Oral testosterone therapy: how effective is it, and what is the actual dose

For most of the modern history of testosterone therapy, the oral route was the one you couldn't use. The only oral testosterone approved in the United States was methyltestosterone, which survives the liver because of a C-17 alkyl group — and that same modification is what made it hepatotoxic, producing cholestasis, peliosis hepatis and hepatic tumours.^[3]

Testosterone undecanoate solves the problem a different way. Formulated in a self-emulsifying delivery system, more than 97% of the absorbed drug travels through the intestinal lymphatic system rather than the portal vein, so it reaches the circulation without a first pass through the liver.^[3] That is also why the capsules must be taken with food: without a meal, absorption is unreliable.^[8]

This matters for how you read the safety data. The liver signal that haunted oral androgens is a property of 17-alpha-alkylation, not of oral dosing as such. In the two-year extension study there were no clinically significant changes in liver function tests.^[2]

The question "does it work" has to be split in two, because the trials only answered the first half.

The trials were designed to show that oral testosterone undecanoate raises testosterone into the normal range. On that endpoint the answer is yes. In the phase 3 registration trial — 221 hypogonadal men aged 18–65, randomised 3:1 against a topical testosterone comparator, open-label — 87% of men on oral testosterone undecanoate achieved an average concentration in the eugonadal range, with a mean average concentration of roughly 489 ng/dL.^[1] Peak concentrations stayed at or below 1500 ng/dL in 90.7% of men.^[1]

A 12-month extension carried 86 men into a second year, with 69 completing 24 months of continuous therapy. Mean total testosterone was 617 ± 427 ng/dL, and every domain of the psychosexual questionnaire improved significantly from baseline.^[2] The standard deviation deserves attention: ±427 ng/dL around a mean of 617 is a wide spread, which is the statistical way of saying that individual men land in very different places on the same dose. This is a titrated drug for a reason.

A second oral testosterone undecanoate product was studied in a 6-month, single-arm phase 3 trial of 155 men (139 in the efficacy analysis), starting at 200 mg twice daily with meals and titrating between 100 and 800 mg daily. It reported 87.8% of all treated men and 96.1% of completers reaching a normal 24-hour mean testosterone, with an average of 452 ng/dL at day 90.^[4] Read that one with the caveats it deserves: single-arm, no comparator, manufacturer-sponsored, and completer analyses always flatter a drug.

Here is the half the trials did not answer. No study has randomised men to oral testosterone versus injections or gel and followed hard outcomes — fractures, cardiovascular events, mortality, or even sustained symptom relief measured against another route. Restoring a number is not the same as improving a life, and anyone telling you oral testosterone is "better than injections" is extrapolating past the evidence.

JATENZO starts at 237 mg twice daily — one capsule in the morning and one in the evening, each with food. From there the dose is titrated between a floor of 158 mg twice daily and a ceiling of 396 mg twice daily, using capsules of 158 mg, 198 mg and 237 mg.^[8][9]

The titration step is the part patients most often get wrong. Serum testosterone is measured 6 hours after the morning dose, and not until at least seven days after starting treatment or changing the dose.^[8][9] A level drawn at the wrong time on this drug is not just imprecise — it is uninterpretable, because the concentration curve peaks roughly four hours after a dose.^[3] Hematocrit is checked about every three months.^[8]

None of this is a protocol to run yourself. It is a scheduled prescription medicine, and the titration only works inside a monitoring relationship with a physician. Peak Human's testosterone replacement therapy entry covers the wider intervention, including the other routes of administration.

This is the honest cost of the oral route, and it is on the label. Ambulatory blood pressure monitoring in the phase 3 trial showed a 4.9 mmHg rise in systolic pressure on oral testosterone undecanoate versus 0.2 mmHg on topical testosterone.^[1] The Canadian product monograph puts it at an average of 4.9/2.5 mmHg systolic/diastolic after four months.^[9] The two-year data described a 3–6 mmHg systolic increase.^[2] The second oral product reported a smaller 24-hour systolic rise of 1.7 mmHg (95% CI 0.3–3.1) at 120 days.^[4]

A 4.9 mmHg shift sounds small and, for one man with a blood pressure of 112/70, it probably is. For a 58-year-old already running 138/88 on two agents, it is not nothing. Both products carry a warning that they are not recommended in men with uncontrolled hypertension, and blood pressure should be measured periodically on treatment.^[8]

Worth knowing what changed in 2025: the FDA removed the class-wide boxed warning about cardiovascular risk from all testosterone products in February 2025, following the TRAVERSE trial and the postmarketing ambulatory blood pressure studies, while adding product-specific blood pressure information.^[7] JATENZO's separate boxed warning about blood pressure increases was removed from its US label in July 2025, with blood pressure retained as a standard warning.^[8] TRAVERSE itself randomised 5,204 men aged 45–80 with hypogonadism and cardiovascular disease or high risk to transdermal testosterone gel or placebo; major adverse cardiac events occurred in 7.0% versus 7.3% (HR 0.96, 95% CI 0.78–1.17), meeting non-inferiority. Atrial fibrillation (3.5% vs 2.4%), acute kidney injury (2.3% vs 1.5%) and pulmonary embolism (0.9% vs 0.5%) were all more common on testosterone.^[6]

Two honest caveats about applying TRAVERSE here: it studied a gel, not an oral capsule, and it was a safety trial in men with confirmed hypogonadism — it is not evidence that testosterone is safe or useful for men whose levels are normal.

Testosterone therapy of any route is a treatment for diagnosed hypogonadism — a deficiency or absence of endogenous testosterone confirmed on testing — not a performance enhancer and not a default response to fatigue.^[9] The regulators kept the limitation-of-use language about age-related declines in place even while removing the cardiovascular boxed warning, which tells you where the evidence still isn't.^[7]

What is genuinely new is availability. Health Canada authorized JATENZO on 12 December 2025,^[9] a month after authorizing a second oral testosterone undecanoate in November 2025, with Canadian availability expected through 2026.^[10] For men who have declined injections or struggled with transference risk from gels, a pill is a real addition to the menu. It is not a reason to start therapy you didn't otherwise need. Our wider hormone optimization guidance covers where this fits.