Abstract Study question Does late follicular phase stimulation yield similar outcomes compared with conventional early follicular phase stimulation in a GnRH antagonist protocol in oocyte donors? Summary answer Late follicular phase stimulation is not inferior in terms of number of oocytes compared to early follicular phase stimulation in a GnRH antagonist protocol. What is known already In patients undergoing fertility preservation for medical reasons, late follicular phase stimulation has been effectively used, resulting in similar numbers of total and mature oocytes obtained, oocyte maturation rate, mature oocyte yield, and fertilization rates compared to conventional early follicular phase ovarian stimulation. Because of LH suppression by endogenous progesterone in the luteal phase, there is less need for the use of a GnRH antagonist. Study design, size, duration This is an open label, phase 3, non-inferiority, randomized controlled trial using a two-arm design with 1:1 allocation ratio. The study included 71 oocyte donors between 18 and 36 years, with a regular menstrual cycle length and BMI 19-35 kg/m2, who underwent ovarian stimulation between November 2018 and May 2022. Patients were allocated to either early follicular start (Group A, n = 36), or to late follicular start (Group B, n = 35). Participants/materials, setting, methods In Group A, patients followed a fixed GnRH antagonist protocol with r-FSH 225IU daily. In Group B, r-FSH 225IU daily was initiated when a dominant follicle and late follicular hormonal values were observed, a GnRH antagonist was added if serum LH-levels were >10IU/L. The primary outcome was number of cumulus-oocyte-complexes (COCs). Secondary endpoints included number of mature oocytes, consumption of gonadotropins, duration of ovarian stimulation, days of GnRH antagonist used, and cost of the stimulation. Main results and the role of chance Using an intention-to-treat analysis, the total number of oocytes did not differ between Group A and Group B (17.7±10.0 vs 17.2±9.1, p = 0.82, difference 0.49 ,95% CI (-4.04 to 5.03)). In the per protocol analysis, after excluding 4 patients, there was no difference between Group A and Group B (18.2±9.7 vs 18.8±7.8, p = 0.62, difference -0.6, 95% CI (-4.9 to 3.7)).The number of mature oocytes did not differ between Group A and Group B (14.1±8.1 vs 12.7±8.5, p = 0.48). In none of the treatment arms OHSS was observed. The duration of stimulation was shorter in Group A than in Group B (10.0±1.6 vs 10.9±1.5 days, p = 0.01). The total amount of r-FSH used was lower in group A than in Group B (2240.7±313.9 IU vs 2453.9±330.1 IU p = 0.008). A GnRH antagonist was used for approximately 6 days in Group A, while in group B, only in one patient a GnRH antagonist was prescribed for 4 days (6.0±1.4 days vs 0.13±0.7 days p < 0.001). There was a significant difference in total medication cost per cycle between both protocols (1147.9±182.8 € in Group A, vs 979.9±129.0 € in group B, p < 0.001), i.e. a cost reduction of 15% for Group B as compared with Group A. Limitations, reasons for caution A limitation of the study is the lack of embryology data. Because of adaptations in the oocyte donation program in our center over the years, this study contains a mix of fresh and frozen donation cycles, leading to a very heterogenous group, making correct interpretation of the embryology data difficult. Wider implications of the findings Late follicular phase stimulation is as efficient as early follicular phase stimulation in terms of number of oocytes. It is patient-friendly, with reduced cost and reduced number of injections. Trial registration number NCT03767218