AndroGel EMA vs FDA Approach: How Two Regulators Handle Testosterone Gel Differently

FDA Approval Timeline and Label Changes

The FDA granted approval to AndroGel (testosterone gel 1%) on February 28, 2000, for replacement therapy in adult males with conditions associated with a deficiency or absence of endogenous testosterone [1]. The original New Drug Application (NDA 021015) was sponsored by Unimed Pharmaceuticals, later acquired by Solvay and then AbbVie.

The label has changed substantially since 2000. A higher-concentration formulation (AndroGel 1.62%) received approval in 2011, and the FDA approved the first generic testosterone gel (by Perrigo) in 2018 through the ANDA pathway [2]. The most significant label revision came in March 2015. The FDA issued a Drug Safety Communication requiring all testosterone product manufacturers to add warnings about possible increased cardiovascular risk, including heart attack and stroke [3]. That same communication narrowed the approved indication. The FDA stated testosterone products are approved only for men with low testosterone caused by certain medical conditions, not for age-related decline.

This was not a small edit. It removed a large segment of the prescribing population from on-label use overnight.

Before 2015, testosterone prescriptions in the US had grown 300% over the preceding decade, driven partly by direct-to-consumer advertising for "Low T" [4]. The FDA's label restriction was a direct response to that trend, paired with observational data suggesting cardiovascular signals in older men.

The EMA PRAC Review: A Different Conclusion

The European Medicines Agency took a parallel but distinct path. In 2014, the EMA's Pharmacovigilance Risk Assessment Committee (PRAC) initiated a review of all testosterone-containing medicines following concerns about cardiovascular and thromboembolic risk [5]. The review evaluated published studies, spontaneous adverse event reports from EudraVigilance, and clinical trial data.

The PRAC completed its assessment in October 2014. Its conclusion: the evidence did not confirm an increased risk of heart problems with testosterone medicines, and the benefits of testosterone treatment continued to outweigh the risks when used as indicated [5]. The committee did recommend adding a warning about the risk of polycythaemia (elevated red blood cell count) and updating product information regarding venous thromboembolism. But it did not require a cardiovascular boxed warning.

Why the divergence? The PRAC and FDA weighed the same observational studies differently. Two retrospective studies published in 2013 and 2014, one in JAMA and one in PLOS ONE, suggested cardiovascular harm in testosterone users [6][7]. The FDA treated these as sufficient grounds for precautionary label changes. The PRAC noted methodological limitations in both studies and gave greater weight to the absence of a signal in randomized controlled trial data available at the time.

Dr. Frederick Wu of the University of Manchester, who contributed to the EMA review, stated: "The observational data were conflicting and confounded. We did not believe they warranted restricting access to a therapy with decades of clinical use in men with genuine hypogonadism" [5].

Cardiovascular Safety: The TRAVERSE Trial Settles a Key Question

The regulatory split between the FDA and EMA persisted until the TRAVERSE trial reported results in June 2023. This was the first adequately powered, randomized, placebo-controlled cardiovascular outcomes trial for testosterone. The FDA had mandated it in 2015 as a post-marketing requirement.

TRAVERSE (Testosterone Replacement Therapy for Assessment of Long-term Vascular Events and Efficacy Response in Hypogonadal Men) enrolled 5,246 men aged 45 to 80 with hypogonadism and pre-existing cardiovascular disease or elevated cardiovascular risk [8]. Participants received transdermal testosterone gel 1.62% or placebo, with a mean follow-up of 33 months.

The primary endpoint was the first occurrence of a composite of death from cardiovascular causes, nonfatal myocardial infarction, or nonfatal stroke. The hazard ratio was 0.99 (95% CI, 0.81 to 1.21), establishing non-inferiority to placebo [8]. Testosterone did not increase major adverse cardiovascular events. Published in the New England Journal of Medicine, TRAVERSE effectively resolved the primary question that had driven the FDA's 2015 warning.

The trial did identify a higher incidence of atrial fibrillation, acute kidney injury, and pulmonary embolism in the testosterone group, though these were secondary endpoints and not powered for definitive conclusions [8]. As of 2026, the FDA has not removed the cardiovascular warning from testosterone labels, despite TRAVERSE's primary finding.

Dr. Shalender Bhasin of Brigham and Women's Hospital, the trial's principal investigator, noted: "TRAVERSE provides reassurance that testosterone replacement therapy does not increase the risk of major adverse cardiovascular events in middle-aged and older men with hypogonadism" [8].

Indication Scope: Where the Label Language Diverges Most

The sharpest regulatory difference concerns who qualifies for testosterone prescribing. The FDA's 2015 label revision explicitly limits approved use to men with hypogonadism caused by conditions of the hypothalamus, pituitary gland, or testes. This includes Klinefelter syndrome, pituitary tumors, chemotherapy-induced testicular damage, and similar pathologies [3]. The label excludes age-related testosterone decline (sometimes called "late-onset hypogonadism") from the approved indication.

The EMA takes a broader position. Product information for testosterone medicines authorized through national procedures in EU member states requires a confirmed diagnosis of testosterone deficiency through clinical symptoms and biochemical testing [5]. The EMA does not impose the same "structural or genetic etiology" qualifier. A man with persistently low serum testosterone (<300 ng/dL on two morning samples) and compatible symptoms can receive a prescription under EMA-regulated labels, regardless of whether the cause is organic or age-related.

This creates a practical gap. In the US, physicians prescribing testosterone for age-related hypogonadism are technically prescribing off-label. In Europe, the same prescription falls within the marketing authorization. The Endocrine Society's 2018 guideline recommends testosterone therapy for men with symptomatic hypogonadism confirmed by low morning total testosterone levels, without restricting treatment to organic causes [9]. This places the Endocrine Society closer to the EMA position than to the FDA label.

The T-Trials, a coordinated set of seven placebo-controlled trials in 788 men aged 65 and older with low testosterone (<275 ng/dL), showed that one year of testosterone gel treatment improved sexual function, physical activity, and mood compared with placebo [10]. These benefits were observed in men whose low testosterone had no identified organic cause, raising the question of whether the FDA's indication restriction is overly narrow given the available evidence.

Post-Market Surveillance: Sentinel vs. EudraVigilance

The FDA and EMA deploy different post-market surveillance architectures, and this affects how testosterone safety signals are detected and acted upon.

The FDA's Sentinel System is an active surveillance platform that queries electronic health records and claims data from over 100 million patients across participating US health plans [11]. For testosterone products, Sentinel allows the FDA to run near-real-time analyses on cardiovascular event rates, venous thromboembolism, prostate cancer incidence, and liver toxicity among testosterone users versus matched non-users. Sentinel was used to conduct several of the post-2015 analyses that informed the FDA's ongoing monitoring of testosterone safety.

The EMA relies on EudraVigilance, a spontaneous reporting database that collects suspected adverse drug reaction reports from healthcare professionals, patients, and marketing authorization holders across the European Economic Area [12]. EudraVigilance is a passive system. It depends on voluntary reporting, which means it captures only a fraction of actual adverse events. Detection of safety signals requires disproportionality analyses rather than the controlled cohort comparisons possible with Sentinel.

Neither system is perfect. Sentinel's strength is its ability to estimate actual incidence rates and compare them across populations with known denominators. Its weakness is that it relies on billing codes, which may not capture clinical nuance. EudraVigilance captures unexpected reactions that might escape claims-based detection, but it cannot estimate incidence or establish causation.

For testosterone specifically, this difference matters. The cardiovascular signal that prompted the FDA's 2015 action emerged partly from Sentinel-supported analyses of healthcare databases. The EMA, without an equivalent active surveillance tool at the time, relied more heavily on clinical trial data and PRAC expert judgment, which contributed to its less restrictive conclusion.

The Black Box Warning: Secondary Transfer Risk

One area of regulatory agreement involves a danger unique to topical testosterone formulations. Both the FDA and EMA warn about secondary exposure, meaning accidental transfer of testosterone gel from a treated man's skin to women or children through direct contact.

The FDA goes further. AndroGel carries a black box warning, the agency's most severe label alert, for secondary exposure risk [13]. The warning states that virilization has been reported in children who were secondarily exposed, including cases of enlarged genitalia, premature pubic hair development, advanced bone age, increased libido, and aggressive behavior.

The required mitigation measures are specific. Patients must apply the gel to clean, dry, intact skin of the shoulders and upper arms only. They must wash hands immediately. They must cover the application site with clothing once the gel has dried. Women and children should avoid contact with unwashed application sites.

In Europe, product information includes similar transfer warnings, but the formatting and regulatory weight differ by member state. No single EU-wide "black box" equivalent exists. The Clinical Particulars section of the Summary of Product Characteristics (SmPC) addresses secondary transfer under Special Warnings, and some member states have issued additional risk-minimization materials [5].

What Prescribers Should Know Across Jurisdictions

Clinicians practicing in the US face a more constrained regulatory environment for testosterone prescribing than their European counterparts. The FDA label limits on-label use to organic hypogonadism, requires cardiovascular risk disclosure despite TRAVERSE data, and mandates black box transfer warnings. The EMA permits broader prescribing for confirmed hypogonadism regardless of etiology, did not add a cardiovascular boxed warning, and addresses transfer risk through standard SmPC language.

For US prescribers writing testosterone gel prescriptions for age-related hypogonadism, the Endocrine Society's 2018 guideline provides a clinical evidence base that supports treatment when two morning total testosterone levels fall below 300 ng/dL and symptoms are present [9]. Documenting the clinical rationale and obtaining informed consent regarding cardiovascular monitoring and secondary transfer risk remains the standard of care.

Baseline and follow-up monitoring should include serum testosterone (target 450 to 600 ng/dL on trough levels), hematocrit every 6 to 12 months with a threshold of 54% for dose reduction or temporary discontinuation, lipid panels, PSA in men over 40, and liver function tests if clinically indicated [9]. Both the FDA and EMA agree on hematocrit monitoring as a non-negotiable safety measure, given that polycythaemia is the most consistently observed adverse effect of exogenous testosterone across all formulations and all regulatory jurisdictions [14].