Question: Can implantation failure (IF) be predicted prior to in vitro fertilization (IVF)? A pilot, non-interventional, clinical study Prospective, controlled, cohort study

The investigators hypothesize that TNFα and MFG-E8 cooperatively maintain the integrity of the normal endometrium, and that in patients with IF or with recurrent pregnancy loss (RPL) of unexplained origin, excessive TNFα increases the maternal shedding of MFG-E8, disrupting the normal protective effect of this protein, resulting in damage of the endometrial epithelium and impairing trophoblast invasion. The investigators propose that these molecules can be measured in local tissue (endometrium) as well as in serum as a reflection of increased inflammation and can therefore be used as markers of implantation and its failure. The investigators hypothesis is that TNFα is up-regulated in serum of women with implantation defects (IF and RM of unexplained origin) and this causes perturbation of MFG-E8 secretion. These results will provide first evidence demonstrating endometrial dysfunctions resulting from over-expression of pro-inflammatory molecules. The release of significant quantities of these molecules into serum broadens the maternal-fetal interface beyond the uterus and into the maternal circulation. The characterization of these molecules is essential to better understand, not only their biological effects, but also their potential as prognostic and diagnostic biomarkers for detection of IF. More importantly, these molecules could be potential therapeutic targets (perhaps intrauterine infusion of MFG-E8 for tissue repair, and/or oral administration of TNFα antagonists) to improve implantation outcomes. Two recent studies have reported the successful therapeutic use of recombinant human MFG-E8 (rhMFG-E8) in animals: one was used to decrease intestinal injury after whole body radiation (Ajakaiye et al, 2012), and the other was to mitigate inflammation and tissue injury after hemorrhagic stroke (Wang et al, 2012). Also, TNFα inhibitors have been shown to significantly increase IVF outcome in infertile patients (Winger et al, 2009 and 2008). In summary, these two pivotal studies show promising clinical uses of rhMFG-E8 in tissue repair/remodeling by decreasing apoptosis, and provide basis for their use as candidates for further clinical development in reproductive health.

Objectives: Here, the investigators seek to delineate novel diagnostic methods with the ultimate long term goal of designing directed therapies to improve embryo implantation competence.

• The specific aim is to quantify and correlate MFG-E8 and TNFα in serum and in endometrial biopsies of women in 3 groups: controls, IF and RM.
• The ultimate goal of this study is to provide data leading to a simple and quick test to measure soluble serum markers of IF/RM in order to discriminate prospectively between these groups of patients.
• To translate these findings into the use of modulators of these inflammatory molecules to improve pregnancy rates in patients with predicted implantation defects. It is widely known that "classic" markers of endometrial receptivity such as those provided by an invasive endometrial biopsy (with Hematoxilin-Eosin staining, or immuhistochemistry, or even with microarray technology) are not reliable to establish clear-cut diagnosis due to cycle-to-cycle variations and overlap of results.

Inclusion Criteria:

• women 21-35y
• with no contraindication to pregnancy
• on no hormonal contraception
• normal uterine cavity
• absence of hydrosalpinges
• normal ovarian reserve (FSH<13 mIU/.ml or AMH>1 ng/ml or normal antral follicular count>10 total follicles <10 mm in diameter, American Society for Reproductive Medicine-ASRM 2012).
• Three groups of patients will be enrolled: 10 fertile controls (C1-10: women who are participating in the donor egg program as egg donors), 10 with unexplained implantation failure (IF1-10: patients who have failed IVF cycles -failure of implantation following 2 or more cycles with transfer of embryos of good quality-at least 2 cleaving embryos with 6-8 cells, grades 1- 3 (scale 1-highest and 5-poorest morphology score), and normal uterus, and 10 with recurrent unexplained first trimester miscarriages (RM1-10: 2 consecutive miscarriages under 10 weeks, after spontaneous or IVF conceptions).

Exclusion Criteria:

• contraindication to pregnancy
• use of hormonal contraception
• abnormal uterine cavity
• hydrosalpinges
• abnormal karyotype.
• women who are pregnant or planning to get pregnant in the cycle in which they are participating in this study
• women who have had a hysterectomy (previous removal of your uterus)
• have a current medical condition, which, in the opinion of the investigator, would result in hazards to the subject, should she participate in the study or have a current medical condition which would potentially confound the study results.

women 21-35 years old attending an infertility clinic in an academic center