In the present study, we compared the OS outcomes of two LH activity preparations- recombinant LH and HP-hMG, in which the LH activity is derived mostly from hCG. Patients treated with rFSH + rLH yielded significantly higher numbers of mature oocytes and fertilized oocytes, with non-significantly lower pregnancy rate per transfer (15% vs 29%, respectively. P = 0.3) compared to those treated with HP-hMG.
LH and hCG are heterodimeric glycoprotein hormones sharing approximately 85% structural identity. Both of them bind to a mutual LH/choriogonadotropin receptor (LH/CGR), a G protein-coupled receptor with an extra-cellular binding domain to which LH and hCG bind in distinct, specific regions. The distinct molecular structure of the beta-LH and beta-hCG subunits results in different conformational changes of the LH/CGR, leading to activation of different intracellular cascades (5). The post receptor, intra-cellular reaction, includes an activation of adenylate cyclase, which increases the intracellular pool of cAMP and results in steroidogenesis, and an extracellular signal-regulated protein kinases 1 and 2 (ERK1/2) and AKT, with putative roles in cell proliferation, differentiation and survival (7). In vitro studies found that hCG is five times more potent than LH with respect of cAMP production and steroidogenesis pathway, while LH strongly activates ERK1/2 and AKT pathways, thus induces proliferation and anti-apoptotic effect (6, 15). Although in vitro models provided the evidence of hormone- specific actions, whether they influence differently on in vivo OS response remains unclear and published evidence of the difference between rLH and hCG during OS is surprisingly scarce.
The main findings of our study were higher number of mature oocytes retrieved and higher number of oocytes fertilized in the rFSH + rLH compared with the HP-hMG treatment cycles. These differences confirm that rLH and hCG poses different influences also in vivo during OS cycles. The higher number of mature oocytes may be due to the synergistic effect of FSH and LH on follicular growth, proliferation and maturation. Moreover, the addition of LH shifts granulosa cells from proapoptotic to proliferative pathways, in contrast to hCG which has proapoptotic effect mediated by relatively higher intracellular cAMP concentrations (16, 17). Gomez-Palomares et al evaluated the OS outcomes in women over 38 years old supplemented with rFSH + rLH or hMG(18). Similar to our results, they found a higher rate of mature oocytes in the group treated with rFSH + rLH. Pacchiarotti et al in their randomized controlled trial comparing rFSH + rLH with hMG, also found higher number of oocytes retrieved and higher number of mature oocytes in the rFSH + rLH cycles, although the rate of matured oocytes per overall oocytes yield was higher in the hMG cycles (19). Unlike our study that included down regulation by GnRH antagonist, Pacchiarotti et al used the long GnRH agonist down regulation protocol, which may influence differently the initial follicular reaction to rLH or hCG.
When comparing the IVF treatment outcomes of rFSH + rLH to HP-hMG, we must also consider the differences in the type of FSH molecules, which may be related to the variation observed between the two treatments. The higher mature oocytes yield demonstrated in the rFSH + rLH treatment cycles, may be derived from the greater effectiveness of the rFSH isoform compared with urinary FSH, rather than from the effect induced by rLH, as it is well-established that rFSH leads to higher follicular recruitment compared with HP-hMG (1, 8, 20).
Another interesting finding of our study was higher progesterone levels in the rFSH + rLH cycles compared with the HP-hMG cycles. This finding might be explained by the higher number of follicles in the rFSH + rLH treatment cycles. However, we must also consider the different effect of rLH and hCG on the steroidogenesis process as an alternative explanation for the different endocrine profile. Previous studies comparing progesterone levels following OS with rFSH or HP-hMG have revealed similar findings; rFSH stimulation alone was associated with higher progesterone levels at the end of stimulation, even after adjusting for ovarian response (1, 8, 21, 22). In accordance to our study, a study by Sebag-Peyrelecade et al which compared the progesterone levels of rFSH + rLH treatment with those of hMG treatment, revealed that supplementation of rLH was not sufficient to decrease the progesterone levels to those observed in the hMG treatment cycles, unrelated to the degree of ovarian response(21). Therefore, we may deduce that the lower progesterone levels measured in the hMG cycles are attributed to the hCG content.
In the present study, we did not find any differences regarding the rate of mature oocytes nor the number of top-quality embryos between the rFSH + rLH and HP-hMG treatments, despite of the initial higher number of fertilized oocytes in the rFSH + LH treatment cycles. This observation suggests that although rFSH + rLH is more potent in follicular genesis and oocytes yield, HP-hMG may favor embryonal maturation. In accordance with our findings, previous studies that compared the use ofHP-hMG vs rFSH, such as the MERiT and the MEGASET trials, have found that HP-hMG results in higher proportion of top-quality embryos(8, 20).
Of not, although not statistically significant, the pregnancy rate per fresh transfer was higher among the HP-hMG treatment cycles compared to the rFSH + rLH cycles. This difference in pregnancy rate may be a consequence of the higher progesterone levels during the rFSH + rLH cycles, which negatively influence the endometrial receptivity and the essential synchronization between the embryo and the endometrium. Nonetheless, because the present study is not sufficiently powered to detect differences in pregnancy rate, this data should be viewed with caution.
A major strength of our study is that we compared the different LH activity preparations in the same cohort of patients. The fact that all women that participated in our study had two consecutive treatment cycles using rFSH + rLH in one cycle and hMG in the other, helps to eliminate multiple bias factors and to attribute the study results to the different treatment preparations.
A limitation of our study is the lack of data regarding clinical pregnancy or live birth rates. Nonetheless, one may say that when the question of investigation is the effect of LH activity products on OS, the primary outcome has to be the first measurable parameter of gonadotropin influence, i.e., the ovarian response. Moreover, the majority of our study cohort comprised of patients which underwent IVF treatment for social fertility preservation. In this section of patients, we are most interested in achieving an optimal number of mature oocytes.
In conclusion, our study suggests that gonadotropins preparations have different influence on OS outcome, proving the necessity of tailoring a specific gonadotropin regimen when assembling a treatment protocol. rFSH + rLH resulted in higher number of matured oocytes and fertilized oocytes, while the lack of difference regarding the number of top-quality embryos between the preparations might suggest an encouraging effect of hMG on oocytes and embryo quality. These treatment characteristics, derived by different LH preparation, reflect the physiological role of these molecules as previously indicated by in vitro data.