To compare the cumulative live birth rate (CLBR) per oocyte retrieval cycle of a conventional progestin-primed ovarian stimulation (cPPOS) regimen with a flexible progestin-primed ovarian stimulation (fPPOS) regimen in poor ovarian response patients, according to POSEIDON criteria.Poor ovarian response women, according to POSEIDON criteria, who underwent the first PPOS protocol for in vitro fertilization/intracytoplasmic sperm injection (IVF/ICSI) between January 2018 and December 2020 were included. The fPPOS group involved 113 participants, and the cPPOS group included 1119 participants. In the cPPOS group, medroxyprogesterone acetate (MPA) (10 mg/d) was administrated on the gonadotropin injection the same day as gonadotropin injections in the cPPOS group, while MPA was started either on the day when the leading follicle with mean diameter > 12mm was present and/or serum E2 was >300 pg/mL in the fPPOS protocol group. The primary outcome was CLBR.The fPPOS protocol had higher CLBR per oocyte retrieval cycle compared to the cPPOS group, even without a statistically significant difference (29.6% vs. 24.9%, p = 0.365). The fPPOS group had fewer numbers of retrieved oocytes (2.87 ± 2.03 vs. 3.76 ± 2.32, p < 0.001) but a higher MII oocyte rate (89.8% vs. 84.7%, p = 0.016). In addition, the number of available embryos in the two groups was comparable (1.37 ± 1.24 vs. 1.63 ± 1.38, p = 0.095). There were five women in the fPPOS group, and 86 women in the cPPOS group had a premature LH surge (4.2% vs. 6.8%, p = 0.261). In the fPPOS group, there was one instance of premature ovulation, while in the cPPOS group, there were six occurrences of premature ovulation (0.8 vs. 0.5%, p = 1.000).The novel fPPOS protocol appears to achieve higher CLBR even without significant differences and with MPA consumption compared with cPPOS protocol in low-prognosis patients.