hormones researchDHEA and Testosterone: The Hormone That Peaks at 25 and What Happens After

At age 22, most men are producing peak levels of a steroid they've never heard of. By 45, they've lost half of it. By 70, roughly 80% is gone — a steeper age-related decline than testosterone itself. That steroid is DHEA, the most abundant circulating steroid hormone in the human body, and it's the subject of one of the most reliably overstated claims in the performance world: that it meaningfully raises testosterone.

The reality is more interesting than the marketing.

DHEA matters — but not primarily as a testosterone precursor. It matters as a longevity biomarker, a stress axis indicator, and a clinically validated predictor of all-cause mortality. Where most content leads with the "natural T booster" angle and leaves out the evidence, this piece goes the other direction: here is what the 42 randomized trials actually show, where the genuine clinical utility lies, and how to think about your own levels.

What DHEA Is and Why Its Decline Is Significant#

Dehydroepiandrosterone (DHEA) is produced primarily in the zona reticularis of the adrenal glands, with smaller amounts from the gonads and brain. In circulation, the vast majority exists as DHEA-S (DHEA sulfate) — a sulfated, stable storage form with a half-life of 10–20 hours, compared to the ~20-minute half-life of unconjugated DHEA. When labs test your DHEA level, they're almost always measuring DHEA-S, which is the clinically relevant marker.

DHEA-S peaks in the early twenties, typically around ages 20–25, at levels of approximately 400–500 µg/dL in men. After that, the decline is linear and relentless: approximately 2–3% per year. By age 45, most men are down 40–50% from their peak. By age 70–80, DHEA-S levels are roughly 80% below young-adult values — men in their late seventies retain only about 20% of what they had at their hormonal peak.

This is not a secondary decline. In terms of percentage loss relative to peak, DHEA-S falls faster and further than testosterone, cortisol, and most other adrenal steroids. The decline is so consistent across populations that DHEA-S has been proposed as a biological clock — a running index of adrenal aging.

Why does the adrenal gland produce progressively less DHEA with age? The primary mechanism involves reduced activity of CYP17A1, the enzyme that converts pregnenolone into DHEA via both 17α-hydroxylase and 17,20-lyase activity. Zone reticularis cells also undergo a form of structural atrophy with aging — the tissue that makes DHEA shrinks while the cortisol-producing zones largely maintain their function. The result: cortisol production stays relatively preserved; DHEA production falls. This shift has downstream consequences for the cortisol/DHEA ratio — covered in detail below.

The Conversion Problem: How DHEA Becomes Testosterone (And Why It's Complicated)#

DHEA's path to testosterone involves a series of enzymatic conversions:

1. DHEA → Androstenedione via 3β-HSD (3β-hydroxysteroid dehydrogenase). This is the rate-limiting step in androgen production from DHEA — 3β-HSD activity gates how much flux enters the downstream pathway.
2. Androstenedione → Testosterone via 17β-HSD (17β-hydroxysteroid dehydrogenase).
3. Testosterone → Estradiol via aromatase (CYP19A1), active in adipose tissue, liver, and brain.
4. Testosterone → DHT via 5α-reductase, primarily in prostate, scalp, and skin.

On paper, this looks like DHEA is a meaningful testosterone precursor, and supplementing it should raise testosterone levels. The clinical picture is considerably messier.

The key concept here is intracrinology — a term coined by endocrinologist Fernand Labrie to describe the local production and action of sex steroids in peripheral tissues. An estimated 30–50% of male androgens are not secreted by the testes at all; they're synthesized locally in peripheral tissues from circulating DHEA-S as a substrate. Adipose tissue, muscle, skin, liver, and the brain all express the enzymes needed to desulfate DHEA-S and convert it through the androgen pathway.

This local synthesis has two important implications for clinical practice:

First, it means that circulating serum testosterone is an incomplete picture of androgen exposure in many tissues. Some of the androgen biology attributed to testosterone may actually be driven by local DHEA conversion.

Second, it explains why serum testosterone response to DHEA supplementation is so inconsistent in clinical trials. The conversion that actually matters — in prostate, muscle, bone — occurs locally and doesn't substantially raise circulating levels. When a study shows DHEA increases serum DHEA-S but produces only a modest testosterone increase, that doesn't mean DHEA was biologically inert; the conversion may have occurred peripherally, at the tissue level.

This nuance is almost entirely absent from the supplement marketing around DHEA, which treats serum testosterone as the only meaningful endpoint and declares DHEA successful or failed based solely on that number.

The DHEA/Cortisol Ratio: The Stress Biomarker You're Not Tracking#

Both cortisol and DHEA originate from the same upstream precursor: pregnenolone. Under healthy conditions, DHEA-S circulates at roughly 5–10 times the concentration of cortisol — an intentional balance built into the stress-response architecture.

Under chronic stress, the HPA axis upregulates cortisol production. Pregnenolone gets diverted toward cortisol synthesis at the expense of the DHEA pathway. Over weeks and months of sustained HPA activation, DHEA-S levels fall while cortisol remains elevated — the ratio widens. This is not hypothetical stress physiology; it's a measurable phenomenon tracked in burnout research, trauma medicine, and aging studies.

DHEA has mild anti-glucocorticoid properties. It competes with cortisol at the glucocorticoid receptor and appears to buffer some of cortisol's more immunosuppressive effects. When DHEA falls relative to cortisol, this protective buffering is reduced. Research linking elevated cortisol/DHEA ratios to accelerated cognitive decline, increased frailty, compromised immune function, and higher all-cause mortality spans multiple independent research groups.

Three stages describe the pattern in men under chronic occupational or lifestyle stress:

• Acute phase (weeks): Cortisol rises; DHEA may briefly increase as part of a counter-regulatory response
• Adaptation phase (months): Cortisol remains persistently elevated; DHEA progressively declines; ratio widens
• Exhaustion phase (years): Both cortisol and DHEA can collapse; this pattern is associated with clinical burnout, severe adrenal dysfunction, and significantly elevated disease risk

For men in their 30s and 40s — the core PrimalPrime audience — the practical implication is this: if you're under chronic high stress, your DHEA-S decline may be significantly faster than the 2–3% annual average. Tracking the ratio matters more than tracking either number in isolation.

Testing the DHEA-S/cortisol ratio requires a morning cortisol draw taken at the same time as DHEA-S. A ratio below 3:1 (DHEA-S to cortisol) in a man under 50 is a flag worth investigating. The intervention, in this scenario, is not primarily supplementation — it's load management, sleep restoration, and training periodization. You cannot supplement your way out of a cortisol pattern driven by lifestyle.

What the Randomized Trials Actually Show#

The DHEA literature contains more than 40 placebo-controlled randomized trials. Reading the headlines, you would expect clear answers. The data is more ambiguous.

The dose-response meta-analysis: Qin et al. (2021, Steroids) pooled 42 RCTs across 55 study arms. DHEA supplementation significantly raised plasma testosterone — with a mean increase of +28.02 ng/dL (95% CI: 21.44–34.60). Men specifically showed an average increase of approximately +21 ng/dL. For reference, the normal male testosterone range is 300–1,000 ng/dL. An increase of 21 ng/dL is real but clinically modest — roughly equivalent to going from 380 to 401 ng/dL. Higher doses (>50 mg/day) produced larger increases; younger participants responded more than older participants; and effects were more pronounced at shorter durations (<12 weeks), suggesting possible down-regulation with sustained use.

The landmark null result: Nair et al. (2006, NEJM) — the longest well-powered RCT in this space — randomized 87 elderly men to 75 mg/day DHEA or placebo for two years. DHEA-S and testosterone were restored to young-adult levels. The primary endpoints — body composition, muscle strength, VO2 max, insulin sensitivity, and quality of life — showed no significant differences from placebo in men. This was the study that Peter Attia subsequently cited when he concluded that "in males, DHEA has no utility and has no effect on testosterone" (though a stricter reading of the data shows testosterone did rise; it just didn't translate into clinical benefit).

The elderly meta-analysis: Corona et al. (2013, JCEM) pooled 25 placebo-controlled RCTs in elderly men (N=1,353). DHEA was associated with a modest reduction in fat mass (SMD: −0.35), but this effect disappeared after adjusting for the testosterone and estradiol elevations DHEA produced. No significant effects were found for lipid metabolism, glycemic parameters, bone health, sexual function, or quality of life.

The DHEAge study: Baulieu et al. (2000, PNAS) was the first large RCT (280 participants, ages 60–79) to show any meaningful benefit from DHEA. At 50 mg/day for one year, participants showed improved bone mineral density (primarily in women), better skin quality, and modest libido improvement — significant only in women over 70. Men saw minimal benefit in any measured domain. The study established that sex matters enormously in DHEA response.

The honest summary: DHEA supplementation raises serum testosterone in men — that part is clear. The effect size is modest (approximately 21 ng/dL). The clinical translation of that testosterone increase — better body composition, strength, sexual function, cognition — has not been convincingly demonstrated in any adequately powered RCT. The literature is not a complete null; it's an honest small-effect story in a biological context where individual variation is high.