🧬 Aging Wiki

testosterone-replacement-therapy

Properties1
aliasesTRT, androgen replacement therapy, testosterone therapy, ART

16 min read

Testosterone Replacement Therapy

Testosterone replacement therapy (TRT) is the administration of exogenous testosterone — or its ester prodrugs — to restore physiological androgen levels in men with documented testosterone deficiency (hypogonadism). TRT for classical hypogonadism (primary or secondary, with clear pituitary/gonadal pathology) is well-evidenced and FDA-approved. The contested space is its use in age-related testosterone decline (“andropause,” low-normal T, “Low T”) in older men without classical disease — where the indication is weaker, benefits more modest, and cardiovascular/thromboembolic signals require careful consideration.

This page covers the therapy class: indications, formulations, major trials, and aging-context risk-benefit synthesis. Detailed compound pharmacology (PK, mechanism, aromatization) lives on testosterone.

A neutral framing is essential: TRT is neither a broadly “pro-aging” intervention (it clearly benefits men with genuine hypogonadism) nor a validated geroprotector (no trial has demonstrated lifespan extension or slowing of biological-aging clocks in the broader population).

Indications and the Diagnostic Threshold Debate

Classical (pathological) hypogonadism

Classical hypogonadism — primary (testicular failure: Klinefelter syndrome, orchitis, radiation) or secondary (pituitary/hypothalamic: Kallmann syndrome, hyperprolactinemia, hemochromatosis) — carries unambiguous indications for TRT. In these men, testosterone is physiologically absent or severely deficient, and restoration produces clear benefits in sexual function, bone mineral density, body composition, mood, and erythropoiesis. The FDA-approved indication is confined to this category.

Serum total testosterone declines approximately 1–2% per year after age 30, with free (bioavailable) testosterone declining faster as sex hormone-binding globulin (SHBG) rises with age 1. By age 70–79, roughly 20–30% of men have total testosterone below 300 ng/dL, the most commonly used clinical threshold. This decline — sometimes called andropause or late-onset hypogonadism — is not equivalent to classical hypogonadism. Most older men with low-normal testosterone do not have identifiable pituitary or gonadal pathology; instead, they have blunted hypothalamic GnRH pulsatility, reduced Leydig cell responsiveness, and altered sleep architecture. See leydig-cells and lh for cell-biology detail.

The diagnostic threshold controversy: there is no single validated lower limit of normal testosterone that defines clinical hypogonadism in older men. The American Urological Association (AUA) uses <300 ng/dL total T; the Endocrine Society uses <264–300 ng/dL (with confirmatory repeat measurement); European guidelines often use <11 nmol/L (~317 ng/dL). Free testosterone thresholds vary by assay. The key clinical issue is symptom burden in context of biochemical deficiency — symptoms (low libido, fatigue, erectile dysfunction, reduced vitality) are nonspecific and overlap extensively with depression, sleep apnea, obesity, and other comorbidities. The T-Trials consortium explicitly required a total testosterone <275 ng/dL plus at least one cardinal symptom for enrollment 2.

The “anti-aging clinic” / direct-to-consumer Low T market prescribes TRT broadly to men in the 300–500 ng/dL range based on symptoms alone, without excluding other causes. This practice is outside guideline-concordant care, lacks strong evidence, and is the contested space where the risk-benefit calculation is most uncertain.

Formulations

FormulationRouteFrequencyPK profileNotes
Testosterone cypionate / enanthateIM injectionEvery 1–2 weeksPeaks 2–5 days, trough before next dose; supraphysiologic peaks commonMost common in US; inexpensive; hematocrit monitoring important
Testosterone undecanoate (Aveed)IM injectionEvery 10–14 weeksSteady pharmacokinetics; no supraphysiologic peaksFDA-approved; requires 30-min post-injection observation (POME risk)
Testosterone gel / transdermal (AndroGel, Testim, Fortesta)TopicalDailyMimics diurnal variation; transfer risk to partners/childrenMost commonly prescribed in US; steady-state T achieved
Testosterone patches (Androderm)TransdermalDailyGood PK, skin irritation commonLess used; higher adherence difficulty
Testosterone pellets (Testopel)SubcutaneousEvery 3–6 monthsVery stable; minor surgical procedureGrowing in DTC clinics; no dose titration possible mid-cycle
Testosterone nasal gel (Natesto)Intranasal3x/dayEpisodic absorption; faster return of spermatogenesisMay preserve fertility better than other routes

TRAVERSE enrolled predominantly gel-treated men; formulation choice was at investigator discretion. T-Trials used a topical testosterone gel (1%) standardized across sites 2.

Key Trials

T-Trials (2016–2018): Coordinated efficacy evidence

The Testosterone Trials (T-Trials) were seven parallel, placebo-controlled, double-blind RCTs conducted across 12 US sites in 790 men aged ≥65 years with total testosterone <275 ng/dL and at least one predefined symptom. All men received testosterone gel 1% or placebo for 12 months. The coordinated design generated evidence across multiple domains simultaneously 2.

Primary results (Snyder 2016 NEJM): Testosterone treatment produced meaningful improvements in sexual function (sexual desire, sexual activity) compared to placebo. Physical performance (walking distance, grip strength) and vitality improvements did not reach statistical significance in the primary analysis, though subgroups with mobility limitation showed modest functional gains. No cognitive benefit was observed in the cognitive trial 2.

T-Trials domain-specific findings (verified claims only):

DomainResultCaveat
Sexual function / libidoStatistically significant improvementSmall-to-moderate effect size; many men require PDE5i regardless
Physical function (6MWT)Modest improvement in walking distanceNot statistically significant in Physical Function Trial alone (P=0.20; OR 1.42); statistically significant when all T-Trials participants combined (OR 1.77; P=0.003); effect sizes small (0.06–0.15)
Vitality / energyModest, not statistically significant
Bone mineral densityIncreased vertebral trabecular BMD (HR-pQCT)Driven substantially by aromatization to estradiol (see below)
Anemia (unexplained)Hemoglobin increase; resolution of mild anemiaClinically meaningful in the anemia sub-trial
Cognitive functionNo significant improvement

Coronary plaque sub-study (Budoff 2017): In the T-Trials cardiovascular imaging sub-study (n=138, men with baseline plaque by coronary CT), testosterone treatment was associated with a greater increase in coronary artery plaque volume compared to placebo over 12 months 3. This was not a MACE endpoint trial — plaque volume is a surrogate — but the signal raised cardiovascular concern and motivated TRAVERSE. needs-replication in larger MACE-powered study (done → TRAVERSE)

TRAVERSE (2023): Cardiovascular safety at scale

The TRAVERSE trial was a Phase 4, double-blind, placebo-controlled, event-driven non-inferiority RCT designed by the FDA to resolve the cardiovascular safety question. n=5,246 men aged 45–80 years with hypogonadism (total T <300 ng/dL on two measurements), pre-existing cardiovascular disease or elevated CV risk, randomized 1:1 to testosterone 1.62% gel or placebo, median follow-up 33 months 4.

Primary endpoint (MACE: cardiovascular death, non-fatal MI, non-fatal stroke):

  • Testosterone: 7.0% vs Placebo: 7.3%; HR 0.96 (95% CI 0.78–1.17)
  • Non-inferiority margin p<0.001 — testosterone was non-inferior to placebo for MACE

Secondary / safety signals (concerning):

OutcomeFinding
Atrial fibrillationHigher in testosterone arm (statistically significant; specific HR in paper body — abstract confirms significance)
Pulmonary embolismHigher in testosterone arm (statistically significant; specific HR in paper body — abstract confirms significance)
Acute kidney injuryHigher in testosterone arm
Sexual functionImproved
Anemia (erythrocytosis)Higher hematocrit / polycythemia rates in testosterone arm

The atrial fibrillation and pulmonary embolism signals are class effects requiring monitoring and are not explained away by the MACE non-inferiority result. The MACE non-inferiority is reassuring for the primary MACE composite but does not exclude meaningful individual harms. A separate sub-analysis of the TRAVERSE arrhythmia and thromboembolic signals confirmed statistical significance 5.

Hudson 2023 meta-analysis (Lancet Healthy Longevity)

A systematic review with individual participant data (IPD) meta-analysis — 17 studies, n=3,431 via IPD; 35 studies, n=5,601 in the combined two-stage aggregate analysis — examined symptom subgroup benefit in men with total T <12 nmol/L (350 ng/dL) 6. Testosterone produced statistically significant improvement in sexual function: IIEF-15 erectile function subscore mean difference 2.14 (95% CI 1.40–2.89), which meets the minimal clinically important difference for mild erectile dysfunction. AMS quality-of-life score was significantly improved; Beck Depression Inventory (BDI) did not improve significantly, and no significant association between testosterone benefit and age, baseline T level, diabetes, or BMI was found for IIEF-15 score. Critically, men with lower baseline IIEF-15 score had lower absolute function during TRT, meaning older men and those with obesity are less likely to reach adequate sexual function despite statistically similar increments. The analysis was restricted to men with total T <12 nmol/L — conclusions do not extrapolate to men with higher baseline testosterone.

2025 meta-analysis: Long-term cardiovascular safety

A meta-analysis of 23 RCTs, n=9,280 men with hypogonadism or low-to-low-normal testosterone (Braga et al. 2025) confirmed no statistically significant difference in all-cause mortality (RR 0.85; 95% CI 0.60–1.19), cardiovascular mortality (RR 0.85; 95% CI 0.65–1.12), stroke (RR 1.00; 95% CI 0.67–1.50), or myocardial infarction (RR 0.94; 95% CI 0.69–1.28) between TRT and placebo. Cardiac arrhythmias were significantly elevated in the testosterone arm (RR 1.53; 95% CI 1.20–1.97; p<0.01), consistent with TRAVERSE 7.

Bhasin 2026 (JAMA Internal Medicine): TRT in prostate cancer survivors

A 2026 phase 2 RCT (PMID 42113507, Bhasin et al.; n=136, 68 per arm; 125 completers) evaluated testosterone cypionate 100 mg IM weekly for 12 weeks in hypogonadal men with low-grade prostate cancer (Gleason 6 or Gleason 7 [3+4]) who had undetectable PSA for ≥2 years post-radical prostatectomy and total T <275 ng/dL. The primary efficacy endpoint was sexual activity (not biochemical recurrence — that was the pre-specified safety endpoint). TRT significantly increased sexual activity vs placebo (between-group difference 0.91 daily events; 95% CI 0.56–1.26; P<0.001) and improved sexual desire, body composition, loaded stair-climbing power, and VO2 peak. No participant in either group experienced biochemical recurrence over the 12-week treatment period. The trial was neither large enough nor long enough to evaluate clinical recurrence or long-term safety; results do not apply to men with high-grade prostate cancer or those treated with androgen deprivation therapy or radiation 8. needs-replication — a larger, longer confirmatory trial is needed.

Aging-Relevant Effects

Muscle mass and physical function

Testosterone increases lean body mass and muscle cross-sectional area via AR-mediated mTORC1 activation, satellite cell proliferation, and suppression of myostatin signaling. Effect sizes are real but modest at physiological replacement doses in older men: meta-analyses report ~1–3 kg increase in lean mass over 6–12 months. Functional strength and physical performance gains are smaller and less consistent than lean mass changes — the phenotypic gap between anabolism and functional benefit is important for aging-relevant claims. See sarcopenia for the muscle aging context and frailty for functional independence framing.

Bone mineral density

T-Trials demonstrated increased vertebral trabecular bone mineral density with TRT. The mechanism is substantially mediated via aromatization — testosterone is converted to 17β-estradiol by aromatase (cyp19a1) in osteoblasts, adipose tissue, and brain; estradiol acting via esr1 is the dominant driver of trabecular bone maintenance and epiphyseal closure in men (not direct AR agonism). Men treated with non-aromatizable androgens (e.g., oxandrolone) show less bone benefit than those treated with testosterone, confirming this. See osteoporosis for the bone aging context.

Sexual function and libido

The most consistent and clinically meaningful benefit of TRT across T-Trials and the Hudson 2023 meta-analysis. Men with total testosterone <12 nmol/L (~350 ng/dL) show statistically significant and patient-relevant improvements in erectile function (IIEF-15 erectile function subscore mean difference 2.14; 95% CI 1.40–2.89) — meeting the minimal clinically important difference for mild erectile dysfunction 6. The magnitude of improvement was not significantly modified by age, baseline testosterone level, BMI, or diabetes status; however, older men and those with obesity have lower absolute post-treatment scores, meaning they are less likely to reach adequate sexual function despite similar incremental gains.

Mood and depression

Modest improvements in depressive symptoms were noted in TRAVERSE (Bhasin et al. 2024, JCEM sub-analysis — not in footnotes; unsourced for this specific claim). Hudson 2023 found no significant improvement in BDI psychological symptoms during TRT vs placebo 6. The signal is not robust enough to recommend TRT as a first-line antidepressant.

Erythropoiesis and anemia

TRT reliably increases hemoglobin and hematocrit via stimulation of the EPO pathway in the kidney. This is beneficial in men with unexplained mild anemia — a specific sub-population from T-Trials showed clinically meaningful hemoglobin improvement 2. However, the same mechanism causes erythrocytosis (polycythemia) — the principal monitorable adverse effect of TRT — defined as hematocrit >54%. Erythrocytosis increases thrombotic risk and may partially explain the pulmonary embolism signal in TRAVERSE. Guidelines recommend stopping or dose-reducing TRT when hematocrit exceeds 54%.

Cognitive function

No benefit demonstrated. The T-Trials cognitive sub-study (n=493, average age 72 years) found no significant improvement in any cognitive domain with 12 months of TRT vs placebo 2. No subsequent large trial has overturned this finding.

Prostate cancer risk: the saturation model

Classic teaching held that testosterone “feeds” prostate cancer; this led to decades of TRT avoidance in any man with prostate history. The saturation model (Morgentaler, 2007 onward) reframes this: androgen receptors are saturated at relatively low testosterone concentrations (~150 ng/dL); above this threshold, incremental testosterone does not drive further prostate cancer growth. Consistent with this, population-level data and the T-Trials / TRAVERSE populations have not demonstrated increased prostate cancer incidence with TRT in appropriately monitored men. PSA monitoring every 3–6 months during the first year of TRT is standard. Men with active prostate cancer remain a contraindication; survivors with undetectable PSA post-treatment are an evolving area (see Bhasin 2026 above).

Biological-Aging Evidence: Does TRT Slow Aging?

No evidence that TRT extends lifespan or slows biological aging in humans. No RCT has used mortality, all-cause lifespan, or validated biological-aging clock (epigenetic clock, DunedinPACE) as a primary or pre-specified secondary endpoint. The “anti-aging clinic” marketing of TRT as a geroprotector is unsupported by the trial evidence. The key gap: needs-human-replication for biological-aging-specific outcomes.

Mechanistic rationale is mixed and not unidirectional:

  • Testosterone activates IGF-1/mTORC1 axis (anabolic; cross-sectional aging association is reduced mTORC1, not increased) — could be interpreted as anti-aging in the muscle-atrophy context or pro-aging in the nutrient-sensing context
  • Aromatization to estradiol supports bone and potentially cardiovascular function via esr1 (modestly pro-longevity direction)
  • Erythrocytosis and atrial fibrillation signals (pro-thrombosis) argue against net cardiovascular geroprotection
  • No DrugAge entry for testosterone as a lifespan-extending compound in model organisms

The hypothesis that age-related testosterone decline constitutes a driver of aging (a form of intercellular communication failure) rather than an epiphenomenon is biologically coherent but not proven. andropause is documented on its own page.

Honest Bottom Line

PopulationEvidence qualityVerdict
Classical hypogonadism (primary/secondary)Strong (multiple RCTs)Clear indication; benefits outweigh risks with monitoring
Older men with T <275 ng/dL + cardinal symptomsModerate (T-Trials, TRAVERSE, Hudson 2023)Modest sexual function and bone benefits; erythrocytosis, AF, PE risks require monitoring; no cognitive or anti-aging benefit
Older men with “low-normal” T (275–400 ng/dL)WeakBenefits attenuated; AF/PE risks remain; guideline-discordant without symptom burden
Anti-aging use in eugonadal menNone / unsupportedNot evidence-based; risks without documented benefits

The reader’s hypothesis that androgen therapies are net “pro-aging” does not match the full evidence profile. In men with genuine hypogonadism, TRT provides bounded, real benefits (sexual function, bone, anemia) with manageable risks. In eugonadal older men without hypogonadism, TRT is not supported and does carry cardiovascular/thromboembolic and erythrocytosis risks — in that context the net balance leans unfavorable. TRT is not a geroprotector and is not cleared for anti-aging indications. The two populations must be distinguished.

Limitations and Gaps

  • long-term-unknown — TRAVERSE median follow-up was 33 months; very long-term cardiovascular effects (10+ years) are unknown.
  • needs-human-replication — no RCT has assessed TRT effect on biological-aging clocks, all-cause mortality in healthy aging populations, or frailty as a primary endpoint.
  • dose-response-unclear — optimal testosterone target range for aging men (if any benefit beyond sexual function exists) is debated; most trials target mid-normal range (400–700 ng/dL).
  • contradictory-evidence — atrial fibrillation signal is consistent across TRAVERSE and meta-analyses, but mechanism is unclear (direct arrhythmogenic vs. indirect via erythrocytosis/viscosity); the PE signal plausibly mediated by erythrocytosis but causality unproven.
  • needs-replication — T-Trials coronary plaque increase (Budoff 2017) was a surrogate; not replicated as MACE events in TRAVERSE, but the MACE non-inferiority does not exclude plaque progression as an intermediate harm.
  • Formulation heterogeneity across trials limits direct comparison.
  • Women’s TRT (for female sexual dysfunction) is a separate, less well-powered evidence base not covered on this page.

R16 Intervention Matrix

  • Hallmarks linked: altered-intercellular-communication (testosterone decline = quantitative shift in endocrine signalling), stem-cell-exhaustion (satellite cell + erythroid progenitor support)
  • Mechanism class: androgen-receptor-agonist — documented in frameworks/intervention-classes.md § androgen-receptor-agonist (added 2026-06-03)
  • Matrix row check: target-hallmarks contains [[altered-intercellular-communication]] — matches the hallmark filename exactly âś“

Cross-References

  • testosterone — verified compound page; full pharmacology, aromatization, PK, and trial-level effect sizes
  • andropause — process page for age-related testosterone decline
  • leydig-cells — cell type responsible for testosterone synthesis
  • lh — pituitary regulator of testicular testosterone production
  • cyp19a1 — aromatase; mediates testosterone → estradiol conversion
  • esr1 — estrogen receptor alpha; mediates bone and cardiovascular effects of aromatized testosterone
  • sarcopenia — phenotype target (muscle mass/strength)
  • frailty — phenotype target (physical function)
  • osteoporosis — phenotype target (bone mineral density)
  • cardiovascular-aging — cardiovascular risk context (TRAVERSE)
  • hormone-replacement-therapy — female counterpart (systemic estrogen/progestogen)

Footnotes

Footnotes

  1. doi:10.1210/jcem.86.2.7280 · Harman SM et al. (BLSA) · J Clin Endocrinol Metab 2001;86(2):724-731 · observational · n=890 men (Baltimore Longitudinal Study of Aging) · longitudinal testosterone decline ~1% per year total; free T falls faster due to SHBG rise · model: humans ↩

  2. doi:10.1056/NEJMoa1506119 · Snyder PJ et al. (T-Trials Investigators) · N Engl J Med 2016;374:611-624 · rct · n=790 men ≥65 yr with total T <275 ng/dL · 12-month testosterone gel vs placebo; coordinated 7-trial design; primary outcomes: sexual function, physical function, vitality · model: humans ↩ ↩2 ↩3 ↩4 ↩5 ↩6

  3. doi:10.1001/jama.2016.21043 · Budoff MJ, Ellenberg SS, Lewis CE et al. (T-Trials) · JAMA 2017;317(7):708-716 · rct · n=138 men (T-Trials CV sub-study) · 12-month coronary CT angiography plaque volume; testosterone vs placebo · greater plaque volume increase in testosterone arm · model: humans surrogate-endpoint — MACE not assessed ↩

  4. doi:10.1056/NEJMoa2215025 · Lincoff AM, Bhasin S, Flevaris P et al. · N Engl J Med 2023;389(2):107-117 · rct · n=5,246 men aged 45–80 yr with hypogonadism + elevated CV risk · median follow-up 33 months; testosterone 1.62% gel vs placebo; primary endpoint MACE (non-inferiority) · HR 0.96 (95% CI 0.78–1.17); AF and PE significantly higher in testosterone arm · model: humans ↩

  5. doi:10.33963/v.phj.111620 · Kardiologia Polska 2026 · review/analysis · TRAVERSE secondary safety signals: atrial fibrillation, acute kidney injury, pulmonary embolism; confirmed statistical significance · model: humans ↩

  6. doi:10.1016/S2666-7568(23)00169-1 · Hudson J et al. · Lancet Healthy Longev 2023;4(10):e561-e572 · systematic review + IPD meta-analysis (17 studies, n=3,431) + aggregate meta-analysis (35 studies, n=5,601) · testosterone vs placebo in men with total T <12 nmol/L; IIEF-15 erectile function subscore mean difference 2.14 (95% CI 1.40–2.89); AMS quality-of-life significantly improved; BDI did not improve significantly; sexual function benefits not significantly modified by age, baseline T, BMI, or diabetes · model: humans ↩ ↩2 ↩3

  7. doi:10.1007/s40256-025-00737-w · Braga MAP et al. · Am J Cardiovasc Drugs 2025;25(6):767-777 · meta-analysis · 23 RCTs, n=9,280 men ≥40 yr with hypogonadism or low-to-low-normal T (≤14 nmol/L), ≥12 months follow-up · no significant difference in all-cause mortality (RR 0.85; 95% CI 0.60–1.19), CV mortality, stroke, or MI; cardiac arrhythmias significantly elevated (RR 1.53; 95% CI 1.20–1.97; p<0.01) · model: humans ↩

  8. doi:10.1001/jamainternmed.2026.1343 · Bhasin S et al. · JAMA Intern Med 2026 May 11 (online ahead of print) · phase 2 rct · n=136 (68 TRT / 68 placebo; 125 completers) · men ≥40 yr with Gleason 6 or 7 [3+4] prostate cancer, undetectable PSA ≥2 yr post-radical prostatectomy, total T <275 ng/dL · testosterone cypionate 100 mg IM weekly for 12 weeks vs placebo · primary efficacy: sexual activity (between-group difference 0.91 daily events; 95% CI 0.56–1.26; P<0.001) · safety: no biochemical recurrence in either group · also improved body composition, stair-climbing power, VO2 peak vs placebo · insufficient duration/size for long-term safety; does not apply to high-grade cancer or ADT/radiation-treated men · model: humans needs-replication ↩

Graph View

Backlinks