Indeed, safety comes first for the use of GLP-1RAs

Abstract

Sir,

Thank you for the opportunity to respond to a letter to the editor by Viganò et al. regarding our recently published review article in Acta Obstetricia et Gynecologica Scandinavica (AOGS).¹ They felt that the issue of possible teratogenicity and clinical safety has not been sufficiently addressed by our review.²

Glucagon-like peptide-1 (GLP-1), an incretin hormone, plays a crucial role in maintaining glucose homeostasis, insulin sensitivity, and body weight under normal physiological circumstances. Thus, a normally functioning GLP-1 system would evidently have a direct positive impact on female reproductive health, predominantly regulating hormonal balance and optimal ovarian function that could support natural conception. Disruptions in the GLP-1 system, as seen in conditions like obesity and polycystic ovary syndrome (PCOS), can interfere with female reproductive health. GLP-1 receptor agonists (GLP-1RAs) mimic the action of GLP-1 and have been proven effective in improving glycemic control, showing remarkable potential in managing obesity and PCOS. Furthermore, the occurrence of numerous unplanned pregnancies during GLP-1RAs treatment underscores the need to understand the mechanism of GLP-1 (RAs) on female fertility and its safety during the pre-implantation period, on exposed fetuses, and long-term effects on children after birth.

In our recent review, we addressed this knowledge gap by outlining the effect of GLP-1RAs on female fertility, especially in obese and PCOS patients, with a specific focus on endometrium and implantation.² Based on initial animal model studies and clinical studies in women, the GLP-1RAs treatment could be promising for alleviating female infertility. However, as mentioned in the key message of our review, “it is crucial to understand the limited evidence available, and their use should be approached with caution, considering potential side effects.”

Viganò et al. raised a concern that our review should have better addressed the safety concerns related to potential teratogenicity and adverse developmental outcomes,¹ citing some animal studies that have shown evidence of adverse outcomes after exposure to GLP-1RA during pregnancy. However, a recent observational population-based cohort study among women with type 2 diabetes exposed to GLP-1RA from 90 days before pregnancy to the end of the first trimester did not show a significantly higher risk of major congenital malformations beyond the baseline risk associated with type 2 diabetes.³ Similarly, no major birth defects were reported in another cohort study among the women exposed to GLP-1RA in early pregnancy, either for diabetes or obesity treatment.⁴ Nevertheless, ambiguous study design and data analysis of recent clinical studies have resulted in debatable conclusions on teratogenicity, embryonic or fetal death, and other pregnancy-associated complications.⁵ Additionally, due to the paucity of evidence in humans regarding the precise washout period for GLP-1RA and their potential adverse effects on fetal development, their use is not recommended during pregnancy or when planning to conceive. Despite this, unplanned pregnancies may occur among women who conceive during GLP-1RA treatment that results in fetal exposure.

Although we have mentioned the safety concerns in our review, we did not emphasize the abovementioned arguments because our principal focus was the effect of GLP-1RAs on endometrium and embryo implantation, calling for further research in this emerging field and not offering clinical recommendations. Our review encourages utilizing multi-omics and single-cell approaches to explore the underlying mechanisms of GLP-1 and its agonists in early stages of conception and highlights the potential of utilizing patient-specific in vitro endometrial models and blastocyst-like structures, blastoids, which would provide preliminary insights into the biological effects of GLP-1RA treatments during the early stages of implantation and pregnancy.

In this context, studies aimed at understanding the impact of GLP-1RA on the cellular composition and molecular features of blastoids, as well as their interaction with the endometrium, could help reveal the potential origin of the teratogenic effects of this drug, if any. Like a zygote developing into a blastocyst, human naive pluripotent stem cells can self-organize into 3D blastoids, which are composed of three founding lineages: epiblast, primitive endoderm (hypoblast), and trophectoderm cells.⁶ The trophectoderm cells adjacent to the epiblast, known as the polar trophectoderm, interact with the maternal uterine epithelium both in vivo and in vitro, establishing the dimensions for further development. Blastoids can even be used to model early gastrulation during an additional 14 days of 3D in vitro culture⁷ that later primes organogenesis. Aberrations in these complex processes can result in anatomical and developmental anomalies. Therefore, it is crucial to initiate teratogenicity studies using human in vitro embryogenesis, implantation, and gastrulation models to assess the impact of GLP-1RA on embryogenesis and its potential teratogenic effects.

We fully agree that safety comes first for the use of GLP-1RAs and would like to emphasize that advanced in vitro models, along with molecular and cellular studies, provide a crucial tool to unravel the involvement of GLP-1 in the early stages of human reproduction. Such studies, conducted in an ethically responsible manner, can help to confirm the safety of GLP-1RA use during the periconceptional period and early pregnancy or warn about its potential teratogenic effects.