The Synthroid brand of levothyroxine improved thyroid-stimulating hormone (TSH) levels by a significantly greater margin compared with generic versions of the synthetic T4 (LT4) hormones for patients with hypothyroidism.
The Synthroid brand of levothyroxine improved thyroid-stimulating hormone (TSH) levels by a significantly greater margin compared with generic versions of the synthetic T4 (LT4) hormones for patients with hypothyroidism, according to a retrospective claims database analysis presented at the American Association of Clinical Endocrinologists (AACE) 26th Annual Scientific and Clinical Congress.
"When compared with generic LT4, Synthroid brand LT4 was associated with significantly improved TSH lab outcomes compared with generic levothyroxine," according to the authors, who worked for AbbVie, the developer of Synthroid.
The retrospective analysis was the first to compare generic and branded versions of levothyroxine in a real-world population. The study was undertaken as even small variations in narrow therapeutic index (NTI) drugs, like levothyroxine, can alter efficacy and safety.
Patient data for the study were selected from the Clinformatics Data Mart database, which contains information on 67 million individuals in the United States. Eligible patients had ≥1 claim for hypothyroidism. Those with thyroid cancer were excluded from the trial. Overall, 28,034 patients were selected for the generic group versus 14,017 in the Synthroid arm.
The mean age of patients in each group was 54 years, and approximately 85% were female. Most patients lived in the south (68%). During follow-up, the TSH was 1.75 mIU/L in the generic group and 1.80 mIU/L for the Synthroid-treated patients. The normal TSH reference range used for the study was 0.3 to 4.12 mIU/L.
The researchers noted that fewer patients were inadequately treated in the Synthroid group versus the generic (20.9% vs 22.6%). Overall, patients receiving generic levothyroxine were 11% more likely to be inadequately treated compared with the branded group (odds ratio [OR], 0.89; 95% CI, 0.65-0.94; P <.0001).
In various sensitivity analyses looking at different TSH ranges, the absolute difference between the groups fluctuated from 1.2% to 2.7%, with each of these assessments showing statistical significance. In matched patients with ≥2 lab outcomes reported, the absolute difference was 2.1% and there was a 14% lower rate of inadequate treatment with Synthroid versus the generic (OR, 0.86; P <.0001). In those treated only between 2013 and 2016, the absolute difference was 2.7% with a 14% lower likelihood of inadequate treatment for Synthroid (OR, 0.86; P <.0001).
"Results were robust to various scenario analyses which showed that changing key parameters of the analyses would not affect the outcome," the authors wrote. "Of note, the greatest differences in proportion of inadequately treated patients were observed in the most robust (≥2 lab outcomes reported) and relevant (2013—2016, time period that covers period after definition of inadequate response was published) analysis scenarios examined."
Across the full analysis, the researchers noted that 59 patients would need to be treated with Synthroid versus the generic to avoid a single inadequate treatment. Across sensitivity analyses, the number needed to treat varied from 37 (those treated between 2013 and 2016) to 83 (a TSH range of 0.4 to 4.0 mIU/L).
As claims data, the authors noted that patient-reported outcomes were lacking. Moreover, they noted that information on treatment simply represented what had been billed and not necessarily whether the medication was taken. Additionally, claims data do not include in-depth patient information, such as smoking status.
"Further research confirming these results and exploring the clinical impact of inadequate treatment is needed," the authors concluded.
Hepp Z, Wang S, Espaillat R. Comparative Effectiveness of Synthroid® vs Generic Levothyroxine on TSH Lab Outcomes: A Retrospective Claims Database Analysis. Presented at: 26th AACE Annual Scientific and Clinical Congress; May 3—7, 2017, Austin, TX. Abstract 1063.