Low-Dose Testosterone in Women: FAERS Safety Signals, FDA Status, and What the Evidence Actually Shows

Why There Is No FDA-Approved Testosterone for Women

The United States has no approved prescription testosterone formulation indicated for women. That is the single most important regulatory fact you need to understand before discussing any safety database.

Intrinsa, a 300-microgram-per-day testosterone patch designed specifically for postmenopausal women with HSDD, completed two key trials (INTIMATE SM1 and SM2) showing statistically significant increases in satisfying sexual events and desire scores compared with placebo. Despite that efficacy signal, the FDA's Reproductive Health Drugs Advisory Committee voted 14-to-1 against approval in December 2004, citing concerns about long-term cardiovascular and breast-cancer safety data. No US sponsor has submitted a new drug application for a female-indicated testosterone product since.

The practical result: the roughly estimated 1-3 million US women who use testosterone for sexual function or hormonal support are doing so entirely through compounded preparations, off-label use of male-formulation products (gels, creams, injections) at fractions of the labeled male dose, or products obtained outside the US. None of those routes come with a package insert written for a female patient.

Why the Label Gap Matters for Safety Monitoring

When a drug lacks an approved indication and a formal label, the FDA's Adverse Event Reporting System (FAERS) becomes the primary post-market safety signal generator. The problem is that FAERS captures spontaneous reports, meaning under-reporting is the rule, not the exception. Estimates suggest fewer than 10 percent of serious adverse events reach FAERS for any given drug. For compounded testosterone specifically, the compound's lot number, concentration, and base formulation may not be recorded consistently, making signal interpretation genuinely difficult.

A second structural problem: most FAERS reports for testosterone in women are coded under male-formulation NDC numbers or listed simply as "testosterone" without specifying dose or route, so female-specific signals can be buried in a data set dominated by male-use patterns and supraphysiologic doses.

What FAERS Reports Actually Show for Women

Mining FAERS for female testosterone signals requires restricting the query to female sex, reproductive-age and postmenopausal age groups, and reported doses below the lowest labeled male dose (typically below 50 mg/day for topical gels). Published disproportionality analyses and FDA FAERS case series point to a consistent cluster of signals.

Androgen Excess: Acne, Hirsutism, and Voice Changes

Androgenic side effects represent the most frequently appearing signals in the female testosterone FAERS literature. A 2021 disproportionality analysis of the FAERS database through Q3 2020 identified acne (reporting odds ratio 4.2), hirsutism (ROR 6.1), and voice alteration (ROR 8.9) as statistically elevated in women coded as testosterone users relative to background drug reports. Voice deepening is of particular concern because it may be irreversible even after the drug is stopped; the laryngeal cartilage responds to androgens in a way that does not reliably reverse with dose reduction.

These signals are dose-dependent and directly tied to one of the core clinical challenges of compounded testosterone: concentration variability. A 2019 analysis of compounded testosterone products found concentration deviations of up to 170 percent from labeled potency in some preparations, meaning a woman prescribed a 1% cream at 10 mg per application might receive substantially more or less than intended.

Polycythemia and Hematologic Signals

Testosterone stimulates erythropoietin production and direct erythroid precursor proliferation. In men using testosterone replacement, polycythemia (hematocrit above 52 percent) is a well-characterized dose-dependent effect. FAERS reports for women include cases of elevated hematocrit, venous thromboembolism, and one cluster of pulmonary embolism cases in women using injectable testosterone formulations at doses above the physiologic female range.

The critical point: the thrombotic risk appears to track supraphysiologic serum levels, not physiologic replacement. No RCT in women using doses targeting the premenopausal range (approximately 15 to 70 ng/dL) has demonstrated a statistically significant VTE signal, but those trials were not powered to detect rare events and ran for 24 weeks or less.

Cardiovascular and Lipid Signals

The FAERS record for cardiovascular events in female testosterone users is sparse but not reassuring. Adverse lipid changes, specifically reductions in HDL cholesterol, appear in case reports and small open-label series. A meta-analysis of 36 randomized trials in postmenopausal women by Islam et al. found that testosterone therapy was associated with a statistically significant reduction in HDL (-3.97 mg/dL, 95% CI -6.03 to -1.91) compared with placebo or estrogen alone, raising a theoretical atherogenic concern. No trial has been large enough or long enough to translate that lipid change into a hard cardiovascular endpoint in women.

Injection-Site and Implant Signals

Pellet implants and injectable testosterone formulations carry a distinct FAERS signature. Pellet extrusion, infection at the implant site, and fibrosis are documented. More clinically significant: pellets cannot be removed once inserted, so if a woman develops androgenic side effects, dose reduction is impossible until the pellet dissolves over three to six months. Several FAERS narratives describe women with persistent voice changes and androgenic alopecia after pellet use, with no corrective option available during that window. Transdermal routes, which can be dose-adjusted or discontinued immediately, are strongly preferred in the 2019 Global Consensus Position Statement for precisely this reason.

The 2019 Global Consensus: What It Says About Safety

The 2019 Global Consensus Position Statement on the Use of Testosterone Therapy for Women, endorsed by The Menopause Society, the British Menopause Society, the Endocrine Society, and nine other international bodies, is the closest thing to an authoritative label women and their clinicians have. It represents a systematic review of 282 papers through 2018.

The Consensus statement specifies the following safety framework, which WomanRx editors propose as a practical monitoring checklist for clinicians and informed patients:

Before starting:

• Serum total testosterone (mass spectrometry-based assay preferred; immunoassay less accurate at low female levels)
• Hematocrit and lipid panel
• Confirmation that the indication is HSDD in a postmenopausal or surgically menopausal woman (the only population with adequate RCT data)
• Ruling out other contributors to low desire: relationship factors, depression, pain disorders, other medications including SSRIs and oral contraceptives

During treatment (every 3-6 months):

• Serum testosterone: target the mid-to-upper premenopausal reference range, not supraphysiologic levels
• Hematocrit
• Clinical assessment for androgenic signs (acne, hair changes, voice)

Stop if:

• Serum testosterone rises above the upper limit of the premenopausal range on two consecutive measurements
• Any androgenic side effects emerge and do not resolve with dose reduction
• Hematocrit exceeds 52 percent

The Consensus explicitly states: "There is a lack of long-term safety data" beyond 24 months for any outcome, including breast cancer and cardiovascular disease, and that testosterone should not be recommended outside the context of a clinical diagnosis of HSDD with proper evaluation.

Sex-Specific Pharmacology: Why Female Testosterone Is Not Just "Less Male Testosterone"

Women metabolize testosterone differently than men, and this is not a minor footnote.

Pharmacokinetics in Women

Women have lower sex hormone-binding globulin production rates at baseline and a smaller volume of distribution for androgens relative to body size, meaning serum free testosterone rises more steeply per unit of exogenous dose compared with men. Studies using radiolabeled testosterone in healthy premenopausal women show that transdermal absorption is also influenced by skin thickness, subcutaneous fat distribution, and, critically, estrogen status: estrogen increases SHBG, which binds testosterone and reduces free fraction.

This means a postmenopausal woman not on concurrent estrogen therapy may experience a higher free testosterone effect from the same total testosterone dose than a woman on combined hormone therapy. Compounding pharmacies rarely adjust for this variable.

The Menstrual Cycle Matters (for Premenopausal Users)

Testosterone peaks around ovulation in the natural cycle, reaching approximately 70 ng/dL in the follicular-to-mid-cycle transition. Women in their reproductive years who are prescribed testosterone for PCOS-related androgen concerns, low libido postpartum, or other indications should be aware that exogenous testosterone will be layered on top of a fluctuating endogenous level. The RCT data supporting testosterone use are almost entirely from postmenopausal women. Premenopausal use is largely extrapolated, and this gap should be stated plainly.

Life-Stage Differences in Risk Profile

Reproductive years: Testosterone use is investigational outside of PCOS management. Androgenic side effects may be more pronounced in women with higher baseline androgen sensitivity (women with PCOS). Anovulation may occur at supraphysiologic levels.

Perimenopause: Testosterone levels fall by approximately 50 percent from the mid-reproductive peak by age 40 to 45, often before estrogen decline becomes symptomatic. Some perimenopause specialists prescribe low-dose testosterone for loss of libido during this stage, but The Menopause Society's 2022 position on hormone therapy notes that perimenopause-specific testosterone RCT data remain insufficient to make a formal recommendation.

Postmenopause: This is the only life stage with adequate RCT evidence. The APHRODITE trial, a 52-week randomized placebo-controlled study in naturally postmenopausal women with HSDD not on estrogen, found a statistically significant improvement in the frequency of satisfying sexual events with a 300-mcg/day testosterone patch versus placebo, with no significant difference in androgenic adverse events at that dose.

Pregnancy, Lactation, and Contraception

Testosterone is absolutely contraindicated in pregnancy. This is not a relative precaution. Exogenous androgens at any dose cause virilization of a female fetus, including clitoral enlargement, labial fusion, and ambiguous genitalia. The effects are dose-dependent but can occur at clinically used doses. Testosterone is classified as FDA Pregnancy Category X, meaning the known fetal risks outweigh any possible benefit.

Any woman of reproductive potential who is prescribed testosterone must use reliable contraception. The contraception conversation is non-negotiable and should happen before the first dose, not at a follow-up visit. Combined hormonal contraceptives are an option but increase SHBG, which reduces free testosterone and may blunt the therapeutic effect. Progestin-only methods or non-hormonal methods (copper IUD) may better preserve free testosterone fraction for therapeutic purposes, though this choice should be individualized.

Lactation: Testosterone transfers into breast milk. The degree of infant exposure is not well quantified in human data. Given the potential for androgenic effects in a nursing infant, testosterone use during breastfeeding is not recommended. Women in the postpartum period asking about testosterone for low libido should be counseled that low desire after childbirth is multifactorial (prolactin elevation, sleep deprivation, estrogen suppression from lactation, relationship adjustment) and that non-pharmacologic approaches should be the first step. If testosterone is considered, breastfeeding should be discontinued first.

Postpartum thyroiditis note: Postpartum thyroiditis affects approximately 5 to 10 percent of postpartum women and is a common contributor to postpartum fatigue and low libido that can be confused with androgen deficiency. Thyroid function should be checked before attributing postpartum symptoms to low testosterone.

Who This Is Right For and Who It Is Not

More likely to benefit

• Postmenopausal women (natural or surgical) with a confirmed clinical and biochemical diagnosis of HSDD, after excluding other causes
• Women in whom HSDD persists despite adequate estrogen therapy (if applicable)
• Women who have been counseled on the absence of an FDA-approved product, the reliance on compounded formulations, and the monitoring requirements

Less likely to benefit or at higher risk

• Women currently pregnant or planning pregnancy in the near term
• Women with active or recent hormone-receptor-positive breast cancer (insufficient safety data; testosterone can aromatize to estradiol)
• Women with polycythemia vera or baseline hematocrit above 50 percent
• Women with untreated severe acne or active androgenic alopecia
• Premenopausal women without a clear diagnostic framework and testosterone level confirming true insufficiency (the evidence base does not exist for this group)
• Women preferring pellet implants over transdermal formulations, given the inability to reverse dose if side effects emerge

What "No Approved Label" Means for Your Informed Consent

When a clinician prescribes compounded low-dose testosterone, you are entitled to understand several things clearly:

The compound is not FDA-approved. The concentration in your specific preparation may vary from what is prescribed; a 2019 review of compounding pharmacy quality found meaningful potency deviations in a significant minority of tested samples. The safety data beyond two years are absent. The FAERS signals, while not definitive, identify androgen excess and hematologic changes as real monitoring priorities.

You should ask your prescriber: What serum testosterone level are we targeting? How often will we check it? What is the plan if I develop acne or notice voice changes? What is the stopping rule?

If those questions do not get specific, numerical answers, that is clinically meaningful information.

Monitoring Protocol: A Practical Summary Table

| Timepoint | Test | Target / Action threshold | |---|---|---| | Baseline | Total testosterone (LC-MS/MS preferred) | Document pre-treatment level | | Baseline | Hematocrit | Stop or defer if >50% | | Baseline | Fasting lipid panel | Establish baseline HDL | | 6 weeks | Total testosterone | Should be in mid-premenopausal range; stop if >upper limit | | 3 months | Clinical androgenic signs | Dose reduce or stop if present | | 6 months | Hematocrit, lipids, testosterone | Ongoing; hematocrit >52% = stop | | Annually | Above plus clinical HSDD reassessment | Is benefit sustained? |