Combined assessment of polymorphisms in the LHCGR and FSHR genes predict chance of pregnancy after in vitro fertilization.

Abstract

Can gonadotrophin receptor variants separately or in combination, be used for the prediction of pregnancy chances in in vitro fertilization (IVF) trials? The luteinizing hormone/human chorionic gonadotrophin receptor (LHCGR) variant N312S and the follicle-stimulating hormone receptor (FSHR) variant N680S can be utilized for the prediction of pregnancy chances in women undergoing IVF. The FSHR N680S polymorphism has been shown to affect the ovarian response in response to gonadotrophin treatment, while no information is currently available regarding variants of the LHCGR in this context. Cross-sectional study, duration from September 2010 to February 2015. Women undergoing IVF were consecutively enrolled and genetic variants compared between those who became pregnant and those who did not. The study was subsequently replicated in an independent sample. Granulosa cells from a subset of women were investigated regarding functionality of the genetic variants. Women undergoing IVF (n = 384) were enrolled in the study and genotyped. Clinical variables were retrieved from medical records. For replication, an additional group of n = 233 women was utilized. Granulosa cells from n = 135 women were isolated by flow cytometry, stimulated with Follitropin alpha or Menotropin, and the downstream targets 3',5'-cyclic adenosine monophosphate (cAMP) and inositol 1,4,5-trisphosphate (IP3) measured with enzyme-linked immunosorbent assay. Women homozygous for serine (S) in both polymorphisms displayed higher pregnancy rates than women homozygous asparagine (N) (OR = 14.4, 95% CI: [1.65, 126], P = 0.016). Higher pregnancy rates were also evident for women carrying LHCGR S312, regardless of FSHR variant (OR = 1.61, 95% CI: [1.13, 2.29], P = 0.008). These women required higher doses of FSH for follicle recruitment than women homozygous N (161 versus 148 IU, P = 0.030). When combining the study cohort with the replication cohort (n = 606), even stronger associations with pregnancy rates were noted for the combined genotypes (OR = 11.5, 95% CI: [1.86, 71.0], P = 0.009) and for women carrying LHCGR S312 (OR = 1.49, 95% CI: [1.14, 1.96], P = 0.004). A linear significant trend with pregnancy rate and increasing number of G alleles was also evident in the merged study population (OR = 1.34, 95% CI: [1.10, 1.64], P = 0.004). A lower cAMP response in granulosa cells was noted following Follitropin alpha stimulation for women homozygous N in both polymorphisms, compared with women with other genotypes (0.901 pmol cAMP/mg total protein versus 2.19 pmol cAMP/mg total protein, P = 0.035). Due to racial differences in LHCGR genotype distribution, these results may not be applicable for all populations. Despite that >250 000 cycles of gonadotrophin stimulations are performed annually worldwide prior to IVF, it has not been possible to predict neither the pregnancy outcome, nor the response to the hormone with accuracy. If LHCGR and FSHR variants are recognized as biomarkers for chance of pregnancy, more individualized and thereby more efficient treatment modalities can be developed. This work was supported by Interreg IV A, EU (grant 167158) and ALF governments grant (F2014/354). Merck-Serono (Darmstadt, Germany) supported the enrollment of the subjects. The authors declare no conflict of interest.