WO1999058974A1

WO1999058974A1 - Method and test device for assessing cervical ripeness at term with highly phosphorylated isoforms of igfbp-1

Info

Publication number: WO1999058974A1
Application number: PCT/FI1999/000390
Authority: WO
Inventors: Eeva-Marja Rutanen; Tytti KÄRKKÄINEN
Original assignee: Medix Biochemica Oy AB
Current assignee: Medix Biochemica Oy AB
Priority date: 1998-05-12
Filing date: 1999-05-10
Publication date: 1999-11-18
Anticipated expiration: 2000-11-12
Also published as: FI981049A7; AU4042699A; FI981049A0

Abstract

The present invention is related to improved methods and test devices for assessing cervical ripeness, predicting susceptibility of delivery and/or determining the cause and origin of the minor amounts of IGFBP-1 present in cervical samples with immunoassays capable of differentiating phosphorylated and nonphosphorylated isoforms of insulin-like growth factor binding protein 1 (IGFBP-1). The invention is also related to the use of specific binding substances, especially antibodies capable of recognizing phosphorylated isoforms of IGFBP-1 for predicting susceptibility of delivery and for properly timing the induction of labor in preterm, term and postterm pregnancies.

Description

1

METHOD AND TEST DEVICE FOR ASSESSING CERVICAL RIPENESS AT TERM WITH HIGHLY PHOSPHORYLATED ISOFORMS OF IGFBP-1

The Technical Field of the Invention

The present invention is related to improved methods and test devices for assessing cervical ripeness at term, and for predicting susceptibility of delivery and/or determining the cause and origin of the minor amounts of IGFBP-1 present in cervical samples with immunoassays capable of differentiating phosphorylated and nonphosphorylated isoforms of insulin- like growth factor binding protein 1 (IGFBP-1) . The invention is also related to the use of specific binding substances capable of recognizing highly phosphorylated decidua-derived isoforms of IGFBP-1 different from amniotic fluid-derived phosphorylated isoforms for predicting the risk of preterm delivery and for properly timing the labor induction in preterm, term and postterm pregnancies.

The Background of the Invention

In the United States, elective induction of labor is performed in as many as 25% of pregnancies. Prediction of success of induction has been traditionally based on the Bishop scores, although the specificity of this prediction is poor. There is, therefore, a great need for a biochemical test to predict cervical ripeness and success of labor induction more accurately.

Especially, the development of immunochromatographic and related techniques has resulted in new tools for devising rapid tests that can be reliably performed at the bedside or chair-side by doctors and nurses, without need for laboratory services. Such tests are described for example in the following patents or patent applications US 5,712,170, US 4,552,839, US 4,361,537, US 4,373,932, WO 86/04683, WO 86/03839, EP 212 599, EP 291 194, EP 284 232, EP 225 054. In many fields everyday diagnostic practice has greatly benefited from the development and availability of such tests. Also in the field of obstetrics the development of immunochromatographic test devices has enabled collection of data, which in turn have made it possible to gain deeper insight in the processes occurring during pregnancy and consequently to provide better diagnostic tools.

The causative and regulating factors in cervical ripening are largely unknown, although the impact of the release of prostaglandin÷ E2 (PGE2) from the amnion and choriodecidual tissue is acknowledged. This has led to the use of exogenous PGE2 for cervical ripening. In addition, the cytokines, such as interleukin- (IL-) -1, IL-6, IL-8, tumor necrosis factor-alpha and fetal fibronectin (FFN) , are involved in cervical ripening and initiation of labor. Produced in the choriodecidual interface, FFN leaks into the cervix and vagina, becomes detectable usually 1-2 weeks before the onset of labor, and predict cervical ripening and the success of labor induction.

Methods and devices used to determine the presence of FFN in the cervical secretion have been developed (EP 563 165, EP 316 919) and used for indicating e.g. a risk for preterm birth (EP 642 667, WO 93/24838) and to rule out rupture of membranes in women at risk for imminent delivery (EP 680 607) .

Insulin- like growth factor binding protein- 1 (IGFBP-1; previously also called placental protein-12) is a 25 kD protein synthesized and secreted by the fetal and adult liver and maternal decidua. In the maternal circulation, IGFBP-1 increases during pregnancy and it is a major protein in the amniotic fluid from the second-trimester of pregnancy until term.

The detection of the high concentration levels of IGFBP-1 in cervical and vaginal samples has been shown to be a reliable diagnostic tool for detecting the rupture of the fetal membranes. Methods and test devices using IGFBP-1 have been disclosed e.g. in the patents US 5,554,504 and US 5,597,700.

Occasionally, small amounts of IGFBP-1, in contrast to the high concentration levels of IGFBP-1 found in connection with premature rupture of membranes, can be detected in the cervix in women with apparently intact membranes (Rutanen E-M et al . , Clin Chim Acta 1996; 253: 91-101). These small amounts of IGFBP-1 have been shown to correlate with susceptibility of delivery. Based on said findings, methods and test devices for diagnosing susceptibility of delivery were developed (WO 94/00765) . However, some uncertainties related to the presence of small amounts of IGFBP-1 exist, because the presence may derive either from a minimal leakage of amniotic fluid above the lower pool of the amniotic cavity or from the decidua, which becomes detached from the chorion during cervical ripening (lams JD et al . , Am J Obstet Gynecol 1995:173:141-5).

In spite of the uncertainties related to the origin of small amount of IGFBP-1 the methods and test devices for predicting susceptibility of delivery have proved to be useful tools for the obstetrician in making important diagnostic decisions, such as admittance to hospital for immediate measures or follow up studies or alternatively, for a safe remittance to home .

However, it would be desirable that the uncertainties related to the cause and origin of IGFBP-1 could be removed and a more specific and reliable method and test device for predicting susceptibility of delivery and to differentiate said condition from premature rupture of membranes could be provided.

Thus, the objective of the present invention is to provide a method and test device by the aid of which the problem, i.e. the uncertainty of the cause and origin of IGFBP-1 could be solved. By solving said problem an improved method and test device for predicting susceptibility of delivery could be provided and suspected PROM could be differentiated from susceptibility of delivery.

A solution to problem was the finding made by the present inventors which indicated that amniotic fluid mostly contains nonphosphorylated and/or less phosphorylated IGFBP-1, whereas cervical ripeness and susceptibility of delivery should be diagnosed with methods and test devices assessing the presence of phosphorylated isoforms of IGFBP-1.

The Summary of the Invention

The characteristic features of the present invention are as defined in the claims.

A Brief Description of the Drawings

Figure 1: IGFBP-1 concentrations in ripe (Bishop scores > 6) and unripe (Bishop scores ≤ 5) cervices and changes in cervical IGFBP-1 concentrations at 6-hour intervals after consecutive application of PGE2 gel in women with unripe cervix measured by immunoenzymometric assay 1 (monoclonal antibody 6305 as the solid phase) . The first cervical swab specimens were taken before the first cervical assessment. Horizontal lines denote medians with the end of each line denoting ripe cervix allowing amniotomy. Each square denotes treatment failure followed by a cesarean section.

Figure 2: IGFBP-1 concentrations in ripe (Bishop scores ≥ 6) and unripe (Bishop scores < 5) cervices and changes in cervical IGFBP-1 concentrations at 6-hour intervals after consecutive application of PGE2 gel in women with unripe cervix measured by immunoenzymometric assay 2 (monoclonal antibody 6303 as the solid phase) . The first cervical swab specimens were taken before the first cervical assessment. Horizontal lines denote medians with the end of each line denoting ripe cervix allowing amniotomy. Each square denotes treatment failure followed by a cesarean section.

Figure 3: Receiver operating characteristic curves: sensitivity (true-positive rate) and 100-specificity (false-positive rate) for various values of cervical IGFBP-1 concentrations (μg/L) measured by two immunoenzymometric assays, assay 1 (Figure 3a) and assay 2 (Figure 3b) in predicting cervical maturity (Bishop scores >5) . The diagonal line represents the line of unity, i.e. the expected curve if there were no relation between the predictor and outcome.

The Detailed Description of the Invention

Definitions

Most terms used in the present specification and claims have the same meaning as they generally have in the field of prenatal diagnostics, obstetrics and immunochemistry as set forth in books, review articles and laboratory handbooks. Some terms are, however, used more extensively and might have a meaning which somewhat differs from the general use in said fields. Thus, some of these terms are defined more closely below.

The term "susceptibility of delivery" is correlated to the ripeness of cervix and indicates that labor will soon start spontaneously or that it can be started artificially without causing undue and prolonged suffering and pains to the mother and child. Alternatively it means that delivery is imminent. In early pregnancies the possibility to assess cervix ripeness can be used to predict a risk of preterm delivery. In preterm, full term and post term pregnancies, the assessment of cervical ripeness can be used for improved timing of e.g. artificial induction of labor.

The term "cervical ripeness or maturity" means the natural maturing process preceeding delivery and which is assessed e.g. by palpation as Bishop scores.

The term PROM means premature rupture of membranes. The premature rupture of membrane results in escape of amniotic fluid and the condition can be diagnosed by different means as described above including immunochemical test methods, which recognize the presence of elevated levels of IGFBP-1.

The term "essentially blood free cervical sample" means that the cervical sample is visually blood free. Naturally, the essentially blood free condition can be determined with appropriate tests and determination methods, but there is no need for auxiliary tests to ensure essential freeness of blood.

The term "immunochemical test method" includes any conventional or new quantitative, semiquantitative and or qualitative immunoassays, including such with detection based on enzyme markers, radioactive markers, fluorescent markers or visible particulate markers. The immunochemical tests can be performed as homogenous or heterogenous tests in liquid or dry state or combinations thereof. The tests include latex-agglutination-tests, immunochromatographic immunometric assays, ELISA, EIA, RIA, FIA, etc. Especially, the term "immunochemical test method" means rapid, one step-, no wash-, yes/no-, bedside or chair-side immunoassays, which do not require elaborate equipments and laboratory facilities. Such tests including immunochromatographic based on the lateral flow principle and immunometric test based on the flow through principle as well as other convenient immunochemical methods and test devices are described for example in the following patents or patent applications US 5,712,170, US 4,552,839, US 4,361,537, US 4,373,932, WO 86/04683, WO 86/03839, EP 212 599, EP 291 194, EP 284 232, EP 225 054. The list is in no way limiting and the teaching of said patents and patent applications is herewith incorporated by reference into the present specification.

Those skilled in the art can easily adapt the techniques disclosed in the above references as well as in the multitude of available publications, laboratory handbooks, etc to provide a multitude of different methods and test devices depending upon the actual need. The prerequisite is that by the aid of the test method and test device it is possible to assess the presence or absence of the highly phosphorylated isoform of IGFBP-1 and to differentiate those from the nonphosphorylated or less phosphorylated IGFBP-1 present in amniotic fluid.

In fact, a higher proportion of less phophorylated IGFBP-1 is an indication that the IGFBP-1 originates from amniotic fluid (assay 1) , whereas a higher proportion of the highly phosphorylated IGFBP-1 than less phosphorylated IGFBP-1 indicates that the IGFBP-1 originates from cervix (assay 2) .

The term "IGFBP-1" means the major secretory protein of human decidua (Julkunen M et al . , FEBS Lett 1988; 236: 295-302; Rutanen E-M et al . , Endocrinology 1985; 116: 1304-9) and its concentration in the amniotic fluid is 100 to 1000 -fold higher than that in serum (Rutanen E-M et al . , Am J Obstet Gynecol 1982; 144 : 460-3) .

IGFBP-1 exists in five electrophoretically and chromatographically distinct isoforms. IGFBP-1 is phosphorylated in serine residues (Ser¹⁹¹, Ser¹¹^, Ser¹^, containing 70%, 5% and 25% of incorporated phosphate when tested with ³²P-labelled IGFBP-1) . (Frost & Tseng, 266 (1991) , 18082-18088); Jones et al . , 268 (1993), 1125-1131. The phosphorylated isoforms also have greater affinity towards IGF.

The synthesis and biochemical characterization especially of phosphorylated isoforms of IGFBP-1 has been described e.g. in the following publications, Frost RA and Tseng L, J Biol Chem 1991,266:18082-18088; Westwood M et al . , J Clin Endocrinol Metab 1994; 79:1735-1741; Martina NA et al . , J Clin Endocrinol Metab 1997; 82: 1874-1898. The phosphorylation of IGFBP-1 is reported to increase affinity for the ligand, IGF-I (Jones JI, Proc Natl Acad Sci USA 1991; 88:7481-7485) .

Based on the results obtained by the present inventors and described in the publication (Nuutila M. et al, An Isoform of Insulin-Like Growth Factor Binding Protein-1 in Cervical Secretion, Different from those in the Amniotic Fluid Predicts Cervical Ripeness, 1998; Submitted), nhe contents of which herewith is incorporated by reference, it was concluded that an isoform of IGFBP-1, different from that in the amniotic fluid, predominates in the cervix and reflects cervical ripeness .

The term "small amount of IGFBP-1" means the amounts of IGFBP-1 detectable in cervical secretion of women with intact membranes at term. In a negative initial PROM TEST, having a detection limit adjusted to be around 25 - 50 μg/L, the cervical IGFBP-1 concentration was < 50 μg/L in 63 of 64 samples. The values given are specific for the PROM TEST developed by Oy Medix Biochemica Ab for diagnosing PROM. The specific values for detecting limits should be determined individually for each test used.

The term "elevated level of IGFBP-1" means "high concentration level" of different IGFBP-1 :s found in the cervix sample containing amniotic fluid.

The term "specific binding substance" means e.g. antibodies, receptors and/or ligands, but especially antibodies capable of recognizing the highly phosphorylated isoform of IGFBP-1 and differentiating them from antibodies specifically recognizing less phosphorylated or nonphosphorylated IGFBP-1. In the present invention the most preferred binding substance is MAb 6303, which recognizes the highly phosphorylated isoform of IGFBP-1.

The Mabs 6305 and 6303, which have been shown to recognize differentially different phosphoforms of IGFBP-1 (Westwood M et al. J Clin Endocrinol Metab 1994; 79: 1735-41).

Mab 6305 detects both nonphosphorylated and some phosphoforms of IGFBP-1, but fails to bind the highly phosphorylated isoform. Thus, Mab 6305 is used in the IEMA TEST, (Medix Biochemica) and in the PROM TEST (Medix Biochemica) designed for the diagnosis of ruptured fetal membranes (Rutanen Ξ-M et al., Clin Chim Acta 1996; 253: 91-101).

The term "test device" includes any test strips, dishes, test tubes, etc., which allow an immunoassay capable of assessing the presence or absence of phosphorylated isoforms of IGFBP-1 to be preformed thereon or therein.

The term "test kit" means a package combination comprising one or more test devices for assessing phosphorylated isoforms of IGFBP-1 and optionally the nonphosphorylated isoform of IGFBP-1 either in the same or a separate test device as well as optional reagents, sampling devices as well as instructions .

The General Description of the Invention

The phosphorylation status of IGFBP-1 in different body fluids and tissues is known to vary (Westwood M, J Clin Endocrinol Metab 1994; 79: 1735-41; Martina NA et al . , J Clin Endocrinol Metab 1997; 82: 1874-1898. For example in amniotic fluid the nonphosphorylated isoform predominates, but the phosphorylated isoforms exist as well. Human decidual cells are known to secrete predominantly the phosphorylated isoform of IGFBP-1, 10

but the phosphorylation degree of IGFBP-1 is known to vary during gestation.

The present inventors performed a study designed to clarify the origin of IGFBP-1 in cervical secretion in women with apparently intact membranes and to examine whether IGFBP-1 in cervical secretion reflects cervical ripeness in order to evaluate whether the assessment of insulin- like growth factor binding protein-1 (IGFBP-1) in cervical secretion could serve in prediction of cervical ripeness.

The studies performed indicate that during early pregnancy, nonphosphorylated IGFBP-1 is the predominant isoform in decidua and maternal serum. However, in the late stages of gestation it is primarily the most highly phosphorylated form that is present, as in non-pregnant serum. As gestation progresses all of the phosphorylated isoforms are observed in decidua, and all but the most phosphorylated form can be detected in amniotic fluid. Antibody that recognizes the nonphosphorylated isoform of IGFBP-1 was selected for the Medix Biochemica PROM-test as the specificity-determining antibody, to ensure that the test favoured the isoform always found in amniotic fluid.

This implies that the isoforms of IGFBP-1 in cervical secretion are different from those in the amniotic fluid. Based on the results it was concluded that an isoform of IGFBP-1, different from those in the amniotic fluid, predominates in the cervix and could reflect cervical ripeness .

As a result of the findings a rapid test for detecting the highly phosphorylated form of IGFBP-1 was developed. Said test which allows the detection and differentiation of different phosphorylated and nonphosphorylated isoforms of IGFBP-1 allows a more accurate assessment of cervical maturity and enables the development of an improved method for predicting 11

the risk for preterm delivery and exacter timing for the induction of delivery. A method for a better differentiation between premature rupture of membranes (PROM) and susceptibility of delivery could also be developed because highly phosphorylated form does not occur in amniotic fluid.

In summary, an isoform of IGFBP-1, different from those in the amniotic fluid, is detectable in cervical secretion, where its concentration rises with cervical ripening. The assessment of the phosphorylated IGFBP-1 isoform in cervical secretion in patients with intact membranes (negative PROM TEST) provides an additional tool for evaluation of cervical ripeness and for prediction of the prognosis of labor induction in preterm and term pregnancies. Rapid bed-side or chair-side test designed for this purpose can easily be adapted for clinical use.

The applicability and feasibility of phosphorylated isoforms of IGFBP-1 for assessing cervical ripeness and differentiating between premature rupture of membranes and predicting susceptibility of delivery is described in more detail in the following illustrative, nonrestrictive examples based on which the man skilled in the art can easily develop different methods and test devices wherein the determination of phosphorylated isoforms of IGFBP-1 provides an improved and convenient diagnostic tool for obstetricians.

Materials and methods

Patients

Sixty- four parturients with a single live fetus in vertex presentation determined for induction of labor between 36 and 42 gestation weeks were recruited into the study, which was approved by the institutional review board. Gestational age was based on fetal sonographic measurements (biparietal diameter and femur length) at 17 to 18 weeks of gestation. Each of the women gave her written informed consent. All women 12

were examined by the same doctor (MN) , and the ripeness of the cervix was evaluated by means of Bishop scores (Bishop EH. Obstet Gynecol 1964; 24: 266-8), 29 patients had Bishop scores > 6, indicating a ripe cervix, and 35 had Bishop scores

< 5 as indication of an unripe cervix. All had intact membranes and a negative PROM TEST (Medix Biochemica, Kauniainen, Finland) (Rutanen E-M et al . , Clin Chim Acta 1996; 253: 91-101) with no evidence of active labor, either subjectively or by cardiotochography (CTG) recording.

Each parturient with a ripe cervix experienced amniotomy, whereas each of the women with an unripe cervix received an intravaginal application of PGE2 gel (2.0 mg Minprostin^R, Pharmacia & Upjohn, Puurs, Belgium) for cervical ripening; this was repeated every 6 hours if the Bishop scores remained

< 6; Bishop scores were redetermined at each time-point. Both groups were carefully followed during ripening or induction, and CTG was recorded continuously for 30 minutes before and 2 hours after application of PGE2 as well as continuously during the labor until delivery.

Sampling

Two dacron swab samples were taken simultaneously by speculum examination from the external os of the cervix of all 64 women before cervical assessment and amniotomy or application of PGE2 gel. In the women with unripe cervix, cervical sampling was repeated before reapplication of PGE2 gel or before amniotomy (Bishop scores > 6) at 6, 12 and 18 hours. Each sample was extracted in 0.5 mL of assay buffer as described previously (Rutanen E-M et al . , Clin Chim Acta 1996; 253: 91-101) . One specimen was used immediately for the PROM TEST (Medix Biochemica), and the result was read in 2-5 minutes. Thereafter, both tubes were frozen and stored at -20 °C until the IGFBP-1 concentration in the remaining buffer was measured. In addition, 39 amniotic fluid samples were collected at term and stored at -20 °C until analyzed for 13

comparison of IGFBP-1 in cervical secretion and amniotic fluid.

Immunochemical tests

One set of cervical swab samples and amniotic fluid samples were analyzed for the concentration of IGFBP-1 by the immunoenzymometric assay, with monoclonal antibody (Mab) 6305 as the solid phase (assay 1) (IGFBP-1 IEMA TEST, Medix Biochemica) as described previously (Rutanen E-M et al . , Clin Chim Acta 1996; 253: 91-101). Mab 6305 isolated from placental extract detects the nonphosphorylated and lesser phosphorylated isoforms of IGFBP-1 (Rutanen E-M, et al . Biochem Biophys Res Commun 1988; 152: 208-15; Westwood M et al. J Clin Endocrinol Metab 1994; 79: 1735-41). The detection limit of the assay is 0.4 μg/L, and the intra- and interassay coefficients of variation were 3.4% and 7.4%, respectively.

Another set of specimens was also extracted to PROM TEST

Specimen Extraction Solution (0.5 mL) , and the IGFBP-1 concentration was measured with a new immunoenzymometric assay which uses Mab 6303 (Medix Biochemica) as the solid phase

(assay 2) . Mab 6303 recognized the highly phosphorylated isoform of IGFBP-1, which is not detected by Mab 6505

(Westwood M et al. J Clin Endocrinol Metab 1994; 79: 1735-41) .

Microtiter wells (Nunc, Maxisorb, Roskilde, Denmark) were coated with 150 μL of antibody in 50 mM phosphate buffer (10 μg/mL) , pH 6.0. After overnight incubation, the wells were washed with PBS-Tween (0.05%), blocked with 0.5% bovine albumin (BSA) in 50 mM phosphate buffer, dried, and used for the assay. In the assay, 50 μL of assay buffer (50 mM Na2HPθ4 ,

50 mM NaCl, 50 mM NaSCN, 10 mM EDTA, 0.05%

3- [ (cholamidopropyl) dimethylammonio] -1-propane sulphonate

[CHAPS], 0.3% BSA, 0.03% Tween-20, 2μL/mL normal mouse serum and 0.1% Proclin 300, pH 8.5), 50 μL of standard (affinity purified phosphoform of IGFBP-1 purified from decidua extract,

A280¹^ ⁼ 13.6), or extracted cervical sample were pipetted 14

into each well. After 30 minutes of incubation in a horizontal shaker, the wells were washed five times with 0.05% PBS-Tween-20, 100 μL of another antibody labeled with horse radish peroxidase (HRP) (6301-HRP) diluted 1:50 in the assay buffer was added to each well and incubated for 15 minutes in a shaker. The wells were washed as earlier, and 100 μL of substrate (2 , 2 ' -azino-di- [3-ethyl-benz-thiazoline sulphonate (6)] ABTS) , Kirkegaard-Perry, Denmark) was added. After 15 minutes, the reaction was stopped by 50 μL of 4% oxalic acid. Absorbances were measured at 415 nm on a microplate reader, and results were read from the standard curve. The detection limit of the assay is 0.3 μg/L, and intra- and interassay variations were 4.6% and 6.4%, respectively. Linearity of diluted decidual cytosols and amniotic fluid samples in the assay buffer was 94-132% and recovery 100-128%, and PG does not interfere with this assay (data not shown) .

Methods for analysing the test results

The end-points for the study included comparison of the IGFBP-1 concentration given by the two assays and comparison of cervical IGFBP-1 concentrations in the unripe and ripe cervix, as well as the changes in the cervical IGFBP-1 levels after consecutive PGE2 applications. If amniotomy could not: be performed (Bishop scores < 6) after 18 hours and 3 applications of PGE2 gel, the treatment was judged as a failure .

The Mann-Whitney test was used for the comparison of the cervical IGFBP-1 levels in women with unripe and ripe cervix. The Wilcoxon signed rank test was used to analyze the differences in IGFBP-1 levels before and after PGE2 treatment. The Chi-square test was used for four-fold tablets with Yates' correction when appropriate. Continuous variables with normal distribution were compared by the two-tailed t-test. A P value < 0.05 was considered significant. The receiver-operating characteristic (ROC) curves were established by plotting the 15

sensitivity (true-positive rate) against the 100-specificity (false-positive rate) (Zweig MH and Campbell G. Clin Chem 1993; 39: 561-77) .

Results

The clinical characteristics of the women in the two study groups were otherwise comparable but the amniotomy group had fewer nulliparous women and higher Bishop scores (Table I) . All women had a negative PROM TEST (Medix Biochemica) (Rutanen E-M et al., Clin Chim Acta 1996; 253: 91-101) result at the baseline .

The amniotomy-to-vaginal delivery time in women with a ripe cervix was significantly shorter (5.6±2.9 hours; mean ±SD) than that in women with an initially unripe cervix (8.1±4.3 hours) (P=0.01) . A cesarean section (CS) was performed in two cases for failed cervical ripening (Figures 1 and 2), in three cases for failure to progress, and in two cases for fetal asphyxia. The neonatal outcome in all these women was uneventful (Table I) .

In each cervical sample the IGFBP-1 concentration measured by assay 2 was higher than that measured by assay 1 (Figures 1 and 2) . A significant positive correlation was detected between these two measurements (r²=0.75) .

At baseline, the cervical IGFBP-1 concentrations in women with ripe cervix were significantly higher (assay 1: median 9.8 μg/L [range 0.4-92.0]; assay 2: 27.0 μg/L [1.3-102.0]) than in women with unripe cervix (assay 1: 2.5 μg/1 [0.4-45.0]; assay 2: 6.6 μg/L [0.3-53.0] (P=0.003 and 0.004, respectively) (Figures 1 and 2) . These differences were independent of the parity (nulliparous vs. parous) , maternal age, gestational age, or indication of induction of labor (data not shown) .

The cervical ripening (increasing Bishop scores) following 16

application of PGE2 led to a rise in the cervical IGFBP-1 concentrations (Figures 1 and 2) . The median level of IGFBP-1

6 hours after PGE2 administration was significantly higher

(assay 1: 20.3 μg/L; assay 2: 51.0 μg/L) (P=.0001) than the basal levels, and amniotomy could be performed in 30 women

(Figures 1 and 2) . Five women received a second dose of PGE2 , and amniotomy could be performed 6 hours later in 2 women

(Figures 1 and 2) . The remaining 3 women received still a third application of PGE2 gel, and 18 hours after the start of the cervical ripening amniotomy could still be performed for one woman (Figures 1 and 2) . The IGFBP-1 concentration obtained at baseline or immediately before amniotomy failed to predict the amniotomy-to-vaginal delivery time (data not shown) .

In contrast to the cervical samples, the IGFBP-1 concentration in each sample of the amniotic fluid (n=39) was 1.9 to 5.9 times greater by assay 1 than that by assay 2 (Table II) .

An IGFBP-1 concentration 6 μg/L with assay 1 and 26 μg/L with assay 2 predicted cervical ripeness (Bishop scores > 6) with a sensitivity of 52% and 53%, respectively. The false-positive rate in both cases was 20% (Figures 3a and 3b) .

17

Table I

Patient characteristics in the two study groups (mean ± SD)

Women with unripe Women with ripe cervix (Bishop cervix (Bishop scores ≤5) ; n=35 scores ≤6) ; n=29

Maternal age 30.2 (5.3) 30.7 (4.7) NS (yr) *

Gestational age 39.4 (1.5! 39.8 (1.4^' NS (wk) *

Percent 66.7% 27.5% 005 primiparous**

Bishop scores 3. !2-5; 6.8 0001 at entry (range) *

Indications for induction Hypertensive pregnancy complication 1 6

Diabetes 3 3

Oligohydramnion 3 4

Post term 3 3

Cholestasis of pregnancy 4 2

Other 4 11

Birth weight (g) 3508 (±650) 3725 (±514) NS

Umbilical artery 7.24 (0.07) 7.25 (0.06! NS pH*

* t-test ** chi-square test Table II

IGFBP-1 concentrations in amniotic fluid measured by assay 1 (detecting nonphosphorylated and lesser phosphorylated isoforms of IGFBP-1) and by assay 2 (recognizing the highly phosphorylated isoform of IGFBP-1) .

IGFBP-1 (μg/L)

Range Median 95% Cl

Assay 1 3000-138000 32000 32323-56523

Assay 2 750-30000 10000 8589-13496

Cl : Confidence interval

Claims

19We claim:

1. An improved method for assessing cervical ripeness at term, predicting susceptibility of delivery including the risk for preterm delivery and/or for correct timing of induction of labor in preterm, term and/or post term pregnancies and/or for determining the cause and origin of minor amounts of IGFBP-1, c h a r a c t e r i z e d in that the test is performed by the assessing immunochemically from an essentially blood free cervical secretion sample the presence or absence of highly phosphorylated decidua-derived isoforms of insulin-like growth factor binding protein 1 (IGFBP-1) .

2. The method of claim 1, c h a r a c t e r i z e d in that for excluding premature rupture of membranes (PROM) , an immunochemical test method recognizing the presence of less phosphorylated or nonphosphorylated IGFBP-1 is performed from the same cervical sample before or simultaneously with the assessment of cervical ripeness.

3. The method of claims 1-2, c h a r a c t e r i z e d in that it comprises the steps of :

(a) collecting a cervical sample;

(b) excluding the occurrence of visually detectable blood in the cervical sample;

(c) excluding the occurrence of PROM by carrying out an immunochemical test which recognizes the presence of less phosphorylated or nonphosphorylated IGFBP-1;

(d) contacting a visually blood free sample from step b) with a specific binding substance capable of recognizing the highly phosphorylated isoform of IGFBP-1; and

(e) detecting the presence of said highly phosphorylated 20

isoform of IGFBP-1.

4. The method of claims 1-3, c h a r a c t e r i z e d in that a bed- or chair-side diagnosis is performed by collecting the sample with a solid absorbing sampling device and contacting the sampling device directly or indirectly with the immunochemical test device.

5. The method of claim 1, c h a r a c t e r i z e d in that for the differentiation of PROM and susceptibility of delivery, an assessment of the phosphorylated isoforms of IGFBP-1 and/or the nonphosphorylated IGFBP-1 is performed simultaneously from the same essentially blood free cervical sample using the same immunochromatographic test device.

6. A test device for assessing cervical ripeness, predicting susceptibility of delivery including the risk for preterm delivery and/or for correctly timing induction of labor in preterm, term and/or post term pregnancies and for determining the cause and origin of minor amounts of IGFBP-1, c h a r a c t e r i z e d in that the test device comprises at least one specific binding substance which is capable of recognizing the presence or absence of phosphorylated isoforms of IGFBP-1 in an essentially blood free cervical sample.

7. The test device according to claim 6, c h a r a c t e r i z e d in that it in addition to comprising at least one specific binding substance, which recognizes the phosphorylated isoforms of IGFBP-1 also comprises at least one specific binding substance capable of recognizing the presence of nonphosphorylated IGFBP-1 for exclusion of PROM.

8. A test kit, c h a r a c t e r i z e d in that it comprises at least one test device according to claim 6 and optionally at least one complementary test device to be used parallely with the test device according to claim 6, which complementary test device comprises at least one specific binding substance 21

capable of recognizing an nonphosphorylated IGFBP-1 for exclusion of PROM.

9. The use of binding substances recognizing decidua-derived highly phosphorylated isoforms of IGFBP-1 for predicting susceptibility of delivery including the risk of preterm delivery by assessing the presence or absence of said decidua-derived highly phosphorylated isoforms of IGFBP-1 from a cervical secretion sample.

10. The use of binding substances recognizing decidua-derived highly phosphorylated isoforms of IGFBP-1 for correctly timing the induction of labor in preterm, term and/or post term pregnancies by assessing the presence or absence of phosphorylated isoforms of IGFBP-1 from cervical secretions samples .

11. The use of specific binding substances recognizing decidua-derived highly phosphorylated isoforms of IGFBP-1 in combination with specific binding substances for nonphosphorylated IGFBP-1 for determining the cause and/or origin of minor amounts of IGFBP-1 in order to differentiate between PROM and susceptibility of delivery.