The relationship between androgen levels and depression, particularly in men, is an area of contradictory reports and intense debate. Although low testosterone levels in men have been associated with increasing age and comorbid conditions of ageing, the evidence to support an association between low testosterone and depressive symptoms is conflicting . Similarly, placebo-controlled trials examining the impact of testosterone replacement therapy on depressive symptoms have yielded mixed results in regards to efficacy , as well as concerns for adverse events . However, a recent large randomized placebo-controlled trial in symptomatic older men with low testosterone levels suggested that testosterone therapy may improve mood and lower the severity of depressive symptoms without increasing the risk of adverse events compared with placebo .
As such, the literature continues to evolve in a number of related arenas, including the impact of androgen levels on depression, the impact of testosterone replacement therapy on depression and the impact of androgen deprivation therapy (ADT) on depression. Here, we summarize and synthesize the recently published literature in these important domains of investigation.
ANDROGEN MEASUREMENTS AND DEPRESSION
There remains controversy regarding whether measured androgen levels predict depression. Three recent studies sought to further address this question. First, Asselmann et al. reported on a prospective longitudinal general population study in Northeastern Germany involving 993 men and 980 women. At baseline, serum concentrations of total testosterone, androstenedione and sex-hormone binding globulin (SHBG) were assessed. Androgens were measured using liquid chromatography-tandem mass spectrometry. The authors subsequently examined 12-month, lifetime and incident depressive disorders, which included major depressive disorder and dysthymia, as defined by DSM-IV criteria. Trained clinical interviewers conducted a face-to-face retrospective assessment of depressive disorders. Androgen levels were examined as continuous variables for association with the binary outcome of a depressive disorder. Incident depressive disorders were present in 6% of men and 13% of women. The authors found that total testosterone, androstenedione and SHBG, overall and when stratified by gender, did not predict depressive disorder. For example, the 12-month odds of any depressive disorder per unit change in total testosterone in men was 7% [odds ratio (OR), 1.07; 95% confidence interval (95% CI), 0.73–1.59]. A primary strength of this study was the utilization of a prospective general population cohort to assess changes in androgen levels. The evaluation for depressive disorders was conducted in a retrospective fashion, which may have impacted the accurate reporting of depressive disorders.
Second, Kische et al. utilized data from a large nationally representative study in Germany that included 6493 primary care patients including 2653 men and 3840 women. Testosterone levels were measured at baseline and at 1-year follow-up using an immunoassay-based approach. Depressive symptoms were assessed using the Depression Screening Questionnaire (DSQ) and examined as a binary and continuous variable. The authors found that baseline testosterone did not have an association with prevalent or incident depression, including when stratifying by gender. Interestingly, men with a higher testosterone level at 1 year, compared with baseline, had a lower burden of depressive symptoms (β-coefficient per unit change in testosterone: −0.17; 95% CI: −0.31 to −0.04) and a lower risk of incident depressive symptoms (OR per unit change in testosterone: 0.84; 95% CI, 0.72–0.98) at 4-year follow-up. No associations with change in testosterone levels and depressive symptoms were observed in women.
Third, Korenman et al. undertook a retrospective medical record review in an urban academic health system to determine whether low testosterone levels in men not secondary to primary testicular failure or inciting comorbid condition (e.g. endocrine disorders) contributed to an increased rate of depression. They utilized ICD-9 codes to determine whether patients carried a diagnosis of hypogonadism or depression. Individuals classified as having hypogonadism using billing codes were also required to have evidence in the electronic medical record of low testosterone levels in the absence of follicle stimulating hormone or luteinizing hormone elevation. They included only men aged 18–40 years in their analysis in order to decrease the likelihood that individuals included had comorbidities known to cause low testosterone levels. They queried data in 105 816 individuals and ultimately analysed 186 individuals with low testosterone and 101 437 controls. Cases had a depression rate of 22.6% compared with 13.4% in controls and an antidepressant medication utilization rate of 22.6% compared with 6.6% in controls. After adjustments for clinical and demographic factors, there was a 13% increased odds of a depression diagnosis or antidepressant medication use in patients with low testosterone compared with controls (OR, 1.13; 95% CI, 1.09–1.17). Overall, men with low testosterone aged 18–40 years without evidence of primary testicular failure or coexisting contributory comorbidity was exceedingly rare (0.2%), but may represent a patient population at high risk for depression.
Taken together, neither of the studies assessing androgen levels support an association between single time-point androgen levels and depressive symptoms. The finding by Kische et al. that changes in testosterone levels in men may impact depressive symptoms is notable, particularly when considering the potential impact of testosterone modifying medications. Importantly though, the study by Kische et al. does not establish a temporal relationship, and therefore, the direction of effect in their study is unknown.
TESTOSTERONE REPLACEMENT THERAPY AND DEPRESSION
As the potential role of androgen levels and depression continues to be evaluated, a large number of randomized and nonrandomized studies have previously been conducted to evaluate the impact of testosterone replacement therapy on depressive symptoms. Two meta-analyses were recently published that synthesized the available literature on testosterone replacement therapy and depression.
First, Elliott et al.[8▪] conducted a systematic review and meta-analysis of randomized controlled trials and nonrandomized studies that evaluated the impact of at least 3 months of testosterone replacement therapy in men with low testosterone on subsequent evaluations of depression and depressive symptoms. They identified 12 randomized controlled trials including 852 participants that evaluated nine distinct testosterone replacement therapies. Medications were given for between a 12-week and 3-year duration and studies were reported from 1998 to 2015. All 12 studies used a validated depression scale or inventory to evaluate depression. Overall, they found that, compared with placebo, treatment with a testosterone replacement therapy improved depression and or depressive symptoms (standardized mean difference, –0.23; 95% CI, –0.44 to –0.01). They did not find any statistically significant differences in the effect on depression for any individual type of testosterone replacement therapy compared with placebo.
Second, Walther et al. conducted a systematic review and meta-analysis of randomized controlled trials in men with low and normal testosterone levels who reported depressive symptoms on psychometrically validated depression scales to determine whether testosterone replacement therapy alleviates depressive symptoms. They included data from 27 randomized controlled trials reported from 2000 to 2016 in 1890 men. They reported a significant reduction in depressive symptoms across all studies (effect estimate, 0.21; 95% CI, 0.10–0.32). The authors also report that they did not observe that the effect statistically significantly differed in men with normal versus low testosterone and therefore that testosterone replacement therapy may also be an effective treatment for depressive symptoms in men with normal testosterone levels. However, this lack of statistical difference between groups may be secondary to inadequate power to detect a difference rather than a true lack of difference. The effect in men with normal testosterone levels was not statistically significant (effect estimate, 0.21; 95% CI, -0.23 to 0.64) and further studies are needed to determine whether testosterone replacement therapy is an effective treatment for depression in men with normal testosterone levels.
ANDROGEN DEPRIVATION THERAPY AND DEPRESSION
In contrast to examining the potential benefits of testosterone replacement therapy on depression, the literature regarding the detrimental impact of ADT in the treatment of prostate cancer on depression continues to grow. With over a million new diagnoses of prostate cancer each year, and approximately 50% of men with prostate cancer ultimately utilizing ADT, the impact of ADT on adverse outcomes, including depression, represents an important area of inquiry for public health [9▪]. A number of recent studies provide further evidence for an association between the use of ADT and depression.
First, Gagliano-Jucá et al. identified 37 men with prostate cancer and followed them prospectively for a 6-month period to assess changes in clinical pain, depression, anxiety levels and quality of life before and after initiating ADT. These men received a gonadotropin-releasing hormone (GnRH) agonist (22.5 mg of leuprolide acetate) every 3 months with a planned intervention of at least 6 months in addition to the androgen receptor antagonist bicalutamide during the first month of therapy. Forty men who had previously undergone a prostatectomy for localized prostate cancer and were in remission were analysed as a non-ADT control group. Depression was evaluated using the Patient Health Questionnaire (PHQ-9), which is a nine-item scale that includes an evaluation of depressive symptoms (’Feeling down, depressed, or hopeless’) experienced in the 2 weeks prior to evaluation. Men with moderate-to-severe depression at baseline per the PHQ-9 were excluded. The authors found that serum testosterone levels, as measured by liquid chromatography-tandem mass spectrometry, significantly decreased in the ADT group (mean = 13 ± 8 ng/dl), but remained unchanged in the non-ADT group (mean = 468 ± 185 ng/day) during the study period. When examining the average effect over the 24 months of intervention using data from baseline, 6 , 12 and 24 weeks, depression scores significantly worsened in the ADT group compared with the non-ADT group (effect size, 0.93; 95% CI, 0.04–1.82; P = 0.042). They did not find an association between ADT use and anxiety or quality of life, but the study may have been underpowered to fully assess these outcomes.
Second, Downing et al. conducted a population-based study among men in the UK who were still alive 18–42 months after prostate cancer diagnosis. Men with any stage of disease at diagnosis were included. Individuals were identified using cancer registry information and data on participating individuals were collected using a postal-based survey. Participants completed the five-level EuroQol five dimensions questionnaire (EQ-5D-5L) to assess mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. Specifically, they were asked if they were ‘Feeling depressed’ (answer options: no, very small, moderate, big problem) in the previous 4 weeks. Thirty-five thousand eight hundred and twenty-three (61%) of 58 930 queried men responded to the survey with disease stage known for 30 733 (86%), including 64% with stage I or II, 23% with stage III and 13% with stage IV disease. Among ADT users, 11% (95% CI, 11–12) reported depression as a ‘moderate’ or ‘big problem’ compared with 7% (95% CI, 6–7) among those not taking ADT. The authors also reported higher rates of depression with increasing disease stage. As rates of depression were not adjusted for stage, it is unclear from this study whether ADT causally impacts depression or whether higher rates of depression in the ADT group are secondary to confounding by disease stage.
Third, Thomas et al. examined 656 men in the prospective COMPARE (Comprehensive, Multicenter, Prostate Adenocarcinoma) registry, which includes 150 geographically diverse sites in the United States. Specifically, they examined men who experienced biochemical recurrence after radiation therapy, radical prostatectomy without or without radiation therapy, or ADT in combination with radical prostatectomy or radiation therapy. Each individual's perception of whether they were depressed was assessed by a physician-completed questionnaire at the time of biochemical recurrence. Specifically, physicians were asked, ‘Does the patient have any other complaints’ and they were provided with a list of symptoms, which included depression. Overall, 44 men (7%) reported depression, including 3% for radiation alone, 6% for radical prostatectomy and 9% for ADT. Men treated with ADT had higher odds of reporting depression after adjustment for age, baseline comorbidities, T stage and Gleason score compared with radiation therapy alone (OR, 3.48; 95% CI, 1.51–10.52; P = 0.027). No difference in reported depression rates was observed when comparing therapies that did not include ADT. A strength of this study is that patients were relatively homogenous: men at the time of biochemical recurrence. It is unclear how the compared groups differed at enrolment, and therefore, it is unknown whether baseline differences between the groups may have contributed to the observed effect, independent of ADT.
Fourth, Ng et al. conducted a retrospective analysis using claims-based data in the Australian universal healthcare system. Specifically, they used a 10% sample of the Pharmaceutical Benefits Scheme (PBS) dispensing claims data, which includes approximately 3 million Australians, over a 12-year period from 2003 through 2014. They compared men with prostate cancer who had received ADT to men without any evidence of having received antineoplastic or hormonal therapy. They found that men with prostate cancer who underwent ADT had a greater risk of depression than 1 : 10 matched controls. As the comparison group did not have evidence of any malignancy, this observed difference in the rate of depression may be an artefact of comparing depression rates in men with and without cancer, rather than secondary to the use of ADT.
Finally, Nead et al.[9▪] conducted a systematic review and meta-analysis to examine the association of ADT and depression. They included studies that reported depression among individuals with prostate cancer exposed to a course of ADT versus a lesser-exposed group (e.g. any-ADT versus no ADT; continuous ADT versus intermittent ADT). Included studies were reported from 2006 to 2016 and utilized a variety of metrics to delineate depression status including billing codes and depression inventories. In a meta-analysis of 18 studies among 168 756 individuals, they found that ADT use conferred a 41% increased risk of depression (risk ratio, 1.41; 95% CI, 1.18–1.70; P < 0.001) that was consistent when limiting their analysis only to prospective studies (risk ratio, 1.41; 95% CI, 1.18–1.70; P < 0.001) and those examining only localized disease (risk ratio, 1.85; 95% CI, 1.20–2.85; P = 0.005). Importantly, when limiting their analysis to a clinical diagnosis of depression, they also found a statistically significant result.
Recent studies have made significant contributions to our understanding of the relationship between androgen levels and depression. The impact of baseline and single time point androgen levels on depressive outcomes remains unclear. Conversely, the evidence base for a beneficial effect of testosterone replacement therapy on depressive symptoms among men with low testosterone levels continues to grow and solidify. Finally, there is convincing evidence that ADT in the treatment of prostate cancer contributes to depression and healthcare providers should be alert for and have a low threshold to intervene on depressive symptoms among patients undergoing ADT. Future research should focus on interventions and strategies to provide a clinically meaningful reduction in depressive symptoms among men with decreased androgen levels.
Financial support and sponsorship
Dr. Nead is supported through a Gerber Foundation Novice Research Grant and by the Marlene Shlomchik Fellowship in Cancer Research through the Abramson Cancer Center.
Conflicts of interest
There are no conflicts of interest.
REFERENCES AND RECOMMENDED READING
Papers of particular interest, published within the annual period of review, have been highlighted as:
1. Walther A, Breidenstein J, Miller R. Association of testosterone
treatment with alleviation of depressive symptoms in men: a systematic review and meta-analysis. JAMA Psychiatry 2019; 76:31–40.
2. Amanatkar HR, Chibnall JT, Seo BW, et al. Impact of exogenous testosterone
on mood: a systematic review and meta-analysis of randomized placebo-controlled trials. Ann Clin Psychiatry 2014; 26:19–32.
3. Xu L, Freeman G, Cowling BJ, Schooling CM. Testosterone
therapy and cardiovascular events among men: a systematic review and meta-analysis of placebo-controlled randomized trials. BMC Med 2013; 11:108.
4. Snyder PJ, Bhasin S, Cunningham GR, et al. Effects of testosterone
treatment in older men. N Engl J Med 2016; 374:611–624.
5. Asselmann E, Kische H, Haring R, et al. Prospective associations of androgens and sex hormone-binding globulin with 12-month, lifetime and incident anxiety and depressive disorders in men and women from the general population. J Affect Disord 2019; 245:905–911.
6. Kische H, Pieper L, Venz J, et al. Longitudinal change instead of baseline testosterone
predicts depressive symptoms. Psychoneuroendocrinology 2018; 89:7–12.
7. Korenman SG, Grotts JF, Bell DS, Elashoff DA. Depression
in nonclassical hypogonadism in young men. J Endocr Soc 2018; 2:1306–1313.
8▪. Elliott J, Kelly SE, Millar AC, et al. Testosterone
therapy in hypogonadal men: a systematic review and network meta-analysis. BMJ Open 2017; 7:e015284.
9▪. Nead KT, Sinha S, Yang DD, Nguyen PL. Association of androgen
deprivation therapy and depression
in the treatment of prostate cancer
: a systematic review and meta-analysis. Urol Oncol 2017; 35:664e1- e9.
10. Gagliano-Juca T, Travison TG, Nguyen PL, et al. Effects of androgen
deprivation therapy on pain perception, quality of life, and depression
in men with prostate cancer
. J Pain Symptom Manage 2018; 55:307–317. e1.
11. Downing A, Wright P, Hounsome L, et al. Quality of life in men living with advanced and localised prostate cancer
in the UK: a population-based study. Lancet Oncol 2019; 20:436–447.
12. Thomas HR, Chen MH, D’Amico AV, et al. Association between androgen
deprivation therapy and patient-reported depression
in men with recurrent prostate cancer
. Clin Genitourin Cancer 2018; 16:313–317.
13. Ng HS, Koczwara B, Roder D, Vitry A. Development of comorbidities in men with prostate cancer
treated with androgen
deprivation therapy: an Australian population-based cohort study. Prostate Cancer
Prostatic Dis 2018; 21:403–410.