Do GLP1 receptor agonists increase thyroid cancer risk?

Use of glucagon-like peptide 1 (GLP1) receptor agonists does not contribute to an elevated risk of thyroid cancer, as shown in a Scandinavian cohort study with a mean follow-up of nearly 4 years.

The main analysis comparing GLP1 receptor agonists with dipeptidyl peptidase 4 (DPP4) inhibitors reveal an upper limit of the confidence interval (CI) that is consistent with no more than a 31-percent increase in the relative risk.

This study was conducted in Denmark, Norway, and Sweden from 2007 to 2021 and compared patients who initiated GLP1 receptor agonists with those who started DPP4 inhibitors and, in further analysis, those treated with sodium-glucose cotransporter 2 (SGLT2) inhibitors.

Mean follow-up time was 3.9 years in the GLP1 receptor agonist arm and 5.4 years in the DPP4 inhibitor arm. Thyroid cancer developed in 76 of 145,410 patients treated with GLP1 receptor agonists and in 184 of 291,667 of those treated with DPP4 inhibitors (incidence rate, 1.33 vs 1.46 events per 10,000 person-years. [BMJ 2024;385:e078225]

Treatment with GLP1 receptor agonists showed no significant correlation with a higher risk of thyroid cancer (hazard ratio [HR], 0.93, 95 percent CI, 0.66‒1.31; rate difference, ‒0.13, 95 percent CI, ‒0.61 to 0.36 events per 10,000 person-years). For medullary thyroid cancer, however, the HR was 1.19 (95 percent CI, 0.37‒3.86).

When comparing with the SGLT2 inhibitors group, patients in the GLP1 receptor agonist group showed an HR of 1.16 (95 percent CI, 0.65‒2.05) for thyroid cancer.

Potential adverse event

“Although pharmacovigilance studies have found increased reporting rates for thyroid cancer with GLP1 receptor agonists, disproportionality analyses are designed to detect potential safety signals but are not intended to make causal conclusions,” said principal researcher Björn Pasternak, Clinical Epidemiology Division, Department of Medicine, Solna, Karolinska Institutet, Stockholm, Sweden. [Front Pharmacol 2022;13:925377; Gastroenterology 2011;141:150-156]

“Given that thyroid cancer is mentioned in the product label as a potential adverse event, spontaneous reporting might have been driven by physician and public awareness,” he added.

In a recent French study utilizing the national health insurance system database, the results showed a significant association between thyroid cancer and use of GLP1 receptor agonists (HR, 1.46, 95 percent CI, 1.23‒1.74). [Diabetes Care 2023;46:384-390]

However, the comparator in the said study was nonuse of GLP1 receptor agonists instead of an active one, which could have resulted in potential confounding. Moreover, point estimates had a similar magnitude for the duration of drug use <1 year, 1‒3 years, and >3 years.

“Given that a potential effect of GLP1 receptor agonists on thyroid cancer is unlikely to emerge after short term use, these findings might indicate the presence of confounding,” Pasternak said.

“An alternative explanation for the increased risk observed in the French study, emerging early and staying at similar magnitude with increasing duration of use, could be detection bias,” he added. [Diabetes Care 2023;46:249-251]

“Our additional analysis indicated a nominally increased risk restricted to the first year after starting treatment, which might be consistent with an increased detection of thyroid cancer among patients using GLP1 receptor agonists,” Pasternak said.