Kresch · Sexual medicine 2023 · randomized controlled trial · n=75

Efficacy and safety outcomes of a compounded testosterone pellet versus a branded testosterone pellet in men with testosterone deficiency: a single-center, open-label, randomized trial.

Cited 5 times in the scientific literature.

Level 2 - randomized trial

Individual randomized controlled trial

PubMed 36936900 · doi:10.1093/sexmed/qfad007 · record verified 2026-08-29

What was done

In a prospective, phase 3, single-center, randomized noninferiority trial, 75 men with testosterone deficiency were randomized 1:1 to receive either FDA-approved branded testosterone pellets (Testopel: 10 pellets of 75 mg; n = 33 analyzed) or compounded testosterone pellets (E100: 8 pellets of 100 mg; n = 42 analyzed). Participants were followed at 2, 4, and 6 months with morning laboratory tests (prior to 10:00 AM) measuring serum testosterone, estradiol, hematocrit, and prostate-specific antigen (PSA), alongside adverse event monitoring.

What was found

Serum testosterone levels were similar between the Testopel and E100 groups at 2, 4, and 6 months, with 82% of participants dropping to <300 ng/dL by the end of the trial. Adverse event rates were comparable between groups, including elevations in PSA, estradiol, and hematocrit. Dropouts were mostly related to persistent testosterone deficiency symptoms and serum testosterone <300 ng/dL, with similar rates in both arms. Specific numerical values for hormone concentrations and exact adverse event rates were not provided in the abstract.

Why it matters

Compounded subcutaneous testosterone pellets demonstrated comparable efficacy and safety to an FDA-approved branded formulation over 6 months. It also highlights that pellet-induced testosterone levels decline below therapeutic targets before 6 months in a vast majority of patients.

Limits

The study was conducted at a single institution with a small sample size (n = 75), limiting external validity. Exact numerical values for serum hormone levels and adverse event counts were not reported in the abstract. Follow-up was restricted to 6 months without long-term multi-cycle data.

Cited by