Patients who are predisposed to medullary thyroid cancer should be advised against the use of GLP1 agents.
Over a mean follow-up of 3.9 years, the use of glucagon-like peptide 1 (GLP1) receptor agonists was not associated with an increased risk of thyroid cancer, according to a study published in The BMJ.1
Of the 145,410 patients treated with GLP1 and 291,667 treated with dipeptidyl peptidase 4 (DPP4) inhibitor treatment, 76 and 184, respectively, developed thyroid cancer. Therefore, investigators did not find an association between GLP1 use and thyroid cancer (HR, 0.93; 95% CI, 0.66-1.31).
“Our study covers a broad group of patients and provides strong support that GLP-1 analogues are not associated with an increased risk of thyroid cancer,” Björn Pasternak, principal researcher at the Department of Medicine at Karolinska Institutet, said in a statement in the press release.2
This was a cohort study that analyzed administrative registers in Denmark from 2007 to 2021, Norway from 2010 to 2018, and Sweden from 2007 to 2021. Patients between 18 and 84 years were eligible for the analysis if they were new users of GLP1 or DPP4. The study followed patients from the start of treatment use to an outcome event defined as emigration, death, or end of study period. Use of treatment was defined as the observation from the first filled prescription onwards.
In the GLP1 group, the mean age was 57.5 years and 53.2% were men. Between the GLP1 and DPP4 groups, the baseline characteristics were well-balanced. The most common GLP1 use was liraglutide (Victoza; 57.3%), semaglutide (Oempic; 32.9%), dulaglutide (Trulicity; 4.9%), exenatide (Byetta; 4.1%), and lixisenatide (Lyxumia; 0.9%). Of note, 25% of patients were followed for 6 years or longer.
The mean follow-up for those using DPP4 inhibitors was 5.4 years. The proportional hazards assumption found no significant interaction in follow-up time and use of a GLP1 (P = .82).
The incidence rate for the 76 and 184 patients who were diagnosed with thyroid cancer was 1.33 and 1.46 per 10,000 person-years. The most common cancer subtype was papillary followed by follicular, medullary, and other types. There was no increased risk of subtype diagnosis with the use of GLP1.
“We cannot rule out that the risk of certain subtypes of thyroid cancer is increased in smaller patient groups that we could not study here,” Peter Ueda, assistant professor at the Department of Medicine at Karolinska Institutet, said in the press release. He noted that those who are more susceptible to these subtypes of cancers should avoid GLP1 treatment.
There was an analysis conducted for different lag periods after starting GLP1. For a 1-year lag period the HR was 0.83 (95% CI, 0.56-1.22), and for a 2-year lag period the HR was 0.90 (95% CI, 0.58-1.38).
For those who used an alternative as-treated definition for drug use, the analysis found a median follow-up time of 1.8 years with 90-day gap and tail periods (HR, 1.37; 95% CI, 0.84-2.23), 2.6 years for 1-year gap and tail periods (HR, 1.04; 95% CI, 0.69-1.57), and 2.9 years with 2-year gap and tail periods (HR, 0.99; 95% CI, 0.67-1.46).
The post-hoc analysis looked at the impact of competing risk of death ratios for thyroid cancer at 5 years follow-up with death as a censoring event (HR, 0.80; 95% CI, 0.58-1.14) and for death as a competing event (HR, 0.82; 95% CI, 0.59-1.16). At 10 year of follow-up, the ratios for thyroid cancer for death as a censoring event (HR, 1.00; 95% CI, 0.72-1.37) and as a competing event (HR, 1.01; 95% CI, 0.74-1.39) were also assessed.