMCID: PMC Chipperfield K, Fletcher J, Millar J, Brooker J, Smith R, Frydenberg M, et al. Predictors of depression, anxiety and quality of life in patients with prostate cancer receiving androgen deprivation therapy. Psycho-oncology doi: /pon PubMed PMID: Jones CU, Hunt D, McGowan DG, Amin MB, Chetner MP, Bruner DW, et al. Radiotherapy and short-term androgen deprivation for localized prostate cancer. The New England journal of medicine. 2011;365(2): doi: /NEJMoa PubMed PMID: Pirl WF, Siegel GI, Goode MJ, Smith MR. Depression in men receiving androgen deprivation therapy for prostate cancer: a pilot study. Psycho-oncology. 2002;11(6): doi: /pon.592. PubMed PMID: Jayadevappa R, Malkowicz SB, Chhatre S, Johnson JC, Gallo JJ. The burden of depression in prostate cancer. Psycho-oncology. 2012;21(12): doi: /pon PubMed PMID: Chipperfield K, Brooker J, Fletcher J, Burney S. The impact of physical activity on psychosocial outcomes in men receiving androgen deprivation therapy for prostate cancer: a systematic review. Health psychology : official journal of the Division of Health Psychology, American Psychological Association. 2014;33(11): doi: /hea PubMed PMID: Llorente MD, Burke M, Gregory GR, Bosworth HB, Grambow SC, Horner RD, et al. Prostate Cancer: A Significant Risk Factor for Late-Life Suicide. The American! %*

51 Journal of Geriatric Psychiatry. 2005;13(3): doi: / Parahoo K, McDonough S, McCaughan E, Noyes J, Semple C, Halstead EJ, et al. Psychosocial interventions for men with prostate cancer. The Cochrane database of systematic reviews. 2013;12:CD doi: / CD pub3. PubMed PMID: Irani J, Salomon L, Oba R, Bouchard P, Mottet N. Efficacy of venlafaxine, medroxyprogesterone acetate, and cyproterone acetate for the treatment of vasomotor hot flushes in men taking gonadotropin-releasing hormone analogues for prostate cancer: a double-blind, randomised trial. The Lancet Oncology. 2010;11(2): doi: /S (09) PubMed PMID: Ahmadi-Davis S, Velasco R, Stewart JT. Goserelin-induced depression in a man with prostate cancer. Psychosomatics. 2014;55(6): doi: /j.psym PubMed PMID: Sharpley CF, Bitsika V, Wootten AC, Christie DR. Differences in major depressive disorder and generalised anxiety disorder symptomatology between prostate cancer patients receiving hormone therapy and those who are not. Psycho-oncology. 2014;23(12): doi: /pon PubMed PMID: Shahinian VB, Kuo YF, Freeman JL, Goodwin JS. Risk of the "androgen deprivation syndrome" in men receiving androgen deprivation for prostate cancer. Archives of internal medicine. 2006;166(4): doi: /archinte PubMed PMID: ; PubMed Central PMCID: PMC Society AC. What are the key statistics about prostate cancer? [cited /20/15]. Available from:,--./ : : <4-7=>4<9?=<40.5:3-7-4@ @A4B@3-7-=3-= Potosky AL, Warren JL, Riedel ER, Klabunde CN, Earle CC, Begg CB. Measuring complications of cancer treatment using the SEER-Medicare data. Medical care. 2002;40(8 Suppl):IV doi: /01.MLR PubMed PMID: Reynolds K, Pietrzak RH, El-Gabalawy R, Mackenzie CS, Sareen J. Prevalence of psychiatric disorders in U.S. older adults: findings from a nationally representative survey. World psychiatry : official journal of the World Psychiatric Association. 2015;14(1): doi: /wps PubMed PMID: ; PubMed Central PMCID: PMC Sharpley CF, Bitsika V, Christie DR. Diagnosing 'male' depression in men diagnosed with prostate cancer: the next step in effective translational psycho-oncology interventions? Psycho-oncology. 2014;23(9): doi: /pon PubMed PMID: Donovan KA, Walker LM, Wassersug RJ, Thompson LM, Robinson JW. Psychological effects of androgen-deprivation therapy on men with prostate cancer and their partners. Cancer. 2015;121(24): doi: /cncr PubMed PMID: Newby TA, Graff JN, Ganzini LK, McDonagh MS. Interventions that may reduce depressive symptoms among prostate cancer patients: a systematic review and metaanalysis. Psycho-oncology. 2015;24(12): doi: /pon PubMed PMID: ! %"

52 118. Jung B, Stoll C, Feick G, Prott FJ, Zell J, Rudolph I, et al. Prostate cancer patients' report on communication about endocrine therapy and its association with adherence. J Cancer Res Clin Oncol doi: /s PubMed PMID: Sharpley CF, Bitsika V, Christie DR. Factor structure of a combined measure of major depressive disorder and male depression in prostate cancer patients. Psychooncology doi: /pon PubMed PMID: ! %#

53 Tables and Figures Part 1.1 Table 1: Demographic and Clinical Features(20) Demographic Characteristics Clinical Characteristics Pathologic Characteristics Age (years) Prostate Specific Antigen (ng/ml) Gleason T/N stage Range 34-92y Range Score pt3b-4 pt3b-4 pt2 pt3a N0 N1 Mean 59.5 Mean 5.2 Median 60.0 Median 5.0 IQR IQR !6 3+4 " % % % % % % % % Race Clinical T Stage Total White 8, % T1c 9, % Black 1, % T2a % Upgrade (Gleason 7-10) Other % No 5, % Marital Status Total Number of Cores Biopsied (n=5581) Yes 4, % Married 7, % Range 1 to 99 Upstage (pt3-4/n0-1) Single % Median 12 No 9, % Separated % Mean 12 Yes % Widowed % IQR Domestic Percent Positive Cores % Partner (n=5581) Unknown % Range 1-100% Residency Type Median 24% Metropolitan 9, % Mean 30% Non % IQR 12-42% metropolitan % % % % 6 0.1% % % % Total % % % 10274

54 Part 1.1 Table 2: Predictive Factors for Upgrading and Upstaging, Multivariable Regression Analysis Upgrading Upstaging Primary AOR 95% Confidence Interval p value Primary AOR 95% Confidence Interval p value Age <0.001 Age <0.001 Black Black PSA <0.001 PSA <0.001 Percent Positive Cores <0.001 Percent Positive Cores <0.001 Secondary Secondary Age> <0.001 Age> <0.001 Black Black PSA> <0.001 PSA> <0.001 Percent Positive Cores >25% <0.001 Percent Positive Cores >25% <0.001

55 Part 1.1 Figure 1: Incidence of upgrading and upstaging among low-risk prostate cancer patients!

56 Part 1.1 Figure 2: Incidence of upgrading and upstaging stratified by percept of positive biopsy cores,'# $$%"# $'# $'%)# &"%'# &&# +'$,'-*(./(+%*0'-*)( &'#!'# "'#!"# *)%+# -*"%$.# *"%,/"$.# "$%*/$'.# 0$'.# # *!%'# *'# $%&# (%'# (%+# '#!"#$%&'&(!"()*%#'&(

57 Part 1.1 Figure 3: Risk of Upgrading Stratified by PSA and Percent Total Cores Positive of 5,581 Patients with Complete Clinical Data Percent Total Cores Positive!12.5% % % >50% All PSA < All 13.6% N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N=5581 Figure 4: Risk of Upstaging Stratified by PSA and Percent Total Cores Positive of 5,581 Patients with Complete Clinical Data Percent Total Cores Positive!12.5% % % >50% Any PSA < % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N= % N=5581

58 Part 1.2 Table 1: Demographic and clinical features, compared by clinical risk group(26) Favorable- Low-risk <50% Low-Risk!50% Intermediate Variable Risk Unfavorable- Intermediate Risk High-Risk n % n % n % n % n % Total 4,485 1,122 2,843 4,008 2,444 Age at diagnosis (years) Median IQR Mean Race (self-reported) White % % % % % Black % % % % % Asian % % % % % Native America/Alaskan % 3 0.3% 9 0.3% % % Hispanic % % % % % Biopsy Gleason Score! % % % % % % 0 0.0% % % % % 0 0.0% 0 0.0% % % 8 to % 0 0.0% 0 0.0% 0 0.0% % Prostate Specific Antigen (ng/ml) Median IQR Mean Range >98.0 Percent of Total Cores Positive Median 16.7% 62.5% 25.0% 50.0% 44.4% IQR % % % % % Mean 20.0% 67.5% 24.3% 53.8% 48.7% Range % % % 0-100% 0-100% <0.001 <0.001 <0.001 <0.001 <0.001

59 Part 1.2 Table 2: Multivariable Logistic Regression to measure the association between clinical risk group and the presence of occult pt3-4 or Gleason score 8-10 at prostatectomy, by NCCN clinical risk-groups and Zumsteg defined intermediate-risk subgroups, favorable and unfavorable(26) NCCN clinical Risk-groups AOR 95% CI p-value AOR 95% CI p-value Low-risk <50% Reference <0.001 Low-risk!50% < <0.001 Intermediate-risk <0.001 Reference High-risk < <0.001 Zumsteg Clinical Risk-groups AOR 95% CI p-value AOR 95% CI p-value AOR 95% CI p-value Low-risk <50% Reference < <0.001 Low-risk!50% < <0.001 Favorableintermediate risk <0.001 Reference <0.001 Unfavorableintermediate risk < <0.001 Reference High-risk < < <0.001! $%

60 Part 1.2 Figure 1: Proportion of patients within each clinical risk-group with each pathologic outcome at prostatectomy(26)! $&

61 Part 1.2 Figure 2: Percent with pathologic T3a-T4 disease or Gleason 8-10 at prostatectomy(26)! $'