An analysis of nearly 50 randomized controlled trials has found no link between treatment with glucagon-like peptide (GLP)-1 receptor agonists, either for type 2 diabetes or weight management, and cancer risk.
The findings may serve as “a very important piece of information for clinicians,” said the study’s senior author Cho-Han Chiang, MD, MMSc, from Harvard Medical School.
He explained that initial safety reports suggested there might be an increased risk for thyroid and pancreatic cancer with GLP-1 receptor agonist use.
“Notably, in our meta-analysis we did not show such a signal across the broad spectrum of GLP-1 receptor agonists,” Chiang told Inside Precision Medicine.
Chiang and colleagues also found no link between the treatments, which are increasingly used for glycemic control and weight management in patients with type 2 diabetes or overweight or obesity, and a reduced risk for obesity-related cancers, even though several population-based observational studies have suggested otherwise.
The investigators reviewed data from 48 randomized placebo-controlled trials that included thyroid, pancreatic, colorectal, gastric, esophageal, liver, gallbladder, breast, ovarian, endometrial, or kidney cancer, and multiple myeloma or meningioma as outcomes.
Overall, the studies included 94,245 participants and most often evaluated the use of semaglutide (more commonly known as Ozempic or Wegovy; 42%) or liraglutide (Saxenda or Victoza; 21%).
The researchers report in the Annals of Internal Medicine that, during a median 70 weeks of follow-up, treatment with a GLP-1 receptor agonist had no statistically significant effect, either positive or negative, on the risk for any of the cancers they studied.
They calculated with moderate certainty that, for thyroid cancer, GLP-1 receptor agonist treatment could be associated with one fewer to nine more cases per 10,000 patients treated.
For pancreatic cancer, the risk ranged from nine fewer to six more cases per 10,000, for breast cancer the range was 10 fewer to 12 more cases per 10,000, while for kidney cancer it was five fewer to 13 more per 10,000.
The analysis also revealed with low certainty that GLP-1 receptor agonists may have little or no effect on colorectal, esophageal, liver, gallbladder, ovarian, or endometrial cancer, multiple myeloma, or meningioma. The effect on gastric cancer was designated “very uncertain,” due to imprecision and inconsistency among the studies.
The findings were consistent between studies of semaglutide and tirzepatide and across subgroups stratified by follow-up duration, population, GLP-1 receptor agonist class, weight loss profile, dose, and duration of action.
Chiang said: “It is interesting that our study did not show a reduction in cancer risk with GLP-1 receptor agonist use when previous observational studies have demonstrated beneficial effects of these treatments in cancer prevention/reduction in cancer risk.”
However, he stressed that “the discrepancy likely stems from the fact that these are two different types of study design–randomized controlled trial data in our case which is typically less susceptible to confounding bias due to randomization versus observational data in these previous studies which are susceptible to confounding bias due to the lack of randomization.”
Chiang continued: “When comparing GLP-1 receptor agonist groups with non-GLP-1 receptor agonist groups, there are often inherent differences between the two that are hard to adjust or control for in observational studies. For example, patients who had a GLP-1 receptor agonist might have had better access to medicines or healthcare than those who had non-GLP-1 receptor agonist, or they might be inherently more health conscious or healthier.”
He also pointed out that many of the randomized controlled trials (RCTs) in the current analysis did not have a long follow-up time, which limits the ability to detect cancer occurrences. This could be another reason why the results were different to those in observational studies, which had longer follow-up. In addition, the trials included in the meta-analysis were not designed to assess cancer outcomes.
Chiang concluded that “longer-term studies are needed to clarify potential risks or benefits associated with GLP-1 receptor agonists,” adding that his team will continue to monitor emerging data from both RCTs and observational studies.
