Glucagon-like Peptide-1 (GLP-1) Receptor Agonists and Cancer Prevention: Methodological Pitfalls in Observational...

Obesity, commonly approximated by body mass index (BMI) ≥ 30 kg/m², is causally associated with at least 13 cancer types (obesity-related cancers) but it is unclear whether weight loss interventions among obese individuals result in risk reduction of cancer. Recently, several trials of short-term use of glucagon-like peptide-1 (GLP-1) receptor agonists, originally designed for use as anti-diabetes medications, have shown substantial weight loss outcomes compared with placebo. Notably, high-dose semaglutide is associated with approximately 15% weight loss in individuals with obesity without diabetes. With these impressive results, hypotheses are emerging that these drugs might have a role in the prevention of obesity-related cancers, mediated through weight loss.

The aim of this opinion paper is to critically appraise the methodological challenges and pitfalls associated with studying the question 'does weight loss through use of GLP-1 receptor agonists reduce cancer risk' through observational studies, and exemplify this through critique of a recent study published in this journal.

We modified the ROBINS-E framework for assessing risk of bias, identifying seven methodological criteria specific to this research question, against which data should be interpreted. These include adequate adjustment for key parameters of body fatness; immortal time bias; treatment allocation bias; survival bias; cumulative drug dose effect; sufficient sojourn time between drug intervention and cancer presentation; and treatment effect specific to obesity-related cancers. We found that 6 out of 7 methodological criteria were at high risk of bias.

Cancer prevention through GLP-1 receptor agonist use should be explored; however, there are several methodological challenges to overcome in understanding this link before it can inform clinical practice and policy.