Granulocyte colony-stimulating factor priming improves embryos and pregnancy rate in patients with poor ovarian reserve: a randomized controlled trial

Study design

Our prospective open-label randomized clinical trial investigated whether G-CSF priming preceding ART enhanced preantral follicle growth, thus increasing the ART pregnancy rate in patients with poor ovarian reserve. Between May 19, 2014 and November 26, 2018, a total of 465 patients sought ART treatment at Women’s Clinic Jinno; 111 met study inclusion criteria. Eleven declined participation, leaving 100 to be enrolled and randomly assigned to groups undergoing or not undergoing G-CSF priming prior to standard ART. Randomization involved patients drawing from a box containing group assignments in sealed envelopes mixed 1:1 (Fig. 1). Neither patients nor investigators were blinded to resulting assignments.

Fig. 1figure 1

Participant flow diagram. The rate of live delivery among patients was 32% (16/50) in the G-CSF group, significantly higher than 14% (7/49) in controls (chi-squared test). G-CSF, granulocyte colony-stimulating factor; OPU, oocyte pick-up; IVF, in vitro fertilization; ICSI, intracytoplasmic sperm injection; ET, embryo transfer. a All 3 of these patients conceived with spontaneous ovulation following menstruation after the initial administration of G-CSF

To more clearly detect G-CSF-related improvement of ovarian reserve, we limited our study to patients with mildly to moderately decreased ovarian reserve, excluding patients with severe diminution. We chose a serum AMH concentration lower than 2 ng/mL as an inclusion criterion for our study, given that 93% of infertile women in their forties have been found to have values below 2 ng/mL [5]. An earlier study found 1.9 ng/mL to be the median serum AMH concentration among nulliparous 38-year-old Japanese women, a population on the verge of a precipitous decline in fertility [6].

Based on such considerations, our inclusion criteria were age between 20 and 42 years; no more than 1 prior oocyte retrieval attempt; serum AMH concentration below 2 ng/mL; day-3 serum FSH concentration below 30 IU/L; a medical history free of serious allergic disease, severe hepatic, renal, or heart disease, or uterine infertility; and a male partner without azoospermia.

All enrolled patients initially received conventional infertility treatments in 2 consecutive cycles. (Since no patient in either group had bilateral tubal occlusion, we believed that a chance of pregnancy using conventional treatment existed.) During these 2 cycles the G-CSF group, but not the control group, underwent subcutaneous administration of G-CSF (100 μg of lenograstim; Neutrogin, Chyugai Pharmaceuticals, Tokyo, Japan) during the early luteal phase, based upon basal body temperature records, vaginal ultrasonographic findings, and, if necessary, serum progesterone determinations (Fig. 1). Controls received no placebo. Conventional infertility treatments included sexual intercourse or intrauterine insemination with or without ovarian stimulation by clomiphene citrate or a recombinant FSH regimen.

Both groups consisted of patients who had failed to conceive with conventional infertility treatments and then underwent ART using controlled ovarian stimulation. Embryos were transferred 2, 3, or 5 days after in vitro fertilization (IVF) or intracytoplasmic sperm injection (ICSI). Remaining embryos were cryopreserved at the blastocyst stage. When IVF/ICSI and fresh embryo transfer (ET) failed to result in delivery, cryopreserved embryos were thawed and transferred (cryopreserved ET) in subsequent spontaneous cycles. During actual ART treatment using fresh or cryopreserved ET, neither group received G-CSF.

Clinical pregnancy and ongoing pregnancy were diagnosed by ultrasonographic detection of a gestational sac and fetal heart movements, respectively. Abortion and delivery were defined as pregnancy loss before 22 weeks and birth after 22 weeks, respectively. Clinical pregnancy rates by IVF/ICSI-fresh ET (per ovarian stimulation) were compared between groups as the primary outcome measure. Cumulative live delivery rates, follicular growth, fertilization, and embryonic development were compared between groups as secondary outcome measures.

To elucidate stimulatory effects of G-CSF priming on preantral follicle growth, pre- vs. post-treatment changes in serum AMH concentration were compared between groups. Serum samples were obtained within 4 months preceding study enrollment and just before initiating ovarian stimulation for ART (Fig. 1).

To achieve a power of 0.8 and an α error of 0.05, the minimum number of participants required to identify a difference between hypothetical pregnancy rates of 10% and 25% in control and G-CSF groups was 113 patients per group, so we planned to recruit those numbers. However, interim assessment halfway through the expected trial duration already showed statistically significant benefit from G-CSF priming that was greater than expected. We therefore ended our trial with 50 enrollees per group, considering the ethical importance of offering potential benefits of G-CSF priming to control patients in a timely manner.

Informed consent was obtained from all patients. The study was approved by the Ethics Committees of both the Inagi Municipal Hospital and Women’s Clinic Jinno, and registered with the UMIN in Japan (UMIN000013956).

Assisted reproductive technology

Follicular development was stimulated with the long protocol as described previously [7]. Briefly, buserelin acetate (Buserecur; Fuji Pharmaceuticals, Tokyo, Japan) at 900 µg per day was administered nasally from the mid-luteal phase until hCG administration. Human menopausal gonadotropin, (hMG, 300 IU i.m.; Ferring Pharmaceuticals, Tokyo, Japan) was administered daily from day 3. Human chorionic gonadotropin (hCG, 10 000 IU i.m.; Mochida Pharmaceuticals, Tokyo, Japan) was administered when the dominant follicle reached a diameter of 19 mm.

Oocytes were collected transvaginally 36 h after hCG administration and inseminated as described previously [8]. ICSI was performed when the male partner had severe infertility (sperm count < 5 × 106 per mL and/or motility < 20%). Oocytes were considered fertilized when 2 pronuclei were observed 17 to 19 h after insemination or ICSI. At 2, 3, or 5 days after oocyte retrieval, embryos were transferred to the uterus according to number and quality of developing embryos for each patient. Progesterone (25 mg i.m.) was administered daily after ET.

Redundant embryos were cultured for 5 to 6 days after IVF/ICSI to the blastocyst stage and cryopreserved using a vitrification method. Thawed cryopreserved blastocysts were transferred to uteri on luteal day 5 of a spontaneous natural cycle, as described previously [7]. For luteal support, 5000 IU of hCG was administered on luteal days 5, 7, and 9.

Evaluations of hormones and killer-cell immunoglobulin-like receptor genotypes

Serum concentrations of AMH, follicle-stimulating hormone (FSH), luteinizing hormone (LH), 17β-estradiol (E2), prolactin (PRL), testosterone (T), thyroid-stimulating hormone (TSH), free thyronine (FT3), free thyroxine (FT4), and dehydroepiandrosterone sulfate (DHEA-S) were measured by enzyme chemiluminescent immunoassays on cycle day 3 within 3 months before study enrollment.

Sensitivities and intra- and interassay coefficients of variation were 0.01 ng/mL (1.4%, 0.8%) for AMH, 0.06 IU/L (2.3%, 1.0%) for FSH, 0.11 IU/L (6.6%, 3.4%) for LH, 5.0 pg/mL (0.7%, 0.9%) for E2, 0.10 ng/mL (1.2%, 1.4%) for PRL, 0.03 ng/mL (2.5%, 3.9%) for T, and 2 μg/dL (6.5%, 2.7%) for DHEA-S.

DNA was genotyped for 16 KIR genes using PCR-SSOP (sequence-specific oligonucleotide probe) using a commercial kit (LABType KIR SSO Genotyping Test; One Lambda, Canoga Park, CA, USA) and Luminex 100 technology (Austin, TX, USA) as previously described [9].

Statistical analysis

IBM SPSS Statistics Version 27 (IBM, Tokyo, Japan) was used for statistical analyses. Normality was tested by the Shapiro–Wilk test. If data were not normally distributed, analysis was performed using the Mann–Whitney U test or the Wilcoxon matched-pairs signed rank test as appropriate. If data were normally distributed, unpaired t tests or paired t tests were performed as appropriate. Data also were analyzed using the chi-squared test, Fisher’s exact test, or multiple logistic regression analysis as appropriate. P values below 0.05 were considered to indicate significance. Whenever appropriate, results are presented as the mean ± standard deviation (SD).

留言 (0)

沒有登入
gif