Frontiers in Endocrinology · Oct 30, 2026
Preconception Liraglutide Exposure, Endocrine-Metabolic Changes, and Reproductive Outcomes with Clinician-Selected Letrozole or Clomiphene in Anovulatory PCOS: A Prospective Observational Cohort Study
By Chengcheng Liu, Haohan Wang, Qiong Wan et al.
Background: Polycystic ovary syndrome (PCOS) combines reproductive and metabolic dysfunction, with insulin resistance contributing to hyperandrogenism and anovulatory infertility. Glucagon-like peptide-1 receptor agonists such as liraglutide may improve metabolic and endocrine parameters before fertility treatment, but prospective evidence linking a liraglutide-containing preconception program with subsequent ovulation-induction outcomes remains limited. Methods: In this single-center prospective observational cohort study at Yichang Central People's Hospital, China, 154 women with anovulatory PCOS entered a preconception liraglutide phase and 138 who reached analyzable ovulation induction were included in the comparative cohort. All 138 analyzable participants were retained, comprising 92 who received clinician-selected letrozole and 46 who received clomiphene. Covariate balance was assessed outcome-blindly with standardized mean differences. The primary endocrine endpoint was within-participant change in free androgen index (FAI), and the primary reproductive endpoint was clinical pregnancy. Results: Among all 154 liraglutide-exposed participants, 72 (46.8%) experienced at least one treatment-emergent adverse event and 4 (2.6%) discontinued because of treatment-limiting gastrointestinal events; no serious adverse events were recorded. The median paired FAI change was -3.36 percentage points (95% bootstrap CI -4.01 to -2.94; p = 2.16 x 10^-24). Per-cycle ovulation occurred in 264/357 (73.9%) letrozole cycles and 134/208 (64.4%) clomiphene cycles. Clinical pregnancy occurred in 38/92 (41.3%) versus 11/46 (23.9%) participants (score RR 1.73, 95% CI 1.01 to 3.12; p = 0.045); the adjusted RR was 1.72 (95% CI 0.98 to 3.00; p = 0.058). Live birth did not differ clearly (29.3% vs 19.6%; RR 1.50, 95% CI 0.80 to 2.96). Time to clinical pregnancy was shorter with letrozole (adjusted HR 2.18, 95% CI 1.11 to 4.29). Conclusion: Broad endocrine-metabolic changes were observed during the liraglutide-containing preconception program. Subsequent letrozole was associated with higher per-cycle ovulation and a shorter time to clinical pregnancy than clomiphene, but the adjusted clinical-pregnancy estimate remained imprecise and no clear live-birth difference was demonstrated. These observational findings are hypothesis-generating and do not establish a causal effect of liraglutide.