IVF (in vitro fertilization) cycles fails more often than they succeed. Surprisingly very little effort is invested in defining the reasons for failure and possibly finding ways to improve the success on the next cycle. The investigators believe that the main reasons for failure are related to oocyte quality and indirectly to the follicle response for a particular patient. The investigators have developed a panel of biomarkers to assess the faulty follicular conditions leading to lower oocyte quality. Using these markers would indicate if a given cycle was characterized by over growth, over-luteinization, early or late trigger. Indeed our transcriptomics analysis has identified biomarkers of follicles still in their growth phase at trigger or follicles that have already begun luteinisation compare to follicle that are at the optimal level of differentiation. Measuring these biomarkers would allow making a better diagnostic for a given patient and potentially explaining reasons for failure. The system would also become adjustable to variable COS (control ovarian stimulation) and individual clinical practices. It is important to realize that this is applicable to almost all cycle failure and can be done on a pool of follicular cells when none of the oocytes obtained has led to a pregnancy. This does not resolve uterine problems but often these are caused by hormonal conditions established by the ovary or the ovarian treatment. This technology can be applied in all IVF clinics as no special equipment is required. It would be particularly valuable in clinics where a number of cycles is limited due to funding, or in clinic where a package of 3 cycles is proposed to the patient. The patient interest to have a custom treatment increases at each failing cycle as well as the doctors' interest to succeed. This technology is not clinically validated yet and would require a period of testing where participating clinics will collect the samples for a retrospective analysis (presence of biomarkers of follicular problems vs outcome) then in a prospective analysis where the diagnostic is used in a sub-set of patient to modulate the second/third cycle compared the outcome to patient with no diagnostic. The increase in pregnancy rate or cumulative pregnancy rate should reach a minimum of 10 and 25 % respectively to indicate a significant value.

IVF (in vitro fertilization) cycles fails more often than they succeed. Surprisingly very little effort is invested in defining the reasons for failure and possibly finding ways to improve the success on the next cycle. We believe that the main reasons for failure are related to oocyte quality and indirectly to the follicle response for a particular patient. We have developed a panel of biomarkers to assess the faulty follicular conditions leading to lower oocyte quality. Using these markers would indicate if a given cycle was characterized by over growth, over-luteinization, early or late trigger. Indeed our transcriptomics analysis has identified biomarkers of follicles still in their growth phase at trigger or follicles that have already begun luteinisation compare to follicle that are at the optimal level of differentiation. Measuring these biomarkers would allow making a better diagnostic for a given patient and potentially explaining reasons for failure. The system would also become adjustable to variable COS (control ovarian stimulation) and individual clinical practices. It is important to realize that this is applicable to almost all cycle failure and can be done on a pool of follicular cells when none of the oocytes obtained has led to a pregnancy. This does not resolve uterine problems but often these are caused by hormonal conditions established by the ovary or the ovarian treatment.

This technology can be applied in all IVF clinics as no special equipment is required. It would be particularly valuable in clinics where a number of cycles is limited due to funding, or in clinic where a package of 3 cycles is proposed to the patient. The patient interest to have a custom treatment increases at each failing cycle as well as the doctors' interest to succeed.

This technology is not clinically validated yet and would require a period of testing where participating clinics will collect the samples for a retrospective analysis (presence of biomarkers of follicular problems vs outcome).

Five Canadian IVF clinics have agreed to provide a minimum of 200 anonymized samples with information on the cycle/patient medical status. From these samples, RNA will be extracted and the gene expression level will be measured for 3 housekeeping genes and 21 markers derived from previous studies (Hamel et al 2008 and 2010, Nivet et al 2015 submitted).

The genes leading to the best predictive value of negative outcome will then be classified according to our understanding of ovarian physiology as key indicator for generating a diagnostic. The read-outs provided to the clinic would then be one of the 5 following: Too early trigger, too late trigger, too high stimulation (FSH), too high differentiation (LH) or time spread response (a combination of immature and over mature follicles)

Inclusion Criteria:

• female Infertility

Exclusion Criteria:

• PCO (polycystic ovary) syndrome women over 42 male factor resulting in sperm samples less than 5 millions motile.

IVF (In vitro fertilisation) patients infertile patients