麻豆传媒

More thyroid cancers were detected soon after starting a GLP-1 receptor agonist than other diabetes drugs, a secondary analysis of a target trial emulation of a comparative effectiveness study found.

Of over 350,000 adults with type 2 diabetes, thyroid cancer risk was significantly higher within the first year after GLP-1 agonist initiation compared with SGLT2 inhibitors, DPP-4 inhibitors, or sulfonylureas (HR 1.85, 95% CI 1.11-3.08), reported Rozalina McCoy, MD, MS, of the University of Maryland School of Medicine in Baltimore, and colleagues.

But when the analysis was broadened beyond year one -- now looking at any length of time since treatment initiation -- GLP-1 agonist use wasn't significantly tied with thyroid cancer compared with the three other diabetes drug classes (HR 1.24, 95% CI 0.88-1.76), they wrote in .

Overall, a thyroid cancer diagnosis occurred in 0.17% of the GLP-1 agonist group, 0.17% of the SGLT2 inhibitor group, 0.20% of the sulfonylurea group, and 0.23% of the DPP-4 inhibitor group.

Risk for thyroid cancer was heightened for GLP-1 agonist users in an overall as-treated analysis that censored patients when therapy was discontinued or another medication was added (HR 2.07, 95% CI 1.10-3.95).

"We found that [GLP-1 drug use] did increase the likelihood of being diagnosed with thyroid cancer, but that appears to be driven not by the greater likelihood of developing cancer, but rather by the greater likelihood of looking for it," McCoy told 麻豆传媒.

"Figuring out if GLP-1 RAs [receptor agonists] cause thyroid cancer is extremely important," she added, noting that more and more Americans have turned to this class of diabetes and obesity medication, which may have additional benefits beyond these indications.

"Clinical trials cannot answer this question because they cannot enroll enough patients nor follow them long enough to see rare or delayed drug effects," said McCoy. Observational studies have yielded mixed data, in part due to studies examining different populations and using different designs, including variation in when some studies start looking for thyroid cancer -- some right after patients start GLP-1 receptor agonist treatment and some waiting 6 or more months, said McCoy.

Concerns were first sparked after rodent models displayed an increased risk for thyroid C-cell tumors including medullary thyroid cancer in a dose- and treatment duration-dependent manner. This prompted all labels of GLP-1 agents to include a boxed warning against use in patients with a personal or family history of this type of cancer.

But fear from this boxed warning may also have prompted clinicians to start looking for thyroid cancer in these patients, "despite the fact that this fear is not grounded in evidence," said McCoy.

"Either patients notice something in the neck, or we start doing neck exams after patients start treatment with a GLP-1 RA -- which I do not think we need to do -- but we end up getting an ultrasound and finding low-risk thyroid cancers," she said.

"This tells us that the increase in the risk of thyroid cancer diagnosis that we have been concerned about is real. We see it in the data," McCoy noted.

"But it is not due to the fact that these patients develop thyroid cancer -- it is too soon for that, especially because the cancers that are diagnosed are not ones that patients die from, meaning that they are low-risk and presumably slow-growing -- but rather because once they are treated with GLP-1 RA, we start looking for thyroid cancer and, when we look for it, we find it," she added.

The study included 40,956 people with type 2 diabetes newly started on a GLP-1 agonist, 74,313 on a DPP-4 inhibitor, 52,049 on an SGLT2 inhibitor, and 187,960 on a sulfonylurea. Average age of participants was 65.3 and 49.3% were female. Median follow-up ranged from 660 days for GLP-1 agonist users up to 1,245 for the DPP-4 inhibitor group. New thyroid cancer diagnoses were identified by ICD-9/10 codes.

GLP-1 agonist initiators had a significantly higher likelihood of undergoing thyroid ultrasonography at 6 and 12 months, suggesting detection bias was behind the increased cancer diagnosis rate observed in the first year of therapy, McCoy and colleagues noted.

While the metabolic advantages of GLP-1 agonists may outweigh the possible risk for thyroid cancer in some patients, this is not the case for patients at high-risk for medullary thyroid cancer, the researchers pointed out.

One limitation was the study's inability to distinguish cases of medullary thyroid cancer from other thyroid cancer subtypes. Participants were also restricted to those using GLP-1 agonists for diabetes, not obesity or cardiovascular risk reduction.

Comment