Drug-target Mendelian randomization using genetic variants in the GLP1R gene region as instruments—analyzed across 21 cancer types via public GWAS consortia—found that genetically proxied GLP-1RA exposure was associated with reduced breast cancer risk (log OR = −0.343, p = 6.46×10⁻⁴) and prostate cancer risk (log OR = −0.004, p = 2.25×10⁻⁴), while simultaneously increasing head and neck cancer risk (log OR = 0.350) and thyroid cancer risk (log OR = 1.016, p = 9.12×10⁻⁴). Mediation analyses attributed only ~3–4% of these effects to BMI or HbA1c changes, suggesting largely BMI-independent pathways. UK Biobank Cox regression directionally confirmed the associations, though without reaching statistical significance.
These findings arrive at a critical moment: with semaglutide and tirzepatide prescribed to tens of millions globally, the oncological profile of GLP-1 receptor activation is urgently under-characterized. The thyroid signal is not entirely surprising—rodent studies long flagged C-cell hyperplasia with GLP-1RA exposure, prompting existing contraindications in patients with medullary thyroid carcinoma history—but this MR analysis extends concern to human-relevant genetic proxies across broader thyroid cancer phenotypes. The breast and prostate risk reductions are clinically intriguing and may reflect direct receptor-mediated anti-proliferative signaling beyond weight loss alone, given the weak BMI mediation. Critical limitations apply: MR instruments proxy lifetime receptor activity, not therapeutic dosing windows; effect sizes for prostate cancer appear negligible despite significance; and UK Biobank validation fell short of significance. This is confirmatory for thyroid concern and genuinely novel for the breast/prostate protective signal—warranting prospective pharmacovigilance stratified by cancer subtype.