Depressive Syndromes in Men With Hypogonadism in the TRAVERSE Trial: Response to Testosterone-Replacement Therapy

Bhasin, Shalender; Seidman, Stuart; Travison, Thomas G; Pencina, Karol M; Lincoff, A Michael; Nissen, Steven E; Miller, Michael G; Flevaris, Panagiotis; Li, Xue; Wannemuehler, Kathleen A; Pope, Harrison G

doi:10.1210/clinem/dgae026

Abstract

Context

The effect of testosterone on depressive symptoms in men with hypogonadism remains incompletely understood.

Objective

We assessed the effects of testosterone-replacement therapy (TRT) in improving depressive symptoms in hypogonadal men with and without depressive symptoms enrolled in the TRAVERSE cardiovascular safety trial.

Methods

A randomized, placebo-controlled, double-blind study was conducted at 316 trial sites. Participants included men, aged 45 to 80 years, with 2 fasting testosterone levels less than 300 ng/dL, 1 or more hypogonadal symptoms, cardiovascular disease (CVD), or increased risk of CVD. We evaluated 3 subgroups of participants: (1) men with rigorously defined, late-life-onset, low-grade persistent depressive disorder (LG-PDD, previously “dysthymia”); (2) all men with significant depressive symptoms (Patient Health Questionnaire-9 Score >4); and (3) all randomly assigned men. Intervention included 1.62% transdermal testosterone or placebo gel. Outcome measures included the proportions of participants (1) meeting criteria for LG-PDD or (2) with significant depressive symptoms; and changes in depressive symptoms, energy, sleep quality, and cognition in testosterone-treated vs placebo-treated men in the 3 subgroups.

Results

Of 5204 randomly assigned participants, 2643 (50.8%) had significant depressive symptoms, but only 49 (1.5%) met rigorous criteria for LG-PDD. Among those with LG-PDD, there was no significant difference in any outcome measure between the TRT and placebo groups, possibly reflecting low statistical power. In men with significant depressive symptoms (n = 2643) and in all randomly assigned participants (n = 5204), TRT was associated with modest but significantly greater improvements in mood and energy but not cognition or sleep quality.

Conclusion

Depressive symptoms are common in middle-aged and older men with hypogonadism but LG-PDD is uncommon. TRT is associated with small improvements in mood and energy in hypogonadal men with and without significant depressive symptoms.

hypogonadism, depressive symptoms, dysthymia/low-grade persistent depressive disorder, major depressive disorder, energy/fatigue, sleep and cognition

The role of testosterone in the pathogenesis and treatment of depressive symptoms and depressive disorders in men remains incompletely understood. Low mood or depressive symptoms are common in men with hypogonadism (1-3). Early-morning luteinizing hormone and testosterone secretion is attenuated in men with melancholic depression (4); however, studies have not shown consistently low testosterone levels in men with major depressive disorders (MDDs). Epidemiological studies find a more consistent association between low testosterone levels and mid- or late-life onset, persistent low-grade depressive disorder (LG-PDD), previously known as dysthymia (5, 6). Older men with LG-PDD have lower testosterone levels than age-matched, “nondepressed” men in the general population (5, 6).

The mood-elevating effects of testosterone and its potential use for the treatment of depressive disorders have been postulated since the 1940s (7). In the Testosterone Trials (TTrials) (8), testosterone treatment was associated with modest improvements in depressive symptoms in older men with hypogonadism (8); however, the participants were not selected based on the presence of a depressive symptoms or a depressive disorder. A meta-analysis of testosterone trials that included a heterogeneous group of men reporting depressive symptoms, some of whom were hypogonadal and some eugonadal, found modest improvements in depressive symptoms with testosterone treatment (9). The Testim Registry In the US (TRIUS), an uncontrolled observational registry study of men with hypogonadism receiving testosterone-replacement therapy (TRT) in a clinical setting, also reported modest improvements in depressive symptoms with TRT (1). TRT administered alone or as an augmentation therapy with traditional antidepressant drugs has not been consistently effective in treating MDDs in men (10-12). Observational studies (5, 6) and small randomized trials (13, 14) have generally suggested that TRT might be particularly effective for depressive symptoms in men with LG-PDD and hypogonadism. However, no randomized trial has evaluated the efficacy of TRT in improving depressive symptoms or inducing remission of depression in hypogonadal men with rigorously defined LG-PDD and hypogonadism.

The Testosterone Replacement therapy for Assessment of long-term Vascular Events and efficacy ResponSE in hypogonadal men (TRAVERSE) Study was designed to compare the effects of TRT and placebo on major adverse cardiovascular events (15, 16). The design and the primary findings of the TRAVERSE study have been published (15, 16). The study provided a unique opportunity to determine the prevalence of depressive symptoms as well as rigorously defined LG-PDD in middle-aged and older men with hypogonadism. The TRAVERSE Depression Substudy was nested within the parent trial to evaluate the effect of TRT in inducing remission and improving depressive symptoms in middle-aged and older men with hypogonadism and LG-PDD. A secondary aim was to evaluate the efficacy of TRT in improving depressive symptoms, energy, sleep, and cognition in middle-aged and older men with hypogonadism, who displayed significant depressive symptoms as defined by a Patient Health Questionnaire-9 (PHQ-9) score greater than 4 but who did not meet all the criteria for LG-PDD. Another secondary aim was to assess whether TRT improved depressive symptoms, energy, sleep quality, and cognition in all randomly assigned participants regardless of depressive status.

Materials and Methods

Eligibility Criteria

As reported previously (15, 16), the TRAVERSE trial enrolled men, aged 45 to 80 years, with 2 fasting, early-morning testosterone levels less than 300 ng/dL, 1 or more symptoms of hypogonadism, and preexisting cardiovascular disease (CVD) or increased risk of CVD. Men in whom TRT might increase the risk of harm, such as those with a history of prostate or breast cancer, prostate-specific antigen greater than 3 ng/mL, severe lower urinary tract symptoms, or erythrocytosis, were excluded.

We studied 3 groups of TRAVERSE participants: (1) the randomly assigned participants who met the prespecified rigorous criteria for LG-PDD (n = 49); (2) all randomly assigned participants who had significant depressive symptoms indicated by a PHQ-9 score greater than 4 (n = 2643); and (3) all randomly assigned participants (n = 5204). To be eligible for the LG-PDD substudy, the participants were required to consent to participation in the substudy, meet all of the eligibility criteria for the parent trial, and fulfill additional criteria for the diagnosis of LG-PDD (Table 1). Specifically, the men were required to (a) meet the Diagnostic and Statistical Manual of Mental Disorders-5 criteria for PDD (17, 18); (b) exhibit low-grade depressive symptoms (PHQ-9 score > 4 but < 15 and Geriatric Depression Scale [GDS] score between 5 and 9); (c) not have severe depression (PHQ-9 score ≥15, GDS-15 score > 9, a score of 3 on PHQ-9 items 1 and 2, or a self-reported history of MDDs); and (d) report onset of the depressive disorder after age 40 years. We excluded individuals who lacked persistent symptoms as evidenced by an answer “no” to the question, “Over the past 2 years, have you been feeling sad or depressed more days than not even if you felt OK sometimes?” and “yes” to the question, “Over the past 2 years, can you remember a period lasting 2 months or more when you were not feeling depressed much of the time?”. Finally, participants considered at increased suicide risk (PHQ-9 item 9 score >0) were excluded.

Table 1.

Open in new tab

Numbers and proportions of TRAVERSE participants meeting one or more inclusion and exclusion criteria for mid- or late-life, low-grade persistent depressive disorder

Description	n	N (%)
Full analysis set (randomly assigned participants in the parent trial included in the analysis)	5204	5204 (100%)
Consented to participate in the PDD substudy (% of full analysis set of 5204)		3359 (65%)
Met any of the following inclusion criteria for LG-PDD on PHQ-9 scores:
Total PHQ-9 score > 4	1810
PHQ-9 item 2 score of 1 or 2	1262
PHQ-9 item 10 score of “somewhat difficult” or greater^a	1403
Total eligible for LG-PDD based on PHQ-9 criteria (% of 3359 consenting participants)		2109 (63%)
Ineligible because of any of the following exclusion criteria for LG-PDD:
Total PHQ-9 score > 15	−151
PHQ-9 item 1 score of 3	−238
PHQ-9 item 2 score of 3	−109
PHQ-9 item 9 score of >0	−17
Ineligible—incomplete PHQ-9 questionnaire	−4
Total ineligible by out-of-range or missing PHQ-9 score (% of 2109 eligible participants)		−336 (10%)
Ineligible based on any of the following sequential additional questions involving missing questionnaires, major depressive disorder, onset before age 40, or symptoms not persistent:
Missing additional questionnaires	−296
Do you have any history of a major depressive episode?—answered YES	−273
Have you had any mood disorder prior to age 40?—answered YES	−107
Have you been treated for a mood disorder for longer than 6 months (eg, antidepressant medication, psychotherapy)?—answered YES	−165
Given your best guess: over the past 2 years, have you been feeling sad or depressed more days than not even if you felt ok sometimes?—answered NO	−690
Again, given your best guess, over the past 2 years can you remember a period of time lasting 2 months or more when you were not feeling depressed much of the time?—YES	−111
Again, given your best guess, how old were you when you were last feeling okay, that is, not feeling depressed most days?—<40 years	−26
Have you ever been diagnosed by a doctor or mental health professional with a major depressive disorder or with bipolar disorder?—answered YES	−2
Have you ever felt so bad that you sought help from a mental health professional and had psychotherapy for more than 6 months?—answered YES	−5
Have you ever been prescribed an antidepressant for a mood problem, and took it for more than 6 months?—answered YES	−8
Have you ever felt so bad that you attempted suicide?—answered YES	−1
Total ineligible based on responses to the above questions (% of the 2109 eligible participants)		−1684
Ineligible because of out-of-range or missing GDS-15 scores:
Incomplete GDS-15 questionnaire	0
Total GDS-15 score <5	−28
Total GDS-15 score >9	−12
Total ineligible because of out-of-range or missing GDS-15 scores		−40
Total enrolled (% of 3359 consenting participants)		49 (1.5%)

Description	n	N (%)
Full analysis set (randomly assigned participants in the parent trial included in the analysis)	5204	5204 (100%)
Consented to participate in the PDD substudy (% of full analysis set of 5204)		3359 (65%)
Met any of the following inclusion criteria for LG-PDD on PHQ-9 scores:
Total PHQ-9 score > 4	1810
PHQ-9 item 2 score of 1 or 2	1262
PHQ-9 item 10 score of “somewhat difficult” or greater^a	1403
Total eligible for LG-PDD based on PHQ-9 criteria (% of 3359 consenting participants)		2109 (63%)
Ineligible because of any of the following exclusion criteria for LG-PDD:
Total PHQ-9 score > 15	−151
PHQ-9 item 1 score of 3	−238
PHQ-9 item 2 score of 3	−109
PHQ-9 item 9 score of >0	−17
Ineligible—incomplete PHQ-9 questionnaire	−4
Total ineligible by out-of-range or missing PHQ-9 score (% of 2109 eligible participants)		−336 (10%)
Ineligible based on any of the following sequential additional questions involving missing questionnaires, major depressive disorder, onset before age 40, or symptoms not persistent:
Missing additional questionnaires	−296
Do you have any history of a major depressive episode?—answered YES	−273
Have you had any mood disorder prior to age 40?—answered YES	−107
Have you been treated for a mood disorder for longer than 6 months (eg, antidepressant medication, psychotherapy)?—answered YES	−165
Given your best guess: over the past 2 years, have you been feeling sad or depressed more days than not even if you felt ok sometimes?—answered NO	−690
Again, given your best guess, over the past 2 years can you remember a period of time lasting 2 months or more when you were not feeling depressed much of the time?—YES	−111
Again, given your best guess, how old were you when you were last feeling okay, that is, not feeling depressed most days?—<40 years	−26
Have you ever been diagnosed by a doctor or mental health professional with a major depressive disorder or with bipolar disorder?—answered YES	−2
Have you ever felt so bad that you sought help from a mental health professional and had psychotherapy for more than 6 months?—answered YES	−5
Have you ever been prescribed an antidepressant for a mood problem, and took it for more than 6 months?—answered YES	−8
Have you ever felt so bad that you attempted suicide?—answered YES	−1
Total ineligible based on responses to the above questions (% of the 2109 eligible participants)		−1684
Ineligible because of out-of-range or missing GDS-15 scores:
Incomplete GDS-15 questionnaire	0
Total GDS-15 score <5	−28
Total GDS-15 score >9	−12
Total ineligible because of out-of-range or missing GDS-15 scores		−40
Total enrolled (% of 3359 consenting participants)		49 (1.5%)

Abbreviations: GDS-15, Geriatric Depression Scale; LG-PDD, low-grade persistent depressive disorder; PHQ-9, Patient Health Questionnaire-9; TRAVERSE, Testosterone Replacement therapy for Assessment of long-term Vascular Events and efficacy ResponSE in hypogonadal men.

^aQuestion 10 reads as follows: If you checked off any problems, how difficult have those problems made it for you to do your work, care for things at home, or get along with other people? The response choices were: not difficult at all, somewhat difficult, very difficult, and extremely difficult. We required “somewhat difficult” or greater for the subject to meet eligibility.

Table 1.

Open in new tab

Numbers and proportions of TRAVERSE participants meeting one or more inclusion and exclusion criteria for mid- or late-life, low-grade persistent depressive disorder

Description	n	N (%)
Full analysis set (randomly assigned participants in the parent trial included in the analysis)	5204	5204 (100%)
Consented to participate in the PDD substudy (% of full analysis set of 5204)		3359 (65%)
Met any of the following inclusion criteria for LG-PDD on PHQ-9 scores:
Total PHQ-9 score > 4	1810
PHQ-9 item 2 score of 1 or 2	1262
PHQ-9 item 10 score of “somewhat difficult” or greater^a	1403
Total eligible for LG-PDD based on PHQ-9 criteria (% of 3359 consenting participants)		2109 (63%)
Ineligible because of any of the following exclusion criteria for LG-PDD:
Total PHQ-9 score > 15	−151
PHQ-9 item 1 score of 3	−238
PHQ-9 item 2 score of 3	−109
PHQ-9 item 9 score of >0	−17
Ineligible—incomplete PHQ-9 questionnaire	−4
Total ineligible by out-of-range or missing PHQ-9 score (% of 2109 eligible participants)		−336 (10%)
Ineligible based on any of the following sequential additional questions involving missing questionnaires, major depressive disorder, onset before age 40, or symptoms not persistent:
Missing additional questionnaires	−296
Do you have any history of a major depressive episode?—answered YES	−273
Have you had any mood disorder prior to age 40?—answered YES	−107
Have you been treated for a mood disorder for longer than 6 months (eg, antidepressant medication, psychotherapy)?—answered YES	−165
Given your best guess: over the past 2 years, have you been feeling sad or depressed more days than not even if you felt ok sometimes?—answered NO	−690
Again, given your best guess, over the past 2 years can you remember a period of time lasting 2 months or more when you were not feeling depressed much of the time?—YES	−111
Again, given your best guess, how old were you when you were last feeling okay, that is, not feeling depressed most days?—<40 years	−26
Have you ever been diagnosed by a doctor or mental health professional with a major depressive disorder or with bipolar disorder?—answered YES	−2
Have you ever felt so bad that you sought help from a mental health professional and had psychotherapy for more than 6 months?—answered YES	−5
Have you ever been prescribed an antidepressant for a mood problem, and took it for more than 6 months?—answered YES	−8
Have you ever felt so bad that you attempted suicide?—answered YES	−1
Total ineligible based on responses to the above questions (% of the 2109 eligible participants)		−1684
Ineligible because of out-of-range or missing GDS-15 scores:
Incomplete GDS-15 questionnaire	0
Total GDS-15 score <5	−28
Total GDS-15 score >9	−12
Total ineligible because of out-of-range or missing GDS-15 scores		−40
Total enrolled (% of 3359 consenting participants)		49 (1.5%)

Description	n	N (%)
Full analysis set (randomly assigned participants in the parent trial included in the analysis)	5204	5204 (100%)
Consented to participate in the PDD substudy (% of full analysis set of 5204)		3359 (65%)
Met any of the following inclusion criteria for LG-PDD on PHQ-9 scores:
Total PHQ-9 score > 4	1810
PHQ-9 item 2 score of 1 or 2	1262
PHQ-9 item 10 score of “somewhat difficult” or greater^a	1403
Total eligible for LG-PDD based on PHQ-9 criteria (% of 3359 consenting participants)		2109 (63%)
Ineligible because of any of the following exclusion criteria for LG-PDD:
Total PHQ-9 score > 15	−151
PHQ-9 item 1 score of 3	−238
PHQ-9 item 2 score of 3	−109
PHQ-9 item 9 score of >0	−17
Ineligible—incomplete PHQ-9 questionnaire	−4
Total ineligible by out-of-range or missing PHQ-9 score (% of 2109 eligible participants)		−336 (10%)
Ineligible based on any of the following sequential additional questions involving missing questionnaires, major depressive disorder, onset before age 40, or symptoms not persistent:
Missing additional questionnaires	−296
Do you have any history of a major depressive episode?—answered YES	−273
Have you had any mood disorder prior to age 40?—answered YES	−107
Have you been treated for a mood disorder for longer than 6 months (eg, antidepressant medication, psychotherapy)?—answered YES	−165
Given your best guess: over the past 2 years, have you been feeling sad or depressed more days than not even if you felt ok sometimes?—answered NO	−690
Again, given your best guess, over the past 2 years can you remember a period of time lasting 2 months or more when you were not feeling depressed much of the time?—YES	−111
Again, given your best guess, how old were you when you were last feeling okay, that is, not feeling depressed most days?—<40 years	−26
Have you ever been diagnosed by a doctor or mental health professional with a major depressive disorder or with bipolar disorder?—answered YES	−2
Have you ever felt so bad that you sought help from a mental health professional and had psychotherapy for more than 6 months?—answered YES	−5
Have you ever been prescribed an antidepressant for a mood problem, and took it for more than 6 months?—answered YES	−8
Have you ever felt so bad that you attempted suicide?—answered YES	−1
Total ineligible based on responses to the above questions (% of the 2109 eligible participants)		−1684
Ineligible because of out-of-range or missing GDS-15 scores:
Incomplete GDS-15 questionnaire	0
Total GDS-15 score <5	−28
Total GDS-15 score >9	−12
Total ineligible because of out-of-range or missing GDS-15 scores		−40
Total enrolled (% of 3359 consenting participants)		49 (1.5%)

Abbreviations: GDS-15, Geriatric Depression Scale; LG-PDD, low-grade persistent depressive disorder; PHQ-9, Patient Health Questionnaire-9; TRAVERSE, Testosterone Replacement therapy for Assessment of long-term Vascular Events and efficacy ResponSE in hypogonadal men.

^aQuestion 10 reads as follows: If you checked off any problems, how difficult have those problems made it for you to do your work, care for things at home, or get along with other people? The response choices were: not difficult at all, somewhat difficult, very difficult, and extremely difficult. We required “somewhat difficult” or greater for the subject to meet eligibility.

Participant Allocation

The participants were assigned using a concealed randomization with stratification for preexisting CVD to receive either 1.62% transdermal testosterone gel or a matching placebo gel daily for the duration of the study. Testosterone dose was titrated using a structured protocol based on on-treatment testosterone and hematocrit level to maintain testosterone levels between 350 and 750 ng/dL and hematocrit less than 54% (15, 16). The study participants, staff, and assessors were blinded to the study intervention.

Study Outcomes

Among men who met the criteria for LG-PDD, the PHQ-9 and GDS-15 were administered at baseline and at months 6, 12, 18 (PHQ-9 only), and 24 of the study. In all randomly assigned participants, mood was assessed using the mood domain of the Hypogonadism Impact of Symptoms Questionnaire (HIS-Q) at baseline and at months 6, 12, and 24 (19).

The PHQ-9 is a reliable and valid measure of depressive symptoms and severity (20) that has been used widely in the diagnosis of depressive disorders and for depression screening (21); higher scores indicate more severe depressive symptoms. Total scores of 5, 10, 15, and 20 represent cut points for mild, moderate, moderately severe, and severe depression, respectively. Question 9 of the PHQ-9 is a single screening question to evaluate suicide risk (20). The 15-item GDS-15 is a validated short form of the GDS that is used to screen and diagnose depression in older adults (22); higher scores indicate more severe depressive symptoms. GDS-15 scores of 0 to 4 are considered normal, 5 to 9 represent mild to moderate depression, and scores greater than 9 represent moderately severe to severe depression. Mood, energy, cognitive function, and sleep quality were evaluated using the HIS-Q, a validated, psychometrically robust instrument to evaluate hypogonadal symptoms that includes 28 questions addressing 5 domains (sexual, mood, energy, sleep, and cognition) (23). A higher HIS-Q composite score or domain/subdomain score indicates worse status.

The primary outcome measure of the LG-PDD substudy was the proportion of men who achieved remission at 6, 12, and 24 months. Remission of LG-PDD was defined as (a) PHQ-9 score less than 4 and GDS-15 score <5, and (b) answer “no” to the question “Give your best guess: Over the past 6 months, have you been feeling sad or depressed more days than not, even if you felt okay sometimes?” Secondary outcome measures were change from baseline in depressive symptoms on the PHQ-9, GDS-15 and HIS-Q scales.

In our secondary studies, the HIS-Q was administered to all participants regardless of depressive status. The outcome measures were change from baseline to months 6, 12, and 24 in depressive symptoms, ascertained using the HIS-Q mood domain and changes in sleep, energy/vitality, and cognition ascertained by the relevant domains of the HIS-Q. A higher HIS-Q composite score or domain/subdomain score indicates worse status.

As reported previously (12, 13), serum concentrations of testosterone, dihydrotestosterone (DHT), and estradiol were measured by liquid chromatography–tandem mass spectrometry (LC-MS/MS) assays in samples collected in the morning after an overnight fast in a central laboratory (LabCorp Inc) that is certified by the Hormone Standardization Program for Testosterone of the Center for Disease Control and Prevention. The lower limit of quantitation (LLOQ) for the testosterone assay was 2 ng/dL; interassay coefficients of variation (CV) at 2, 43, 194, and 913 ng/dL were 2.4%, 1.3%, 1.4%, and 1.7%, and the intra-assay CV at the corresponding concentrations were 8.6%, 5.6%, 5.4%, and 5.1%, respectively. For the estradiol LC-MS/MS assay, the LLOQ was 0.5 pg/mL, the interassay and intra-assay CVs in quality control pools with estradiol concentrations at the LLOQ, low, medium, and high range were 7.3%, 7.5%, 5.8%, 8.1% and: 14.1%, 2.5%, 1.8%, and 2.0%, respectively. For the DHT assay, the LLOQ was 10 ng/dL, and the interassay and intra-assay CVs in quality control pools with DHT concentrations in the low, medium, and high male range were 3.1%, 3.0%, and 1.6%, and 3.0%, 2.9%, and 2.6% respectively.

Statistical Analyses

The analyses were performed using the intent-to-treat principle; all available data at scheduled time points were included regardless of treatment adherence. The prespecified outcome analyses were performed in 3 analytic samples: (1) the randomly assigned participants who met the criteria for LG-PDD (n = 49); (2) all randomly assigned participants who reported significant depressive symptoms (n = 2643); and (3) all randomly assigned participants in the full analysis set (FAS) (n = 5204).

The risk ratio of remission of LG-PDD in the TRT vs placebo group was estimated by a repeated-measures generalized estimating equation Poisson regression model with fixed effects for treatment, visit, treatment-visit interaction, preexisting CVD, and an unstructured working correlation matrix to account for repeated measures at multiple visits. The estimates of the risk ratio of TRT over placebo together with 95% CIs and an associated generalized score statistic P value testing the null hypothesis of no difference between the TRT and placebo across all time points were derived from the model.

Change from baseline in PHQ-9, GDS-15, HIS-Q Mood, Energy, and Cognitive scores and sex hormones were analyzed using a linear mixed-effects model with fixed effects for treatment, visit, treatment-visit interaction, baseline value, and prior CVD status, allowing for intrasubject correlation using an unstructured covariance matrix.

The association between standardized mean postrandomization change scores (HIS-Q Mood and Energy) and mean change in sex hormones in testosterone-treated men was analyzed using linear regression, adjusted for preexisting CVD. The sample size for the PDD substudy was not prespecified but based on the number of men in the parent trial who met the prespecified definition.

Results

Study Population

Enrollment took place between May 23, 2018, and February 1, 2022, and end-of-study visits were conducted between May 31, 2022, and January 19, 2023. As reported previously (15, 16), in April 2019, a preplanned look at the blinded data from the first 2669 patients showed a pooled primary event rate below the projected rate of 1.5% per year. Therefore, the enrollment of patients who qualified via cardiovascular risk factors was stopped May 31, 2019 (16); after this date, only those with preexisting CVD were enrolled in the study. Fifty-five percent of study participants had preexisting CVD (16).

Among 5246 randomized participant IDs, 42 were attributed to duplicate enrollment. After excluding these, the FAS included 5204 participants, of whom 2643 reported significant depressive symptoms (PHQ-9 > 4). A total of 3359 (65%) individuals consented to participate in the LG-PDD substudy and underwent additional screening to determine if they met the prespecified criteria for LG-PDD (see Table 1 and Fig. 1); 3310 out of these 3359 participants were excluded from the LG-PDD substudy because of the reasons listed in Table 1, including out-of-range or missing PHQ-9 scores, missing additional questionnaires, lack of persistent depressive symptoms, major depressive disorder or treatment for a mood disorder, age of onset younger than 40 years, or out-of-range or missing GDS-15 scores. Only 49 participants (1.5%) met all the rigorously defined criteria for LG-PDD; of these, 26 were randomly assigned to the testosterone and 23 to the placebo group.

Figure 1.

The 3 subgroups of patients enrolled in the TRAVERSE study who were analyzed for the effect of testosterone-replacement therapy (TRT) on depressive symptoms and other prespecified outcome measures.

Open in new tab Download slide

Baseline characteristics were similar between the testosterone and placebo groups in participants with LG-PDD, as well as in men with significant depressive symptoms (PHQ-9 score > 4) (Table 2). The mean (SD) age of the participants with significant depressive symptoms (PHQ-9 score > 4) was 62.2 (7.8) years; mean body mass index 35.3 (5.8); PHQ-9 score 9.1 (3.9); GDS-15 score 7.0 (1.3); HIS-Q energy domain score 59.9 (20.6); HIS-Q mood domain score 44.6 (15.8); HIS-Q sleep domain score 45.0 (17.4); and 56% had prior CVD. The mean (SD) age of the participants with LG-PDD was 62.7 (6.6) years, mean body mass index 36.4 (6.5), PHQ-9 score 8.6 (3.3); GDS-15 score 7.0 (1.3); HIS-Q energy domain score 64.8 (16.3); HIS-Q mood domain score 52.2 (13.7); HIS-Q sleep domain score 51.7 (16.2); and 47% had prior CVD.

Table 2.

Open in new tab

Baseline characteristics of participants randomly assigned in the TRAVERSE trial who had significant depressive symptoms (Patient Health Questionnaire-9 score > 4) and in participants who met the criteria for mid- or late-life low-grade persistent depressive disorder (previously “dysthymia”)

	TRAVERSE full analysis set with PHQ-9 > 4		Depression study in men meeting definition of LG-PDD
Variable	TRT (N = 1308)	Placebo (N = 1335)	TRT (N = 26)	Placebo (N = 23)
Age, y	62.2 (7.7)	62.1 (7.8)	62.2 (6.7)	63.3 (6.7)
Age group, y
45-<65	765 (58.5%)	792 (59.3%)	18 (69.2%)	11 (47.8%)
65+	543 (41.5%)	543 (40.7%)	8 (30.8%)	12 (52.2%)
Race group
White	1064 (81.3%)	1069 (80.1%)	23 (88.5%)	17 (73.9%)
Black/African American	208 (15.9%)	220 (16.5%)	3 (11.5%)	6 (26.1%)
Other	36 (2.8%)	46 (3.4%)	0 (0.0%)	0 (0.0%)
Ethnicity
Hispanic/Latino	198 (15.1%)	230 (17.3%)	2 (7.7%)	2 (8.7%)
Not Hispanic/Latino	1109 (84.9%)	1103 (82.7%)	24 (92.3%)	21 (91.3%)
Missing	1 (−)	2 (−)
Body mass index	35.3 (5.7) (n = 1307)	35.3 (6.0)	36.6 (5.7)	36.3 (7.4)
Prior CVD	750 (57.3%)	742 (55.6%)	10 (38.5%)	13 (56.5%)
At risk for CVD	558 (42.7%)	593 (44.4%)	16 (61.5%)	10 (43.5%)
Prior testosterone use	3 (0.2%)	6 (0.4%)	0 (0.0%)	0 (0.0%)
Hypertension	1203 (92.0%)	1230 (92.1%)	26 (100.0%)	17 (73.9%)
Dyslipidemia	1190 (91.0%)	1177 (88.2%)	26 (100.0%)	21 (91.3%)
Current alcohol use
Yes	698 (53.4%)	730 (54.8%)	15 (57.7%)	16 (69.6%)
Missing	1 (−)	4 (−)
Current nicotine use
Yes	281 (21.5%)	289 (21.6%)	5 (19.2%)	4 (17.4%)
Missing	1 (−)
Sitting systolic blood pressure, mm Hg	132.6 (15.8)	132.9 (15.7)	131.8 (15.0)	129.8 (15.7)
Sitting diastolic blood pressure, mm Hg	79.6 (9.9)	79.9 (9.7)	78.2 (9.6)	80.1 (9.3)
Testosterone, ng/dL	219.0 (46.6)	217.2 (49.2)	203.3 (47.1)	197.5 (56.7)
Dihydrotestosterone, ng/dL	16.0 (7.8) (n = 1248)	15.7 (8.6) (n = 1279)	14.7 (7.3) (n = 25)	13.1 (5.3) (n = 22)
Estradiol, pg/mL	20.9 (8.1) (n = 1241)	21.0 (8.6) (n = 1275)	22.1 (7.9) (n = 23)	21.5 (9.6) (n = 20)
HIS-Q total score	51.9 (10.4) (n = 1215)	51.5 (10.4) (n = 1231)	55.2 (6.7)	57.4 (9.5)
HIS-Q mood domain score	44.7 (15.7) (n = 1231)	44.5 (15.9) (n = 1241)	50.3 (13.4)	54.3 (14.1)
HIS-Q cognition domain score	42.9 (16.0) (n = 1229)	41.7 (15.9) (n = 1240)	45.5 (16.9)	43.8 (16.7)
HIS-Q energy domain score	60.6 (20.1) (n = 1231)	59.2 (21.0) (n = 1238)	60.9 (14.5)	69.2 (17.5)
GDS-15 score			6.7 (1.3)	7.3 (1.1)
PHQ-9 score	9.1 (3.9)	9.0 (3.9)	7.8 (2.7)	9.5 (3.6)

	TRAVERSE full analysis set with PHQ-9 > 4		Depression study in men meeting definition of LG-PDD
Variable	TRT (N = 1308)	Placebo (N = 1335)	TRT (N = 26)	Placebo (N = 23)
Age, y	62.2 (7.7)	62.1 (7.8)	62.2 (6.7)	63.3 (6.7)
Age group, y
45-<65	765 (58.5%)	792 (59.3%)	18 (69.2%)	11 (47.8%)
65+	543 (41.5%)	543 (40.7%)	8 (30.8%)	12 (52.2%)
Race group
White	1064 (81.3%)	1069 (80.1%)	23 (88.5%)	17 (73.9%)
Black/African American	208 (15.9%)	220 (16.5%)	3 (11.5%)	6 (26.1%)
Other	36 (2.8%)	46 (3.4%)	0 (0.0%)	0 (0.0%)
Ethnicity
Hispanic/Latino	198 (15.1%)	230 (17.3%)	2 (7.7%)	2 (8.7%)
Not Hispanic/Latino	1109 (84.9%)	1103 (82.7%)	24 (92.3%)	21 (91.3%)
Missing	1 (−)	2 (−)
Body mass index	35.3 (5.7) (n = 1307)	35.3 (6.0)	36.6 (5.7)	36.3 (7.4)
Prior CVD	750 (57.3%)	742 (55.6%)	10 (38.5%)	13 (56.5%)
At risk for CVD	558 (42.7%)	593 (44.4%)	16 (61.5%)	10 (43.5%)
Prior testosterone use	3 (0.2%)	6 (0.4%)	0 (0.0%)	0 (0.0%)
Hypertension	1203 (92.0%)	1230 (92.1%)	26 (100.0%)	17 (73.9%)
Dyslipidemia	1190 (91.0%)	1177 (88.2%)	26 (100.0%)	21 (91.3%)
Current alcohol use
Yes	698 (53.4%)	730 (54.8%)	15 (57.7%)	16 (69.6%)
Missing	1 (−)	4 (−)
Current nicotine use
Yes	281 (21.5%)	289 (21.6%)	5 (19.2%)	4 (17.4%)
Missing	1 (−)
Sitting systolic blood pressure, mm Hg	132.6 (15.8)	132.9 (15.7)	131.8 (15.0)	129.8 (15.7)
Sitting diastolic blood pressure, mm Hg	79.6 (9.9)	79.9 (9.7)	78.2 (9.6)	80.1 (9.3)
Testosterone, ng/dL	219.0 (46.6)	217.2 (49.2)	203.3 (47.1)	197.5 (56.7)
Dihydrotestosterone, ng/dL	16.0 (7.8) (n = 1248)	15.7 (8.6) (n = 1279)	14.7 (7.3) (n = 25)	13.1 (5.3) (n = 22)
Estradiol, pg/mL	20.9 (8.1) (n = 1241)	21.0 (8.6) (n = 1275)	22.1 (7.9) (n = 23)	21.5 (9.6) (n = 20)
HIS-Q total score	51.9 (10.4) (n = 1215)	51.5 (10.4) (n = 1231)	55.2 (6.7)	57.4 (9.5)
HIS-Q mood domain score	44.7 (15.7) (n = 1231)	44.5 (15.9) (n = 1241)	50.3 (13.4)	54.3 (14.1)
HIS-Q cognition domain score	42.9 (16.0) (n = 1229)	41.7 (15.9) (n = 1240)	45.5 (16.9)	43.8 (16.7)
HIS-Q energy domain score	60.6 (20.1) (n = 1231)	59.2 (21.0) (n = 1238)	60.9 (14.5)	69.2 (17.5)
GDS-15 score			6.7 (1.3)	7.3 (1.1)
PHQ-9 score	9.1 (3.9)	9.0 (3.9)	7.8 (2.7)	9.5 (3.6)

Continuous outcomes are presented as mean (SD) and categorical as numbers and percentages. Race category “other” includes Asian, Multi-race, American Indian/Alaskan Native, Native Hawaiian, or other Pacific Islander.

To convert serum total testosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply testosterone concentration in nanograms per deciliter by 0.0347. To convert estradiol concentrations from picogram per milliliter to picomoles per liter, multiply estradiol concentrations in picogram per milliliter by 3.67. To convert dihydrotestosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply dihydrotestosterone concentrations in nanograms per deciliter by 0.0344.

Abbreviations: CVD, cardiovascular disease; GDS-15, Geriatric Depression Scale; HIS-Q, Hypogonadism Impact of Symptoms Questionnaire; LG-PDD, mid- or late-life onset, low-grade persistent depressive disorder; PHQ-9, Patient Health Questionnaire-9; TRT, testosterone-replacement therapy.

Table 2.

Open in new tab

Baseline characteristics of participants randomly assigned in the TRAVERSE trial who had significant depressive symptoms (Patient Health Questionnaire-9 score > 4) and in participants who met the criteria for mid- or late-life low-grade persistent depressive disorder (previously “dysthymia”)

	TRAVERSE full analysis set with PHQ-9 > 4		Depression study in men meeting definition of LG-PDD
Variable	TRT (N = 1308)	Placebo (N = 1335)	TRT (N = 26)	Placebo (N = 23)
Age, y	62.2 (7.7)	62.1 (7.8)	62.2 (6.7)	63.3 (6.7)
Age group, y
45-<65	765 (58.5%)	792 (59.3%)	18 (69.2%)	11 (47.8%)
65+	543 (41.5%)	543 (40.7%)	8 (30.8%)	12 (52.2%)
Race group
White	1064 (81.3%)	1069 (80.1%)	23 (88.5%)	17 (73.9%)
Black/African American	208 (15.9%)	220 (16.5%)	3 (11.5%)	6 (26.1%)
Other	36 (2.8%)	46 (3.4%)	0 (0.0%)	0 (0.0%)
Ethnicity
Hispanic/Latino	198 (15.1%)	230 (17.3%)	2 (7.7%)	2 (8.7%)
Not Hispanic/Latino	1109 (84.9%)	1103 (82.7%)	24 (92.3%)	21 (91.3%)
Missing	1 (−)	2 (−)
Body mass index	35.3 (5.7) (n = 1307)	35.3 (6.0)	36.6 (5.7)	36.3 (7.4)
Prior CVD	750 (57.3%)	742 (55.6%)	10 (38.5%)	13 (56.5%)
At risk for CVD	558 (42.7%)	593 (44.4%)	16 (61.5%)	10 (43.5%)
Prior testosterone use	3 (0.2%)	6 (0.4%)	0 (0.0%)	0 (0.0%)
Hypertension	1203 (92.0%)	1230 (92.1%)	26 (100.0%)	17 (73.9%)
Dyslipidemia	1190 (91.0%)	1177 (88.2%)	26 (100.0%)	21 (91.3%)
Current alcohol use
Yes	698 (53.4%)	730 (54.8%)	15 (57.7%)	16 (69.6%)
Missing	1 (−)	4 (−)
Current nicotine use
Yes	281 (21.5%)	289 (21.6%)	5 (19.2%)	4 (17.4%)
Missing	1 (−)
Sitting systolic blood pressure, mm Hg	132.6 (15.8)	132.9 (15.7)	131.8 (15.0)	129.8 (15.7)
Sitting diastolic blood pressure, mm Hg	79.6 (9.9)	79.9 (9.7)	78.2 (9.6)	80.1 (9.3)
Testosterone, ng/dL	219.0 (46.6)	217.2 (49.2)	203.3 (47.1)	197.5 (56.7)
Dihydrotestosterone, ng/dL	16.0 (7.8) (n = 1248)	15.7 (8.6) (n = 1279)	14.7 (7.3) (n = 25)	13.1 (5.3) (n = 22)
Estradiol, pg/mL	20.9 (8.1) (n = 1241)	21.0 (8.6) (n = 1275)	22.1 (7.9) (n = 23)	21.5 (9.6) (n = 20)
HIS-Q total score	51.9 (10.4) (n = 1215)	51.5 (10.4) (n = 1231)	55.2 (6.7)	57.4 (9.5)
HIS-Q mood domain score	44.7 (15.7) (n = 1231)	44.5 (15.9) (n = 1241)	50.3 (13.4)	54.3 (14.1)
HIS-Q cognition domain score	42.9 (16.0) (n = 1229)	41.7 (15.9) (n = 1240)	45.5 (16.9)	43.8 (16.7)
HIS-Q energy domain score	60.6 (20.1) (n = 1231)	59.2 (21.0) (n = 1238)	60.9 (14.5)	69.2 (17.5)
GDS-15 score			6.7 (1.3)	7.3 (1.1)
PHQ-9 score	9.1 (3.9)	9.0 (3.9)	7.8 (2.7)	9.5 (3.6)

	TRAVERSE full analysis set with PHQ-9 > 4		Depression study in men meeting definition of LG-PDD
Variable	TRT (N = 1308)	Placebo (N = 1335)	TRT (N = 26)	Placebo (N = 23)
Age, y	62.2 (7.7)	62.1 (7.8)	62.2 (6.7)	63.3 (6.7)
Age group, y
45-<65	765 (58.5%)	792 (59.3%)	18 (69.2%)	11 (47.8%)
65+	543 (41.5%)	543 (40.7%)	8 (30.8%)	12 (52.2%)
Race group
White	1064 (81.3%)	1069 (80.1%)	23 (88.5%)	17 (73.9%)
Black/African American	208 (15.9%)	220 (16.5%)	3 (11.5%)	6 (26.1%)
Other	36 (2.8%)	46 (3.4%)	0 (0.0%)	0 (0.0%)
Ethnicity
Hispanic/Latino	198 (15.1%)	230 (17.3%)	2 (7.7%)	2 (8.7%)
Not Hispanic/Latino	1109 (84.9%)	1103 (82.7%)	24 (92.3%)	21 (91.3%)
Missing	1 (−)	2 (−)
Body mass index	35.3 (5.7) (n = 1307)	35.3 (6.0)	36.6 (5.7)	36.3 (7.4)
Prior CVD	750 (57.3%)	742 (55.6%)	10 (38.5%)	13 (56.5%)
At risk for CVD	558 (42.7%)	593 (44.4%)	16 (61.5%)	10 (43.5%)
Prior testosterone use	3 (0.2%)	6 (0.4%)	0 (0.0%)	0 (0.0%)
Hypertension	1203 (92.0%)	1230 (92.1%)	26 (100.0%)	17 (73.9%)
Dyslipidemia	1190 (91.0%)	1177 (88.2%)	26 (100.0%)	21 (91.3%)
Current alcohol use
Yes	698 (53.4%)	730 (54.8%)	15 (57.7%)	16 (69.6%)
Missing	1 (−)	4 (−)
Current nicotine use
Yes	281 (21.5%)	289 (21.6%)	5 (19.2%)	4 (17.4%)
Missing	1 (−)
Sitting systolic blood pressure, mm Hg	132.6 (15.8)	132.9 (15.7)	131.8 (15.0)	129.8 (15.7)
Sitting diastolic blood pressure, mm Hg	79.6 (9.9)	79.9 (9.7)	78.2 (9.6)	80.1 (9.3)
Testosterone, ng/dL	219.0 (46.6)	217.2 (49.2)	203.3 (47.1)	197.5 (56.7)
Dihydrotestosterone, ng/dL	16.0 (7.8) (n = 1248)	15.7 (8.6) (n = 1279)	14.7 (7.3) (n = 25)	13.1 (5.3) (n = 22)
Estradiol, pg/mL	20.9 (8.1) (n = 1241)	21.0 (8.6) (n = 1275)	22.1 (7.9) (n = 23)	21.5 (9.6) (n = 20)
HIS-Q total score	51.9 (10.4) (n = 1215)	51.5 (10.4) (n = 1231)	55.2 (6.7)	57.4 (9.5)
HIS-Q mood domain score	44.7 (15.7) (n = 1231)	44.5 (15.9) (n = 1241)	50.3 (13.4)	54.3 (14.1)
HIS-Q cognition domain score	42.9 (16.0) (n = 1229)	41.7 (15.9) (n = 1240)	45.5 (16.9)	43.8 (16.7)
HIS-Q energy domain score	60.6 (20.1) (n = 1231)	59.2 (21.0) (n = 1238)	60.9 (14.5)	69.2 (17.5)
GDS-15 score			6.7 (1.3)	7.3 (1.1)
PHQ-9 score	9.1 (3.9)	9.0 (3.9)	7.8 (2.7)	9.5 (3.6)

Continuous outcomes are presented as mean (SD) and categorical as numbers and percentages. Race category “other” includes Asian, Multi-race, American Indian/Alaskan Native, Native Hawaiian, or other Pacific Islander.

To convert serum total testosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply testosterone concentration in nanograms per deciliter by 0.0347. To convert estradiol concentrations from picogram per milliliter to picomoles per liter, multiply estradiol concentrations in picogram per milliliter by 3.67. To convert dihydrotestosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply dihydrotestosterone concentrations in nanograms per deciliter by 0.0344.

Abbreviations: CVD, cardiovascular disease; GDS-15, Geriatric Depression Scale; HIS-Q, Hypogonadism Impact of Symptoms Questionnaire; LG-PDD, mid- or late-life onset, low-grade persistent depressive disorder; PHQ-9, Patient Health Questionnaire-9; TRT, testosterone-replacement therapy.

Among men with significant depressive symptoms, the mean (SD) treatment duration was 20.7 (14.1) and 21.3 (13.9) months and the mean follow-up duration was 32.6 (12.4) and 33.0 (12.1) in the TRT and placebo groups, respectively. Among those with LG-PDD, the mean (SD) treatment duration was 19.5 (13.9) and 26.2 (15.0) months and the mean follow-up duration was 33.6 (12.2) and 38.1 (7.9) in the TRT and placebo groups, respectively.

Effects of Testosterone-Replacement Therapy on Outcomes in Men With Low-Grade Persistent Depressive Disorder

The relative risk ratio for LG-PDD remission did not differ significantly between the TRT and placebo groups (Fig. 2). The changes from baseline in PHQ-9, GDS-15, and HIS-Q scores were numerically greater in testosterone-treated than in placebo-treated men but the differences were not significantly different between the 2 groups, likely due to the small sample size (Fig. 3).

Figure 2.

Number (percentage) of participants with low-grade persistent depressive disorder (LG-PDD) at baseline experiencing remission at various time points during the intervention period. Frequencies and relative risks of LG-PDD remission in the testosterone-replacement therapy (TRT) group relative to the placebo group and the associated 95% CIs at each visit in LG-PDD substudy participants are shown by treatment group and time point. The risk ratio of PDD remission in the TRT vs placebo group was estimated by a repeated-measures generalized estimating equation (GEE) Poisson regression model with fixed effects for treatment, visit, treatment-visit interaction, and preexisting cardiovascular disease using an unstructured covariance matrix to account for repeated measures at multiple visits. The omnibus test P value shown in the figure is a test of the null hypothesis of no difference between TRT and placebo groups across all time points.

Open in new tab Download slide

Figure 3.

Estimated change from baseline in depressive symptom scores using the Patient Health Questionnaire-9 (PHQ-9) (upper panel), Geriatric Depression Scale (GDS-15) (middle panel), and Hypogonadism Impact of Symptoms Questionnaire (HIS-Q) mood score (lower panel) in low-grade persistent depressive disorder (LG-PDD) substudy participants. Linear mixed models were fit with fixed effects for treatment, visit, and treatment-visit interaction, baseline value, and preexisting cardiovascular disease and a random per-subject repeated-measures effect with an unstructured covariance matrix. The least significant means and the corresponding 95% CIs are shown by treatment group and time point. Each omnibus test P value is from a test of the null hypothesis of no difference between the testosterone-replacement therapy and placebo groups across all time points.

Open in new tab Download slide

Effects of Testosterone-Replacement Therapy in All Randomly Assigned Men With Significant Depressive Symptoms (Patient Health Questionnaire-9 Score > 4)

Men with significant depressive symptoms randomly assigned to TRT showed significantly greater improvement in depressive symptoms than those randomly assigned to placebo, as assessed using the HIS-Q mood domain score (between-group LS mean difference [95% CI] −0.70 [−1.9 to 0.5] at month 6; −2.3 [−3.6 to −1.0] at month 12; and −0.9 [−2.5 to 0.6] at month 24; omnibus test P = .008) (Fig. 4, upper panel). Higher scores indicate more depressive symptoms. TRT also was associated with greater improvements in the energy domain score more than placebo (between-group difference −1.9 [−3.6 to −0.2] at month 6; −3.1 [−5.0 to −1.2] at month 12; and −1.8 [−4.0 to 0.4] at month 24; omnibus test P = .010; Fig. 5, middle panel). A lower score indicates a higher energy level. However, the changes in cognition and sleep scores did not differ significantly between the testosterone and placebo groups (Figs. 6 and 7, middle panel). Among all men with a baseline PHQ-9 score greater than 4 who were randomly assigned to TRT, mean changes in HISQ energy score were positively associated with mean changes in HISQ mood score (r² = 0.194; P <.0001).

Figure 4.

Estimated change from baseline in mood score (Hypogonadism Impact of Symptoms Questionnaire [HIS-Q]) in randomly assigned participants with a baseline Patient Health Questionnaire-9 (PHQ-9) score greater than 4 (upper panel) and all randomly assigned participants (lower panel). Linear mixed models were fit with fixed effects for treatment, visit, and treatment-visit interaction, baseline value, and preexisting cardiovascular disease and a random per-subject repeated-measures effect with an unstructured covariance matrix. The least significant means and the corresponding 95% CIs are shown by treatment group and time point. Each omnibus test P value is from a test of the null hypothesis of no difference between the testosterone-replacement therapy and placebo groups across all time points.

Open in new tab Download slide

Figure 5.

Estimated change from baseline in energy score (Hypogonadism Impact of Symptoms Questionnaire [HIS-Q]) in A, persistent depressive disorder (PDD) substudy participants; B, randomly assigned participants with a baseline Patient Health Questionnaire-9 (PHQ-9) score greater than 4; and C, all randomly assigned participants. Linear mixed models were fit with fixed effects for treatment, visit, and treatment-visit interaction, baseline value, and preexisting cardiovascular disease and a random per-subject repeated-measures effect with an unstructured covariance matrix. The least significant means and the corresponding 95% CIs are shown by treatment group and time point. Each omnibus test P value is from a test of the null hypothesis of no difference between the testosterone-replacement therapy and placebo groups across all time points.

Open in new tab Download slide

Figure 6.

Estimated change from baseline in cognitive score (Hypogonadism Impact of Symptoms Questionnaire [HIS-Q]) in persistent depressive disorder (PDD) substudy participants (upper panel), randomly assigned participants with a baseline Patient Health Questionnaire-9 (PHQ-9) score greater than 4 (middle panel), and all randomly assigned participants (lower panel). Linear mixed models were fit with fixed effects for treatment, visit, and treatment-visit interaction, baseline value, and preexisting cardiovascular disease and a random per-subject repeated-measures effect with an unstructured covariance matrix. The least significant means and the corresponding 95% CIs are shown by treatment group and time point. Each omnibus test P value is from a test of the null hypothesis of no difference between the testosterone-replacement therapy and placebo groups across all time points.

Open in new tab Download slide

Figure 7.

Estimated change from baseline in sleep domain score (Hypogonadism Impact of Symptoms Questionnaire [HIS-Q]) in persistent depressive disorder (PDD) substudy participants (upper panel), randomly assigned participants with a baseline Patient Health Questionnaire-9 (PHQ-9) score greater than 4 (middle panel), and all randomly assigned participants (lower panel). Linear mixed models were fit with fixed effects for treatment, visit, and treatment-visit interaction, baseline value, and preexisting cardiovascular disease and a random per-subject repeated-measures effect with an unstructured covariance matrix. The least significant means and the corresponding 95% CIs are shown by treatment group and time point. Each omnibus test P value is from a test of the null hypothesis of no difference between the testosterone-replacement therapy and placebo groups across all time points.

Open in new tab Download slide

In a post hoc analysis of men (n = 208 with at least 1 follow-up) with a baseline PHQ-9 score of 15 or higher (moderately severe or severe depressive symptoms), the change in depressive symptoms as measured by HIS-Q mood score did not differ significantly between the testosterone and placebo-treated men (between-group difference [95% CI] at 6 months 0.5 [−3.8 to 4. 8]; −2.0 [−7.2 to 3.2] at 12 months, and 1.4 [−4.5 to 7.2] at 24 months; omnibus test P = .7). In contrast, among men with mild to moderate depressive symptoms (PHQ-9 score 5-14), TRT was associated with greater improvements in depressive symptoms than placebo (between-group difference [95% CI] −0.8 [−2.0 to 0.5] at 6 months, and −2.3 [−3.6 to −0.9] at 12 months, and −1.2 [−2.7 to 0.4]; omnibus test P = .012).

The numbers (incidences) of men with completed (1 [0.1%] in each group; P = .988) or attempted suicide (0 [0%] in TRT and 2 [0.1%] in placebo group; P = .161) and new diagnoses of psychiatric disorders in men with a baseline PHQ-9 score greater than 4 (24 [1.8%]) in the TRT group and 23 [1.7%] in the placebo group; P = .828) were small and did not differ between groups.

Effects of Testosterone-Replacement Therapy in All Randomly Assigned Participants

In all randomly assigned participants in TRAVERSE, TRT was associated with a small but significantly greater improvement in mood than placebo treatment (mean between-group least significant [LS] difference −1.0 [−1.8 to −0.2] at 6 months; −1.6 [−2.5 to −0.7] at 12 months; −1.3 [−2.3 to −0.2]; omnibus test P = .003] (see Fig. 4, lower panel). TRT also significantly improved energy in all participants in TRAVERSE compared to placebo treatment (mean [95% CI] between-group LS difference: −2.2 [−3.4 to −1.0] at 6 months; −2.7 [−4.0 to −1.4] at 12 months, and −1.9 [−3.4 to −0.4]; omnibus test P < .001; see Fig. 5, lower panel). The cognition and sleep scores did not differ significantly between the TRT and placebo groups (see Figs. 6 and 7, lower panel). Among all men randomly assigned to TRT, mean changes in HIS-Q energy score were positively associated with mean changes in HIS-Q mood score (r² = 0.203; P < .0001).

The numbers (incidences) of randomly assigned participants with completed (1 [<0.1%] in each group; P = .999) or attempted suicide (0 [0%] in TRT and 2 [0.1%] in placebo group; P = .157) or new diagnoses of psychiatric disorders (48 [1.8%] in the TRT and 41 [1.6%] in the placebo group; P = .452) did not differ between groups.

Hormone Levels

Serum testosterone, DHT, and estradiol levels increased significantly more in the TRT group than in the placebo group in men with LG-PDD, men with PHQ-9 scores greater than 4 (Table 3), and in all randomly assigned men (reported previously) (16).

Table 3.

Open in new tab

Changes from baseline in total testosterone, dihydrotestosterone, and estradiol levels in randomly assigned men with low-grade persistent depressive disorder or with Patient Health Questionnaire-9 score greater than 4

Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	Change from baseline		Treatment difference		P
Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	LS mean	95% CI	LS mean	95% CI	P
Testosterone	LG-PDD	3	Testosterone	23	198.5 (45.8)	417.9 (224.3)	213.5	(149.2 to 277.7)	213.6	to	<.001
			Placebo	20	193.5 (59.2)	192.4 (85.2)	−0.2	(−69.3 to 68.9)		to
		6	Testosterone	22	201.6 (44.7)	370.1 (227.2)	170.6	(105.0 to 236.2)	177.3	(82.0 to 272.6)
			Placebo	20	200.0 (57.9)	191.5 (72.2)	−6.7	(−75.6 to 62.2)
		12	Testosterone	17	209.8 (42.3)	426.0 (298.9)	220.8	(146.5 to 295.2)	219.0	(116.4 to 321.7)
			Placebo	19	204.2 (56.2)	202.2 (89.6)	1.8	(−68.9 to 72.5)
		18	Testosterone	13	206.4 (44.6)	310.9 (155.8)	99.1	(14.4 to 183.9)	51.8	(−64.2 to 167.8)
			Placebo	15	212.4 (55.4)	255.9 (143.8)	47.4	(−31.8 to 126.5)
		24	Testosterone	11	204.6 (50.1)	301.5 (121.9)	91.7	(−0.4 to 183.8)	68.2	(−53.4 to 189.8)
			Placebo	15	208.6 (54.3)	226.2 (110.5)	23.5	(−55.6 to 102.5)
	PHQ-9 > 4	3	Testosterone	1204	219.9 (46.1)	412.0 (250.4)	193.2	(182.5 to 203.9)	176.0	(161.0 to 191.0)	<.001
			Placebo	1249	217.0 (48.8)	235.4 (103.5)	17.2	(6.7 to 27.7)
		6	Testosterone	1104	219.9 (46.0)	430.2 (309.6)	211.7	(198.4 to 225.1)	193.2	(174.5 to 211.9)
			Placebo	1148	217.4 (48.9)	236.5 (97.0)	18.5	(5.4 to 31.6)
		12	Testosterone	900	219.5 (46.1)	421.9 (263.0)	203.3	(189.7 to 216.8)	169.2	(150.2 to 188.1)
			Placebo	949	216.3 (48.4)	251.4 (139.6)	34.1	(20.9 to 47.3)
		18	Testosterone	764	220.5 (44.3)	436.9 (321.8)	210.9	(194.2 to 227.7)	180.0	(156.4 to 203.5)
			Placebo	788	216.8 (48.5)	244.7 (121.7)	31.0	(14.5 to 47.5)
		24	Testosterone	630	221.0 (43.7)	397.1 (254.3)	171.3	(156.5 to 186.2)	136.0	(115.1 to 156.9)
			Placebo	646	217.3 (47.0)	251.0 (113.0)	35.3	(20.6 to 50.0)
			Placebo	138	216.2 (47.5)	271.6 (146.7)	55.9	(29.1 to 82.7)
Dihydrotestosterone, ng/dL	LG-PDD	12	Testosterone	16	16.1 (7.1)	58.0 (40.6)	45.5	(31.5 to 59.4)	46.2	(26.7 to 65.7)	<.001
			Placebo	18	12.3 (4.9)	12.5 (6.7)	−0.7	(−13.9 to 12.4)
		36	Testosterone	11	12.6 (6.4)	42.3 (38.0)	25.5	(9.8 to 41.2)	26.7	(3.9 to 49.5)
			Placebo	10	11.7 (4.1)	12.5 (4.8)	−1.2	(−17.4 to 15.0)
	PHQ-9 > 4	12	Testosterone	939	16.3 (7.9)	59.3 (46.4)	42.9	(40.8 to 45.0)	40.6	(37.7 to 43.6)	<.001
			Placebo	979	15.7 (8.8)	17.9 (11.3)	2.2	(0.2 to 4.3)
		36	Testosterone	494	16.6 (8.2)	55.3 (52.4)	37.7	(34.6 to 40.9)	33.8	(29.3 to 38.3)
			Placebo	501	15.7 (9.0)	19.6 (14.0)	3.9	(0.8 to 7.1)
Estradiol, pg/mL	LG-PDD	12	Testosterone	14	21.5 (7.2)	33.2 (21.7)	11.9	(4.0 to 19.8)	13.1	(2.3 to 23.9)	.034
			Placebo	16	21.5 (9.0)	20.0 (8.2)	−1.2	(−8.6 to 6.3)
		36	Testosterone	9	18.0 (6.5)	25.0 (16.2)	5.3	(−2.8 to 13.3)	7.4	(−3.7 to 18.5)
			Placebo	9	22.1 (8.1)	19.4 (4.8)	−2.1	(−9.7 to 5.4)
	PHQ-9 > 4	12	Testosterone	930	21.1 (8.0)	29.5 (16.7)	8.4	(7.5 to 9.2)	8.2	(7.0 to 9.3)	<.001
			Placebo	960	21.1 (8.5)	21.3 (10.7)	0.2	(−0.6 to 1.0)
		36	Testosterone	490	21.4 (8.0)	30.3 (20.0)	8.8	(7.4 to 10.2)	6.4	(4.5 to 8.4)
			Placebo	501	21.0 (8.6)	23.1 (14.1)	2.4	(1.0 to 3.7)

Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	Change from baseline		Treatment difference		P
Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	LS mean	95% CI	LS mean	95% CI	P
Testosterone	LG-PDD	3	Testosterone	23	198.5 (45.8)	417.9 (224.3)	213.5	(149.2 to 277.7)	213.6	to	<.001
			Placebo	20	193.5 (59.2)	192.4 (85.2)	−0.2	(−69.3 to 68.9)		to
		6	Testosterone	22	201.6 (44.7)	370.1 (227.2)	170.6	(105.0 to 236.2)	177.3	(82.0 to 272.6)
			Placebo	20	200.0 (57.9)	191.5 (72.2)	−6.7	(−75.6 to 62.2)
		12	Testosterone	17	209.8 (42.3)	426.0 (298.9)	220.8	(146.5 to 295.2)	219.0	(116.4 to 321.7)
			Placebo	19	204.2 (56.2)	202.2 (89.6)	1.8	(−68.9 to 72.5)
		18	Testosterone	13	206.4 (44.6)	310.9 (155.8)	99.1	(14.4 to 183.9)	51.8	(−64.2 to 167.8)
			Placebo	15	212.4 (55.4)	255.9 (143.8)	47.4	(−31.8 to 126.5)
		24	Testosterone	11	204.6 (50.1)	301.5 (121.9)	91.7	(−0.4 to 183.8)	68.2	(−53.4 to 189.8)
			Placebo	15	208.6 (54.3)	226.2 (110.5)	23.5	(−55.6 to 102.5)
	PHQ-9 > 4	3	Testosterone	1204	219.9 (46.1)	412.0 (250.4)	193.2	(182.5 to 203.9)	176.0	(161.0 to 191.0)	<.001
			Placebo	1249	217.0 (48.8)	235.4 (103.5)	17.2	(6.7 to 27.7)
		6	Testosterone	1104	219.9 (46.0)	430.2 (309.6)	211.7	(198.4 to 225.1)	193.2	(174.5 to 211.9)
			Placebo	1148	217.4 (48.9)	236.5 (97.0)	18.5	(5.4 to 31.6)
		12	Testosterone	900	219.5 (46.1)	421.9 (263.0)	203.3	(189.7 to 216.8)	169.2	(150.2 to 188.1)
			Placebo	949	216.3 (48.4)	251.4 (139.6)	34.1	(20.9 to 47.3)
		18	Testosterone	764	220.5 (44.3)	436.9 (321.8)	210.9	(194.2 to 227.7)	180.0	(156.4 to 203.5)
			Placebo	788	216.8 (48.5)	244.7 (121.7)	31.0	(14.5 to 47.5)
		24	Testosterone	630	221.0 (43.7)	397.1 (254.3)	171.3	(156.5 to 186.2)	136.0	(115.1 to 156.9)
			Placebo	646	217.3 (47.0)	251.0 (113.0)	35.3	(20.6 to 50.0)
			Placebo	138	216.2 (47.5)	271.6 (146.7)	55.9	(29.1 to 82.7)
Dihydrotestosterone, ng/dL	LG-PDD	12	Testosterone	16	16.1 (7.1)	58.0 (40.6)	45.5	(31.5 to 59.4)	46.2	(26.7 to 65.7)	<.001
			Placebo	18	12.3 (4.9)	12.5 (6.7)	−0.7	(−13.9 to 12.4)
		36	Testosterone	11	12.6 (6.4)	42.3 (38.0)	25.5	(9.8 to 41.2)	26.7	(3.9 to 49.5)
			Placebo	10	11.7 (4.1)	12.5 (4.8)	−1.2	(−17.4 to 15.0)
	PHQ-9 > 4	12	Testosterone	939	16.3 (7.9)	59.3 (46.4)	42.9	(40.8 to 45.0)	40.6	(37.7 to 43.6)	<.001
			Placebo	979	15.7 (8.8)	17.9 (11.3)	2.2	(0.2 to 4.3)
		36	Testosterone	494	16.6 (8.2)	55.3 (52.4)	37.7	(34.6 to 40.9)	33.8	(29.3 to 38.3)
			Placebo	501	15.7 (9.0)	19.6 (14.0)	3.9	(0.8 to 7.1)
Estradiol, pg/mL	LG-PDD	12	Testosterone	14	21.5 (7.2)	33.2 (21.7)	11.9	(4.0 to 19.8)	13.1	(2.3 to 23.9)	.034
			Placebo	16	21.5 (9.0)	20.0 (8.2)	−1.2	(−8.6 to 6.3)
		36	Testosterone	9	18.0 (6.5)	25.0 (16.2)	5.3	(−2.8 to 13.3)	7.4	(−3.7 to 18.5)
			Placebo	9	22.1 (8.1)	19.4 (4.8)	−2.1	(−9.7 to 5.4)
	PHQ-9 > 4	12	Testosterone	930	21.1 (8.0)	29.5 (16.7)	8.4	(7.5 to 9.2)	8.2	(7.0 to 9.3)	<.001
			Placebo	960	21.1 (8.5)	21.3 (10.7)	0.2	(−0.6 to 1.0)
		36	Testosterone	490	21.4 (8.0)	30.3 (20.0)	8.8	(7.4 to 10.2)	6.4	(4.5 to 8.4)
			Placebo	501	21.0 (8.6)	23.1 (14.1)	2.4	(1.0 to 3.7)

To convert serum total testosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply testosterone concentration in nanograms per deciliter by 0.0347. To convert estradiol concentrations from picogram per milliliter to picomoles per liter, multiply estradiol concentrations in picogram per milliliter by 3.67. To convert dihydrotestosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply dihydrotestosterone concentrations in nanograms per deciliter by 0.0344.

Abbreviations: LG-PDD, low-grade persistent depressive disorder; PHQ-9, Patient Health Questionnaire-9.

Table 3.

Open in new tab

Changes from baseline in total testosterone, dihydrotestosterone, and estradiol levels in randomly assigned men with low-grade persistent depressive disorder or with Patient Health Questionnaire-9 score greater than 4

Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	Change from baseline		Treatment difference		P
Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	LS mean	95% CI	LS mean	95% CI	P
Testosterone	LG-PDD	3	Testosterone	23	198.5 (45.8)	417.9 (224.3)	213.5	(149.2 to 277.7)	213.6	to	<.001
			Placebo	20	193.5 (59.2)	192.4 (85.2)	−0.2	(−69.3 to 68.9)		to
		6	Testosterone	22	201.6 (44.7)	370.1 (227.2)	170.6	(105.0 to 236.2)	177.3	(82.0 to 272.6)
			Placebo	20	200.0 (57.9)	191.5 (72.2)	−6.7	(−75.6 to 62.2)
		12	Testosterone	17	209.8 (42.3)	426.0 (298.9)	220.8	(146.5 to 295.2)	219.0	(116.4 to 321.7)
			Placebo	19	204.2 (56.2)	202.2 (89.6)	1.8	(−68.9 to 72.5)
		18	Testosterone	13	206.4 (44.6)	310.9 (155.8)	99.1	(14.4 to 183.9)	51.8	(−64.2 to 167.8)
			Placebo	15	212.4 (55.4)	255.9 (143.8)	47.4	(−31.8 to 126.5)
		24	Testosterone	11	204.6 (50.1)	301.5 (121.9)	91.7	(−0.4 to 183.8)	68.2	(−53.4 to 189.8)
			Placebo	15	208.6 (54.3)	226.2 (110.5)	23.5	(−55.6 to 102.5)
	PHQ-9 > 4	3	Testosterone	1204	219.9 (46.1)	412.0 (250.4)	193.2	(182.5 to 203.9)	176.0	(161.0 to 191.0)	<.001
			Placebo	1249	217.0 (48.8)	235.4 (103.5)	17.2	(6.7 to 27.7)
		6	Testosterone	1104	219.9 (46.0)	430.2 (309.6)	211.7	(198.4 to 225.1)	193.2	(174.5 to 211.9)
			Placebo	1148	217.4 (48.9)	236.5 (97.0)	18.5	(5.4 to 31.6)
		12	Testosterone	900	219.5 (46.1)	421.9 (263.0)	203.3	(189.7 to 216.8)	169.2	(150.2 to 188.1)
			Placebo	949	216.3 (48.4)	251.4 (139.6)	34.1	(20.9 to 47.3)
		18	Testosterone	764	220.5 (44.3)	436.9 (321.8)	210.9	(194.2 to 227.7)	180.0	(156.4 to 203.5)
			Placebo	788	216.8 (48.5)	244.7 (121.7)	31.0	(14.5 to 47.5)
		24	Testosterone	630	221.0 (43.7)	397.1 (254.3)	171.3	(156.5 to 186.2)	136.0	(115.1 to 156.9)
			Placebo	646	217.3 (47.0)	251.0 (113.0)	35.3	(20.6 to 50.0)
			Placebo	138	216.2 (47.5)	271.6 (146.7)	55.9	(29.1 to 82.7)
Dihydrotestosterone, ng/dL	LG-PDD	12	Testosterone	16	16.1 (7.1)	58.0 (40.6)	45.5	(31.5 to 59.4)	46.2	(26.7 to 65.7)	<.001
			Placebo	18	12.3 (4.9)	12.5 (6.7)	−0.7	(−13.9 to 12.4)
		36	Testosterone	11	12.6 (6.4)	42.3 (38.0)	25.5	(9.8 to 41.2)	26.7	(3.9 to 49.5)
			Placebo	10	11.7 (4.1)	12.5 (4.8)	−1.2	(−17.4 to 15.0)
	PHQ-9 > 4	12	Testosterone	939	16.3 (7.9)	59.3 (46.4)	42.9	(40.8 to 45.0)	40.6	(37.7 to 43.6)	<.001
			Placebo	979	15.7 (8.8)	17.9 (11.3)	2.2	(0.2 to 4.3)
		36	Testosterone	494	16.6 (8.2)	55.3 (52.4)	37.7	(34.6 to 40.9)	33.8	(29.3 to 38.3)
			Placebo	501	15.7 (9.0)	19.6 (14.0)	3.9	(0.8 to 7.1)
Estradiol, pg/mL	LG-PDD	12	Testosterone	14	21.5 (7.2)	33.2 (21.7)	11.9	(4.0 to 19.8)	13.1	(2.3 to 23.9)	.034
			Placebo	16	21.5 (9.0)	20.0 (8.2)	−1.2	(−8.6 to 6.3)
		36	Testosterone	9	18.0 (6.5)	25.0 (16.2)	5.3	(−2.8 to 13.3)	7.4	(−3.7 to 18.5)
			Placebo	9	22.1 (8.1)	19.4 (4.8)	−2.1	(−9.7 to 5.4)
	PHQ-9 > 4	12	Testosterone	930	21.1 (8.0)	29.5 (16.7)	8.4	(7.5 to 9.2)	8.2	(7.0 to 9.3)	<.001
			Placebo	960	21.1 (8.5)	21.3 (10.7)	0.2	(−0.6 to 1.0)
		36	Testosterone	490	21.4 (8.0)	30.3 (20.0)	8.8	(7.4 to 10.2)	6.4	(4.5 to 8.4)
			Placebo	501	21.0 (8.6)	23.1 (14.1)	2.4	(1.0 to 3.7)

Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	Change from baseline		Treatment difference		P
Test	Population	Mo	Study arm	No.	Baseline Mean (SD)	Visit Mean (SD)	LS mean	95% CI	LS mean	95% CI	P
Testosterone	LG-PDD	3	Testosterone	23	198.5 (45.8)	417.9 (224.3)	213.5	(149.2 to 277.7)	213.6	to	<.001
			Placebo	20	193.5 (59.2)	192.4 (85.2)	−0.2	(−69.3 to 68.9)		to
		6	Testosterone	22	201.6 (44.7)	370.1 (227.2)	170.6	(105.0 to 236.2)	177.3	(82.0 to 272.6)
			Placebo	20	200.0 (57.9)	191.5 (72.2)	−6.7	(−75.6 to 62.2)
		12	Testosterone	17	209.8 (42.3)	426.0 (298.9)	220.8	(146.5 to 295.2)	219.0	(116.4 to 321.7)
			Placebo	19	204.2 (56.2)	202.2 (89.6)	1.8	(−68.9 to 72.5)
		18	Testosterone	13	206.4 (44.6)	310.9 (155.8)	99.1	(14.4 to 183.9)	51.8	(−64.2 to 167.8)
			Placebo	15	212.4 (55.4)	255.9 (143.8)	47.4	(−31.8 to 126.5)
		24	Testosterone	11	204.6 (50.1)	301.5 (121.9)	91.7	(−0.4 to 183.8)	68.2	(−53.4 to 189.8)
			Placebo	15	208.6 (54.3)	226.2 (110.5)	23.5	(−55.6 to 102.5)
	PHQ-9 > 4	3	Testosterone	1204	219.9 (46.1)	412.0 (250.4)	193.2	(182.5 to 203.9)	176.0	(161.0 to 191.0)	<.001
			Placebo	1249	217.0 (48.8)	235.4 (103.5)	17.2	(6.7 to 27.7)
		6	Testosterone	1104	219.9 (46.0)	430.2 (309.6)	211.7	(198.4 to 225.1)	193.2	(174.5 to 211.9)
			Placebo	1148	217.4 (48.9)	236.5 (97.0)	18.5	(5.4 to 31.6)
		12	Testosterone	900	219.5 (46.1)	421.9 (263.0)	203.3	(189.7 to 216.8)	169.2	(150.2 to 188.1)
			Placebo	949	216.3 (48.4)	251.4 (139.6)	34.1	(20.9 to 47.3)
		18	Testosterone	764	220.5 (44.3)	436.9 (321.8)	210.9	(194.2 to 227.7)	180.0	(156.4 to 203.5)
			Placebo	788	216.8 (48.5)	244.7 (121.7)	31.0	(14.5 to 47.5)
		24	Testosterone	630	221.0 (43.7)	397.1 (254.3)	171.3	(156.5 to 186.2)	136.0	(115.1 to 156.9)
			Placebo	646	217.3 (47.0)	251.0 (113.0)	35.3	(20.6 to 50.0)
			Placebo	138	216.2 (47.5)	271.6 (146.7)	55.9	(29.1 to 82.7)
Dihydrotestosterone, ng/dL	LG-PDD	12	Testosterone	16	16.1 (7.1)	58.0 (40.6)	45.5	(31.5 to 59.4)	46.2	(26.7 to 65.7)	<.001
			Placebo	18	12.3 (4.9)	12.5 (6.7)	−0.7	(−13.9 to 12.4)
		36	Testosterone	11	12.6 (6.4)	42.3 (38.0)	25.5	(9.8 to 41.2)	26.7	(3.9 to 49.5)
			Placebo	10	11.7 (4.1)	12.5 (4.8)	−1.2	(−17.4 to 15.0)
	PHQ-9 > 4	12	Testosterone	939	16.3 (7.9)	59.3 (46.4)	42.9	(40.8 to 45.0)	40.6	(37.7 to 43.6)	<.001
			Placebo	979	15.7 (8.8)	17.9 (11.3)	2.2	(0.2 to 4.3)
		36	Testosterone	494	16.6 (8.2)	55.3 (52.4)	37.7	(34.6 to 40.9)	33.8	(29.3 to 38.3)
			Placebo	501	15.7 (9.0)	19.6 (14.0)	3.9	(0.8 to 7.1)
Estradiol, pg/mL	LG-PDD	12	Testosterone	14	21.5 (7.2)	33.2 (21.7)	11.9	(4.0 to 19.8)	13.1	(2.3 to 23.9)	.034
			Placebo	16	21.5 (9.0)	20.0 (8.2)	−1.2	(−8.6 to 6.3)
		36	Testosterone	9	18.0 (6.5)	25.0 (16.2)	5.3	(−2.8 to 13.3)	7.4	(−3.7 to 18.5)
			Placebo	9	22.1 (8.1)	19.4 (4.8)	−2.1	(−9.7 to 5.4)
	PHQ-9 > 4	12	Testosterone	930	21.1 (8.0)	29.5 (16.7)	8.4	(7.5 to 9.2)	8.2	(7.0 to 9.3)	<.001
			Placebo	960	21.1 (8.5)	21.3 (10.7)	0.2	(−0.6 to 1.0)
		36	Testosterone	490	21.4 (8.0)	30.3 (20.0)	8.8	(7.4 to 10.2)	6.4	(4.5 to 8.4)
			Placebo	501	21.0 (8.6)	23.1 (14.1)	2.4	(1.0 to 3.7)

To convert serum total testosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply testosterone concentration in nanograms per deciliter by 0.0347. To convert estradiol concentrations from picogram per milliliter to picomoles per liter, multiply estradiol concentrations in picogram per milliliter by 3.67. To convert dihydrotestosterone concentrations in nanograms per deciliter to nanomoles per liter, multiply dihydrotestosterone concentrations in nanograms per deciliter by 0.0344.

Abbreviations: LG-PDD, low-grade persistent depressive disorder; PHQ-9, Patient Health Questionnaire-9.

Discussion

A majority of middle-aged and older men with hypogonadism enrolled in the TRAVERSE Trial (50.8%) reported significant depressive symptoms (PHQ-9 score > 4) but surprisingly, only 1.5% of those consenting to the substudy met the rigorous criteria for LG-PDD, substantially fewer than the prevalence reported in previous studies (19, 24-26). The prevalence of depressive symptoms in the enrolled TRAVERSE participants was high and similar to that reported in the TRIUS Registry study of men with hypogonadism (1). A majority of men with depressive symptoms had low to moderate depressive symptoms, and a substantial fraction did not have persistent symptoms. However, nearly 9.9% had moderately severe or severe depressive symptoms (PHQ-9 score ≥15, and GDS-15 score > 9) and 15% self-reported a history of MDDs. These findings suggest that men diagnosed with hypogonadism should be evaluated for depressive symptoms and depressive disorders because depression adversely affects quality of life, ability to function in daily life, and increases the risk of substance use and self-harm.

Because the number of consenting participants meeting the rigorous definition of LG-PDD (1.5%) was substantially lower than we had projected, the LG-PDD substudy did not have sufficient power to detect meaningful treatment effects. However, in a large sample of participants with significant depressive symptoms (PHQ-9 score >4), TRT was associated with a small but significant improvement in mood and energy, consistent with the findings of some other studies (8, 27). Small but significantly greater improvements in mood and energy were also observed in all randomly assigned, testosterone-treated participants compared to placebo-treated men regardless of the presence of depressive symptoms. It is notable that the depressive symptoms did not improve in men who had severe or moderately severe depressive symptoms, and post hoc analysis suggested that the depressive symptoms improved in men with mild to moderate depressive symptoms at baseline. There was a substantial placebo effect on the mood and energy scores. The clinical significance of the small but statistically significant improvements in mood and energy levels needs further investigation; however, the changes in energy were positively associated with changes in mood level.

The present study has both strengths and limitations. In contrast to prior observational studies and randomized trials that recruited heterogeneous populations of participants and did not distinguish between depressive symptoms, such as dysphoric mood and anhedonia, and clinical depressive disorders, such as MDDs and LG-PDD (8, 27), the present study implemented a standardized, rigorous protocol for defining LG-PDD. The participants met the Endocrine Society's criteria for hypogonadism, including 1 or more hypogonadal symptoms and 2 fasting, early-morning testosterone levels of less than 300 ng/dL, measured using LC-MS/MS in a laboratory certified by the Hormone Standardization Program for Testosterone of the Centers for Disease Control and Prevention. The trial had the attributes of good trial design: parallel groups; masked, stratified, randomized allocation of participants; placebo control; and blinding of participants as well as outcome assessors. Participants’ mood and energy were evaluated using HIS-Q, a validated, psychometrically robust instrument designed specifically for men with hypogonadism (23). The sample size for secondary aims (men with significant depressive symptoms and all randomly assigned participants with hypogonadism) is greater and the follow-up duration (2 years) longer than any previous trial.

Because very few participants met the rigorous definition of LG-PDD, the trial lacked sufficient statistical power to determine the efficacy of TRT in improving depressive symptoms in LG-PDD. The trial took place during the COVID-19 pandemic, which affected retention. As reported previously (16), the nonretention rates were high though similar in the 2 intervention groups and comparable to those reported in trials in other symptomatic conditions (28, 29) or in men receiving TRT in clinical settings (30); however, nonadherence would tend to bias the results toward null. These data apply only to middle-aged and older men with hypogonadism and should not be extrapolated to women, transgender, and gender-diverse people, or to supraphysiologic doses of testosterone. The cause of hypogonadism was not determined; therefore, we do not know what proportion of enrolled men with hypogonadism had organic hypogonadism due to known causes. Because the trial enrolled men with hypogonadism with preexisting CVD or with 3 or more risk factors for CVD, the enrolled participants had higher rates of diabetes and other cardiovascular risk factors than in the general population, not dissimilar from the men with hypogonadism receiving TRT in large health-care systems in the United States (31). The on-treatment serum testosterone levels were lower than in some other trials but the mean increment in serum testosterone levels above baseline was not dissimilar from that expected from an average dose of 61-mg transdermal testosterone daily (32).

The trial's findings have clinical implications. Middle-aged and older men with hypogonadism should be evaluated for depressive symptoms because of the high prevalence of depressive symptoms and disorders in this population. However, TRT does not appear to represent an effective treatment option for most men with clinical depressive disorders. The trial's findings support the professional society guidelines on the treatment of male hypogonadism with respect to the effects of TRT on mood (2).

Conclusion

Depressive symptoms are common in middle-aged and older men with hypogonadism, but the prevalence of rigorously defined LG-PDD is low in these men. TRT modestly improves mood and energy in men with hypogonadism with and without depressive symptoms but the treatment effect is small.

Acknowledgments

The authors thank the members of the data monitoring committee, which included the following members: John H. Alexander, MD, MHSc (chairman), Duke Clinical Research Institute, Duke University, Durham, NC; William Bremner, MD, PhD, University of Washington, Seattle, WA; Eric Klein, MD, Cleveland Clinic, Cleveland, OH; Darren K. McGuire, MD, MHSc, University of Texas Southwestern Medical Center, Dallas, TX; Janet Wittes, PhD, Wittes LLC, Washington, DC; Renato D. Lopes, MD, PhD (observer, nonvoting), Duke Clinical Research Institute, Durham, NC; Andrew Armstrong, MD, ScM (ad hoc consult), Duke University Medical Center, Duke Cancer Institute Center for Prostate and Urologic Cancers, Durham, NC.

Funding

This work was supported by a consortium of testosterone manufacturers led by AbbVie, Inc (North Chicago, IL) with additional financial support provided by Endo Pharmaceuticals (Malvern, PA), Acerus Pharmaceuticals Corporation (Ontario, Canada), and Upsher-Smith Laboratories, LLC (Maple Grove, MN). S.B. was supported in part by the Boston Claude D. Pepper Older Americans Independence Center (5P30AG31679).

Author Contributions

S.B., K.W., K.M.P., and T.G.T. had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

The design and analysis plan for the TRAVERSE Depression Substudy were crafted by the TRAVERSE Depression Substudy Committee led by the Research Program in Men's Health at the Brigham and Women's Hospital, Boston, Massachusetts. The funder reviewed and made suggestions to the Depression Substudy design and statistical analysis plan, and oversaw the conduct, data collection, and management of the trial. The data analysis and interpretation of the data were performed by the statisticians associated with the Depression Substudy Committee (K.A.W., K.M.P., T.G.T.). The first author (S.B.) wrote the first manuscript draft and all coauthors reviewed and approved the manuscript. The decision to submit the manuscript for publication was made by the academic leadership of the TRAVERSE Depression Substudy Committee. The funder reviewed the manuscript, made suggestions on the content, and approved the final version of the manuscript.

Disclosures

S.B. reports receiving research grants paid to and managed by his institution from the National Institute on Aging, the National Institute of Child Health and human Development—National Center for Rehabilitation Research, PCORI, AbbVie, Metro International Biotech, FPT, and Transition therapeutics; receiving consulting fees from Novartis, Versanis, and Besins Healthcare; and equity interest in FPT and Xyone. These conflicts are overseen by the Massachusetts General Brigham's Office of Industry Interaction and managed in accordance with institutional policies. A.M.L. reports receiving research grants from AbbVie, AstraZeneca, CSL Behring, Eli Lilly, Esperion, and Novartis; and consulting fees from Akebia, Ardelyx, Becton-Dickson, Eli Lilly, Endologix, Fibrogen, Glaxo-SmithKline, Medtronic, Neovasc, Novo Nordisk, Provention Bio, and ReCor. S.E.N. reports receiving grants to perform clinical trials from Abbvie, AstraZeneca, Amgen, Bristol Myers Squibb, Eli Lilly, Esperion Therapeutics Inc, Medtronic, MyoKardia, New Amsterdam Pharmaceuticals, Novartis, and Silence Therapeutics. P.F., M.G.M., and X.L. were employees and stockholders of AbbVie, Inc. The other authors report no conflicts of interest.

Data Availability

Original data generated and analyzed during this study are included in this article.

Current Affiliation

P.F. is currently an employee of Eli Lilly and Company, Indianapolis, Indiana.

Clinical Trial Information

ClinicalTrials.gov identifier: NCT03518034 (registered May 8, 2018).

References

1

Khera

M

,

Bhattacharya

RK

,

Blick

G

,

Kushner

H

,

Nguyen

D

,

Miner

MM

.

The effect of testosterone supplementation on depression symptoms in hypogonadal men from the testim registry in the US (TRiUS)

.

Aging Male

.

2012

;

15

(

1

):

14

‐

21

.

2

Bhasin

S

,

Brito

JP

,

Cunningham

GR

, et al.

Testosterone therapy in men with hypogonadism: an endocrine society clinical practice guideline

.

J Clin Endocrinol Metab

.

2018

;

103

(

5

):

1715

‐

1744

.

3

Wu

FC

,

Tajar

A

,

Beynon

JM

, et al.

Identification of late-onset hypogonadism in middle-aged and elderly men

.

N Engl J Med

.

2010

;

363

(

2

):

123

‐

135

.

4

Schweiger

U

,

Deuschle

M

,

Weber

B

, et al.

Testosterone, gonadotropin, and cortisol secretion in male patients with major depression

.

Psychosom Med

.

1999

;

61

(

3

):

292

‐

296

.

5

Shores

MM

,

Sloan

KL

,

Matsumoto

AM

,

Moceri

VM

,

Felker

B

,

Kivlahan

DR

.

Increased incidence of diagnosed depressive illness in hypogonadal older men

.

Arch Gen Psychiatry

.

2004

;

61

(

2

):

162

‐

167

.

6

Seidman

SN

,

Araujo

AB

,

Roose

SP

, et al.

Low testosterone levels in elderly men with dysthymic disorder

.

Am J Psychiatry

.

2002

;

159

(

3

):

456

‐

459

.

7

Altschule

MD

,

Tillotson

KJ

.

The use of testosterone in the treatment of depressions

.

N Engl J Med

.

1948

;

239

(

27

):

1036

‐

1038

.

8

Snyder

PJ

,

Bhasin

S

,

Cunningham

GR

, et al.

Effects of testosterone treatment in older men

.

N Engl J Med

.

2016

;

374

(

7

):

611

‐

624

.

9

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

(

1

):

31

‐

40

.

10

Seidman

SN

,

Spatz

E

,

Rizzo

C

,

Roose

SP

.

Testosterone replacement therapy for hypogonadal men with major depressive disorder: a randomized, placebo-controlled clinical trial

.

J Clin Psychiatry

.

2001

;

62

(

6

):

406

‐

412

.

11

Pope

HG

Jr.,

Amiaz

R

,

Brennan

BP

, et al.

Parallel-group placebo-controlled trial of testosterone gel in men with major depressive disorder displaying an incomplete response to standard antidepressant treatment

.

J Clin Psychopharmacol

.

2010

;

30

(

2

):

126

‐

134

.

12

Orengo

CA

,

Fullerton

L

,

Kunik

ME

.

Safety and efficacy of testosterone gel 1% augmentation in depressed men with partial response to antidepressant therapy

.

J Geriatr Psychiatry Neurol

.

2005

;

18

(

1

):

20

‐

24

.

13

Shores

MM

,

Kivlahan

DR

,

Sadak

TI

,

Li

EJ

,

Matsumoto

AM

.

A randomized, double-blind, placebo-controlled study of testosterone treatment in hypogonadal older men with subthreshold depression (dysthymia or minor depression)

.

J Clin Psychiatry

.

2009

;

70

(

7

):

1009

‐

1016

.

14

Seidman

SN

,

Orr

G

,

Raviv

G

, et al.

Effects of testosterone replacement in middle-aged men with dysthymia: a randomized, placebo-controlled clinical trial

.

J Clin Psychopharmacol

.

2009

;

29

(

3

):

216

‐

221

.

15

Bhasin

S

,

Lincoff

AM

,

Basaria

S

, et al.

Effects of long-term testosterone treatment on cardiovascular outcomes in men with hypogonadism: rationale and design of the TRAVERSE study

.

Am Heart J

.

2022

;

245

:

41

‐

50

.

16

Lincoff

AM

,

Bhasin

S

,

Flevaris

P

, et al.

Cardiovascular safety of testosterone-replacement therapy

.

N Engl J Med

.

2023

;

389

(

2

):

107

‐

117

.

17

Uher

R

,

Payne

JL

,

Pavlova

B

,

Perlis

RH

.

Major depressive disorder in DSM-5: implications for clinical practice and research of changes from DSM-IV

.

Depress Anxiety

.

2014

;

31

(

6

):

459

‐

471

.

18

Wakefield

JC

.

DSM-5: proposed changes to depressive disorders

.

Curr Med Res Opin

.

2012

;

28

(

3

):

335

‐

343

.

19

Beekman

AT

,

Deeg

DJ

,

Smit

JH

, et al.

Dysthymia in later life: a study in the community

.

J Affect Disord

.

2004

;

81

(

3

):

191

‐

199

.

20

Kroenke

K

,

Spitzer

RL

,

Williams

JB

.

The PHQ-9: validity of a brief depression severity measure

.

J Gen Intern Med

.

2001

;

16

(

9

):

606

‐

613

.

21

Costantini

L

,

Pasquarella

C

,

Odone

A

, et al.

Screening for depression in primary care with patient health questionnaire-9 (PHQ-9): a systematic review

.

J Affect Disord

.

2021

;

279

:

473

‐

483

.

22

Montorio

I

,

Izal

M

.

The geriatric depression scale: a review of its development and utility

.

Int Psychogeriatr

.

1996

;

8

(

1

):

103

‐

112

.

23

Gelhorn

HL

,

Bodhani

AR

,

Wahala

LS

, et al.

Development of the hypogonadism impact of symptoms questionnaire short form: qualitative research

.

J Sex Med

.

2016

;

13

(

11

):

1729

‐

1736

.

24

Devanand

DP

.

Dysthymic disorder in the elderly population

.

Int Psychogeriatr

.

2014

;

26

(

1

):

39

‐

48

.

25

Murphy

JA

,

Byrne

GJ

.

Prevalence and correlates of the proposed DSM-5 diagnosis of chronic depressive disorder

.

J Affect Disord

.

2012

;

139

(

2

):

172

‐

180

.

26

Steffens

DC

,

Skoog

I

,

Norton

MC

, et al.

Prevalence of depression and its treatment in an elderly population: the cache county study

.

Arch Gen Psychiatry

.

2000

;

57

(

6

):

601

‐

607

.

27

Giltay

EJ

,

Tishova

YA

,

Mskhalaya

GJ

,

Gooren

LJ

,

Saad

F

,

Kalinchenko

SY

.

Effects of testosterone supplementation on depressive symptoms and sexual dysfunction in hypogonadal men with the metabolic syndrome

.

J Sex Med

.

2010

;

7

(

7

):

2572

‐

2582

.

28

Nissen

SE

,

Yeomans

ND

,

Solomon

DH

, et al.

Cardiovascular safety of celecoxib, naproxen, or ibuprofen for arthritis

.

N Engl J Med

.

2016

;

375

(

26

):

2519

‐

2529

.

29

Rossouw

JE

,

Anderson

GL

,

Prentice

RL

, et al.

Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women's health initiative randomized controlled trial

.

JAMA

.

2002

;

288

(

3

):

321

‐

333

.

Google Scholar

PubMed

OpenURL Placeholder Text

WorldCat

30

Schoenfeld

MJ

,

Shortridge

E

,

Cui

Z

,

Muram

D

.

Medication adherence and treatment patterns for hypogonadal patients treated with topical testosterone therapy: a retrospective medical claims analysis

.

J Sex Med

.

2013

;

10

(

5

):

1401

‐

1409

.

31

Jasuja

GK

,

Bhasin

S

,

Reisman

JI

, et al.

Who gets testosterone? Patient characteristics associated with testosterone prescribing in the veteran affairs system: a cross-sectional study

.

J Gen Intern Med

.

2017

;

32

(

3

):

304

‐

311

.

32

Bhasin

S

,

Travison

TG

,

O'Brien

L

, et al.

Contributors to the substantial variation in on-treatment testosterone levels in men receiving transdermal testosterone gels in randomized trials

.

Andrology

.

2018

;

6

(

1

):

151

‐

157

.

Abbreviations

CV
coefficient of variation

CVD
cardiovascular disease

DHT
dihydrotestosterone

FAS
full analysis set

GDS-15
Geriatric Depression Scale

HIS-Q
Hypogonadism Impact of Symptoms Questionnaire

LC-MS/MS
liquid chromatography–tandem mass spectrometry

LG-PDD
low-grade persistent depressive disorder

LLOQ
lower limit of quantitation

MDDs
major depressive disorders

PHQ-9
Patient Health Questionnaire-9

TRAVERSE
Testosterone Replacement therapy for Assessment of long-term Vascular Events and efficacy ResponSE in hypogonadal men

TRIUS
Testim Registry In the US

TRT
testosterone-replacement therapy

This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/pages/standard-publication-reuse-rights)

Download all slides

Month:	Total Views:
January 2024	111
February 2024	148
March 2024	89
April 2024	80
May 2024	73
June 2024	167
July 2024	202
August 2024	69
September 2024	93
October 2024	71
November 2024	50
December 2024	102
January 2025	189
February 2025	178
March 2025	264
April 2025	190

Article Contents

Depressive Syndromes in Men With Hypogonadism in the TRAVERSE Trial: Response to Testosterone-Replacement Therapy

Abstract

Materials and Methods

Eligibility Criteria

Participant Allocation

Study Outcomes

Statistical Analyses

Results

Study Population

Effects of Testosterone-Replacement Therapy on Outcomes in Men With Low-Grade Persistent Depressive Disorder

Effects of Testosterone-Replacement Therapy in All Randomly Assigned Men With Significant Depressive Symptoms (Patient Health Questionnaire-9 Score > 4)

Effects of Testosterone-Replacement Therapy in All Randomly Assigned Participants

Hormone Levels

Discussion

Conclusion

Acknowledgments

Funding

Author Contributions

Disclosures

Data Availability

Current Affiliation

Clinical Trial Information

References

Abbreviations

Citations

Views

Altmetric

Email alerts

See also

Commentary

Citing articles via

Latest

Most Read

Most Cited

Article Contents

Depressive Syndromes in Men With Hypogonadism in the TRAVERSE Trial: Response to Testosterone-Replacement Therapy

Abstract

Materials and Methods

Eligibility Criteria

Participant Allocation

Study Outcomes

Statistical Analyses

Results

Study Population

Effects of Testosterone-Replacement Therapy on Outcomes in Men With Low-Grade Persistent Depressive Disorder

Effects of Testosterone-Replacement Therapy in All Randomly Assigned Men With Significant Depressive Symptoms (Patient Health Questionnaire-9 Score > 4)

Effects of Testosterone-Replacement Therapy in All Randomly Assigned Participants

Hormone Levels

Discussion

Conclusion

Acknowledgments

Funding

Author Contributions

Disclosures

Data Availability

Current Affiliation

Clinical Trial Information

References

Abbreviations

Citations

Views

Altmetric

Email alerts

See also

Commentary

Citing articles via

Latest

Most Read

Most Cited

This Feature Is Available To Subscribers Only