{"paper_id":"6bb8096d-e832-4567-82a6-a1c7c5a4f618","body_text":"Endometriosis is characterised by the presence of endometrial glands and stroma\noutside the endometrial cavity and the uterine musculature. While its pathogenesis\nremains unclear, this disease has been extensively studied in recent years and is\none of the leading causes of the impairment of female fertility. Moreover, assisted\nreproduction techniques (ART) are one of the methods capable of reversing\ninfertility in women with endometriosis.\nRecently, some investigators described the same reproductive results in patients with\nendometriosis submitted to IVF compared to non-endometriotic patients, despite the\nnumber of related abnormalities in those infertile patients with endometriosis\ndescribed in the literature ( Giudice, 2010 ;\n Opøien  et al. \n2012 ;  Barbosa et al., 2014 ).\nSoluble human leukocyte antigen-G (sHLA-G) is a molecule described to be linked to\nimmunological tolerance of the semi-allogeneic fetus at the maternal-fetal interface\nduring pregnancy as well as the activation of NK cells and even embryo quality\nresponse ( Hunt  et al ., 2000 ;\n Noci et al., 2005 ;  Jee et al., 2011 ;  Rizzo et al.,\n2011 ;  Wunder et al., 2013 ).\nsHLA-G has an important role in the activation of natural killer (NK) cells,\nMoreover, some authors associated sHLA-G in patients with endometriosis (mainly, the\nmore severe forms) and adenomyosis ( Hunt  et\nal. , 2000 ;  Wang et al.,\n2008 ;  Rached et al., 2019 ).\nHowever, the literature is completely absent in terms of the role of sHLA-G in\ninfertile patients with endometriosis and the concentration of this molecule in the\nfollicular fluid during controlled ovarian stimulation for IVF in this group of\nwomen.\nEndometriosis is not associated with altered embryo quality. However, some recente\npapers described that granulosa (cumulus) cell function could be altered in these\nwomen with peritoneal endometriosis submitted to IVF ( De Conto  et al.,  2017 ;  2021 ;  Caran et al.,\n2021 ).\nsHLA-G was also described as a marker for oocyte and embryo quality, and some authors\nlinked this molecule to pregnancy rates ( Rebmann\n et al.,  2010 ). Furthermore, follicular fluid is a\ncrucial environment to analyse and describe the association of sHLA-G and some\nreproductive outcomes in infertile patients with endometriosis ( Wunder  et al.,  2013 ).\nConsidering the link of endometriosis with granulosa cell dysfunction and the\npossible role of sHLA-G as a marker for oocyte quality, the rationale of this study\nis to investigate the association between the concentrations of sHLA-G detected in\nthe follicular fluid of infertile patients with peritoneal endometriosis submitted\nto IVF compared to non-endometriotic patients (tubal factor) as the primary outcome\nand its effect in some important reproductive outcomes.\n\nA cross-sectional study was carried out from November 2016 to April 2021 at the\nHuman Reproduction Center.\nOne hundred-nine patients seeking infertility treatment (IVF) for the first time\nwere included.\nEligibility criteria used in the study were age ≤ 40 years, presence of\nboth ovaries, hormone levels of TSH, FSH and PRL within the reference values and\nindication of in vitro fertilization procedure due to tubal factor or\nendometriosis, previously diagnosed by video laparoscopy, which was done at\nleast six months before IVF (peritoneal endometriosis confirmed by biopsy and\ncauterised).\nExclusion criteria were: ovarian hyperstimulation syndrome in the evaluated\ncycle, autoimmune disease, polycystic ovary syndrome, early luteinisation,\nendometrioma and presence of blood after follicular fluid centrifugation.\nMoreover, male partner semen analysis should be normal during semen preparation\nfor IVF.\nWe considered 109 patients with eligibility criteria, 13 patients were excluded;\nfinally, 96 patients were included in this research, as demonstrated in  Figure 1 . We subdivided into two groups,\naccording to the cause of infertility-that is, with peritoneal endometriosis\n(n=46) and without endometriosis (n=50) -to assess the association between\nsHLA-G levels and embryonic quality.\nFigure 1 Follicular Fluid concentration of sHLA-G (IU/ml): a) in peritoneal\nendometriosis vs control group (non-endometriosis),\n p =0.682; b)comparing per follicle sHLA-G\nconcentration between both groups,  p =0.955 and (c)\nper embryo sHLA-G concentration ( p =0.857), Mann-\nWhitney-U test.\nFollicular Fluid concentration of sHLA-G (IU/ml): a) in peritoneal\nendometriosis vs control group (non-endometriosis),\n p =0.682; b)comparing per follicle sHLA-G\nconcentration between both groups,  p =0.955 and (c)\nper embryo sHLA-G concentration ( p =0.857), Mann-\nWhitney-U test.\nThe presence or absence of endometriosis was verified by laparoscopy and biopsy,\nperformed in the last 6 months before in vitro fertilization, and all included\npatients were diagnosticated with peritoneal endometriosis phenotype according\nto the European Society of Human Reproduction and Embryology ( Kennedy  et al.,  2005 ).\nThe control group was formed by patients with only tubal fator as the infertility\netiology. All patients in this group underwent laparoscopy, male analysis,\nhormonal screening for thyroid and propactin, regular menstrual cycle (21-35\ndays), age ≤ 40 years and presence of both ovaries.\nHormonal dosages TSH, FSH and PRL were requested for the patients to be evaluated\nbefore beginning the cycle of assisted reproduction and were determined by\nchemiluminescence immunoassay (Siemens Om-MA Immulite 2000, Munich, Germany).\nSerum levels of AMH were determined by and enzyme-linked immunosorbent assay\n(Beckman Coulter, Inc., Brea, CA, USA).\nThese results were transcribed from the medical records, and only those whose\nserum measurement was performed on the third day of the cycle were accepted.\nAll patients were submitted to ovarian stimulation according to standard\nprotocols. The therapeutic regimen for oocyte stimulation consisted of a GnRH\nantagonist for pituitary suppression with the administration of recombinant\nfollicle-stimulating hormone (recombinant FSH) for ovarian stimulation.\nFollicular growth was monitored by transvaginal ultrasound, and when at least\nthree follicles reached a diameter ≥ 17mm, the administration of\nchorionic gonadotropin (hCG) was determined to induce ovulation. Transvaginal\nfollicular aspiration was performed 36 hours after the administration of hCG\nunder routine intravenous sedation.\nAt the time of oocyte laboratory recovery, follicular fluid was collected from a\npool of punctured follicles (we only included in this study follicles superior\nto 17 mm with MII oocytes) for each cycle of the respective patient, centrifuged\nfor 5 minutes at 1300 rpm, and, in the absence of blood, stored at -20°C for\nposterior analysis.\nConventional IVF (in vitro fertilization) was performed 3 hours post\npuncture.\nEmbryo score (GES, Graduated Embryo Score) criteria were used for assessing\nEmbryonic Quality and Calculating the Average Embryo Score Generated). The\nevaluation of embryonic quality was carried out in three stages, from which\nscores ranging from 0 to 100 points were obtained. An embryo with 100 points is\nclassified according to this criterion with the best morphological quality\n( Fisch  et al., \n2001 ).\nThe first moment of evaluation occurs 16-18 hours after IVF. At this moment,\nfertilization is confirmed by visualising the female and male pro-nuclei (PNs)\nand the presence of 2 polar corpuscles in the perivitelline space. The second\nmoment occurs 25-27 hours after IVF and evaluates the early cleavage (presence\nof two cells), and the third moment of evaluation occurs 64-67 hours after IVF.\nIn these three moments, the following parameters were evaluated: cell division\nand the presence of fragmentation with an embryo with maximum quality presenting\n6 to 8 blastomeres of uniform size and no fragmentation at this time.\nAfter classifying the scores for each embryo generated from each cycle, the\narithmetic mean between these scores was performed. In this study, the quality\nof the embryo was assessed by the average embryo score (EMEG) whose calculation\nwas performed as follows: EMEG = (sum of scores of N embryos generated)/N\n(numbers of embryos generated).\nThe levels of sHLA-G in the follicular fluid were measured by the enzyme-linked\nimmunosorbent assay (ELISA) technique using the ELISA Kit (RD194070100R, Bio\nVenda, Czech Republic). The kit with the anti-sHLA-G monoclonal antibody can\nidentify and measure the soluble isoforms HLA-G1 quantitatively (by proteolytic\ncleavage) and HLA-G5.\nFollowing the recommendations of the kit manufacturer, the samples were not\ndiluted. The 16-20 hour incubation period between the FF sample and the\nanti-HLA-G monoclonal antibodies is an advantage of the kit since the literature\ndescribes considerably shorter periods, and a more extended incubation period\nallows greater test sensitivity (with a detection limit of 0.6 Units/ml). The\nplates were read at 450 nm, and the concentration of each sample analysed was\ncalculated by the ELISA (Biotek ELX 800) that builds a calibration curve formed\nbetween the absorbance of the sample (Y-axis) and the concentration of the\ncalibrators and samples tested (X-axis). The measured HLA-G concentration value\nis given in units per mL.\nThe Ethics Committee approved this study under the number: 26453514.1.0000.5327.\nSTROBE guidelines for observational studies were utilised for this research\n( von Elm  et al., \n2008 ).\nData analysis was performed using chi-square or Fisher’s exact tests for\ncategorical data. Continuous variables were compared with Student’s t-test for\nparametric data, and Mann-Whitney-U was used for non-parametric data. Multiple\ncomparisons were made using linear regression to address potential confounders.\nFurthermore, for correlations, we utilised Spearmen/Pearson tests.\nSeventy patients were required to have an 80% chance of detecting, as significant\nat the 5% level, a decrease in the primary outcome measure from 4.35 in the\ncontrol group to 3 in the experimental group ( Rizzo  et al.,  2007 ).\nWe also analysed the s-HLAG for follicles (dividing sHLA-G/number of follicles)\nand sHLA-G per embryo (total number of embryo/sHLA-G). These analyses were\nperformed using the statistical program Statistical Package for the Social\nScience (SPSS) 21.0, and the data analysis was considered statistically\nsignificant when  p <0.05.\n\nBoth groups were comparable in terms of age, AMH and reproductive outcomes ( Table 1 ). Pregnancy rates were also not\ndifferent between the groups ( p =0.945) as the embryo score per\ntransfer or the median embryo obtained per patient ( Table 1 ). The association (Spearman test) between the following\nvariables was analysed: sHLA-G and AMH, age, number of collected MII, number of\nfollicles, number of embryos and embryo score. All analyses did not demonstrate\nstatistical significance ( p >0.05). Moreover, the multivariable\nanalysis reinforced that sHLA-G was not related to any reproductive variable or\noutcome (age, number of follicles, number of follicles > 17mm, pregnancy rates,\nAMH, FSH and number of mature oocytes).\nClinical characteristics and reproductive outcomes of included patients.\nVariables showed as mean±standard deviation (parametric\ndata a ) or\nmedian and 95%CI (non-parametric data b ).\nt-student test;\nMann-Whitney;\nChi-square test.\nPerforming the Mann-Whitney-U statistical test, the amount of sHLA-G measured in the\ngroup of women with peritoneal endometriosis and women without endometriosis was\nanalysed, and we found no significant difference ( p =0.682). We also\ncompared s-HLAG per follicle and per embryo between both groups of patients, and all\nanalyses were not significant ( p =0.682, 0.955 and 0.857,\nrespectively) ( Figure 1 ).\n\nWe demonstrated that peritoneal endometriosis was not associated with an altered\nsHLA-G follicular fluid concentration. Moreover, sHLA-G was not related to embryo\nquality, pregnancy rates, AMH, age or endometriosis.\nEndometriosis is a heterogeneous disease associated with infertility and one of the\nmain causes of IVF. Furthermore, several mechanisms involving granulosa cells were\ndescribed for this group of patients, involving prolactin secretion, decreased\nanti-Mullerian hormone and dysfunction in BMP-6 and SMAD4 in cumulus cells linked to\nperitoneal endometriosis ( Cunha-Filho  et\nal. , 2001 ;  2003 ;  De Conto et al., 2017 ; 2020). Nowadays the\neffect of endometriosis during an IVF cycle is disputed and has recently been\nquestioned by several authors, but it is essential to understand the role of this\ndisease during the IVF process to enhance pregnancy rates ( Opøien  et al.,  2012 ;  Barbosa et al., 2014 ;  González-Comadran et al., 2017 ;  Caran et al., 2021 ). Therefore, we decided to better understand\nthe granulosa cell (follicular compartment) of patients with peritoneal\nendometriosis submitted to IVF, analysing sHLA-G as a marker for oocyte/embryo\ncompetence and development ( Rizzo  et\nal.,  2007 ;  Jee et al.,\n2011 ;  Wunder  et al., \n2013 ). We included only peritoneal endometriosis because this specific\nphenotype was already described ( Santulli  et\nal.,  2016 ) as the most linked to infertility compared to the\nother phenotypes of endometriosis (endometrioma/deep).\nIn this study, we also analysed the association of sHLA-G with endometriosis and with\nseveral potential markers of oocyte development and reproductive outcomes. The fact\nthat sHLA-G is not related to endometriosis comes following recent papers showing\nthat endometriosis per se is not a detrimental factor for patients submitted to IVF\n( Barbosa  et al.,  2014 ;\n González-Comadran et al., 2017 ).\nFurthermore, sHLA-G seems to be an independent factor in oocyte competence/embryo\nquality, as demonstrated in a multicenter study ( Rebmann  et al.,  2010 ).\nAlso, previous studies that showed an association between HLA system and\nendometriosis focused on the peritoneal fluid compartment and the immunological role\nof HLA molecules in patients with endometriosis. Deep endometriosis is more linked\nto several immunological abnormalities, and HLA molecules may play a role in the\nactivation of NK cells in this group of patients ( Bylińska  et al.,  2018 ;  Rached et al., 2019 ;  Ścieżyńska et al.,\n2019 ). We have demonstrated that the immunological profile of peritoneal\nendometriosis is different from other phenotypes (deep and ovarian endometriosis),\nwhich could partially explain our results ( D’Hooghe\n et al.,  2001 ;  Glitz\net al., 2009 ;  Andreoli et al.,\n2011 ;  Carmona et al., 2012 ).\nFollicular fluid sHLA-G concentration was associated with oocyte competence and\nembryo fertilization but not with good-quality embryos ( Jee  et al.,  2011 ). Besides, when performing\noocyte maturation in close contact with cumulus oophorus complex (COCs), these cells\nproduced sHLA-G during the oocyte maturation process, and sHLA-G was not detected in\nsupernatants in the culture of COCs with immature oocytes, suggesting that sHLA-G is\npart of the oocyte maturation process but is not the limiting factor in that process\n( Rizzo  et al., \n2009 ).\nWe included strict criteria for embryo quality ( Fisch\n et al.,  2001 ) to improve our statistical power and\nunderstand the relationship between sHLA-G and embryo development. As mentioned\nbefore and proved by our multifactorial analysis, sHLA-G is an independent factor\n(not associated with other reproductive parameters) that has been linked to embryo\ncompetence and even pregnancy rates after IVF by one study ( Rebmann  et al.,  2010 ). However, these authors\nstressed that embryo morphology is better than sHLA-G to select embryos.\nHowever, another group of authors did not find any association between sHLA-G in\npregnant or not pregnant women after IVF ( Wunder\n et al.,  2013 ). Furthermore, our study, when\ncorrelating the average score of the generated embryos and the concentrations of\nsHLA-G in the follicular fluid, concluded that the analysis of sHLA-G in FF is a\nparameter that is not associated with embryonic quality. This result corroborates\nwith some studies in the literature. Future work could potentially utilise\nartificial intelligence or neural networks related to sHLA-G to improve embryo\nselection and better classify/choose the best embryo for transfer ( Bormann  et al.,  2020 ;  VerMilyea et al., 2020 ).\nOur study had several limitations. First, we included only peritoneal endometriosis.\nNext, some reproductive outcomes need more subjects to be analysed, and for some\noutcomes, our number included patients was limited (pregnancy rate, for example), we\ncalculated our sample size based on follicular fluid sHLA-G. Besides, sHLA-G is an\nunpractical tool, supplemented by embryo score and recently the advent of artificial\nintelligence. We included only superficial (peritoneal) endometriosis to perform a\nhomogeneous group and increase our external validation. The main objective of this\nstudy was to investigate sHLA-G. For this purpose, we calculated the sample size,\nand our research included a sufficient number of patients. However, for reproductive\noutcomes, we need to include more patients, since analysing reproductive outcomes\ntypically requires systematic reviews or international databases ( Barbosa  et al.,  2014 ;  González-Comadran et al., 2017 ). The\nusefulness of sHLA-G during daily life in a human reproduction clinic is limited and\nobsolete, but it still does not invalidate this utility in terms of a research\nprotocol.\nIn this study, when analysing the sHLA-G, a kit with a detection limit of 0.6\nUnits/ml was used and therefore offered a greater sensitivity than the ELISA Kits\ndescribed in the literature. The increase in this limit is directly related to the\nlong incubation time (16-20 hrs) of the antibodies in contact with the FF samples,\nmaking it possible to detect sHLA-G in all the analysed samples in our study. Note\nthat the importance of ELISA sHLA-G accuracy was already stressed by others ( Dahl & Hviid, 2012 ). We, therefore,\nemphasise the need for a more sensitive ELISA test both for studies with\nmeasurements in FF and embryo cultures.\n\nWe conclude that the level in the follicular fluid of sHLA-G in patients with\nsuperficial endometriosis was not altered compared to tubal factor patients\nsubmitted to IVF. Thus, we accept the null hypothesis that the concentrations of\nsHLA-G in follicular fluid in women with and without endometriosis do not\ndiffer.","source_license":"CC0","license_restricted":false}