Recombinant follicle-stimulating hormone and recombinant luteinizing ...

Download PDF
13 downloads 835 Views 268KB Size Report
Comment
Mar 3, 2015 - ... and Gynaecology, University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh City, Vietnam 2IVFAS, ... School of Medicine, Vietnam National University HCMC. ...... Reprod Biomed Online 2004;8:635–643.
Human Reproduction, Vol.30, No.5 pp. 1188 –1195, 2015 Advanced Access publication on March 3, 2015 doi:10.1093/humrep/dev038

ORIGINAL ARTICLE Reproductive endocrinology

Recombinant follicle-stimulating hormone and recombinant luteinizing hormone versus recombinant follicle-stimulating hormone alone during GnRH antagonist ovarian stimulation in patients aged ≥35 years: a randomized controlled trial T.N.L. Vuong 1,2,*, H.T. Phung 2, and M.T. Ho2,3 1 Department of Obstetrics and Gynaecology, University of Medicine and Pharmacy at Ho Chi Minh City, Ho Chi Minh City, Vietnam 2IVFAS, An Sinh Hospital, Ho Chi Minh City, Vietnam 3Research Center for Genetics and Reproductive Health, School of Medicine, Vietnam National University HCMC, Ho Chi Minh City, Vietnam

*Correspondence address. Department of Obstetrics and Gynaecology, University of Medicine and Pharmacy at Ho Chi Minh City, 217 Hong Bang Street, District 5, Ho Chi Minh City, Vietnam. Tel: +84-903008889; E-mail: [email protected]

Submitted on October 17, 2014; resubmitted on January 19, 2015; accepted on February 9, 2015

study question: Does luteinizing hormone (LH) supplementation improve live birth rate after in vitro fertilization (IVF) in patients aged ≥35 years receiving a gonadotrophin-releasing hormone (GnRH) antagonist protocol? summaryanswer: There was no difference in live birth rate with use of LH during IVF in patients aged ≥35 years undergoing IVF treatment using a GnRH antagonist protocol. what is known already: Use of GnRH analogues as part of a controlled ovarian hyperstimulation protocol during IVF treatment cycles decreases the amount of LH available to developing follicles. The role of LH supplementation for improving outcomes in patients undergoing controlled ovarian hyperstimulation as part of assisted reproduction treatments, particularly those involving a GnRH antagonist protocol, is unclear. It has been suggested that higher risk patients (e.g. age ≥35 years, poor ovarian reserve) may benefit from LH supplementation.

study design, size, duration: This single-centre, randomized controlled trial was conducted from 1 October 2012 to 30 June 2014. A total of 240 women aged ≥35 years undergoing IVF received ovarian stimulation using a GnRH antagonist protocol, with recombinant folliclestimulating hormone (r-FSH; Gonal-Fw) starting from cycle day 2 or 3. GnRH antagonist (Cetrotidew) was administered on Day 5 of r-FSH administration. On Day 6, patients in the LH supplementation group were switched to r-FSH/r-LH (Pergoverisw) 150/75 IU/day. Randomization to study treatments was performed in blocks of 4 via a computer-generated random number list. participants/materials, setting, methods: Of the 240 patients randomized to treatment, 120 received r-FSH/r-LH and 120 received r-FSH. Patients were recruited from the IVFAS, An Sinh Hospital, Ho Chi Minh City, Vietnam. main results and the role of chance: Live birth rate did not differ significantly (P . 0.05) between r-FSH/r-LH and r-FSH recipients (16.7 versus 17.5%; between-group difference 0.8, 95% confidence interval [CI] 29.5, 11.2). In addition, there were no significant differences between the r-FSH/r-LH and r-FSH groups with respect to the number of oocytes retrieved, implantation rate, miscarriage rate and clinical pregnancy rate.

limitations, reasons for caution: The open-label design could have introduced bias, and the relatively small sample size may have allowed detection of only the most common adverse events. In addition, the study was likely to be underpowered based on differences between the response rate assumptions used in the sample size calculation and the actual response rate during the study.

& The Author 2015. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected]

r-FSH/r-LH versus r-FSH in IVF patients aged ≥35 years

1189

wider implications of the findings: The results of this study found no additional benefit from adding LH supplementation to ovarian stimulation with a GnRH antagonist protocol in women aged ≥35 years, and add to the body of evidence in this area. However, findings across studies are still inconsistent and additional research is needed before any clear recommendations for clinical practice can be made.

study funding/competing interests: This study was supported by the Research Center for Genetics and Reproductive Health, School of Medicine, Vietnam National University HCMC. The authors state that they have no financial or commercial conflicts of interest. trial registration number: The trial was registered with clinicaltrials.gov (NCT02244866). Key words: luteinizing hormone supplementation / recombinant follicle-stimulating hormone / recombinant luteinizing hormone / ovarian stimulation / in vitro fertilization

Introduction Ovarian stimulation plays an important role in achieving optimal follicular and endometrium development and improving outcomes in in vitro fertilization (IVF) treatment cycles. However, ovarian stimulation protocols in IVF inhibit endogenous luteinizing hormone (LH) release. LH, along with follicle-stimulating hormone (FSH), is fundamental to normal follicular development (Mochtar et al., 2007). LH also plays an important role in oocyte maturation during normal menstrual cycles (Hillier, 2001). Approximately 10 –12% of patients do not respond to currently-used ovarian stimulation protocols, and this is thought to be due to lack of LH (Hill et al., 2012). As a result, LH supplementation during ovarian stimulation for IVF is necessary and is part of the treatment protocol at many IVF centres. However, there is still debate about which specific patient groups might benefit from LH supplementation and the optimal treatment regimen (Ferraretti et al., 2004). According to a consensus on LH supplementation among IVF experts in the Asia-Pacific region (Wong et al., 2011), patients for whom a benefit of LH supplementation has been confirmed include those with central ovarian failure (group 1 with no ovulation according to WHO classification), a history of poor response to ovarian stimulation (,4 oocytes with standard ovarian stimulation, minimum FSH dose of 300 IU/day), and suboptimal ovarian response in the current treatment cycle (no follicles .10 mm, estradiol ,200 pg/ml and endometrium ,6 mm on stimulation day 6). Futhermore, it has been noted that patients aged ≥35 years may obtain benefit from LH supplementation. This group is at risk of poor or suboptimal ovarian response due to reduced ovarian reserve. Moreover, LH bioactivity often falls below the threshold in patients aged ≥35 years (Wong et al., 2011). There is limited data on the use of LH supplementation in patients aged .35 years, particularly in combination with a gonadotrophinreleasing hormone (GnRH) antagonist protocol, and most studies only report ongoing pregnancy rate, providing insufficient evidence to support routine LH supplementation in these patients. Published studies to date have reported conflicting results. Some studies have reported better implantation and/or ongoing pregnancy rates when patients aged .35 years had LH supplementation added to the ovarian stimulation protocol (Humaidan et al., 2004; Marrs et al., 2004; Matorras et al., 2009; Bosch et al., 2011). In contrast, Ko¨nig et al. (2013) and Fa´bregues et al. (2006) concluded that LH supplementation during the second half of the follicular phase had no effect on ovarian response, implantation rate, and ongoing pregnancy rate in IVF recipients aged ≥35 years. In one meta-analysis, the live birth rate was not affected by LH supplementation, both overall and in patients aged

.35 years (Kolibianakis et al., 2007), whereas others have reported that r-LH may improve clinical pregnancy rates overall (Lehert et al., 2014), in women aged ≥35 years (Hill et al., 2012) or in those with poor ovarian reserve (Lehert et al., 2014). In Vietnam, LH supplementation in IVF has been utilized since 2004, primarily for patients with poor response and suboptimal ovarian response in the treatment cycle. A randomized controlled clinical trial with recombinant LH supplementation in ovarian stimulation protocols for patients with suboptimal ovarian response reported significantly improved clinical pregnancy rates in patients who received supplementation compared with those who did not (Vuong et al., 2004). Local data suggest that the proportion of patients receiving IVF treatment who are aged ≥35 years is quite large (23.7%). Given that age has long been identified as a critical factor in response to controlled ovarian hyperstimulation (DeCherney and Berkowitz, 1982; Templeton et al., 1996), it is clear that strategies to improve the success rate in patients aged ≥35 years undergoing IVF are required. One such option might be LH supplementation as part of the ovarian stimulation protocol for women aged ≥35 years, although further studies of this approach are required to better define efficacy. On the basis that existing data on the effects of adding LH supplementation during IVF in older patients are conflicting and the limited amount of information available on the live birth rate in this setting, the current study was designed to compare the effects of LH supplementation (recombinant follicle-stimulating hormone [r-FSH]/r-LH) with non-LH supplementation (r-FSH) on live birth rate and other IVF outcomes in patients aged ≥35 years undergoing IVF treatment using a GnRH antagonist protocol for ovarian stimulation.

Materials and Methods This randomized controlled trial (NCT02244866) was conducted from 1 October 2012 to 30 June 2014 at IVFAS, An Sinh Hospital, Ho Chi Minh City, Vietnam. The end-of-study period was defined as a negative pregnancy test according to routine clinical practice, clinical pregnancy confirmed by ultrasound scan performed 6 – 7 weeks after embryo transfer, or delivery of a baby (for patients who had ongoing pregnancy).

Ethical approval The Institutional Review Board (IRB) and Ethics Committee approved the study protocol on 12 September 2012 (IRB reference number: 10/ QD-CGRH-NCKH and DT TP.HCM). All patients provided written informed consent to participate in the study, which was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice.

1190

Study population All patients undergoing routine assisted cycles during the trial period were invited to participate, and all had basic investigations before ovarian stimulation so that their eligibility for enrolment in the study could be determined. This included medical history, anti-Mu¨llerian hormone (AMH) measurement, antral follicle count (AFC) and gynaecological ultrasound. To be eligible for enrolment, subjects had to be starting treatment with r-FSH according to the decision of the investigator and in accordance with the indication and recommendations detailed in the drug data sheet, aged ≥35 years at the time of r-FSH dosing, have a body mass index (BMI) ,28 kg/m2, had ≤3 previous IVF attempts, receiving a GnRH antagonist protocol, and willing and able to comply with the protocol requirements for the duration of the study. Patients were excluded from the study if they were already participating in another interventional clinical trial, had polycystic ovary syndrome (PCOS), were WHO group 1, had uterine abnormalities such as uterine bicornuate, uterine cavity adhesion, and/or had endocrine disorders such as hyperprolactinaemia and thyroid disorders.

Vuong et al.

Luteal phase support was provided with vaginal progesterone (Crinonew 8% gel, Merck-Serono, Germany), 2 applicators/day. A pregnancy test was done 14 days after embryo transfer. A beta hCG level of .5 mIU/ml was considered positive. Fetal ultrasound was performed 3 weeks after a positive pregnancy test.

Outcome measures The primary end-point of the study was live birth rate. Secondary end-points were clinical pregnancy rate, embryo implantation rate, miscarriage rate, duration of stimulation, total number of r-FSH units used, estradiol concentrations on the hCG-administered day, endometrial thickness on the hCG-administered day, premature LH surge rate (.10 IU/l), number of oocytes retrieved, number of embryos, number of good embryos, and the number of patients with a premature rise in progesterone (.1.5 ng/ml) on the day of hCG administration, ovarian hyperstimulation syndrome or cycle cancellation due to poor response (,3 follicles grown ≥8 days after ovarian stimulation injection).

Statistical analysis Sample size and randomization Based on a 19.1% difference in the clinical pregnancy rate with and without LH supplementation reported in a previous study of LH supplementation in poor responders (Vuong et al., 2004), it was determined that 109 patients in each group would be required to detect a significant difference between treatments with 90% power and a 2-sided significance level of 0.05. The recruitment target was 120 subjects per group (240 in total) to allow for dropouts. Eligible subjects were randomized using computer-generated random number list in blocks of 4 to either the r-FSH/r-LH or r-FSH treatment arm; treatment allocation was generated by an administration and provided to study staff in a sealed envelope. A physician enrolled each patient and then, after the patient had signed the informed consent form, a treatment group was assigned by a midwife who had opened the sealed envelope.

IVF protocol and LH supplementation Ovarian stimulation was performed by using a GnRH antagonist protocol; r-FSH (Gonal Fw, Merck-Serono, Germany) was administered on Day 2 or Day 3 of the menstrual cycle. The first r-FSH dose was individualized for each patient based on the following criteria: AFC ≤6, dose 300 IU/day; AFC 7 – 15, dose 225 IU/day; and AFC ≥16, dose 150 IU/day. GnRH antagonist (Cetrotidew, Merck-Serono, Germany) was administered on Day 5 of r-FSH administration. r-LH was supplemented from Day 6 of r-FSH administration; patients in the r-FSH/r-LH group were switched to r-FSH/r-LH (Pergoverisw; Merck Serono, Germany) 150/75 IU/day and the r-FSH group continued to receive r-FSH at a dosage determined by the treating physician. In both groups, the r-FSH dose could be titrated based on physician judgment if the number of follicles was inappropriate (,3 follicles of ≥8 mm on Day 6 of stimulation). The cost of the r-FSH/r-LH and r-FSH preparations used in the study was the same; therefore there was no additional cost to patients for the addition of r-LH. Monitoring of follicular development was performed using a vaginal ultrasound probe, and quantifying LH, estradiol and progesterone, starting from Day 5 of r-FSH, based on the current clinical practice at the hospital. Ovum pick-up (OPU) was performed 36 h after administration of hCG 250 mg (Ovitrelle, Merck-Serono, Germany). Embryos were transferred on Day 2. Embryo transfer was cancelled if patients had ovarian hyperstimulation syndrome (OHSS) (Mathur et al. 2005), an unfavourable endometrium (endometrial thickness of ≤6 mm or ≥16 mm on ultrasound, fluid in cavity or endometrial polyp), progesterone level ≥1.5 ng/ml on the day of hCG administration, or no embryo; in addition, some patients did not want to undergo transfer of fresh embryos.

Data collection was conducted using questionnaires, and data processing was done using SPSS 19.0 (IBM SPSS Statistics, IBM Corporation, USA). All randomized patients were included in the intent-to-treat analysis. Shapiro– Wilk test of normality was used to test for normal distribution of quantity variables. Values are reported as median (quartile 1 [Q1], quartile 3 [Q3]) for nonnormally distributed variables, mean + standard deviation (SD) for normally distributed variables or percentage. Between-group comparisons used parametric tests for normally distributed variables and non-parametric tests for not normally distributed variables. Comparison of two rates was made using the Chi-square test. Comparison of two means used Tukey’s method. Comparison of median values used bootstrapping with 1000 replications. Multivariate analysis of logit model was conducted to evaluate the association between live birth and risk factors such as age, BMI, number of treatment cycles, AMH and AFC. Univariate analysis of logit model was conducted to determine the change in live birth rate as age increased. A P-value of ,0.05 was considered statistically significant.

Results A total of 240 patients were enrolled in the study, of whom 120 patients were assigned to the r-FSH/r-LH group and 120 to the r-FSH group. A Consolidated Standards of Reporting Trials (CONSORT) flow diagram is shown in Fig. 1. Baseline characteristics of patients in the two treatment groups were similar (Table I). Ovarian stimulation characteristics were essentially similar in the two treatment groups, except for cycle cancellation due to poor response, which occurred more often in r-FSH versus r-FSH/ r-LH recipients (Table II).

Live birth rate The live birth rate did not differ significantly between the r-FSH/r-LH and r-FSH groups (Table III). Multivariate analysis did not identify any significant associations between live birth and risk factors such as age, BMI, number of treatment cycles, AMH and AFC. Post hoc analysis showed that for each 1-year increase in age, the live birth rate reduced by 15% (odds ratio 0.85, 95% CI 0.74, 0.89; P ¼ 0.026).

Ovarian stimulation and embryo transfer There were no statistically significant differences between the r-FSH/ r-LH and r-FSH groups with respect to secondary study end-points,

r-FSH/r-LH versus r-FSH in IVF patients aged ≥35 years

1191

Figure 1 Consolidated Standards of Reporting Trials (CONSORT) diagram showing flow of patients through the study. OHSS, ovarian hyperstimulation syndrome.

including number of oocytes retrieved, embryos, good embryos and cryopreserved embryos (Table III). Embryo transfer outcomes were also similar in the two treatment groups (Table III).

Side effects No side effects were reported during the study and there were no cases of ovarian hyperstimulation syndrome (OHSS).

Discussion This study showed that LH supplementation during IVF stimulation using a GnRH antagonist protocol had no significant effect on the primary outcome, live birth rate, in women aged ≥35 years. In addition, no

benefits of LH supplementation were observed with respect to the number of oocytes retrieved, implantation rate and clinical pregnancy rate. LH supplementation in ovarian stimulation remains a controversial issue. Currently, most IVF centres prescribe different forms of LH supplementation (e.g. r-LH, human menopausal gonadotrophins [hMG] or human chorionic gonadotrophin [hCG]) for specific patient subgroups, depending on whether or not they personally consider that the small amounts of LH present after down-regulation are sufficient to sustain theca and granulosa cell stimulation. As a result, the addition of LH is often based more on physicians’ experience or opinion than highquality published evidence. For women aged ≥35 years, the results of this study showed no overall improvement in implantation and clinical pregnancy rates with the

1192

Vuong et al.

addition of LH to GnRH antagonist ovarian stimulation during IVF. These findings are similar to a number of published studies conducted in the same older patient population (Marrs et al., 2004; Fa´bregues et al., 2006; Ko¨nig et al., 2013). In addition, the conclusion of one systematic review was that available evidence did not support the addition of r-LH both overall and in the subgroup of women .35 years of age (Kolibianakis et al., 2007). Conversely, recent systematic reviews of r-LH supplementation in any women undergoing IVF have suggested that r-FSH/ r-LH improves the clinical pregnancy rate compared with r-FSH overall (risk ratio [RR] 1.09; 95% CI 1.01– 1.18) (Lehert et al., 2014), in women aged ≥35 years (OR 1.37, 95% CI 1.03 –1.83) (Hill et al., 2012) and in those with poor ovarian response (OR 1.3, 95% CI 1.05 –1.62) (Lehert et al., 2014). Some individual studies have also

Table I Baseline patient characteristics and demographic data. r-FSH (n 5 120)

r-FSH/r-LH (n 5 120)

........................................................................................ Age, years

38 (36, 40)

Number of treatment cycles BMI, kg/m2

38 (36, 40)

1 (1, 2)

1 (1, 2)

19.2 (17.9, 20.2)

19.2 (18.2, 20.7)

AMH, ng/ml

1.6 (1.0, 3.2)

1.6 (1.2, 2.8)

AFC

6.0 (4.0, 7.7)

7.0 (5.0, 10.2)

Cycles with reduced ovarian reservea, n (%)

59 (49.2)

49 (40.8)

Values are median (Q1, Q3), or percent patients. Statistical comparisons were made using Wilcoxon rank sum test, except for number of cycles with reduced ovarian reserve, which was compared using Pearson’s Chi-squared test. a Reduced ovarian reserve is defined as AMH ≤1.1 ng/ml or AFC ,6. AFC, antral follicle count; AMH, anti-Mu¨llerian hormone; BMI, body mass index; r-FSH, recombinant follicle-stimulating hormone; Q1, quartile 1; Q3 quartile 3; r-LH, recombinant luteinizing hormone.

reported results that conflict with those of the current study in women aged ≥35 years (Humaidan et al., 2004; Matorras et al., 2009). One factor that may have contributed to different results in the meta-analysis by Lehert and colleagues is that they included non-randomized studies as well as randomized clinical trials. Live birth rate, the primary end-point in our study, has not been reported in many studies, particularly those focusing on older patients. Meta-analysis findings for the IVF population as a whole found that only 8 studies reported live birth rate, while rates for an additional 31 studies were imputed based on available data (Lehert et al., 2014). Based on these calculations, which included non-randomized studies, the live birth rate RR value for r-FSH/r-LH versus r-FSH in the ITT population was 1.11 (95% CI 1.01 –1.21). This contrasts with the lack of significant difference between treatment groups in the current study. Absolute values for clinical pregnancy rate in the current study (22.5% in both groups) are slightly lower, but correspond well with those reported in a similar studies of patients aged ≥35 years (28 and 30% for r-FSH/r-LH and r-FSH, respectively) (Ko¨nig et al., 2013). In a subgroup of patients aged 36 –39 years in another recent study (Bosch et al., 2011), the ongoing pregnancy rate was higher, although not significantly so, in r-FSH/r-LH versus r-FSH recipients (33.5 versus 25.3%), and absolute values are relatively close to those reported in the other two studies. There are number of possible reasons for the differences between our findings and those of some other studies and systematic reviews. We initiated LH in the second half of the follicular phase, based on a previously studied r-LH supplementation protocol for patients with poor response to ovarian stimulation at our institution (Vuong et al., 2004), and similar to Fa´bregues et al. (2006) who also reported no influence of r-LH supplementation on pregnancy outcomes in women aged ≥35 years. Ko¨nig et al. also used the same timing of r-LH administration in their study and reported similar clinical pregnancy rates: 22.5% in both groups in the current study, and 28 and 30% for r-FSH/r-LH and r-FSH, respectively, in the Ko¨nig study. There was some evidence of a

Table II Ovarian stimulation characteristics (ITT population). r-FSH

r-FSH/r-LH

............................................ ............................................ n or n/N

Median (Q1, Q3) or percent

n or n/N

Between-group difference (95% CI)

P-value

Median (Q1, Q3) or percent

............................................................................................................................................................................................. Duration of stimulation, days Total FSH dose, IU

116 116

10 (9, 11) 2550 (2025, 3000)

114 114

10 (9, 11) 2700 (2175, 3000)

0 (21, 0)c

0.156a c

0.147a

c

297 (2553, 272)

0.304a

0 (21.0, 0.5)c

0.620a

2150 (2300, 150)

Estradiol concentration on day of hCG administration, pg/ml

102

1529 (973, 3241)

109

1632 (1119, 4108)

Endometrial thickness on day of hCG administration, mm

102

11 (10, 12)

109

11 (9, 13)

Cycles with premature progesterone rise

9/120

7.5%

7/120

5.8%

1.7 (25.5, 8.8)

0.604b

Cycles with premature LH surge

2/120

1.7%

1/120

0.8%

0.8 (22.8, 4.4)

0.561b

Cycle cancellation due to poor Response*

14/120

11.7%

5/120

4.2%

7.5 (0.0, 15.1)

0.031b

CI, confidence interval; FSH, follicle-stimulating hormone; hCG, human chorionic gonadotrophin; ITT, intent-to-treat; LH, luteinizing hormone; Q1, quartile 1; Q3 quartile 3; r, recombinant. *Includes cycles cancelled ≥3 days after r-LH was started (Day 8 of stimulation). a Wilcoxon rank sum test. b Pearson’s Chi-squared test. c Bootstrap method with 1000 replications.

r-FSH/r-LH versus r-FSH in IVF patients aged ≥35 years

1193

Table III Ovarian stimulation and embryo transfer outcomes (ITT population). r-FSH

....................................... n or n/N

Value

r-FSH/r-LH

.......................................

n or n/N

Between-group difference (95% CI)

P-value

Value

............................................................................................................................................................................................. Ovarian stimulation Oocytes, n Embryos, n Good embryos Embryos cryopreserved

102 102 101

8 (5, 11) 4 (3, 6) 1 (1, 2)

109

7 (4, 11)

109

3 (2, 6)

107

1 (0, 2.5)

1 (21, 3)c

0.516a

c

0.321a

c

0.519a

1 (21, 2)

0 (21, 1) c

96

0 (0, 2)

106

0 (0, 2)

0 (0, 0)

No. transferred

93

2 (2, 3)

100

2 (1, 3)

0 (21.0, 1.0)c

Clinical pregnancy rate

27/120

22.5%

27/120

22.5%

0.0 (210.6, 10.6)

Implantation rate

93

11.13 + 20.42

100

12.47 + 34.63

Miscarriage rate

6/120

5.0%

7/120

Ovarian hyperstimulation

0/120

0.0%

Live birth rate

21/120

17.5%

0.866a

Embryo transfer 0.272a 1.0b

21.34 (29.48, 6.81)

0.746d

5.8%

21.8 (28.0, 4.7)

0.554b

0/120

0.0%

0.0 (0.0, 0.0)

–

20/120

16.7%

0.8 (29.5, 11.2)

0.864b

Values are reported as median (Q1, Q3), mean + standard deviation (implantation rate only) or percent patients. CI, confidence interval; ITT, intent-to-treat; No., number; Q1, quartile 1; Q3 quartile 3, r-FSH, recombinant follicle-stimulating hormone; r-LH, recombinant luteinizing hormone. a Wilcoxon rank sum test; b Pearson’s Chi-squared test; c Bootstrap method with 1000 replications; d Tukey’s method.

nonsignificant benefit for LH supplementation in a subgroup of older patients in the Bosch study, which initiated r-FSH/r-LH treatment starting on cycle day 1. Younger women (age 20 –36 years) have been shown to produce higher quality embryos with addition of r-LH to r-FSH on the first day of a GnRH antagonist stimulation protocol (Wiser et al., 2011). However, administration of r-LH later in the cycle (Day 6– 8) has also been associated with higher pregnancy or birth rates during r-LH supplementation in subgroups of patients aged ≥35 years in some studies (Humaidan et al., 2004; Marrs et al., 2004; Matorras et al., 2009). In addition, subgroup data from the Kolibianakis meta-analysis showed that the live birth rate was similar in r-FSH/r-LH versus r-FSH recipients irrespective of the time r-LH was added during the follicular phase (Kolibianakis et al., 2007), casting doubt on whether this factor had an influence on our results. GnRH antagonists are increasingly being used for ovarian stimulation because this approach effectively suppresses gonadotrophin release while avoiding an LH surge. It is possible that the category of GnRH analogue used for inhibition of the premature LH surge could influence the comparative effects of r-FSH/r-LH and r-FSH. The majority of studies investigating the value of adding LH supplementation have used GnRH agonist long protocols (Marrs et al., 2004; Fa´bregues et al., 2006; Matorras et al., 2009). In the current study, a GnRH antagonist protocol was used and no effect of LH supplementation on live birth rate was documented. In the setting of GnRH antagonist use, the findings of Bosch et al. (2011) on the effects of adding r-LH supplementation in older women were in contrast to the current study, with a clinically-relevant increase in ongoing pregnancy rate reported. Furthermore, meta-analysis data do not support the proposition that the type of GnRH analogue used has an influence on live birth rate during r-FSH/r-LH or r-FSH treatment (Kolibianakis et al., 2007). Another potential explanation for the differences in findings between studies is ethnic differences between the enrolled populations; previous

data were obtained in European-based cohorts whereas our study included Asian patients who have been shown to have worse outcomes after assisted reproductive technologies compared with Caucasian women (Purcell et al., 2007). The success of IVF procedures decreases as age increases, and there is a marked reduction in pregnancy rates from the age of 35 years onwards. The general aging process contributes to this decline, but reductions in natural levels of LH during the ovarian stimulation process may also have a role. Baseline data show that the patients in our study had relatively poor prognosis. Nearly half the population (46.3%) were aged ≥38 years, the proportion of patients with reduced ovarian reserve based on AMH (Ferraretti et al., 2011) or AFC (Vuong et al., 2013) criteria was 32.5%, and 21.7% of patients had both negative risk factors. Finally, the lack of difference in pregnancy outcomes between the two treatment groups in our study could be because low LH levels during ovarian stimulation may in fact not be detrimental for IVF outcome, something that has been suggested by other researchers (Bosch et al., 2005; Ramachandran et al., 2014). The results of this study suggest a number of important topics for future research, including continued investigation to more clearly define the role of supplementation with r-FSH/r-LH in patients aged ≥35 years, poor ovarian reserve, or both; and investigation of whether amending the protocol to initiate LH supplementation on Day 1 of the ovarian stimulation instead of Day 6 as in this study has any impact on its efficacy. Such studies would help to better answer the question as to whether LH supplementation is beneficial to higher risk patients, who are increasingly presenting for assisted reproductive therapies and present an important challenge to clinicians. This study has a number of strengths. Firstly, it has a randomized design, improving the robustness of the findings. Secondly, it fills a gap in the literature because there are relatively few published studies investigating LH supplementation in combination with a GnRH antagonist

1194 ovarian stimulation protocol (most have been done using a GnRH agonist long protocol), or in patients aged ≥35 years. In addition, it provides data specific to Asian patients, a group that have lower ovarian response compared with Caucasian women (Purcell et al., 2007) who comprise the majority of subjects in other clinical trials. Another important feature of this trial compared with other studies investigating the same question, is that we reported live birth rate rather than just the ongoing pregnancy rate. There are also a number of limitations to this study. The open-label design could have introduced bias into the results, and the relatively small sample size means that it may have only been possible to detect the most common adverse events. Perhaps the most important issue is that the study is probably underpowered. The sample size calculation was based on clinical pregnancy rate data from a previous study conducted in a similar patient population (Vuong et al., 2004). This was deemed appropriate because Asian data were thought to more readily apply to our Asian patient population. However, the 19.1% between-group difference in pregnancy rate in that study was quite large and probably overestimated the expected beneficial effects of LH supplementation. The actual between-group difference observed in the current study mean that the power is substantially less than that calculated when planning the study. In addition, it would have been preferable to calculate sample size based on live birth rate, but the absence of existing data on this end-point in a relevant (i.e. Asian) population meant that this was not possible. In conclusion, this study found no overall difference in live birth rate, clinical pregnancy rate or other ovarian stimulation outcomes with LH supplementation in women aged ≥35 years undergoing controlled ovarian stimulation for IVF with a GnRH antagonist protocol.

Acknowledgements The authors would like to thank Nicola Ryan, independent medical writer for her English medical writing assistance and Duc HT, MD, National Hospital of Can Tho & Ton Duc Thang University, Vietnam for his invaluable help in data analysis.

Authors’ roles T.N.L.V. was involved in study design, execution, analysis, manuscript drafting, critical discussion and final approval of the manuscript. H.T.P. and M.T.H. were involved in study design, execution, critical discussion and final approval of the manuscript.

Funding The study’s administrative work was supported by the Research Center for Genetics and Reproductive Health, School of Medicine Vietnam National University HCMC.

Conflict of interest The authors state that they have no financial or commercial conflicts of interest to report.

Vuong et al.

References Bosch E, Escudero E, Crespo J, Simo´n C, Remohı´ J, Pellicer A. Serum luteinizing hormone in patients undergoing ovarian stimulation with gonadotropinreleasing hormone antagonists and recombinant follicle-stimulating hormone and its relationship with cycle outcome. Fertil Steril 2005; 84:1529–1532. Bosch E, Labarta E, Crespo J, Simo´n C, Remohı´ J, Pellicer A. Impact of luteinizing hormone administration on gonadotropin-releasing hormone antagonist cycles: an age-adjusted analysis. Fertil Steril 2011;95:1031–1036. DeCherney AH, Berkowitz GS. Female fecundity and age. N Engl J Med 1982; 306:424– 426. Fa´bregues F, Creus M, Pen˜arrubia J, Manau D, Vanrell JA, Balasch J. Effects of recombinant human luteinizing hormone supplementation on ovarian stimulation and the implantation rate in down-regulated women of advanced reproductive age. Fertil Steril 2006;85:925 – 931. Ferraretti AP, Gianaroli L, Magli MC, D’angelo A, Farfali V, Montanaro N. Exogenous luteinizing hormone in controlled ovarian hyperstimulation for assisted reproductive techniques. Fertil Steril 2004;82:1251– 1526. Ferraretti AP, La Marca A, Fauser BCJM, Tarlatzis B, Nargund G, Gianaroli L, ESHRE working group on Poor Ovarian Response Definition. ESHRE consensus on the definition of ‘poor response’ to ovarian stimulation for in vitro fertilization: the Bologna criteria. Hum Reprod 2011;26: 1616 – 1624. Hill MJ, Levens ED, Levy G, Ryan ME, Csokmay JM, DeCherney AH, Whitcomb BW. The use of recombinant luteinizing hormone in patients undergoing assisted reproductive techniques with advanced reproductive age: a systematic review and meta-analysis. Fertil Steril 2012;97:1108–1114. Hillier SG. Gonadotropic control of ovarian follicular growth and development. Mol Cell Endocrinol 2001;179:39– 46. Humaidan P, Bungum M, Bungum L, Ydling Andersen C. Effects of recombinant LH supplementation in women undergoing assisted reproduction with GnRH agonist down-regulation and stimulation with recombinant FSH; an opening study. Reprod Biomed Online 2004;8:635–643. Kolibianakis EM, Kalogeropoulou L, Griesinger G, Papanikolaou EG, Papadimas J, Bontis J, Tarlatzis BC. Among patients treated with FSH and GnRH analogues for in vitro fertilization, is the addition of recombinant LH associated with the probability of live birth? A systematic review and meta-analysis. Hum Reprod Update 2007;13:445–452. Ko¨nig TE, van der Houwen LEE, Overbeek A, Hendriks ML, BeutlerBeemsterboer SN, Kuchenbecker WKH, Renckens CNM, Bernardus RE, Schats R, Homburg R et al. Recombinant LH supplementation to a standard GnRH antagonist protocol in women of 35 years or older undergoing IVF/ICSI: a randomized controlled multicentre study. Hum Reprod 2013;28:2804–2812. Lehert P, Kolibianakis EM, Venetis CA, Schertz J, Saunders H, Arriagada P, Copt S, Tarlatzis B. Recombinant human follicle-stimulating hormone (r-hFSH) plus recombinant luteinizing hormone versus r-hFSH alone for ovarian stimulation during assisted reproductive technology: systematic review and meta-analysis. Reprod Biol Endocrinol 2014;12:17. Marrs R, Meldrum D, Muasher S, Schoolcraft W, Werlin L, Kelly E. Randomized trial to compare the effect of recombinant human FSH (follitropin alfa) with or without recombinant human LH in women undergoing assisted reproduction treatment. Reprod Biomed Online 2004;8:175 – 182. Mathur R, Evbuomwan I, Jenkins J. Prevention and management of ovarian hyperstimulation syndrome. Curr Obstet Gynaecol 2005;15:132 – 138. Matorras R, Prieto B, Exposito A, Mendoza R, Crisol L, Herranz P, Burgue´s S. Mid-follicular LH supplementation in women aged 35 – 39 years undergoing ICSI cycles: a randomized controlled study. Reprod Biomed Online 2009;19:879 – 887.

r-FSH/r-LH versus r-FSH in IVF patients aged ≥35 years

Mochtar MH, Van der Veen F, Ziech M, van Wely M. Recombinant luteinizing hormone (rLH) for controlled ovarian hyperstimulation in assisted reproductive cycles. Cochrane Database Syst Rev 2007; 2:CD005070. Purcell K, Schembri M, Frazier LM, Rall MJ, Shen S, Croughan M, Grainger DA, Fujimoto VY. Asian ethnicity is associated with reduced pregnancy outcomes after assisted reproductive technology. Fertil Steril 2007; 87:297– 302. Ramachandran A, Jamdade K, Kumar P, Adiga SK, Bhat RG, Ferrao SR. Is there a need for luteinizing hormone (LH) estimation in patients undergoing ovarian stimulation with gonadotropin-releasing hormone (GnRH) antagonists and recombinant follicle-stimulating hormone (rFSH)? J Clin Diagn Res 2014;8:90 – 92. Templeton A, Morris JK, Parslow W. Factors that affect outcome of in-vitro fertilisation treatment. Lancet 1996;348:1402 – 1406.

1195 Vuong TNL, Ho MT, Pham VT, Nguyen TNP. The addition of recombinant LH in poor-responder IVF patients [abstract]. Presented at 18th World Congress on Fertility and Sterility, May 23 –28, 2004; Montreal, Quebec, Canada (Abstract 307.6). Vuong TNL, Nguyen KL, Ho MT, Wong PC, Howles CM. Anti-Mu¨llerian hormone versus antral follicle stimulating count for defining the starting dose of FSH. Reprod Biomed Online 2013;27:390 – 399. Wiser A, Hourvitz A, Yinon Y, Levron J, Dor J, Elizur SE. Recombinant human luteinizing hormone supplementation may improve embryo quality in in vitro fertilization/intracytoplasmic sperm injection cycles with gonadotropinreleasing hormone antagonist protocol. Open J Obstet Gynecol 2011;1:31–35. Wong PC, Qiao J, Ho C, Ramaraju GA, Wiweko B, Takehara Y, Nadkarni PV, Cheng LC, Chen HF, Suwajanakorn S et al. Current opinion on use of luteinizing hormone supplementation in assisted reproduction therapy: an Asian perspective. Reprod Biomed Online 2011;23:81– 90.