|Ahead of print publication
Metabolic syndrome, levels of androgens, and changes of erectile dysfunction and quality of life impairment 1 year after radical prostatectomy
Yann Neuzillet1, Mathieu Rouanne1, Jean-François Dreyfus2, Jean-Pierre Raynaud3, Marc Schneider4, Morgan Roupret5, Sarah Drouin5, Marc Galiano6, Xavier Cathelinau6, Thierry Lebret1, Henry Botto1
1 Department of Urology, Hospital Foch, UVSQ-Paris-Saclay University, Suresnes 92150, France
2 Department of Clinical Research and Innovation, Hospital Foch, UVSQ-Paris-Saclay University, Suresnes 92150, France
3 Sorbonne University, Paris 75013, France
4 Department of Urology, Hospital Louis Pasteur, Colmar 68000, France
5 Department of Urology, Hôpital Pitié-Salpêtrière, Assistance Publique-Hôpitaux de Paris, Sorbonne University, Paris 75013, France
6 Department of Urology, Montsouris Institute, Paris-Descartes University, Paris 75014, France
|Date of Submission||28-Apr-2020|
|Date of Acceptance||17-Nov-2020|
|Date of Web Publication||09-Feb-2021|
Department of Urology, Hospital Foch, UVSQ-Paris-Saclay University, Suresnes 92150
Source of Support: None, Conflict of Interest: None
Robust data evaluating the association of preoperative parameters of the patients with quality of life after radical prostatectomy are lacking. We investigated whether clinical and biological preoperative characteristics of the patients were associated with impaired patient-reported quality of life (QoL) and sexual outcomes 1 year after radical prostatectomy. We evaluated patient-reported outcomes among the 1343 men participating in the AndroCan trial (NCT02235142). QoL and erectile dysfunction (ED) were assessed before and 1 year after radical prostatectomy using validated self-assessment questionnaires (Aging Male's Symptoms [AMS] and the 5-item abridged version of the International Index of Erectile Function [IIEF5]). At baseline, 1194 patients (88.9%) accepted to participate. A total of 750 (55.8%) patients answered the 1-year postoperative questionnaires. Out of them, only 378 (50.4% of responders) provided answers that could be used for calculations. One year after prostatectomy, ED had worsened by 8.0 (95% confidence interval [CI]: 7.3–8.7; P < 0.0001) out of a maximum of 20. The global AMS score has worsened by 2.8 (95% CI: 1.7–3.8; P < 0.0001). ED scores 1 year postsurgery were positively correlated with preoperative age and percentage of fat mass, and negatively correlated with total cholesterol, dehydroepiandrosterone (DHEA), and androstenediol (D5); AMS were poorly correlated with preoperative parameters. QoL and sexual symptoms significantly worsened after radical prostatectomy. Baseline bioavailable testosterone levels were significantly correlated with smaller changes on AMS somatic subscores postprostatectomy. These findings may be used to inform patients with newly diagnosed prostate cancer.
Keywords: metabolic syndrome; quality of life; radical prostatectomy; sexual outcomes; testosterone deficiency
Article in PDF
|How to cite this URL:|
Neuzillet Y, Rouanne M, Dreyfus JF, Raynaud JP, Schneider M, Roupret M, Drouin S, Galiano M, Cathelinau X, Lebret T, Botto H. Metabolic syndrome, levels of androgens, and changes of erectile dysfunction and quality of life impairment 1 year after radical prostatectomy. Asian J Androl [Epub ahead of print] [cited 2021 Jun 12]. Available from: https://www.ajandrology.com/preprintarticle.asp?id=309010
| Introduction|| |
Prostate cancer is one of the most commonly diagnosed cancer types worldwide and one of the leading causes of cancer death among men. Most prostate cancer patients are localized and about two-thirds of these patients will be alive 10 years later. Therefore, the aim of any primary treatment besides maximizing survival is to preserve quality of life (QoL).,,,, Since the adverse effects of primary treatments can negatively affect disease-specific QoL especially due to sexual function impairment, it is important to delineate how these issues may be perceived by patients once the immediate effects of surgery have waned.
We recently reported some preoperative results on the QoL and erectile dysfunction (ED) in patients from a cohort study (AndroCan) involving men with localized prostate cancer about to undergo robotic radical prostatectomy (not published). The present paper reports the changes noted on QoL and ED during the first postoperative year and the 1 year after surgery and how they relate to preoperative clinical/demographic signs and symptoms, symptoms and biological parameters that are generally considered as pertaining to the metabolic syndrome, and levels of circulating sexual hormones.
| Participants and Methods|| |
AndroCan trial (ClinicalTrials.gov identifier: NCT02235142) is a prospective longitudinal cohort study conducted in 4 academic institutions, on newly diagnosed patients with localized prostate cancer referred for robot-assisted radical prostatectomy., Demographic and clinical data were collected on the day before surgery. Circulating androgens were measured prior to surgery, in accordance with the Endocrine Society guidelines and assayed in a single central laboratory.
Patients who received previous local treatment or systemic treatment that could interfere with hormonal status were excluded. The study protocol was approved by the locally competent institutional (CPP Ile-de-France VIII Ethic Committee, Boulogne-Billancourt, France; approval number 130207) review board and conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from each patient.
All patients had to complete, prior to surgery and one year later, the Aging Male's Symptom (AMS) scale and the 5-item abridged version of the International Index of Erectile Function (IIEF5) self-assessment questionnaire., For the latter, the patient could indicate that he was willing to answer but unable to do so due to the absence of a partner or the lack of sexual life. Item scores were adjusted so that a score of 0 coded no symptom and a score of 4, the higher severity. The IIEF5 and AMS scores and subscores were only calculated if at least 80% of items had been properly completed.
To determine if a dichotomous variable was associated with differences on IIEF5 and AMS scores and subscores, comparisons were done on the scores for the 2 levels of each variable using bootstrap confidence intervals and permutation t-tests to avoid artifacts that might be linked to the nonnormal distribution of variables. Since the 1-year postoperative sample was about 3 times smaller than the preoperative one, in order to allow comparisons between factors in the twin articles, we retained a bilateral significance threshold of 0.05 but also outlined as trends values between 0.05 and 0.085. Spearman's nonparametric coefficient was used to test the correlations. Again, P values between 0.05 and 0.085 were outlined.
Robust multiple regression was used to define optimal parsimonious models for the IIEF5 and AMS scores and score changes from pre- to 1-year postsurgery. “Parsimonious” in this context is defined by a model with less than 10 independent variables, all of them being individually significant. Huber's procedure with a tuning constant of 1.345 and fixed regression coefficients was used. The following categorical variables were initially entered into the model: ethnicity, history of cancer, biopsy staging (the tumor, node, and metastasis [TNM] classification), biopsy dominant grade (3 and less vs 4 and more), presence of diabetes mellitus, cardiovascular disorder, high blood pressure, hypogonadism (bioavailable testosterone [BT] <0.8 ng ml-1 or serum total testosterone [TT] <3 ng ml-1), anatomopathological dominant grade for the removed prostate (III or IV), and node involvement. The following continuous variables were initially included in the model: age, height, weight, body mass index (BMI), percentage of fat mass, waist circumference, biopsy Gleason score, prostate-specific antigen (PSA), percentage of high-density lipoprotein (HDL) cholesterol, total cholesterol, triglycerides, blood sugar, number of symptoms that may pertain to a metabolic syndrome, number of concomitant medication, follicle-stimulating hormone (FSH), luteinizing hormone (LH), sex hormone-binding globulin (SHBG), BT, TT, dihydrotestosterone (DHT), dehydroepiandrosterone (DHEA), DHEA-sulfate, androstenediol (D5), androstenedione (D4), estrone (E1), estradiol (E2), weight of removed prostate, preoperative IIEF5 score, preoperative AMS somatic, psychological, sexual subscores, and preoperative AMS global score. Since this was a hypothesis-generating study, no correction (such as Bonferroni's) was applied to probability values; a bilateral probability lower than 0.05 was considered significant. All calculations were done using NCSS 2020 (NCSS LLC, Kaysville, UT, USA).
| Results|| |
Population cohort study
A total of 750 patients (55.8% of all patients) answered the 1-year postoperative questionnaires. Out of them, only 378 (28.2% of all patients and 50.4% of responders) provided answers that could be used for calculations; 335 patients (25.0% of all cases, 32.9% of those answering preoperatively [n = 1019 patients]) provided suitable answers to the 4 questionnaires (2 preoperative and 2 in one year after surgery). Those who were responding at 1 year postsurgery, even if they indicated that they could not provide suitable scores for some items, were found, to be significantly younger, to have less concomitant medications, higher BT, and less ED and sexual problems preoperatively than those who elected not to answer [Supplementary Table 1 [Additional file 1]]. Among those who were responding to the self-assessment questionnaires 1 year postsurgery, those who were unable to provide suitable scores for erectile function due to the unavailability of sexual experiences were found, on their preoperative parameters to be significantly older, to have significantly lower BT, DHEA, and D5 and to have significantly more diabetes and erectile and sexual troubles [Supplementary Table 2 [Additional file 2]].
Demographic and clinical characteristics of patients
The mean age of the sample was 64.6 (range: 41.5–79.1) years at the time they completed the 1-year postsurgery questionnaire. Among the 378 respondents, 93 (24.6%) had indicated that, one year before, prior to surgery, they suffered from diabetes mellitus and 76 patients (20.1%) have a cardiovascular disorder. Some characteristics of the global sample are displayed in [Supplementary Table 3 [Additional file 3]].
One year after prostatectomy, ED (i.e., the global IIEF5 score) had increased (worsened) by 8.0 (95% CI: 7.3–8.7; P < 0.0001) out of a maximum of 20. The global AMS score has increased (worsened) by 2.8 (95% CI: 1.7–3.8; P < 0.0001). Among subscores, the somatic subscore did not change significantly whereas the psychological subscore significantly decreased (improved) by a mean of 0.6 (95% CI: 0.3–1.0; P = 0.0003) and the sexual subscore worsened significantly by a mean of 3.6 (95% CI: 3.1–4.0; P < 0.0001).
Baseline parameters associated with QoL 1 year postsurgery
Non-Caucasian had significantly higher scores for erectile dysfunction, and the sexual AMS subscore and the global score for aging symptoms. If they were free of concomitant treatment at baseline, their health-related QoL scores and subscores and IIEF5 score were significantly lower 1 year postsurgery. Not surprisingly, patients with more aggressive forms of cancer as ascertained by pathology fared significantly worse on QoL one year after surgery [Supplementary Table 1].
ED scores 1 year postsurgery were positively correlated with (worsened by) preoperative age and percentage of fat mass and negatively with (improved by) total cholesterol, DHEA, and D5; aging male symptoms were poorly correlated with preoperative parameters (<10% of significant correlation; [Table 1]).
|Table 1: Correlation of preoperative parameters and 1-year postoperative quality of life and erectile dysfunction scores|
Click here to view
Preoperative predictors from 303 patients allowed calculating a model that explained 14.4% of the 1-year postoperative score on IIEF5. Height and BMI were independently negative predictors of this score, while waist circumference and preoperative IIEF5 were positively correlated with it. The 3 preoperative subscores of the AMS were negatively correlated with the IIEF5 postoperative score, while the global AMS score was a positive correlate [Table 2].
|Table 2: Multiple regression model for the 5-item abridged version of the International Index of Erectile Function (baseline preoperative variables used as independent predictors)|
Click here to view
Using predictors from 335 patients, a model was developed from preoperative predictors, which explained 41.2% of the variance of the AMS global score 1 year after surgery. BMI, E2, preoperative IIEF, AMS somatic subscore, AMS psychological subscore, and pathological aggressiveness were positive predictors of the score, while weight was an independent negative correlate [Table 3].
|Table 3: Multivariate model 1-year postoperative Aging Male's Symptoms global score (baseline preoperative variables used as independent predictors)|
Click here to view
QoL evolution from pre-operation to 1 year postsurgery
Factors related to score evolution were studied. For the change of the IIEF5 score, there were a few parameters that showed significant differences: presence of diabetes mellitus (worsening by 8 points against worsening by 6 points), high blood pressure (worsening by 9 points against 7 points), and TNM (worsening by 7 points for TNM1, and by 9 points for TNM2). For the change of AMS, lower waist circumference, lack of hypertension, 3-dominant grade for biopsy, normal BT and TT, hypogonadism, and lower cancer aggressiveness were significantly associated with a smaller deterioration of QoL [Supplementary Table 4 [Additional file 4]].
Correlates of score changes were subsequently defined. Height, preoperative IIEF5 score, and preoperative sexual and global AM scores were positively and significantly correlated with erectile function deterioration. Height, preoperative IIEF, AMS somatic, psychological and sexual subscore, and AMS global scores were significantly and positively correlated with AMS deterioration. On the contrary, E2 levels were significantly negatively associated with AMS deterioration [Table 4].
|Table 4: Correlates of changes in erectile dysfunction and Aging Male's Symptoms scores from baseline, preoperative, to 1 year postsurgery|
Click here to view
Finally, multivariate models were done. The model for IIEF5 score changes only explains 14.0% of the difference variance whereas the model for AMS changes in global score explains 28.0% of the difference variance [Table 5]. In the latter case, preoperative weight and preoperative global AMS score were independent correlates of an increased magnitude of the difference (QoL improvement), whereas preoperative BMI, and E2, preoperative IIEF5, AMS somatic and psychological subscores, and cancer aggressiveness were independent correlates of QoL worsening.
|Table 5: Multiple regression for the change in the 5-item abridged version of the International Index of Erectile Function scores and in Aging Male's Symptoms global score (from preoperation to 1 year postoperation)|
Click here to view
| Discussion|| |
To our knowledge, we report for the first time an evaluation of aging male symptoms and circulating sexual hormones together with QoL and sexual function, 1 year after surgery in a cohort of newly diagnosed prostate cancer patients treated by radical prostatectomy. There was a significant alteration of sexual outcomes 1 year after surgery, and this effect is one of the main reasons for the alteration of their QoL. Recent analyses from a large prospective cohort and from the ProtecT trial identified similar patterns of adverse effects in patients with clinically localized prostate cancer following radical prostatectomy., In the study reported by Chen and colleagues, radical prostatectomy (n = 469 mostly robot assisted) was associated with sexual dysfunction and urinary incontinence over two years. In our study, we evaluated QoL and ED, and outcomes with their relation to baseline levels of circulating sexual hormones, and of clinical and biological parameters. Baseline testosterone levels and metabolic syndrome parameters were independent factors associated with impaired QoL and sexual functioning 1 year after surgery in men aged 40–80 years, found to have a localized prostate cancer requiring robot-assisted surgery. Correlations of clinical factors and preoperative testosterone and its precursors and metabolites could also be ascertained.
There was a clear worsening of ED (loss of a mean 1.5 points per item) and QoL (loss of about one-sixth of a level per item for each AMS item). Some preoperative parameters are associated with the 1-year outcome on ED and QoL, namely concomitant medications, a surrogate marker of good/poor health status, and the results of anatomopathological examination of the resected prostate. With regard to sexual hormones, FSH, LH, DHEA, D5, and DHEA-sulfate are significantly correlated with IIEF5, AMS psychological subscore, or AMS global score. E2 also appears as a significant independent factor of IIEF5 one year after surgery. On the contrary, AMS somatic subscore does not seem to be significantly associated with any androgen level.
Low levels of bioavailable and total testosterone were found to be significantly associated with a higher deterioration of the AMS sexual subscore; preoperative ED and moreover QoL preoperative values were significantly and positively associated with the changes in IIEF5 and AMS scores and subscores (the higher the initial levels, the larger the changes from pre- to postoperative for the relevant score). E2 is the only androgen that is significantly correlated with changes in QoL subscores – the lower its preoperative levels, the larger the changes from pre- to postoperative QoL.
In addition, lower levels of BT are associated with a lower AMS somatic subscore, whereas such a relation is lacking for TT. Baseline parameters that are generally related to a metabolic syndrome are associated with a few significant differences on QoL and/or ED.
Reduced testosterone levels have been increasingly recognized to be a risk factor for high-grade prostate cancer. In the AndroCan trial, we observed that baseline testosterone deficiency, i.e., low total testosterone and/or low bioavailable testosterone, was independently associated with higher prostate cancer aggressiveness.
The physiological mechanisms underlying the detrimental effect of testosterone deficiency in the early stages of prostate cancer are still debated. Population-based studies have shown that obesity, diabetes mellitus, and metabolic syndrome could be linked to aggressive prostate cancer, the risk of biochemical recurrence, and prostate cancer mortality., Interestingly, recently reported data showed that testosterone therapy following radical prostatectomy reduced biochemical recurrence by about 50%.
It has also been recently suggested that prostate cancer risk might be reduced by testosterone therapy owing to improvements in metabolic syndrome components, such as elevated blood sugar levels or elevated insulin levels., The complex relationship of testosterone and prostate cancer allows speculations on the possible positive effect (notably on sexual outcomes) of supplementation of testosterone in patients with low baseline testosterone even if this approach challenges the current paradigms.
Previously, questionnaire studies evaluating QoL and sexual function have been reported in advanced and metastatic cancer patients., Even in these patients, the response rate to mailings confirms that QoL and sexual function in particular continue to be major issues for patients with an active disease and cancer survivors. In our study, we noted a slight but significant reduction (improvement) of the psychological subscore (irritability, nervousness, and anxiety) of the AMS (0.58, 95% CI: 0.25–0.91; P = 0.0003) 1 year after radical prostatectomy. This result is in line with the improved emotional and cognitive functioning 12 months after radical prostatectomy reported in a prospective longitudinal cohort (n = 209) of prostate cancer patients treated with radical prostatectomy, possibly consequent to patients' relief as the expected that the malignant tumor had been removed. One should be aware that this improvement was not sufficient to compensate for the sexual dysfunction as shown by the worsening of the global AMS score. Physicians may use these data to better individualize counseling when baseline risk factors are not decisive.
Unfortunately, our data do not provide a straightforward and definite answer on the possible relationship between cancer aggressiveness and QoL. The cancer aggressiveness, dominant grade 4 on the anatomopathological examination of the resected prostate, independently increases (worsens) the AMS global score. Although statistically significant, this effect is moderate with increases of 2 points of the 1-year AMS global score, i.e., about 10%; this impairment of QoL may be a direct effect of cancer or a consequence of more aggressive cancer treatments being administered to such patients.
One limitation of our study is not to connect our results on QoL and ED to current levels of androgens. However, this would have severely hampered our initial recruitment as most patients would not have agreed to come back as inpatients, even for one night, one year after surgery, just to have blood collected. In addition, our own pilot study and the 3 published ones do show that confounding factors are so important in nomad assessment that the latter are extremely unlikely to clarify the results based on the recommended assessment done preoperatively.
Another severe limitation is the large number of patients that did not answer the questionnaires after a 1-year follow-up. In fact, the IIEF5 score is actually designed as a screening tool and thus uses quite direct questions about erectile function. It was retained because of its brevity to limit the number of intimate questions. Unfortunately, the wording of the items is probably too direct and led about half of the responders to state they could not provide a suitable answer. Thus, our final sample was reduced to about one-third of our initial sample. If such a study was to be reproduced, one should consider amending the questionnaire in order to assess ED in ways that would not deter participants with limited or no recent sexual encounters. The loss of such a large proportion of patients makes it difficult to establish if some parameters are truly not associated with QoL or ED or whether the absence of a significant association is related to the vastly reduced power of the experiment. Nevertheless, the size of our analyzable patients' sample (that is 350 at 1 year) gives us some protection in terms of power. Last but not least, no standard procedure for nerve sparing surgery was stated in the study protocol as this type of surgery is highly operator dependent. However, the recommendation to use such a procedure was reiterated at each monitoring visit. Nevertheless, it cannot be assumed that such a procedure was used and was successful in all study participants. While we have undertaken to produce videos to standardize such an approach in a forthcoming study, many surgeons were highly reluctance to switch to an unfamiliar procedure and this issue is likely to be a recurrent problem in future studies.
Regarding to methodology, since the study is ongoing, we are not yet able to determine if one year is adequate to observe period for the whole effects on QoL. Prior trials showed that there may be little change in sexual outcomes after 1 year.,, Finally, concomitant medications during the 1-year postoperative period may have affected patients' QoL and ED independently of radical prostatectomy. Limited data are available on this topic.
| Conclusions|| |
In this cohort of men with localized prostate cancer, general QoL and sexual symptoms were, as expected, significantly affected by some demographic/clinical characteristics. TT and BT had different effects on our outcome parameters. In addition, testosterone precursors and metabolites also show differential effects on the same outcome parameters. These findings may be used to inform patients with newly diagnosed prostate cancer. Further study of the influence of concomitant medications on QoL/ED would be desirable.
| Author Contributions|| |
YN drafted the manuscript and helped with data acquisition. JFD drafted the manuscript and performed statistics. JPR helped with establishing the study design and reviewing the manuscript. M Rouanne drafted the manuscript and helped with data acquisition. MS, M Roupret, SD, MG, XC, and TL helped with data acquisition and reviewing the manuscript. HB established the study design and helped with drafting the manuscript. All authors read and approved the final manuscript.
| Competing Interests|| |
All authors declared no competing interests.
| Acknowledgments|| |
The authors wish to thank the subjects and health-care staff who participated in this study. This study was funded by the Foch Foundation, a nonprofit institution, and a grant of the French Ministry of Health/DGOS/CRC3F.
Supplementary Information is linked to the online version of the paper on the Asian Journal of Andrology website.
| References|| |
Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin
2020; 70: 7–30.
Sieh W, Lichtensztajn DY, Nelson DO, Cockburn M, West DW, et al
. Treatment and mortality in men with localized prostate cancer: a population-based study in California. Open Prost Cancer J
2013; 6: 1–9.
Barocas DA, Alvarez J, Resnick MJ, Koyama T, Hoffman KE, et al
. Association between radiation therapy, surgery, or observation for localized prostate cancer and patient-reported outcomes after 3 years. JAMA
2017; 317: 1126–40.
Chen RC, Basak R, Meyer AM, Kuo TM, Carpenter WR, et al
. Association between choice of radical prostatectomy, external beam radiotherapy, brachytherapy, or active surveillance and patient-reported quality of life among men with localized prostate cancer. JAMA
2017; 317: 1141–50.
Donovan JL, Hamdy FC, Lane JA, Mason M, Metcalfe C, et al
. Patient-reported outcomes after monitoring, surgery, or radiotherapy for prostate cancer. N Engl J Med
2016; 375: 1425–37.
Lardas M, Liew M, van den Bergh RC, De Santis M, Bellmunt J, et al
. Quality of life outcomes after primary treatment for clinically localised prostate cancer: a systematic review. Eur Urol
2017; 72: 869–85.
Nolte S, Liegl G, Petersen MA, Aaronson NK, Costantini A, et al
. General population normative data for the EORTC QLQ-C30 health-related quality of life questionnaire based on 15,386 persons across 13 European countries, Canada and the Unites States. Eur J Cancer
2019; 107: 153–63.
Neuzillet Y, Raynaud JP, Dreyfus JF, Radulescu C, Rouanne M, et al
. aggressiveness of localized prostate cancer: the key value of testosterone deficiency evaluated by both total and bioavailable testosterone: AndroCan Study results. Horm Cancer
2019; 10: 36–44.
Sanda MG, Cadeddu JA, Kirkby E, Chen RC, Crispino T, et al
. clinically localized prostate cancer: AUA/ASTRO/SUO guideline. Part II: recommended approaches and details of specific care options. J Urol
2018; 199: 990–7.
Handelsman DJ, Wartofsky L. Requirement for mass spectrometry sex steroid assays in the Journal of Clinical Endocrinology and Metabolism
. J Clin Endocrinol Metab
2013; 98: 3971–3.
Rosen RC, Cappelleri JC, Smith MD, Lipsky J, Peña BM. Development and evaluation of an abridged, 5-item version of the International Index of Erectile Function (IIEF-5) as a diagnostic tool for erectile dysfunction. Int J Impot Res
1999; 11: 319–26.
Heinemann LA, Saad F, Zimmermann T, Novak A, Myon E, et al
. The Aging Males' Symptoms (AMS) scale: update and compilation of international versions. Health Qual Life Outcomes
2003; 1: 15.
Montgomery D, Peck E, Vining G. Introduction to Linear Regression Analysis. Oxford: Wiley Bleackwell; 2012. p67–133.
Neal DE, Metcalfe C, Donovan JL, Lane JA, Davis M, et al
. Ten-year mortality, disease progression, and treatment-related side effects in men with localised prostate cancer from the ProtecT randomised controlled trial according to treatment received. Eur Urol
2020; 77: 320–30.
1Kasper JS, Liu Y, Giovannucci E. Diabetes mellitus and risk of prostate cancer in the health professionals follow-up study. Int J Cancer
2009; 124: 1398–403.
Lee J, Giovannucci E, Jeon JY. Diabetes and mortality in patients with prostate cancer: a meta-analysis. Springerplus
2016; 5: 1548.
Lavalette C, Trétarre B, Rebillard X, Lamy PJ, Cénée S, et al
. Abdominal obesity and prostate cancer risk: epidemiological evidence from the EPICAP study. Oncotarget
2018; 9: 34485–94.
Edward AT, Huynh LM, Towe MM, See K, El Khatib F, et al
. Is there a role for testosterone replacement therapy in reducing biochemical recurrence following radical prostatectomy? J Clin Oncol
2019; 37: 5085.
Cheetham TC, An J, Jacobsen SJ, Niu F, Sidney S, et al
. Association of testosterone replacement with cardiovascular outcomes among men with androgen deficiency. JAMA Intern Med
2017; 177: 491–9.
Lopez DS, Huang D, Tsilidis KK, Khera M, Williams SB, et al
. Association of the extent of therapy with prostate cancer in those receiving testosterone therapy in a US commercial insurance claims database. Clin Endocrinol (Oxf)
2019; 91: 885–91.
Downing A, Wright P, Hounsome L, Selby P, Wilding S, et al
. Quality of life in men living with advanced and localised prostate cancer in the UK: a population-based study. Lancet Oncol
2019; 20: 436–47.
Rouanne M, Massard C, Hollebecque A, Rousseau V, Varga A, et al
. Evaluation of sexuality, health-related quality-of-life and depression in advanced cancer patients: a prospective study in a Phase I clinical trial unit of predominantly targeted anticancer drugs. Eur J Cancer
2013; 49: 431–8.
Shin DW, Lee SH, Kim TH, Yun SJ, Nam JK, et al
. Health-related quality of life changes in prostate cancer patients after radical prostatectomy: a longitudinal cohort study. Cancer Res Treat
2019; 51: 556–67.
Potosky AL, Harlan LC, Stanford JL, Gilliland FD, Hamilton AS, et al
. Prostate cancer practice patterns and quality of life: the Prostate Cancer Outcomes Study. J Natl Cancer Inst
1999; 91: 1719–24.
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5]