Follitropin Alpha versus Follitropin Beta in IVF/ICSI Cycle: A Retrospective Cohort Study.

The use of Gn has experienced human menopausal gonadotropins, urinary-derived gonadotropins, and recombinant human follicle-stimulating hormone (r-FSH), and is widely used in the process of in vitro fertilization and embryo transfer (IVF-ET). 1 In 1988, recombinant DNA technology was used to develop the gonadotropin for higher purity gonadotropin, 2 Gonal-F was successfully developed by transforming subunits of human Follitropin beta genes coding α and β into expression vectors into Chinese hamster ovary cells, 3 licensed in Europe in 1996. In 1997, Puregon, another recombinant human FSH, was approved. 4 Advocation of recombinant DNA technology has promoted the development of high purity and high-specific r-FSH, according to the composition subunit, composed of two non-covalently connected, different glycoproteins, respectively designed as α subunit and β base. These α and β subunits were encoded by different genes, 92 and 111 amino acids, respectively. There are currently three r-FSH products on the market, FSH α, FSH β and FSH δ, and a fourth product, FSH ε, is reported to be under development. In a systematic review and meta-analysis, 7 randomized controlled trials (RCTs) were included to compare the effects of personalized r-FSH δ with traditional r-FSH α/β in gonadotropin releasing hormone (GnRH) antagonist IVF/ICSI cycles. Neither group demonstrated significant differences in clinical pregnancy rates, ongoing pregnancy rates, or live birth rates. 5 r-FSHα (Gonal-F) and r-FSHβ (Puregon) in IVF/intracytoplasmic sperm injection (ICSI) are widely used, with many disclosed, safe and clinical evidence and experience. 6 , 7 The two preparations are similar to pituitary and urinary derived FSH, but show slight differences in carbohydrate side chain structure, containing more alkali and less acidic hormones than urinary derived gonadotrophin preparations. 2 r-FSHα and r-FSHβ have been compared in previous studies, and although there are some differences between the two pharmacokinetics, they have shown similar safety and efficacy in clinical trials. 8 It has also been suggested that the differences in the isomer profiles of the two r-FSH preparations may affect their clinical efficacy. 9 In a recent study, the dose of r-FSHβ was lower than that of r-FSHα, and no significant difference was observed in oocyte production, but the clinical pregnancy rate was lower in the r-FSHβ group than that in the r-FSHα group. These discrepancies may be related to the individual characteristics of the subjects and the different procedures of ovulation induction. 10 In this retrospective cohort study, we aim to examine the ovarian responsiveness and cost of two different r-FSHs in Chinese women. Since they were authorized by the European Medicines Agency, there has been debate over the efficacy and safety of two r-FSH preparations. 8 , 11–13 Various clinical parameters of r-FSHα and r-FSHβ have been assessed, but few studies have included CLBR as the primary outcome. The CLBR has been paid increasingly attention, and has been used as the main pregnancy outcome of IVF/ICSI. The ultimate goal of assisted reproductive technology (ART) is to successfully conceive and deliver a live baby, and CLBR is of even greater clinical significance for the patient’s physical, mental, and economic well-being, as well as for the clinician’s treatment plan. Therefore, we conducted this retrospective study to analyze all cycles including fresh and frozen embryo transfer after one ovarian stimulation cycle of ovum pick-up, pregnancy outcomes including CLBR were compared between Gonal-F and Puregon groups to explore more optimized medication regimens for ovarian stimulation.

The flow chart in Figure 1 represents the study population. 2864 cycles out of a total of 6364 fulfilled the inclusion criteria. 1313 were identified as Puregon (Group A) and 1551 as Gonal-F (Group B). In Table 1 , the differences in duration of infertility (p = 0.255), gravidity (p = 0.050), infertility type (p = 0.229), BMI (p = 0.465) and basal E 2 (p = 0.489) were not statistically significant (p > 0.05). There were significant differences in age (p < 0.001), etiology of infertility (p < 0.001), parity (p = 0.012), AMH (p = 0.032), basal FSH (p = 0.037), basal luteinizing hormone (p = 0.001), and AFC (p = 0.009). Nevertheless, as shown in Table 1 , there was no longer a significant difference between the two groups after PSM. The distributions of PS and standardized difference were shown in Figure 2 . Table 1 Baseline Characteristics Characteristic Before Propensity Score Matching After Propensity Score Matching Puregon (Group A) Gonal F (Group B) P-value Puregon (Group A) Gonal F (Group B) P-value Patients 1313 1551 1160 1160 Female age at oocyte retrieval (years; mean±SD) 32.1±4.7 31.3±4.4 <0.001 a 32.0±4.7 31.9±4.5 0.593 a Etiology of infertility <0.001 c 0.083 c  Tubal factor 1140 (86.8) 1293 (83.4) 1007 (86.8) 990 (85.3)  Male factor 47 (3.6) 52 (3.4) 43 (3.7) 41 (3.5)  PCOS 90 (6.9) 177 (11.4) 81 (7.0) 110 (9.5)  Endometriosis 36 (2.7) 29 (1.9) 29 (2.5) 19 (1.6) Duration of infertility (years; median (IQR)) 3(2,4) 3 (2,4) 0.255 b 3 (2, 4) 3 (2, 4) 0.571 b Gravidity (median (IQR)) 1 (0,2) 1 (0,2) 0.050 b 1 (0, 2) 1 (0, 2) 0.929 b Parity (median (IQR)) 0 (0,1) 0 (0,0) 0.012 b 0 (0, 1) 0 (0, 1) 0.507 b Infertility type 0.229 c 0.739 c  Primary infertility 607 (46.2) 752 (48.5) 540 (46.6) 548 (47.2)  Secondary infertility 706 (53.8) 799 (51.5) 620 (53.4) 612 (52.8) BMI (kg/m 2 ; mean (SD)) 23.6±3.8 23.5±3.9 0.465 a 23.5±3.8 23.5±3.9 0.650 a AMH (ng/ml; median (IQR)) 2.6 (2.1, 3.5) 2.7 (2.2, 3.5) 0.032 b 2.7 (2.2, 3.5) 2.6 (2.2, 3.4) 0.667 b bFSH (mIU/ml; mean (SD)) 7.4±2.8 7.2±2.7 0.037 a 7.3±2.6 7.3±2.8 0.745 a bLH (mIU/ml; median (IQR)) 4.7 (3.3, 6.6) 5.1 (3.7, 6.9) 0.001 b 4.9 (3.4, 6.8) 5.0 (3.5, 6.6) 0.710 b bE 2 (pg/ml; mean±SD) 40.7±14.9 40.4±14.5 0.489 a 40.2±14.1 40.5±14.9 0.680 a Number of antral follicles (median (IQR)) 12 (8, 19) 13 (9, 18) 0.009 b 13 (8, 19) 13 (8, 19) 0.746 b Notes : Data are presented as numbers (%) unless otherwise noted. Data are presented as mean±SD, median (IQR) or n (%). a t -test for Equality of Means. b Independent-Samples Mann-Whitney U-Test. c X 2 -test. Abbreviation : IQR, interquartile range; BMI, body mass index; AMH, anti-müllerian hormone; FSH, follicle stimulating hormone; LH, luteinizing hormone; GnRH-a, gonadotropin releasing hormone agonist; GnRH-ant, gonadotropin releasing hormone antagonist; 2PN, double pronuclear fertilization; Gn, gonadotropin; IU, international units. Figure 1 The flow chart of study enrolment. Figure 2 Distribution of propensity score ( a and b ) and standardized difference ( c and d ) before and after matching for Puregon and Gonal-F groups. Baseline Characteristics Notes : Data are presented as numbers (%) unless otherwise noted. Data are presented as mean±SD, median (IQR) or n (%). a t -test for Equality of Means. b Independent-Samples Mann-Whitney U-Test. c X 2 -test. Abbreviation : IQR, interquartile range; BMI, body mass index; AMH, anti-müllerian hormone; FSH, follicle stimulating hormone; LH, luteinizing hormone; GnRH-a, gonadotropin releasing hormone agonist; GnRH-ant, gonadotropin releasing hormone antagonist; 2PN, double pronuclear fertilization; Gn, gonadotropin; IU, international units. The flow chart of study enrolment. Distribution of propensity score ( a and b ) and standardized difference ( c and d ) before and after matching for Puregon and Gonal-F groups. In Table 2 , after PSM, the total amount of Gn in Puregon group was significantly higher than that in Gonal-F group (2400 (1925, 3075) vs 2325 (1800, 3037.5), p = 0.038), however, the total cost of Gn usage in Puregon group was significantly lower (5327.9¥ (4273.4, 6812.5) vs 7547.2¥ (5843.0, 9860.1), p 0.05), as shown in Table 2 . Table 2 Ovarian Stimulation Outcomes Characteristic Before Propensity Score Matching After Propensity Score Matching Puregon (Group A) Gonal F (Group B) P-value Puregon (Group A) Gonal F (Group B) P-value Patients 1313 1551 1160 1160 Ovarian stimulation protocol <0.001 c 0.772 c  GnRH-a long protocol 961 (73.2) 1245 (80.3) 874 (75.3) 880 (75.9)  GnRH-ant protocol 352 (26.8) 306 (19.7) 286 (24.7) 280 (24.1) No of days of COS (mean±SD) 8.4±2.8 9.1±2.9 <0.001 a 8.6±2.7 8.6±2.7 0.741 a Total Gn dose administered (IU; median (IQR)) 2450 (1950, 3150) 2250 (1725, 2925) <0.001 b 2400 (1925, 3075) 2325 (1800, 3037.5) 0.038 b Total cost of Gn usage (RMB; median (IQR)) 5438.9 (4328.9, 6992.8) 7303.8 (5599.5, 9494.9) <0.001 b 5327.9 (4273.4, 6812.5) 7547.2 (5843.0, 9860.1) <0.001 b Serum LH on OPU (mIU/ml; median (IQR)) 1.4 (0.9,2.4) 1.4 (0.9,2.3) 0.928 b 1.4 (0.9, 2.3) 1.5 (0.9, 2.4) 0.609 b Serum E 2 on OPU (pg/ml; median (IQR)) 3715.0 (2180.2, 4839.0) 3941.0 (2364.0, 4856.0) 0.011 b 3773.0 (2304.0, 4839.0) 3840.5 (2279.5, 4856.0) 0.351 b Serum progesterone on OPU (ng/ml; median (IQR)) 1.2 (0.8, 1.7) 1.1 (0.8, 1.5) 0.001 b 1.2 (0.8, 1.7) 1.1 (0.8, 1.6) 0.162 b Ovulation trigger 0.656 c 0.590 c  hCG 1204 (91.7) 1415 (91.2) 1072 (92.4) 1065 (91.8)  GnRH-a and hCG (dual trigger) 109 (8.3) 136 (8.8) 88 (7.6) 95 (8.2) No of oocytes retrieved (median (IQR) 13 (8,19) 14 (9,20) 0.012 b 13 (9, 19) 13 (8, 20) 0.946 b Grade I embryos on D3 (median (IQR)) 1 (0, 2) 1 (0,2) 0.006 b 1 (0, 2) 1 (0, 2) 0.791 b Grade II embryos on D3 (median (IQR)) 2 (1,4) 3 (1,4) 0.002 b 3 (1, 4) 2 (1, 4) 0.923 b No of blastocysts available for transfer (median (IQR)) 2 (1, 3) 2 (1, 3) 0.400 b 2 (1, 3) 2 (1, 3) 0.919 b Total number of normally fertilized eggs (median (IQR)) 8 (4, 11) 8 (5, 12) 0.008 b 8 (5, 12) 8 (4, 12) 0.754 b Day of transfer 0.119 c 0.580 c  Day 3 377 (84.2) 394 (80.2) 319 (83.7) 324 (82.2)  Day 5 71 (15.8) 97 (19.8) 62 (16.3) 70 (17.8) No. of embryos transferred 0.204 c 0.148 c  SET 96 (21.4) 89 (18.1) 74 (19.4) 61 (15.5)  DET 352 (78.6) 402 (81.9) 307 (80.6) 333 (84.5) Ovarian stimulation cycle outcomes 0.162 c 0.567 c  Cycle with fresh embryo transfer 448 (34.1) 491 (31.7) 381 (32.8) 394 (34.0)  Cycle with freeze-all strategy 865 (65.9) 1060 (68.3) 779 (67.2) 766 (66.0) OHSS 58 (4.4%) 79 (5.1%) 0.398 49 (4.2%) 67 (5.3%) 0.086 Notes : Data are presented as mean±SD, median (IQR) or n (%). a t -test for Equality of Means. b Independent-Samples Mann-Whitney U-Test. C X 2 -test. Abbreviation : IQR, interquartile range; LH, luteinizing hormone; GnRH-a, gonadotropin releasing hormone agonist; GnRH-ant, gonadotropin releasing hormone antagonist; Gn, gonadotropin; IU, international units; hCG, human chorionic gonadotropin; SET, single embryo transfer; DET, double embryo transfer; OHSS, ovarian hyperstimulation syndrome. Ovarian Stimulation Outcomes Notes : Data are presented as mean±SD, median (IQR) or n (%). a t -test for Equality of Means. b Independent-Samples Mann-Whitney U-Test. C X 2 -test. Abbreviation : IQR, interquartile range; LH, luteinizing hormone; GnRH-a, gonadotropin releasing hormone agonist; GnRH-ant, gonadotropin releasing hormone antagonist; Gn, gonadotropin; IU, international units; hCG, human chorionic gonadotropin; SET, single embryo transfer; DET, double embryo transfer; OHSS, ovarian hyperstimulation syndrome. As shown in Table 3 , before and after PSM, the difference in pregnancy outcomes between the two groups of fresh embryo transfer was not detected. However, in frozen-thawed embryo transfer (FET) cycles, we found significantly reduced rates of positive pregnancy (64.3% vs 69.9%; P=0.007), clinical pregnancy (60.6% vs 65.5%; P=0.020), LBR (55.1% vs 60.4%; P=0.015) and CLBR (52.5% vs 57.8%; P=0.005) in the Puregon group compared with the Gonal-F group before matching. Nonetheless, the differences between the two groups, particularly for the primary outcome CLBR (52.8% vs 55.7%, P=0.169), were not statistically significant after matching. Table 3 Pregancy Outcomes Characteristic Before Propensity Score Matching After Propensity Score Matching Puregon (Group A) Gonal F (Group B) P-value Puregon (Group A) Gonal F (Group B) P-value Fresh ET 448 491 381 394 Positive pregnancy per ET 221 (49.3) 252 (51.3) 0.542 193 (50.7) 196 (49.7) 0.800 Clinical pregnancy rate per ET 207 (46.2) 235 (47.9) 0.612 178 (46.7) 182 (46.2) 0.883 Miscarriage rate per ET 43 (19.5) 39 (15.5) 0.254 40 (20.7) 32 (16.6) 0.296 Ectopic pregnancy rate per ET 8 (3.6) 13 (5.2) 0.418 8 (4.1) 10 (5.1) 0.653 Live birth rate per ET 170 (37.9) 200 (40.7) 0.383 145 (38.1) 154 (39.1) 0.769 FET 942 1146 848 833 Positive pregnancy rate per ET 606 (64.3) 801 (69.9) 0.007 543 (64.0) 565 (67.8) 0.101 Clinical pregnancy rate per ET 571 (60.6) 751 (65.5) 0.020 507 (59.8) 524 (62.9) 0.189 Miscarriage rate per ET 79 (13.0) 90 (11.2) 0.304 71 (13.1) 83 (14.7) 0.437 Ectopic pregnancy rate per ET 8 (1.3) 15 (1.9) 0.418 6 (1.1) 11 (1.9) 0.254 Live birth rate per ET 519 (55.1) 696 (60.4) 0.015 468 (55.2) 482 (57.9) 0.269 Cumulative live birth rate 689 (52.5) 896 (57.8) 0.005 613 (52.8) 646 (55.7) 0.169 Notes : Data are presented as numbers (%) unless otherwise noted. Abbreviations : ET, embryo transfer FET; frozen embryo transfer. Pregancy Outcomes Notes : Data are presented as numbers (%) unless otherwise noted. Abbreviations : ET, embryo transfer FET; frozen embryo transfer. As shown in Table 4 , multivariable logistic regression analysis, allowing adjustment for relevant confounders, revealed that female age, AMH, and number of oocytes retrieved were the only variables significantly associated with CLBR (OR 0.74, 95% CI 0.61–0.89, P=0.002; OR 1.60, 95% CI 1.08–2.37, P=0.019 and OR 1.47, 95% CI 1.10–1.96, P=0.008, respectively), while the type of Gn was not significantly associated with CLBR (OR 1.01; 95% CI 0.87–1.17; P = 0.912). Table 4 Multivariable Logistic Regression for Clbr Crude Model a Adjusted Model b OR (95% CI) P-value OR (95% CI) P-value Type of gonadotrophin  Gonal F (Follitropin β) Reference Reference Reference Reference  Puregon (Follitropin α) 0.98 (0.84–1.13) 0.745 1.01 (0.87–1.17) 0.912 Female age (< 37 yrs., ref) 0.70 (0.58–0.85) <0.001 0.74 (0.61–0.89) 0.002 BMI (< 28 kg/m 2 , ref) 0.82 (0.66–1.02) 0.079 0.84 (0.67–1.04) 0.115 AMH (<1.2ng/ml, ref) 2.09 (1.50–2.93) <0.001 1.60 (1.08–2.37) 0.019 Antral follicle count (AFC) 1.01 (1.00–01.02) 0.022 0.99 (0.98–1.01) 0.376 Basic FSH 0.96 (0.94–0.99) 0.005 0.99 (0.96–1.02) 0.504 Etiology of infertility  Tubal factor Reference Reference Reference Reference  Male factor 1.37 (0.93–2.04) 0.116 0.98 (0.62–1.57) 0.946  PCOS 1.38 (1.09–1.77) 0.013 0.74 (0.50–1.11) 0.146  Endometriosis 0.81 (0.50–1.31) 0.389 0.62 (0.33–1.17) 0.139 Ovarian stimulation protocol (GnRH-ant protocol, ref) 1.08 (0.91–1.29) 0.353 1.01 (0.85–1.22) 0.886 Ovulation trigger drugs (hCG, ref) 1.10 (0.84–1.43) 0.485 1.09 (0.96–1.22) 0.174 No of oocytes retrieved (< 5, ref) 1.80 (1.41–2.30) <0.001 1.47 (1.10–1.96) 0.008 Notes : a No adjustments for other covariates. b Adjusted for female age (<37 yrs., ≥37 yrs.), BMI (<28 kg/m 2 , ≥28 kg/m 2 ), AFC, AMH (<1.2ng/ml, ≥1.2ng/ml), basic FSH, etiology of infertility, ovarian stimulation protocol (GnRH-a long protocol, GnRH-ant protocol), ovulation trigger drugs (hCG, dual trigger), and number of oocytes retrieved (<5, ≥5). Abbreviations : BMI, body mass index; FSH, follicle stimulating hormone; PCOS, polycystic ovary syndrome; OR, odds ratio; CI, confidence interval. Multivariable Logistic Regression for Clbr Notes : a No adjustments for other covariates. b Adjusted for female age (<37 yrs., ≥37 yrs.), BMI (<28 kg/m 2 , ≥28 kg/m 2 ), AFC, AMH (<1.2ng/ml, ≥1.2ng/ml), basic FSH, etiology of infertility, ovarian stimulation protocol (GnRH-a long protocol, GnRH-ant protocol), ovulation trigger drugs (hCG, dual trigger), and number of oocytes retrieved (<5, ≥5). Abbreviations : BMI, body mass index; FSH, follicle stimulating hormone; PCOS, polycystic ovary syndrome; OR, odds ratio; CI, confidence interval.

A retrospective cohort study of 2864 IVF/ICSI patients at the Affiliated Hospital of Shandong University of Traditional Chinese Medicine (SDUTCM) from July 2015 to June 2021 was conducted. Reproductive Ethics Committee of the Affiliated Hospital of SDUTCM approved this study (No. SDUTCM-RME2021-0212). Informed consent was obtained from all subjects. Inclusion criteria: (1) age 21–45 at oocyte retrieval; (2) ovarian stimulation with GnRH agonist (GnRH-a) long protocol and GnRH antagonist (GnRH-ant) long protocol; (3) Patients undergoing their first oocyte retrieval cycle; (5) The endometrial preparation protocol for all subsequent frozen-thawed embryo transfer cycles was artificial cycle. Exclusion criteria: 1) A history of uterine abnormalities, either congenital or acquired; 2) patients undergoing blastocyst biopsy for preimplantation genetic testing (PGT); 3) failure to transfer all embryos following one egg retrieval cycle; 4) patients who underwent in vitro maturation (IVM); 5) embryos derived from previous ovarian stimulation cycles (ie, cryopreserved oocytes and/or donor oocytes); 6) incomplete case and follow-up data. All patients were divided into two groups according to the ovarian stimulation drugs. A total of 1313 patients who received Puregon injections (Group A) and 1551 patients who received Gonal-F injections (Group B) were included in the ovarian stimulation cycles. For ovarian stimulation, all participants underwent either the pituitary down-regulated GnRH-a protocol or a GnRH-ant protocol comprising daily injections of r-FSH (Gonal-F or Puregon). The initial daily dose of r-FSH (112.5–300U) was determined by patients’ age, anti-Mullerian hormone (AMH), basic antral follicle count (AFC), body mass index (BMI), and basic serum FSH. Follicle size and hormone levels were monitored during ovarian stimulation, which led to adjustments in the dose. Transvaginal ultrasound (TVUS) is essential for monitoring the ovarian response and serum level of steroid hormones, which allows adjusting the dosage of Gn and promoting oocyte maturation. After hCG administration or GnRH-a combined hCG administration, cumulus oocyte complexes were collected by TVUS-guided needle aspiration 36–37 hours later depending on the size of follicles (at least two dominant follicles of ≥ 18mm diameter) and serum estradiol levels (over 300 pg/ml). Following ovum collection, oocytes are fertilized with conventional insemination (standard IVF) or ICSI, as indicated. Following insemination, the presence of two clearly distinct pronuclei containing nuclei was evaluated as a sign of normal fertilization 16–18 hours later. Under ultrasound supervision, embryo transfer was routine. Patient age, previous failed attempts, and embryo quality were considered when determining the number of embryos to be transferred. Cryopreservation of surplus embryos and blastocysts of high quality was performed using the fast freezing procedure described previously. 14 Embryos that reached at least the six-cell stage on Day 3 with less than 20% fragmentation were considered to be of good quality. AA, AB, BA or BB quality trophectoderm and inner cell mass scores were required for Day 5 blastocysts to be considered good quality. Regardless of the developmental stage or number of embryos transferred, our center consistently adheres to the principle of transferring one high-quality Day 3 embryo, two suboptimal Day 3 embryos, or one high-quality Day 5 blastocyst. 15–17 Until the 10th gestational week, intramuscular progesterone (20mg, Zhejiang Xianju Pharmaceutical Co., Ltd.) was administered at a daily dose of 40mg or 90mg of vaginal progesterone (8% Crinone; Merck-Serono), with oral progesterone (20mg, Dydrogesterone, Abbott Biologicals B.V.) twice daily. LPS was discontinued without pregnancy. In this study, the primary outcome was the CLBR, which was calculated as a percentage of the number of patients who delivered for the first time at 24 weeks or more during an ovarian stimulation cycle (including fresh and frozen ETs) divided by the number of patients undergoing all oocyte retrieval cycles. Following fresh ETs, the secondary outcome measures included LBR, positive pregnancy rate, clinical pregnancy rate, ectopic pregnancy rate, and miscarriage rate. Serum β-HCG was measured 14 days after embryo transfer. By TVUS, clinical pregnancy is defined as an intrauterine gestational sac with a fetal heartbeat detected after 7 weeks; an ectopic pregnancy is defined as the development of the fertilized egg outside of the uterus; miscarriage is defined as termination of pregnancy at less than 28 weeks of gestation with a fetus weighing less than 1000g; a live birth occurs at a minimum of 24 weeks or 500 grams. SPSS 26.0 (SPSS Inc., Chicago, IL, USA) was used to analyze all data. Data normality was assessed using the Shapiro-Wilk test. A student t -test or a Mann-Whitney U-test, as appropriate, was used to analyze the quantitative variables based on their normality. Frequency and percentage were used to express qualitative variables. The χ 2 or Fisher’s exact test was performed to compare categorical variables. For all statistical tests, a two-tailed probability of 0.05 was applied. Moreover, PSM adjustments were made to account for potential confounding factors and selection biases. Using the Logit model, the personal propensity score (PS) was calculated. A nearest-neighbor random matching algorithm was used to pair the Puregon group with the Gonal-F group in a ratio of 1:1. In the PS, the caliper width was 0.01 standard deviation. With CLBR as the dependent variable, multivariable logistic regression analysis was conducted for ovarian stimulation medication type (Follitropin α versus Follitropin β) as the primary independent variable. In the analysis, female age, BMI, AFC, AMH, basic FSH, ovulation trigger drugs, and number of retrieved oocytes were considered as potential contributing factors. A logistic regression model was constructed simultaneously for all variables. An odds ratio (OR) with a 95% confidence interval is presented for CLBR following IVF/ICSI.

In conclusion, the Gonal-F and Puregon employed in ovarian stimulation protocols results in similar CLBRs for infertile patients undergoing IVF/ICSI. Patients should be informed that Gonal-F is a suitable option for patients who wish to decrease Gn dosage of injections in daily clinical practice. Puregon can allow patients and IVF clinics to minimize unneeded stress and expenses while also providing them more flexibility. Larger studies, as well as an economic assessment of the costs involved, are needed to validate the current findings.

To the best of our knowledge, only a few studies have compared the CLBR of r-FSH in infertile patients undergoing IVF/ICSI. 18–20 We expanded the sample size to include more patients and compared CLBR, pregnancy outcomes, ovarian stimulation and embryo laboratory outcomes of Puregon and Gonal-F groups in IVF/ICSI patients. The results showed that there was no significant difference in the CLBR between the two groups. It is nevertheless interesting to note that despite similar outcomes, fewer amounts of Gn were used in the Gonal-F group, and reduced funds were spent on Gn in the Puregon group. The main goal of ART is to achieve healthy live births. Although various research have indicated the CLBR as a major endpoint, 18 , 21 , 22 both studies on the two groups of recombinant human FSH mentioned above were implemented prior to 2014. 23 , 24 Statistics on the cumulative probability of pregnancy and live birth are extremely beneficial to patients, however cumulative data on the likelihood of live birth are insufficient at present. 25 The objective of ovarian stimulation with r-FSH preparations is to produce numerous oocytes to permit in vitro fertilization and, eventually, a live baby. As a result, we opted to adopt the CLBR as the main endpoint in this trial, therefore providing additional data support for future studies. In our study, there was no significant differences between the two groups in CLBR. In a retrospective study, lien et al evaluated CLBR after ovarian stimulation with Ovaleap (Follitropin α) and Puregon (Follitropin β), respectively, in ICSI patients undergoing a GnRH-ant protocol. The results of this study did not detect a statistically significant difference in CLBR between the two groups, which is consistent with our findings. 18 Several randomized controlled trials (RCTs) have been conducted to evaluate the impact of various r-FSH dosages on IVF/ICSI treatment outcomes, and researchers have identified a dose-effect relationship between the total dose of Follitropin β and the number of oocytes retrieved. 26 Raoul Orvieto and colleagues found that r-FSH β and r-FSH α resulted in comparable number of oocytes retrieved after ovarian stimulation, but r-FSH β tended to require less dosage, apparently with greater recruitment of follicle capacity. 27 This conclusion seems to contradict our findings. In this large retrospective study, we found no statistically significant difference in the number of oocytes retrieved between the two groups, but the Gonal-F group used less total dosage than the Puregon group. Therefore, Gonal-F may be a friendlier choice for patients who are afraid of receiving injections. With the development of ART, treatment selection is not only based on drug efficacy and safety, but also cost considerations, which are determined in part by drug dosage. 28 Economic benefits have always been an important measure for patients and clinicians. A pharmacoeconomic evaluation of IVF-ET based on the Dutch healthcare system found that Follitropin α may be a cost-effective option compared to u-FSH and HP-HMG. 29 Because of differences in cost, the decision to use Follitropin α or Follitropin β to stimulate the ovaries in low income families may have a significant impact on economic stress and treatment discontinuation rates. 30 In our present study, Gn days between two groups after PSM was no statistically significant difference, while the total dose of Gonal-F was less than that of Puregon. However, the cost of live birth based on Gonal-F therapy was found to be higher than that of Puregon (7547.2 RMB and 5327.9 RMB, respectively). Thus, Puregon has better cost-effectiveness to some extent, more meaningful to relieve economic pressure of patients, especially those with low income. Compared with the previous retrospective study, our advantage lies in expanding the sample size, increasing the reliability and persuasiveness of the results of this study. And considering the effects of confounding factors and selection bias, we established a 1:1 PS matching to limit bias and obtain relatively reliable data. Furthermore, the economic and psychological burden of economic cost-effectiveness on patients was considered, as was the detailed information of patients missing from previous studies, such as duration of infertility, gravidity, parity, basal serum reproductive endocrine hormones, AFC, AMH, and so on. This will act as a guidance for IVF-ET clinical medication selection. Despite efforts to control selection bias, there are limitations of retrospective studies. This study did not assess the usage or handling faults of both pen injectors, nor did it provide additional practical information on self-injection. An easy-to-use pen-based device can reduce patient stress, as handling errors can lead to overdose or underdose, as well as adverse events or delays in treatment.