Androgen deprivation therapy and depression in the prostate cancer patients: review of risk and pharmacological management

Abstract Purpose: Despite the effectiveness of androgen deprivation therapy in advanced prostate cancer, serious neuropsychiatric consequences in androgen deprivation therapy (ADT)-treated patients, mainly depression, have been concerning and gained more attention recently. This narrative review aims to shed light on the risk and pharmacological management of ADT-induced depression in PCa patients. Methods: We searched PubMed, Scopus and Google Scholar databases using MESH keywords “Prostate cancer OR prostate neoplasm” AND “Depression” AND “Androgen Deprivation Therapy” AND “antidepressants”. Search was limited to English and studies conducted on humans. Studies’ titles and abstracts were screened, and further information were obtained from the text, if necessary, to decide whether studies are to be included in this review. Results: Our review revealed 23 studies confirming the occurrence and worsening of depressive symptoms in ADT-treated patients, which frequently require pharmacological interventions; whereas 10 studies indicated otherwise. All studies were prospective, retrospective, cross-sectional or case reports. Based on the incidence of depression provided by the observational studies, the average among ADT-treated patients was 18.23% (range: 2.1–46.9%), while it was 8.42% (range: 1.4–23.3%) in the non-ADT patients. Although several treatments have been used for depression in cancer patients, current knowledge lacks observational and controlled studies as well as clinical guidelines that demonstrate efficacy and safety of antidepressants and guide clinicians to the appropriate treatment in these patients, respectively. On the other side, a few clinical studies have been published regarding the efficacy of selective serotonin reuptake inhibitors, selective serotonin and norepinephrine reuptake inhibitors and/or saftey on other ADT associated adverse effects. Conclusions: Our work supports the recent attention towards mood issues as an adverse effect of ADT, and that greater awareness of this is warranted among clinicians. Clinical studies published regarding the use of antidepressants for other ADT associated adverse effects established the foundation that can be adopted to examine these therapies on ADT-induced depression.


Introduction
Androgen deprivation therapy (ADT) represents the cornerstone of treatment for advanced and metastatic prostate cancer (PCa) [1]. Although it has shown an improvement in the disease-free survival (time to clinical failure or death from any cause) when used as neoadjuvant or adjuvant therapy [2-4], several adverse events have been attributed to it including gynecomastia, loss of libido, sexual dysfunction, hot flashes (HFs), sarcopenia, osteopenia, osteoporosis, metabolic syndrome and cardiovascular-related adverse events (i.e. CHF and MI) [5-8]. Furthermore, ADT was shown to predispose patients to higher risk of neurocognitive [9,10] and/or psychological complications, particularly depression [5,[11][12][13][14][15][16][17][18]. This created a speculation that depression could potentially occur as a result of low androgens levels and/or activity [19][20][21][22][23][24][25], which is the major mechanism underlying ADT in PCa. Despite the current medications used for depression in cancer patients, studies that were published recently revealed unclear results regarding their effectiveness in the PCa patients [26][27][28][29][30][31], and no research has been conducted so far to investigate their effectiveness in ADT-treated PCa patients. The aims of this review are: � To highlight the risk of depression with PCa, � To review the current state of evidence surrounding ADT-induced depression in the PCa patients, and � To discuss pharmacotherapies used to treat depression associated with ADT in PCa patients.
To achieve these aims, authors systematically searched PubMed, Scopus and Google Scholar databases using MESH keywords "Prostate cancer OR prostate neoplasm" AND "Depression" AND "Androgen Deprivation Therapy" AND "antidepressants." Search was limited to English and studies conducted on humans. Studies' titles and abstracts were screened, and further information were obtained from the text, if necessary, to decide whether studies are to be included in this review.

Prevalence of depression in cancer patients
Depression affects millions of people globally and represents one of the most common psychiatric sequelae in cancer patients [32]. It manifests in low mood, insomnia, fatigue, loss of pleasure in activities, and hopelessness. The prevalence of depression among cancer patients widely ranges from 2% to 60 % due to the influence of several factors, such as type of cancer, type of treatment, treatment phase, health status of the patient, socio-demographic characteristics, method of depression assessment, settings where patients were assessed, and the overall quality of reporting studies [32][33][34][35][36].

Recognition and consequences of depression in cancer patients
When depression coexists with cancer, it could lead to an overutilization of healthcare resources, resulting in a significant medico-economic burden [37]. More importantly, it can compromise the patients' health and interfere with their survivorship [38]. For instance, it has been shown as a risk factor for suicide-related mortality among cancer patients [39]. A meta-analysis revealed that depression led to 39% increase in the risk of mortality among cancer patients [40]. On the other hand, a significant prolongation in the patients' lives was associated with an early detection and treatment of depression, as it was demonstrated by an approximately twenty-nine-month increase in the survival time among women diagnosed with metastatic breast cancer [41].
As a result, primary and secondary prevention of depression has become a cancer care priority. Yet, inadequate attention towards depression recognition and treatment has been reported in this category of patients, evident by the 50-60% of depression cases with cancer remaining undiagnosed and untreated [32,42,43]. This could be attributed to multiple reasons ranging from the lack of spontaneous expression of depressive symptoms (somatic and cognitive) by these patients (e.g. only 4% of depressed cancer patients expressed feeling of guilt and failure vs. 57% of depressed non-neoplastic patients), difficulty in exploring depressive symptoms by clinicians as they may vary throughout the trajectory of disease and type of therapy, to the significant challenge in determining whether their root cause is cancer, cancer therapies or other comorbidities [42,44].

Treatment of prostate cancer
Prostate cancer (PCa) is the second most common non-cutaneous malignancy in males worldwide [45]. It was shown to be highly prevalent among western men [46]. Multiple therapeutic modalities have been established to treat it including but not limited to watchful waiting, active surveillance (AS), prostatectomy (RP), external beam radiotherapy (XRT), chemotherapy (CT), immunotherapy, and pharmacological castration or androgen deprivation therapy (ADT). urological and sexual functions would further deteriorate their condition and elicit depression. Recently, a meta-analysis that included data of 4494 PCa patients from 27 studies was conducted to estimate the prevalence of depression among these patients [52]. Despite the utilized treatments, the prevalence was found to reach 17%, 15%, and 18% in the pre-treatment, on-treatment, and post-treatment groups, respectively. Overall, it should be noted that the mean age of patients in this study was 66.3 years (57.5 À 73.2 years), and approximately 73% had a localized PCa. Thus, these results must be cautiously interpreted when it comes to the elderly and, more particularly, those with metastatic PCa.
Concerning the age, one study showed higher depression rates among older PCa patients (>75 years) compared to the younger patients [38]. On the other hand, another study reported no difference in depression rates between younger patients (�65 years) and those over 65 years [53]. Despite that, the impact of age on developing depression among PCa is still controversial and requires further investigation given the fact that the majority of men diagnosed with PCa are at advanced ages [47].

Impact of the stage of prostate cancer on depression in prostate cancer patients
Although the aforementioned meta-analysis [52] lacked demonstrating the correlation between metastatic disease and depression, a previous study showed that patients with advanced stages were significantly more depressed compared to those with localized disease [53]. This is reasonable given the fact that these patients experience severe symptoms, higher rates of treatment failure, and shorter life expectancy.

Impact of comorbidities on depression in prostate cancer patients
Lastly, the presence of comorbidities, such as cardiovascular disease, diabetes, and arthritis, in men with PCa was found to increase the risk of depression compared to those free of comorbidities [54].
Collectively, it is important to highlight that age, disease stage, depression history, and other comorbidities could influence the correlation between any PCa therapy and depression. However, for this review, all these factors will be emphasized on when discussing each study in order to demonstrate the actual correlation between ADT and depression in PCa patients.

Androgen and its relationship with neurocognition
In non-human primates and rodents, androgen receptors (ARs) are abundantly expressed in the hippocampus, amygdala, hypothalamus, and cerebral cortex [55,56], areas in the brain that are linked to the control of the mood and pathogenesis of depression [57][58][59]. ARs expression pattern in the human brain appears to be similar to other mammals [55], although this remains to be further researched. Testosterone is lipophilic in nature [55] and has been shown to cross the blood brain barrier (BBB) and exists in the aforementioned regions of the human brain [60,61]. Furthermore, its role in mood regulation has been closely related to the ARs expression [55]. Low level of free testosterone in men was correlated with reduced synaptic density and attenuated activity of the hippocampus and amygdala that predisposed to an increased risk of depression [23,56,62-65], whereas testosterone supplementation was shown to boost the mood of the hypogonadal men [63,66-71]. More importantly, recent research has demonstrated an activation in the serotonergic system in the brain of male rodents [72], non-human primates [73], and humans [74] secondary to an increase in the circulating or free testosterone, which supports the notion that testosterone has a positive feedback effect on the central serotonergic system. This has yielded a general assumption regarding the association between reduced testosterone levels and developing depression in the ADT-treated PCa patients (see Figure 1). Below is a detailed discussion of the most important studies published on this topic (summarized in Tables  1 and 2).

Studies showing no rise in the risk of depression with androgen deprivation therapy
In 2003, Salminen and colleagues assessed the effect of ADT [antiandrogen (Flutamide) and luteinizing hormone releasing hormone agonist (LHRH analog)] on emotional, social, and cognitive functions as well as the mood in newly diagnosed PCa patients, who had no history or signs of any neurological or psychiatric disorder and had been determined to administer ADT and XRT for up to 12 months (n ¼ 25) [75]. The PCa patients were prospectively compared with healthy matched controls (n ¼ 52) with an overall mean age of approximately 65 years (46-77 years) at baseline, 6 and 12 months. Although 84% of the patients had  In 2008, Pril and colleagues studied depression and fatigue in men diagnosed with locally advanced or recurrent non-metastatic PCa [mean age: 62 years (46-84); n ¼ 48], who were treated with ADT (either leuprolide (n ¼ 23) or bicalutamide (n ¼ 25)) for 12 months [76]. Patients in each treatment group were evaluated for fatigue using fatigue severity scale (FSS), Figure 1. Schematic demonstration of ADT-induced depression in prostate cancer patients. In healthy men, normal testosterone level is believed to maintain centrally serotonergic and adrenergic neuronal activity, hence functioning properly with no depression symptoms developed in these individuals. However, ADT use in PCa results in lowering testosterone levels, that is believed to attenuate these neurons (either through affecting the levels of these neurotransmitters or androgen receptors), hence developing depression among theses patients. ADT: Androgen deprivation therapy; DA: Dopamine; GnRH: Gonadotropin-releasing hormone; LHRH: Luteinizing hormone releasing; NE: Norepinephrine; 5-HT: 5-hydroxytryptamine or serotonin. The figure was created with BioRender.com. and depression symptoms using BDI at baseline, 6 and 12 months. Despite the significantly increased fatigue over the study period, ADT appeared to have no effect on depression reflected by non-significant change in BDI scores across the three timepoints when the whole sample or each group's results at baseline were compared to the respected 6-or 12-month results.
Timilshina and colleagues conducted a study similar to Salminen et al., with an additional arm including non-ADT treated PCa patients (ADT vs. non-ADT vs. control; n ¼ 85, 86, 86, respectively) [77]. All arms had matched age [mean age: 69.1 years (60-80)] and psychiatric disorders rates at baseline. PCa patients in both groups had a non-metastatic disease and undergone RP, XRT, both, or remained untreated at their inclusion. Most of the ADT users were treated for at least 1 year and all participants were evaluated for depression at baseline, 3, 6, and 12 months using Geriatric depression score (GDS). Overall, no significant differences have been observed neither in depression symptoms nor rates even after considering the history of depression and length of ADT usage among PCa participants.
In 2013, Wiechno and colleagues compared the effect of adjuvant ADT after XRT (n ¼ 88) to XRT alone (n ¼ 61) on the severity and diagnosis of depression among PCa patients [78]. ADT receiving patients were significantly younger [median 67 vs. 72 years (50-85)] and had higher proportion of those diagnosed with T3 tumors compared to XRT alone group, who were prominently diagnosed with T2 tumors. Using the Hospital Anxiety and Depression Scale (HADS), the adjuvant ADT and XRT alone groups had similar mean scores of anxieties (6.52 vs. 6.39) and depression (5.93 vs. 5.94), respectively. Moreover, the rate of depression in both groups was 30%. Collectively, this indicates that ADT does not worsen nor does precipitate depression among the PCa patients treated with XRT as primary therapy.
In the same year, Hervouet and colleagues investigated the impact of neoadjuvant or adjuvant ADT [antiandrogen (bicalutamide) with LHRH analog (Goserelin or Leuprolide)] to XRT (n ¼ 28) compared to XRT alone (n ¼ 32) on the symptoms and incidences of depression in non-metastatic PCa patients [79]. The mean age of patient was 70.8 years (56-80) with no significant difference in the diagnosed psychiatric disorders and use of sleep or mood medications at baseline between the two groups. Using three depressive symptoms measuring scales, HADS, BDI-II, and Structured Interview Guide for Hamilton Depression Rating Scale (SIGH-D), the mean depression scores of ADT-XRT group did not significantly change compared to the XRT's at baseline and during the treatment period (1, 2, 4, 6, 8, 12, 16 months). However, pairwise analysis within the ADT-XRT group showed a significantly higher mean of BDI-II score during the 16th month compared to other time points (p < 0.05) except the 12th month, whereas nothing was observed when similar analysis was applied to XRT group. With regards to the depressive disorder incidences, 10.1% of the ADT-XRT group developed depression criteria vs. 13.2% of XRT group, indicating no significant changes between the two treatment approaches. Applying pairwise analysis on the XRT group showed a significant increase in the incidence of depressive disorder in the 12th compared to the 8th month only (12.3% vs. 23%, p ¼ 0.05) that decreased significantly in the 16th (23% vs. 12.3%, p ¼ 0.05). On the other hand, nothing was observed when a similar analysis was applied to ADT-XRT group. Although ADT introduction tended to consistently increase the mean of BDI-II scores in the ADT-XRT group only as shown by the intragroup analysis measured up to the 6th month and compared to the baseline, it did not have any significant impact neither on depression measures nor depression incidences upon comparing to the XRT group (intergroup comparison) within the 6-month time frame. Regarding the effect imposed by the length of ADT use, subintragroup analyses performed on the ADT-XRT group during the 12th and 16th months showed that short term users (< 9 months) tended to have lower but non-significant mean depression scores (SIGH-D) on both time points as well as lower mean HADS scores on the 12th month point only in comparison to the long term users (> 16 months), indicating that longer ADT usage predisposes higher risk of depression. Despite the significantly worse diagnosis of ADT group (advanced stage and elevated PSA), the authors concluded no risk of depression associated with ADT among the PCa patients.
In 2020, Hou� ed� e and colleagues assessed the quality of life prospectively in newly diagnosed PCa patients after 3-year treatment in comparison to healthy controls using QLQ-C30 questionnaire [80]. All PCa patients had no hormonal therapy in the previous year and no relapse in the intervening years. Both patient (n ¼ 367) and control (n ¼ 188) groups were well-matched, including age with a mean of 67.9 years. Patients received one of the following treatment modalities: active surveillance (AS; n ¼ 61), RP (n ¼ 285), XRT (n ¼ 90), brachytherapy or High-intensity Focused Ultrasound (HIFU) (n ¼ 52), ADT (n ¼ 26), chemotherapy (CT, n ¼ 15) or combined care (CC). Although all patients (including those treated with ADT) appeared to have a significantly poorer function and more dyspnea compared to controls (24.3 % vs. 16.3 % and 22 % vs. 12.4 %, respectively), no significant difference in suspected and probable depression, using HADS, was found between patients and controls (14.9 % vs. 12.5 %, respectively p ¼ 0.44) and within the patients' group (AS, RP, EBRT/Brachytherapy/HIFU/ ADT, CC; 13.2 %, 18.6 %, 10.2 %, 11.8 %, respectively p ¼ 0.16).
Recently, Alonso-Quiñones and colleagues evaluated the risk of mild cognitive impairment (MCI) with ADT prospectively in PCa patients every 15 months after recruitment [81]. According to the ADT use and length of treatment, PCa patients were stratified into 3 groups including non-ADT users (n ¼ 174), ADT users <5 years (n ¼ 47) and ADT users �5 years (n ¼ 20), with an overall median age of 78 years, to further identify the correlation between MCI and length of ADT. BDI results indicated no difference in baseline depression between ADT compared to non-ADT patients. Despite the significantly increased comorbidities and a higher proportion of those with metastatic disease among the ADT users, there was no association between MCI and ever using ADT or non-using ADT. More importantly, although ADT use >5 years tended to increase the risk of MCI, no such association was revealed with short-term users.
In another study, Sanchez-Mart� ınez and colleagues evaluated brain functions of PCa patients treated with ADT (n ¼ 33; mean age 70.8 years, treated with leuprorelin or triptorelin) for 1 year prospectively in the first six to twelve months of treatment and in 12 months following the first evaluation [82]. Around two-third had RP previously, whereas 7 patients had metastatic PCa. Despite the significant increase in insomnia after one year of the initial assessment, such association was not seen in depressive symptoms and rates using GDS. Nevertheless, depressive symptoms and rate of impaired cases among the elderly (�75 years) only tended to be worse during the follow-up assessment compared to the younger patients.
Lastly, additional one arm prospective study was conducted by Cinar and colleagues where they performed 3-and 6-month evaluation of the cognitive functions and neuropsychological symptoms of men (mean age of 69 years) diagnosed with metastatic PCa, except to brain, and treated with ADT quarterly (bicalutamide 50 mg daily for at least 10 d then leuprolide every 3 months) [83]. Despite the significant decline of testosterone levels 3-and 6-months post treatment compared to the baseline levels as well as to each other, patients' cognitive functions remained similar. More importantly, their depressive symptoms have not significantly changed as shown by BDI scores, 11.15, 11.78 and 12.00, at baseline, 3 months and 6 months, respectively (p ¼ 0.61).

Studies showing rise in the risk of depression with androgen deprivation therapy
While the previous studies did not show an average increase in depression scores with ADT, this does mean no individual experienced depression. In fact, the impact of smaller subsets of individuals who had increased depression scores may not significantly alter the results of the group as a whole, yet remained questionable. Consequently, additional studies have revisited this association and indicated otherwise (summarized in Table 2).
The first was a case report published by Rosenblatt and Mellow in 1995 that included three patients cases [84]. The first case was about a 76-year-old male with a Gleason score 7, stage D PCa and prescribed leuprolide and flutamide. Two months later, the patient started suffering from depression and was started on pharmacological therapy; however, it was minimally effective. Thus, leuprolide was discontinued, which improved depressive symptoms considerably. After a few months, leuprolide was re-introduced and the patient was switched to another antidepressant; however, depressive symptoms returned. Despite the effort in controlling his depression, the patient decided on his own to discontinue ADT and depression-treating regimens that resulted in a total disappearance of his depression symptoms. The second case was for an 84year-old male diagnosed with Gleason score 5, stage B2 PCa and treated with leuprolide. Nine months later, the patient was diagnosed with depression and prescribed various antidepressants. However, all were not effective nor tolerated. Thus, leuprolide was discontinued, which completely resolved his depression. Noteworthy, when leuprolide was re-introduced later, depression recurred. Therefore, the patient elected to discontinue leuprolide which markedly improved his mood. The third case was about an 81-year-old male with Gleason score 8, stage D PCa. Although he had a history of depression that was pharmacologically controlled, his depressive symptoms did worsen three months after leuprolide initiation, and antidepressant regimen became ineffective. Eventually, leuprolide was terminated and depressive symptoms completely resolved within 2 months.
Brawer MK reported another case for a 52-year-old male with T2B PCa, who achieved castration after RP for approximately 3 years [85]. Later, the patient had a biochemical recurrence, and was administered adjuvant XRT preceded by a 3-month injection of goserelin. After 16 months of the single injection, depression started developing and was not responding to treatment. Testosterone was cautiously initiated, which markedly resolved his symptoms. Later, testosterone was recommended to be continued and tapered gradually to prevent depression relapse.
In 2006, Shahinian and colleagues conducted a comparative, retrospective study on Surveillance, Epidemiology and End-Results (SEER) Medicare data for men diagnosed with PCa between 1992 and 1997 that were either treated with ADT (n ¼ 15,748) or not (n ¼ 34,865) vs. non-cancer controls (n ¼ 50,476) [86]. PCa patients in both groups had received RP, XRT, both, or not received any intervention. Classification under ADT group was considered if patients were to receive GnRH agonists or orchiectomy within 6 months of PCa diagnosis. Those who received ADT or orchiectomy after 6 months of diagnosis were excluded. The rest of PCa patients will be classified under non-ADT group. ADT users tended to be older, had more advanced and aggressive cancers compared to the non-ADT users (median age: 75 years vs. 72 years; stage IV: 20% vs. 2.9% respectively). Interestingly, although the proportion of patients with depressive disorder among PCa non-ADT users was similar to the non-cancer controls (9.5% vs. 9.6%, respectively), it was significantly higher in the ADT users (12.1%; 95% CI, 11.3%-12.8%) compared to both groups. Furthermore, when the analysis was limited to the PCa patients only, the risk of depressive disorders with ADT was significantly higher compared to the other group before and after adjustment, (RR, 1.37; 95% CI, 1.30-1.44; RR, 1.08; 95% CI, 1.02-1.15, respectively).
In 2006, an additional case report was published by Kohen and Koppel regarding a 73-year-old male with no history of depression, who was diagnosed with PCa and determined to be initiated on goserelin injection every 2 weeks [87]. After the 1st injection, the patient started complaining of depressive symptoms and was started on an antidepressant. However, his symptoms continued to worsen especially after the 2nd injection, hence referred to a psychiatrist for further evaluation.
In 2011, van Tol-Geerdink and colleagues investigated prospectively whether depressive symptoms among the non-metastatic PCa patients worsen with neoadjuvant (3 months) and adjuvant (3 months) ADT (goserelin þ bicalutamide) to XRT and whether depression is caused by ADT or higher prognostic risk of the disease [88]. PCa patients (n ¼ 288) with a mean age of 69.5 years were recruited for the study. ADT receiving or hormonal group (n ¼ 198) had a significantly higher prognostic risk compared to the non-hormonal group (n ¼ 89) (T stage > 2: 51% vs. 35%; Gleason score > 7: 22% vs. 8%, PSA > 20: 39% vs. 25%, respectively), yet no difference in the mean of depression scores between the two groups at baseline was observed. Patients were examined for depression at 2 weeks after ADT initiation, just before XRT initiation, 2 weeks after XRT completion and while on adjuvant ADT, and at 6 months after ADT and XRT completion. Just after the end of the neoadjuvant ADT and before starting XRT, the hormonal group had a significantly worsened depression score compared to the non-hormonal group (HADS score: 4 vs. 2.7; p < 0.001), however it was not clinically significant. Overall, although poor prognosis appears to significantly increase the depression risk and prolong depressive effect, depression risk induced by ADT appeared to be fortunately limited to the duration of treatment.
Saini and colleagues have demonstrated the effect of ADT on depression and anxiety among the PCa patients, who had no metastasis or history of neuropsychiatric disorders, undergone locoregional therapy (RP and/or XRT) and had a follow-up on primary therapy of � 6 months [89]. In their 3 months-cross sectional study, patients were divided into either hormonal group (n ¼ 49) who received adjuvant ADT after locoregional therapy and a follow-up group (n ¼ 54), in which patients were observed only after locoregional therapy. ADT composed of LHRH agonist every 3 months with antiandrogens (flutamide or bicalutamide) in the first month only to prevent flare. It should be noted that patients in the hormonal group were significantly older (median: 73 years vs. 70 years), at higher rate of exposure to XRT, and had higher prognostic risk (PSA >20; GS 7-10; stage T3). Although no difference was observed in the anxiety measured by the HADS, there was a significant increase in the depression score of patients undergoing hormonal therapy compared to non-hormonal counterparts (7.3 vs. 4.9, p < 0.002). Additionally, the proportion of patients diagnosed with moderate (M) and severe (S) depression among ADT users was noticed to be higher compared to their counterparts (M: 26.5% vs. 11%; S: 20.4% vs. 7.4%; p < 0.004, respectively). Furthermore, ADT was associated with a 5.5-fold increase in the risk of depression after adjustment of age, Gleason score, tumor stage, social and demographic variables (RR, 5.56; 95% CI, 2.1-14.8).
In 2014, Sharpley CF et al. evaluated depression and anxiety in a cross-sectional study among 526 PCa patients who either had never received ADT (n ¼ 241), had received ADT in the past (n ¼ 168) or were currently on ADT (n ¼ 117) [90]. ADT mainly consisted of LHRH agonist and cyproterone acetate tablets that were administered six months prior to XRT. The mean age of patients was 67.9 years ranging between 26 to 87 years. All patients had confined PCa with/out local lymph node invasion, received XRT with/out RP and were evaluated for depression and anxiety using the Self-Rating Depression Scale (SDS) and Self-Rating Anxiety Scale (SAS) within the last 2 weeks of ADT course, respectively. Although previous ADT users had no significant difference in the mean SDS score compared to never users, current ADT users had a significant higher score compared to the past (37.34 vs. 34.49; p < 0.05) and never users (37.34 vs. 33.43; p < 0.001). Similar observations were reported regarding the SAS score with no significant difference between the previous and never users only, while the current ADT users had a significantly worse score compared to the past (34.14 vs. 30.7; p < 0.001) and never ADT users (34.14 vs. 32.11; p < 0.05). In addition, a great proportion of the current ADT users had clinical anxiety compared to the never users (31.5% vs. 18.3%; p < 0.01). However, that was not observed when they were compared to previous ADT users (31.5% vs. 23.9%) nor between the previous and never ADT users (23.9% vs. 18.3%). More importantly, a significantly great proportion of the current ADT users were clinically depressed compared to other groups (p < 0.01); however, this was absent when the previous ADT were compared to never users (current ADT ¼ 39.1%, past ADT ¼ 25.6%, never ADT ¼ 23.3%). Another cross-sectional, questionnaire-based study by Sharpley's group was published in the same year where they investigated depression and anxiety among the PCa patients using quite similar grouping used previously [91]. One hundred and fifty-seven patients with a mean age of 68 years (range: 49-84 years), confined tumors and different treatment modalities including surgery or XRT with hormonal therapy, or none were included. Self-report Patient Health Questionnaire-9 (PHQ9) was utilized to examine major depressive disorder (MDD), whereas generalized anxiety disorder-10 (GAD10) was used to evaluate anxiety. Patients who were on ADT with/out other therapies were classified under the hormonal group (n ¼ 93), while all others even those with past ADT history were classified under the non-hormonal group (n ¼ 63). Their findings showed that current hormonal therapy increased the risk of depression and anxiety in PCa patients compared to the non-hormonal therapy (mean score of PHQ9: 14.53 vs. 12.48; mean score of GAD: 16.8 vs. 14.64, p < 0.05), confirming their previous conclusion that ADT users among the PCa patients are clinically more depressed and anxious compared to the non-ADT users [90]. Supporting this, Ahmadi-Davis and colleagues reported in the same year a case of 63-year-old man with no depression history, who was diagnosed with Gleason score 8, confined PCa, and treated with XRT and a 2-year course goserelin, administered every 3 months [92]. Six months after the treatment initiation, the patient was diagnosed with depression, which was controlled pharmacologically throughout the treatment course.
Lee and colleagues conducted a longitudinal prospective study in 2015 to investigate the effect of ADT (goserelin or leuprolide) on depression symptomology and rates among non-metastatic PCa patients [93]. Patients were categorized into either ADT group (who were scheduled for ADT for �6 months and had not been treated with LHRH agonist in the past 12 months or antiandrogen in the past 6 months; n ¼ 61) or non-ADT group (who had undergone RP only and not received any testosterone supplementation; n ¼ 61). Non-cancer individuals were also included in the study as matched controls (not received testosterone supplementation; n ¼ 61). All groups were evaluated for depressive symptoms and proportion of participants with clinical depressive symptomology at the initiation of ADT and 6 months later using the Center for Epidemiological Studies Depression Scale (CES-D). It should be noted that ADT group had a significantly higher rate of Gleason score � 7 compared to the non-ADT group. The results at initiation of study indicated that ADT users had significantly higher mean depressive scores compared to the non-ADT users (12.2 vs. 6.19, p ¼ 0.02); however, that was absent upon comparing to the non-cancer participants (12.2 vs. 7.2, p ¼ NS). This goes hand in hand with the proportion of those with clinically depressive symptomology (28%, 5% and 12%, respectively). On the other hand, six months later, the ADT group showed significantly higher mean depressive score compared to the non-ADT and non-cancer groups (14.77 vs. 6.91, p < 0.01; 14.77 vs. 6.16, p < 0.001, respectively). More importantly, although no difference in the rates of antidepressants use between groups at baseline, the rate of antidepressants use in the ADT group after 6 months was significantly higher than the non-ADT and non-cancer groups (15% vs. 2% vs. 7%; p ¼ 0.001), which resonates with the rates of clinically depressive symptomology when ADT groups was compared to the non-ADT (p < 0.01) or to the non-cancer group (p ¼ 0.01) at the same time point of evaluation (39% vs. 9% vs. 11% in the ADT, non ADT, non-cancer participants, respectively).
Later that year, Kao and colleagues published a case of a 60-year-old male with history of schizophrenia who was diagnosed with T2B PCa and determined to undergo ADT as primary therapy with adjuvant XRT [94]. Two days after starting leuproline acetate injection and bicalutamide 50 mg daily, the patient experienced mild depressed mood which worsened after approximately 2 weeks of ADT. The patient was treated with an antidepressant which ameliorated his depressive symptoms within a week and cured his mood disturbance by the end of the third week.
In 2016, Dinh and colleagues conducted a retrospective study on a large set of SEER data for patients diagnosed with stage I-III PCa between 1992 and 2006, age �66 years, never had orchiectomy, with no history of depression in the past 12 months prior to PCa diagnosis and started on ADT within 6 months of their diagnosis, in order to investigate the risk of depression and antipsychotics use with ADT [95]. Patients were stratified based on hormonal therapy to either ADT as primary therapy (with/out RP or XRT; n ¼ 33,882) or non-ADT group (with/out RP, XRT or none; n ¼ 44,670). ADT group was further classified based on ADT duration to 1-6 months, 7-11 months and � 12 months. Noteworthy, patients in ADT group were significantly older, had higher-grade tumors, more comorbidities, and shorter life expectancy. At the end of the study, they observed a significantly higher incidence of depression in the ADT vs. non-ADT group (7.1% vs. 5.2%, p < 0.001), as well as a 23% increase in the risk of depression with ADT when analyses were adjusted for demographic and clinical factors (AHR, 1.23; 95% CI, 1.15-1.31). Similarly, ADT group experienced higher rate and risk of inpatient (2.8% vs. 1.9%, p < 0.001; AHR, 1.29; 95% CI, 1.17-1.41) and outpatient use of antipsychotics (3.4% vs. 2.5%, p < 0.001; AHR, 1.07; 95% CI, 0.97-1.17, p ¼ 0.17) compared to the non-ADT group. With respect to the length of treatment, longer ADT increased the risk of depression (12% vs. 26% vs. 37%; p < 0.001, �6 months vs. 7-11 months vs. �12 months, respectively) as well as the overall cumulative use of antipsychotics (6.1% vs. 7.6% vs. 8% vs. 5.2%; p < 0.001, �6 months vs. 7-11 months vs. �12 months vs. non-ADT, respectively) when the data of ADT users were adjusted for demographic and clinical variables. Furthermore, compared to the non-ADT group, ADTtreated patients had a significant increase in the proportion of inpatients receiving antipsychotics (2.4%, 3%, 3.3% vs. 1.9%; p < 0.001), and outpatients receiving antipsychotics (2.8%, 3.5%, 4.1% vs. 2.5%; p < 0.001, �6 months vs. 7-11 months vs. �12 months vs. non-ADT, respectively). When the ADT group data were adjusted for demographic and clinical variables and compared to the non-ADT data, the risk of inpatient use of antipsychotic increased by 16%, 28% and 47% with �6 months, 7-11 months and �12 months use of ADT (p < 0.001), respectively. The same was observed with respect to the risk of outpatient use of antipsychotics (3% and 20% with 7-11 months and �12 months use of ADT, respectively, p ¼ 0.04).
Another retrospective, longitudinal, Taiwanese health insurance-based study was conducted by Chung and colleagues in 2017 to investigate the correlation between ADT and depression over a follow-up period of 3 years on patients who were newly diagnosed with malignant PCa between 2001 and 2010 [96]. All patients with aged �40 years were included and divided, based on the presence of ADT receiving claim that was dated around the 1st time visit to oncology clinics, into either ADT (n ¼ 868) or non-ADT (n ¼ 846) group. Patients in the ADT group had received LHRH agonists, antiandrogens, and/or estrogen during the study period, but not orchiectomy that resulted in an exclusion from this group. Furthermore, patients diagnosed with any depressive disorder 3 years prior to the date of ADT claim (in the ADT group) or the date of PCa diagnosis (in the control group) were excluded from the study. It should be noted that patients in the ADT group were significantly older (74 vs. 70 years), urban and had lower income compared to the control group. Overall, the depressive incidence was shown to be higher in the ADT compared to the control group (3.9% vs. 1.7%, 14 vs. 7 per 1000 person-year, respectively). Further analysis indicated that ADT had significantly increased the risk of depression in this group relative to the control group (HR, 2.12; CI95%, 1.14-3.95), even after adjusting the analysis for living location, income and comorbidities over the 3-year follow-up (AHR, 1.93; CI95%, 1.03-3.62).
In 2017, Zhang and colleagues investigated prospectively over a 15-month period the risk of depression with ADT among elderly men (age >60 years) diagnosed with localized PCa using the Zung Self-Rating Depression Scale (SDS) that was administered by physicians at baseline, 1, 6, 9, 12 and 15 months [97]. Participants were categorized into three groups: ADT group (n ¼ 50, mean age of 71.6 years) that were treated primarily with ADT only (combination of LHRH agonists þ antiandrogens) without any adjunctive therapy (RP or XRT), RP control group (n ¼ 46, mean age of 69.4 years) who had undergone RP only, and a benign prostatic hypertrophy (BPH) as another control group (n ¼ 50, mean age of 70.2 years). A score of �40 points indicated depression symptoms (lowest is 20 while the highest is 80). All participants had no depressive symptoms nor were diagnosed with depression at baseline. ADT and RP group had a significantly higher PSA >20 ng/ml and T3 stage (p < 0.001) compared to the BPH group. As patients were evaluated at different time points, the ADT group had higher depressive scores and developed significantly more depressive symptoms at 9, 12 and 15 months compared to the counterpart groups (p < 0.001); the same observation was noted as well for the RP compared to the BPH group only (p < 0.001). More importantly, the incidence rates of depression at 12 and 15 months relative to baselines among the ADT recipients were significantly higher than RP and BPH groups, whereas the rates in RP group were significantly more compared to BPH group (12 months: 24% vs. 4.3% vs. 2%, p < 0.000; 15 months: 20% vs. 6.5% vs. 2%, p < 0.007, respectively).
Another interesting work was performed by Gagliano-Juca et al. in 2018 at Harvard medical school, in which part of the work was focused on evaluating the ADT effect on pain perception, anxiety, and depression among localized PCa patients, who were followed prospectively for 6 months [98]. Thirty-seven patients whom decided to administer leuprolide every 3 months (for at least 6 months) with bicalutamide in the first month only were included in the ADT group, while forty matched patients who have undergone RP ± XRT for � 6 months prior to the study were used as controls (non-ADT group). The two groups had a median age of 67 years (53-89 years) with no difference in the rate of antidepressants use secondary to minimal or mild depression and no history of moderate or severe depression measured by PHQ9 at baseline. Patients were followed for 24 weeks and evaluated at baseline, 6, 12 and 24 weeks. Intriguingly, although ADT had no effect on the score and perception of the clinical pain (based on Brief Pain Inventory Questionnaire (BPI)), nor on the anxiety (assessed by the State-Trait Anxiety Inventory (STAI)), it significantly worsened the depression in the ADT compared to the non-ADT group (effect size ¼ 0.93; CI95%, 0.04-1.82; p < 0.04).
In the same year, another observational, multicenter, prospective study (COMPARE) was also published by Thomas and colleagues who investigated the risk of depression with adjuvant ADT on PCa patients (n ¼ 656) whom tumors recurred biochemically after XRT or RP as primary therapy [99]. Most of the patients (77%) were aged between 60 and 80 years, and 80% had T1 and T2 tumors. Patient were categorized based on their definitive therapy into 3 groups: XRT alone (n ¼ 124), RP with/out XRT (n ¼ 269), and adjuvant ADT after RP or XRT (n ¼ 263). The prevalence of depression was found to be 3.2%, 5.9% and 9.1%, respectively. In other words, ADT was significantly associated with higher rate of depression compared to XRT alone (p < 0.03). However, this was absent when it was compared to RP with/out XRT or when the latter was compared to XRT alone. When analyses were adjusted for age and other comorbidities, ADT was associated with higher odds of depression compared to XRT alone (OR, 3.29; CI95%, 1.11-9.76, p < 0.03). However, that was not observed neither in ADT vs. RP with/out XRT nor the latter compared to XRT alone.
In 2019, a prospective study that focused on the association of ADT with depression and depressionlinked cognitive impairment in PCa patients diagnosed with locally advanced and metastatic disease was published by Ceylan and colleagues [100]. Two groups were compared in the study including those with advanced and metastatic disease who either received only ADT for 12 months continuously (n ¼ 72, mean age 67.3 years) or subjected to RP only (n ¼ 72, mean age 66.7 years) with no other adjunctive therapies. None of the patients had psychiatric disease nor a history of antipsychotics use. However, ADT patients had higher PSA values and elevated Gleason scores (p < 0.0009 and p < 0.001, respectively) compared to their counterparts. Hamilton depression rating scale (HAM-D) was utilized to measure depression score at 6 months and 12 months, which were then compared to baseline assessment. Maximum score of the scale is 53 indicating a major depression state while a score below 8 indicated a normal state. Relative to the baseline of each group, the mean HAM-D scores measured at 6 and 12 months were significantly higher in ADT group (6 months: 6.19 vs. 5.11, p ¼ 0.003; 12 months: 6.7 vs. 5.11, p ¼ 0.001, respectively), however that was absent in the control group. When the mean of 12month scores was compared to the 6-month, the ADT but not the control group showed a significantly higher values (p ¼ 0.023 and p ¼ 0.058, respectively). More importantly, looking at the depression incidences at 6 and 12 months, 11 and 14 patients among the ADT patients while 4 and 1 patients in the control group were diagnosed with depression, respectively.
Deka and colleagues investigated retrospectively the risk of depression developed post ADT administration, through focusing on inpatient and outpatient psychiatric services utilization and suicide, among veterans treated with XRT as a definitive therapy for localized PCa [101]. None of the patients received chemotherapy, diagnosed with any psychiatric condition including depression or used psychiatric services within a year before PCa diagnosis. Patients who had undergone RP or did not receive XRT as definitive therapy were excluded. Patients were categorized into two groups: a group treated with XRT alone (n ¼ 25,122) and another group (n ¼ 14,843) treated with XRT and ADT (within the 1st year of diagnosis). ADT users were further divided into two different groups according to the duration of ADT to: �1 year or >1 year. The ADT drugs that patients were exposed to included leuprolide, goserelin, triptorelin, histrelin, degarelix, flutamide, bicalutamide, or nilutamide. Patients in the ADT group were significantly older (median age 67.97 vs. 66.24 years), had lower income, and higher rates of PSA levels/Gleason score/T stage. There was also a significant increase in the rate of antidepressants use, especially selective serotonin and norepinephrine reuptake inhibitors (SNRIs) (5.65% vs. 4.72%), atypical antidepressants (2.20% vs. 1.84%), and serotonin modulators (2.41% vs. 2.06%) among the ADT compared to XRT group. Groups were followed for 10 years (a median of 6.8 years), during which they were compared for depression, inpatient and outpatient clinic services utilization and suicide. By the end of the follow-up period, the unadjusted cumulative incidence rates of depression and outpatient clinic services usage were significantly higher among the ADT compared to the XRT patients (approximately 2.1% vs. 1.4% for depression rates; 22% vs. 17.6% for services usage, all p < 0.001, respectively). Furthermore, the estimated risk of depression and outpatient clinic services usage after applying the multivariable risk regression model was significantly increased by 50% and 21% with ADT (SHR, 1.50; 95% CI, 1.32-1.71 and SHR, 1.21; 95% CI, 1.16-1.27; all p < 0.01, respectively). However, neither inpatient clinic services usage nor suicide were correlated with the use ADT in PCa treated with XRT as definitive therapy.
Tully and colleagues have studied retrospectively in late 2019 the correlation between ADT and risk of depression and dementia among the PCa patients [102] using TRICARE insurance database 2007-2014. To avoid the impact of patients' variables, the study was limited to young-middle age patients (40-64 years) with non-metastatic lesions, never undergone orchiectomy, no pre-existing neurocognitive conditions, no history of antidepressants use, and no use of psychotherapy 12 months prior to or 6 months after PCa diagnosis. PCa patients were categorized under ADT group if they received ADT within 6 months of their diagnosis with PCa (n ¼ 325) whereas non-ADT group included patients who underwent RP ± XRT (n ¼ 8792). It is worth mentioning that 80% of the ADT group received XRT while 5.3% and 94.3% of the non-ADT group received XRT and RP, respectively. Patients were followed up for a median of approximately 10 years, which allowed authors to further investigate the long-term effect of ADT. Overall, ADT was associated with a significant increase in the incidence of depression (12% vs. 7.1%) and dementia (7.4% vs. 3.4%) compared to non-ADT regimen (all p < 0.001). Moreover, subgroup analysis comparing the study group with long-term ADT administration (�12 months) to the matched non-ADT group estimated that ADT carried an increased risk of depression and dementia (all p < 0.001). Furthermore, although the risk of new onset of dementia did not differ, ADT had significantly increased the overall risk of new-onset depression by two folds after adjusting the analysis (AHR, 2.07; 95% CI, 1.32-1.71, p < 0.001) compared to the non-ADT group.
Four additional studies came out in 2020 and 2021 supporting this association. The first study was published by Shin and colleagues who investigated the risk of depression with ADT prospectively [103]. Patients with newly diagnosed PCa, without any other types of cancer or history of psychiatric illness, and had not received any procedures or treatment affecting male hormones were included. Patients were determined to receive either ADT (n ¼ 49) or RP (n ¼ 58). BDI was completed 1 month before the study (T1) and 6 months after receiving the determined intervention (T2). At T1, there was no significant difference in the mean BDI scores between the ADT and RP group (10.27 vs. 10.74; p > 0.784, respectively). However, the score of ADT group has significantly increased after 6 months compared to the control (19.4 vs. 8.98; p < 0.001). This was accompanied with a significantly greater reduction of the testosterone levels resulted from ADT compared to RP when the 6month levels were compared to baseline (T1: 162.10 ng/dL vs. 349.03 ng/dL, p ¼ 0.061; T2: 30.83 ng/ dL vs. 380.71 ng/dL, p < 0.001).
Another prospective study published by Holtfrerich and colleagues investigated the long-term impact of ADT (use up to 20 years) on cognitive function and socioeconomic decision in PCa patient [104]. Participants were divided into three groups including PCa patients with ADT (n ¼ 24), PCa patients without ADT (n ¼ 22) and healthy controls (n ¼ 22). Groups were matched for age (mean age of 68.7, 67.1 and 65.7 years, respectively), education level and time from PCa diagnosis (median of 33 months). Although no significant difference in BDI scores and quality of life was found between non-treated PCa patients and healthy controls, ADT-treated patients had significantly higher BDI scores compared to former and latter (p ¼ 0.002 and p ¼ 0.003, respectively). Also, poorer quality of life was found among ADT-treated patient compared to those who have not received ADT only (p ¼ 0.002).
The third was a questionnaire-based study conducted prospectively by Chen and colleagues on PCa patients who did not undergo RP, orchiectomy or ADT previously and had no history of neuropsychiatric disorders [105]. Patients who had received ADT (leuprolide acetate, goserelin acetate or degarelix acetate) regularly every 2-or 3-month periods were asked to complete the Patient Health Questionnaire-9 (PHQ-9) to diagnose depression. Based on PHQ-9 scores, patients were divided to depression (PHQ-9 � 6; n ¼ 29) and non-depression group (PHQ-9 < 6; n ¼ 42). Although there was no difference in ADT duration between the two groups, a significant positive correlation was established between the duration of ADT and PHQ-9 score in the patients of depression group. In other words, longer duration of ADT predisposes more severe depression symptoms.
Recently, Hoogland and colleagues investigated whether systemic inflammation after ADT (goserelin or leuprolide for � 6 months) is a potential reason of developing fatigue, MCI and depression in PCa patients [106]. PCa patients (n ¼ 47; mean age: 67.6 years) and healthy controls (n ¼ 82; mean age: 68.4) were evaluated for proinflammatory biomarkers, fatigue, depression and MCI before or within 21 d of the intervention, 6 and 12 months later. PCa patients had non-metastatic or asymptomatic metastatic disease and a few of them were treated with RT/brachytherapy besides ADT. Although fatigue was attributed to elevated IL-6 levels over the study period (p < 0.05), this was absent when depression symptoms and MCI were assessed for this correlation. Nevertheless, when depression scores were assessed between the two groups by the Center for Epidemiological Studies Depression Scale, they were found to be at significantly higher levels at baseline and over the 12 months of follow-up in ADT compared to healthy group (9.9 vs. 6.7, p < 0.003; 13.3 vs. 7.7, p < 0.48, respectively).
Overall and based on the incidence of depression provided by 10 studies only [85,88,89,92,94-96, 98,100,101], the average incidence of depression among PCa patients treated with ADT is 18.23% (range: 2.1-46.9%), while it is 8.42% (range: 1.4-23.3%) in the non-ADT treated patients. Cumulatively, these studies support the notion of ADT being a risk for the rise of depression in PCa patients, which could negatively influence their adherence to medications and worsen their conditions.

Antidepressants use for depression induced by androgen deprivation therapy
Because of the negative health consequences attributed to ADT-induced depression in PCa patients, finding the appropriate pharmacological treatment is necessary to improve patient's clinical outcomes. Antidepressants including selective serotonin reuptake inhibitors (SSRIs), SNRIs, tricyclic antidepressants (TCAs), monoamine oxidase inhibitors (MAOIs), selective dopamine and noradrenaline reuptake inhibitors (NDRIs), and atypical antidepressants are commonly used to treat depression, yet their efficacy in ADTtreated PCa patients is highly questionable. In other words, low levels of androgen dependent neurotransmitters, particularly serotonin [71-73], secondary to reduced AR activity after ADT, in the brain's area that control pathogenesis of depression, increases the belief that these treatment modalities may not be appropriate in this group of patients. Despite that, a literature search revealed only five case reports published so far that discuss the use of antidepressants for ADT-induced depression in PCa patients (summarized in Table 3).
The first was published in 1995 about three elderly men diagnosed with locally advanced or metastatic PCa who suffered from depression after ADT initiation [84]. The first patient was treated with bupropion ± lorazepam and was then switched to lorazepam monotherapy then doxepin. The second was prescribed desipramine only, while the third, although had a history of depression controlled by nortriptyline, was started on lorazepam. Despite the dose escalation of antidepressants in all cases, they had minimal effects nd eventually lost their efficacy which led to antidepressants as well as ADT discontinuation.
The second report was published in 2004 on a PCa patient who developed depression post-ADT and was initiated on different antidepressants [85]. Since they were not effective, testosterone was administered which significantly resolved his depression symptoms.
The third study has also shown worsening of depressive symptoms in an ADT-treated PCa patient despite using sertraline 25 mg daily. Later, the patient was referred to a psychiatric clinic for further evaluation and treatment [87].
The fourth study was on a PCa patient diagnosed with ADT-induced depression and initiated on paroxetine 40 mg daily that partially cured his depressive symptoms [92]. The patient was then switched to venlafaxine and titrated up to 150 mg daily which nearly resolved his depression.
The final study was published in 2015 on a schizophrenic patient diagnosed with PCa and treated with ADT. Later, the patient developed depression which was improved after 10 mg daily of escitalopram [94].
Despite the obvious scarcity of research investigating antidepressants efficacy in ADT-induced depression among PCa patients, there are few studies published on their usefulness in hot flashes (HFs) associated to ADT.
In 1998, Loprinzi and colleagues conducted a prospective questionnaire-based study on breast cancer female survivors (n ¼ 23) and ADT-treated PCa patients (n ¼ 5) [age �50 years in 90%] who were complaining of HFs and treated with 12.5 mg venlafaxine twice daily for 4 weeks [107]. By the end of 5th week of study, the questionnaire responses showed that incidence and severity of HFs were significantly reduced by �50% compared to the first week's symptoms.
Similarly, Quella and colleagues investigated in 1999 the effect of venlafaxine 12.5 mg twice daily for 4 weeks in ADT-treated PCa patients (n ¼ 16, mean age 70 years) [108]. Compared to the 1st week with no administered treatment, the 4-week treatment course reduced HFs frequency and severity by �40 % with well toleration of its side effects.
A prospective study assessing the impact of HFs on ADT-treated PCa patients' quality of life (n ¼ 55; mean age 72 years) was published by Nishiyama et al. in 2004 [109]. Thirty-two patients suffered from HFs, 12 of which were treated with fluvoxamine 25 mg daily for 2 weeks after which a significant improvement in their symptoms was reported without developing any side effect.
In 2006, Naoe and colleagues conducted a questionnaire-based study to assess the impact of 4-week treatment with paroxetine 10 mg daily on HFs symptoms in PCa patients (age of 71 years; n ¼ 10; with localized, advanced or metastatic disease) treated with ADT [110]. Patients were assessed before the treatment, and at first, second, third and fourth week of the treatment course. After the 4th treatment week, HFs incidences and severity were significantly reduced by �43 % which resulted in improving their QOL significantly. Seven years later, a randomized, double-blind, placebo-controlled, multicenter trial was conducted by Vitolins and colleagues to assess the efficacy of venlafaxine 75 mg daily with/out Soy powder (20 gm þ 160 mg isoflavones) on HFs in ADT-treated PCa patients (n ¼ 120; median age 69 years) [111]. Patients were assessed at baseline and 12 weeks using a questionnaire. Although frequencies and severity of HFs decreased significantly by the end of study in all groups including placebo, it was found that neither venlafaxine, Soy powder, or combo had significantly differed compared to placebo.
Overall, there is compelling evidence confirming the association between depression and ADT in PCa patients that frequently requires pharmacological treatment. However, the current literature lacks observational as well as controlled studies that elucidate the efficacy and safety of antidepressants in this category of patients. Nevertheless, using the results of studies conducted on their efficacy and safety in ADTinduced HFs could be adopted to guide clinicians throughout the process of investigating their usefulness in treatment of depression in ADT-treated PCa patients.

Other potential strategies useful to treat depression induced by androgen deprivation therapy in prostate cancer patients
Additional strategies to treat depression induced by ADT in PCa patients have been investigated recently. One new strategy was to administer testosterone to these patients. Although testosterone use has been contraindicated in PCa patients for decades, a recent study has shown that offering it for patients with advanced PCa resulted in a favorable outcomes as it improved their cognition and well-being and didn't seem to cause adverse effects at rapid rates [112]. This is supported by the Testosterone Trial, one of the largest randomized control trials of testosterone therapy in men, that demonstrated improvement in the mood with testosterone therapy in hypogonadal men [71]. Another potential strategy that could be adopted is to stop ADT temporarily. However further studies are required to test its effectiveness.

Conclusion
Despite the controversial evidence between ADT and depression among PCa patients, the majority of studies indicate that ADT use in PCa is associated with increased rates of depression. Although some studies showed no average mean increase in depression scores with ADT, this doesn't reflect the smaller subset of individuals who experienced depression but had no significant effect on the score of the group as a whole. Even though, there is no literature addressing the effectiveness of modern pharmacotherapy for depression in this population, which may or may not be as effective in men not on ADT due to known changes in neurotransmitters in the androgen deprived state. Nevertheless, the results associated with the use of SSRIs, SNRIs and testosterone for adverse effects attributed to ADT would pave the way to address the current unmet need in this area.