[Skip to Navigation]
April 16, 1997

Bioequivalence of Generic and Brand-name Levothyroxine Products in the Treatment of Hypothyroidism

Author Affiliations

From the Department of Clinical Pharmacy (Drs Dong, Gambertoglio, White, and Bubp), Division of Biostatistics, Department of Epidemiology and Biostatistics (Dr Hauck and Ms Gee), and Department of Medicine (Dr Greenspan), University of California Medical Center, San Francisco. Dr Hauck is currently with the Division of Clinical Pharmacology, Thomas Jefferson University, Philadelphia, Pa. Dr White is currently with the College of Pharmacy, Washington State University at Spokane. Dr; Bubp is currently with the Biomedical Research Program, San Mateo County General Hospital, San Mateo, Calif.

JAMA. 1997;277(15):1205-1213. doi:10.1001/jama.1997.03540390035032

Objective.  —To compare relative bioavailability of Synthroid, Levoxine (Levoxine has been renamed Levoxyl), and 2 generic levothyroxine sodium preparations.

Design.  —Single-blind (primary investigators blinded), randomized, 4-way crossover trial.

Setting.  —Ambulatory care.

Patients.  —Twenty-Two women with hypothyroidism who were clinically and chemically euthyroid and were receiving levothyroxine sodium, 0.1 or 0.15 mg.

Interventions.  —All patients received each of the 4 levothyroxine products for 6-week periods in the same dosage as their prestudy regimen with no washout period. The order of the drug sequences was randomly determined before study initiation.

Main Outcome Measures.  —Area under the curve, time to peak serum concentrations, and peak serum concentrations of thyroxine, triiodothyronine, and free thyroxine index for all 4 products.

Results.  —All data analyses were completed prior to unblinding of the product codes. No significant differences between the 4 products were found in area under the curve or peak serum concentrations of total thyroxine, total triiodothyronine, or free thyroxine index. Although Synthroid produced a more rapid rise in total serum triiodothyronine concentration and a higher total peak serum triiodothyronine concentration than the other products, these differences were not statistically significant (P=.08). The Food and Drug Administration criterion for relative bioequivalence within 90% confidence intervals (0.8-1.25) was demonstrated (P<.05) for all pairs of products. Relative bioequivalence of 0.95 to 1.07 was demonstrated, tighter than the current bioequivalence criterion for oral formulations.

Conclusions.  —The 4 generic and brand-name levothyroxine preparations studied are different but are bioequivalent by current Food and Drug Administration criteria and are interchangeable in the majority of patients receiving thyroxine replacement therapy. Further investigation is required to determine whether our results are equally applicable to all existing levothyroxine preparations.