Testosterone gel and cream: absorption, DHT and transfer risk

Testosterone gel is the transdermal form used in Canada, and only 1 percent products are marketed here. Swerdloff 2000 in the Journal of Clinical Endocrinology and Metabolism found about 9 to 14 percent of the testosterone applied was bioavailable. The rest stays on the skin, which is why transfer to other people is the main practical issue.

Key takeaways

• Swerdloff 2000 in the Journal of Clinical Endocrinology and Metabolism, in 227 hypogonadal men, found about 9 to 14 percent of applied testosterone was bioavailable, with dihydrotestosterone up 3.6-fold and 4.6-fold in the 50 mg and 100 mg gel groups.
• Mazer 2005 in the Journal of Sexual Medicine found dihydrotestosterone to testosterone ratios 2 to 3-fold higher for gel than patch (p less than 0.0001), though both met bioequivalence criteria for testosterone.
• A companion Mazer 2005 paper extracted a median 0.44 micrograms of testosterone from t-shirts at baseline, 25.4 micrograms with the patch and 6,762.7 micrograms with the gel (all p less than 0.0001).
• Stahlman 2012a in Current Medical Research and Opinion found female partner testosterone rose 86 to 185 percent after 15 minutes of direct abdominal contact, with risk persisting at least 12 hours.
• Kaufman 2011 in the Journal of Sexual Medicine, in 274 men over 182 days, found 81.6 to 82.5 percent reached the eugonadal range versus 28.6 to 37.0 percent on placebo.
• Canada has three gel brands across six marketed drug identification numbers, all 1 percent, and no testosterone cream at any status in the Drug Product Database, verified 14 September 2026.

Who does this evidence apply to?

All of it applies to men with confirmed low testosterone. Health Canada’s product monograph for oral testosterone undecanoate capsules, authorized 12 December 2025, requires deficiency to be confirmed by two separate validated biochemical assays of morning testosterone. None of it applies to men whose testosterone is normal. For the diagnostic steps, see how low testosterone is diagnosed in Canada. What testing your own case needs is a decision for your doctor or nurse practitioner.

How does testosterone gel work, and how much of it gets in?

A gel is applied to the skin, a small fraction of the dose crosses into the circulation, and the remainder stays on the surface or is lost. Swerdloff 2000 in the Journal of Clinical Endocrinology and Metabolism, a long-term pharmacokinetic study in 227 hypogonadal men, put that fraction at about 9 to 14 percent.

At 180 days the average 24-hour serum level was highest in the 100 mg gel group, 1.4-fold higher than the 50 mg gel group and 1.9-fold higher than the patch group. That bioavailability figure matters twice over: the amount applied is far larger than the amount absorbed, and the rest is left on skin and clothing.

Why does testosterone gel raise DHT more than other forms?

Because the drug passes through skin, which contains the enzyme that converts testosterone to dihydrotestosterone. Swerdloff 2000 reported it rose 3.6-fold and 4.6-fold in the 50 mg and 100 mg gel groups.

Mazer 2005 in the Journal of Sexual Medicine, a steady-state comparison of patch against gel, is the clearest evidence on this. Both met bioequivalence criteria for average and maximum testosterone concentration, yet dihydrotestosterone levels and dihydrotestosterone to testosterone ratios were 2 to 3-fold higher for the gel (p less than 0.0001). Variability in time to maximum concentration approached 100 percent for the gel, against 23 and 42 percent for the patch.

Two products can therefore deliver the same testosterone exposure and produce different hormone profiles. That is a finding about the route, not a ranking of forms.

Does testosterone gel raise levels into the normal range?

Yes, in roughly three-quarters to four-fifths of treated men. Kaufman 2011 in the Journal of Sexual Medicine, in 274 men over 182 days, found 81.6 to 82.5 percent of active subjects reached average concentrations in the eugonadal range against 28.6 to 37.0 percent on placebo (p less than 0.0001). Kaufman 2012, the one-year extension, found 77.9 percent still in range at day 364, and Cunningham 2017 in Endocrine Practice found 76.1 percent met the criteria at day 90.

Body weight changes how much is needed rather than whether it works. Dobs 2013 in the Journal of Sexual Medicine found 79.1, 79.5 and 73.8 percent of men reached range across body mass index tertiles, with a higher mean daily dose needed in the highest tertile.

Figures used on this page

Do all testosterone gels behave the same way?

No. Olsson 2014 in Clinical Pharmacology in Drug Development studied 11 testosterone-suppressed healthy men and found two newer gels had 2.6-fold and 1.6-fold greater relative bioavailability than Testogel on day 1. Maximum concentrations were 6.25 and 2.97 ng/mL at about 5 to 6 hours, against 1.71 ng/mL at about 24 hours for Testogel.

That study is small and used healthy men whose own testosterone had been suppressed, not patients. Gel names a route, not a single product.

Can testosterone gel transfer to a partner or a child?

Yes, and this is the best-documented practical risk of the transdermal route. The evidence is two transfer studies in couples, one clothing measurement study and two pediatric case report papers.

What the partner studies measured

Stahlman 2012a in Current Medical Research and Opinion studied 72 dosed men and their female partners with 15 minutes of supervised skin contact. Female testosterone rose 86 to 185 percent from baseline after direct abdominal contact, though levels stayed within the female eugonadal range and returned to baseline within 48 hours. A t-shirt barrier reduced transfer by about 40 to 48 percent when 5.0 g was applied to the abdomen, and prevented transfer when 5.0 g was applied to the upper arms and shoulders, where the rise was about 5 to 11 percent. Risk persisted at least 12 hours after application, and the authors called the design exaggerated conditions of transfer.

Stahlman 2012b, a companion paper, studied 24 couples across application site, clothing and washing. Female exposure rose up to 27 percent at the 2.5 g dose and up to 280 percent at the 5.0 g dose after direct contact at 2 hours. A t-shirt prevented transfer at the 2.5 g dose, and washing the site before contact brought average exposure back to about baseline.

How much testosterone ends up on clothing

Mazer 2005 in the Journal of Sexual Medicine measured what could be extracted from t-shirts worn by treated men. Median amounts were 0.44 micrograms at baseline, 25.4 micrograms for the patch and 6,762.7 micrograms for the gel, all at p less than 0.0001.

The pediatric case reports

Nelson 2013 in the Journal of Pediatric Endocrinology and Metabolism described two pre-pubertal children exposed to a parent’s topical androgen: a 21-month-old boy and a girl aged 3 years and 8 months. The boy had six weeks of pubic hair, phallic growth and a growth spurt, reaching Tanner stage 2 pubic hair and Tanner 3 phallic development with prepubertal testes of 2 mL. Both fathers applied the product to the arms and chest before bed, and both children often slept with their parents. Levels returned to normal once exposure stopped.

Cavender 2010 in the Journal of Sexual Medicine described precocious puberty in a 10-month-old boy, with symptoms receding once the father switched from a topical product to a buccal one. These are case reports, not a rate, and they cannot say how often this happens.

Why did the testosterone patch disappear?

It failed on skin tolerability, and the regulatory record shows the withdrawal was commercial rather than a safety finding. Wang 2000 in the Journal of Clinical Endocrinology and Metabolism reported skin irritation in 5.5 percent of men treated with gel and in 66 percent of men in the permeation-enhanced patch group. Kuhnert 2005 in the European Journal of Endocrinology, in 162 men over 24 weeks, likewise found tolerability much better in the gel groups.

The United States Orange Book entries for the 2 mg and 4 mg patch strengths carry the Federal Register determination that the product was not discontinued or withdrawn for safety or effectiveness reasons. In Canada the two patch drug identification numbers, 02239653 and 02245972, were cancelled post market on 14 April 2022 and 13 December 2021. No testosterone patch is marketed in Canada, the United States, the United Kingdom or Australia, verified 14 September 2026.

Is testosterone cream available in Canada?

No. No testosterone cream dosage form appears in the Health Canada Drug Product Database at any status, verified 14 September 2026, and none ever has. Any Canadian page offering cream as an option is describing something that cannot be dispensed here.

Australia is the only one of the four countries checked with a registered cream for men. AndroForte 5, testosterone 5 percent weight in volume in a 50 mL tube from Lawley Pharmaceuticals, holds ARTG 123764, active since 11 July 2014, and is Pharmaceutical Benefits Scheme item 10378F at a dispensed price of $105.62 with a patient charge of $25.00.

The evidence behind it is one bioequivalence study. Wittert 2016 in Andrology, an open-label crossover trial in 16 androgen-deficient Australian men aged 29 to 73 across two 30-day periods, compared the cream against a 1 percent gel. Cream to gel ratios at day 30 were 0.99 for area under the curve (90 percent confidence interval 0.86 to 1.14), 1.02 for maximum concentration (0.84 to 1.24) and 0.99 for average concentration (0.86 to 1.14), all inside the 80 to 125 percent bioequivalence window, with no significant differences in dihydrotestosterone, estradiol, sex hormone binding globulin, luteinising hormone or follicle stimulating hormone. It establishes bioequivalence, not advantage.

Preparations made up by a pharmacy are a separate category, covered in compounded testosterone in Canada.

Which testosterone gels are available in Canada?

Three brands, six marketed drug identification numbers, all 1 percent, verified in the Drug Product Database on 14 September 2026.

Canada has no 1.62 percent or 2 percent gel, and no 2 percent topical solution, because Axiron, DIN 02382369, was cancelled on 4 December 2017. Several trials above studied products not sold here. See which testosterone products are available in Canada and testosterone products compared across four countries.

What the evidence does not show

• No difference in patient-reported outcomes between gel and patch. Kuhnert 2005, in 162 men over 24 weeks, found questionnaires on sexual function, mood and quality of life did not differ, even though tolerability favoured the gels.
• No evidence that gel is the lowest-risk form for red cell counts. Nackeeran 2022 in the Journal of Urology, pooling 29 randomized trials in 3,393 men, found a mean hematocrit rise versus placebo of 3.0 percent for gel (95 percent confidence interval 1.8 to 4.3), against 1.4 percent for patch and 4.0 percent for intramuscular enanthate or cypionate.
• No transfer rate. The Stahlman studies used supervised contact the authors called exaggerated, and the pediatric papers are case reports.
• No demonstrated clinical advantage for cream over gel, because Wittert 2016 established bioequivalence in 16 men rather than superiority, and no head-to-head trial of the Canadian 1 percent gels against each other.

Frequently asked questions

How much of a testosterone gel dose actually gets into the body?

Swerdloff 2000 in the Journal of Clinical Endocrinology and Metabolism, a study in 227 hypogonadal men, found about 9 to 14 percent of the testosterone applied was bioavailable. The remainder stays on the skin or is lost. That is one reason transfer to other people through skin contact and clothing is the main practical safety issue with this form.

Can testosterone gel transfer to a partner?

Yes. Stahlman 2012a in Current Medical Research and Opinion studied 72 dosed men and their female partners after 15 minutes of supervised skin contact. Female testosterone rose 86 to 185 percent from baseline after direct abdominal contact, although levels stayed in the female eugonadal range and returned to baseline within 48 hours. The design simulated exaggerated conditions of transfer.

Does a t-shirt stop testosterone gel from transferring?

It depends on the site and the amount studied. Stahlman 2012a found a t-shirt reduced transfer by about 40 to 48 percent when 5.0 g was applied to the abdomen, and prevented it when 5.0 g was applied to the upper arms and shoulders. Stahlman 2012b found a t-shirt prevented transfer at the 2.5 g dose, and washing the site before contact returned average exposure to about baseline.

Have children been affected by testosterone gel transfer?

Two case report papers describe it. Nelson 2013 in the Journal of Pediatric Endocrinology and Metabolism described a 21-month-old boy and a girl aged 3 years and 8 months exposed to a father’s topical androgen, with pubic hair, phallic growth and a growth spurt in the boy. Cavender 2010 in the Journal of Sexual Medicine described a 10-month-old boy. Signs receded once exposure stopped.

Is testosterone cream available in Canada?

No. No testosterone cream dosage form appears in the Health Canada Drug Product Database at any status, verified 14 September 2026, and none ever has. Australia is the only one of the four countries checked where a cream for men is registered: AndroForte 5, ARTG 123764, active since 11 July 2014 and Pharmaceutical Benefits Scheme item 10378F.

Related reading: testosterone injections and the evidence, hematocrit by delivery method and what gets monitored on testosterone therapy.

References

1. Swerdloff et al. 2000. Long-term pharmacokinetics of transdermal testosterone gel in hypogonadal men. Journal of Clinical Endocrinology and Metabolism. PMID 11134099. DOI: https://doi.org/10.1210/jcem.85.12.7045
2. Kaufman et al. 2011. Efficacy and safety study of 1.62% testosterone gel. The Journal of Sexual Medicine. PMID 21492400. DOI: https://doi.org/10.1111/j.1743-6109.2011.02265.x
3. Kaufman et al. 2012. One-year open-label extension. The Journal of Sexual Medicine. PMID 22321357. DOI: https://doi.org/10.1111/j.1743-6109.2011.02630.x
4. Dobs et al. 2011. A novel testosterone 2% gel. Journal of Andrology. PMID 21979302. DOI: https://doi.org/10.2164/jandrol.111.014308
5. Dobs et al. 2013. Testosterone 2% gel regardless of body mass index. The Journal of Sexual Medicine. PMID 24283410. DOI: https://doi.org/10.1111/jsm.12411
6. Cunningham et al. 2017. Endocrine Practice. PMID 28225313. DOI: https://doi.org/10.4158/EP161665.OR
7. Olsson et al. 2014. Comparison with Testogel. Clinical Pharmacology in Drug Development. PMID 27129008. DOI: https://doi.org/10.1002/cpdd.110
8. Stahlman et al. 2012. Serum testosterone levels in non-dosed females after secondary exposure to 1.62% testosterone gel: effects of clothing barrier. Current Medical Research and Opinion. PMID 22188558. DOI: https://doi.org/10.1185/03007995.2011.652732
9. Stahlman et al. 2012. Effect of application site, clothing barrier, and application site washing. Current Medical Research and Opinion. PMID 22188557. DOI: https://doi.org/10.1185/03007995.2011.652731
10. Mazer et al. 2005. Transfer of transdermally applied testosterone to clothing: patch versus gel. The Journal of Sexual Medicine. PMID 16422890. DOI: https://doi.org/10.1111/j.1743-6109.2005.20232.x
11. Mazer et al. 2005. Steady-state pharmacokinetics, patch versus gel. The Journal of Sexual Medicine. PMID 16422889. DOI: https://doi.org/10.1111/j.1743-6109.2005.20231.x
12. Nelson et al. 2013. Virilization in two pre-pubertal children exposed to topical androgen. Journal of Pediatric Endocrinology and Metabolism. PMID 23729604. DOI: https://doi.org/10.1515/jpem-2013-0127
13. Cavender et al. 2010. Precocious puberty secondary to topical testosterone transfer. The Journal of Sexual Medicine. PMID 21114766. DOI: https://doi.org/10.1111/j.1743-6109.2010.02082.x
14. Wang et al. 2000. Transdermal testosterone gel improves sexual function, mood, muscle strength, and body composition parameters in hypogonadal men. Journal of Clinical Endocrinology and Metabolism. PMID 10946892. DOI: https://doi.org/10.1210/jcem.85.8.6747
15. Kuhnert et al. 2005. Scrotal versus non-scrotal gel versus patch. European Journal of Endocrinology. PMID 16061839. DOI: https://doi.org/10.1530/eje.1.01964
16. Wittert et al. 2016. An open-label, phase 2, single centre, randomized, crossover design bioequivalence study of AndroForte 5 testosterone cream and Testogel 1% testosterone gel in hypogonadal men: study LP101. Andrology. PMID 26754331. DOI: https://doi.org/10.1111/andr.12129
17. Morton et al. 2020. A National Analysis of Temporal Changes in Prescribing of Testosterone Replacement Therapy. The World Journal of Men’s Health. PMID 32777869. DOI: https://doi.org/10.5534/wjmh.190166
18. Nackeeran et al. 2022. The Effect of Route of Testosterone on Changes in Hematocrit: A Systematic Review and Bayesian Network Meta-Analysis of Randomized Trials. The Journal of Urology. PMID 34445892. DOI: https://doi.org/10.1097/JU.0000000000002188
19. Health Canada. Drug Product Database. Queried 14 September 2026. https://health-products.canada.ca/dpd-bdpp/search/
20. Health Canada. Product Monograph, oral testosterone undecanoate capsules. Authorized 12 December 2025. https://pdf.hres.ca/dpd_pm/00082762.PDF
21. United States Food and Drug Administration. Drugs@FDA and the Orange Book. Queried 14 September 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
22. Therapeutic Goods Administration. Australian Register of Therapeutic Goods. Queried 14 September 2026. https://www.tga.gov.au/resources/artg
23. Pharmaceutical Benefits Scheme. Item listings. Queried 14 September 2026. https://www.pbs.gov.au

This page is educational information, not medical advice. Testosterone is a prescription medication and a controlled substance in Canada. Talk to your doctor or pharmacist about your own situation.