Key takeaways
- 5-alpha-reductase converts testosterone to DHT irreversibly, and DHT cannot be aromatised to estradiol.
- The conversion happens inside target tissues, so a serum DHT level predicts hair loss poorly.
- DHT is required for external genital development, prostate growth and pattern hair loss, but not for muscle, voice or libido.
- 5-alpha-reductase inhibitors lower PSA, so a result on one cannot be read against the usual reference.
- The same enzyme makes neurosteroids, which is the most plausible mechanism for the mood and sexual side effects a minority report.
Testosterone gets the attention, but in several tissues it is largely a delivery vehicle. What binds the androgen receptor in the prostate, the scalp follicle and the skin is dihydrotestosterone, made locally from testosterone by an enzyme called 5α-reductase. That one conversion step explains why two men with identical testosterone levels can have completely different hair — and why the drugs that block it help two specific problems while producing effects well outside them.
The conversion step
5α-reductase reduces the 4,5 double bond on testosterone's A-ring, producing dihydrotestosterone (Russell & Wilson, Annu Rev Biochem 1994). Two features of that reaction matter clinically.
First, it is irreversible. Once testosterone has become DHT there is no route back and no route sideways — DHT cannot be aromatised to estradiol, because aromatase needs the double bond that has just been removed. The conversion commits that molecule permanently to pure androgen signalling.
Second, it happens mostly inside target tissues rather than in the bloodstream. The enzyme is expressed where DHT is needed, so the concentration inside a prostate or a hair follicle bears little relationship to the concentration in serum. That is why a serum DHT level is a much weaker predictor of hair loss than people expect: the relevant exposure is local, and local expression is largely genetic.
Why DHT is the more powerful androgen
Compared with testosterone, DHT:
- binds the androgen receptor with roughly 3-5× higher affinity
- dissociates from it more slowly, so the receptor stays occupied longer per binding event
- stabilises the active receptor conformation more effectively, which improves the efficiency of downstream transcription
- cannot be aromatised, so none of it is diverted into oestrogen signalling
Affinity is only part of it; the slow off-rate is arguably more important, because a receptor that stays bound produces more transcriptional output per molecule. The net result is that a tissue expressing 5α-reductase experiences considerably stronger androgen signalling than one that does not, at the same circulating testosterone.
Type 1 and type 2
| Type 1 (SRD5A1) | Type 2 (SRD5A2) | |
|---|---|---|
| Main sites | Skin, sebaceous glands, liver, brain | Prostate, seminal vesicles, hair follicles, external genitalia, liver |
| Clinical relevance | Sebum, body hair, neurosteroid synthesis | Prostate growth, scalp hair loss, genital development |
| Blocked by finasteride | Minimally | Yes |
| Blocked by dutasteride | Yes | Yes |
The type 1 entry for neurosteroid synthesis is easy to skim past and is the key to the side-effect section below. 5α-reductase does not only act on testosterone; it is also the first step in converting progesterone to allopregnanolone, a neurosteroid acting on GABA-A receptors. An enzyme inhibitor does not know which substrate you wanted it to leave alone.
What congenital deficiency taught us
The cleanest evidence comes from people born without functional type 2 enzyme. Genetic males with 5α-reductase type 2 deficiency have normal testosterone but under-virilised external genitalia at birth, because external genital development is DHT-dependent — while internal structures, which are testosterone-dependent, form normally. At puberty, when testosterone rises sharply, they virilise substantially: muscle develops and the voice deepens (Imperato-McGinley et al., 5α-reductase deficiency).
Two things fall out of that natural experiment. DHT is required for external genital development, for pattern hair loss and for prostate enlargement — affected individuals develop neither of the latter two. And DHT is not required for most of what people associate with testosterone: muscle, voice, libido and body composition are largely testosterone-driven. That asymmetry is why blocking DHT is a viable therapy rather than a catastrophe.
Where DHT dominates
- Prostate — the principal androgen driving growth, and therefore benign prostatic hyperplasia
- Scalp hair follicles — in genetically susceptible follicles, DHT shortens the growth phase with each cycle until the follicle miniaturises
- External genitalia — developmental
- Skin — sebum production and body hair
The hair mechanism explains the treatment timeline. DHT does not kill follicles; it progressively shortens the anagen phase, so each cycle produces a thinner hair until the follicle becomes vellus. Because the change accumulates across cycles, blocking DHT halts further miniaturisation quickly but only partially reverses what has happened — and assessing whether treatment is working takes months, not weeks.
The inhibitors
- Finasteride blocks primarily type 2 and reduces serum DHT by around 70%
- Dutasteride blocks both isozymes and reduces it by around 95% (Clark et al., J Clin Endocrinol Metab 2004)
Approved uses are benign prostatic hyperplasia, where reducing prostate volume improves urinary symptoms, and male pattern hair loss, where the aim is to halt progression and recover recently miniaturised follicles.
Two second-order effects are worth anticipating. Blocking the conversion leaves more testosterone available for other routes, including aromatisation, so estradiol tends to rise modestly. And because these drugs shrink the prostate they lower PSA, so a PSA result on a 5α-reductase inhibitor cannot be read against the usual reference without accounting for the drug — anyone using PSA for surveillance needs their prescriber to know. PSA monitoring covers it.
The long-term prostate cancer picture is mixed rather than straightforwardly good: a large prevention trial found fewer prostate cancers detected overall on finasteride, alongside a higher proportion of high-grade tumours among those found — debated on methodological grounds ever since (Thompson et al., N Engl J Med 2003).
The trade-offs, stated honestly
Most men tolerate these medications well. A minority report sexual side effects — reduced libido, erectile difficulty, changes in ejaculation — and a smaller group report mood changes, low motivation or cognitive fog. In some, symptoms have been described as persisting after stopping, the cluster usually referred to as post-finasteride syndrome (Irwig & Kolukula, J Sex Med 2011).
How common and how persistent this is remains contested, and the honest position is that the literature does not settle it (Traish, post-finasteride syndrome review). What is not contested is a plausible mechanism: because 5α-reductase also makes neurosteroids, inhibiting it alters neuroactive steroid profiles, and altered profiles have been measured in affected patients (Melcangi et al., neuroactive steroids in finasteride patients). Allopregnanolone supports GABA signalling; less of it is a reasonable candidate for anxiety, low mood and disturbed sleep.
The practical consequence is not "do not take it". It is that the decision should be made knowing the enzyme has more than one job, and that a new mood or sexual symptom after starting deserves to be taken seriously rather than attributed elsewhere.
On testosterone therapy
Men on testosterone therapy convert more of it to DHT, in proportion to the rise in substrate. Three consequences follow:
- Accelerated pattern hair loss in susceptible men — usually an unmasking rather than a creation, the same trajectory arriving sooner. Whether TRT causes hair loss covers the distinction.
- Prostate effects requiring ongoing PSA monitoring, although a large randomised safety trial did not show testosterone therapy causing prostate cancer (Lincoff et al., N Engl J Med 2023).
- Skin changes — oilier skin and acne, which track DHT more than testosterone.
Combining a 5α-reductase inhibitor with testosterone therapy is a genuine trade rather than a free addition: you are deliberately blocking one arm of a hormone you are deliberately supplementing. Watch estradiol alongside it, since less conversion to DHT leaves more substrate for aromatisation — estradiol management on TRT and the real side effects of TRT cover it.
The clinical insight: DHT is not an unwanted by-product. It is the dominant androgen in four tissues and irrelevant in most others, which is exactly why blocking it is both useful and consequential. You are trading prostate and scalp benefit against a systemic reduction in an enzyme that also makes neurosteroids — a reasonable trade for many men, a poor one for some, and one that should be made explicitly rather than by default.
Bottom line
5α-reductase irreversibly converts testosterone into DHT, a more potent androgen that binds harder, releases more slowly and cannot become oestrogen. Two differently distributed isozymes explain why finasteride and dutasteride behave differently. Congenital deficiency shows what DHT is genuinely required for — external genital development, prostate growth, pattern hair loss — and what it is not, which is most of what people think of as testosterone's effects. Inhibitors work for BPH and hair loss, lower PSA in a way that must be accounted for, and carry a side-effect profile best explained by the enzyme's second job in neurosteroid synthesis. It is a patient-specific decision, and it belongs with a physician who will monitor the downstream markers. The 60-second assessment is where that starts.
Educational content, not medical advice. Laboratory interpretation and any treatment decision are made by a licensed physician after individual evaluation. Individual results vary.
