WO2007125317A1

WO2007125317A1 - Fertility test based on measuring inhibin b, amh, and fsh

Info

Publication number: WO2007125317A1
Application number: PCT/GB2007/001514
Authority: WO
Inventors: William Ledger
Original assignee: Life Style Choices Limited
Priority date: 2006-04-27
Filing date: 2007-04-26
Publication date: 2007-11-08
Also published as: GB2451396A; GB0821417D0

Abstract

We disclose an assay to determine the fertility of a female animal subject that combines the detection of three hormones and kits for use in the assay.

Description

FERTILITY TEST BASED ON MEASURING INHIBIN B, AMH, AND FSH

The invention relates to a diagnostic test to determine the fertility of an animal subject.

It is becoming common in developed societies that couples are deferring pregnancy until later in life and it is becoming common for women to have their first child in their late 3O's or even 40's. This is due to a number of factors such as establishing a career, wanting a stable relationship, wanting to achieve financial security and being unsure of the desire for parenthood. However, it is known that female fertility decreases with age and that conception in later years can be problematic. It is estimated that the chance of becoming pregnant in any one month is about 20% in women below the age of 30, but only 5% in women over 40. Even with advanced treatments such as in vitro fertilisation, age remains an important predictor of whether a woman will become pregnant and what her chances of miscarriage are. Currently there is no reliable predictive test to determine the fertility of a woman. Typically the tests available to couples include semen analysis, and routine gynaecological procedures that include X-ray hysterosalpingogram or laparoscopy each of which will only be predicative in severe cases of failure to conceive. An important indicator of fertility is how long the woman has been trying to conceive.

The control of reproduction is mediated by the combined effects of hormones which regulate the female ovulation and menstruation. The hypothalamus and pituitary gland secrete the hormones follicle stimulating hormone (FSH) and luteinizing hormone (LH). These hormones control the growth of oocytes in the ovaries and the secretion of the female hormones oestradiol and progesterone by the ovaries. Two further hormones, inhibin B and anti-Mullerian hormone (AMH) are also associated with female reproductive biology.

Bovine inhibin B is two subunit protein complex of a molecular weight of approximately 56kD. The subunits are around 44kD and 14kD. Bovine inhibin B has been isolated from a gonadal source and characterised. Antibodies to bovine inhibin B have been raised and used^" to detect the protein in sera; see US5, 102, 807. The genes encoding human inhibin B subunits have been cloned and are described in EP0222491. The human hormone is also is a two subunit hormone referred to as ά and β subunits. The processed ά subunit is 134 amino acids and the β subunit is 350 amino acids in length. Inhibin B has also been isolated from porcine sources. It is also a two subunit hormone of 32kD (14 and 18kD subunits). The function of inhibin B is to suppress the secretion of FSH, and not LH, at the level of the pituitary gland.

AMH is also referred to as Mullerian Inhibiting Substance and is expressed by prepubertal Sertoli cells and post natal granulosa cells and causes regression of the Mullerian duct. A cDNA encoding AMH has been cloned from fetal bovine testicular tissue; see Picard et al (1986) PNAS, 83, 5464-5468) and is a homodimer of two subunits the complex of approximately 62kD. The human hormone is a similar size. AHM is known to be a measure of ovarian reserve and has been suggested as a means to control female fertility; see WOO 1/08695. Monoclonal antibodies to AMH have also been produced and characterised; see US4, 792, 601.

FSH is produced and secreted by the anterior pituitary gland. In the ovary FSH stimulates the growth of immature Graafian follicles to maturation and as the follicle matures it begins to release inhibin B which inhibits the secretion of FSH. FSH is a glycoprotein comprising two subunits; the ά unit is 92 amino acids in length and the β subunit is 118 amino acids in length. Human FSH has been cloned and characterised, for example see US5, 156, 957.

This disclosure relates to a test that combines the detection of each of inhibin B, AMH and FSH to provide a predictive test that allows a woman to forecast her ovarian reserve for the following two years. Women can therefore make an informed decision as to whether, or how long, they can potentially delay before trying to conceive. The test also allows the prediction of when a woman is to begin her menopause several years before this occurs giving them the opportunity to plan their lives accordingly.

According to an aspect of the invention there is provided an assay to determine the fertility of a female animal subject comprising the detection of each of inhibin B, anti- Mullerian hormone and follicle stimulating hormone in an isolated biological sample to determine the concentration of the respective hormones to provide a data output and the analysis of the data out put to provide a measure of the fertility of said female subject. According to an aspect of the invention there is provided an assay to determine the fertility of a female animal subject comprising the steps of: i) providing an isolated biological sample from the subject; ii) detecting the presence in said biological sample of at least three polypeptides wherein said polypeptides are encoded by a nucleic acid molecule selected from the group consisting of: a) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1; b) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with inhibin B; c) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 2; d) a nucleic acid molecule that hybridises the a nucleic acid molecule in (c) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with anti- mullerian hormone; e) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 3; f) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with follicle stimulating hormone; iii) collating the detection of each of said polypeptides into a data analysable form; and optionally iv) providing an output for the analysed data.

In a preferred method of the invention the output from the assay is analysed by application of the algorithm: Ln(100/[FSH])*ln([InB])*ln([AMH*10]) * 10/Age

to provide a measure of the ova retained by the ovaries of said female subject.

By applying this algorithm the assay provides an ovarian reserve index which is a measure of the fertility of the female subject.

In a preferred method of the invention said polypeptide is encoded by a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1, Figure 2 and Figure 3. Preferably said polypeptide is encoded by a nucleic acid molecule that consists of the nucleic acid sequence as represented in Figure 1, Figure 2 and Figure 3.

In a preferred method of the invention said polypeptides are detected by the binding of a ligand that specifically binds said polypeptide wherein the binding of said ligand is detected and is a direct or indirect measure of the presence of said polypeptide in said biological sample.

In a preferred method of the invention said ligand is an antibody, or active binding part thereof, that specifically binds said polypeptide.

In a preferred method of the invention said animal subject is a mammal; preferably said mammal is human.

In a further preferred method of the invention said biological sample comprises serum. Preferably said biological sample is a blood sample.

According to a further aspect of the invention there is provided a method to determine whether a female subject is likely to suffer premature menopause comprising the steps of: i) providing an isolated biological sample from the subject; ii) detecting the presence in said biological sample of at least three polypeptides wherein said polypeptides are encoded by a nucleic acid molecule selected from the group consisting of: a) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1; b) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with inhibin

B; c) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 2; d) a nucleic acid molecule that hybridises the a nucleic acid molecule in (c) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with anti- mullerian hormone; e) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 3; f) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with follicle stimulating hormone; iii) collating the detection of each of said polypeptides into a data analysable form; and optionally iv) providing an output for the analysed data.

In a preferred method of the invention the output from the assay is analysed by application of the algorithm:

Ln(100/[FSH])*ln([InB])*ln([AMH*10]) * 10/Age

to provide a measure of the ova retained by the ovaries of said female subject.

According to a further aspect of the invention there is provided a method to determine whether a female subject would benefit from in vitro fertility treatment comprising the steps of: i) providing an isolated biological sample from the subject; ii) detecting the presence in said biological sample of at least three polypeptides wherein said polypeptides are encoded by a nucleic acid molecule selected from the group consisting of: a) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1; b) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with inhibin B; c) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 2; d) a nucleic acid molecule that hybridises the a nucleic acid molecule in (c) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with anti-mullerian hormone; e) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 3; f) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with follicle stimulating hormone; iii) collating the detection of each of said polypeptides into a data analysable form; and optionally iv) providing an output for the analysed data.

In a preferred embodiment of the invention said output is compared to output from a female subject(s) that have a normal ovarian reserve index; preferably said output is compared to a female subject that successfully completed in vitro fertilisation.

Ln(100/[FSH])*ln([InB])*ln([AMH*10]) * 10/Age to provide a measure of the ova retained by the ovaries of said female subject.

According to a further aspect of the invention there is provided a kit comprising: an antibody that binds follicle stimulating hormone; an antibody that binds anti-Mullerian hormone and an antibody that binds inhibin B.

In a preferred embodiment of the invention said kit further comprises a storage container for the collection and storage of an isolated biological sample; and optionally a skin puncture device.

In a further preferred embodiment of the invention the kit is contained within a container. Preferably said container is sealed to protect the kit components. Preferably said kit is for the isolation of blood samples.

Throughout the description and claims of this specification, the words "comprise" and "contain" and variations of the words, for example "comprising" and "comprises", means "including but not limited to", and is not intended to (and does not) exclude other moieties, additives, components, integers or steps.

Throughout the description and claims of this specification, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.

Features, integers, characteristics, compounds, chemical moieties or groups described in conjunction with a particular aspect, embodiment or example of the invention are to be understood to be applicable to any other aspect, embodiment or example described herein unless incompatible therewith.

An embodiment of the invention will now be described by example only and with reference to the following Figures; Figure Ia is the nucleic acid sequence of human inhibin B; Figure Ib is the amino acid sequence of inhibin B

Figure 2a is the nucleic acid sequence of human AMH; Figure 2b is the amino acid sequence of human AMH; and

Figure 3a is the nucleic acid sequence of human FSH; Figure 3b is the amino acid sequence of human FSH; and

Figure 4a is a ROC curve illustrating measurement of FSH concentration and IFV cycle cancellation; Figure 4b is a ROC curve illustrating measurement of estradiol concentration and cycle cancellation; Figure 4c is a ROC curve illustrating measurement of inhibin B concentration on day 2 (post GnRH agonist administration) and cycle cancellation; Figure 4d is a ROC curve illustrating measurement of inhibin B concentration on day 3 (post GnRH agonist administration) and cycle cancellation; Figure 4e is a ROC curve illustrating measurement of AFC concentration and poor response to IVF treatment; Figure 4f is a ROC curve illustrating mean ovarian volume and cycle cancellation; and Figure 4g is a ROC curve illustrating measurement AMH concentration and cycle cancellation.

Materials and Methods

Subjects

A total of 84 women (age 26-44) undergoing IVF/ICSI treatment at the Assisted

Conception Unit, The Jessop Wing, Sheffield, UK were studied prospectively between March 2004 and June 2005.

In order to be included the women had to meet one or more of the following criteria:

(i) a previously raised basal FSH >10 IU/L

(ii) >39 years of age (iii) a previous poor response to stimulation during an IVF cycle.

A previous poor response was defined as the collection of <4 oocytes or cycle cancellation. Cycle cancellation was defined as the failure to reach oocyte collection. Informed consent was obtained from all women. The author's internal Ethics committee approved the study.

23 of the women had measurements taken in a subsequent cycle to compare the intercycle variability of the different measures.

Study Protocol

On day 2 of the menstrual cycle all patients had blood collected for measurement of serum levels of E2, FSH, LH, inhibin B and AMH. At the same visit an ultrasound was performed using a 6mHz Toshiba (Toshiba, Sterling, UK) transvaginal probe. An AFC and ovarian volume measurements were done. An AFC was defined as the total number of antral follicles 2- 10mm in both ovaries. Ovarian volume was calculated using the formula for the volume of an ellipsoid (π/6 x length x width x height). The mean volume was then determined. Ovaries with cysts >15mm present were excluded from the analysis of ovarian volume. Women with only 1 ovary present were also excluded. Buserelin acetate (Aventis Pharma Ltd, West Mailing, Kent, UK) 0.5mg was then administered subcutaneously.

24 hours later (cycle day 3) a second blood sample was collected to measure E2 and inhibin B. The rise in estradiol levels was calculated by dividing day 3 E2 levels by day 2

E2 levels to obtain a ratio ; the GAST test (Winslow et al. 1991). The difference in E2 levels between day 3 and 2 was calculated by subtracting the day 2 from the day 3 level (Padilla et al. 1990).

Following the second venepuncture buserelin (0.5mg) was continued as a daily injection and daily injections of rFSH 250IU (Puregon®; Organon, Cambridge, UK) were commenced (Short GnRH agonist protocol). When 3 or more follicles >17mm were observed hCG (Pregnyl; Organon, Cambridge, UK) 10 000 IU was administered and transvaginal oocyte recovery performed -36 hours later. Embryo transfer was performed 2 or 3 days following oocyte recovery. Luteal support was provided with progesterone pessaries (Cyclogest, Alphapharma, Barnstaple, UK) 400mg/day.

Patients were advised not to proceed with stimulation in a particular treatment cycle if their baseline FSH level was >15IU/1. In that circumstance they were advised to return with their next menstrual period in order to have a repeat measurement performed and if <15IU/1 they then commenced stimulation.

Cycle cancellation was recommended if 3 or less mature (>14mm) follicles developed in response to stimulation and the lead follicle was already 18mm or more. However some couples chose to proceed with oocyte collection against medical advice. If a couple were advised to cancel the cycle an intra-uterine insemination (IUI) was offered if appropriate (patent tubes and adequate sperm concentration and motility).

A clinical pregnancy was defined as the presence of foetal cardiac activity on ultrasound between 6-8 weeks gestation. An ongoing pregnancy was defined as the presence of foetal cardiac activity beyond 12 weeks gestation. Hormonal Assays FSH, LH and E2 levels were determined using an automated multi-analysis system with chemiluminescence detection (Adiva Centaur; Bayer, Newbury, UK). For FSH functional sensitivity was 0.3 IU/Land intra- and inter-assay variability were <3%. For estradiol, functional sensitivity was 26 pmol/L and intra- and inter-assay variability were <11 and 7% respectively.

Inhibin B samples were assayed in duplicate using a commercial ELISA kit (Oxford Bio- Innovation Ltd, Oxford, UK) according to the manufacturer's protocol. The sensitivity of the assay is 15 pg/mL. The intra- and inter-assay variability were <7%. Anti-Mullerian Hormone samples were assayed in duplicate using a commercial ELISA kit (Lnmunotech, Beckman Coulter UK Ltd, High Wycombe, Buckinghamshire, UK) according to the manufacturer's protocol. The sensitivity of the assay is 0.24ng/ml. The intra- and inter-assay variability were <5 and 8% respectively. Statistical Analysis Data were analysed with the SPSS (SPSS Inc, Chicago, USA) and the Medcalc (Medcalc Software, Mariakerke, Belgium) programs. The area under the receiver operating characteristic curve (ROC AUC) was computed to assess the predictive accuracy of the various tests, yielding values from 0.5 (no predictive power) to 1.0 (perfect prediction). Using the results of the ROC an appropriate cut-off level for each test was defined and the sensitivity and specificity of that cut-off determined. The chi-square for that cut-off against several outcomes (cycle cancellation, poor response and pregnancy) was then determined. The Likelihood Ratio (LR) was calculated for that cut-off. Logistic regression The paired t test was used to compare the inter-cycle variability of the parameters Examples

13 of the 84 patients (15%) had their cycle cancelled because of a poor response. 60 patients had an embryo transfer (71%). 48 patients (57%) had a poor response defined as the collection of 4 oocytes or less or cycle cancellation.

Table 1: Baseline, treatment and outcome characteristics

Variable Overall Cycle TVOPU P

Group Cancelled Performed n=84 n=13 n=71

Baseline

Age (years ± SD) 37.3 + 3.9 37.5 ± 5.0 37.3 ± 3.8 0.9

Primary infertility (%) 57 (68) 9 (69) 48 (68) 0.9

Cause of infertility (%)

Unexplained 30 (36) 8 (61) 22 (31)

Male factor 32 (38) 4 (31) 28 (39)

Tubal 13 (15) 0 (0) 13 (100)

Endometriosis 9 (11) 1 (8) 8 (11) 0.1

Duration of infertility (years 4.6 ±3.7 4.3 ±5.1 4.6 ± 3.4 0.7

± SD)

Highest FSH recorded (IU/1) 10.0 ± 2.9 12.3 ± 2.5 9.6 ± 2.7 0.00 1

Treatment

Duration of stimulation 9.3+2.0 8.9±2.5 9.4+ 1.9 0.5

(days±SD)

Total amount rFSH 2321+496 2231+633 2334+471 0.5

(IUtfcSD)

PeakE2 level (pmol/l±SD) 5497+3148 NA 5497+3148 Outcome

No. oocytes (±SD) 4.7 ± 2.0 NA 4.7 ± 2.0

Fertilisation rate (%) 67± 25 NA 67 ± 25

No of embryos transferred 1.9 ± 0.4 NA 1.9 ± 0.4

(±SD)

Values are presented as mean ± SD or %

P values are for comparison between cancelled and not cancelled patients.

Baseline, treatment and cycle characteristics are presented in Table 1. The patients who were cancelled did not differ significantly in their age nor in the duration, type or cause of infertility compared to those that were not cancelled. The mean age was 37 years (Range = 26-44; SD ± 3.9).

They did differ in their previously highest recorded FSH; women who were cancelled had a higher mean FSH level previously recorded (P=0.001). Of note mean FSH levels were high in all groups with the overall mean being 10 IU/1. Many clinics would use a level of lOIU/1 to deicide on whether to proceed with treatment. This indicates that we were reasonably inclusive when making decisions on whether to proceed with treatment. It also emphasises that this is a unique group of patients and not a good prognosis group.

All patients received a similar duration and amount of rFSH (Table 1). All patients had a fixed dose of 250IU rFSH (Puregon) each day. There were no differences in the duration of stimulation or amount of rFSH used between those who were and were not cancelled (Table 1). Patients who were cancelled were given an adequate trial of stimulation and cancellation was not merely due to opting out early.

Cycle outcomes are presented for the patients who did make it to oocyte recovery in Table 1. The mean number of oocytes collected was low at 4.7 again emphasising the nature of this group. The fertilisation in this study (67%) was equivalent to that of the clinic average for this time period.

Table 2: Study variables

Variable Overall Cycle TVOPU P

Group Cancelled Performed n=84 n=13 n=71

Baseline FSH (IU/1) 8.3± 2.3 9.2 ±2.2 8.1 ±2.3 0.1

(NS)

Baseline E2 (pmol/1) 167 ±62 145 ± 54 171 ±62 0.2

(NS)

E2 level day 3 (pmol/1) 405 ±179 303 ± 96 424 ±185 0.03

Inhibin B baseline 58 ±55 28±43 64 ±55 0.03

(Pg/ml)

Inhibin day 3 (pg/ml) 116±88 50 ±58 129±87 0.002

AMH 1.6 ±0.92 0.93 ± 0.6 1.72 ±0.92 0.004

Antral Follicle Count 7.4 ±3.5 5.2 ±2.5 7.8 ±3.5 0.02

Mean ovarian volume 8.0 ±3.9 5.1 ±2.5 8.4 ±3.9 0.01

(ml)

Values are presented as mean ± SD

P values are for comparison between cancelled and not cancelled patients.

The ovarian reserve markers studied are presented in table 2. The mean baselines between the two groups varied significantly for most of the factors studied. Only baseline FSH and E2 levels did not differ between the two groups

Table 3: Performance of the different variables in predicting IVF cycle cancellation

Variable ROC Cut- Sensitiv Specific LR (CI P

AUC off ity ity 95%)

Baseline FSH (IUZL) OM ≥ΪO 69 73 (λ6^~

GAST E2 level day 3 0.70 <355 85 59 2.1(1.2 to 0.04

(pmol/1) 2.8)

Ratio E2 rise day 3 0.60 <2.0 54 64 1.5 0.6 to 3

Δ E2 day 2 to 3 0.69 <195 77 60 1.9(1.1 to 0.04

(pmol/1) 2.8)

Inhibin B baseline 0.67 <30 77 69 2.5(1.5-3.8) 0.00

(pg/ml) 2 day 3 (pg/ml) 0.78 <60 77 79 3.7(1.9-6.3) 0.00

1 sum (pg/ml) 0.77 <65 69 79 3.3(1.5-5.9) 0.00

2

AMH (ng/ml) 0.78 <1.25 85 65 2.3 (1.5 to 0.00

3.3) 1

Antral follicle count 0.74 <5 75 76 2.5(1.1-4.9) 0.1

Mean ovarian volume 0.78 <5.3 78 75 3.1(1.6-5.3) 0.00

(ml) 1 Table 3 depicts ROC AUCs, sensitivity, specificity, LRs and statistical significance for the ovarian reserve markers studied in relation to cycle cancellation. The variables that were predictive of cycle cancellation included the day 3 estradiol level, inhibin B (both day 2 and 3 levels), AMH and mean ovarian volume. The exceptions were the baseline FSH, ratio of E2 rise (GAST) and, unexpectedly, the AFC.

Table 4: Performance of the different variables in predicting a poor response using the same cut-off as determined by the ROC AUC for cycle cancellation Variable CuT Sensitiv Specific LR (CI P off ity ity 95%)

Baseline FSH (IUZL) ≥ΪO 31 83 Ϊ3 OJ

GAST E2 level day 3 <355 57 66 1.6 (0.9 to 0.2

(pmol/1) 2.8)

Ratio E2 rise day 2 <2.0 55 83 2.9(1.4 to 0.01 to 3 6.3)

Δ E2 day 2 to 3 <195 57 66 1.6 (0.9 to 0.2

(pmol/1) 2.8)

Inhibin B baseline <30 50 75 2.3 (1.2 to 0.009

(pg/ml) 4.5) day 3 (pg/ml) <60 44 90 4.2 (1.5 to 0.004

12.7) sum (pg/ml) <65 44 90 3.9 (1.4 to 0.01

11.7) AMH (ng/ml) <1.25 63 75 2.3 (1.3 to 0.002

4.4) Antral follicle count <5 45 86 3.1 (1.2 to 0.04

8.3) Mean ovarian volume <5.3 44 79 2.1 (1.0 to 0.06 (ml) 4.3)

Table 4 depicts the various ovarian reserve markers studied as predictors of a poor response to simulation. The cut-off used to assess each factor was the same as that determined by the ROC curve for cycle cancellation. Inhibin B (day 2 and 3 levels), AMH and mean ovarian volume were all good predictors of both cycle cancellation and a poor response to stimulation. Although the E2 rise (GAST) and the AFC were not predictive of cycle cancellation they were predictive of a poor response. Just as baseline FSH levels were not predictive of cycle cancellation, neither were they predictive of a poor response.

Day 3 estradiol levels were not on their own predictive of a poor response but the ratio of the rise from day 2 was helpful.

The various markers of ovarian reserve will now be discussed separately.

FSH

20 patients were enrolled in the study because of a previously raised FSH level. The highest FSH measure previously recorded was documented for each patient. The mean previously highest FSH level for all patients was 10 IU/L (SD ± 2.9) [Table I]. The previously highest recorded FSH for any patient was 17.4 IU/L.

As FSH was used as one of the criteria for enrolment in the study the baseline levels were elevated as would be expected. The mean baseline FSH level for the cycle studied was 8.3 (SD ± 2.3) [Table 2]. The highest baseline FSH in this study was 14.2 IU/1. The baseline FSH was not a good predictor of cycle cancellation with the area under the ROC curve measuring 0.64 (CI 95% = 0.46 to 0.81) [Table 3]. The ROC chose a cut-off of 10 IU/1 which is also a commonly defined cut-off for a raised FSH in clinical practice. At this cutoff the sensitivity and specificity of a baseline FSH level were 69 and 73% respectively with a LR = 2.6 (CI 95% ; P=0.6).

The previously highest recorded FSH was a better predictor of cycle cancellation with the area under the ROC curve being 0.78 (CI 95% 0.67-0.87). At a cut-off level of 10.7 the sensitivity and specificity of the highest FSH level at predicting cycle cancellation were 84.6 and 69 % respectively. The LR of cycle cancellation with an FSH level previously recorded as above 10.7 was 2.7 (95% CI = 1.7 to 4.1; P=0.001). However as a raised FSH was one of the enrolment criteria these results should be treated with caution. They are merely confirming that a raised FSH was a reasonable criterion for study entry; a self- fulfilling prophesy.

When the same FSH cut-off (10.7IU/L) was used to predict a poor response it performed less well with an area under the ROC curve of 0.59 (95% CI = 0.48 to 0.70). LR = 1.7 (95% CI =; P= 0.06) [Table 4]. Neither the baseline FSH level nor the previously highest recorded FSH were able to predict the likelihood of an ongoing pregnancy (P>0.1 for both). [Table 6]

Baseline E2 andE2 Response (GAST)

The mean baseline E2 level was 167 pmol/1 (SD ± 62). [Table 2] Baseline E2 levels were not predictive of cycle cancellation, poor response or pregnancy.

The mean E2 level on day 3 (24 hours post GnRH agonist) was 405 pmol/1 (SD ± 179). [Table 2] Day 3 E2 levels were predictive of cycle cancellation with an area under the ROC curve of 0.70 (CI 95% = 0.60 to 0.80). At a cut-off of 355 pmol/1 the sensitivity & specificity were 85% and 59% respectively (LR = 2.1; 95% CI = 1.2 to 2.8; P=0.04). [Table 3] However a day 3 E2 levels was not predictive of a poor response when a cut-off of 355pmol/L was used (LR = 1.6; CI 95% = 0.9 to 2.8; P= 0.2). [Table 4] Nor could it predict ongoing pregnancy (P=0.4). [Table 6]

The ratio of the baseline estradiol level (day 3) to that following GnRH agonist on day 3(GAST) was a poor predictor of cycle cancellation with an area under the ROC curve of 0.57 (CI 95% = 0.40 to 0.74). [Table 3] It was marginally better at predicting a poor response (ROC curve area = 0.68; CI 95%= 0.57 to 0.80) with a sensitivity and specificity of 55% and 83% respectively when a cut-off of <2 was used. (LR =2.9; CI 95% = 1.4 to 6.3; P=0.01). [Table 4]

The change in E2 levels between day 2 and 3 was predictive of cycle cancellation with an area under the ROC curve of 0.69 (CI 95% = 0.56 to 0.82). [Table 3] If the E2 level rose <195 pmol/L then the LR of cycle cancellation was 1.9 (CI 95% = 1.1 to 2.8; P = 0.04) with a sensitivity and specificity of 77 and 60% respectively. But this cut-off was not helpful in predicting a poor response (LR = 1.6: CI 95% = 0.9 to 2.8; P=0.2). [Table 4]

The GAST (both ratio of E2 change and absolute change) was poorly predictive of clinical or ongoing pregnancy. [Table 6] If the E2 level rose <2 times between baseline and post GnRH agonist stimulation ongoing pregnancy was no more likely than if the E2 rise was greater than 2 times (P=0.3). Similarly if the E2 level rose <195pmol/L between day 2 and 3 pregnancy was not less likely to occur than if the rise was greater than this level (P=O.1).

Inhibin B

The mean baseline inhibin B level was 58 pg/mL (SD ± 55). The mean inhibin B level on day 3 was 116 pg/mL (SD ± 88). The mean sum of inhibin B was 173 pg/mL (SD ± 135). [Table 2]

Baseline inhibin B levels were predictive of cycle cancellation with the area under the ROC curve being 0.67 (CI 95% = 0.57 to 0.77). At a cut-off of 30 pg/mL the sensitivity and specificity were 77 and 69% respectively with a LR of 2.5 (CI 95% = 1.5 to 3.8; P=0.002). [Table 3] A baseline inhibin B level <30pg/mL was also predictive of a poor response (LR = 2.3; CI 95% = 1.2 to 4.5; P=0.009). [Table 4]

Inhibin B levels on day 3 (post GnRH agonist administration) were also predictive of cycle cancellation with an area under the ROC curve of 0.78 (CI 95% = 0.67 to 0.87). At a cut-off of 60pg/mL the sensitivity and specificity were 77 and 79% respectively (LR = 3.7: CI 95% =1.9 to 6.3; P =0.001). [Table 3] This level was also useful in predicting a poor response to treatment (LR = 4.2: CI 95% = 1.5 to 12.7; P=0.004). [Table 4]

The sum of inhibin B (day 2 plus day 3) was also helpful in predicting cycle cancellation with an area under the ROC curve of 0.77 (CI 95% 0.66 to 0.86). At a cut-off level of 65 pg/mL the sensitivity and specificity were 69 and 79% respectively with a LR of 3.3 (CI 95% = 1.5 to 5.9; P= 0.002). [Table 3] The inhibin B sum was also helpful in predicting a poor response to stimulation (LR = 3.9: CI 95% = 1.4 to 11.7; P=0.01). [Table 4]

Inhibin B levels were not able to predict pregnancy. Baseline, day 3 and the inhibin sum were all unable to discriminate between those who did and did not become pregnant (P >0.5 for all). [Table 6]

AMH

The mean AMH was 1.6 ng/ml (SD ± 0.92). [Table 2] AMH levels were predictive of cycle cancellation with the area under the ROC being 0.78 (CI 95% = 0.68 to 0.86). Using a cut-off level of 1.25 ng/ml the sensitivity and specificity were 85 and 65 % respectively with a LR = 2.3 (CI 95% = 1.5 to 3.3; P=0.001). [Table 3]

Antral Follicle Count

The mean AFC in this group of patients was 7.4 (SD ± 3.5). [Table 2] The AFC was predictive of a poor response but not of cycle cancellation. The area under the ROC curve was 0.74 (CI 95% = 0.63 to 0.83) for predicting cycle cancellation. At a cut-off level of 5 the sensitivity and specificity were 75 and 76% respectively with a LR = 2.5 (CI 95% = 1.1 to 4.9; P= 0.1). [Table 3]

An AFC <5 was better at predicting poor response with a LR of 3.1 (CI 95%= 1.2 to 8.3; P=0.04). [Table 4]

An AFC <5 was not able to predict those women who would not fall pregnant from treatment (P=0.3). [Table 6]

Ovarian Volume

The mean ovarian volume was 8.OmL (SD ± 3.9). [Table 2] Ovarian volume was a useful predictor of cycle cancellation with an area under the ROC of 0.78 (CI 95%= 0.66 to 0.87). At a cut-off of 5.3mL the sensitivity and specificity were 78 and 75% respectively with a LR = 3.1 (CI 95% = 1.6 to 5.3; P = 0.001). [Table 3] A mean ovarian volume <5.3 was less predictive of a poor response (LR = 2.1: CI 95% = 1.0 to 4.3; P=0.06) [Table 4].

Mean ovarian volume was unable to discriminate between those who would and would not become pregnant (P=O.1). [Table 6]

Recruitment Factors

20 women were enrolled who had a raised baseline FSH measurement. 20 women were enrolled who were aged 40 years or more at the time of treatment. 24 women had had a previous poor response to stimulation. 20 women had more than one reason for expecting they would have a poor response. [Table 5] Counter-intuitively although cycle cancellation rates were lowest in those women with a previous poor response to stimulation so were pregnancy rates although this did not reach significance. A previously raised FSH or the presence of 2 or more factors significantly increased the risk of cycle cancellation (P = 0.006). But this did not significantly affect pregnancy rates.

Table 5: Recruitment factors as predictors of cycle cancellation and pregnancy

Reason for enrolment No. of Cancelled Ongoing Pregnancy patients (%)^a (%)^b

Age >39 years 20 1 (5) 4 (20)

Previous poor response 24 0 (0) 1 (0)

Previous raised FSH 20 6 (43) 2 (10)

(>10IU/l)

More than one factor 20 6 (43) 2 (10) present

^a P= 0.006 for various enrolment factors as predictors of cycle cancellation ^b P= 0.4 for various enrolment factors as predictors of ongoing pregnancy Pregnancy

Table 6: Performance of the different variables in predicting ongoing pregnancy

Variable CutSensitiv Specific LR (CI P off ity ity 95%)

Baseline FSH (IUZL) >10 44 76 1.4 ( 0.9 to 0.2

2.9) (NS)

GAST E2 level day 3 <355 67 55 1.7 (0.8 to 0.4

(pmol/1) 4.7) (NS)

Ratio E2 rise day 3 <2.0 67 64 0.6 (0.3 to 0.3 to 3 1.3) (NS)

Δ E2 day 2 to 3 <195 78 58 1.8 LR 0.1

(pmol/1) (positive test) (NS)

= 0.545988

(0.375974 to

0.983959)

Inhibin B baseline <30 33 61 1.2 (0.5 to 0.8

(Pg/ml) 3.3) (NS) day 3 (pg/ml) <60 38 70 1.2 (0.4 to 1.0

4.4) (NS) sum (pg/ml) <65 38 70 1.9 (0.5 to 0.5

11.3) (NS)

AMH (ng/ml) <1.25 56 57 1.3 0.5

(NS)

Antral follicle count <5 44 70 1.5 0.3

(NS)

Mean ovarian volume <5.3 67 71 1.6 (0.9 to 0.1

(ml) 3.8) (NS)

'No factor reached significance

Using the same cut-offs for each variable that were determined to predict cycle cancellation LRs were calculated for the prediction of pregnancy. None of the markers of ovarian reserve were able to predict pregnancy. This emphasises that these markers are almost certainly measuring oocyte quantity not quality. [Table 6]

Table 7: Cycle and pregnancy outcomes

Group No. of Clinical Ongoing patients Pregnancy (%) Pregnancy (%) (%)

Cancellation with no 2 (2) 0 0 treatment

Converted to RJI 11 (13) 1 (9) 1 (9)

Oocyte retrieval but no ET 3 (4) 0 0

Oocyte retrieval and ET 68 (81) 11 (16) 8 (12)

Overall per cycle started 84 (100) 12 (14) 9 (11)

13 of the 84 patients (15%) had their cycle cancelled because of a poor response. Of those that cancelled 11 patients were suitable and opted for IUI treatment. 1 of these patients achieved an ongoing pregnancy (9.1%). [Table 7]

68 patients had an embryo transfer (81%). The clinical and ongoing pregnancy rates per cycle started were 14.3% (12/84) and 10.7% (9/84) respectively. The clinical and ongoing pregnancy rates per embryo transfer were 16% (11/68) and 12% (8/68) respectively. The miscarriage rate was 25% (3/12). [Table 7]

Table 8: Ongoing Pregnancy Rate in Different Patient Groups

Group Number Total (%) pregnant

Age >39 years 4 16 20

Previous poor response 1 23 4

Previous raised FSH 2 20 10

Raised FSH and >39 years 1 13 7

Raised FSH and <39 years 5 25 20

P = 0.4 Pregnancy rates were not statistically different between the different recruitment groups. Patients who had had a previous poor response had the lowest pregnancy rates overall 1/23 (4%). Women aged 40 years or more and were known to have a raised FSH also had a low pregnancy rate (1/13). Younger women with a raised FSH performed respectably in comparison to other groups in this study as other authors have previously found. [Table 8]

As pregnancy rates were low in all of these patients it is important not to over emphasise this data. The primary outcome of this study was always going to be cycle cancellation rather than pregnancy.

Table 9: Pregnancy rates in relation to response to stimulation

Response Cycle outcome Number Ongoing Pregnancy

(%)

Poor Response Cancelled no treatment 2 0

IUI 11 1 (9)

Oocyte retrieval <4 35 5 (14) eggs

ALL poor responders 48 6 (12)

Adequate 36 3 (8) response

P=OJ

Any patient who had 3 or less mature follicles (>14mm) with a lead follicle at 18mm was advised that proceeding with oocyte recovery would entail a risk of there being no embryo for transfer. However ultimately the final decision was left up to the couple as to whether they wished to take that risk. As already mentioned 11 patients opted for intra-uterine insemination. However some patients were not suitable for IUI treatment or preferred to proceed with IVF. 16 patients opted to proceed with oocyte retrieval in this situation and did surprisingly well. 13/16 (81%) had an embryo for transfer and 6/16 (37.5%) fell pregnant. However 3/16 (19%) did not have an embryo for transfer after proceeding in this situation. In this group of patients a poor response to stimulation did not reduce the likelihood of pregnancy. 8% (3/36) adequate responders achieved a pregnancy versus 12% (6/48) poor responders (P=0.7). 14% patients who had 4 or less oocytes retrieved achieved an ongoing pregnancy despite nearly 20% of these patients not making it to embryo transfer. However these were not an unselected group of patients and the overall pregnancy rates were low.

Inter-cycle Variation

23 patients had measures taken in a subsequent cycle to compare inter-cycle variation of the different parameters. All of the second cycles took place within one year of the original cycle. The only parameter that showed significant variation between cycles was the baseline inhibin B level. [Table 10]

Table 10: Inter-cycle variability of predictors

Variable Cycle 1 (mean ± Cycle 2 (mean ± P

SD) SD)

Baseline E2 (pmol/1) 182 ± 87 145± 61 0.06

E2 level day 3 (pmol/1) 430 ± 61 370 ± 203 0.1

Baseline FSH (IU/1) 8.9 ± 3.5 9.6 ± 4.0 0.2

Inhibin B baseline 67 ± 56 42 ± 45 0.04

(pg/ml)

Inhibin B day 3 (pg/ml) 173 ± 139 112 ± 83 0.08

AMH 1.36 ± 0.78 1.38 ± 0.91 0.8

Antral follicle count 7.5 ± 3.4 6.6 ± 2.5 0.4

Mean ovarian volume 4.9 ± 2.3 4.6 ± 2.4 0.6

(ml)

Claims

1. An assay to determine the fertility of a female animal subject comprising the detection of each of inhibin B, anti-Mullerian hormone and follicle stimulating hormone in an isolated biological sample to determine the concentration of the respective hormones to provide a data output and the analysis of the data out put to provide a measure of the fertility of said female subject.

2. An assay to determine the fertility of a female animal subject according to claim 1 comprising the steps of: i) providing an isolated biological sample from the subject; ii) detecting the presence in said biological sample of at least three polypeptides wherein said polypeptides are encoded by a nucleic acid molecule selected from the group consisting of: a) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1; b) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with inhibin B; c) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 2; d) a nucleic acid molecule that hybridises to a nucleic acid molecule in (c) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with anti-mullerian hormone; e) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 3; f) a nucleic acid molecule that hybridises to a nucleic acid molecule in (e) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with follicle stimulating hormone; g) collating the detection of each of said polypeptides into a data analysable form; and optionally h) providing an output for the analysed data.

3. An assay according to claim 1 or 2 wherein the output from the assay is analysed by application of the algorithm:

Ln(100/[FSH])*ln([InB])*ln([AMH*10]) * 10/Age

to provide a measure of the ova retained by the ovaries of said female subject.

4. An assay according to any of claims 1-3 wherein said polypeptide is encoded by a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1, Figure 2 and Figure 3. Preferably said polypeptide is encoded by a nucleic acid molecule that consists of the nucleic acid sequence as represented in Figure 1, Figure 2 and Figure 3.

5. An assay according to any of claims 1-4 wherein said polypeptides are detected by the binding of a ligand that specifically binds said polypeptide wherein the binding of said ligand is detected and is a direct or indirect measure of the presence of said polypeptide in said biological sample.

6. An assay according to claim 5 wherein said ligand is an antibody, or active binding part thereof, that specifically binds said polypeptide.

7. An assay according to any of claims 1-6 wherein said animal subject is a mammal.

8. An assay according to claim 7 wherein said mammal is human.

9. An assay according to any of claims 1-8 wherein said biological sample comprises serum.

10. An assay according to claim 9 wherein said biological sample is a blood sample.

11. An assay to determine whether a female subject is likely to suffer premature menopause comprising the steps of: i) providing an isolated biological sample from the subject; ii) detecting the presence in said biological sample of at least three polypeptides wherein said polypeptides are encoded by a nucleic acid molecule selected from the group consisting of: a)a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1 ; b) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with inhibin B; c) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 2; d) a nucleic acid molecule that hybridises the a nucleic acid molecule in (c) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with anti-mullerian hormone; e) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 3; f) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with follicle stimulating hormone; ii) collating the detection of each of said polypeptides into a data analysable form; and optionally iii) providing an output for the analysed data.

12. An assay according to claim 11 wherein the output from the assay is analysed by application of the algorithm:

Ln(100/[FSH])*ln([InB])*ln([AMH*10]) * 10/Age

to provide a measure of the ova retained by the ovaries of said female subject.

13. An assay to determine whether a female subject would benefit from in vitro fertility treatment comprising the steps of: i) providing an isolated biological sample from the subject; ii) detecting the presence in said biological sample of at least three polypeptides wherein said polypeptides are encoded by a nucleic acid molecule selected from the group consisting of: a) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 1; b) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with inhibin B; c) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 2; d) a nucleic acid molecule that hybridises the a nucleic acid molecule in (c) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with anti-mullerian hormone; e) a nucleic acid molecule comprising a nucleic acid sequence as represented in Figure 3; f) a nucleic acid molecule that hybridises the a nucleic acid molecule in (a) under stringent hybridisation conditions and that encodes a polypeptide that has the biological activity associated with follicle stimulating hormone; iii) collating the detection of each of said polypeptides into a data analysable form; and optionally iv) providing an output for the analysed data.

14. An assay according to claim 13 wherein said output is compared to output from a female subject(s) that have a normal ovarian reserve index.

15. An assay according to claim 13 or 14 wherein said output is compared to a female subject that successfully completed in vitro fertilisation.

16. An assay according to any of claims 13-15 wherein the output from the assay is analysed by application of the algorithm:

Ln(100/[FSH])*ln([InB])*ln([AMH*10]) * 10/Age

to provide a measure of the ova retained by the ovaries of said female subject.

17. A kit comprising: an antibody that binds follicle stimulating hormone; an antibody that binds anti-Mullerian hormone and an antibody that binds inhibin B.

18. A kit according to claim 17 wherein the kit further comprises a storage container for the collection and storage of an isolated biological sample; and optionally a skin puncture device.

19. A kit according to claim 17 and 18 wherein the kit is contained within a container.

20. A kit according to claim 19 wherein said container is sealed to protect the kit components.

21. A kit according to any of claims 17-20 further comprising instructions for use of the kit and analysis of the data generated by the kit when used.