AndroGel Delayed-Onset Side Effects: What Takes Weeks or Months to Appear

Which AndroGel effects actually arrive late?

A testosterone gel 1% label recommends checking hematocrit before treatment, again at 3 to 6 months, and then annually [2]. These are 1% gel data; TRAVERSE tested a 1.62% testosterone gel [1]. Hematocrit changes were observed at 1 and 3 months in a testosterone gel trial [5], blood pressure had risen by week 16 in a 1% gel study [2], and mean PSA rose over the first 6 months in a 1% gel trial [2].

Two product notes matter when reading the data. AndroGel comes as a 1% gel and a 1.62% gel, and a testosterone gel 1% label states that dosing for the 1% gel differs from the 1.62% gel, which has its own prescribing information [2]. The large cardiovascular trial described below used the 1.62% gel [1].

Hematocrit: the clearest timeline of any late effect

The 1% gel label says to check hematocrit before starting, that re-evaluating it 3 to 6 months after starting is appropriate, and then annually. If hematocrit becomes elevated, the label says to stop therapy until it falls to an acceptable concentration, and notes that an increase in red blood cell mass may increase the risk of thromboembolic events [2].

The timing is visible in a randomized 6-month trial of daily testosterone gel in older men with mobility limitation: hemoglobin and hematocrit rose 7% to 10%, with erythropoietin up and ferritin and hepcidin down at 1 and 3 months. By 6 months erythropoietin and hepcidin had moved back toward baseline while hemoglobin and hematocrit stayed above it, which the authors described as a new set point [5].

How large the shift is depends on the formulation. In a network meta-analysis of 29 placebo-controlled trials in 3,393 men, gel raised mean hematocrit by 3.0% versus placebo (95% CI 1.8 to 4.3), intramuscular testosterone enanthate or cypionate by 4.0% (2.9 to 5.1), and patch by 1.4% [6]. Every formulation raised mean hematocrit significantly versus placebo [6].

Across formulations, men on testosterone replacement had a 315% greater risk of erythrocytosis, defined as hematocrit above 0.52, than controls; all formulations raise hemoglobin and hematocrit, injectables the most [7]. Whether a raised hematocrit translates into more thrombotic events remains unresolved in the review literature [8].

Blood pressure climbs within the first months

In a 4-month ambulatory blood pressure study of the 1% gel, 24-hour systolic and diastolic pressure rose 1.9 and 1.3 mm Hg from baseline by week 16 in 169 men with usable recordings. In men already taking blood pressure medicine, the rise was 3.0 and 2.2 mm Hg. The 1% gel label does not recommend use in men with uncontrolled hypertension [2]. Over 36 months in TRAVERSE, systolic pressure rose 1.0 mm Hg in the testosterone gel group while it fell 0.5 mm Hg on placebo, a between-group difference of 1.5 mm Hg [2]. In February 2025, FDA required a blood pressure warning for testosterone products that lacked one, after ambulatory monitoring studies confirmed increases across all testosterone products [9].

PSA and prostate changes

In the 1% gel trials, mean PSA rose 0.26 ng/mL over the first 6 months, with no further statistically significant mean increase from 6 through 36 months. Individual patients did rise: about 18% of the 162 men in the 3-year extension met the protocol criterion of more than twice baseline PSA or any single value above 6 ng/mL, and 23 of the 29 who met it did so by month 12. Four men had PSA above 6 ng/mL, and prostate cancer was found on biopsy in two of them [2].

The same 1% gel label warns that men with benign prostatic hyperplasia are at increased risk of worsening signs and symptoms, and that men treated with androgens may be at increased risk of prostate cancer, and it says to evaluate patients for prostate cancer before and during treatment [2]. In June 2026, FDA requested label updates revising prostate cancer and benign prostatic hyperplasia safety information for testosterone products [4].

Sperm production and fertility

The 1% gel label warns that exogenous androgens may lead to azoospermia [2]. Exogenous testosterone inhibits pituitary luteinizing hormone and, at large doses, follicle-stimulating hormone, which suppresses sperm production.

The recovery timeline comes from contraceptive research rather than from men on replacement. In an integrated analysis of 30 studies in 1,549 healthy men with normal testosterone levels who used androgen-based contraception, the median time for sperm concentration to return to 20 million/mL was 3.4 months; 67% reached that threshold within 6 months, 90% within 12, 96% within 16 and 100% within 24. Recovery was faster with shorter-acting testosterone, shorter treatment, older age and a higher baseline sperm count [10].

Two retrospective case series describe treatment to restore sperm production. In 49 men with azoospermia or severe oligospermia while on testosterone, human chorionic gonadotropin 3,000 units every other day plus clomiphene, tamoxifen, anastrozole or FSH restored or improved spermatogenesis in 47 (95.9%), at an average of 4.6 months [11]. In a second series of 66 men given hCG plus a selective estrogen receptor modulator, 70% reached a total motile count above 5 million within 12 months [12]. The Endocrine Society guideline recommends against starting testosterone therapy in men planning fertility in the near term [3].

Gynecomastia, acne and application-site reactions

Breast enlargement was reported in 1%, 0% and 3% of men taking 50, 75 and 100 mg in the 180-day trial, and in 2.5% of men in the 3-year extension [2]. Aromatase converts androgens to estrogens [13]. Estrogens stimulate breast tissue growth, while androgens inhibit it [14].

Acne was reported in 1%, 3% and 8% at 50, 75 and 100 mg over 180 days, and application-site reactions in 5%, 3% and 4%. No participant in that trial discontinued because of a skin reaction, and none was severe enough to require treatment [2].

Secondary exposure to children and partners

This is the boxed warning. Virilization has been reported in children secondarily exposed to testosterone gel, and children should avoid contact with unwashed or unclothed application sites [2]. Reported signs include enlargement of the penis or clitoris, pubic hair, increased erections and libido, aggressive behavior and advanced bone age. In most cases the signs regressed once exposure stopped, but in a few, enlarged genitalia did not fully return to an age-appropriate size and bone age stayed modestly greater than chronological age [2]. The 1% gel label says the gel should be promptly discontinued until the cause of virilization is identified, and it is contraindicated in women who are pregnant [2].

Mood and sleep

In the 180-day trial, depression was reported in 1%, 0% and 1% of men at 50, 75 and 100 mg, and emotional lability in 0%, 3% and 3%. Nervousness was reported in 0%, 3% and 1%, while hostility and anxiety appear among reactions reported in fewer than 1% of patients [2]. The 1% gel label also states that sleep apnea may occur in men with risk factors [2].

Other warning signs the label describes

The 1% gel label directs evaluation of leg pain, edema, warmth and redness for deep vein thrombosis, and of acute shortness of breath for pulmonary embolism [2]. Edema, with or without congestive heart failure, may occur in men with existing heart, kidney or liver disease [2]. The label lists changes in body hair distribution and a significant increase in acne in adult women as signs of virilization from secondary exposure [2].