Preovulatory progesterone concentration associates significantly to follicle number and LH concentration but not to pregnancy rate

Reproductive BioMedicine Online (2011) 23, 187 195 www.sciencedirect.com www.rbmonline.com ARTICLE Preovulatory progesterone concentration associates significantly to follicle number and LH concentration but not to pregnancy rate Claus Yding Andersen a, *, Leif Bungum b, Anders Nyboe Andersen c, Peter Humaidan d a Laboratory of Reproductive Biology, University Hospital of Copenhagen, University of Copenhagen, Section 5712, Blegdamsvej 9, DK-2100 Copenhagen, Copenhagen, Denmark; b Reproductive Medicine Center, Skåne University Hospital, Lund University, Malmö, Sweden; c The Fertility Clinic, University Hospital of Copenhagen, University of Copenhagen, Copenhagen, Denmark; d The Fertility Clinic, Skive Regional Hospital, Skive, Denmark * Corresponding author. E-mail address: yding@rh.dk (C Yding Andersen). Claus Yding Andersen is professor in human reproductive physiology at University of Copenhagen, Denmark and heads the Laboratory of Reproductive Biology at the University Hospital of Copenhagen, Denmark. His main interests are cryopreservation of gonadal tissue, ovarian endocrinology and human embryonal stem cells. He has published almost 200 scientific papers and has given a large number of international presentations. Abstract Using data from a large prospective randomized controlled trial that evaluated the effect of recombinant LH (rlh) co-administration for ovarian stimulation, the present study assessed whether progesterone concentration on the day of human chorionic gonadotrophin (HCG) administration was associated with pregnancy outcome. Progesterone concentration was measured on stimulation day 1 and on the day of HCG administration in 475 patients who underwent IVF/intracytoplasmic sperm injection treatment following ovarian stimulation with gonadotrophin-releasing hormone (GnRH) agonist and recombinant FSH with or without rlh administration from day 6 of stimulation. There was no significant association between the late-follicular-phase progesterone concentration and the clinical pregnancy rate. However, progesterone concentration was strongly associated with the number of follicles and retrieved oocytes. Late-follicular-phase LH concentration also showed a significant positive association with progesterone concentration (P = 0.018). Administration of rlh during ovarian stimulation did not affect progesterone concentration. The present study does not support an association between progesterone concentration on the day of HCG administration and the probability of clinical pregnancy in women undergoing ovarian stimulation with GnRH agonists and gonadotrophins for assisted reproduction treatment. Instead, late-follicular-phase progesterone concentration appears to be governed by the number of preovulatory follicles and LH concentration. RBMOnline ª 2011, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. KEYWORDS: ovarian stimulation, pregnancy outcome, progesterone elevation 1472-6483/$ - see front matter ª 2011, Reproductive Healthcare Ltd. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.rbmo.2011.04.003

188 C Yding Andersen et al. Introduction Control of human ovarian steroidogenesis is invariably linked to the cyclic variations of gonadotrophins that occur during the menstrual cycle. FSH and LH trigger the conversion of inactive steroid precursors, like cholesterol to oestradiol, during the follicular phase and to progesterone during the luteal phase. In contrast to the fine-tuned control and feedback mechanisms that exist between FSH and LH and steroid secretion during the natural menstrual cycle, ovarian stimulation introduces concentrations of gonadotrophins out of the normal range, which, in addition to the development of multiple follicles, also affects steroidogenesis. In connection with ovarian stimulation, each individual follicle is affected and results in a collective augmented output of ovarian steroids, primarily oestradiol and progesterone. The control over oestradiol secretion is well described by the two-cell two-gonadotrophin concept in which theca cells specifically undertake androgen synthesis, which then becomes converted to oestradiol exclusively in the granulosa cell compartment. In contrast, control of progesterone production especially in connection with ovarian stimulation is less well described. Both theca and granulosa cells have the capacity to produce progesterone and expression of the enzymes converting cholesterol to pregnenolone (P450 side-chain cleavage enzyme, P450scc) and pregnenolone to progesterone (3-b-hydroxysteroid dehydrogenase, 3bHSD) both appear to be up-regulated by FSH and LH (Payne and Hales, 2004; Yu et al., 2005). So although progesterone is naturally produced by both theca and granulosa cells and is present in the circulation during the follicular phase of the human menstrual cycle, the precise control of secretion, especially in connection with ovarian stimulation, is not known in detail. While a wide range of oestradiol concentration in the late follicular phase accommodates successful pregnancies, there is somewhat more controversy as to the late-follicular-phase concentration of progesterone. Although serum progesterone concentration in the late follicular phase is considerably lower than that observed in connection with the mid-cycle surge of gonadotrophins or as induced by human chorionic gonadotrophin (HCG) for ovulation induction, even subtle increases in relatively low progesterone concentrations have been suspected to negatively affect ongoing pregnancy rates in connection with ovarian stimulation, possibly through an advancement of endometrial maturation (Fanchin et al., 1997a; Smitz et al., 2007). A number of studies have reported a negative association between progesterone concentration and clinical pregnancy rate (Azem et al., 2008; Bosch et al., 2003, 2010; Check et al., 1993; Fanchin et al., 1993, 1997b; Harada et al., 1996; Shulman et al., 1996), while many others found no association (Abuzeid and Sasy, 1996; Bustillo et al., 1995; Check, 1994; Doldi et al., 1999; Edelstein et al., 1990; Givens et al., 1994; Hofmann et al., 1993; Levy et al., 1995; Martinez et al., 2004; Miller et al., 1996; Moffitt et al., 1997; Niu et al., 2008; Silverberg et al., 1991; Ubaldi et al., 1995; Urman et al., 1999; Venetis et al., 2007). Collectively, in 2007 a meta-analysis concluded that there was no statistical association between the late-follicular-phase progesterone concentrations and the ongoing pregnancy rates (Venetis et al., 2007). Despite this meta-analysis several recent studies have reopened the debate on the effect of late-follicular-phase progesterone concentration on pregnancy outcome by providing new data that suggests an association between progesterone and pregnancy outcome (Bosch et al., 2010), and by suggesting progesterone to be a precursor of human ovarian androgen production (Fleming and Jenkins, 2010). Further, as pointed out by Venetis et al. (2007), there is no consensus on the cut-off value to define increased progesterone concentration and there are differences in measuring progesterone, which at these relatively low concentrations can vary considerably within an assay and from one assay to another. In addition, there is a lack of knowledge on how different gonadotrophin preparations, different pituitary down-regulation regimens and FSH and LH in themselves affect progesterone production during the follicular phase. Collectively, these data suggest that the regulation of progesterone production in connection with ovarian stimulation and its impact on the subsequent chance of becoming pregnant remains an unresolved issue. The aim of the present study was to retrospectively analyse data from a large, prospective, multicentre, randomized controlled trial on the effects of adding recombinant LH (rlh) to recombinant FSH (rfsh) during the final days of ovarian stimulation for IVF/intracytoplasmic sperm injection (ICSI) in order to explore whether late-follicular-phase serum progesterone concentration was associated with pregnancy rates. This study could also focus on whether the addition of LH in the second half of the follicular phase influenced serum progesterone concentration on the day of HCG. Materials and methods Blood samples included in the present study came from an original prospective, randomized, controlled, investigator-driven study to evaluate the effect of rlh given from day 6 of stimulation in combination with rfsh for ovarian stimulation, the so-called Nordic LH study (Nyboe Andersen et al., 2008). The original study involved 22 centres: 10 in Denmark, two in Finland, four in Norway and six in Sweden and all clinical data from the study have been previously published (Nyboe Andersen et al., 2008). The study was approved by the local ethics committees of each participating country. All participants gave written, informed consent. In brief, the inclusion criteria included: female age <40 years with an indication for IVF or ICSI, a regular (21 35 day) menstrual cycle and basal serum FSH concentration of <10 IU/l on menstrual cycle day 2 5. Ovarian stimulation was performed as follows: pituitary down-regulation was achieved using nafarelin (Synarela; Pfizer), 200 lg, administered three times daily for 14 days; thereafter the dosage of nafarelin was decreased to 200 lg twice daily. After confirmation of successful down-regulation, 526 women were randomized 1:1 to receive rfsh (GONAL-f; Serono) alone (n = 261) or rfsh plus rlh (Luveris; Serono; n = 265) from day 6 of stimulation. The starting dose of rfsh was 150 IU/day for women aged 35 years (n = 426) and 225 IU/day for patients aged >35 years (n = 100). Dosing

Late-follicular-phase progesterone does not impact subsequent pregnancy rate 189 was fixed for the first 6 days and then individualized if appropriate. The dose of rlh was fixed at 75 IU/day for women aged 35 years (n = 216) and 150 IU/day for patients aged >35 years (n = 49). Administration of recombinant HCG (Ovitrelle; Serono) for ovulation induction, oocyte retrieval, fertilization via conventional IVF or ICSI, embryo culture and transfer techniques were performed according to local procedures. Data were collected from one cycle (with possible fresh embryo transfer) per patient. Retrospective data analysis In the original study a total of 526 patients were included. Of these patients 496 underwent oocyte aspiration and 452 received embryo transfer. Each patient was only included once. The present study included a total of 475 patients in whom blood samples were available on the day of HCG administration. Of these a total of 419 received embryo transfer. For the present study, blood samples were collected on day 1, prior to exogenous FSH administration, and on the day of HCG administration. The last blood sample was obtained on the day of HCG administration in the evening. Serum samples were prepared locally and stored at 20 C. All samples were kept in tubes with airtight caps and all were stored in the same freezer after collection from the different centres. Samples did not experience unforeseen thawing during storage. Serum progesterone concentration was measured using a chemiluminescence immunometric assay (Capio Diagnostik, Copenhagen, Denmark). The detection limit of the assay was 0.6 nmol/l and it was characterized by inter- and intra-assay variation of 8.5% and 4.3%, respectively, for a sample containing 40 nmol/l progesterone and 13.8% and 5.8%, respectively, for a sample containing 3.5 nmol/l progesterone. The laboratory is accredited and the analysis was performed in a good laboratory practice environment. The company only identified each sample by a single continuous number and was unaware of the purpose of the study. The original clinical trial was performed in year 2004 and 2005 and the progesterone concentration was measured in year 2007. The concentration of progesterone on the day of HCG administration (i.e. ovulation induction) was related to the number of oocytes retrieved, the number of follicles >10 mm on the day of HCG administration and the total consumption of FSH. Patients were stratified into subgroups according to serum progesterone concentration on the day of HCG administration and related to the frequency of biochemical and clinical pregnancies. A biochemical pregnancy was defined by a plasma HCG concentration of >10 IU/l on day 14 following embryo transfer and a clinical pregnancy was confirmed 3 weeks after a positive HCG test, using transvaginal ultrasonography. A clinical pregnancy was defined as a live pregnancy with an intrauterine gestational sac and fetal heartbeat. Statistical analyses Student s t-test was used to compare progesterone concentrations in two groups. When the number of groups exceeded two, ANOVA was used. The proportion of clinical pregnancies was calculated in groups spanning the progesterone concentrations 0 2, 2 3, 3 4, 4 5 and >5 nmol/l with 95% confidence intervals using binomial regression analysis. A P-value <0.05 was accepted as statistically significant. Results Stored serum samples were available from 475 of the 526 patients (90%) who were randomized to treatment in the original study. Demographic and disease characteristics of the 475 patients were similar to those of the original study (Table 1). The average progesterone concentration on the day of ovulation induction (day of HCG administration) did not differ between those who received embryo transfer and those who did not (mean ± SD, 4.38 ± 3.90 nmol/l, (n = 419), versus 3.99 ± 2.16 nmol/l, (n = 56), respectively). In the current study a total of 247 patients received only rfsh and 228 received rfsh plus rlh for ovarian stimulation. The mean concentration of progesterone was similar between the two groups on day 1 of stimulation and on the day of ovulation induction, with an increase of about 2 nmol/l for the length of the stimulation period in both groups (Table 2). It was found that women below the age of 35 years tended to have slightly higher concentrations of progesterone as compared with those above 35 years of age. Again, both age groups showed an increase of about 2 nmol/l for the length of the stimulation period (Table 3). The concentration of progesterone on the day of ovulation induction, irrespective of whether or not rlh was added, showed a strong positive association with the number of oocytes retrieved and the number of follicles Table 1 Demographic and infertility characteristics of patients included in the retrospective analysis. Characteristic FSH (n=247) FSH + LH (n=228) Age (years) 31.8 ± 4.0 31.8 ± 3.9 >35 years 46 (19) 44 (19) 35 years 201 (81) 184 (81) Duration of infertility 3.0 ± 2.0 3.2 ± 2.1 (years) Cycle number 1 173 (70) 145 (64) 2 46 (19) 58 (25) 3 28 (11) 25 (11) Cause of infertility Tubal 41 (17) 37 (16) Male 105 (43) 100 (44) Anovulation 5 (2) 4 (2) Endometriosis 9 (4) 8 (4) Unexplained 66 (27) 59 (26) Others 4 (2) 6 (3) Combined 17 (7) 14 (6) Values are mean ± SD or n (%).

190 C Yding Andersen et al. Table 2 Progesterone concentration in women randomized to receive either LH or no LH in connection with exogenous stimulation with FSH. No LH Number 247 228 Stimulation day 1 2.29 ± 007 2.24 ± 0.06 Day of ovulation induction 4.38 ± 0.22 4.29 ± 0.26 Values are mean nmol/l ± SEM unless otherwise stated. There were no statistically significant differences between the two groups (Student s t-test). Table 3 women. Progesterone concentration in relation to age of No LH Below the age of 35 years Number 201 184 Stimulation day 1 2.34 ± 0.06 2.31 ± 0.07 Day of ovulation induction 4.43 ± 0.28 4.34 ± 0.33 Above the age of 35 years Number 46 44 Stimulation day 1 2.15 ± 0.20 2.03 ± 0.08 Day of ovulation induction 4.24 ± 0.27 4.11 ± 0.30 Values are mean nmol/l ± SEM unless otherwise stated. There were no statistically significant differences between the two groups (Student s t-test). observed on sonography (Table 4). Actually the number of oocytes and follicles increased more than 50% comparing LH LH the two flanking groups. Interestingly, the total consumption of rfsh was unrelated to progesterone concentration on the day of HCG administration (Table 4). The original clinical trial found that the total FSH consumption was unrelated to the use of rlh (Nyboe Andersen et al., 2008). The concentration of LH was measured in the original study (Nyboe Andersen et al., 2008), where it was found that concentrations remained similar whether or not rlh was administered except for the last day of stimulation (i.e. day of HCG administration), where concentrations were significantly higher in the group who received rlh (1.57 versus 2.14 IU/l, respectively). The present study found a significant positive association between progesterone concentration on the day of ovulation induction and LH concentration (P = 0.018; Table 5), despite the similar concentrations of progesterone when the two groups as a whole were evaluated (Table 2). The number of oocytes retrieved was not significantly associated with the circulatory concentration of LH (Table 5). The biochemical and clinical pregnancy rates in relation to late-follicular-phase progesterone concentration are given in Table 6. Each data point shows the number of patients with progesterone concentration above 1, 2, 3, 4, 5, 6 or 7 nmol/l, respectively, and the corresponding pregnancy rates. The biochemical and clinical pregnancy rates were not related to the progesterone concentration and appeared overall to be constant. When the clinical pregnancy rate was related to different intervals of progesterone concentration (Fig. 1), no statistical significant correlation between the two parameters was found. Using a progesterone concentration of 4 nmol/l as a cut-off point, 40% of the patients had concentrations exceeding 4 nmol/l. The outcomes of patients having progesterone concentration 4 nmol/l or >4 nmol/l were sim- Table 4 Progesterone concentration on the day of ovulation induction in relation to number of oocytes retrieved, number of follicles >10 mm and total FSH consumption. Progesterone concentration (nmol/l) <1 1 2 2 3 3 4 4 5 >5 ANOVA Patients a 2 34 108 116 66 113 Oocytes retrieved 6.5 ± 0.5 6.8 ± 0.6 7.6 ± 0.4 9.2 ± 0.5 9.7 ± 0.6 10.3 ± 0.5 P < 0.0001 Follicles >10 mm 10.0 ± 4.0 11.3 ± 1.1 13.9 ± 0.8 15.1 ± 0.9 15.6 ± 0.8 17.8 ± 0.8 P = 0.0007 FSH consumption (IU) 2432 ± 324 1995 ± 144 2226 ± 101 2192 ± 85 2147 ± 124 2264 ± 90 NS Values are mean ± SEM unless otherwise stated. a In 36 cycles the number of oocytes and/or number of follicles were not recorded properly. NS = not statistically significant. Table 5 LH concentration on the day of ovulation induction in relation to progesterone concentration and number of oocytes retrieved. LH concentration (IU/l) <1 1 2 2 3 3 4 4 5 >5 ANOVA Progesterone (nmol /l) 3.7 ± 0.3 3.9 ± 0.1 4.9 ± 0.6 4.5 ± 0.3 5.3 ± 0.6 6.0 ± 1.5 P = 0.018 Oocytes 8.5 ± 0.4 9.1 ± 0.3 9.4 ± 0.5 9.2 ± 0.9 8.9 ± 1.1 8.6 ± 1.0 NS NS = not statistically significant.

Late-follicular-phase progesterone does not impact subsequent pregnancy rate 191 Table 6 Biochemical and clinical pregnancy rates of patients, undergoing ovarian stimulation for IVF/intracytoplasmic sperm injection, stratified according to late-follicular-phase serum progesterone concentration. Total Late-follicular-phase serum progesterone concentration (nmol/l) 1 2 3 4 5 6 7 Per patient included Patients 475 472 432 306 187 124 80 45 Biochemical pregnancy rate 189 (40) 188 (40) 172 (40) 131 (43) 88 (47) 51 (41) 31 (39) 20 (44) Clinical pregnancy rate 146 (31) 144 (31) 129 (30) 98 (32) 64 (34) 35 (28) 24 (30) 16 (36) Per embryo transfer Patients 419 416 379 276 171 112 72 39 Biochemical pregnancy rate 189 (45) 187 (45) 171 (45) 130 (47) 87 (51) 50 (45) 30 (42) 19 (49) Clinical pregnancy rate 146 (35) 144 (35) 129 (34) 98 (36) 64 (37) 35 (31) 24 (33) 16 (41) Values are n or n (%). Figure 1 Progesterone concentration at discrete intervals in relation to the clinical pregnancy rate. No statistically significant association between the clinical pregnancy rate and the progesterone concentration in the late follicular phase was found. ilar, as the clinical pregnancy rate was 33% when the progesterone concentration was 4 nmol/l and 37% for progesterone >4 nmol/l. Using a progesterone concentration of 4.77 nmol/l (corresponding to 1.5 ng/ml) as a cut-off point, the clinical pregnancy rate was similar in the two groups (35% in each of the two groups) with a little less than 30% of the patients present in the group with higher progesterone concentration. The average progesterone concentration of women receiving either rlh or not, in relation to whether they achieved a biochemical pregnancy is shown in Table 7. A few outliers had a progesterone concentration above 10 nmol/l. They were considered to have ovulated prematurely and the number of follicles developed and oocytes retrieved were in the same range as the rest of the patients. After removal of these patients, there was a significantly increased concentration of progesterone (P < 0.05) in the group who achieved a positive HCG test as compared with those who did not in the group that only received rfsh for ovarian stimulation. Discussion The present study demonstrated that there was no association between the late-follicular-phase serum progesterone Table 7 Progesterone concentrations in women randomized to receive either LH or no LH during ovarian stimulation in relation to human chorionic gonadotrophin test results. No LH LH Negative HCG Positive HCG Negative HCG Positive HCG Number 146 101 139 88 Stimulation day 1 2.32 ± 0.10 a 2.25 ± 0.09 a 2.15 ± 0.06 b 2.39 ± 0.11 b Day of ovulation induction 4.10 ± 0.23 a 4.82 ± 0.43 a 4.04 ± 0.16 a 4.69 ± 0.63 a Excluding progesterone values >10 nmol/l on the day of ovulation induction Number 142 100 139 87 Day of ovulation Induction 3.78 ± 0.14 b 4.26 ± 0.18 b 4.04 ± 0.16 a 4.00 ± 0.19 a Values are mean nmol/l ± SEM. Significance testing by Student s t-test. Data based on available data on stimulation day 1. Patients who did not undergo embryo transfer were included in the group with a negative HCG test. HCG= Human chorionic gonadotrophin a Not statistically significant. b P < 0.05.

192 C Yding Andersen et al. concentration and the biochemical and clinical pregnancy rates obtained after ovarian stimulation for IVF/ICSI using a long gonadotrophin-releasing hormone (GnRH) agonist protocol and stimulation with rfsh and rlh. Instead, a strong significant association was found between the number of developing follicles/number of retrieved oocytes and serum progesterone concentration, suggesting that each individual follicle contributes to the collective concentration observed in the circulation. Equally important, the late-follicular-phase serum LH concentration also exhibited a significant positive association with progesterone concentration. However, although there was a significantly increased late-follicular-phase concentration of LH in the group of patients that received rlh as compared with those who only received rfsh, the association between progesterone and LH was not strong enough to reveal any difference in progesterone concentration between these two groups. Further, it was of interest that the total FSH consumption in connection with ovarian stimulation was unrelated to the late-follicular-phase progesterone concentration, which contrasts with previous studies (Bosch et al., 2003, 2010). Collectively, the present results showed that the pregnancy rate was unrelated to the late-follicular-phase concentration of progesterone, which mainly seemed to be determined by the number of developing follicles and, to some degree, by the late-follicular-phase concentration of LH. It is widely accepted that women who are stimulated with conventional ovarian stimulation and who develop many follicles have a relatively high pregnancy rate and the present results are in accordance with this. Indeed, the highest pregnancy rate was found in the group of patients who had the highest late-follicular-phase progesterone concentrations (i.e. >7 nmol/l) and thus developed many follicles. The present study is in agreement with the most recent meta-analysis on several points (Venetis et al., 2007); (i) there was no association between the late-follicular-phase progesterone concentration and pregnancy rates; (ii) in the group of patients receiving GnRH agonists for pituitary down-regulation there was a significant association between the number of oocytes retrieved and the late-follicular-phase progesterone concentration; and (iii) it was not possible to demonstrate an association between the total FSH consumption during ovarian stimulation and the late-follicular-phase progesterone concentration. This is, however, in contrast to a recent retrospective study by Bosch et al. (2010). Further, the study of Bosch et al. (2010) found a strong significant negative association between pregnancy rates and late-follicular-phase progesterone concentration irrespective of the type of ovarian stimulation protocol (i.e. GnRH agonist or antagonist). The authors used no inclusion or exclusion criteria. Thus data from all patients were analysed retrospectively irrespective of cycle number, indication and stimulation protocol in an attempt to mimic everyday clinical experience. Only a small fraction, approximately 6% of all the patients, had a high progesterone concentration on the day of HCG administration (i.e. >4.77 nmol/l, corresponding to 1.5 ng/ml) (Bosch et al., 2010). Thus, it may be speculated that the high-responder group reflects a special group of patients, perhaps those who did not initially respond adequately (either during a given stimulation cycle or in a previous cycle), and therefore received enhanced gonadotrophin stimulation to promote follicular development or because they were low-responder patients (Bosch et al., 2010). The high-responder group is characterized by an enhanced gonadotrophin stimulation, which may have caused granulosa cells to produce more progesterone. This is, however, likely to be secondary to an inadequate response reflecting that these patients for other reasons possessed a poorer pregnancy prognosis. Patients in the present study were less likely to receive an enhanced exogenous gonadotrophin stimulus, since they were normal-responder women; the majority of women were in their first or second cycle and a fixed protocol for the first half of the follicular phase was used. Furthermore, the study by Bosch et al. (2010) was unable to account for the LH activity present in these patients, which in the present study was shown to be significantly associated with late-follicular-phase progesterone concentration. Some of the patients with high progesterone concentration in the study of Bosch et al. (2010) may simply be patients with a higher LH tonus despite pituitary down-regulation. The cut-off value to define high late-follicular-phase progesterone concentration (i.e. 4.77 nmol/l) in the study by Bosch et al. (2010) resulted in about 6% of patients being classified as having a high concentration. If a similar cut-off value is used in the present study, almost 30% were included in the high-responder group. This trend is also reflected in the overall late-follicular-phase progesterone concentration, which in the study by Bosch et al. (2010) was 2.67 nmol/l, whereas it was 4.34 nmol/l (or 4.01 nmol/l if values above 10 nmol/l were excluded) in the current study. This suggests that the two assays used for progesterone measurement in the two studies did not result in similar results with these relatively low concentrations of progesterone. However, the assay used in the present study was exactly the same as the one used in the so-called MERIT trial (Nyboe Andersen et al., 2006) comparing rfsh with highly purified human menopausal gonadotrophin. Moreover, the assay was performed by exactly the same laboratory, which was an accredited laboratory that performed the assay in a good laboratory practice environment, albeit with a relatively high variability, which is often the case when measuring such low concentrations of progesterone. In the study by Nyboe Andersen et al. (2006) the authors reported that elevated progesterone concentrations above 4 nmol/l were associated with a poorer pregnancy outcome. Although this difference is difficult to explain, it is noticeable that the FSH consumption in the rfsh group in the study by Nyboe Andersen et al. (2006) received a starting dose of 225 IU/day (total dose of 2385 IU FSH). In the present study the age group below 35 years of age only received 150 IU rfsh per day (total median dose of 1950 IU in this group), which is about a 20% reduction in FSH consumption. Maybe the higher progesterone concentrations observed in the study by Nyboe Andersen et al. (2006) reflected a higher drive on granulosa cell progesterone synthesis, which furthermore may compensate for the higher LH-like activity that was present in the other group of patients receiving highly purified human menopausal gonadotrophin, a drug containing HCG as LH activity. This may serve to illustrate that both FSH and LH-like activity has an effect on follicular progesterone synthesis.

Late-follicular-phase progesterone does not impact subsequent pregnancy rate 193 In the present study women above the age of 35 years received an increased dose of rlh as compared with those aged below 35 years from day 6 of stimulation. On the day of HCG administration this resulted in significantly increased concentrations of LH in the older patient group. Although this study found an overall significant positive association between late-follicular-phase LH concentration and progesterone concentration, the older patient group showed progesterone concentration similar to the younger patient group. This is likely to reflect that the number of developing follicles in the older patients is reduced leading to a lower progesterone concentration in circulation and the relatively small increase in LH concentration between the younger and the older age groups was unable to compensate for this, thus explaining why no difference in progesterone concentration was detected. Further, the increased use of rfsh in the older age group may also stimulate more follicles and more progesterone synthesis in the cohort of follicles, which is relatively smaller than in the younger age group. A number of studies have used the term premature luteinization to describe the late-follicular-phase rise in progesterone and it has been assumed that the granulosa cells had prematurely started to luteinize. However, this is most likely not the case, since progesterone is an end-product in human ovarian steroidogenesis that naturally occurs in the follicular phase of the menstrual cycle and an increase in intrafollicular progesterone concentration occurs in parallel to an increase in follicular diameter (Schneyer et al., 2000; Yding Andersen et al., 2010). Accumulation of ovarian-derived progesterone is regulated by the amount being synthesized and the amount being further converted into down-stream metabolites. Both theca and granulosa cells have the capacity to synthesize progesterone, since both cell types express P450scc and 3bHSD (Payne and Hales, 2004). The down-stream processing of progesterone is performed by P450c17 (17R-hydroxylase/17,20-lyase cytochrome P450) enzyme, which is exclusively located in the theca cells (Smyth et al., 1993). This enzyme catalyzes two reactions: 17a-hydroxylation and 17,20 bond scission. Although it is only one enzyme, these two reactions are independently regulated, so that different requirements are needed for each reaction to be performed (Miller, 2008). These requirements are different from species to species, but in humans it appears that P450c17 hardly catalyzes the 17,20 scission when the starting material is derivatives from the D4 pathway of ovarian steroidogenesis (i.e. progesterone or 17-hydroxyprogesterone). Actually, it appears that the vast majority of all androgens produced by the ovary in humans go through the D5 pathway (Auchus et al., 1998; Brock and Waterman, 1999; Miller 2008), which is the pathway involving pregnenolone, 17-hydroxypregnenolone and dehydroepiandrosterone. In essence, the human P450c17 preference for D5 metabolites excludes the theca cells from converting progesterone into androgens, because progesterone is not back-converted into pregnenolone from the D4 pathway. Once the progesterone has been produced by either the theca or granulosa cells it cannot be converted into androgens and subsequent oestradiol. If anything it might be converted into 17-hydroxyprogesterone, which also possesses progesterone activity. Thus, it may be a misconception to believe that granulosa cell-derived progesterone can be converted into androgens in the theca cell compartment as recently suggested (Bosch et al., 2010; Fleming and Jenkins, 2010). In humans, the presence of progesterone in the follicular phase of the menstrual cycle in humans is a natural phenomenon and a moderate concentration does not indicate early luteinization (Schneyer et al., 2000). On the contrary, it is known that progesterone has physiological responsibilities in the follicular phase of the human menstrual cycle, as for instance progesterone is mandatory for the mid-cycle surge of gonadotrophins to be initiated. Given the difference in the results of a large number of studies and given the fact that most studies have not accounted for the presence of LH activity, it appears plausible that progesterone concentration in the late follicular phase is secondary to effects exerted by the combined action of LH and FSH on both theca and granulosa cells and that the progesterone concentration as such does not reflect the subsequent reproductive outcome. In conclusion, in this large retrospective analysis, preovulatory progesterone concentration was significantly associated with LH concentration but unrelated to biochemical or clinical pregnancy rates in women undergoing ovarian stimulation for IVF/ICSI. Further, the results showed that exogenously added rlh did not affect late-follicular-phase progesterone concentration. Collectively, these results question whether reduced endometrial receptivity, induced by elevated progesterone concentration in the late follicular phase, at least in connection with pituitary downregulation using GnRH agonist, has any influence on the reproductive outcome in patients undergoing IVF/ICSI. Acknowledgements The technical assistance by Tiny Roed is gratefully acknowledged. Analysis of the progesterone concentration in the serum samples was performed through an unrestricted grant from Merck-Serono. References Abuzeid, M.I., Sasy, M.A., 1996. Elevated progesterone levels in the late follicular phase do not predict success of in vitro fertilization-embryo transfer. Fertil. Steril. 65, 981 985. Auchus, R.J., Lee, T.C., Miller, W.L., 1998. Cytochrome b5 augments the 17,20-lyase activity of human P450c17 without direct electron transfer. J. Biol. Chem. 273, 3158 3165. Azem, F., Lessing, J.B., Malcov, M., Ben-Yosef, D., Almog, B., Amit, A., 2008. Does high serum progesterone level on the day of human chorionic gonadotropin administration affect pregnancy rate after intracytoplasmic sperm injection and embryo transfer? Gynecol. Endocrinol. 24, 368 372. Bosch, E., Labarta, E., Crespo, J., Simón, C., Remohí, J., Jenkins, J., Pellicer, A., 2010. Circulating progesterone levels and ongoing pregnancy rates in controlled ovarian stimulation cycles for in vitro fertilization: analysis of over 4000 cycles. Hum. Reprod. 25, 2092 2100. Bosch, E., Valencia, I., Escudero, E., Crespo, J., Simon, C., Remohi, J., Pellicer, A., 2003. Premature luteinization during gonadotropin-releasing hormone antagonist cycles and its relationship with in vitro fertilization outcome. Fertil. Steril. 80, 1444 1449. Brock, B.J., Waterman, M.R., 1999. Biochemical differences between rat and human cytochrome P450c17 support the

194 C Yding Andersen et al. different steroidogenic needs of these two species. Biochemistry 38, 1598 1606. Bustillo, M., Stern, J.J., Coulam, C.B., 1995. Serum progesterone at the time of human chorionic gonadotrophin does not predict pregnancy in in-vitro fertilization and embryo transfer. Hum. Reprod. 10, 2862 2867. Check, J.H., 1994. Predictive value of serum progesterone levels for pregnancy outcome? Fertil. Steril. 62, 1090 1091. Check, J.H., Lurie, D., Askari, H.A., Hoover, L., Lauer, C., 1993. The range of subtle rise in serum progesterone levels following controlled ovarian hyperstimulation associated with lower in vitro fertilization pregnancyrates is determined by the source of manufacturer. Eur. J. Obstet. Gynecol. Reprod. Biol. 52, 205 220. Doldi, N., Marsiglio, E., Destefani, A., Gessi, A., Merati, G., Ferrari, A., 1999. Elevated serum progesterone on the day of HCG administration in IVF is associated with a higher pregnancy rate in polycystic ovary syndrome. Hum. Reprod. 14, 601 605. Edelstein, M.C., Seltman, H.J., Cox, B.J., Robinson, S.M., Shaw, R.A., Muasher, S.J., 1990. Progesterone levels on the day of human chorionic gonadotropins administration in cycles with gonadotropin-releasing hormone agonist suppression are not predictive of pregnancy outcome. Fertil. Steril. 54, 853 857. Fanchin, R., de Ziegler, D., Taieb, J., Hazout, A., Frydman, R., 1993. Premature elevation of plasma progesterone alters pregnancy rates of in vitro fertilization and embryo transfer. Fertil. Steril. 59, 1090 1094. Fanchin, R., Righini, C., Olivennes, F., Ferreira, A.L., de Ziegler, D., Frydman, R., 1997a. Consequences of premature progesterone elevation on the outcome of in vitro fertilization: insights into a controversy. Fertil. Steril. 68, 799 805. Fanchin, R., Hourvitz, A., Olivennes, F., Taieb, J., Hazout, A., Frydman, R., 1997b. Premature progesterone elevation spares blastulation but not pregnancy rates in in vitro fertilization with coculture. Fertil. Steril. 68, 648 652. Fleming, R., Jenkins, J., 2010. The source and implications of progesterone rise during the follicular phase of assisted reproduction cycles. Reprod. Biomed. Online 21, 446 449. Givens, C.R., Schriock, E.D., Dandekar, P.V., Martin, M.C., 1994. Elevated serum progesterone levels on the day of human chorionic gonadotropins administration do not predict outcome in assisted reproduction cycles. Fertil. Steril. 62, 1011 1017. Harada, T., Katagiri, C., Takao, N., Toda, T., Mio, Y., Terakawa, N., 1996. Altering the timing of human chorionic gonadotropin injection according to serum progesterone (P) concentrations improves embryo quality in cycles with subtle P rise. Fertil. Steril. 65, 594 597. Hofmann, G.E., Bentzien, F., Bergh, P.A., Garrisi, G.J., Williams, M.C., Guzman, I., Navot, D., 1993. Premature luteinization in controlled ovarian hyperstimulation has no adverse effect on oocyte and embryo quality. Fertil. Steril. 60, 675 679. Levy, M.J., Smotrich, D.B., Widra, E.A., Sagoskin, A.W., Murray, D.L., Hall, J.L., 1995. The predictive value of serum progesterone and 17-OH progesterone levels on in vitro fertilization outcome. J. Assist. Reprod. Genet. 12, 161 166. Martinez, F., Coroleu, B., Clua, E., Tur, R., Buxaderas, R., Parera, N., Barri, P.N., Balasch, J., 2004. Serum progesterone concentrations on the day of HCG administration cannot predict pregnancy in assisted reproduction cycles. Reprod Biomed Online 8, 183 190. Miller, W.L., 2008. Steroidogenic enzymes. In: Flück, C.E., Miller, W.E. (Eds.), Disorders of the human Adrenal cortex, vol. 13. Endocr. Dev., Basel Karger, pp. 1 18. Miller, K.F., Behnke, E.J., Arciaga, R.L., Goldberg, J.M., Chin, N.W., Awadalla, S.G., 1996. The significance of elevated progesterone at the time of administration of human chorionic gonadotropin may be related to luteal support. J. Assist. Reprod. Genet. 13, 698 701. Moffitt, D.V., Queenan Jr., J.T., Shaw, R., Muasher, S.J., 1997. Progesterone levels on the day of human chorionic gonadotropin do not predict pregnancy outcome from the transfer of fresh or cryopreserved embryos from the same cohort. Fertil. Steril. 67, 296 301. Niu, Z., Feng, Y., Zhang, A., Sun, Y., Zhang, H., 2008. Progesterone levels on oocyte retrieval day can predict the quantity of viable embryos but not pregnancy outcome of intracytoplasmic sperm injection. Gynecol. Endocrinol. 24, 452 458. Nyboe Andersen, A., Humaidan, P., Fried, G., Hausken, J., Antila, L., Bangsbøll, S., Rasmussen, P.E., Lindenberg, S., Bredkjaer, H.E., Meinertz, H.Nordic LH study group, 2008. Recombinant LH supplementation to recombinant FSH during the final days of controlled ovarian stimulation for in vitro fertilization. A multicentre, prospective, randomized, controlled trial. Hum Reprod. 23, 427 434. Nyboe Andersen, A., Devroey, P., Arce, J.C., 2006. Clinical outcome following stimulation with highly purified hmg or recombinant FSH in patients undergoing IVF: a randomized assessor-blind controlled trial. Hum. Reprod. 21, 3217 3227. Payne, A.H., Hales, D.B., 2004. Overview of steroidogenic enzymes in the pathway from cholesterol to active steroid hormones. Endocr. Rev. 25, 947 970, Review. PMID: 15583024. Schneyer, A.L., Fujiwara, T., Fox, J., Welt, C.K., Adams, J., Messerlian, G.M., Taylor, A.E., 2000. Dynamic changes in the intrafollicular inhibin/activin/follistatin axis during human follicular development: relationship to circulating hormone concentrations. J. Clin. Endocrinol. Metab. 85, 3319 3330. Shulman, A., Ghetler, Y., Beyth, Y., Ben-Nun, I., 1996. The significance of an early (premature) rise of plasma progesterone in in vitro fertilization cycles induced by a long protocol of gonadotropin releasing hormone analogue and human menopausal gonadotropins. J. Assist. Reprod. Genet. 13, 207 211. Silverberg, K.M., Burns, W.N., Olive, D.L., Riehl, R.M., Schenken, R.S., 1991. Serum progesterone levels predict success of in vitro fertilization/embryotransfer in patients stimulated with leuprolide acetate and human menopausal gonadotropins. J. Clin. Endocrinol. Metab. 73, 797 803. Smitz, J., Andersen, A.N., Devroey, P., Arce, J.C., 2007. Endocrine profile in serum and follicular fluid differs after ovarian stimulation with HP-hMG or recombinant FSH in IVF patients. Hum. Reprod. 22, 676 687. Smyth, C.D., Miró, F., Whitelaw, P.F., Howles, C.M., Hillier, S.G., 1993. Ovarian thecal/interstitial androgen synthesis is enhanced by a follicle-stimulating hormone-stimulated paracrine mechanism. Endocrinology 133, 1532 1538. Ubaldi, F., Smitz, J., Wisanto, A., Joris, H., Schiettecatte, J., Derde, M.P., Borkham, E., Van Steirteghem, A., Devroey, P., 1995. Oocyte and embryo quality as well as pregnancy rate in intracytoplasmic sperm injection are not affected by high follicular phase serum progesterone. Hum Reprod 10, 3091 3096. Urman, B., Alatas, C., Aksoy, S., Mercan, R., Isiklar, A., Balaban, B., 1999. Elevated serum progesterone level on the day of human chorionic gonadotropin administration does not adversely affect implantation rates after intracytoplasmic sperm injection and embryo transfer. Fertil. Steril. 72, 975 979. Venetis, C.A., Kolibianakis, E.M., Papanikolaou, E., Bontis, J., Devroey, P., Tarlatzis, B.C., 2007. Is progesterone elevation on the day of human chorionic gonadotrophin administration associated with the probability of pregnancy in in vitro fertilization? A systematic review and meta-analysis. Hum. Reprod. Update 13, 343 355. Yding Andersen, C., Schmidt, K.T., Kristensen, S.G., Rosendahl, M., Byskov, A.G., Ernst, E., 2010. Concentrations of AMH and

Late-follicular-phase progesterone does not impact subsequent pregnancy rate 195 inhibin-b in relation to follicular diameter in normal human small antral follicles. Hum. Reprod. 25, 1282 1287. Yu, F.Q., Han, C.S., Yang, W., Jin, X., Hu, Z.Y., Liu, Y.X., 2005. Activation of the p38 MAPK pathway by follicle-stimulating hormone regulates steroidogenesis in granulosa cells differentially. J. Endocrinol. 186, 85 96, PMID: 16002539. Declaration: The authors report no financial or commercial conflicts of interest. Received 17 November 2010; refereed 7 April 2011; accepted 7 April 2011.