Abstract Study question What were the effects of free androgen, DHEAS on pregnancy and neonatal outcomes and the cut-off value of East Asian population with PCOS ? Summary answer PCOS patients with biochemical hyperandrogenism show unsatisfactory clinical PR and CLBR when undergoing assisted reproductive technology (ART). What is known already Polycystic ovary syndrome (PCOS) is a prevalent endocrine disorder associated with infertility and pregnancy complications. The pathogenesis of PCOS and its impact on reproductive function may be influenced by different source of androgens, including testosterone, free androgen, dehydroepiandrosterone sulfate (DHEAS). However, the differential effects of these androgen on pregnancy and neonatal outcomes and the cut-off value of East Asian population with PCOS remain unclear. Study design, size, duration A retrospective cohort study was conducted at the Reproductive Medicine Center of the First Affiliated Hospital of Sun Yat-sen University from January 2015 to November 2022, involving 636 cycles of in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI). Subgroup analyses were performed using cut-off values of 6.4 for free androgen index (FAI), 9.5 μmol/L for DHEAS. Pregnancy and neonatal outcomes were compared between groups. Restricted cubic spline (RCS) was used to identify significant cut-off values affecting pregnancy. Participants/materials, setting, methods This retrospective cohort study included 1647 PCOS patients. The diagnosis of PCOS was based on the Rotterdam criteria.Exclusion criteria included patients with congenital adrenal hyperplasia, androgen-secreting masses, oocyte donation cycles, and PGT cycles. Main results and the role of chance Higher FAI levels (>6.4) were associated with decrease in clinical pregnancy rate (PR) (58.91% vs. 51.47%, p = 0.064), live birth rate (LBR) (42.42% vs. 32.35%, p = 0.011), cumulative pregnancy rate (CPR) (p = 0.032), and cumulative live birth rate (p = 0.032). When DHEAS levels exceeded 9.5 μmol/L, there was a significant decrease in clinical PR (60.33% vs. 49.55%, P = 0.008), LBR (42.73% vs. 32.73%, P = 0.012). Negative correlations were also observed between DHEAS levels and cumulative pregnancy rate (p = 0.031) and cumulative live birth rate (CLBR) (p = 0.031). Both FAI and DHEAS elevated is associated with the lowest clinical pregnancy rate (45.95%). Conversely, when solely FAI is elevated, the pregnancy rate increases to 58.10%, while an elevation in DHEAS alone is associated with a pregnancy rate of 61.44%, both of which are lower than when neither FAI nor DHEAS are elevated(62.73%). The live birth rates exhibit a similar trend(30.00%vs40.00%vs41.83%vs44.48%). RCS revealed a significant decrease in CPR and CLBR when DHEA levels exceeded 7.69 umol/L, while the cut-off value of FAI was 6.36 for CPR and CLBR. Limitations, reasons for caution Firstly, the data used to develop the model were obtained from a single center with a relatively small sample size. Secondly, the patients included in this study were predominantly of slim physique with a lower proportion of obesity. Wider implications of the findings Both adrenal-derived DHEAS and ovarian-derived FAI are responsible to the unfavorable pregnancy outcomes in lean PCOS women. Trial registration number 2020B1212060029