Current opinion on use of luteinizing hormone supplementation in assisted reproduction therapy: an Asian perspective

Transcription

1 Reproductive BioMedicine Online (2011) 23, REVIEW Current opinion on use of luteinizing hormone supplementation in assisted reproduction therapy: an Asian perspective Peng Cheang Wong a, *, Jie Qiao b, Clement Ho c, Gottumukkala A Ramaraju d, Budi Wiweko e, Yuji Takehara f, Prashant V Nadkarni g, Li-Chang Cheng h, Hsin-Fu Chen i, Somchai Suwajanakorn j, Thi Ngoc Lan Vuong k, for the Asia Pacific Fertility Advisory Group a Department of Obstetrics and Gynaecology, National University Hospital, Singapore; b Reproductive Medicine Center, Peking University Third Hospital, Beijing, China; c Assisted Reproductive Technology (ART) Laboratory, Hong Kong Special Administrative Region; d Krishna IVF Clinic, Maharanipeta, Visakhapatnam, India; e Department of Obstetrics and Gynecology, Faculty of Medicine, University of Indonesia Dr. Cipto Mangunkusumo General Hospital, Jakarta, Indonesia; f Advanced Medical Research Institute of Fertility, Kato Ladies Clinic, Tokyo, Japan; g KL Fertility and Gynaecology Centre, Kuala Lumpur, Malaysia; h Thomson Fertility Centre, Thomson Medical Centre, Singapore; i Department of Obstetrics/ Gynecology, National Taiwan University Hospital, Taipei, Taiwan; j Department of Obstetrics and Gynecology, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand; k Department of Obstetrics and Gynaecology, University of Medicine and Pharmacy, Ho Chi Minh City, Viet Nam * Corresponding author. address: peng_cheang_wong@nuhs.edu.sg (PC Wong). Peng Cheang Wong is Professor in the Department of Obstetrics and Gynaecology at the Yong Loo Lin School of Medicine, National University of Singapore and Senior Consultant and Head in the Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynaecology, National University Hospital (NUH). Prof Wong is also Director of the Assisted Reproductive Technologies programme in NUH where he is actively involved in postgraduate education. He sits on the editorial board of the Journal of Paediatrics, Obstetrics and Gynaecology and is a reviewer for journals such as the Annals of Academy of Medicine, Singapore and the Singapore Medical Journal. Abstract LH and FSH have complementary functions in ensuring optimal oocyte maturation and ovulation. In women undergoing assisted reproduction technology protocols with gonadotrophin-releasing hormone analogues, LH and FSH concentrations are reduced. While FSH use in assisted reproduction technology is well established, there is no published consensus on the need for exogenous LH in Asian patients. Having reviewed the concept of the LH therapeutic window and differences between recombinant human LH (r-hlh) and human menopausal gonadotrophin, a consensus was reached on which patient subgroups may benefit from LH supplementation. Adjuvant r-hlh gives clinicians precise control over the dose of LH bioactivity administered to target the therapeutic window. The use of r-hlh is recommended in women with poor response in a previous cycle or suboptimal follicular progression in a current cycle by day 6 8 of stimulation. r-hlh should also be considered in women at risk of suboptimal response, specifically age >35 years. Other risk markers that suggest the need for LH supplementation, which include baseline/day-6 serum /$ - see front matter ª 2011, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. doi: /j.rbmo

2 82 PC Wong et al. LH and anti-müllerian hormone concentrations, antral follicle count and LH polymorphisms require further research and verification. For measurement of LH response adequacy, the monitoring of follicular progression, oestradiol concentrations and endometrial thickness is recommended. RBMOnline ª 2011, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: Asia, GnRH agonist, IVF, ovarian stimulation, recombinant LH Introduction The most commonly used protocol in assisted reproductive technology consists of daily injections of recombinant human FSH (r-hfsh) to induce multiple follicle growth in the ovaries, a process called ovarian stimulation. This treatment is often combined with daily injections of a gonadotrophin-releasing hormone (GnRH) agonist or antagonist to prevent a premature LH surge. The pituitary down-regulation that is achieved with agonists or the block of GnRH receptors by antagonists at mid-cycle means that the growing follicles are developing in an environment of very low or even no LH. It has been shown that growing follicles become increasingly sensitive to, and ultimately dependent on, the presence of LH for their development (Shoham, 2002). Some assisted reproduction practitioners thus advocate add-back LH to the mid-follicular phase in ovarian stimulation cycles. Other practitioners deem add-back LH to be unnecessary, justifying that the small amounts of LH present after down-regulation are sufficient to sustain theca and granulosa cell stimulation. Several studies have attempted to answer the question of whether adding LH to ovarian stimulation improves assisted reproduction results; a Cochrane review on this subject was published recently (Mochtar et al., 2009). The primary rationale for convening a working group to review the available data on possible adjunctive treatments that could improve ovarian response to assisted reproduction technology was that the majority of the data published on the benefits of adding LH to ovarian stimulation do not originate in Asia. Furthermore, in the Asian context, it has become increasingly clear that the average age of couples seeking assisted reproduction technology is rising. A survey of the group s members indicated that more than 50% of their patients are above age 35. It is therefore imperative to further customize the stimulation protocols currently being used in the region according to the individual needs of the Asian population. The first adjunctive treatment considered was adjuvant recombinant human LH (r-hlh) in addition to ovarian stimulation. In an attempt to reach an Asian consensus and to provide guidance on specific patient populations where there is good evidence for the combination of LH and FSH, a review of the recent studies was undertaken on the use of LH in assisted reproduction technology, including some non-randomized controlled trials (RCT) omitted from the Cochrane review. The current study decided on a consensus rather than a meta-analysis because current published studies comparing r-hfsh versus r-hlh in ovarian stimulation are either too heterogenous or too few in number to conduct an adequate meta-analysis. It was estimated that to detect a 5% difference in clinical pregnancy with 90% power would require more than 3000 patients. For these reasons, this study focused on drafting a consensus statement as a suitable way to provide some guidance to clinicians in Asia. Physiology As an oocyte matures within a follicle, a myriad of complex hormonal events take place, particularly during the follicular phase. The two gonadotrophins LH and FSH play distinct but complementary roles in ensuring follicular growth and ovulation. These two gonadotrophins are active in the final weeks of the development of a mature surviving oocyte, with initiation of folliculogenesis occurring independently of gonadotrophic stimulation (McGee and Hsueh, 2000). Secretion of LH and FSH in an orderly fashion is therefore crucial to the final development of a mature oocyte that can result in pregnancy. The ovarian thecal and granulosa cells are the principal sites of LH bioactivity, although LH receptors are also present in extra-gonadal sites such as the uterus. The only cells in the female body known to have FSH receptors are granulosa cells (McGee and Hsueh, 2000). According to the two-cell, two-gonadotrophin theory, only FSH is essential for triggering antral follicle formation and follicular growth, but some LH is essential in the pre-antral stage (follicle size <10 mm) to stimulate secretion of androgens by thecal cells (Shoham, 2002). The androgens diffuse into the granulosa cells where the aromatase enzyme system augmented by FSH converts them to oestrogen (Falck, 1959; Shoham, 2002). This synergism between LH and FSH is thus essential for steroidogenesis and to develop the subsequent capacity of the follicle to ovulate and luteinize when exposed to the mid-cycle LH surge. By around days 7 9 (follicle diameter about mm), granulosa cells stimulated by the effect of FSH begin developing LH receptors in preparation for the final stages of follicle maturation (Hillier, 2009a). As such, LH plays an increasingly important role after day 6 in regulating the final stages of oocyte maturation. The LH surge (days 13 14) induces resumption of meiosis I in the oocyte, early luteinization of the granulosa and theca cells, initiation of the synthesis of progesterone and the production of prostaglandins within the follicle (Berger and Taymor, 1971). These two substances are essential to allow rupture of the follicular wall and eventual liberation of the oocyte about h after the LH surge. The role of r-hlh and the LH therapeutic window r-hfsh replacement has been used in ovarian stimulation protocols since its approval in 1994 (Devroey et al., 1994).

3 Luteinizing hormone supplementation: an Asian perspective 83 There is increasing interest in exploring the role of LH and its optimal use in relation to assisted reproduction. LH in association with FSH is recommended for stimulating follicular development in women with severe luteinizing LH and FSH deficiency. In clinical trials, these patients were defined by an endogenous serum LH concentration <1.2 IU/l (Humaidan et al., 2002; O Dea et al., 2008; Shoham, 2002). According to the threshold theory of LH in ovarian function (Hillier et al., 1994), the ovarian follicle requires a minimal amount of LH for steroidogenesis (<1% of receptors attached by LH). There is also a ceiling beyond which excessively high concentrations of LH may actually suppress granulosa aromatase activity and inhibit cell growth (Loumaye et al., 2003). The LH ceiling is dependent on timing of the menstrual cycle but for optimal follicle development, this concentration is typically 1.2 IU/l and 5 IU/l (O Dea et al., 2008). In the clinical situation, the LH therapeutic window is best observed in two patient groups where there may be a severe endogenous deficiency of LH. Hypogonadotrophic hypogonadism (hypo-hypo) patients are the first and probably the most extensively studied group of patients with an endogenous severe LH and FSH deficiency. The use of r-hlh in ovarian follicular development is well supported by extensive studies with excellent safety and efficacy and is standard treatment for clinicians (Bosch, 2009; European r-hlh Study Group, 1998). There is a second group of patients whose endogenous LH secretion is profoundly suppressed with GnRH analogues (agonists or antagonists) during ovarian stimulation who may develop severe endogenous LH deficiency. The role of r-hlh in this group of patients needs further evaluation. Severe endogenous LH deficiency Women with hypo-hypo have impaired pituitary neuroendocrine function that results in abnormally low LH and FSH concentrations. Such patients are typically amenorrhoeic and have small ovaries with follicles in arrested development due to the lack of effective hypothalamic pituitary activity. These women do not have sufficient endogenous LH for optimal follicular growth and steroidogenesis when treated with FSH alone (Couzinet et al., 1988) and will typically benefit from FSH and LH for optimal follicular development (European r-hlh Study Group, 1998). Hypophysectomized animal studies and studies in hypo-hypo women confirmed that r-hfsh was able to increase follicular growth in a dose-dependent manner, but was ineffective in stimulating synthesis of oestradiol due to the extremely low or near-undetectable endogenous LH concentrations (Mannaerts et al., 1991; Shoham et al., 1993). Shoham (2002) also extensively reviewed the role of LH in hypophysectomized women and in women with isolated gonadotrophin deficiency and concluded that although r-hfsh increased the growth of multiple ovarian follicles, serum LH, androstenedione and oestradiol concentrations remained low or showed only negligible increase. The reviews confirmed that LH is physiologically essential for oestradiol synthesis and that replacement to above the minimal threshold concentration will be necessary in women with inherent gonadotrophin insufficiency. Severe LH deficiency due to suppression by GnRH analogues in assisted reproduction technology Women treated with GnRH analogues (agonists or antagonist), for example, during ovarian stimulation in IVF may similarly experience severely reduced LH and FSH concentrations due to oversuppression of endogenous LH and FSH pituitary secretion. In some patients, the suppressed baseline LH concentrations may reach those described in studies in hypo-hypo women (i.e. <1.2 IU/l). However, for women undergoing ovarian stimulation, not every patient requires exogenous LH replacement, as endogenous production of small amounts of LH may remain sufficient for thecal cell function. In selected patients whose endogenous LH is low after GnRH agonist treatment, reports have documented poorer outcomes among those patients who have a lower LH concentration or a sharper fall in LH from baseline concentrations (Humaidan et al., 2002; Lahoud et al., 2006; Nakagawa et al., 2008). A retrospective analysis (n = 200) found that normogonadotrophic women undergoing long GnRH agonist protocol IVF cycles and treated with r-hfsh were five times more likely to suffer early pregnancy loss if LH serum concentrations on stimulation day 8 were below 0.5 IU/l (P < 0.005) (Westergaard et al., 2000). This was further supported by a recent two-treatment arm RCT (Pezzuto et al., 2010) that compared r-hfsh versus r-hfsh combined with r-hlh, in a long agonist assisted reproduction technology cohort with day-6 LH concentrations <0.5 IU/l. There were no differences between the groups in the number of oocytes retrieved (6.37 ± 2.67 versus 7.32 ± 1.99, respectively); however a significantly higher number of mature oocytes were obtained from the group receiving r-hlh (136 versus 93, P < 0.05) and fertilized oocytes (92% versus 69%, P < 0.001). Clinical pregnancy rate was 5% for r-hfsh alone compared with 22% with r-hfsh plus r-hlh (P < 0.05). However, given the small number of patients in the Pezzuto study, it is concluded that this is an area where further research is needed to identify whether LH concentrations are useful markers for LH deficiency. This conclusion was also reached by Alviggi et al. (2006). There may be a role for other biomarkers, for example, LH polymorphisms such as the LH beta variant (Alviggi et al., 2009) in further identifying LH-deficient patients. Another way to identify the severely LH-deficient patient is to look at the ovarian response to FSH stimulation. As described earlier, a certain minimum LH concentration is necessary for adequate thecal cell function and subsequent oestradiol synthesis in the granulosa cells. The consequent rise in oestradiol concentration is essential for endometrial proliferation and corpus luteum formation in anticipation of a fertilized oocyte, implantation and embryo development in pregnancy. It therefore stands to reason that LH concentrations that are too low will increase the likelihood of unsuccessful implantation or early pregnancy loss (Shoham, 2002). Supplementation of LH may benefit selected women with LH deficiency and suboptimal ovarian response during assisted reproduction technology as measured by clinical end-points such as oestradiol concentrations and follicular development. While it may be tempting to supplement all patients with LH to reap its benefits, it is critical to take note of the

4 84 PC Wong et al. Table 1 Below LH threshold (<1.2 IU/l) Impaired follicular development Concept of LH therapeutic window. Inadequate thecal androgen synthesis and hence reduced granulosa aromatization to oestrogen No full oocyte maturation Within LH therapeutic window (1.2 5 IU/l) Optimal follicular growth and development Full oocyte maturation Above LH ceiling (>5 IU/l) LH receptor down-regulation Suppression of granulosa cell proliferation Follicular atresia (non-dominant follicles) Premature luteinization (preovulatory follicle) Adapted from Balasch and Fábreques (2002) and O Dea et al. (2008). ceiling effect of LH supplementation. Early overexposure of LH in ovarian stimulation can result in premature follicle luteinization of small follicles and follicular atresia leading either to cycle cancellation due to follicle maturation arrest or to poor-quality oocytes, all of which translates into severely compromised outcomes. Table 1 summarizes the impact of LH concentrations within and outside the therapeutic window. Exogenous LH supplementation: molecular and functional differences between recombinant human LH and human menopausal and chorionic gonadotrophins The concept of an LH therapeutic window has implications for the choice of LH supplementation in assisted reproduction technology. Endogenous LH production is pulsatile and occurs in response to the pulsatile release of GnRH from the hypothalamus. The most physiological way to maintain LH concentrations is to utilize endogenous LH secretion, and this explains why some clinicians use a short GnRH agonist or microflare protocol in some of their older or poor-responder patients (Loutradis et al., 2008). If an exogenous source of LH is needed, physicians have a choice of either urinary human menopausal gonadotrophin (HMG) or r-hlh. Since LH has a precise therapeutic window, the physician needs to have precise control over the activity of exogenous LH administered. r-hlh is analogous to endogenous LH and characterized by high purity, precision of dosing and consistency. When administered by subcutaneous injection, r-hlh has a terminal half-life of 24 h (le Cotonnec et al., 1998) and exhibits modest accumulation with an accumulation ratio of 1.6 ± 0.8. Human menopausal and chorionic gonadotrophins Until recently, the only available source of exogenous LH activity has been HMG, a urine-derived preparation containing both FSH and LH, which comprises about 5% of the total protein content. Despite significant improvements in processing, different HMG preparations are subject to wide variation in LH quantity and bioactivity and with increased purification, more LH is lost. For this reason, human chorionic gonadotrophin (HCG) is often added in an attempt to boost the LH bioactivity to meet the required LH activity range (FSH:LH ratio = 1:1) as stated in the pharmacopoeia. This may result in the end-product having much more HCG than LH activity. The Van Hell bioassay of HMG only detects LH bioactivity and does not distinguish LH from HCG. Analysis of one HMG product (Menopur) showed that the content of HCG was more than 10 times higher than LH (Baer and Loumaye, 2003). Another analysis (van de Weijer et al., 2003) also showed that about 95% of the LH receptor bioactivity in one HMG product (Menopur) was attributed to HCG, with less than 5% contributed by pure LH. It should be noted that HCG is not normally present in women except during pregnancy and malignancy. Functional and molecular differences between HCG and LH HCG has higher binding affinity and longer half-life HCG has a higher binding affinity to the LH receptor, with 1 IU HCG having biological activity approximately 6 8 times greater than 1 IU LH. Given by subcutaneous injection, HCG exhibits a longer serum half-life (30 h), thus leading to the possibility of significant accumulation over time (Filicori et al., 2002). In their paper comparing the addition of 50 IU of HCG per day from day 8 of stimulation, Filicori et al. documented the accumulation of HCG leading to HCG concentrations of 10.2 ± 3 IU/dl in patients given 50 IU HCG/day from day 8 of stimulation. In the same paper, while demonstrating that he could substitute FSH with 50, 100 or 200 IU of HCG activity from day 8, Filicori also noted that nine out of the 30 patients treated with HCG had a negative treatment outcome, i.e. they did not achieve ovulation (Filicori et al., 2002). HCG/HMG may induce LH receptor internalization A study showed that, in rats injected with HCG, the ovaries showed LH receptor down-regulation lasting up to 72 h, just as prolonged GnRH agonist stimulation leads to downregulation (Menon et al., 2006). Another recent study comparing stimulation with HMG versus r-hfsh in women undergoing IVF/intracytoplasmic sperm injection treatment also found statistically significantly reduced expression of LH receptor messenger RNA in ovarian granulosa cells in the HMG group, which was associated with altered expression of genes and proteins involved in steroidogenesis in preovulatory granulosa cells (Grøndahl et al., 2009). These studies show that HCG is not equivalent to LH and there are effects at the level of LH receptor internalization, which may explain why there is tolerance or a lack of effect. Finally, it is known that there is a LH ceiling associated with the use of r-hlh. Hugues et al. (2005) established

5 Luteinizing hormone supplementation: an Asian perspective 85 Table 2 Evidence supporting recommendations guiding use of recombinant human LH in assisted reproduction patient subpopulations. Variables Prior poor response Ongoing suboptimal response At risk of suboptimal response Age >35 years Day-6 LH <0.5 IU/l GnRH long agonist Lisi et al. (2003) De Placido et al. (2004, 2005) Marrs et al. (2004) Pezzuto et al. (2010) Mochtar et al. (2009) Ferraretti et al. (2004) Humaidan et al. (2004) Barrenetxea et al. (2008) Matorras et al. (2009) GnRH antagonist De Placido et al. (2006) Bosch et al. (2011) Consensus Recommended Recommended Recommended More research needed GnRH = gonadotrophin-releasing hormone. that in World Health Organization type II anovulatory women undergoing follicular stimulation, this LH ceiling was >60 lg or 1325 IU r-hlh/day when these women showed atresia of non-dominant follicles, elevated progesterone concentrations and reduced clinical pregnancy rates. Recombinant human LH r-hlh has recently become available in many Asian countries. When used to augment r-hfsh, r-hlh is associated with high purity, precision of dosing and consistency. When administered by subcutaneous injection, r-hlh has a terminal half-life of 24 h (le Cotonnec et al., 1998). r-hlh is structurally and functionally analogous to endogenous human LH. In a RCT with patients with a suboptimal response to stimulation with a long GnRH agonist stimulation protocol that compared adding higher doses of r-hfsh versus adding r-hlh or HMG, those given r-hlh (n = 54) had higher live-birth rates (40.7%) than those given HMG (18%). The r-hlh group also had higher implantation rates (Ferraretti et al., 2004). The authors speculated that the differences between the two preparations can produce different biological effects in women who exhibit a suboptimal response to FSH and concluded that using r-hlh is justified if better outcomes can be achieved. Interestingly, a retrospective observational study of 1637 antagonist cycles found that protocols using r-hfsh/r-hlh in combination with ganirelix acetate used lower doses of gonadotrophins to achieve higher implantation and clinical pregnancy rates compared with r-hfsh/hmg in patients older than 38 years (Thornton et al., 2008). Exogenous LH supplementation in Asia: consensus and recommendations This review recognizes that there are wide variations in ovarian response in patients and that treatment regimes tailored to individual patient phenotypes and based on sound paracrine principles are more likely to achieve positive outcomes. During ovarian stimulation, increasing attention needs to be paid to the paracrine environment as more is learnt about the impact of these factors on steroidogenesis and cell growth and apoptosis. Some of the emerging modalities to improve oocyte quality through paracrine modulation include: (i) preconditioning with FSH; (ii) follicular priming with LH to achieve synergy with FSH; and (iii) promoting preovulatory maturation through follicular coasting with LH (Hillier, 2009b). This review also recognizes the importance of optimizing per-cycle cumulative pregnancy rates by improving oocyte counts, quality and improving endometrial receptivity. For this reason it is important at this juncture to critically examine the role of adjuvant r-hlh in assisted reproduction technology. Asian assisted reproduction practitioners make use of both long agonist and antagonist protocols for ovarian stimulation; experience with the former is greater. Published literature on the beneficial effects of exogenous LH in patients with previous suboptimal response or low baseline serum LH concentrations is more extensive in long agonist protocols (Lisi et al., 2003; De Placido et al., 2004, 2005; Franco et al., 2009; Pezzuto et al., 2010). Table 2 lists and summarizes the group s consensus recommendations and the supporting studies, which are discussed in-depth below. The recommendations based on literature review and expert opinion regarding exogenous LH apply mainly to the use of GnRH agonists. The evidence for addition of r-hlh to r-hfsh in antagonist protocols for ovarian stimulation is still being accumulated and more data from future studies is awaited. Patient subpopulations where there is substantial evidence of a benefit of adding r-hlh in ovarian stimulation (poor responders) There is substantial evidence of a benefit of adding r-hlh to women who have a poor response to ovarian stimulation, including: (i) poor response in a previous cycle (Lisi et al., 2003; Mochtar et al., 2009); and (ii) a suboptimal ovarian response with suboptimal follicular progression in a current cycle by day 6 8 (Bosch, 2009; De Placido et al., 2005; Pezzuto et al., 2010). Currently, there is a lack of consensus in Asia on the definition of a prior poor response. This review defines prior poor response as an oocyte count of less than four from a previous stimulation cycle. Other criteria of prior poor response include r-hfsh dose >3000 IU per completed stimulation cycle (Kailasam et al., 2004) and/or less than 800 pg/ml oestradiol on the day of HCG injection. The definition of poor response necessarily means that the diagnosis is based on information from one prior cycle. In an ongoing cycle, there is a group of patients who, after using suppressive GnRH analogues (either agonists or

6 86 PC Wong et al. antagonists), develop severe LH deficiency and exhibit a suboptimal ovarian follicular response by day 6 8. This review defines this suboptimal ovarian response as: (i) having no follicle >10 mm by day 6 (De Placido et al., 2005); (ii) low oestradiol concentration <200 pg/ml by day 6 (Vuong et al., 2004); and (iii) poor progression or slowing of follicle growth, i.e., previously 1 2 mm progression/day slowing to less than 2 mm in 3 days (Bosch, 2009; De Placido et al., 2005; Pezzuto et al., 2010). There is an opportunity in this group of patients to salvage the ongoing cycle through r-hlh supplementation. When compared with increasing r-hfsh dose, adding r-hlh on day 8 was associated with a better cumulative implantation rate (14.2 versus 10.5, P < 0.05) and cumulative pregnancy rate (37.2 versus 29.3, P < 0.05) (De Placido et al., 2005). This review recommends the use of LH supplementation in the ongoing cycle for: (i) patients with a history of prior poor response (Lisi et al., 2003; Mochtar et al., 2009); and (ii) patients who exhibit a suboptimal response during long agonist ovarian stimulation protocols. (De Placido et al., 2005; Pezzuto et al., 2010). There is a possibility to optimize the ovarian response in both these patient groups through the addition of r-hlh. Patient subpopulations where there is evidence of a benefit of adding r-hlh in ovarian stimulation (at risk of hyporesponse) Following on the conclusions for the role of r-hlh in poor responders, patients likely to be at risk of suboptimal response were considered. This review concludes that there are several studies published in peer-reviewed journals that support the use of r-hlh adjuvant treatment in the following at-risk patients who have markers suggestive of suboptimal ovarian response: (i) women aged >35 years undergoing ovarian stimulation with long GnRH agonist protocol there are three RCT supporting the use of r-hlh in addition to r-hfsh in assisted reproduction treatment with long luteal-phase agonist protocols (Humaidan et al., 2004; Marrs et al., 2004; Matorras et al., 2009); and (ii) women aged >35 years undergoing ovarian stimulation with GnRH antagonist protocol there are two RCT that demonstrate benefit of addition of r-hlh to r-hfsh based stimulation (Bosch et al., 2011; De Placido et al., 2006). Based on these clinical studies and personal clinical experience, this review concurred that adjuvant r-hlh starting on either day 1 of stimulation or day 6 8 may be beneficial in patients older than 35 years in long agonist or antagonist protocol ovarian stimulation. Monitoring of the response to add-back LH will depend on clinical tests and equipment available. The monitoring of follicular progression, oestradiol concentrations and endometrial thickness as a clinical measure of LH response is suggested. Patient subpopulations where r-hlh is probably not needed or not shown to improve clinical pregnancy Based on a meta-analysis of RCT comparing r-hfsh versus r-hfsh plus r-hlh ovarian stimulation, which showed that for unselected patients undergoing assisted reproduction technology there is no difference between the two treatments in live-birth rate (odds ratio (OR) 0.92, 95% confidence intervals (CI) ), this review does not recommend adding LH to unselected patients (age <35 years). However, it should be noted that the authors also mention that their conclusion should be interpreted with caution as the number of subjects (n = 701) was insufficient to reach statistical significance (Kolibianakis et al., 2007). Patient subpopulations where r-hlh may be of benefit but further research will be needed to quantify the benefit There is interest in the use of biomarkers to identify patients at risk of LH deficiency. Some putative biomarkers include: (i) LH concentrations either at baseline or midfollicular this review concludes that there is currently insufficient evidence either supporting or refuting the use of LH concentrations to decide which patients should receive r-hlh, and while one study (Pezzuto et al., 2010) shows a benefit in patients with day-6 mid-follicular LH concentrations <0.5 U/l, this finding needs to be corroborated by larger RCT; and (ii) anti-müllerian hormone (AMH) or antral follicle count (AFC) there is an increasing body of research on the use of AFC and AMH concentrations to predict ovarian response and a retrospective study (n = 80) in Indonesian women found AMH to be a good predictor of ovarian response in IVF (Wiweko, 2010). But as far as is known, there are currently no prospective studies in Asian women that allow the recommendation of using either AFC or AMH reliably to predict who may need adjuvant r-hlh treatment in first-time ovarian stimulation patients. Further research is required to elucidate the utility of LH polymorphisms to identify patients who may benefit from r-hlh addition to r-hfsh in assisted reproduction technology. Alviggi et al. (2009) has published on the possible role of variant LH polymorphisms in patients exhibiting suboptimal response to long agonist protocol ovarian stimulation. This may be an interesting area for further studies especially in Asia where the prevalence and types of LH variants are not well described. This review therefore recommends that further research be done in patients at risk of poor ovarian response based on the following biomarkers: (i) AFC <6 in both ovaries; (ii) AMH concentration <1.5 ng/ml; and (iii) LH polymorphisms. Dose and timing of initiation of r-hlh The dose of r-hlh in hypo-hypo patients as stated in the summary of product characteristics for r-hlh is 75 IU combined with 150 IU r-hfsh, i.e. a 2:1 ratio of FSH to LH. In patients undergoing assisted reproduction technology with prevention of LH surge using GnRH analogues, most of the published studies on the combination of r-hlh and r-hfsh in suboptimal responders used r-hlh doses of IU daily combined with r-hfsh doses of IU. In a study that compared either 75 IU or 150 IU r-hlh with r-hfsh (follitropin a or follitropin b) in suboptimal responders with r-hfsh alone in normal responders, significantly more oocytes were retrieved from the 150 IU r-hlh plus r-hfsh group (De Placido et al., 2004). In a patient with

7 Luteinizing hormone supplementation: an Asian perspective 87 Table 3 Consensus on recommended use of LH in Asian women undergoing long gonadotrophin-releasing hormone agonist protocols. Patient category Indication Supporting citation(s) Substantial evidence of benefit of r-hlh in addition to r-hfsh Some evidence of benefit of r-hlh in addition to r-hfsh Further research is needed to determine benefit of r-hlh Prior poor responders, defined as oocyte count <4 in previous cycle Mid-follicular (day 6) suboptimal response on long agonist No follicles >10 mm Oestradiol <200 pg/ml Endometrial thickness <6mm Age >35 years started on ovarian stimulation with either the long agonist or antagonist protocol Biomarkers e.g. variant LH Low baseline serum LH <1.2 IU/l Low antral follicle count Low anti-müllerian hormone Mochtar et al. (2009) and Lisi et al. (2003) De Placido et al. (2004, 2005) and Ferraretti et al. (2004) Vuong et al. (2004) Long agonist: Humaidan et al. (2004), Marrs et al. (2004) and Matorras et al. (2009) Antagonist: De Placido et al. (2006) and Bosch et al. (2011) r-hfsh = recombinant human FSH; r-hlh = recombinant human LH. suboptimal response, this review suggests initiating 150 IU r-hlh combined with 300 IU r-hfsh on either day 1 or 6 of stimulation. Where appropriate, r-hfsh may be titrated to a maximum dose of 450 IU based on clinical response. Regarding the timing of initiation of r-hlh in ovarian stimulation, in some protocols patients start on r-hlh from day 1 of stimulation and for others patients start on days 6 8. Currently, there is no evidence supporting either day 1 or day 6 8 for starting r-hlh. However, in theory there may be a benefit to starting patients on day 1 if a clinician wants to maximize the benefit of increased ovarian androgen production. LH supplementation from day 1 may increase circulating androgen concentrations and it has been demonstrated that increased androgens in combination with FSH can act synergistically to promote FSH receptor mrna expression, follicular development and steroidogenesis (Weil et al., 1999). Table 3 summarizes recommendations for LH use in patients on long GnRH agonist ovarian stimulation protocols. Use of adjuvant r-hlh in poor/suboptimal responders to ovarian stimulation in antagonist protocols Most of the above recommendations are based on data drawn mainly from studies using a long agonist protocol. There are fewer studies examining the situation with respect to antagonist protocols, although in theory, their use should follow similar recommendations for poor responders as in long agonist protocols. An RCT that compared patients undergoing assisted reproduction technology with either r-hfsh alone or with the addition of r-hlh from stimulation day 1 found benefit in the subgroup of patients older than 35 years (Bosch et al., 2011). In patients older than 35 years, the implantation rate was significantly higher in the r-hfsh plus r-hlh group (26.7% versus 18.6%, P < 0.05). Ongoing pregnancy rates per started cycle were 33.5% (r-hfsh plus r-hlh) versus 25.3% (OR 1.49, 95% CI ). In a study of 133 poor responders comparing the GnRH antagonist cetrorelix plus r-hlh to GnRH agonist triptorelin flare-up protocol, the former was associated with greater mean number of metaphase-ii oocytes (5.73 ± 3.57 versus 4.64 ± 2.23, respectively, P < 0.05) (De Placido et al., 2006). A RCT in oocyte donors that compared r-hfsh versus r-hfsh combined with r-hlh in an antagonist protocol showed a significantly higher metaphase-ii oocyte count (80% versus 71%, P < 0.05), fertilization rates (83% versus 71%, P < 0.05), grade 1 embryos (17% versus 3%, P < 0.05) and implantation rates (35% versus 15%, P < 0.05), in recipients whose embryos originated from donors receiving added r-hlh than in donors receiving GnRH antagonist and r-hfsh alone (Acevedo et al., 2004). Future research will be required to lend further support to the use of adjuvant r-hlh in patients on antagonist ovarian stimulation protocols. Conclusions The complex interplay of LH and FSH and their complementary actions are critical to optimal follicle development and eventual ovulation. Secretion of the two gonadotrophins must occur in an orderly fashion. FSH is essential for stimulating antral follicle growth. There is a minimum effective dose for LH and, according to the LH threshold concept, a maximum dose. A minimum threshold concentration of LH is essential in the mid-follicular phase for steroidogenesis in the thecal cells to ensure adequate oestradiol synthesis

8 88 PC Wong et al. within the granulosa cells. In the mid-cycle phase, a surge in LH is required for final maturation and follicle rupture and ovulation. However, if the LH surge above a certain ceiling concentration occurs prematurely, the resulting liberated oocyte may be immature, of poor quality and hence not conducive to a successful pregnancy. It should be recognized that the LH threshold evolves through the follicular cycle and conceptually should become higher as the follicles grow. The LH therapeutic window is best observed in two patient groups, namely hypo-hypo patients and patients who are profoundly suppressed by down-regulation with either GnRH agonists or antagonists in assisted reproduction technology. It is now recognized that with GnRH analogue protocols in ovarian stimulation, levels of LH bioactivity in some patients (e.g. age >35) may be reduced to below the threshold, who thereby need adjuvant r-hlh. The strongest predictive factor for need of exogenous LH in assisted reproduction technology is a prior poor or suboptimal response to ovarian stimulation. In the case of Asian women, this is defined as an oocyte count of less than four and this review recommends that these women should be considered for exogenous LH in the next cycle. Another important group who benefit from adjuvant r-hlh in addition to r-hfsh are women who exhibit suboptimal ovarian response during ovarian stimulation as characterized by: (i) no follicle >10 mm by day 6 8; (ii) low oestradiol (<180 pg/ml) by day 6; and (iii) poor progression or slowing of follicle growth, with previously 1 2 mm progression per day slowing to less than 2 mm in 3 days. These ongoing suboptimal response patients would benefit from addition of r-hlh to salvage the ovarian response to stimulation. There is increasing evidence that age is an important marker of deficient LH bioactivity in women undergoing assisted reproduction technology, with multiple studies showing benefit in women aged above 35 years. This review recommended adding 75 IU r-hlh per day in these patients from day 6. These recommendations apply only to the use of r-hlh which is analogous to endogenous LH in terms of bioactivity, purity and consistency and is the most physiological replacement in terms of binding affinity and half-life. There is also a substantial body of clinical evidence from studies on r-hlh in assisted reproduction technology. HCG has much greater binding affinity and a half-life in excess of 30 h giving rise to significant accumulation in vivo. For these reasons, and because the evidence for HCG in HMG versus r-hfsh is not clear, these recommendations pertain only to r-hlh. These recommendations apply to GnRH agonist protocols where there is substantial evidence for the use of adjuvant r-hlh. While there are studies supporting the use of r-hlh in addition to r-hfsh in GnRH antagonist protocols, these are fewer in number. This is an area that warrants further research. Acknowledgements The Asia Pacific Fertility Treatment Advisory Group meeting was made possible through a grant from Merck Serono. UBM Medica provided writing assistance and editorial support, which was funded by Merck Serono. All contributing authors reviewed and approved the manuscript prior to submission for publication. References Acevedo, B., Sanchez, M., Gomez, J.L., Cuadros, J., Ricciarelli, E., Hernández, E.R., Luteinizing hormone supplementation increases pregnancy rates in gonadotropin-releasing hormone antagonist donor cycles. Fertil. Steril. 82, Alviggi, C., Clarizia, R., Mollo, A., Ranieri, A., De Placido, G., Outlook: who needs LH in ovarian stimulation? Reprod. Biomed. Online 12, Alviggi, C., Clarizia, R., Pettersson, K., Mollo, A., Humaidan, P., Strina, I., Coppola, M., Ranieri, A., D Uva, M., De Placido, G., Suboptimal response to GnRHa long protocol is associated with a common LH polymorphism. Reprod. Biomed. Online 18, Baer, G., Loumaye, E., Comparison of recombinant human luteinising hormone (r-hlh) and human menopausal gonadotropin (hmg) in assisted reproductive technology. Curr. Med. Res. Opin. 19, Balasch, J., Fábreques, F., Is luteinizing hormone needed for optimal ovulation induction? Curr. Opin. Obstet. Gynecol. 14, Barrenetxea, G., Agirregoika, J.A., Jimenez, M.R., Lopez de Larruzea, A., Ganzabal, T., Carbonero, K., Ovarian response and pregnancy outcome in poor-responder women: a randomized controlled trial on the effect of luteinizing hormone supplementation on in vitro fertilization cycles. Fertil. Steril. 8, Berger, M.J., Taymor, M.L., The role of luteinizing hormone in human follicular maturation and function. Am. J. Obstet. Gynecol. 111, Bosch, E., Recombinant human FSH and recombinant human LH in a 2:1 ratio combination: a new tool for ovulation induction. Expert Rev. Obstet. Gynecol. 4, Bosch, E., Labarta, E., Crespo, J., Simón, C., Remohí, J., Pellicer, A., Impact of luteinizing hormone administration on gonadotropin-releasing hormone antagonist cycles: an age-adjusted analysis. Fertil. Steril. 95, Couzinet, B., Lestrat, N., Brailly, S., Forest, M., Schaison, G., Stimulation of ovarian follicular maturation with pure follicle stimulating hormone in women with gonadotropin deficiency. J. Clin. Endocrinol. Metab. 66, De Placido, G., Alviggi, C., Mollo, A., Strina, I., Ranieri, A., Alviggi, E., Wilding, M., Varricchio, M.T., Borrelli, A.L., Conforti, S., Effects of recombinant LH (rlh) supplementation during controlled ovarian hyperstimulation (COH) in normogonadotrophic women with an initial inadequate response to recombinant FSH (rfsh) after pituitary downregulation. Clin. Endocrinol. 60, De Placido, G., Alviggi, C., Perino, A., Strina, I., Lisi, F., Fasolino, A., De Palo, R., Ranieri, A., Colacurci, N., Mollo, A., Italian Collaborative Group on Recombinant Human Luteinizing Hormone, Recombinant human LH supplementation versus recombinant human FSH (rfsh) step-up protocol during controlled ovarian stimulation in normogonadotrophic women with initial inadequate ovarian response to rfsh. A multicentre, prospective, randomized controlled trial. Hum. Reprod. 20, De Placido, G., Mollo, A., Clarizia, R., Strina, I., Conforti, S., Alviggi, C., Gonadotropin-releasing hormone (GnRH) antagonist plus recombinant luteinizing hormone vs. a standard GnRH agonist short protocol in patients at risk for poor ovarian response. Fertil. Steril. 85, Devroey, P., Mannaerts, B., Smitz, J., Coelingh Bennink, H., Van Steirteghem, A., Clinical outcome of a pilot efficacy study

9 Luteinizing hormone supplementation: an Asian perspective 89 on recombinant human follicle-stimulating hormone (Org 32489) combined with various gonadotropin-releasing hormone agonist regimens. Hum. Reprod. 9, European Recombinant Human LH Study Group, Recombinant human luteinizing hormone (LH) to support recombinant human follicle-stimulating hormone (FSH)-induced follicular development in LH-and FSH-deficient anovulatory women: a dose finding study. J. Clin. Endocrinol. Metab. 83, Falck, S., Site of the production of estrogen in rat ovary as studies in micro-transplants. Ph.D. Thesis, University of Lund, Sweden. Ferraretti, A.P., Gianaroli, L., Magli, M.C., D angelo, A., Farfalli, V., Montanaro, N., Exogenous luteinizing hormone in controlled ovarian hyperstimulation for assisted reproduction techniques. Fertil. Steril. 82, Filicori, M., Cognini, A.E., Tabarelli, C., Pocognoli, P., Taraborrelli, S., Spettoli, D., Ciampaglia, W., Stimulation and growth of antral ovarian follicles by selective LH activity administration in women. J. Clin. Endocrinol. Metab. 87, Franco Jr., J.G., Baruffi, R.L., Oliveira, J.B., et al., Effects of recombinant LH supplementation to recombinant FSH during induced ovarian stimulation in the GnRH-agonist protocol: a matched case-control study. Reprod. Biol. Endocrinol. 7 (58), 1 8. Grøndahl, M.L., Borup, R., Lee, Y.B., Myrhøj, V., Meinertz, H., Sørensen, S., Differences in gene expression of granulosa cells from women undergoing controlled ovarian hyperstimulation with either recombinant follicle-stimulating hormone or highly purified human menopausal gonadotropin. Fertil. Steril. 91, Hillier, S.G., Current concepts of the roles of follicle stimulating hormone and luteinizing hormone in folliculogenesis. Hum. Reprod. 9, Hillier, S.G., 2009a. The science of ovarian ageing how might knowledge be translated into practice? In: Bewley, S., Ledger, W., Nikolaou, D. (Eds.), Reproductive Ageing. RCOG Press, London, pp Hillier, S.G., 2009b. Paracrine support of ovarian stimulation. Mol. Hum. Reprod. 15, Hugues, J.N., Soussis, J., Calderon, I., Balasch, J., Anderson, R.A., Romeu, A., on behalf of the Recombinant LH Study Group, Does the addition of recombinant LH in WHO group II anovulatory women over-responding to FSH treatment reduce the number of developing follicles? A dose-finding study. Hum. Reprod. 20, Humaidan, P., Bungum, L., Bungam, M., Andersen, C., Ovarian response and pregnancy outcome related to mid-follicular LH levels in women undergoing assisted reproduction with GnRH agonist down-regulation and recombinant FSH stimulation. Hum. Reprod. 17, Humaidan, P., Bungum, M., Bungam, L., Yding 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 8, Kailasam, C., Keay, S.D., Wilson, P., Ford, W.C., Jenkins, J.M., Defining poor ovarian response during IVF cycles, in women aged <40 years, and its relationship with treatment outcome. Hum. Reprod. 19, Kolibianakis, E.M., Kalogeropoulou, L., Griesinger, G., Papanikolaou, E.G., Papadimas, J., Bontis, J., Tarlatzis, B.C., 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 metaanalysis. Hum. Reprod. Update 13, Lahoud, R., Al-Jefout, M., Tyler, J., Ryan, J., Driscoll, G., A relative reduction in mid-follicular LH concentrations during GnRH agonist IVF/ICSI cycles leads to lower live birth rates. Hum. Reprod. 21, Loumaye, E., Engrand, P., Shoham, Z., Hillier, S.G., Baird, D.T., Clinical evidence for an LH ceiling effect induced by administration of recombinant human LH during the late follicular phase of stimulated cycles in World Health Organization type I and type II anovulation. Hum. Reprod. 18, Lisi, F., Rinaldi, L., Fishel, S., Lisi, R., Pepe, G., Picconeri, M.G., Use of recombinant follicle-stimulating hormone (Gonal F) and recombinant luteinizing hormone (Luveris) for multiple follicular stimulation in patients with a suboptimal response to in vitro fertilization. Fertil. Steril. 79, le Cotonnec, J.Y., Loumaye, E., Porchet, H.C., Beltrami, V., Munafo, A., Pharmacokinetic and pharmacodynamic interactions between recombinant human luteinizing hormone and recombinant human follicle-stimulating hormone. Fertil. Steril. 69, Loutradis, D., Vomvolaki, E., Drakakis, P., Poor responder protocols for in-vitro fertilization: options and results. Curr. Opin. Obstet. Gynecol. 20, Mannaerts, B., de Leeuw, R., Geelen, J., Van Ravestein, A., Van Wezenbeek, P., Schuurs, A., Kloosterboer, H., Comparative in vitro and in vivo studies on the biological characteristics of recombinant human follicle stimulating hormone. Endocrinology 129, 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 8, Matorras, R., Prieto, B., Exposito, A., Mendoza, R., Crisol, L., Herranz, P., Burgués, S., Mid-follicular LH supplementation in women aged years undergoing ICSI cycles: a randomized controlled study. Reprod. Biomed. Online 19, McGee, E.A., Hsueh, A.J., Initial and cyclic recruitment of ovarian follicles. Endocr. Rev. 21, Menon, K.M.J., Nair, A.K., Wang, L., A novel post-transcriptional mechanism of regulation of luteinizing hormone receptor expression by an RNA binding protein from the ovary. Mol. Cell. Endocrinol. 246, Mochtar, M.H., van der Veen, F., Ziech, M., van Wely, M., Musters, A., Recombinant Luteinizing Hormone (rlh) for controlled ovarian hyperstimulation in assisted reproductive cycles. Cochrane Database Syst. Rev. 1, Nakagawa, K., Ohgi, S., Nakashima, A., Horikawa, T., Sugiyama, R., Saito, H., The ratio of late-follicular to mid-follicular phase LH concentrations efficiently predicts ART outcomes in women undergoing ART treatment with GnRH-agonist long protocol and stimulation with recombinant FSH. J. Assist. Reprod. Genet. 25, O Dea, L., O Brien, F., Currie, K., Hemsey, G., Follicular development induced by recombinant luteinizing hormone (LH) and follicle-stimulating hormone (FSH) in anovulatory women with LH and FSH deficiency: evidence of a threshold effect. Curr. Med. Res. Opin. 24, Pezzuto, A., Ferrari, B., Coppola, F., Nardelli, G.B., LH supplementation in down-regulated women undergoing assisted reproduction with baseline low serum LH levels. Gynecol. Endocrinol. 26, Shoham, Z., The clinical therapeutic window for luteinizing hormone in controlled ovarian stimulation. Fertil. Steril. 77, Shoham, Z., Mannaerts, B., Insler, V., Coelingh-Bennink, H., Induction of follicular growth using recombinant human follicle-stimulating hormone in two volunteer women with hypogonadotropic hypogonadism. Fertil. Steril. 59,