Biomarker pairs for predicting preterm birth

Abstract

Disclosed are isolated biomarker pairs selected from the group consisting of IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein the biomarker pairs show a change in a reversal value between a pregnant female animal at risk for preterm birth and a term birth control. Also provided are methods of determining the probability of preterm birth in a pregnant female animal using the isolated biomarker pairs.

Description

[0001] This application is a divisional application with the same title as the parent invention. The parent invention's Chinese application number is 201680048762.6, its international application number is PCT / US2016 / 038198, and its application date is June 17, 2016.

[0002] This application claims priority to U.S. Provisional Patent Application No. 62 / 290,796, filed February 3, 2016; U.S. Provisional Patent Application No. 62 / 387,420, filed December 24, 2015; and U.S. Provisional Patent Application No. 62 / 182,349, filed June 19, 2015, the entire contents of each of which are incorporated herein by reference.

[0003] This invention generally relates to the field of precision medicine, and more specifically, to compositions and methods for determining the probability of preterm birth in pregnant female animals. Background of the Invention

[0004] According to the World Health Organization, an estimated 15 million babies are born prematurely each year (before the 37th full week of gestation). Reliable data indicate that preterm birth rates are increasing in almost all countries. See also: World Health Organization; March of Dimes; The Partnership for Maternal, Newborn & Child Health; Save the Children. Born too soon: the global action report on preterm birth ISBN 9789241503433 (2012). An estimated one million infants die each year from complications of preterm birth. Globally, preterm birth is the leading cause of neonatal death (infants born within the first four weeks), while the second leading cause of death is pneumonia in children under five. Most survivors face lifelong disabilities, including learning disabilities and visual and hearing problems.

[0005] According to reliable data, preterm birth rates range from 5% to 18% of births in 184 countries. (Blencowe et al., “National, regional, and worldwide estimates of preterm birth”). The Lancet, 9;379(9832):2162-72 (2012). Preterm birth remains a global problem despite over 60% occurring in Africa and South Asia. Countries with the highest preterm birth rates include Brazil, India, Nigeria, and the United States. Of the 11 countries with preterm birth rates exceeding 15%, all but two are located in sub-Saharan Africa. In the poorest countries, on average, 12% of infants are born prematurely, compared to 9% in higher-income countries. Within a country, poorer families face a higher risk. Over three-quarters of premature infants can be saved through feasible and cost-effective care, such as prenatal steroid injections to strengthen the infant's lungs in mothers at risk of preterm birth.

[0006] Preterm infants face a higher risk of death and a variety of health and developmental problems than full-term infants. Complications include acute respiratory problems, gastrointestinal problems, immunological problems, central nervous system problems, hearing and visual problems, as well as long-term motor problems, cognitive problems, behavioral problems, socio-emotional problems, health problems, and growth problems. The birth of a preterm infant can also impose significant emotional and financial costs on families and involve public sector services such as health insurance, education, and other social support systems. The greatest risk of death and morbidity is for infants born in the earliest possible stages of pregnancy. However, those born closer to full term represent the largest majority of preterm infants and still experience more complications than full-term infants.

[0007] To prevent preterm birth in women less than 24 weeks of gestation with an ultrasound showing a dilated cervix, a surgical procedure known as cervical cerclage can be used, in which the cervix is ​​closed with a strong suture. For women less than 34 weeks of gestation in an active preterm delivery, hospitalization may be necessary, and medication may be administered to temporarily stop preterm labor and / or promote fetal lung development. If a pregnant woman is determined to be at risk of preterm birth, healthcare providers can implement a variety of clinical strategies, which may include prophylactic pharmacological treatments such as 17-α-hydroxyprogesterone caproate (Makena) injections and / or vaginal progesterone gel, cervical vaginal suppositories, restriction of sexual activity and / or other physical activity, and modification of chronic conditions that increase the risk of preterm birth (such as diabetes and hypertension).

[0008] There is an urgent need to identify women at risk of preterm birth and provide them with appropriate prenatal care. Closer prenatal monitoring and preventative interventions can be arranged for women identified as high-risk. Current strategies for risk assessment are based on obstetric and medical history, as well as clinical examination; however, these strategies only identify a low proportion of women at risk of preterm delivery. Currently, a prior history of spontaneous preterm birth (sPTB) is the single strongest predictor of subsequent PTB. The probability of a second PTB after one prior sPTB is 30-50%. Other maternal risk factors include: race, low maternal body mass index, and short cervical length. Studies of amniotic fluid, cervical-vaginal fluid, and serum biomarkers predicting sPTB have shown abnormalities in multiple molecular pathways in women who ultimately deliver preterm. Reliable early preterm birth risk identification methods will enable the arrangement of appropriate monitoring and clinical management to prevent preterm delivery. Such monitoring and management can include: more frequent prenatal care visits, continuous cervical length measurement, enhanced education on early signs and symptoms of preterm labor, lifestyle interventions for modifiable risk behaviors (such as smoking cessation), cervical vaginal suppositories, and progesterone therapy. Finally, reliable prenatal identification of preterm labor risk is also crucial for the cost-effective allocation of monitoring resources.

[0009] Despite extensive research on identifying women at risk, PTB prediction algorithms based solely on clinical and demographic factors or using measured serum or vaginal biomarkers have not resulted in clinically useful tests. A more accurate method is needed to identify women at risk during their first pregnancy and at a sufficiently early stage of pregnancy to enable clinical intervention. This invention addresses this need by providing compositions and methods for determining whether a pregnant woman is at risk of preterm birth. Related advantages are also provided. Invention Overview

[0010] This invention provides compositions and methods for predicting the probability of preterm birth in pregnant female animals.

[0011] This invention provides isolated biomarkers selected from the groups described in Table 26. The biomarkers of this invention can predict the risk of preterm birth in pregnant female animals. In some embodiments, the isolated biomarkers are selected from IBP4, SHBG, PSG3, LYAM1, IGF2, CLUS, IBP3, INHBC, PSG2, PEDF, CD14, and APOC3.

[0012] This invention provides alternative peptides for isolated biomarkers selected from the groups described in Table 26. In some embodiments, the alternative peptides for the isolated biomarkers are selected from the alternative peptide groups described in Table 26. The biomarkers and their alternative peptides described in this invention can be used in methods for predicting the risk of preterm birth in pregnant female animals. In some embodiments, the alternative peptides correspond to isolated biomarkers selected from IBP4, SHBG, PSG3, LYAM1, IGF2, CLUS, IBP3, INHBC, PSG2, PEDF, CD14, and APOC3.

[0013] This invention provides stable isotopically labeled standard peptides (SIS peptides) corresponding to alternative peptides selected from the groups described in Table 26. The biomarkers, their alternative peptides, and SIS peptides described in this invention can be used in methods for predicting the risk of preterm birth in pregnant female animals. In some embodiments, the SIS peptides correspond to alternative peptides of isolated biomarkers selected from IBP4, SHBG, PSG3, LYAM1, IGF2, CLUS, IBP3, INHBC, PSG2, PEDF, CD14, and APOC3.

[0014] The present invention provides isolated biomarker pairs selected from the isolated biomarkers listed in Table 26, wherein the biomarker pairs show a variation in the ratio between pregnant females at risk of preterm birth and full-term controls.

[0015] This invention provides isolated biomarker pairs selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein the biomarker pairs show a change in ratio between pregnant females at risk of preterm birth and full-term controls.

[0016] This invention provides isolated biomarker pairs selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein the biomarker pairs show changes in ratios between pregnant females at risk of preterm birth and full-term controls.

[0017] In one embodiment, the present invention provides isolated biomarker pairs selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein said biomarker pairs show reversible value changes between pregnant females at risk of preterm birth and full-term controls.

[0018] In one embodiment, the present invention provides isolated biomarker pairs selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein said biomarker pairs show reversible value changes between pregnant females at risk of preterm birth and full-term controls.

[0019] In one embodiment, the present invention provides a composition comprising alternative peptide pairs corresponding to biomarker pairs selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein the biomarker pairs exhibit reversible changes in value between pregnant females at risk of preterm birth and full-term controls. In one embodiment, each alternative peptide in the composition pair comprises a stable isotope-labeled standard peptide (SIS peptide).

[0020] In one embodiment, the present invention provides a composition comprising alternative peptide pairs corresponding to biomarker pairs selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein the biomarker pairs exhibit reversible changes in value between pregnant females at risk of preterm birth and full-term controls. In one embodiment, each alternative peptide in the composition comprises a stable isotope-labeled standard peptide (SIS peptide).

[0021] In specific embodiments, the present invention provides isolated biomarker pairs IBP4 / SHBG, wherein the biomarker pairs exhibit reversible value changes between pregnant females at risk of preterm birth and full-term controls. In other embodiments, the present invention provides isolated biomarker pairs IBP4 / SHBG, wherein the biomarker pairs exhibit higher ratios in pregnant females at risk of preterm birth compared to full-term controls.

[0022] In one embodiment, the present invention provides a composition comprising an alternative peptide pair corresponding to the biomarker pair IBP4 / SHBG, wherein the biomarker pair exhibits a higher ratio in pregnant females at risk of preterm birth compared to a full-term control. In one embodiment, each alternative peptide in the composition pair comprises a stable isotope-labeled standard peptide (SIS peptide).

[0023] In other embodiments, the present invention provides sets of at least two pairs of biomarkers selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein each pair shows a reversible value change between pregnant females at risk of preterm birth and full-term controls. In one embodiment, the sets of biomarkers derived from alternative peptides for each biomarker comprise stable isotope-labeled standard peptides (SIS peptides).

[0024] In other embodiments, the present invention provides at least two pairs of biomarkers selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein each pair shows a reversible value change between pregnant females at risk of preterm birth and full-term controls. In one embodiment, the pair of biomarkers derived from alternative peptides for each biomarker comprises a stable isotope-labeled standard peptide (SIS peptide).

[0025] In other embodiments, the present invention provides a set of at least two pairs of alternative peptides, each pair corresponding to a pair of biomarkers selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein each pair shows a reversible value change between pregnant females at risk of preterm birth and full-term controls. In one embodiment, each alternative peptide in the set comprises a stable isotope-labeled standard peptide (SIS peptide).

[0026] In other embodiments, the present invention provides sets of at least two pairs of alternative peptides, each pair corresponding to a pair of biomarkers selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein each pair shows a change in reversible value between pregnant females at risk of preterm birth and full-term controls. In one embodiment, each alternative peptide in the set contains a stable isotope-labeled standard peptide (SIS peptide).

[0027] In other embodiments, the present invention provides a set of at least two pairs of alternative peptides, each pair of alternative peptides corresponding to a pair of biomarkers selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein at least one pair shows a reversible change in value between pregnant females at risk of preterm birth and full-term controls. In one embodiment, the composition comprises a stable isotope-labeled standard peptide (SIS peptide) for each pair of alternative peptides.

[0028] In other embodiments, the present invention provides a group of at least two pairs of alternative peptides, each pair of alternative peptides corresponding to a pair of biomarkers selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein at least one pair shows a reversible change in value between pregnant females at risk of preterm birth and full-term controls. In one embodiment, the composition comprises a stable isotope-labeled standard peptide (SIS peptide) for each alternative peptide.

[0029] In other embodiments, the present invention provides a set of at least two pairs of alternative peptides, each pair of alternative peptides corresponding to a pair of biomarkers selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein the calculated score of the set derived from at least two pairs of biomarkers shows a change in value between pregnant female animals and full-term controls. In one embodiment, the composition comprises a stable isotope-labeled standard peptide (SIS peptide) for each alternative peptide.

[0030] In other embodiments, the present invention provides a set of at least two pairs of alternative peptides, each pair of alternative peptides corresponding to a pair of biomarkers selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, wherein the calculated score of the set derived from at least two pairs of biomarkers shows a change in value between pregnant female animals and full-term controls. In one embodiment, the composition comprises a stable isotope-labeled standard peptide (SIS peptide) for each alternative peptide.

[0031] In one embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the ratio of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal, the biomarker pair being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS to determine the probability of preterm birth in the pregnant female animal. In some embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2 .

[0032] In one embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the ratio of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal, the biomarker pair being selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1 to determine the probability of preterm birth in the pregnant female animal. In some embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2In some embodiments, the method includes an initial step of obtaining a biological sample. In some embodiments, the method includes detecting, measuring, or quantifying the SIS alternative peptide for each biomarker.

[0033] In some embodiments, determining the probability of preterm birth in pregnant female animals includes an initial step of forming a probability / risk index by measuring the ratio of isolated biomarkers selected from preterm pregnancy groups and full-term pregnancy groups with known gestational age at birth. In other embodiments, a preterm birth risk index is formed by measuring the ratio of IBP4 / SHBG in preterm and full-term pregnancy groups, where gestational age at birth is recorded. In some embodiments, determining the probability of preterm birth in pregnant female animals includes measuring the IBP4 / SHBG ratio and comparing that value with an index that yields a preterm birth risk using the same isolation and measurement techniques to determine if IBP4 / SHBG falls within the index group.

[0034] In one embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal value of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal, the biomarker pair being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS to determine the probability of preterm birth in the pregnant female animal. In some embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2 In some embodiments, the method includes an initial step of obtaining a biological sample. In some embodiments, the method includes detecting, measuring, or quantifying the SIS alternative peptide for each biomarker.

[0035] In one embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal value of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal, the biomarker pair being selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1 to determine the probability of preterm birth in the pregnant female animal. In some embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2 In some embodiments, the method includes an initial step of obtaining a biological sample. In some embodiments, the method includes detecting, measuring, or quantifying the SIS alternative peptide for each biomarker.

[0036] In another embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring changes in the reversal values ​​of at least two pairs of biomarkers selected from a biological sample obtained from the pregnant female animal, the biomarker pairs being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS to determine the probability of preterm birth in a pregnant female animal. In another embodiment, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring changes in the reversal values ​​of at least two pairs of biomarkers selected from biological samples obtained from the pregnant female animal, the biomarker pairs being IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1 to determine the probability of preterm birth in pregnant female animals. In some embodiments, the reversal values ​​show a relative change in the intensity of a single biomarker between the pregnant female animal and a full-term control and indicate the probability of preterm birth in the pregnant female animal. In other embodiments, the measurement step includes measuring alternative peptides of biomarkers in a biological sample obtained from the pregnant female animal. In some embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2 In some embodiments, the method includes an initial step of obtaining a biological sample. In some embodiments, the method includes detecting, measuring, or quantifying the SIS alternative peptide for each biomarker.

[0037] In one embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal value of a biomarker pair consisting of IBP4 and SHBG in a biological sample obtained from the pregnant female animal to determine the probability of preterm birth in the pregnant female animal. In some embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2 In some embodiments, the method includes an initial step of obtaining a biological sample. In some embodiments, the method includes detecting, measuring, or quantifying the SIS alternative peptide for each biomarker.

[0038] In one embodiment, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring the inversion value of a biomarker pair consisting of the ratio of IBP4 to SHBG (IBP4 / SHBG) in a biological sample obtained from the pregnant female animal to determine the probability of preterm birth in the pregnant female animal, wherein a higher ratio in the pregnant female animal indicates an increased risk of preterm birth compared to a full-term control. In other embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2 In some embodiments, the method includes an initial step of obtaining a biological sample. In some embodiments, the method includes detecting, measuring, or quantifying the SIS alternative peptide for each biomarker.

[0039] In one embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal values ​​of biomarkers IBP4 and SHBG in a biological sample obtained from the pregnant female animal to determine the probability of preterm birth in the pregnant female animal. In some embodiments, the pregnant female animal has a body mass index (BMI) greater than 22 and less than or equal to 37 kg / m². 2 In some embodiments, the method includes an initial step of obtaining a biological sample. In some embodiments, the method includes detecting, measuring, or quantifying the SIS alternative peptide for each biomarker.

[0040] This invention also provides a method for detecting isolated biomarker pairs in pregnant female animals, wherein the isolated biomarker pairs are selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, P EDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, the method comprising the steps of: a. obtaining a biological sample from the pregnant female animal; b. detecting the presence of the separated biomarker pair in the biological sample by contacting the biological sample with a first capture reagent that specifically binds to a first member of the pair and a second capture reagent that specifically binds to a second member of the pair; and detecting the binding between the first biomarker of the pair and the first capture reagent, and between the second member of the pair and the second capture reagent.

[0041] In one embodiment, the present invention provides a method for detecting IBP4 and SHBG in pregnant female animals, the method comprising the steps of: a. obtaining a biological sample from the pregnant female animal; b. detecting the presence of IBP4 and SHBG in the biological sample by contacting the biological sample with a capture reagent that specifically binds IBP4 and a capture reagent that specifically binds SHBG; and c. detecting the binding between IBP4 and the capture reagent and between SHBG and the capture reagent. In one embodiment, the method includes measuring the reversal value of the biomarker pair. In other embodiments, the presence of a change in the reversal value between the pregnant female animal and a full-term control indicates the probability of preterm birth in the pregnant female animal. In one embodiment, the sample is obtained between 19 and 21 weeks of gestational age. In other embodiments, the capture reagent is selected from antibodies, antibody fragments, nucleic acid-based protein binding reagents, small molecules, or variants thereof. In other embodiments, the method is performed by an assay selected from enzyme immunoassay (EIA), enzyme-linked immunosorbent assay (ELISA), and radioimmunoassay (RIA).

[0042] The present invention also provides a method for detecting isolated biomarker pairs in pregnant female animals, said isolated biomarker pairs being selected from IBP4 / SHBG, IBP4 / PSG3, IBP4 / LYAM1, IBP4 / IGF2, CLUS / IBP3, CLUS / IGF2, CLUS / LYAM1, INHBC / PSG3, INHBC / IGF2, PSG2 / LYAM1, PSG2 / IGF2, PSG2 / LYAM1, PEDF / PSG3, PEDF / SHBG, PEDF / LYAM1, CD14 / LYAM1, and APOC3 / LYAM1, said method comprising the steps of: a. obtaining a biological sample from said pregnant female animal; and b. detecting the presence of said isolated biomarker pairs in said biological sample, which includes performing a proteomics workflow consisting of mass spectrometry quantification on said sample.

[0043] In one embodiment, the present invention provides a method for detecting IBP4 and SHBG in pregnant female animals, the method comprising the steps of: a. obtaining a biological sample from the pregnant female animal; and b. detecting the presence of the isolated biomarker pair in the biological sample, comprising performing a proteomics workflow consisting of mass spectrometry quantification on the sample.

[0044] In some embodiments, the reversal value indicates a relative change in the intensity of a single biomarker between pregnant female animals and full-term controls, and it indicates the probability of preterm birth in pregnant female animals. In other embodiments, the measurement step includes measuring a substitute peptide for the biomarker in a biological sample obtained from the pregnant female animal. In one embodiment, a preterm birth risk index is formed by measuring the IBP4 / SHBG ratio in preterm and full-term pregnancy groups, where gestational age at birth is recorded. Then, in clinical practice, the IBP4 / SHBG ratio measured in a single pregnancy is compared with an index derived using the same separation and measurement techniques to determine if IBP4 / SHBG falls within the index group.

[0045] Other features and advantages of the invention will be apparent from the detailed description and claims. Attached Figure Description

[0046] Figure 1 Blood sampling window. Shows the individual reversal performance across the entire blood sampling window. Reversals shown: IBP4 / SHBG; VTNC / VTDB; IBP4 / SHBG; VTNC / SHBG; IBP4 / SHBG; CATD / SHBG; PSG2 / ITIH4; CHL1 / ITIH4; PSG2 / C1QB; PSG2 / FBLN3; HEMO / IBP6; HEMO / PTGDS.

[0047] Figure 2 .GABD discovery cases, review cases, and verification cases.

[0048] Figure 3. Protein expression during pregnancy. Based on known protein functions and understanding of proteins / pathways unaffected by preterm birth, a variety of proteins can be analyzed. Figure 3 shows the expression of pregnancy-related proteins during pregnancy. At the gestational age shown, these proteins and their networks were unaffected by preterm birth pathology.

[0049] Figure 4 Protein expression during pregnancy. Figure 4 An enlarged version of the graph shown in Figure 3 is displayed, which relates to placental-specific growth hormone.

[0050] Figure 5. Protein pathology during pregnancy. Insulin-like growth factor binding protein 4 (IBP4) was overexpressed by at least 10% during the blood sampling window of 19–21 weeks. Sex hormone binding protein (SHBG) was underexpressed by at least 10% during the blood sampling window of 19–21 weeks.

[0051] Figure 6 . Review selection criteria. Figure 6 This demonstrates the standards for implementing high-performance, clinically and analytically robust preterm birth testing.

[0052] Figure 7 Monte Carlo Cross-Validation (MCVV). MCCV is a conservative method for estimating how well a classifier will classify independent sets of samples drawn from the same population (e.g., PAPR).

[0053] Figure 8 Analysis of [IBP4] / [SHBG CHL1 CLUS]. Only relative to IBP4 / SHBG, CHL1 and CLUS showed a performance improvement of 0.03.

[0054] Figure 9 Power and sample size analysis. Power and sample size analysis predicted the likelihood that, at the sample size threshold and performance estimate, the study would be driven to reject the null hypothesis (AUC = 0.5).

[0055] Figure 10 Pregnancy clock and delivery time. Several analytes that rise during pregnancy but remain unchanged in PTB cases and controls can be used to determine gestational age using biochemical methods. Biochemical dates can be used to confirm the date based on the last menstrual period or ultrasound date, or prior to subsequent determination of sPTB risk, TTB, or GAB prediction.

[0056] Figure 11 Classifier development. Figure 11 The standard for developing classifiers is shown.

[0057] Figure 12 Discover pathway coverage in the assay. Figure 12 The distribution of proteins via pathways is shown.

[0058] Figure 13 The PCA analysis of the data detected changes throughout the blood draw window, thus demonstrating that highly multiplexed assays are sensitive to gestational age.

[0059] Figure 14 Hierarchical clustering of proteins measured in the sample was discovered.

[0060] Figure 15 Branching of placental-specific proteins within a large cluster. The right-hand inset shows gene modules expressed during pregnancy as identified by Thompson, while the left-hand inset shows that serum-proteomics assays reproduced the expression of these modules. (Thompson et al., Genome Res. 12(10): 1517-1522 (2002).)

[0061] Figure 16 Disordered protein PreTRM TM sample.

[0062] Figure 17 Biology of sex hormone-binding protein (SHGB) highlighted. SHBG is expressed in placental cells (right panel). SHBG may be responsible for controlling free testosterone and estrogen levels in the fetal compartments of the placenta (left panel).

[0063] Figure 18 Interactions of IBP4, IGF2, PAPP-A, and PRG2. IBP4 is a negative regulator of IGF2. IBP4 does not interact with IGF2 via PAPPA-mediated proteolysis. Low levels of PAPPA have been involved in IUGR and PE. Elevated IBP4 levels are an indicator of inhibited IGF2 activity. PTB cases exhibit inhibited PAPPA and PRG2 levels and elevated IBP4 levels.

[0064] Figure 19 Insulin-like growth factor binding protein 4 (IBP4). IBP4 is upregulated in cases of placental dysplasia (PTB). In early pregnancy, IGF2 stimulates proliferation, differentiation, and EVT invasion. IGF activity is essential for normal placental formation and fetal growth. IBP4 mediates the autocrine and paracrine control of IGF2 activity at the maternal-fetal junction. The activity of IGF2 expressed in the cytotrophoblast is balanced by IBP produced by decidual cells. Elevated IBP4 and decreased IGF2 in the first trimester are associated with placental dysfunction (e.g., IUGR / SGA).

[0065] Figure 20. Correlation between MS and ELISA for IBP4, SHBG, and CHL1. For important analytes, mass spectrometry and ELISA showed good agreement. The agreement between two uncorrelated platforms confirms the reliability of analyte measurements.

[0066] Figure 21 PTB classification of IBP4 / SHBG in samples from GABD (Gas-Induced Abnormalities) at 19–21 weeks of gestation. Samples from GABD at 19–21 weeks of gestation were categorized into high and low BMI groups. In the high BMI group, the reversal value of IBP4 / SHBG was higher due to lower SHBG values. In the low BMI group, the separation between cases and controls was greater.

[0067] Figure 22 Suppressed SHBG levels in PTB cases at low BMI. Linear fit of serum SHBG levels throughout GABD in PAPR subjects. SHBG levels are suppressed by high BMI. SHBG levels increase throughout pregnancy. PTB cases have decreased SHBG levels at low BMI, which then accelerates throughout pregnancy.

[0068] Figure 23 The distribution of study participants in PAPR was summarized.

[0069] Figure 24 The ROC curves and corresponding AUC values ​​of the validation sample groups divided by BMI using the IBP4 / SHBG predictor factor classification are shown.

[0070] Figure 25 The positive predictive value (PPV) of incidence rate modulated is displayed, which is a clinical risk measure as a function of predictor scores. The relationship between predictor scores and PPV allows for the determination of the sPTB risk probability for any unknown subject. The topmost line (purple) on the risk curve corresponds to GAB < 35 0 / 7 weeks; the second line from the top (red) corresponds to GAB between 35 0 / 7 and 37 0 / 7 weeks; the third line from the top (green) corresponds to GAB between 37 0 / 7 and 39 0 / 7 weeks; and the fourth line from the top (blue) corresponds to GAB ≥ GAB 39 0 / 7 weeks.

[0071] Figure 26 This shows the delivery rates for high- and low-risk groups as events in the Kaplan-Meier analysis. High and low risk are defined as higher or lower than those from... Figure 25 The average population SPTB risk (=14.6%) in the data is 2× the relative risk.

[0072] Figure 27 The ROC curves for predictor performance are shown for subject combinations using blinded verification and validation analyses within the optimal BMI and GA intervals. The ROC curve for the combination samples corresponds to an AUROC of 0.72 (p = 0.013).

[0073] Figure 28 The results showed that 44 proteins were upregulated or downregulated in the overlapping 3-week GA intervals and passed through the analysis filter.

[0074] Figure 29 The best performing inverse IBP4 / SHBG was observed to be AUROC = 0.74 during the interval from 19 0 / 7 to 21 6 / 7.

[0075] Figure 30 The mean AUROC obtained from 2,000 bootstrapping iterations was 0.76. In the review sample, the blinded IBP4 / SHBGAUROC performance was 0.77 for all subjects and 0.79 for subjects categorized by BMI, which is in good agreement with the performance found in the findings. After blinded review, the findings and review samples were pooled for bootstrapping performance determination.

[0076] Figure 31 shows the separation of sPTB cases vs. controls obtained from MS vs. ELISA scores.

[0077] Figure 32 shows the immunoassay versus MS ROC analysis without BMI limitations.

[0078] Figure 33 shows the immunoassay versus MS ROC analysis for BMI greater than 22 and less than or equal to 37.

[0079] Figure 34 shows the correlation between IBP4 / SHBG scores obtained from MS and ELISA within GABD133-146 for subjects categorized by BMI (left inset) and all subjects (right inset).

[0080] Figure 35 shows the ELISA and MS separation of controls and cases (BMI classification).

[0081] Figure 36 shows the ELISA and MS separation of controls and cases (all BMIs).

[0082] Figure 37 The comparison of SHBG measurements using the Abbott Architect CMIA (semi-automated immunoassay analyzer) and the Sera Prognostics' proteomic analysis method (which includes immune consumption of samples, enzymatic digestion, and analysis on an Agilent 6490 mass spectrometer) is shown.

[0083] Figure 38 The comparison of SHBG measurements is shown using a Roche cobas e602 analyzer (a semi-automated immunoassay analyzer) and a Sera Prognostics proteomic analysis method (which includes immune consumption of samples, enzymatic digestion, and analysis on an Agilent 6490 mass spectrometer).

[0084] Figure 39 The comparison of SHBG measurements using the Abbott Architect CMIA and Roche cobas e602 analyzers (both semi-automated immunoassay analyzers) is shown.

[0085] Figure 40 The domain and structural features of the longest isoform of the IBP4 protein (Uniprot: P22692) are shown. The IBP4QCHPALDGQR(aa, 214-223) peptide (SEQ ID NO: 2) is located within the thyroglobulin type 1 domain. IBP4 has a single N-linked glycosylation site at residue 125.

[0086] Figure 41The position of the QCHPALDGQR peptide (SEQ ID NO:2) in the two IBP4 isoforms (SEQ ID NOS 158 and 159 in order of appearance) is highlighted.

[0087] Figure 42 The domain and structural features of the longest isoform of the SHBG protein (Uniprot: P04278) are shown. The SHBGIALGGLLFPASNLR (aa, 170-183) peptide (SEQ ID NO: 18) is located in the first Lamin G-like domain. SHBG has three glycosylation sites; two N-linked sites at residues 380 and 396; and one O-linked site at residue 36.

[0088] Figure 43 The position of the IALGGLLFPASNLR peptide (SEQ ID NO: 18) in exon 4 of seven SHBG isoforms (in order of appearance: SEQ ID NO: 160, 160, 160, 160, 160, 160 and 161) is highlighted.

[0089] Figure 44 The mean response ratio of IBP4 levels for spTB cases and term-pregnant controls is shown throughout gestational age (GABD) at the time of blood draw. Cross-sectional findings were analyzed using a sliding 10-day window with smoothing. Cases corresponding to the control signal corresponded to a maximum difference of approximately 10%.

[0090] Figure 45 The mean response ratio of SHBG levels for spTB cases and term-pregnant controls is shown throughout gestational age (GABD) at the time of blood draw. Cross-sectional findings were analyzed using a sliding 10-day window with smoothing. Cases corresponding to control signals corresponded to a maximum difference of approximately 10%.

[0091] Figure 46 The IBP4 / SHBG predictor scores are displayed individually for spTB cases and term-pregnancy controls throughout gestational age (GABD) at the time of blood draw. Crosssectional findings were analyzed using a sliding 10-day window with smoothing. The differences were compared to approximately 10% in individual analyte signals. Figure 45 and 46 The maximum difference between the two curves corresponds to approximately 20% variation. These data demonstrate the amplification of the diagnostic signal obtained using the IBP4 / SHBG reversal strategy.

[0092] Figure 47 shows the diagnostic signal amplification caused by the formation of multiple different reversals. To investigate whether the formation of reversals in general amplifies the diagnostic signal, we examined the diagnostic performance of reversals formed by multiple different proteins by ROC analysis. The range of AUC values (sPTB cases versus term controls) using a data set of samples collected between 19 / 0 and 21 / 6 weeks of gestation is shown in the top inset. The nearby box plot shows the range of ROC performance of the individual upregulated and downregulated proteins used to form the relevant reversals. Similarly, the lower inset shows that the p-values of the reversals from the Wilcoxon test (sPTB cases versus term controls) are more significant than the p-values of the corresponding individual proteins.

[0093] Figure 48 Shows the measurement of the individual IBP4 and SHBG response ratios and the calculated coefficient of variation (CV) of the corresponding reversal scores. Pooled control serum samples (pHGS) from pregnant donors with no biological differences were analyzed over multiple batches and over several days. The reversal variation is less than the variation associated with individual proteins. These data suggest that reversal controls are formed for analytical variation occurring during the laboratory handling of samples. Analytical variation is not a biological phenomenon.

[0094] Figure 49 Shows the analytical CVs of multiple reversals and their individual upregulated and downregulated proteins. To investigate whether the formation of reversals in general amplifies the diagnostic signal, we examined the ROC performance (AUC) of high-performance reversals (AUC > 0.6) formed by ratios of multiple proteins. The range of AUC values (sPTB cases versus term controls) using a data set of samples collected between 19 / 0 and 21 / 6 weeks of gestation is shown in the top inset. The nearby box plot shows the range of ROC performance of the individual upregulated and downregulated proteins used to form the relevant reversals. Similarly, the p-values of the reversals from the Wilcoxon test (sPTB cases versus term controls) are more significant than the p-values of the corresponding individual proteins.

[0095] Figure 50 Shows the PreTRM TM score comparison of subjects annotated as medically indicated for preeclampsia relative to other indications.

[0096] Figure 51 Shows a table of measures of the IBP4 / SHBG predictor performance in a validation sample set (22 < BMI <= 37). Using different cutoffs to define cases (less than the cutoff) relative to controls (greater than the cutoff), the sensitivity, specificity, area under the ROC curve (AUC), and odds ratio of the predictor were determined.

[0097] Figure 52A heatmap showing the intensity of reversal with diabetes annotations is displayed. Red arrows indicate diabetes cases. Samples are listed at the bottom, with PTB cases on the right side of the screen and full-term deliveries on the left. The diabetes cases are clustered on the right, demonstrating the potential for establishing diagnostic tests based on biomarkers to predict gestational diabetes.

[0098] Figure 53 The hierarchical clustering of the analyte response ratios is displayed.

[0099] Figure 54 shows differentially expressed proteins that play a role in extracellular matrix interactions.

[0100] Figure 55 shows a kinetic diagram of differentially expressed proteins that play a role in the IGF-2 pathway, exhibiting maximum separation at 18 weeks.

[0101] Figure 56A The interaction between IGF-2, IBP4, PAPP1, and PRG2 proteins that affect the bioavailability of these proteins in sPTB is shown; 56B shows a schematic diagram of intracellular signals preferentially activated by insulin and IGF through binding to IR-B and by insulin through binding to either IR-A or IGF1R.

[0102] Figure 57 shows a kinetic diagram of differentially expressed proteins that play a role in metabolic hormone homeostasis.

[0103] Figure 58 shows a kinetic diagram of differentially expressed proteins that play a role in angiogenesis.

[0104] Figure 59 shows a kinetic diagram of differentially expressed proteins that play a role in innate immunity.

[0105] Figure 60 shows a kinetic diagram of differentially expressed proteins that play a role in blood clotting.

[0106] Figure 61 shows the kinetics of differentially expressed serum / secreted proteins.

[0107] Figure 62 shows the kinetics of differentially expressed PSGs / IBPs.

[0108] Figure 63 shows the kinetics of differentially expressed ECM / cell surface proteins.

[0109] Figure 64 shows the kinetics of differentially expressed complement / acute phase protein-1.

[0110] Figure 65 shows the kinetics of differentially expressed complement / acute phase protein-2.

[0111] Figure 66 shows the kinetics of differentially expressed complement / acute phase protein-3.

[0112] Figure 67 shows the kinetics of differentially expressed complement / acute phase protein-4.

[0113] Figure 68 shows the kinetics of the analytes indicated in insets A through I, based on data from gestational age (GABD) from 17 weeks 0 days to 28 weeks 6 days.

[0114] Figure 69 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0115] Figure 70 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0116] Figure 71 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0117] Figure 72 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0118] Figure 73 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0119] Figure 74 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0120] Figure 75 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0121] Figure 76 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0122] Figure 77 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0123] Figure 78 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from day 0 of week 17 to day 6 of week 28.

[0124] Figure 79 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0125] Figure 80 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0126] Figure 81 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0127] Figure 82 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0128] Figure 83 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0129] Figure 84 shows the kinetic plots of the analytes indicated in insets A through I, based on data from GABD from 17 weeks 0 days to 28 weeks 6 days.

[0130] Figure 85 shows the kinetics of peptide conversion indicated in insets A through G, based on data from GABD from day 0 of week 17 to day 6 of week 28.

[0131] Figure 86 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestational age at birth relative to >=37 0 / 7 weeks.

[0132] Figure 87 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestational age at birth relative to >=37 0 / 7 weeks.

[0133] Figure 88 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestation relative to >=37 0 / 7 weeks of gestation.

[0134] Figure 89 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestational age at birth relative to >=37 0 / 7 weeks.

[0135] Figure 90 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestation relative to >=37 0 / 7 weeks of gestation.

[0136] Figure 91 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestational age at birth relative to >=37 0 / 7 weeks.

[0137] Figure 92 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestation relative to >=37 0 / 7 weeks of gestation.

[0138] Figure 93 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestational age at birth relative to >=37 0 / 7 weeks.

[0139] Figure 94 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <37 0 / 7 weeks of gestational age at birth relative to >=37 0 / 7 weeks.

[0140] Figure 95 shows the kinetics of peptide conversion using a cutoff value of <37 0 / 7 weeks of gestation relative to >=37 0 / 7 weeks of gestation. Insets A through C indicate the peptide conversion kinetics.

[0141] Figure 96 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestational age at birth relative to >=35 0 / 7 weeks.

[0142] Figure 97 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestational age at birth relative to >=35 0 / 7 weeks.

[0143] Figure 98 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestation relative to >=35 0 / 7 weeks of gestation.

[0144] Figure 99 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestation relative to >=35 0 / 7 weeks of gestation.

[0145] Figure 100 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestation relative to >=35 0 / 7 weeks of gestation.

[0146] Figure 101 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestational age at birth relative to >=35 0 / 7 weeks.

[0147] Figure 102 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestational age at birth relative to >=35 0 / 7 weeks.

[0148] Figure 103 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestational age at birth relative to >=35 0 / 7 weeks.

[0149] Figure 104 shows the kinetics of peptide conversion indicated in insets A through I, using a cutoff value of <35 0 / 7 weeks of gestational age at birth relative to >=35 0 / 7 weeks.

[0150] Figure 105 shows the kinetics of peptide conversion using a cutoff value of <35 0 / 7 weeks of gestation relative to >=35 0 / 7 weeks of gestation. Insets A through C indicate this kinetics.

[0151] Figure 106 shows the levels of IBP4 and SHBG, and the reversal value of IBP4 / SHBG, in sPTB cases and controls, respectively.

[0152] Figure 107 shows the correlation between MSD results using commercial ELISA kits and MS-MRM.

[0153] Figure 108 provides a box plot showing examples of reversal that performed well at 19–20 weeks in preterm (PTL) births without PPROM.

[0154] Figure 109 provides a box plot showing examples of reversal in preterm premature rupture of membranes (PPROM) with good performance at 19–20 weeks.

[0155] Figure 110 This is a risk curve showing the relationship between the predictor score (lnIBP4 / SHBG) and the morbidity-modified relative risk (positive predictive value) of spTB, using a cutoff value of <37 0 / 7 weeks of pregnancy versus >=37 0 / 7 weeks. The topmost line (purple) on the risk curve corresponds to spTB (GAB < 35 weeks); the second line from the top (red) corresponds to spTB (35 ≤ GAB ≤ 37 weeks); the third line from the top (green) corresponds to term delivery (37 ≤ GAB ≤ 39 weeks); and the fourth line from the top (blue) corresponds to term delivery (39 weeks ≤ GAB).

[0156] Figure 111 This is a risk curve showing the relationship between the predictor score (lnIBP4 / SHBG) and the morbidity-modified relative risk of spTB (positive predictive value) using a cutoff value of <35 0 / 7 weeks of pregnancy versus >=35 0 / 7 weeks of pregnancy. The topmost line (purple) below the risk curve corresponds to spTB (GAB < 35 weeks); the second line from the top (red) corresponds to spTB (35 ≤ GAB ≤ 37 weeks); the third line from the top (green) corresponds to term delivery (37 ≤ GAB ≤ 39 weeks); and the fourth line from the top (blue) corresponds to term delivery (39 weeks ≤ GAB). Invention Details

[0157] This invention is generally based on the finding that certain proteins and peptides in biological samples obtained from pregnant female animals are differentially expressed in pregnant female animals at high risk of preterm birth, relative to controls. This invention is also specifically based in part on the unexpected finding that the reversal values ​​of the biomarker pairs disclosed herein can be used with high sensitivity and specificity in methods for determining the probability of preterm birth in pregnant female animals. The proteins and peptides disclosed herein as components of ratios and / or reversal pairs are used in pregnant female animals at risk of PTB, in the form of ratios, reversal pairs alone, or as groups of biomarkers / reversal pairs, as biomarkers for classifying test samples, predicting the probability of preterm birth, predicting the probability of full-term birth, predicting gestational age at birth (GAB), predicting time to delivery (TTB), and / or monitoring the development of preventative therapies. The reversal value is the ratio of the relative peak area of ​​an upregulated biomarker to the relative peak area of ​​a downregulated biomarker and is used to normalize the degree of difference and amplify the diagnostic signal. Part of this invention lies in the selection of specific biomarkers that can be used to predict the probability of preterm birth based on the reversal value when the pairs are combined. Therefore, selecting a specific biomarker that, once paired, is informative in the potential novel reversal of the present invention is an inventive act.

[0158] The term "reversal value" refers to the ratio of the relative peak area of ​​an upregulated analyte to the relative peak area of ​​a downregulated analyte and is used to normalize variability and amplify diagnostic signals. Subgroups can be selected based on the performance of each univariate among all possible reversals within a narrow window. As disclosed herein, the ratio of the relative peak area of ​​an upregulated biomarker to the relative peak area of ​​a downregulated biomarker, referred to herein as the reversal value, can be used in pregnant female animals to identify robust and accurate classifiers and predict the probability of preterm birth, the probability of full-term birth, the probability of gestational age at birth (GAB), the time of delivery, and / or to monitor the progress of preventative therapies. Therefore, this invention is partly based on the identification of biomarker pairs, wherein the relative expression reversal of the biomarker pair shows a change in reversal value between PTB and non-PTB. In the methods disclosed herein, the use of biomarker ratios corrects for variability resulting from artificially regulated outcomes after the removal of biological samples from pregnant female animals. This variability can be introduced during sample collection, processing, consumption, digestion, or any other step of a method for measuring the presence of a biomarker in a sample, and the variability is independent of the inherent performance of the biomarker. Therefore, the present invention generally includes the use of reverse pairs in diagnostic or prognostic methods to reduce variability and / or amplify, normalize, or clarify diagnostic signals.

[0159] Although the term reversal value refers to the ratio of the relative peak area of ​​the upregulated analyte to the relative peak area of ​​the downregulated analyte and is used to normalize differences and amplify diagnostic signals, it is also contemplated that the biomarker pairs of the present invention can be measured in any other way, for example, by subtraction, addition, or multiplication of relative peak areas. The methods disclosed herein cover the measurement of biomarker pairs by these other methods.

[0160] This approach is advantageous because it provides the simplest possible classifier, independent of data normalization, which helps avoid overfitting and results in very simple experimental tests that are easy to implement in clinics. The use of biomarker pairs based on changes in reversal values ​​independent of data normalization enables the development of clinically relevant biomarkers disclosed herein. Since the quantification of any single protein is subject to uncertainties arising from measurement variability, normal fluctuations, and individual relevant changes in baseline expression, the identification of biomarker pairs under coordinated, systematic control enables robust methods for individual diagnosis and prognosis.

[0161] This invention discloses groups, methods, and kits for determining the probability of preterm birth in pregnant female animals, including biomarker reversal pairs and associated reversal pairs. A key advantage of this invention is that it allows for the assessment of the risk of developing preterm birth early in pregnancy, enabling timely initiation of appropriate surveillance and clinical management to prevent preterm delivery. This invention is particularly useful for female animals lacking any preterm birth risk factors and those that would otherwise go unidentified and untreated.

[0162] For example, the present invention discloses a method for generating useful results in determining the probability of preterm birth in pregnant female animals by obtaining a dataset associated with a sample, wherein the dataset includes at least quantitative data on the relative expression of pairs of biomarkers identified as showing changes in reversal values ​​indicative of preterm birth, and an analytical method for inputting the dataset into an analytical method for generating useful results in determining the probability of preterm birth in pregnant female animals. As further described below, the quantitative data may include amino acids, peptides, polypeptides, proteins, nucleotides, nucleic acids, nucleosides, sugars, fatty acids, steroids, metabolites, carbohydrates, fats, hormones, antibodies, regions of interest used as substitutes for biological macromolecules, and combinations thereof.

[0163] In addition to the specific biomarkers identified in this disclosure through, for example, accession numbers, sequences, or references in public databases, the invention also contemplates the use of biomarker variants that have at least 90%, 95%, or 97% identity with the illustrated sequences and are now known or subsequently discovered and useful to the methods described herein. These variants may represent polymorphisms, splicing variants, mutations, etc. In this regard, this specification discloses a variety of proteins known in the art within the context of the invention and provides exemplary accession numbers associated with one or more public databases and exemplary references to published journal articles associated with these proteins known in the art. However, those skilled in the art will recognize that other accession numbers and journal articles can be readily identified, which may provide other characteristics of the disclosed biomarkers, and the illustrated references are by no means a limitation on the disclosed biomarkers. As described herein, a variety of techniques and reagents are useful in the methods described in the invention. In the context of the invention, suitable samples include, for example, blood, plasma, serum, amniotic fluid, vaginal secretions, saliva, and urine. In some embodiments, the biological sample is selected from whole blood, plasma, and serum. In a particular embodiment, the biological sample is serum. As described herein, biomarkers can be detected using a variety of assays and techniques known in the art. As further described herein, these assays include, without limitation, mass spectrometry (MS)-based assays, antibody-based assays, and assays combining aspects of both methods.

[0164] Protein biomarkers that are components of the reversal pairs described herein include, for example, insulin-like growth factor binding protein 4 (IBP4), sex hormone binding protein (SHBG), vitrinin (VTNC), group-specific component (vitamin D binding protein) (VTDB), cathepsin D (lysosomal aspartic acid protease) (CATD), pregnancy-specific β-1-glycoprotein 2 (PSG2), endogenous α-trypsin inhibitor heavy chain family member 4 (ITIH4), cell adhesion molecule L1-like protein (CHL1), complement component 1, Q subfraction, B chain (C1QB), fibula protein 3 (FBLN3), hemobinin (HEMO or HPX), insulin-like growth factor binding protein 6 (IBP6), and prostaglandin D2 synthase 21 kDa (PTGDS).

[0165] In some embodiments, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring the reversal values ​​of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal, the biomarker pair being selected from the figures and tables, including... Figure 1 Those pairs listed in any of them.

[0166] In some embodiments, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal values ​​of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal to determine the probability of preterm birth in the pregnant female animal, said biomarker pair being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, HPX / PTGDS.

[0167] This invention provides isolated biomarkers selected from the groups described in Table 26. The biomarkers of this invention can predict the risk of preterm birth in pregnant female animals. In some embodiments, the isolated biomarkers are selected from IBP4, SHBG, VTNC, VTDB, CATD, PSG2, ITIH4, CHL1, C1QB, FBLN3, HPX, and PTGDS. In some embodiments, the isolated biomarkers are selected from IBP4, SHBG, PSG3, LYAM1, IGF2, CLUS, IBP3, INHBC, PSG2, PEDF, CD14, and APOC3.

[0168] This invention provides alternative peptides for isolated biomarkers selected from the groups described in Table 26. In some embodiments, the alternative peptides for the isolated biomarkers are selected from the alternative peptide groups described in Table 26. The biomarkers and their alternative peptides described in this invention can be used in methods for predicting the risk of preterm birth in pregnant female animals. In some embodiments, the alternative peptides correspond to isolated biomarkers selected from IBP4, SHBG, VTNC, VTDB, CATD, PSG2, ITIH4, CHL1, C1QB, FBLN3, HPX, and PTGDS. In some embodiments, the alternative peptides correspond to isolated biomarkers selected from IBP4, SHBG, PSG3, LYAM1, IGF2, CLUS, IBP3, INHBC, PSG2, PEDF, CD14, and APOC3.

[0169] This invention provides stable isotopically labeled standard peptides (SIS peptides) corresponding to alternative peptides selected from the groups described in Table 26. The biomarkers, their alternative peptides, and SIS peptides described in this invention can be used in methods for predicting the risk of preterm birth in pregnant female animals. In some embodiments, the SIS peptide corresponds to an alternative peptide of the isolated biomarker selected from IBP4, SHBG, VTNC, VTDB, CATD, PSG2, ITIH4, CHL1, C1QB, FBLN3, HPX, and PTGDS. In some embodiments, the SIS peptide corresponds to an alternative peptide of an isolated biomarker selected from IBP4, SHBG, PSG3, LYAM1, IGF2, CLUS, IBP3, INHBC, PSG2, PEDF, CD14, and APOC3.

[0170] In some embodiments, the present invention provides isolated biomarker pairs IBP4 / SHBG, wherein the biomarker pairs exhibit reversible value changes between pregnant females at risk of preterm birth and full-term controls. In other embodiments, the present invention provides isolated biomarker pairs IBP4 / SHBG, wherein the biomarker pairs exhibit higher ratios in pregnant females at risk of preterm birth compared to full-term controls.

[0171] In some embodiments, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal value of at least one pair of biomarkers selected from biological samples obtained from the pregnant female animal, the biomarker pair being IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, and CATD / SHBG to determine the probability of preterm birth in the pregnant female animal. In other embodiments, samples are obtained between 19 and 21 weeks of GABD. In other embodiments, samples are obtained between 19 and 22 weeks of GABD.

[0172] In some embodiments, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal value of IBP4 / SHBG in a biological sample obtained from the pregnant female animal to determine the probability of preterm birth in the pregnant female animal. In other embodiments, samples are obtained between 19 and 21 weeks of GABD. In other embodiments, samples are obtained between 19 and 22 weeks of GABD.

[0173] In some embodiments, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring the reversal value of at least one pair of biomarkers selected from biological samples obtained from the pregnant female animal, the biomarker pair being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, HPX / PTGDS to determine the probability of preterm birth in the pregnant female animal, wherein the presence of a change in the reversal value between the pregnant female animal and a full-term control determines the probability of preterm birth in the pregnant female animal. In other embodiments, samples are obtained between 19 and 21 weeks of GABD. In other embodiments, samples are obtained between 19 and 22 weeks of GABD.

[0174] Embodiments of the present invention include an iterative method for determining the probability of preterm birth in pregnant female animals. The method includes measuring the reversal value of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal. The biomarker pair is selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, HPX / PTGDS, and any other biomarker pairs selected from proteins described and / or illustrated herein to determine the probability of preterm birth in the pregnant female animal. The presence of a change in the reversal value between the pregnant female animal and a full-term control determines the probability of preterm birth in the pregnant female animal. The iterative performance of the method described herein includes subsequent measurements of a single sample and obtaining subsequent samples for measurement. For example, if the probability of preterm birth (which can be represented as a risk score) in pregnant female animals is determined to be higher than a given value, the method can be repeated using different reversal pairs from the same sample or the same or different reversal pairs from subsequent samples to further classify the risk of sPTB.

[0175] In addition to specific biomarkers, this invention also discloses biomarker variants having about 90%, about 95%, or about 97% identity with the illustrated sequences. As used herein, variants include polymorphisms, splicing variants, mutations, etc. Although described with reference to protein biomarkers, changes in the reversal value of biomarker pairs can be identified at the protein or gene expression level.

[0176] Other biomarkers may be selected from one or more risk indicators, including (but not limited to) maternal characteristics, medical history, previous pregnancy history, and marital and reproductive history. These other biomarkers may include, for example, previous low birth weight or preterm delivery, multiple spontaneous abortions in the second trimester, previous induced abortion in the first trimester, family and intergenerational factors, infertility, nulliparity, placental abnormalities, cervical and uterine abnormalities, short cervical length measurement, gestational hemorrhage, intrauterine growth restriction, intrauterine diethylstilbestrol exposure, multiple pregnancies, infant sex, short stature, low pre-pregnancy weight, low or high body mass index, diabetes, hypertension, genitourinary tract infection (i.e., urinary tract infection), asthma, anxiety and depression, asthma, hypertension, and hypothyroidism. Demographic risk indicators for preterm birth may include, for example, maternal age, race / ethnicity, marital status (single or non-single), low socioeconomic status, maternal age, occupational physical activity, occupational exposure, and environmental exposure and stress. Other risk indicators may include inadequate prenatal care, smoking, use of marijuana and other illicit drugs, cocaine use, alcohol consumption, caffeine intake, maternal weight gain, dietary intake, sexual activity in late pregnancy, and physical activity during leisure time. (Preterm Birth: Causes, Consequences, and Prevention, Institute of Medicine (US) Committee on Understanding Premature Birth and Assuring Healthy Outcomes; edited by Behrman RE and Butler AS, Washington (DC): National Academies Press (US); 2007). Other risk indicators useful as biomarkers can be identified using learning algorithms known in the art, such as linear discriminant analysis, support vector machine classification, regression feature elimination, predictive analysis of microarrays, logistic regression, CART, FlexTree, LART, random forest, MART, and / or survival analysis regression, which are known to those skilled in the art and are further described herein.

[0177] It should be noted that unless explicitly stated in the context, the singular forms of "an" and "described" as used in this specification and the appended claims include plural references. Thus, for example, a reference to "biomarker" includes a mixture of two or more biomarkers, etc.

[0178] Specifically, the term “about” in relation to a given quantity indicates a deviation of plus or minus 5%.

[0179] Unless expressly stated in the content, as used in this application including the appended claims, the singular forms “a” and “the” include plural references and are used interchangeably with “at least one” and “one or more”.

[0180] As used herein, the terms “comprising,” “including,” “containing,” and any variations thereof are intended to cover non-exclusive inclusions, that is, a process, method, or method-defined composition of a product or substance that includes, comprises, or contains elements or a list of elements, may include not only those elements, but may also include other elements not expressly listed or not inherent to the product or substance composition that is not defined by such process, method, or method.

[0181] As used herein, the term "group" refers to a composition, such as an array or collection, containing one or more biomarkers. The term may also refer to a spectrum or index of the expression type of one or more biomarkers described herein. The number of biomarkers used for a biomarker group is based on the sensitivity and specificity values ​​of a particular combination of biomarker values.

[0182] As used herein and unless otherwise stated, the terms “isolated” and “purified” generally describe a composition of substances that has been removed from its natural environment (e.g., the natural environment if it is naturally occurring) and thus altered from its natural state by human hand, thereby having significantly different characteristics in at least one of its structure, function, and properties. Isolated proteins or nucleic acids differ from the way they exist in nature and include synthetic peptides and proteins.

[0183] The term "biomarker" refers to a biomolecule or fragment whose changes and / or detection can be associated with a specific physical condition or state. The terms "marker" and "biomarker" are used interchangeably throughout this disclosure. For example, the biomarkers of this invention are associated with an increased likelihood of preterm birth. These biomarkers include any suitable analytes, but are not limited to biomolecules, including nucleotides, nucleic acids, nucleosides, amino acids, sugars, fatty acids, steroids, metabolites, peptides, polypeptides, proteins, carbohydrates, fats, hormones, antibodies, regions of interest used as alternatives to biological macromolecules, and combinations thereof (e.g., glycoproteins, ribonucleoproteins, lipoproteins). The term also includes portions or fragments of biomolecules, such as peptide fragments of proteins or polypeptides containing at least 5 consecutive amino acid residues, at least 6 consecutive amino acid residues, at least 7 consecutive amino acid residues, at least 8 consecutive amino acid residues, at least 9 consecutive amino acid residues, at least 10 consecutive amino acid residues, at least 11 consecutive amino acid residues, at least 12 consecutive amino acid residues, at least 13 consecutive amino acid residues, at least 14 consecutive amino acid residues, at least 15 consecutive amino acid residues, at least 5 consecutive amino acid residues, at least 16 consecutive amino acid residues, at least 17 consecutive amino acid residues, at least 18 consecutive amino acid residues, at least 19 consecutive amino acid residues, at least 20 consecutive amino acid residues, at least 21 consecutive amino acid residues, at least 22 consecutive amino acid residues, at least 23 consecutive amino acid residues, at least 24 consecutive amino acid residues, at least 25 consecutive amino acid residues, or more consecutive amino acid residues.

[0184] As used herein, the term "alternative peptide" refers to a peptide selected in an MRM assay configuration for quantification of the biomarker of interest. Quantification of the alternative peptide is best achieved using a stable isotope-labeled standard alternative peptide ("SIS alternative peptide" or "SIS peptide") in conjunction with an MRM assay. The alternative peptide may be synthetic. SIS alternative peptides with a relabeled C-terminus, for example, containing arginine or lysine or any other amino acid, can be synthesized for use as an internal standard in MRM assays. SIS alternative peptides are not naturally occurring peptides and have significantly different structures and properties compared to their naturally occurring counterparts.

[0185] In some embodiments, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring the ratio of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal to determine the probability of preterm birth in the pregnant female animal, the biomarker pair being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, HPX / PTGDS, wherein the presence of a change in the ratio between the pregnant female animal and a full-term control determines the probability of preterm birth in the pregnant female animal. In some embodiments, the ratio may include an upregulated protein in the molecule, a downregulated protein in the denominator, or both. For example, as illustrated herein, IBP4 / SHBG is an upregulated protein in the molecule and a downregulated protein in the denominator, which is defined herein as “reversed”. In cases where the ratio includes upregulated proteins in the molecule or downregulated proteins in the denominator, the upregulated proteins will be used for normalization (e.g., to reduce pre-analytical or analytical variability). In specific cases where the ratio is “reversed,” both amplification and normalization are possible. It should be understood that the methods described in this invention are not limited to reversing subgroups but also cover ratios of biomarkers.

[0186] As used herein, the term "reversal" refers to the ratio of the measured value of an upregulated analyte to the measured value of an downregulated analyte. In some implementations, the analyte value itself is the ratio of the peak area of ​​the endogenous analyte to the peak area of ​​the corresponding stable isotope standard analyte, which is referred to herein as the response ratio or relative ratio.

[0187] As used herein, the term "reversal pair" refers to a pair of biomarkers that show a change in value between compared categories. Detection of reversals in protein concentration or gene expression levels eliminates the need for data normalization or the establishment of population-wide thresholds. In some embodiments, the reversal pair is a separate biomarker pair IBP4 / SHBG, wherein the reversal pair shows a change in reversal value between pregnant females at risk of preterm birth and full-term controls. In other embodiments, the reversal pair IBP4 / SHBG shows a higher ratio in pregnant females at risk of preterm birth compared to full-term controls. Within the definition of any reversal pair, a corresponding reversal pair in which a single biomarker switches between the numerator and denominator is included. Those skilled in the art will understand that such corresponding reversal pairs are equally informative for their predictive power.

[0188] The term "reversal value" refers to the ratio of the relative peak area of ​​an upregulated analyte to the relative peak area of ​​a downregulated analyte and is used to normalize variability and amplify diagnostic signals. Within a narrow window of all possible reversals, subgroups can be selected based on the performance of each univariate. As disclosed herein, the ratio of the relative peak area of ​​an upregulated biomarker to the relative peak area of ​​a downregulated biomarker, referred to herein as the reversal value, can be used in pregnant female animals to identify robust and accurate classifiers and predict the probability of preterm birth, the probability of full-term birth, the gestational age at birth (GAB), the time of delivery, and / or to monitor the progress of preventative therapies.

[0189] This reversal approach is advantageous because it provides the simplest possible classifier, independent of data normalization, which helps avoid overfitting and results in a very simple experimental test that is easy to implement in a clinic. The use of reversal-based biomarker pairs, independent of data normalization as described herein, as a method for identifying clinically relevant PTB biomarkers has enormous potential. Since the quantification of any single protein is subject to uncertainties caused by measurement variability, normal fluctuations, and individual relevant changes in baseline expression, the identification of biomarker pairs that can be under coordinated, systematic control should prove more robust for individual diagnosis and prognosis.

[0190] This invention provides compositions comprising isolated pairs of biomarkers selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein the biomarker pairs exhibit reversible changes in value between pregnant females at risk of preterm birth and full-term controls. In one embodiment, the composition comprises a stable isotope-labeled standard peptide (SIS peptide) derived from an alternative peptide of each of the biomarkers.

[0191] In a specific embodiment, the present invention provides a separate pair of biomarkers consisting of IBP4 and SHBG, wherein the pair shows a reversal value change between pregnant females at risk of preterm birth and full-term controls.

[0192] IBP4 is a member of the insulin-like growth factor binding protein (IBP) family, which negatively regulates insulin-like growth factors IGF1 and IGF2 (Forbes et al., Insulin-like growth factor I and II regulate the lifecycle of trophoblast in the developing human placenta. Am J Physiol, Cell Physiol. 2008; 294(6):C1313–22). IBP4 is expressed in the syncytial trophoblast (Crosley et al., IGFBP-4 and-5 are expressed in first-trimester villi and differentially regulate themigration of HTR-8 / SVneo cells. Reprod Biol Endocrinol. 2014; 12(1):123) and is the major IBP expressed in extravillous trophoblasts (Qiu et al., Significance of IGFBP-4 in the development of fetal growth restriction. J Clin Endocrinol Metab. 2012; 97(8):E1429–39). Compared with full-term pregnancies, maternal IBP4 levels are higher in early pregnancy in pregnancies complicated by fetal growth restriction and preeclampsia. (Qiu et al., ibid., 2012).

[0193] SHBG regulates the availability of unbound steroid hormones with biological activity. Hammond GL. Diverse roles for sex hormone-binding globulin in reproduction. Biol Reprod. 2011; 85(3):431–41. During pregnancy, plasma SHBG levels increase 5–10 times (Anderson DC. Sex-hormone-binding globulin. Clin Endocrinol (Oxf). 1974; 3(1):69–96) and there are signs of extrahepatic expression, including expression in placental trophoblast cells. (Larrea et al. Evidence that human placenta is a site of sexhormone-binding globulin gene expression. J Steroid Biochem Mol Biol. 1993; 46(4):497–505). Physiologically, SHBG levels are negatively correlated with triglyceride, insulin, and BMI. (Simó et al., Novel insights in SHBG regulation and clinical implications. Trends Endocrinol Metab. 2015; 26(7):376–83). The effect of BMI on SHBG levels can partially explain the predictive performance of improvements categorized by BMI.

[0194] Intraamnional infection and inflammation have been associated with PTB, as it has elevated levels of pro-inflammatory cytokines, including TNF-α and IL-1-β. (Mendelson CR. Minireview: fetal-maternal hormonal signaling inpregnancy and labor. Mol Endocrinol. 2009; 23(7):947–54; Gomez-Lopez et al., Immunecells in term and preterm labor. Cell Mol Immunol. 2014; 11(6):571–81). Hepatic SHBG transcription is inhibited by IL1-β and NF-κB-mediated TNF-α signaling (Simó et al., Novel insights in SHBG regulation and clinical implications. Trends Endocrinol Metab. 2015; 26(7):376–83), a pathway involved in the initiation of normal and abnormal labor (Lindstm TM, Bennett PR. The role of nuclear factor kappa B in human labour. Reproduction. 2005; 130(5):569–81). Lower SHBG levels in women destined for spTB can be a result of infection and / or inflammation. Therefore, SHBG may be important for controlling the effects of androgens and estrogens in the placental-fetal unit in response to upstream inflammatory signals.

[0195] In one embodiment, the present invention provides a composition comprising alternative peptide pairs corresponding to biomarker pairs selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein the biomarker pairs exhibit reversible value changes between pregnant females at risk of preterm birth and full-term controls.

[0196] In other embodiments, the present invention provides a set of at least two pairs of biomarkers selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein each pair shows a change in reversible value between pregnant females at risk of preterm birth and full-term controls.

[0197] In other embodiments, the present invention provides a set of at least two pairs of alternative peptides, each pair of alternative peptides corresponding to a pair of biomarkers selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, wherein each pair shows a change in reversible value between pregnant females at risk of preterm birth and full-term controls.

[0198] In one embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal value of at least one pair of biomarkers selected from a biological sample obtained from the pregnant female animal, the biomarker pair being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS to determine the probability of preterm birth in a pregnant female animal.

[0199] In another embodiment, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring changes in reversal values ​​of at least two pairs of biomarkers selected from a biological sample obtained from the pregnant female animal, the biomarker pairs being IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS to determine the probability of preterm birth in the pregnant female animal. In some embodiments, the reversal value shows a change in reversal value between the pregnant female animal and a full-term control and indicates the probability of preterm birth in the pregnant female animal. In some embodiments, the measurement step includes measuring alternative peptides of the biomarkers in a biological sample obtained from the pregnant female animal.

[0200] In one embodiment, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring the reversal values ​​of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal, the biomarker pairs being selected from Tables 1 to 77 of the pregnant female animal. Figures 1 to 111 Use any of the biomarkers listed in the literature to determine the probability of preterm birth in pregnant female animals.

[0201] In other embodiments, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal value of at least one pair of biomarkers in a biological sample obtained from the pregnant female animal, the biomarker pair being selected from the biomarker pairs specified in Tables 27 to 59, 61 to 72, 76 to 77 of the pregnant female animal to determine the probability of preterm birth in the pregnant female animal.

[0202] In other embodiments, the present invention provides a method for determining the probability of preterm birth in a pregnant female animal, the method comprising measuring the reversal values ​​of at least one pair of biomarkers selected from biological samples obtained from the pregnant female animal, the biomarker pair being selected from the biomarkers listed in Table 26 of the pregnant female animal to determine the probability of preterm birth in the pregnant female animal.

[0203] In another embodiment, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring changes in the reversal values ​​of at least two pairs of biomarkers selected from biological samples obtained from the pregnant female animal, the biomarkers being selected from Tables 1 to 77 of the pregnant female animal. Figures 1 to 111 The biomarkers specified in any one of these provisions are used to determine the probability of preterm birth in pregnant female animals. In some embodiments, the reversal value shows a change in reversal value between pregnant female animals and full-term controls and indicates the probability of preterm birth in pregnant female animals. In some embodiments, the measurement step includes measuring alternative peptides of the biomarkers in a biological sample obtained from the pregnant female animal.

[0204] In another embodiment, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring changes in reversal values ​​of at least two pairs of biomarkers selected from biological samples obtained from the pregnant female animal, the biomarker pairs being selected from biomarker pairs specified in Tables 27 to 59, 61 to 72, and 76 to 77 of pregnant female animals to determine the probability of preterm birth in pregnant female animals. In some embodiments, the reversal value shows a change in reversal value between the pregnant female animal and a full-term control and indicates the probability of preterm birth in the pregnant female animal. In some embodiments, the measurement step includes measuring alternative peptides of the biomarkers in biological samples obtained from the pregnant female animal.

[0205] In another embodiment, the present invention provides a method for determining the probability of preterm birth in pregnant female animals, the method comprising measuring changes in reversal values ​​of at least two pairs of biomarkers selected from those specified in Table 26 of pregnant female animals to determine the probability of preterm birth in pregnant female animals. In some embodiments, the reversal values ​​show a change in reversal values ​​between pregnant female animals and full-term controls and indicate the probability of preterm birth in pregnant female animals. In some embodiments, the measurement step includes measuring alternative peptides of the biomarkers in the biological sample obtained from the pregnant female animal.

[0206] Regarding methods for predicting the time of labor, "labor" should be understood as delivery following the onset of spontaneous labor, with or without rupture of the amniotic sac.

[0207] Although described and illustrated with reference to methods for determining the probability of preterm birth in pregnant female animals, the present invention discloses similarly applicable methods for predicting gestational age at birth (GAB), predicting full-term birth, determining the probability of full-term birth in pregnant female animals, and predicting time of parturition (TTB) in pregnant female animals. It will be apparent to those skilled in the art that each of the above methods has specific and numerous applications and benefits for the sake of maternal and fetal health.

[0208] Furthermore, although described and illustrated with reference to methods for determining the probability of preterm birth in pregnant female animals, this invention is similarly applicable to predicting abnormal glucose tests, gestational diabetes, hypertension, preeclampsia, intrauterine growth restriction, stillbirth, fetal growth restriction, HELLP syndrome, oligohydramnios, chorioamnionitis, placenta previa, placental hyperplasia, rupture, placental abruption, placental hemorrhage, premature rupture of membranes, preterm birth, cervical malformation, post-term pregnancy, gallstones, uterine overdistension, and stress. As described in more detail below, based on the condition, such as (e.g.) preeclampsia or gestational diabetes, the classifier described herein is sensitive to medically specified components of PTB.

[0209] In some embodiments, this invention discloses the use of biomarkers, biomarker pairs, and / or reversals of ITIH4 / CSH, as exemplified herein, which are strong predictors of time to labor (TTB). Figure 10 TTB is defined as the difference between GABD and gestational age at birth (GAB). This finding enables prediction of TTB or GAB, either alone or in mathematical combinations of these analytes. In accordance with the method of the invention, cases lacking differences relative to controls, but analytes indicating changes in analyte intensity during pregnancy are useful in the gestational clock. The calibration of multiple analytes that may not be a diagnosis of preterm birth or other conditions can be used to determine gestational age. This gestational clock is valuable for confirmation of the date by another measurement (e.g., the date of last menstrual period and / or ultrasound date), or individually for subsequent and more accurate prediction (e.g., sPTB, GAB, or TTB). These analytes are also referred to herein as “clock proteins,” which can be used to determine gestational age in the absence of other date-determining methods or in combination with other date-determining methods. Table 60 provides a list of clock proteins useful in the gestational clock for predicting TTB and GAB according to the invention.

[0210] In other embodiments, the method for determining the probability of preterm birth in pregnant female animals further includes detecting measurable characteristics of one or more risk indicators associated with preterm birth. In other embodiments, the risk indicators may include candidates such as previous low birth weight or preterm delivery, multiple spontaneous abortions in the second trimester, previous induced abortion in the first trimester, family history and factors existing between generations, infertility, nulliparity, conception, primiparity, multiparity, placental abnormalities, cervical and uterine abnormalities, gestational hemorrhage, intrauterine growth restriction, intrauterine diethylstilbestrol exposure, multiple pregnancies, infant sex (infantsex), short stature, low pre-pregnancy weight, low or high body mass index, diabetes, hypertension, and genitourinary infections.

[0211] "Measurable characteristic" is any property, characteristic, or aspect that can be determined and is associated with the probability of preterm birth in a subject. The term also includes any property, characteristic, or aspect that can be determined and is associated with the prediction of GAB, the prediction of term birth, or the prediction of delivery time in pregnant female animals. For biomarkers, such measurable characteristic may include, for example, the presence, absence, or concentration of the biomarker or a fragment thereof in a biological sample; altered structure, such as the presence or amount of post-translational modifications, such as oxidation at one or more positions on the amino acid sequence of the biomarker; or, for example, the presence of an altered conformation compared to the conformation of the biomarker in a term-birth control subject; and / or the presence, amount, or altered structure of a biomarker that is part of a spectrum of more than one biomarker.

[0212] In addition to biomarkers, measurable characteristics may also include risk indicators, such as maternal characteristics, age, race, ethnicity, medical history, previous pregnancy history, and marital / reproductive history. For risk indicators, measurable characteristics may include, for example, a history of low birth weight or preterm delivery, multiple spontaneous abortions in the second trimester, a previous induced abortion in the first trimester, family history and factors existing between generations, infertility, nulliparity, placental abnormalities, cervical and uterine abnormalities, short cervical length measurement, gestational hemorrhage, intrauterine growth restriction, intrauterine diethylstilbestrol exposure, multiple pregnancies, infant sex (infantsex), short stature, low pre-pregnancy weight / low body mass index, diabetes, hypertension, genitourinary tract infections, hypothyroidism, asthma, low education level, smoking, drug use, and alcohol consumption.

[0213] In some embodiments, the method of the present invention includes the calculation of body mass index (BMI).

[0214] In some embodiments, the disclosed methods for determining the probability of preterm birth encompass the use of mass spectrometry, capture reagents, or combinations thereof to detect and / or quantify one or more biomarkers.

[0215] In other embodiments, the disclosed method for determining the probability of preterm birth in pregnant female animals covers the initial step of providing biological samples from pregnant female animals.

[0216] In some embodiments, the disclosed methods for determining the probability of preterm birth in pregnant female animals include communicating said probability to a healthcare provider. The disclosed methods for predicting preterm birth (GAB), predicting full-term birth, determining the probability of full-term birth in pregnant female animals, and predicting the time of delivery in pregnant female animals similarly encompass communicating said probability to a healthcare provider. As described and illustrated above with reference to determining the probability of preterm birth in pregnant female animals, all embodiments described throughout this invention disclosure are similarly applicable to methods for predicting GAB, predicting full-term birth, determining the probability of full-term birth in pregnant female animals, and predicting the time of delivery in pregnant female animals. Specifically, in this application, biomarkers and groups explicitly listed with reference to methods for preterm birth can also be used in methods for predicting GAB, predicting full-term birth, determining the probability of full-term birth in pregnant female animals, and predicting the time of delivery in pregnant female animals. It will be apparent to those skilled in the art that each of the above methods has specific and numerous applications and benefits for the sake of maternal-fetal health.

[0217] In other embodiments, communication informs the pregnant female animal of subsequent treatment decisions. In some embodiments, methods for determining the probability of preterm birth in pregnant female animals include representing the probability as other features of a risk score.

[0218] In the methods disclosed herein, determining the probability of preterm birth in pregnant female animals encompasses an initial step of forming a probability / risk index by measuring the ratio of isolated biomarkers selected from preterm pregnancy groups and full-term pregnancy groups with known gestational age at birth. For an individual pregnancy, determining the probability of preterm birth in a pregnant female animal includes measuring the ratio of isolated biomarkers using the same measurement methods used in the initial step of generating the probability / risk index, and comparing the measured ratio with the risk index to derive an individualized risk for the individual pregnancy. In one embodiment, a preterm birth risk index is formed by measuring the ratio of IBP4 / SHBG in preterm and full-term pregnancy groups, where gestational age at birth is recorded. Then, in clinical practice, the IBP4 / SHBG ratio measured in a single pregnancy is compared with an index that yields the risk of preterm birth using the same isolation and measurement techniques to determine if IBP4 / SHBG falls within the index group.

[0219] As used herein, the term "risk score" refers to a score that can be assigned based on a comparison between the amount or reversal value of one or more biomarkers in a biological sample obtained from a pregnant female animal and a standard or reference score representing the average amount of one or more biomarkers calculated from a random mix of biological samples obtained from a pregnant female animal. In some embodiments, the risk score may be expressed as the logarithm of the reversal value, i.e., the ratio of the relative intensities of the individual biomarkers. Those skilled in the art will understand that risk scores can be represented based on various data transformations, and that the risk score is expressed as a ratio itself. Furthermore, by paying particular attention to reversal pairs, those skilled in the art will understand that any ratio is equally informative if biomarkers are switched in the numerator and denominator or if relevant data transformations (e.g., subtraction) are applied. Since biomarker levels may not be constant throughout pregnancy, it is necessary to obtain a standard or reference score corresponding to the gestation time point of the pregnant female animal at the time of sample collection. Standard or reference scores can be predetermined and predictive models built so that the comparison is indirect rather than actually performed each time a probability is determined for a subject. The risk score can be a standard (e.g., a numerical value) or a threshold (e.g., a line on a graph). The risk score is related to the upper or lower deviation of the average amount of one or more biomarkers calculated from a randomized mixture of biological samples obtained from pregnant female animals. In some embodiments, if the risk score is greater than a standard or reference risk score, the pregnant female animal may have an increased likelihood of preterm birth. In some embodiments, the magnitude of the risk score of the pregnant female animal, or the amount by which it exceeds the reference risk score, may serve as an indication of or be associated with the risk level of the pregnant female animal.

[0220] As illustrated in this article, PreTRM TM The classifier was defined as the natural logarithm of the SIS normalized intensity of IBP4 peptide conversion (QCHPALDGQR_394.5_475.2(SEQ ID NO:2)) and SHBG peptide conversion (IALGGLLFPASNLR_481.3_657.4(SEQ ID NO:18)). Score = ln(P) 1 n / P 2 n ), where P 1 n and P 2 n These represent the SIS normalized peak areas for IBP4 and SHBG conversions, respectively. SIS normalization is defined as the ratio of the intrinsic peak area to the corresponding SIS peak area: for example, P 1 n =P 1 e / P 1 SIS, where P 1 e = Peak area of ​​IBP4 endogenous transformation, P 1 SIS = Peak area of ​​IBP4 SIS conversion. According to PreTRM TM The identified link between the score distribution and the corresponding incidence rate modulated positive predictive value can be used to assign the sPTB probability to unknown subjects based on the determination of their scores. Figure 25 This relationship or connection is shown in the data, and it links laboratory measurements with clinical predictions.

[0221] Despite PreTRM TM The classifier is defined as the natural log of the SIS normalized intensity of IBP4 peptide conversion (QCHPALDGQR_394.5_475.2 (SEQ ID NO:2)) and SHBG peptide conversion (IALGGLLFPASNLR_481.3_657.4 (SEQ ID NO:18)), but the invention also includes classifiers comprising multiple reversals. Improved performance can be achieved by constructing predictors formed from more than one reversal. In other embodiments, the method of the invention thus includes multiple reversals that have strong predictive performance for, for example, individual GABD windows, preterm premature rupture of membranes (PPROM) versus preterm labor (PTL) without PPROM, fetal sex, and primiparous women versus multiparous women. This embodiment is illustrated in Examples 10 and Table 61 with examples of reversals that produce strong predictive performance in early (e.g., 17–19 weeks) or late (e.g., 19–21 weeks) gestational ages. As illustrated in the example, the performance of a predictor formed by a combination of multiple reversals (SumLog) was evaluated over the entire blood sampling range, and the predictor score was derived from the sum of the log values ​​of each reversal (SumLog). Those skilled in the art may choose other models (e.g., logistic regression) to construct predictors formed by more than one reversal.

[0222] The method of this invention also includes a classifier containing an indicator variable, which selects a reversal or reversal subgroup based on known clinical factors, such as the date of blood draw, fetal sex, pregnancy, and any other known patient characteristics and / or risk factors described in this application. This embodiment is illustrated in Example 10, Tables 61-64, which exemplify reversal performance (17-21 weeks) for the two different phenotypes of sPTB, PPROM and PTL. Similarly, this embodiment is illustrated in Example 10, Tables 76 and 77, and Figures 108 and 109, which exemplify reversal performance (19-21 weeks) for the two different phenotypes of sPTB, premature rupture of membranes (PPROM) and preterm birth (PTL) without PPROM. The method of this invention therefore includes the selection of reversals to establish independent predictors for PPROM and PTL, or to maximize overall performance through a combination of more than one reversal among the single predictors described above. This embodiment is further illustrated in Example 10, Tables 65-68, which illustrate reversal performance (17-21 weeks) for two different types of sPTB, primiparous and multiparous women. This embodiment is further illustrated in Example 10, Tables 69-72 and Figure 106, which illustrate reversal performance (17-21 weeks) for two different types of sPTB based on fetal sex. Although examples of PPROM and PTL, pregnancy and fetal sex are given, the method described in this invention includes a classifier containing an indicator variable that selects a reversal or reversal subgroup based on GABD or any known clinical factors / risk factors described herein or known to those skilled in the art. As an alternative to a classifier containing an indicator variable, this invention also provides a separate classifier suitable for a pregnant woman subgroup based on GABD or any known clinical factors / risk factors described herein or known to those skilled in the art. For example, this embodiment covers consecutive and / or overlapping time windows of GABD using a separate classifier based on the best implementation of reversal for each time window.

[0223] As illustrated in the examples in this article, it can be achieved by using a weight greater than 22 and equal to or less than 37 kg / m³. 2BMI classification is used to improve the predictive performance of the claimed method. Therefore, in some embodiments, the method described in this invention can be practiced using samples obtained from pregnant female animals with a specified BMI. Briefly, BMI is an individual's weight (kg) divided by the square of their height (m). BMI does not directly measure body fat, but studies have shown that BMI is associated with more direct measurements of body fat obtained from skinfold thickness measurements, bioelectrical impedance analysis, densitometric analysis (underwater weighing), dual-energy X-ray absorptiometry (DXA), and other methods. Furthermore, BMI appears to be as strongly associated with a variety of metabolic and disease outcomes as these more direct measurements of body fat. Generally, individuals with a BMI below 18.5 are considered below normal weight, individuals with a BMI equal to or greater than 18.5 to 24.9 are considered normal weight, individuals with a BMI equal to or greater than 25.0 to 29.9 are considered overweight, and individuals with a BMI equal to or greater than 30.0 are considered obese. In some implementations, the predictive performance of the claimed method can be improved by BMI classifications equal to or greater than 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30. In other implementations, the predictive performance of the claimed method can be improved by BMI classifications equal to or less than 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30.

[0224] In the context of this invention, the term "biological sample" encompasses any sample derived from a pregnant female animal and containing one or more biomarkers disclosed herein. Suitable samples in the context of this invention include, for example, blood, plasma, serum, amniotic fluid, vaginal secretions, saliva, and urine. In some embodiments, the biological sample is selected from whole blood, plasma, and serum. In a particular embodiment, the biological sample is serum. As those skilled in the art will understand, a biological sample may include any portion or component of blood, without limitation, T cells, monocytes, neutrophils, erythrocytes, platelets, and microcapsules, such as exosomes and exosome-like microcapsules. In a particular embodiment, the biological sample is serum.

[0225] As used herein, the term "preterm birth" refers to delivery or labor at a gestational age of less than 37 full weeks. Other commonly used subclasses of preterm birth have been established and are denoted as moderate preterm birth (delivery between 33 and 36 weeks of gestation), excessive preterm birth (delivery before 33 weeks of gestation), and severe preterm birth (delivery before 28 weeks of gestation). Those skilled in the art will understand that the cutoffs for representing preterm and term birth, as well as for representing preterm birth subclasses, can be adjusted in practice with regard to the methods disclosed herein, for example, to maximize specific health benefits. In various embodiments of the invention, the cutoff time for describing preterm birth includes, for example, delivery at ≤37 weeks of gestation, delivery at ≤36 weeks of gestation, delivery at ≤35 weeks of gestation, delivery at ≤34 weeks of gestation, delivery at ≤33 weeks of gestation, delivery at ≤32 weeks of gestation, delivery at ≤30 weeks of gestation, delivery at ≤29 weeks of gestation, delivery at ≤28 weeks of gestation, delivery at ≤27 weeks of gestation, delivery at ≤26 weeks of gestation, delivery at ≤25 weeks of gestation, delivery at ≤24 weeks of gestation, delivery at ≤23 weeks of gestation, or delivery at ≤22 weeks of gestation. In some embodiments, the cutoff time for describing preterm birth is ≤35 weeks of gestation. It should be further understood that these adjustments are within the skill set of those skilled in the art and are covered within the scope of the invention disclosed herein. Gestational age is representative of the degree of fetal development and fetal readiness for delivery. Gestational age is generally defined as the length of time between the last normal menstrual period and the date of delivery. However, obstetric measurements and ultrasound estimation can also aid in estimating gestational age. Preterm birth is generally divided into two distinct subgroups. One is spontaneous preterm birth, which occurs spontaneously following premature delivery or premature rupture of membranes, without consideration of subsequent induction or cesarean section. The second is medically indicated preterm births, which are those preterm births that occur after induction or cesarean section, as determined by the woman's caregivers to pose one or more conditions threatening the health or life of the mother and / or fetus. In some embodiments, the methods disclosed herein involve determining the probability of spontaneous or medically indicated preterm birth. In some embodiments, the methods disclosed herein involve determining the probability of spontaneous preterm birth. In other embodiments, the methods disclosed herein involve medically indicated preterm birth. In other embodiments, the methods disclosed herein involve predicting gestational age at birth.

[0226] As used herein, the terms “estimated gestational age” or “estimated GA” refer to a GA determined based on the date of the last normal menstrual period and other obstetric measurements, ultrasound estimates, or other clinical parameters (which include, without limitation, those described in the preceding paragraph). Conversely, the terms “predicted gestational age at birth” or “predicted GAB” refer to a GAB determined based on the methods described herein. As used herein, “full-term delivery” means delivery at a gestational age of 37 full weeks or greater.

[0227] In some embodiments, the pregnant female animal is between 17 and 28 weeks of gestation at the time of biological sample collection, also known as GABD (gestational age at blood draw). In other embodiments, the pregnant female animal is between 16 and 29 weeks of gestation, between 17 and 28 weeks of gestation, between 18 and 27 weeks of gestation, between 19 and 26 weeks of gestation, between 20 and 25 weeks of gestation, between 21 and 24 weeks of gestation, or between 22 and 23 weeks of gestation. In other embodiments, the pregnant female animal is approximately between 17 and 22 weeks of gestation, approximately between 16 and 22 weeks of gestation, approximately between 22 and 25 weeks of gestation, approximately between 13 and 25 weeks of gestation, approximately between 26 and 28 weeks of gestation, or approximately between 26 and 29 weeks of gestation. Therefore, the gestational age of pregnant female animals at the time of biological sample collection can be 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 weeks. In specific embodiments, biological samples are collected between 19 and 21 weeks of gestational age. In specific embodiments, biological samples are collected between 19 and 22 weeks of gestational age. In specific embodiments, biological samples are collected between 19 and 21 weeks of gestational age. In specific embodiments, biological samples are collected between 19 and 22 weeks of gestational age. In specific embodiments, biological samples are collected at 18 weeks of gestational age. In other embodiments, the highest implementation reversal of consecutive or overlapping time windows can be combined in a single classifier to predict the sPTB probability for a wider gestational age window at the time of blood collection.

[0228] As used herein, the term "amount" or "level" refers to the amount of a biomarker that is detectable or measurable in a biological sample and / or control. The amount of a biomarker can be, for example, the amount of a peptide, a nucleic acid, or a fragment or substitute. Alternatively, the term may include combinations thereof. The term "amount" or "level" of a biomarker is a measurable characteristic of that biomarker.

[0229] The present invention also provides a method for detecting pairs of biomarkers isolated from pregnant female animals, said biomarker pairs being selected from Tables 1 to 77 and Figures 1 to 111 The method comprises the steps of: a. obtaining a biological sample from the pregnant female animal; b. detecting the presence of the isolated biomarker pair in the biological sample by contacting the biological sample with a first capture reagent that specifically binds to a first member of the pair and a second capture reagent that specifically binds to a second member of the pair; and detecting the binding between the first biomarker and the first capture reagent of the pair and between the second member of the pair and the second capture reagent.

[0230] The present invention also provides a method for detecting isolated biomarker pairs in pregnant female animals, said biomarker pairs being selected from those specified in Tables 27 to 59, 61 to 72, 76 and 77, said method comprising the steps of: a. obtaining a biological sample from said pregnant female animal; b. detecting the presence of said isolated biomarker pairs in said biological sample by contacting said biological sample with a first capture reagent that specifically binds to a first member of said pair and a second capture reagent that specifically binds to a second member of said pair; and detecting the binding between the first biomarker and the first capture reagent of said pair and between the second member of said pair and the second capture reagent.

[0231] The present invention also provides a method for detecting isolated biomarker pairs in pregnant female animals, wherein the isolated biomarker pairs are selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS. The method comprises the following steps: a. obtaining a biological sample from the pregnant female animal; b. detecting the presence of the isolated biomarker pair in the biological sample by contacting the biological sample with a first capture reagent that specifically binds to a first member of the pair and a second capture reagent that specifically binds to a second member of the pair; and detecting the binding between the first biomarker and the first capture reagent of the pair, and between the second member of the pair and the second capture reagent. In one embodiment, the present invention provides a method for detecting IBP4 and SHBG in pregnant female animals, the method comprising the steps of: a. obtaining a biological sample from the pregnant female animal; b. detecting the presence of IBP4 and SHBG in the biological sample by contacting the biological sample with a capture reagent that specifically binds IBP4 and a capture reagent that specifically binds SHBG; and c. detecting the binding between IBP4 and the capture reagent and between SHBG and the capture reagent. In one embodiment, the method includes measuring the reversal value of the biomarker pair. In other embodiments, the presence of a change in the reversal value between the pregnant female animal and a full-term control indicates the probability of preterm birth in the pregnant female animal. In one embodiment, the sample is obtained between 19 and 21 weeks of gestational age. In other embodiments, the capture reagent is selected from antibodies, antibody fragments, nucleic acid-based protein binding reagents, small molecules, or variants thereof. In other embodiments, the method is performed by an assay selected from enzyme immunoassay (EIA), enzyme-linked immunosorbent assay (ELISA), and radioimmunoassay (RIA).

[0232] The present invention also provides a method for detecting isolated biomarker pairs in pregnant female animals, said isolated biomarker pairs being selected from IBP4 / SHBG, VTNC / VTDB, VTNC / SHBG, CATD / SHBG, PSG2 / ITIH4, CHL1 / ITIH4, PSG2 / C1QB, PSG2 / FBLN3, HPX / IBP4, and HPX / PTGDS, said method comprising the steps of: a. obtaining a biological sample from said pregnant female animal; and b. detecting the presence of said isolated biomarker pairs in said biological sample, which includes performing a proteomics workflow consisting of mass spectrometry quantification on said sample.

[0233] In one embodiment, the present invention provides a method for detecting IBP4 and SHBG in pregnant female animals, the method comprising the steps of: a. obtaining a biological sample from the pregnant female animal; and b. detecting the presence of the isolated biomarker pair in the biological sample, comprising performing a proteomics workflow consisting of mass spectrometry quantification on the sample.

[0234] A typical proteomics workflow includes one or more of the following steps: Melting a serum sample and consuming the 14 most abundant proteins via immunoaffinity chromatography. The consumed serum is then hydrolyzed with a protease, such as trypsin, to obtain peptides. Subsequently, a mixture of SIS peptides is added to the hydrolysate, followed by desalting and LC-MS / MS using a triple quadrupole apparatus operating in MRM mode. Response ratios are derived based on the area ratio of the endogenous peptide peak to the corresponding SIS peptide counterpart peak. Those skilled in the art will understand that other types of MS, such as (e.g.) MALDI-TOF or ESI-TOF, can be used in the methods described herein. Furthermore, those skilled in the art can modify the proteomics workflow, for example, by selecting specific reagents (such as proteases) or omitting or changing the order of certain steps; for instance, immunoremoval may be unnecessary, SIS peptides may be added earlier or later, and stable isotope-labeled proteins may be used instead of peptides as standards.

[0235] The presence or absence (e.g., a reading of presence relative to absence; or a detectable amount relative to an undetectable amount) and / or amount (e.g., a reading of absolute or relative amount, such as absolute or relative concentration) of a biomarker, peptide, polypeptide, protein, and / or fragment thereof, and optionally one or more other biomarkers or fragments thereof, in a sample can be measured using any existing, available, or conventional separation, detection, and quantification method described herein. In some embodiments, the detection and / or quantification of one or more biomarkers includes an assay using a capture reagent. In other embodiments, the capture reagent is an antibody, antibody fragment, nucleic acid-based protein-binding reagent, small molecule, or a variant thereof. In other embodiments, the assay is selected from enzyme immunoassay (EIA), enzyme-linked immunosorbent assay (ELISA), and radioimmunoassay (RIA). In some embodiments, the detection and / or quantification of one or more biomarkers further includes mass spectrometry (MS). In other embodiments, the mass spectrometry is co-immunoprecipitation-mass spectrometry (co-IP MS), where co-immunoprecipitation is a technique suitable for the separation of intact protein complexes, followed by mass spectrometry.

[0236] As used herein, the term "mass spectrometry" refers to a device capable of volatilizing / ionizing analytes to form gaseous ions and determining their absolute or relative molecular weights. Suitable volatilization / ionization methods include matrix-assisted laser desorption / ionization (MALDI), electrospraying, laser / optical, thermal, electrical, nebulization / spraying, etc., or combinations thereof. Suitable forms of mass spectrometry include (but are not limited to) ion trap instruments, quadrupole instruments, electrostatic and magnetic sector field instruments, time-of-flight instruments, time-of-flight tandem mass spectrometers (TOF MS / MS), Fourier transform mass spectrometers, Orbitraps, and hybrid instruments consisting of different combinations of these types of mass spectrometers. In turn, these instruments can be coupled with a variety of other instruments that separate samples (e.g., liquid chromatography or solid-phase adsorption techniques based on chemical or biological properties) and ionize the samples for introduction into the mass spectrometer, including matrix-assisted laser desorption / ionization (MALDI), electrospraying, or nanospray ionization (ESI), or combinations thereof.

[0237] Generally, any mass spectrometry (MS) technique can be used in the methods disclosed herein, which can provide precise information on peptide mass and preferably also provide precise information on fragments and / or (partial) amino acid sequences of the selected peptide (e.g., in tandem mass spectrometry, MS / MS; or in post-source decay, TOF MS). Suitable peptide MS and MS / MS techniques and systems are well known in themselves (see, for example, Methods in Molecular Biology, Vol. 146: "Mass Spectrometry of Proteins and Peptides", edited by Chapman, Humana Press 2000; Biemann 1990. MethodsEnzymol 193:455-79; or Methods in Enzymology, Vol. 402: "Biological Mass Spectrometry", edited by Burlingame, Academic Press 2005) and can be used in practicing the methods disclosed herein. Thus, in some embodiments, the disclosed methods include performing quantitative MS to measure one or more biomarkers. These quantitative methods can be performed automatically (Villanueva et al., Nature Protocols (2006) 1(2): 880-891) or semi-automatically. In specific embodiments, MS can be operatively connected to a liquid chromatography apparatus (LC-MS / MS or LC-MS) or a gas chromatography apparatus (GC-MS or GC-MS / MS). Other methods useful in the context include isotope-coded affinity tag (ICAT), tandem mass spectrometry tag (TMT), or stable isotope labeling (SILAC) of amino acids in cell cultures followed by chromatography and MS / MS.

[0238] As used herein, the terms “multiple reaction monitoring (MRM)” or “selective reaction monitoring (SRM)” refer to MS-based quantitative methods particularly useful for quantifying low-abundance analytes. In SRM experiments, a predetermined precursor ion and one or more fragments thereof are selected via two mass filters of a triple quadrupole and precisely quantified over time. Multiple SRM precursor and fragment ion pairs can be measured on the same chromatographic timescale within the same experiment by rapidly switching between different precursor / fragment pairs to perform MRM experiments. A combination of a series of transitions (precursor / fragment ion pairs) with the retention times of the target analyte (e.g., peptides or small molecules, such as chemical entities, steroids, hormones) can constitute a deterministic assay. A large number of analytes can be quantified in a single LC-MS experiment. The terms “arranged” or “dynamic” associated with MRM or SRM refer to variations in the assay in which transitions of a specific analyte are obtained only within a time window around the expected retention time, thereby significantly increasing the number of analytes that can be detected and quantified in a single LC-MS experiment and contributing to the selectivity of the assay, as retention time is a property dependent on the physical properties of the analyte. Individual analytes can also be monitored with more than one transition. Finally, the assay can include standards corresponding to the analyte of interest (e.g., the same amino acid sequence), but differing in that they contain stable isotopes. Stable isotope standards (SIS) can be introduced into the assay at a precise level and used to quantify the corresponding unknown analyte. Additional levels of specificity are facilitated by the co-elution of the unknown analyte with its corresponding SIS and their transition properties (e.g., the similarity between the ratio of two transition levels of the unknown and the ratio of two transitions of its corresponding SIS).

[0239] Mass spectrometry assays, instruments, and systems suitable for biomarker peptide analysis can include, without limitation, matrix-assisted laser desorption / ionization time-of-flight (MALDI-TOF) MS; MALDI-TOF post-source decay (PSD); MALDI TOF / TOF; surface-enhanced laser desorption / ionization time-of-flight (SELDI-TOF) MS; electrospray ionization mass spectrometry (ESI-MS); ESI-MS / MS; ESI-MS / (MS)n (where n is a positive integer); ESI 3D or linear (2D) ion trap MS; ESI triple quadrupole MS; ESI quadrupole orthogonal TOF (Q-TOF); ESI Fourier transform MS system; silicon-on-silicon desorption / ionization (DIOS); secondary ion mass spectrometry (SIMS); atmospheric pressure chemical ionization mass spectrometry (APCI-MS); APCI-MS / MS; APCI-(MS)n; ion mobility spectrometry (IMS); inductively coupled plasma mass spectrometry (ICP-MS); atmospheric pressure photochemical ionization mass spectrometry (APPI-MS); APPI-MS / MS; and APPI-(MS)n. Peptide ion fragmentation in tandem MS (MS / MS) arrangements can be achieved using methods established in the art, such as collision-induced dissociation (CID). As described herein, the detection and quantification of biomarkers by mass spectrometry can include multiple reaction monitoring (MRM), as described by Kuhn et al. Proteomics 4:1175-86 (2004). During LC-MS / MS analysis, scheduled multiple reaction monitoring (Scheduled MRM) acquisition improved the sensitivity and accuracy of peptide quantification. Anderson and Hunter, Molecular and Cellular Proteomics 5(4):573(2006). As described herein, mass spectrometry-based assays can be advantageously combined with upstream peptide or protein separation or fractionation methods, such as (e.g.) with chromatography and other methods described below. As further described herein, shotgun quantitative proteomics can be combined with SRM / MRM-based assays for high-throughput identification and verification of prognostic biomarkers for preterm birth.

[0240] Those skilled in the art will understand that various methods can be used to determine the amount of a biomarker, including mass spectrometry such as MS / MS, LC-MS / MS, multiple reaction monitoring (MRM) or SRM, and product ion monitoring (PIM), and also antibody-based methods such as immunoassays, such as Western blotting, enzyme-linked immunosorbent assay (ELISA), immunoprecipitation, immunohistochemistry, immunofluorescence, radioimmunoassay, dot blot, and FACS. Therefore, in some embodiments, determining the level of at least one biomarker includes using immunoassays and / or mass spectrometry. In other embodiments, the mass spectrometry method is selected from MS, MS / MS, LC-MS / MS, SRM, PIM, and other such methods known in the art. In other embodiments, LC-MS / MS further includes 1D LC-MS / MS, 2D LC-MS / MS, or 3D LC-MS / MS. Immunoassay techniques and procedures are generally known to those skilled in the art (Price and Newman, Principles and Practice of Immunoassay , 2nd edition, Grove's Dictionaries, 1997; and Gosling, Immunoassays:A Practical Approach (Oxford University Press, 2000). A variety of immunoassay techniques can be used, including competitive and non-competitive immunoassays (Self et al., ). Curr. Opin.Biotechnol .,7:60-65(1996)).

[0241] In other embodiments, the immunoassay is selected from immunoblotting, ELISA, immunoprecipitation, immunohistochemistry, immunofluorescence, radioimmunoassay (RIA), dot blot, and FACS. In some embodiments, the immunoassay is an ELISA. In other embodiments, the ELISA is a direct ELISA (enzyme-linked immunosorbent assay), indirect ELISA, sandwich ELISA, competitive ELISA, multichannel ELISA, ELISPOT technology, and other similar techniques known in the art. The principles of these immunoassay methods are known in the art, for example, by John R. Crowther. The ELISA Guidebook , 1st edition, HumanaPress 2000, ISBN 0896037282. Typically, antibodies are used for ELISA, but they can be performed using any capture reagent that specifically binds to and can detect one or more biomarkers of the present invention. Multichannel ELISA enables the simultaneous detection of two or more analytes within a single compartment (e.g., the wells of a microplate), which is typically performed at multiple array locations (Nielsen and Geierstanger 2004). J Immunol Methods 290:107-20 (2004) and Ling et al., 2007. Expert Rev Mol Diagn7:87-98 (2007)).

[0242] In some embodiments, radioimmunoassay (RIA) can be used in the methods described in this invention to detect one or more biomarkers. RIA is a competition-based assay well known in the art and includes mixing a known amount of a radiolabeled (e.g., 125I or 131I-labeled) target analyte with a specific antibody against said analyte, then adding an unlabeled analyte from the sample and measuring the amount of the replaced labeled analyte (see, e.g., An Introduction to Radioimmunoassay and Related Techniques (Chard T, ed., Elsevier Science, 1995, ISBN 0444821198 for guidance).

[0243] In the methods described herein, detectable markers can be used in the assays described herein for the direct or indirect detection of biomarkers. A variety of detectable markers can be used, and the markers are selected based on the required sensitivity, ease of antibody conjugation, stability requirements, available instruments, and processing specifications. Those skilled in the art are familiar with the selection of suitable detectable markers for the assays of biomarkers based on the methods described herein. Suitable detectable markers include (but are not limited to) fluorescent dyes (e.g., fluorescein, fluorescein isothiocyanate (FITC), Oregon Green, rhodamine, Texas Red, tetrarhodimine isothiocyanate (TRITC), Cy3, Cy5, etc.), fluorescent markers (e.g., green fluorescent protein (GFP), phycoerythrin, etc.), enzymes (e.g., luciferase, horseradish peroxidase, alkaline phosphatase, etc.), nanoparticles, biotin, digoxigenin, metals, etc.

[0244] For mass spectrometry-based analysis, differential labeling using isotopic reagents, such as isotope-encoded affinity tags (ICAT) or more recent variations using isotope-inversely labeled reagents, iTRAQ (Applied Biosystems, Foster City, Calif.), or tandem mass spectrometry tags, TMT (Thermo Scientific, Rockford, IL), followed by multidimensional liquid chromatography (LC) and tandem mass spectrometry (MS / MS) analysis, can provide other methods in the practice of the methods of this invention.

[0245] Chemiluminescent assays using chemiluminescent antibodies can be used for sensitive, non-radioactive detection of protein levels. Antibodies labeled with fluorescent substances are also suitable. Examples of fluorescent substances include, without limitation, DAPI, fluorescein, Hoechst 33258, R-phycocyanin, β-phycoerythrin, R-phycoerythrin, rhodamine, Texas red, and lissamine. Indirect labels include a variety of enzymes well known in the art, such as horseradish peroxidase (HRP), alkaline phosphatase (AP), β-galactosidase, urease, etc. Detection systems using substrates suitable for horseradish peroxidase, alkaline phosphatase, and β-galactosidase are well known in the art.

[0246] For example, a spectrophotometer can be used to detect the color from a chromogenic substrate; a radiation counter can be used to detect radiation, such as a gamma particle counter. 125 Alternatively, fluorescence can be detected using a fluorometer in the presence of light of a specific wavelength, thereby analyzing the signal from the direct or indirect label. For the detection of enzyme-linked antibodies, quantitative analysis can be performed using a spectrophotometer, such as an EMAX microplate reader (Molecular Devices; Menlo Park, Calif.), according to the manufacturer's instructions. If desired, the assays used to practice this invention can be performed automatically or mechanically, and signals from multiple samples can be detected simultaneously.

[0247] In some embodiments, the methods described herein include quantifying biomarkers using mass spectrometry (MS). In other embodiments, the mass spectrometer may be liquid chromatography-mass spectrometry (LC-MS), multiple reaction monitoring (MRM), or selected reaction monitoring (SRM). In other embodiments, the MRM or SRM may also include a scheduled MRM or a scheduled SRM.

[0248] As described above, chromatography can also be used in the practice of the methods described in this invention. Chromatography includes methods for separating chemical substances and generally involves the following process: carrying a mixture of analytes by moving a liquid or gas stream (“mobile phase”) and separating them into different components due to the differential distribution of the analytes between the mobile phase and the stationary phase as they flow through or over a stationary liquid or solid phase (“stationary phase”). The stationary phase can typically be a finely broken solid, a sheet of filter material, or a liquid film on a solid surface, etc. Those skilled in the art will readily recognize that chromatography is a suitable technique for separating compounds of biological origin (e.g., amino acids, proteins, protein or peptide fragments, etc.).

[0249] Chromatography can be column chromatography (i.e., in which a stationary phase is deposited or packed into a column), preferably, liquid chromatography, and more preferably, high-performance liquid chromatography (HPLC) or ultra-high-performance liquid chromatography (UHPLC). Detailed information on chromatographic methods is well known in the art (Bidlingmeyer, Practical HPLC Methodology and Applications (John Wiley & Sons Inc., 1993). Exemplary types of chromatography include, without limitation, high-performance liquid chromatography (HPLC), unrestricted high-performance liquid chromatography (UHPLC), normal-phase HPLC (NP-HPLC), reversed-phase HPLC (RP-HPLC), ion-exchange chromatography (IEC), such as cation or anion exchange chromatography, hydrophilic interaction chromatography (HILIC), hydrophobic interaction chromatography (HIC), size exclusion chromatography (SEC), including gel filtration chromatography or gel permeation chromatography, chromatographic polymerization, affinity chromatography, such as immunoaffinity, immobilized metal affinity chromatography, etc. Chromatography (including one-dimensional, two-dimensional, or multidimensional chromatography) can be used in conjunction with other peptide analysis methods (e.g., downstream mass spectrometry as described elsewhere in this specification) as peptide separation methods.

[0250] Other peptide or polypeptide separation, identification, or quantification methods may optionally be used in conjunction with any of the analytical methods described above for measuring biomarkers in this invention disclosure. These methods include, without limitation, chemical extraction partitioning, isoelectric point focusing (IEF), including capillary isoelectric point focusing (CIEF), capillary isovelocity electrophoresis (CITP), capillary electrochromatography (CEC), etc., one-dimensional polyacrylamide gel electrophoresis (PAGE), two-dimensional polyacrylamide gel electrophoresis (2D-PAGE), capillary gel electrophoresis (CGE), capillary zone electrophoresis (CZE), micelle electrokinetic chromatography (MEKC), free-flow electrophoresis (FFE), etc.

[0251] In the context of this invention, the term "capture agent" refers to a compound that can specifically bind to a target, specifically a biomarker. This term includes antibodies, antibody fragments, nucleic acid-based protein-binding agents (e.g., aptamers, slow-dissociation-rate modified aptamers (SOMAmers)), protein capture agents, natural ligands (i.e., hormones for their receptors or vice versa), small molecules, or variants thereof.

[0252] Capture reagents can be configured to specifically bind to targets, specifically biomarkers. Capture reagents may include (but are not limited to) organic molecules such as peptides, polynucleotides, and other non-polymer molecules that can be identified by a technician. In the embodiments disclosed herein, capture reagents include any reagent that can be used to detect, purify, separate, or enrich targets, specifically biomarkers. Any affinity capture technique known in the art can be used to selectively separate and enrich / concentrate biomarkers as components of complex mixtures of biological culture media used in the disclosed methods.

[0253] Antibody capture reagents that specifically bind to biomarkers can be prepared using any suitable method known in the art. See, for example, Coligan. Current Protocols in Immunology (1991); Harlow & Lane, Antibodies:A Laboratory Manual (1988); Goding, Monoclonal Antibodies:Principles and Practice (2nd edition, 1986). An antibody capture agent can be any immunoglobulin or its derivative, whether naturally occurring or wholly or partially synthetically produced. All its derivatives that maintain specific binding ability are also included in this term. An antibody capture agent has a binding domain that is homologous or substantially homologous to the immunoglobulin binding domain and can be derived from a natural source or be wholly or partially synthetically produced. An antibody capture agent can be a monoclonal antibody or a polyclonal antibody. In some embodiments, the antibody is a single-chain antibody. Those skilled in the art will understand that antibodies can be provided in many different forms, including, for example, humanized, partially humanized, chimeric, chimeric humanized, etc. An antibody capture agent can be an antibody fragment, including (but not limited to) Fab, Fab', F(ab')2, scFv, Fv, dsFv double-chain antibodies and Fd fragments. An antibody capture agent can be produced by any means. For example, an antibody capture agent can be produced enzymatically or chemically by fragmentation of a complete antibody and / or it can be produced from genetic recombination encoding the partial antibody sequence. An antibody capture agent can contain a single-chain antibody fragment. Alternatively, antibody capture reagents may comprise, for example, multiple chains linked together by disulfide bonds; and any functional fragments derived from these molecules, wherein these fragments retain the specific binding properties of the parent antibody molecule. Due to their smaller size as functional components of the complete molecule, antibody fragments may offer advantages over complete antibodies for use in certain immunochemical techniques and experimental applications.

[0254] Suitable capture agents useful in practicing this invention also include aptamers. Aptamers are oligonucleotide sequences that can specifically bind to their targets through a unique stereo (3-D) structure. Aptamers can comprise any suitable number of nucleotides, and different aptamers can have the same or different numbers of nucleotides. Aptamers can be DNA or RNA or chemically modified nucleic acids, and can be single-stranded, double-stranded, or contain double-stranded regions, and can include a higher degree of ordered structure. Aptamers can also be photoaptamers, wherein photoreactive or chemically reactive functional groups are contained within the aptamer to covalently link it to its corresponding target. The use of aptamer capture agents can include the use of two or more aptamers that specifically bind to the same biomarker. Aptamers can include tags. Aptamers can be identified using any known method, including the SELEX (Spiritual Evolution of Ligands) method. Once identified, aptamers can be prepared or synthesized according to any known method, including chemical synthesis and enzymatic synthesis, and can be used in a variety of applications for biomarker detection. (Liu et al.) Curr Med Chem .18(27):4117-25(2011). In practice, useful capture agents also include SOMAmers (slow dissociation rate modified aptamers) known in the art to have improved dissociation rate characteristics. Brody et al., J MolBiol .422(5):595-606(2012). SOMAmer can be generated using any known method, including the SELEX method.

[0255] Those skilled in the art will understand that biomarkers can be modified prior to analysis to improve their resolution or to identify them. For example, biomarkers can be protein-digested prior to analysis. Any protease can be used. Proteases that can cleave biomarkers into a discontinuous number of fragments (such as trypsin) are particularly useful. The fragments produced by digestion serve as fingerprints of the biomarkers, thereby enabling indirect detection. This is particularly useful when biomarkers have similar molecular weights that could be confused with the biomarker in question. Additionally, protein hydrolysis fragmentation is useful for high molecular weight biomarkers because smaller biomarkers are more easily resolved by mass spectrometry. In another instance, biomarkers can be modified to improve detection resolution. For example, neuraminidase can be used to remove terminal sialic acid residues from glycoproteins to improve binding to anion adsorbents and improve detection resolution. In yet another instance, biomarkers can be further differentiated by attaching tags with specific molecular weights that bind specifically to the molecular biomarker. Optionally, after detecting these modified biomarkers, the identity of the biomarkers can be further determined by matching the physical and chemical properties of the modified biomarkers in a protein database (e.g., SwissProt).

[0256] It will also be recognized in the art that biomarkers in a sample can be captured on a substrate for detection. Conventional substrates include 96-well plates or nitrocellulose membranes coated with antibodies for subsequent exploration of protein presence. Alternatively, protein-binding molecules attached to microspheres, microparticles, microbeads, beads, or other particles can be used for biomarker capture and detection. Protein-binding molecules can be antibodies, peptides, peptide-like substances, aptamers, small molecule ligands, or other protein-binding capture agents attached to the particle surface. Each protein-binding molecule can include a unique detectable marker, encoding the marker so that it can be different from detectable markers attached to other protein-binding molecules, thereby enabling the detection of biomarkers in multichannel assays. Examples include (but are not limited to) color-coded microspheres with known fluorescence intensities (see, for example, microspheres produced by Luminex (Austin, Tex.) using xMAP technology); microspheres containing quantum dot nanocrystals, for example, having different ratios and combinations of quantum dot colors (e.g., Qdot nanocrystals produced by Life Technologies (Carlsbad, Calif.); glass-coated metal nanoparticles (see, for example, SERS nanotags produced by Nanoplex Technologies, Inc. (Mountain View, Calif.); barcode materials (see, for example, submicron-sized striped metal bars, such as Nanobarcodes produced by Nanoplex Technologies, Inc.); coded microparticles with color barcodes (see, for example, cellcards produced by Vitra Bioscience, vitrabio.com); and glass microparticles with digitally holographically encoded images (see, for example, those produced by Illumina (San...). (CyVera microbeads produced by Diego, Calif.); combinations of chemiluminescent dyes and dye compounds; and beads of various sizes that can be detected.

[0257] In another aspect, biochips can be used for the capture and detection of biomarkers of the present invention. Various protein biochips are known in the art. These include, for example, protein biochips produced by Packard BioScience Company (Meriden Conn.), Zyomyx (Hayward, Calif.), and Phylos (Lexington, Mass.). Generally, a protein biochip comprises a substrate having a surface. A capture reagent or adsorbent is attached to the substrate surface. Typically, the surface includes a plurality of addressable addresses, each address having a capture reagent bound thereto. The capture reagent can be a biomolecule, such as a peptide or nucleic acid, which captures other biomarkers in a specific manner. Alternatively, the capture reagent can be a chromatographic material, such as anion exchange materials or hydrophilic materials. Examples of protein biochips are well known in the art.

[0258] This invention also discloses a method for predicting the probability of preterm birth, which includes measuring changes in reversal values ​​of biomarkers. For example, a biological sample may be contacted with a group containing one or more polynucleotide binding reagents. The expression of one or more of the detected biomarkers can then be evaluated according to the methods disclosed herein, for example, with or without nucleic acid amplification methods. Skilled practitioners will understand that in the methods described herein, the measurement of gene expression can be automated. For example, a system capable of multiplexing gene expression measurements can be used, for example, simultaneously providing digital readings of the relative abundance of hundreds of mRNAs.

[0259] In some implementations, nucleic acid amplification methods can be used to detect polynucleotide biomarkers. For example, the oligonucleotide primers and probes described herein can be used in amplification and detection methods using nucleic acid substrates isolated by any of the many well-known and established methods (e.g., Sambrook et al., Molecular Cloning, A laboratory Manual, pp. 7.37-7.57 (2nd edition, 1989); Lin et al., in Diagnostic Molecular Microbiology, Principles and Applications, pp. 605-16 (edited by Persing et al., (1993); Ausubel et al., Current Protocols in Molecular Biology (2001 and subsequent updates)). Methods for amplifying nucleic acids include (but are not limited to) (e.g.) polymerase chain reaction (PCR) and reverse transcription PCR (RT-PCR) (see, e.g., U.S. Patent Nos. 4,683,195; 4,683,202; 4,800,159; 4,965,188), and ligase chain reaction (LCR) (see, e.g., Weiss, Science). 254:1292-93 (1991)), chain substitution amplification (SDA) (see, for example, Walker et al., Proc. Natl. Acad. Sci. USA 89:392-396 (1992); U.S. Patent Nos. 5,270,184 and 5,455,166), thermophilic SDA (tSDA) (see, for example, European Patent No. 0 684 315 and U.S. Patent No. 5,130,238; Lizardi et al., BioTechnol. 6:1197-1202 (1988); Kwoh et al., Proc. Natl. Acad. Sci. USA 86:1173-77 (1989); Guatelli et al., Proc. Natl. Acad. Sci. USA The methods described in 87:1874-78 (1990); U.S. Patent Nos. 5,480,784; 5,399,491; and U.S. Patent Publication No. 2006 / 46265.

[0260] In some implementations, the measurement of mRNA in a biological sample can be used as an alternative to the detection of corresponding protein biomarker levels in the biological sample. Therefore, any biomarker, biomarker pair, or biomarker inverse group described herein can also be detected by detecting appropriate RNA. mRNA levels can be measured by reverse transcription quantitative polymerase chain reaction (RT-PCR, followed by qPCR). RT-PCR is used to generate cDNA from mRNA. As the DNA amplification process proceeds, the cDNA can be used in the qPCR assay to generate fluorescence. qPCR can produce absolute measurements, such as the number of mRNA copies per unit cell, compared to a standard curve. RNA blotting, microarrays, invader assays, and RT-PCR combined with capillary electrophoresis have all been used to measure mRNA expression levels in samples. See also Gene Expression Profiling:Methods and Protocols Richard A. Shimkets, ed., Humana Press, 2004.

[0261] Some embodiments disclosed herein relate to diagnostic and prognostic methods for determining the probability of preterm birth in pregnant female animals. The detection of expression levels of one or more biomarkers and / or the determination of biomarker ratios can be used to determine the probability of preterm birth in pregnant female animals. These detection methods can be used, for example, for early diagnosis of conditions to determine whether a subject is susceptible to preterm birth, to monitor the development of preterm birth or the progress of treatment protocols, to assess the severity of preterm birth, to predict preterm birth outcomes and / or the prospect of recovery or term delivery, or to help determine appropriate preterm birth treatment.

[0262] The amount of biomarkers in biological samples can be determined without limitation using the methods described above and any other methods known in the art. The resulting quantitative data is then analyzed using a classification method. In this method, the raw data is manipulated according to an algorithm, which has been predefined using a training data set, for example, as described in the examples provided herein. The algorithm can be generated using the training data set provided herein, or the guidance provided herein can be used to generate the algorithm using different data sets.

[0263] In some embodiments, analyzing measurable features to determine the probability of preterm birth in pregnant female animals includes the use of a predictive model. In other embodiments, analyzing measurable features to determine the probability of preterm birth in pregnant female animals includes comparing the measurable features with reference features. As those skilled in the art will understand, this comparison can be a direct comparison with reference features or an indirect comparison in which reference features have been incorporated into the predictive model. In other embodiments, analyzing measurable features to determine the probability of preterm birth in pregnant female animals includes one or more of the following: linear discriminant analysis models, support vector machine classification algorithms, regression feature elimination models, microarray predictive analysis models, logistic regression models, CART algorithms, flex tree algorithms, LART algorithms, random forest algorithms, MART algorithms, machine learning algorithms, penalized regression methods, and combinations thereof. In a specific embodiment, the analysis includes logistic regression.

[0264] Classification analysis can utilize any of a variety of statistical analysis methods to manipulate quantitative data and prepare for sample classification. Examples of useful methods include linear difference analysis, regression feature elimination, microarray predictive analysis, logistic regression, CART algorithm, FlexTree algorithm, LART algorithm, random forest algorithm, MART algorithm, machine learning algorithms, etc.

[0265] To generate a random forest predicting GAB, those skilled in the art can consider a group of k subjects (pregnant women) whose gestational age (GAB) at birth is known and whose blood samples from the weeks preceding delivery have been measured for N analytes (transformations). The regression tree begins with a root node containing all subjects. The mean GAB for all subjects can be calculated at the root node. The variation in GAB within the root node is high because there is a mixture of women with different GABs. The root node is then divided (assigned) into two branches, each containing women with similar GABs. The mean GAB for subjects in each branch is calculated again. The variation in GAB within each branch will be lower than the variation in the root node because the subgroup of women in each branch has a relatively more similar GAB than in the root node. The two branches are generated by selecting analytes and thresholds for generating branches with similar GABs. Analytes and thresholds are selected from a set of all analytes and thresholds, typically with random subgroups of analytes at each node. The procedure continuously and recursively generates branches to produce leaves (terminal nodes) where subjects have very similar GABs. The predicted GAB in each terminal node is the mean GAB of the subjects in that terminal node. The program generates a single regression tree. Random forests can contain hundreds or thousands of such trees.

[0266] Classification can be performed using a predictive model method that sets a threshold to determine the probability that a sample belongs to a given category. This probability is preferably at least 50%, or at least 60%, or at least 70%, or at least 80% or higher. Classification can also be performed by determining whether a comparison between the obtained data set and a reference data set produces a statistically significant difference. If so, the sample that obtained the data set is classified as not belonging to the reference data set. Conversely, if the comparison is not statistically significant compared to the reference data set, the sample that obtained the data set is classified as belonging to the reference data set.

[0267] The predictive power of a model can be evaluated based on a quality metric that provides a specific value or range of values, such as AUROC (Area Under the ROC Curve) or accuracy. The area under the curve metric is useful for comparing the accuracy of classifiers across the entire data range. A classifier with a larger AUC will have a greater ability to correctly classify unknowns between two groups of interest. In some implementations, the desired quality threshold is a predictive model that classifies samples with an accuracy of at least about 0.5, at least about 0.55, at least about 0.6, at least about 0.7, at least about 0.75, at least about 0.8, at least about 0.85, at least about 0.9, at least about 0.95, or higher. As an alternative metric, the desired quality threshold may represent a predictive model that classifies samples with an AUC of at least about 0.7, at least about 0.75, at least about 0.8, at least about 0.85, at least about 0.9, or higher.

[0268] As is known in the art, the relative sensitivity and specificity of a prediction model can be adjusted to favor a selectivity measure or a sensitivity measure, wherein the two measures are inversely related. Depending on the specific requirements of the test being performed, the limits in the above model can be adjusted to provide a selected level of sensitivity or specificity. One or both of the sensitivity and specificity can be at least about 0.7, at least about 0.75, at least about 0.8, at least about 0.85, at least about 0.9, or higher.

[0269] Initially, raw data can be analyzed by measuring the value of each biomarker, typically repeated three or more times. Data can be manipulated; for example, the raw data can be transformed using a standard curve, and the mean and standard deviation for each patient can be calculated using the average of the three measurements. These values ​​can be pre-transformed in the model, for example, using log-transformation, Box-Cox transformation (Box and Cox transformation). Royal Stat.Soc .,Series B,26:211-246(1964). The data is then fed into a predictive model that classifies the samples according to their condition. The resulting information can be communicated to patients or healthcare providers.

[0270] To generate a preterm birth prediction model, a robust dataset is used in the training group, comprising known control samples and samples corresponding to the preterm birth category of interest. Sample size can be selected using accepted criteria. As discussed above, different statistical methods can be used to obtain a high-precision prediction model. Example 2 provides an instance of this analysis.

[0271] In one implementation, hierarchical clustering is performed in the derivation of the prediction model, where Pearson correlation is used as the cluster metric. One approach is to consider the preterm birth data set as “learning samples” in a “supervised learning” problem. CART is the standard in medical applications (Singer, Recursive Partitioning in the Health Sciences, (1999)) and can be modified by: converting any qualitative feature into a quantitative feature; classifying by the significance level achieved by a sample reuse method evaluated using the Hotelling T2 statistic; and the appropriate application of the lasso method. Indeed, by appropriately using the Gini classification criteria in the regression quality assessment, the prediction problem can be transformed into a regression problem without losing the sight of prediction.

[0272] This method resulted in the so-called FlexTree (Huang, Proc.Nat.Acad.Sci.USA 101:10529-10534 (2004)). FlexTree performs well in simulating and applying various data formats and is useful for the methods advocated in practice. Software-automated FlexTree has been developed. As an alternative, LARTree or LART (Turnbull (2005)) can be used. Classification Trees with Subset Analysis Selection by the Lasso (Stanford University). The names reflect binary trees, such as CART and FlexTree; lasso, as mentioned; and implementations of lasso via so-called LARS, Efron et al. (2004) Annals of Statistics 32:407-451 (2004). Also see, Huang et al. Proc.Natl.Acad.Sci.USA .101(29):10529-34(2004). Other analytical methods that can be used include logistic regression. One method of logistic regression: Ruczinski, Journal of Computational and Graphical Statistics 12:475-512(2003). Logistic regression is similar to CART, where its classifier can be displayed as a binary tree. The difference is that each node has a Boolean statement about the features, which is more conventional than the simple AND statement produced by CART.

[0273] Another approach is the nearest shrunken centroid method (Tibshirani, Proc). Natl.Acad.Sci.USA 99:6567-72 (2002)). This technique is k-means-like, but has the advantage of automatically selecting features (as in lasso) to focus attention on the small number of informative features by reducing cluster centers. This method is available as PAM software and is widely used. Two other sets of algorithms that can be used are Random Forest (Breiman, Machine Learning 45:5-32 (2001)) and MART (Hastie, The Elements of Statistical Learning, Springer (2001)). In the art, these two methods are known as “committee methods”, which involve “voting” on predictors of the outcome.

[0274] To provide an importance ranking, the false discovery rate (FDR) can be determined. First, a set of null distributions with distinct values ​​is generated. In one implementation, the observed spectral values ​​are arranged to generate a sequence of distributions of incidentally obtained correlation coefficients, thereby producing an appropriate set of null distributions of the correlation coefficients (Tusher et al.). Proc.Natl.Acad.Sci.USA 98,5116-21 (2001)). The null distribution set is obtained by: permuting the values ​​of each spectrum of all available spectra; calculating the paired correlation coefficients of all spectra; calculating the probability density function of the correlation coefficients of this permutation; and repeating this process N times, where N is a large number, typically around 300. Using the N-distribution, an appropriate measure (mean, median, etc.) of the count of correlation coefficient values ​​(values ​​of similarity) obtained at a given significance level exceeding the values ​​obtained from the distribution of similarity values ​​observed in experiments is calculated.

[0275] The FDR is the ratio of the expected number of false significant correlations (an estimate of the correlation from the Pearson correlation of the chosen random data set) to the number of correlations (significant correlations) from the empirical data set that are greater than the Pearson correlation of the chosen random data set. This cutoff correlation value can be applied to the correlation between experimental spectra. Using the distribution described above, a confidence level for significance is selected. This is used to determine the minimum value of the correlation coefficient beyond incidentally obtained results. Using this method, a threshold for positive correlation, negative correlation, or both is obtained. Using this threshold, the user can filter observations with paired correlation coefficients and remove those that do not exceed the threshold. Additionally, an estimate of the false positive rate for a given threshold can be obtained. For each of the individual "random correlation" distributions, it is possible to find how many observations fall outside the threshold range. The program provides a count sequence. The mean and standard deviation of the sequence provide the mean number of potential false positives and their standard deviation.

[0276] In alternative analytical approaches, the variables selected in cross-sectional analyses are used individually as predictors in time-event analyses (survival analyses), where the event is the occurrence of preterm birth, and event-free subjects are considered to have been examined at the time of delivery. Considering specific pregnancy outcomes (preterm birth event or no event), the duration of randomization observed in each patient, and the selection of proteomics and other characteristics, parametric methods for analyzing survival may outperform the widely used semi-parametric Cox model. Weibull parameter fitting for survival allows for a single increase, decrease, or constant hazard rate, and also exhibits proportional hazard representation (as with the Cox model) and accelerated failure time representation. All standard tools available for obtaining approximate maximum likelihood assessments of regression coefficients and corresponding functions are available in this model.

[0277] Furthermore, the Cox model can be used, particularly because reducing the number of covariates to a size manageable by lasso significantly simplifies the analysis, making it possible to predict the time of preterm birth using nonparametric or semiparametric methods. These statistical tools are known in the art and can be applied to all aspects of proteomics data. A set of biomarkers, clinical, and genetic data is provided that can be readily identified and contains rich information about the probability of preterm birth and the predicted time of preterm birth events in the pregnant female animals. Additionally, the algorithm provides information about the probability of preterm birth in pregnant female animals.

[0278] Therefore, those skilled in the art will understand that the probability of preterm birth according to the present invention can be determined using quantitative or categorical variables. For example, in practice of the method described in the present invention, categorical data analysis can be performed on the measurable characteristics of each of N biomarkers to determine the probability of preterm birth as a binary classification outcome. Alternatively, the method described in the present invention can analyze the measurable characteristics of each of N biomarkers by initially calculating a quantitative variable, specifically, the predicted gestational age at birth. Subsequently, the predicted gestational age at birth can be used as the basis for predicting the risk of preterm birth. By initially using a quantitative variable and subsequently converting the quantitative variable into a categorical variable, the method described in the present invention considers the continuous region of measurement of the measurable characteristic detection. For example, by predicting gestational age at birth rather than making a binary prediction of preterm birth relative to term birth, it is possible to adjust the treatment of pregnant female animals. For example, an early prediction of gestational age at birth will lead to closer prenatal intervention, i.e., monitoring and treatment, than a prediction closer to term gestational age.

[0279] Among women whose predicted GAB is j days plus or minus k days, p(PTB) can be estimated as the proportion of women in the PAPR clinical trial (see Example 1) whose predicted GAB is j days plus or minus k days but who actually deliver before 37 weeks of gestation. Generally, for women whose predicted GAB is j days plus or minus k days, the probability p (actual GAB < specified GAB) that the actual gestational age at birth will be less than the specified gestational age is estimated as the proportion of women in the PAPR clinical trial whose predicted GAB is j days plus or minus k days but who actually deliver before the specified gestational age.

[0280] During the development of a predictive model, it is desirable to select subgroups of biomarkers, i.e., at least 3, 4, 5, or 6 biomarkers, or even a complete set of biomarkers. Typically, the subgroups of biomarkers selected will provide what is needed for quantitative sample analysis, such as reagent availability and ease of quantification, while maintaining a high-accuracy predictive model. The selection of informative biomarkers used to build the classification model requires defining performance metrics and user-defined thresholds for generating a model with useful predictive capabilities based on these metrics. For example, the performance metrics could be AUC, predictive sensitivity and / or specificity, and the overall accuracy of the predictive model.

[0281] As those skilled in the art will understand, analytical classification methods can utilize any of a variety of statistical analysis methods to manipulate quantitative data and prepare for sample classification. Examples of useful methods, without limitation, include linear differential analysis, regression feature elimination, microarray predictive analysis, logistic regression, CART algorithm, FlexTree algorithm, LART algorithm, random forest algorithm, MART algorithm, and machine learning algorithms. Various methods are used in training the model. The selection of marker subgroups can be either forward or backward selection. It is possible to choose a number of markers that optimizes model performance without using all of them. One approach to defining the optimal number of items is to select the number of items that produces a model with the desired predictive power (e.g., AUC > 0.75, or an equivalent measure of sensitivity / specificity) where the predictive power is obtained by using any combination and number of items for a given algorithm, but not exceeding one standard deviation from the maximum value obtained for that metric.

[0282] In another aspect, the present invention provides a kit for determining the probability of preterm birth. The kit may include one or more reagents for detecting biomarkers, a container for containing a biological sample isolated from a pregnant female animal, and printed instructions for reacting the reagents with the biological sample or a portion of the biological sample to detect the presence or amount of the biomarker isolated in the biological sample. The reagents may be packaged in separate containers. The kit may also include one or more control reference samples and reagents for performing immunoassays.

[0283] The kit may contain one or more containers for the composition contained within the kit. The composition may be in liquid form or may be freeze-dried. Suitable containers for the composition include, for example, bottles, vials, syringes, and test tubes. The containers may be formed from a variety of materials, including glass or plastic. The kit may also include a package insert containing written instructions for determining the probability of preterm birth.

[0284] Based on the foregoing description, it is evident that changes and variations can be made to the invention described herein to suit various uses and conditions. These embodiments are also within the scope of the following claims.

[0285] In this document, the enumeration of elements in any definition of a variable includes the definition of the variable as any single element or combination (or sub-combination) of the listed elements. Similarly, the enumeration of implementations includes implementations as any single implementation or in combination with any other implementation or parts thereof.

[0286] All patents and patent disclosures mentioned in this specification are incorporated herein by reference to the same extent that each individual patent and patent disclosure is specifically and individually indicated and incorporated herein by reference.

[0287] The following embodiments are provided by way of non-limiting description. Example

[0288] Example 1. Establishment of a sample set for the discovery and validation of early biomarkers

[0289] Establish standard procedures for conducting clinical studies of proteomics evaluation for controlling the risk of preterm birth (PAPR). Samples were obtained from women at 11 Institutional Review Board (IRB) approved sites throughout the United States. Serum and plasma samples, along with relevant information on patient demographics, past medical and pregnancy history, current pregnancy, and concurrent drug treatments, were obtained after informed consent was obtained. Postpartum, data related to maternal and infant conditions and complications were collected. Serum and plasma samples were processed according to procedures requiring standardized cryogenic centrifugation, aliquoting of samples into 2-D barcode vials, and subsequent freezing at -80°C.

[0290] Postpartum, preterm birth cases were reviewed individually to determine whether they were spontaneous or medically induced preterm births. Only spontaneously preterm birth cases were used for this analysis. For the discovery of preterm birth biomarkers, serum samples from 86 preterm birth cases and 172 controls were analyzed, covering gestational age at blood draw (GABD) from 17 weeks 0 days (17.0) to 28 weeks 6 days (28.6). For verification purposes, separate sample groups were also analyzed, consisting of serum from 50 preterm birth cases and 100 controls within the same gestational age range. Two controls were matched for each case by GABD, and these controls were selected from a randomized control group matched for the birth distribution reported in the 2012 National Population Report. Procedures were established to ensure that laboratory personnel were unaware of the gestational age at birth and the case vs. control status of the subjects in both sample groups. Data personnel were also unaware of the verification sample groups until the sample analysis was completed.

[0291] High-abundance proteins were removed from serum samples using the Human 14 Multiple Affinity Removal System (MARS 14), which removed the 14 most abundant proteins considered non-informative for identifying disease-related changes in serum proteomics. Filtered samples were removed using a MARS-14 column (4.6 × 100 mm, catalog #5188-6558, Agilent Technologies) according to the manufacturer's specifications. Samples were cooled to 4°C in an autosampler, the removal column was run at room temperature, and the collected fractions were held at 4°C until further analysis. Unbound fractions were collected for further analysis.

[0292] The removed serum sample was reduced with dithiothreitol, alkylated with iodoacetamide, and then hydrolyzed with 5.0 μg of trypsin Gold-Mass Spec (Promega) at 37 °C for 17 h (±1 h). After trypsin digestion, a mixture of 187 stable isotope standard (SIS) peptides was added to the sample, and half of each sample was desalted on an Empore C18 96-well solid-phase extraction plate (3MBioanalytical Technologies). The plate was conditioned according to the manufacturer's specifications. The peptides were washed with 300 μl of 1.5% trifluoroacetic acid and 2% acetonitrile, eluted with 250 μl of 1.5% trifluoroacetic acid and 95% acetonitrile, frozen at -80 °C for 30 min, and then lyophilized to dryness. The lyophilized peptides were reconstituted with 2% acetonitrile / 0.1% formic acid containing three non-human internal standard (IS) peptides. The peptides were separated at 40°C on an Agilent Poroshell 120EC-C18 column (2.1×100mm, 2.7μm) using an acetonitrile gradient of 400μl / min for 30 min and then injected into an Agilent 6490 triple quadrupole mass spectrometer.

[0293] Samples after removal and trypsin hydrolysis were analyzed using a scheduled multiple reaction monitoring method (sMRM). The sMRM assay monitored 898 transformations and measured 259 bioactive peptides and 190 IS peptides (187 SIS + 3 IS), representing 148 proteins. Chromatographic peak integration was performed using Mass Hunter quantitative analysis software (Agilent Technologies).

[0294] Data Analysis

[0295] The analysis of discovery and verification sample data is conducted in two phases. In the first phase, robust biomarkers are identified through the selection of discovery samples and confirmation using independent verification sample groups. In the second phase, the discovery and verification data are combined and used to identify the optimal analytes and analyte groups for classifier development.

[0296] Phase I: Blind Analysis

[0297] Initially, the classifier was developed for gestational ages of 17.0 to 25.6 gestational ages. Using discovery samples, a group of peptides corresponding to 62 proteins was selected based on pre-analysis and analytical criteria. The diagnostic performance of the analytes was evaluated within a series of narrow GABD windows spanning three weeks (with two-week overlap between adjacent windows). Based on concordance in diagnostic performance (upregulation and downregulation in cases relative to controls throughout the GABD), subgroups of 43 analytes were selected for further analysis.

[0298] For each narrow GABD window, a set of reversals is formed using all combinations of up- and down-regulated analytes within the narrow window. The reversal value is the ratio of the relative peak area of ​​the up-regulated analyte to the relative peak area of ​​the down-regulated analyte and is used to normalize the difference and amplify the diagnostic signal. Among all possible reversals within the narrow window, subgroups are selected based on their individual univariate performance (AUC ≥ 0.6).

[0299] For each window, reversal groups of different sizes were formed (sizes 2, 3, 4, 6, and 8). For each group size within a window, Monte Carlo cross-validation (MCCV) was performed by iteratively training and testing a logistic classifier 1000 times on 70% and 30% of the samples, respectively. Subsequently, the optimal group size of 4, determined by the average MCCV AUC, was used to identify candidate reversals that were well implemented for the group. Candidate reversals were identified by their frequency of occurrence on the best-performing logistic classifier at group size 4 in the MCCV analysis. For each window, three reversal frequency tables were generated using a performance metric of AUC or partial AUC (pAUC) with sensitivity ranging from 0.7 to 1, or the correlation between the classifier output score and the time to delivery (TTB) value (the difference in days between GABD and gestational age at birth). The top 15 reversals from each of these reversal lists were selected for further analysis.

[0300] For each GABD narrow window, inversion groups of sizes 2, 3, and 4 are formed from each of the three lists (AUC, pAUC, and TTB), and based on the performance of the MCCV analysis, the top 15 groups for each group size are selected within each window. These top 15 groups of sizes 2, 3, and 4, along with the top 15 inversions from each of the three lists (AUC, pAUC, and TTB) for each window, are used to train a logistic classifier on the discovery samples and generate classification scores for the review samples in a blind manner.

[0301] Third-party statisticians evaluated the performance of all reversal and classifier groups and reported the correlation between AUC, pAUC and TTB of the ROC curves and classifier scores.

[0302] Phase II: Unblinded Analysis

[0303] After unblinding, the discovery and verification data were combined and reanalyzed. Since the expression of diagnostic proteins can change during pregnancy, we examined protein levels associated with GABD. A median smoothing window of + / - 10 days was applied to generate a kinetic plot. Relative protein levels were expressed as the ratio of the area of ​​the endogenous peptide peak to its corresponding SIS standard (relative ratio). Examples of proteins with elevated levels during pregnancy but no difference between PTB cases and controls are shown in Figures 3, 4, and 10. This measurement of protein levels can be used to accurately determine the date of pregnancy (e.g., the pregnancy “clock”). The pregnancy clock predicts gestational age based on the relative abundance (conversion) of one or more proteins. Alternatively, in this same analysis, we identified proteins whose levels changed during GABD but showed differences between PTB cases and controls, Figure 5. These proteins are obvious diagnostic candidates for the development of the PTB classifier. The effect of using the ratio of overexpressed proteins to underexpressed proteins to form a reversal is also illustrated. Figure 8 and21 This clearly led to an increased segregation between PTB cases and controls. Previous analyses have shown that the levels of some analytes may be influenced by pre-pregnancy body mass index (BMI). CLIN. CHEM. 37 / 5, 667-672 (1991); European Journal of Endocrinology (2004) 150 161–171. For this reason, the effect of BMI on segregation was investigated by representing the reversal value during pregnancy only in patients with a BMI less than 35. Figure 21 This leads to further improvements in the separation.

[0304] The reversal selection and classifier development in the merged discovery and verification datasets reflect earlier research. We focus on the third overlapping GABD window (133–153 days) to illustrate the analysis. MCCV analysis was performed to identify candidate reversals. To evaluate group performance, reversal values ​​were merged in a simple LogSum classifier. The LogSum classifier assigns a score to each sample based on the sum of the log values ​​of the relative ratios of each reversal to that sample. The lack of coefficients in this type of classifier helps avoid overfitting. Any person skilled in the art can obtain an equivalent logistic classifier using the same analytes using well-established techniques. The multivariate performance of the three best reversal groups formed by the four proteins is shown as a bar chart of AUC values ​​obtained through cross-validation and in... Figure 8 The ROC curve is shown. Previous analyses have indicated that the levels of some analytes may be affected by pre-pregnancy body mass index (BMI).

[0305] We identified ITIH4 / CSH as a strong predictor of time to delivery (TTB) by using the protein illustrated in this paper and / or reversing it. Figure 10 TTB is defined as the difference between GABD and gestational age at birth (GAB). This has the potential to enable prediction of TTB (or GAB) clinically, either individually or in mathematical combinations of these analytes.

[0306] Example 2. Validation of IBP4 / SHBG sPTB predictor

[0307] This example demonstrates the validation of the IBP4 / SHBG sPTB predictor in asymptomatic women in early pregnancy, as identified in a large maternal serum proteomics study.

[0308] Subjects

[0309] Proteomics assessment of preterm birth risk (PAPR) will be conducted under standardized protocols at 11 Institutional Review Board (IRB) approved sites in the United States (Clinicaltrials.gov ID: NCT01371019). Participants will be recruited between 17 0 / 7 and 28 6 / 7 weeks of gestational age (GA). Pre-defined dates will be established using protocols that provide the best clinically estimated gestational age, confirmed by early ultrasound biometry or ultrasound alone. Body mass index (BMI) will be derived from height and pre-pregnancy self-reported weight. Pregnancies with multiple pregnancies and those with known or suspected major fetal abnormalities will be excluded. Relevant information on participants' demographic characteristics, past medical and pregnancy history, current pregnancy history, and ongoing medications will be collected and entered into an electronic case report form. Postpartum, data on maternal and infant outcomes and complications will be collected. All deliveries will be categorized as term (≥37 0 / 7 weeks GA), spontaneous preterm birth (including premature rupture of membranes), or medically induced preterm birth. As instructed, the principal investigator will clarify any discrepancies at the study site. Make a determination and lock in the data before conducting a validation study.

[0310] Sample collection

[0311] Maternal blood was collected and processed as follows: a 10-minute room temperature clotting period, followed by rapid frozen centrifugation or incubation at 4–8°C on an ice-water bath until centrifugation. The blood was centrifuged within 2.5 hours of collection, and 0.5 ml aliquots of serum were stored at -80°C until analysis.

[0312] Predictor Development Principles

[0313] The development of the IBP4 / SHBG predictor involved independent and sequential discovery, verification, and validation steps consistent with the Institute of Medicine's (IOM) best practice guidelines for "omics" research. (IOM. Evolution of Translation Omics: Lessons Learned and the Path Forward. (Michael CM, Nass SJ, Omenn GS, eds.). Washington, DC: The National Academies Press.; 2012: 1–355.) Analytical validation precedes clinical validation sample analysis and includes evaluation of batch-to-batch and intra-batch accuracy, residues, and limits of detection.

[0314] Independent of discovery and verification, a validation nested case / control analysis was performed on pre-specified sPTB case and control samples. sPTB cases included samples from all nine sites, two of which were unique for validation. Validation cases and controls underwent 100% in-situ documentation verification against each subject's medical records prior to mass spectrometry (MS) serological analysis. This process ensured all subjects met inclusion and exclusion criteria and confirmed the allocation of medical / pregnancy complications and GA at birth for all subjects at the time of sample collection and delivery. Detailed analytical protocols were pre-defined, including the validation study design, analysis plan, and blinding procedures. Staff, except the Director of Clinical Operations (DCO) and the Clinical Data Manager, were unaware of the allocation of subject case, control, and GA data. The data analysis plan included pre-defined validation claims and procedures for dual independent external analyses. Predictor scores, calculated as described below, were determined for all subject samples by an unaware statistician. Independent external statistical analyses were performed on case, control, and GA data linked to predictor scores by the DCO. The area under the receiver operating characteristic curve (AUROC) and significance test results were then transferred back to the DCO. Data transfer incorporated the use of the SUMPRODUCT function (Microsoft Excel 2013) to ensure data integrity. Real-time digital timestamps were applied to analytical data, planning, and reporting to provide a review trail from each participant's data up to the validation results.

[0315] Validation study design

[0316] In the preliminary analysis, sPTB cases were defined as subjects who delivered due to premature rupture of membranes (PPROM) or spontaneous labor at GA <37 0 / 7 weeks. Controls were subjects who delivered at GA ≥37 0 / 7 weeks. Previous discovery and verification analyses investigated 44 candidate biomarkers using serum samples collected across a wide gestational age range (17 0 / 7 to 25 6 / 7 weeks GA) (Supplementary Material). Discovery and verification identified the optimal narrow GA interval (19 0 / 7 to 21 6 / 7 weeks) and two proteins, IBP4 and SHBG, which were used as the best predictors of sPTB by AUROC ratio (IBP4 / SHBG) (Supplementary Material). In discovery and verification, patients without extreme BMI values ​​showed improved classification performance by IBP4 / SHBG (Supplementary Results). Following the discovery and verification analyses, we proceeded with further analysis and clinical validation.

[0317] The validation of sPTB cases involved 18 subjects whose blood samples were collected between 19 0 / 7 and 21 6 / 7 weeks of GA (GABD), from a total of 81 available subjects between 17 0 / 7 and 28 6 / 7 weeks of GA. The control group consisted of two controls matched by GABD for each sPTB case, randomly selected using the R statistical procedure (R 3.0.2) (Team RC. R: a Language and Environment for Statistical Computing. Vienna, Austria; 2014-2015; Matei A, Tillé Y. The R “sampling” package. European Conference on Quality in Survey Statistics. 2006) and X. 2 The test was compared with the distribution of full-term births as listed in the 2012 National Population Statistics Reports (Martin JA, Hamilton BE, Osterman MJ, Curtin SC, Mathews TJ. Births: Final Data for 2012. National Vital Statistics Reports. 2014; 63(09): 1–86). The test yielded a p-value close to 1.0 for the randomly generated control group (groups of 10).

[0318] The primary objective was to validate the performance of the IBP4 / SHBG ratio as a predictor of spTB using AUROC (TeamRC.R: A Language and Environment for Statistical Computing. Vienna, Austria; 2014. 2015; Sing T, Sander O, Beerenwinkel N, Lengauer T.ROCR: Visualizing Classifier Performance in R. Bioinformatics. 2005; 21(20):7881). To control for the overall error rate of multiple tests (α = 0.05), a fixed-order approach (Dmitrienko A, Tamhane AC, Bretz F, eds. Multiple Testing Problems in Pharmaceutical Statistics. Boca Raton, Florida: CRC Press; 2009: 1–320; Dmitrienko A, D'Agostino RB, Huque MF. Key multiplicity issues in clinical drug development. Stat Med. 2012; 32(7): 1079–111. doi: 10.1002 / sim.5642.) was applied to the GABD increment within the optimal interval (19 0 / 7 to 21 6 / 7 weeks GA) for identification in discovery and review with and without BMI partitioning (see Supplementary Material). Significance was evaluated by Wilcoxon-Mann-Whitney statistic for equivalence with AUROC = 0.5 (randomness factor).(Bamber D.The area above the ordinal dominance graph and the area below the receiver operatingcharacteristic graph.Journal of mathematical psychology.1975;12(4):387–415.doi:10.1016 / 0022-2496(75)90001-2;Mason SJ,Graham NE.Areas beneath the relative operating characteristics(ROC)and relative operating characteristics(ROC) levels(ROL)curves: Statistical significance and interpretation. QJR Meteorol Soc. 2002;128(584):2145–2166.doi:10.1256 / 003590002320603584.). For the determination of categorical performance for <37 0 / 7 relative to ≥37 0 / 7 week GA (e.g., <36 0 / 7 relative to ≥36 0 / 7, <35 0 / 7 relative to ≥35 0 / 7), cases and controls are redefined as all subjects less than and equal to / greater than the specific boundaries, respectively.

[0319] Laboratory methods

[0320] A highly multi-pathway multiple response monitoring (MRM) MS assay was developed using systems biology approaches (Supplementary Methods and Results). The assay was validated to quantify proteolytic peptides specific to the predictor proteins IBP4 and SHBG, as well as other controls. Samples were processed in 32 batches comprising clinical subjects (24), pooled serum standards (HGS) from healthy non-pregnant donors (3), pooled serum standards (pHGS) from healthy pregnant donors (3), and phosphate-buffered saline (2) used as treatment controls. For all analyses, high-abundance and non-diagnostic proteins in serum samples were first removed using a MARS-14 immunoremoval column (Agilent Technologies), followed by reduction with dithiothreitol, alkylation with iodoacetamide, and hydrolysis with trypsin. Restable isotope-labeled standard (SIS) peptides were then added to the samples, followed by desalting and analysis by reversed-phase liquid chromatography (LC) / MRM-MS. The SIS peptide is normalized by generating a response ratio (RR), in which the peak area (i.e., conversion) of the peptide fragment ion measured in serum is divided by the peak area of ​​the corresponding SIS conversion when the same serum sample is added.

[0321] IBP4 / SHBG predictor

[0322] The predictor score is defined as the natural logarithm of the ratio of IBP4 to SHBG peptide conversion response ratio:

[0323]

[0324] Where RR is the response ratio measured for each peptide.

[0325] result

[0326] Figure 23The distribution of participants in the PAPR study was summarized. Between March 2011 and August 2013, 5,501 participants were recruited. As pre-determined in the protocol, 410 (6.7%) participants were excluded from the analysis due to receiving progestin therapy after the first trimester of pregnancy. Another 120 (2.2%) participants were excluded due to early discontinuation, and 146 (2.7%) participants lost follow-up. A total of 4,825 participants were available for analysis. There were 533 cases of PTB; 248 (4.7%) were spontaneous and 285 (5.9%) were medically induced PTB. Compared to those who delivered at term, participants with sPTB were more likely to have had one or more prior PTBs and experienced bleeding after 12 weeks of gestation during the pregnancies in this study (Table 1). The characteristics of the sPTB cases and full-term controls selected for validation were not significantly different from each other, except for a significantly larger number of Hispanic controls (47.5% vs. 33.3%, p = 0.035). Similarly, except for the race of the full-term controls, the subjects selected for validation were largely representative of the study cohort as a whole (Table 1).

[0327] Validation Analysis

[0328] In the discovery and verification analyses, the IBP4 / SHBG ratio and the GA interval between 19 0 / 7 and 21 6 / 7 weeks were identified as the best-performing sPTB predictors, using AUROC and GA intervals, respectively (Supplementary Results below). For validation, the IBP4 / SHBG predictor was validated using a pre-determined fixed-order method with and without BMI, based on the best performance identified for the GA interval between 19 1 / 7 and 20 6 / 7 weeks. Without considering BMI, the validation performance was AUROC = 0.67 (p = 0.02) (Supplementary Results). However, as expected, performance was lower with >22 and ≤37 kg / m². 2 The BMI classification improved performance, corresponding to AUROC = 0.75 (p = 0.016, 95% CI 0.56–0.91). Figure 24 A more detailed identification of BMI classifications can be found in the supplemental results. Performance measures of sensitivity, specificity, AUROC, and odds ratio (OR) were determined at different case vs. control boundaries (Table 2). For spTB vs. term (<37 0 / 7 weeks vs. ≥37 0 / 7 weeks), the sensitivity and specificity were 0.75 and 0.74, respectively, and the odds ratio (OR) was 5.04 (95% CI 1.4–18). Results for other boundaries are summarized in Table 2. Lower GA boundaries improved test accuracy.

[0329] Figure 25The positive predictive value (PPV) of morbidity moderated is shown in the figure, which is a clinical risk measure as a function of the predictor score. As the PPV increases from the background value (7.3% for singleton births in the US) to 2× (14.6%) or 3× (21.9%) (dashed line) or higher relative risk ( Figure 25 This resulted in the segmentation of subjects by increasing predictor scores. Figure 25 The box plots show the distribution of IBP4 / SHBG predictor scores for subjects color-coded by GA classification at birth. The earliest sPTB cases (<35 0 / 7 weeks GA) had higher predictor scores than late term controls (≥39 0 / 7 weeks GA), while the scores of late sPTB cases (≥35 0 / 7 to <37 0 / 7 weeks GA) overlapped with those of early term controls (≥37 0 / 7 to <39 0 / 7 weeks GA). Figure 25 Based on the predictor score cutoff corresponding to 2 × relative risk (14.6% PPV), the validation subjects were identified as high or low risk. The delivery rates of the high and low risk groups were then displayed as events in the Kaplan-Meier analysis. Figure 26 According to the analysis, those classified as high-risk generally delivered earlier than those classified as low-risk (p = 0.0004).

[0330] Post-verification analysis

[0331] The performance of predictors was measured using a combination of subjects from the optimal BMI and GA intervals (Supplementary Data below) and validation analyses. ROC curves for the pooled sample groups are shown, corresponding to an AUROC of 0.72 (p = 0.013). Figure 27 ).

[0332] Using an "omics" approach, we developed a BMI range for individuals with BMIs >22 and ≤37 kg / m². 2The ratio of IBP4 / SHBG levels between 19 and 20 weeks of gestation constitutes a maternal serum predictive factor, which identifies 75% of women destined to develop spTB. Prior studies include spTb history (Goldenberg et al., Epidemiology and causes of preterm birth. Lancet. 2008; 371(9606):75–84. doi:10.1016 / S0140-6736(08)60074-4; Petrini et al., Estimate deffect of 17alpha-hydroxyprogesterone caproate on preterm birth in the United States. Obstet Gynecol. 2005; 105(2):267–272) and cervical length measurement (Iams et al., The length of the cervix and the risk of spontaneous premature delivery. National Institute of Child Health and Human Development Maternal Fetal Medicine Unit Network. N Engl J Med. 1996; 334(9):567–72; Hassan et al., Vaginal progesterone reduces the rate of preterm birth in women with a sonographic shortness of breath). Cervix: a multicenter, randomized, double-blind, placebo-controlled trial. Ultrasound Obstet Gynecol. 2011; 38(1):18–31). These are considered to be the best clinical risk measures to date; however, neither individually nor in combination can predict most sPTB.

[0333] Ideally, a spTP prediction tool would be minimally invasive, performed in early pregnancy at the same time as a routine obstetric visit, and would accurately identify those at highest risk. Current “omics” research suggests that perturbations to pregnancy physiology can be detected in maternal serum analytes measured in spTP subjects. Omic discovery studies in PTB include proteomics (Gravett et al., Proteomic analysis of cervical-vaginal fluid: identification of novel biomarkers for detection of intra-amniotic infection. J ProteomeRes. 2007;6(1):89–96; Goldenberg et al., The preterm prediction study: the value of new vs standard risk factors in predicting early and all spontaneous pretermbirths. NICHD MFMU Network. Am J Public Health.1998;88(2):233–8; Gravett et al., Diagnosis of intra-amniotic infection by proteomic profiling and identification of novel biomarkers.JAMA.2004;292(4):462–469; Pereira et al., Insights into the multifactorial nature of preterm birth: proteomic profiling of the maternal serum glycoproteome and maternal serum peptidome womenin preterm labor.Am J Obstet Gynecol.2010;202(6):555.e1–10;32;Pereira et al., Identification of novel protein biomarkers of preterm birth in humancervical-vaginal fluid.JProteome Res.2007; 6(4): 1269–76; Dasari et al., Comprehensive proteomic analysis of human cervical-vaginal fluid. J Proteome Res. 2007; 6(4): 1258–1268; Esplin et al., Proteomic identification of serum peptides predicting subsequent spontaneous preterm birth. Am J Obstet Gynecol. 2010; 204(5): 391.e1–8), transcriptomics (Weiner et al., Human effector / initiator gene sets that regulate myometrial contractility during term and preterm labor. Am J Obstet Gynecol. 2010; 202(5): 474.e1–20; Chim et al., Systematic identification of spontaneous preterm birth-associated RNA transcripts in maternal plasma. PLoS ONE. 2012; 7(4): e34328. Enquobahrie et al., Early pregnancy peripheral blood gene expression and risk of preterm delivery: a nested case control study. BMC Pregnancy Childbirth. 2009; 9(1): 56), genomics (Bezold et al., The genomics of preterm birth: from animal models to human studies. Genome Med. 2013; 5(4): 34; Romero et al., Identification of fetal and maternal single nucleotide polymorphisms in candidate genes that predispose to spontaneous preterm labor with intact membranes. Am J Obstet Gynecol.2010; 202(5):431.e1–34; Swaggart et al., Genomics of preterm birth. Cold Spring Harb Perspect Med. 2015; 5(2):a023127; Haataja et al., Mapping a new spontaneous preterm birth susceptibility gene, IGF1R, using linkage, haplotype sharing, and association analysis. PLoSGenet. 2011; 7(2):e1001293; McElroy et al., Maternal coding variants in complementreceptor 1 and spontaneous idiopathic preterm birth. Hum Genet. 2013; 132(8):935–42) and metabolomics (Menon et al., Amniotic fluid metabolomic analysis in spontaneous preterm birth. Reprod Sci. 2014; 21(6):791–803) methods. However, to date, none of these methods have produced a validated test to reliably predict the risk of spTB in asymptomatic women.

[0334] This invention enables the independent discovery, verification, and validation of analyses while adhering to IOM guidelines for the development of omics testing, and represents the results of large-scale, concurrent clinical studies. It involves constructing large-scale and standardized multi-pathway proteomics assays to explore relevant biological pathways in pregnancy. The study size and relatively wide blood collection window (17 0 / 7 to 28 6 / 7 weeks GA) also allow for the identification of GA intervals with significant protein concentration variations between spTB cases and full-term controls. The use of low-complexity predictor models (i.e., the ratio of two proteins) limits the problem of overfitting.

[0335] The application of proteomics assays and model building led to the identification of a pair of key proteins (IBP4 and SHBG) with consistently excellent predictive performance for sPTB. Despite the challenges in building classifiers for conditions attributable to multiple etiologies, the predictors showed good performance at the cutoff point of <37 0 / 7 weeks GA vs. ≥37 0 / 7 weeks GA, with an AUROC of 0.75. Importantly, the accuracy of the predictors improved for early sPTB (e.g., <35 0 / 7 weeks GA), enabling the detection of those sPTBs with the highest probabilities of disease development. Subjects identified as high-risk for sPTB using the IBP4 / SHBG predictors delivered significantly earlier than those identified as low-risk. Our findings suggest that IBP4 and SHBG may play important functions related to the etiology of sPTB and / or act as confluence points in related biological pathways.

[0336] In most of our research centers, universal transvaginal ultrasound (TVU) measurement of cervical length (CL) is not performed regularly, and this measurement is available to less than one-third of study participants. It would be meaningful to evaluate in future studies whether proteomics predictors could improve CL measurement, or, as an alternative, whether risk classification using the IBP4 / SHBG classifier could identify women who would benefit most from subsequent CL measurements. Finally, it would be interesting to investigate the performance of the aforementioned molecular predictors in conjunction with BMI variables, or possibly in combination with other medical / pregnancy history and sociodemographic characteristics.

[0337] In summary, this study validated a predictive test for spPTB based on serum measurements of IBP4 and SHBG in asymptomatic multiparous and nulliparous women in a completely independent group of subjects. Further research into these proteins, their genetic regulation, and other functions of related pathways may help elucidate the molecular and physiological basis of spPTB. The application of this predictive factor should enable early and sensitive detection of women at risk for spPTB. This could improve pregnancy outcomes through enhanced clinical oversight and accelerate the development of clinical interventions for spPTB prevention.

[0338] Supplementary materials and methods

[0339] Discovery and verification of subjects

[0340] The subjects identified and verified were derived from the PAPR study described above in this embodiment.

[0341] Discovery and review principles

[0342] sPTB cases are as described above in this embodiment. Predictor discovery and verification were performed according to best practices guidelines in omics research (IOM). Evolution of Translation Omics: Lessons Learned and the Path Forward (Michael CM, Nass SJ, Omenn GS, eds.). Washington, DC: The National Academies Press.; 2012: 1–355). Nested case / control analyses used completely independent sample groups. Cases and controls selected for discovery and verification underwent central review of intra-subject data discrepancies; source document verification (SDV) with medical records was not performed. All sPTB cases and controls used for discovery and verification were individually judged by the chief medical officer, and discrepancies were clarified by the PI at the clinical site. Detailed analytical procedures were pre-established, including study design, analysis plan, and verification blinding procedures. Laboratory and data analysts were unaware of the assignment of case, control, and GA data to the verification subjects. Predictor scores, calculated as described below, were assigned to all subjects by an internally unaware statistician. Case, control, and GA data linked to predictor scores via DCO were provided to independent external statisticians for analysis. AUROC results were then transferred back to the DCO. Data transfer utilized the SUMPRODUCT function in Excel (Microsoft Excel 2013) to ensure data integrity. To provide a review trail from participant data to the final review results, digital timestamps were applied to the analyzed data, planning, and reporting.

[0343] Research design for discovery and review

[0344] A total of 86 and 50 subjects, respectively, were identified and reviewed for sPTB cases, with GA (GABD) collected at blood draws between weeks 17 0 / 7 and 28 6 / 7. The subjects used in the identification and review were completely independent of each other and unrelated to those used in the validation. In this embodiment, matched controls were identified for sPTB cases in the identification and review processes described above.

[0345] Incidence analysis

[0346] Following discovery, review, and validation analyses, other full-term controls not used in previous studies were selected from the PAPR database and processed in the laboratory using the MRM-MS assays applied in validation and as described above in this embodiment. Using the sampling package in R Statistical software (version 3.0.3) (Team RC.R: a Language and Environment for Statistical Computing. Vienna, Austria; 2014-2015; Matei A, Tillé Y. The R “sampling” package. European Conference on Quality in Survey Statistics. 2006), 187 subjects were randomly selected from the validated GA blood draw interval and analyzed using univariate statistical analysis (X). 2 The test was used to compare gestational age at birth (GAB) data from the 2012 National Vital Statistics Reports (NVSR). Martin et al., Final Data for 2012. National Vital Statistics Reports. 2014; 63(09): 1–86. Then, a control group that most closely resembled the delivery distribution in the 2012 NVSR was selected for comparison against the BMI distribution in the PAPR study as a whole, based on the best p-value (close to 1, with the minimum acceptable value being 0.950). Univariate statistical analysis (X-test) was used. 2 For BMI data from the PAPR study database, a control group was selected that best approximated the BMI distribution (closest to 1, minimum acceptable value 0.950) and the gestational age distribution in NVSR, and compared with the GABD of validated blood samples. The group best approximating all three distributions was selected as the subject group for the morbidity study. A total of 150 subjects were used to validate the GABD interval and BMI limits for predictor scores of morbidity, including those for review, validation, and morbidity. Figure 25 In this study, the composite dataset was used to obtain the best estimate of the confidence interval for the PPV curve. The confidence interval for PPV was calculated using the normal approximation of the binomial proportion error. Brown et al., Interval estimation for a binomial proportion. Statistical Science. 2001; 16(2):101–133.

[0347] Laboratory methods

[0348] Through iterative application of the following steps, a highly multi-pathway multiple reaction monitoring (MRM) mass spectrometry (MS) assay was generated using systems biology methods: literature management, targeted and non-targeted proteomics discovery, and miniature MRM-MS analysis of subject samples. A mature MRM-MS assay measuring 147 proteins was used in the discovery and verification studies. For all analyses, serum samples were processed in the laboratory as described above in this example. Aliquots of a pooled serum control (pHGS) were used to calculate the intra-batch analytical coefficient of variation (CV) for IBP4 and SHBG.

[0349] General predictor development strategies

[0350] The development strategy avoids overfitting and overcomes the dilution of expected biomarker performance across a wide gestational age range due to the dynamic nature of protein expression during pregnancy. The ratio of upregulated to downregulated analyte intensity was used in predictor development. This “reversal” is similar to the highest-scoring pair and 2-gene classifier strategy. (Geman et al., Classifying gene expression profiles from pair-wise mRNA comparisons. StatAppl Genet Mol Biol. 2004; 3(1):Article19; Price et al., Highly accurate two-gene classifier for differentiating gastrointestinal stromal tumors and leiomyosarcomas. Proc Natl Acad Sci USA. 2007; 104(9):3414–9). Because both proteins undergo the same pre-analysis and analytical processing steps in the “reversal,” this approach allows for amplification of diagnostic signals and self-normalization. As a strategy for normalizing peptide intensity measurements in complex proteomics workflows, reversal is also similar to a recently introduced approach known as “endogenic normalization (EPN).” (Li et al., An integrated quantification method to increase the precision, robustness, and resolution of protein measurement in human plasma samples. Clin Proteomics. 2015; 12(1):3; Li et al., A blood-based proteomic classifier for the molecular characterization of pulmonary nodules. Sci Transl Med. 2013; 5(207):207ra142). The number of candidate analytes used for model building was reduced through analytical criteria. Analytical filters included cutoffs for analytical precision, intensity, signs of interference, sample processing order dependence, and pre-analytical stability. The total number of analytes in any predictor was limited to a single reversal, thus avoiding complex mathematical models. The predictor score was defined as the natural log of the single reversal value, where the reversal itself is the response ratio (defined above in this embodiment). Finally, predictive performance was investigated in narrow-overlap 3-week intervals of pregnancy.

[0351] Receiver operating characteristic curve

[0352] In this embodiment, the AUROC value and associated p-value were calculated in reverse as described above. Bootstrap sampling was performed iteratively by selecting random sample groups with replacement, and the distribution and mean of the predictor AUROC in the merged discovery and verification groups were calculated. (Efron B, Tibshirani RJ. An Introduction to the Bootstrap. Boca Raton, Florida: Chapman and Hall / CRC Press; 1994.) The total number of samples selected at each iteration corresponds to the total number available in the initial mix.

[0353] Supplementary results

[0354] Table 3 summarizes the characteristics of subjects who were identified, reviewed, and validated. The percentage of subjects with one or more prior sPTB cases was higher in the identified sPTB cases than in the reviewed or validated cases, and other characteristics were generally consistent with those in the study.

[0355] Discovery and verification analysis

[0356] Forty-four proteins were upregulated or downregulated during the overlapping 3-week GA intervals and passed through the analysis filter. Figure 28 The reversal consists of the ratio of upregulated to downregulated protein and the predicted performance tested in samples from each of the overlapping 3-week GA intervals. Figure 29 The results show the performance of the reversal subgroup, which is illustrated by representative patterns. Fluctuations in performance are evident: IBP4 / SHBG and APOH / SHBG reversals exhibited better AUROC values ​​in the early window, while ITIH4 / BGH3 and PSG2 / BGH3 peaked in late pregnancy. Figure 24 Some reversals show consistent but modest performance throughout the gestational age range (PSG2 / PRG2). Figure 29 During the interval from 19 0 / 7 to 21 6 / 7, the best performing overall reversal IBP4 / SHBG had an AUROC of 0.74. Figure 29 When the pre-pregnancy BMI is <35 (kg / m²) 2 When dividing the subjects, the AUROC performance of the IBP4 / SHBG predictor increased to 0.79 (Table 4). This is due to its consistently strong performance in early pregnancy (i.e., 17 0 / 7 to 22 6 / 7 weeks gestational age). Figure 29 Based on the expected clinical applications, the IBP4 / SHBG predictor was selected for review analysis.

[0357] In the review sample, the blinded IBP4 / SHBG AUROC performance was 0.77 for all subjects and 0.79 for subjects categorized by BMI, which is in good agreement with the performance obtained in the findings (Table 5). After blind review, the findings and review samples were combined for bootstrap performance determination. A mean AUROC of 0.76 was obtained from 2,000 bootstrap iterations. Figure 30 ).

[0358] BMI validation analysis

[0359] The performance of the IBP4 / SHBG predictor was evaluated at several BMI cutoff values ​​in the validation samples (Table 5). This was achieved by removing extremely high (e.g., >37 kg / m²) predictors. 2 Or low BMI (e.g., ≤22kg / m²) 2 This moderately improved the performance of AUROC measurements. A combined cutoff value of 0.75 was provided (Table 5).

[0360] Example 3. Correlation between mass spectrometry and immunoassay data

[0361] This example shows the results of Myriad RBM screening for identifying IBP4 and other single biomarkers of sPTB in early, mid and late gestational age acquisition windows, (2) the correlation between MS and immunoassay results of SHBG / IBP4, and (3) clinical data related to SHBG as a biomarker of sPTB.

[0362] RBM data

[0363] Briefly, RBM measured 40 cases and 40 controls from PAPR (20 / 20 from the early window, 10 / 10 from the mid-term window, and 10 / 10 from the late window). RBM used Human Discovery MAP 250+ v2.0 (Myriad RBM, Austin, TX). The goal of these analyses was to develop multivariate models to predict PTB using multiple analytes. We used four modeling methods: random forest (rf), boosting, lasso, and logit. We performed a first round of variable selection, where the 15 variables best suited to each method were selected independently. From these 15 variables, the best analytes were selected independently for each of the four modeling methods using backward stepwise selection and area under the ROC curve (AUC) estimation using out-of-bag bootstrap samples. Table 6 shows the maximum number of samples for some multivariate models. Table 7 shows the analyte grading for the early window (GABD weeks 17–22) of different multivariate models. Table 8 shows the analyte grading for the mid-term window (GABD weeks 23-25) using different multivariate models. Table 9 shows the analyte grading for the late-term window (GABD weeks 26-28) using different multivariate models.

[0364] Identify commercially available ELISA kits related to mass spectrometry data.

[0365] Briefly, the comparison of ELISA versus MS included multiple studies using PAPR samples and ranging in size from 30 to 40 subjects. Each ELISA was performed according to the manufacturer's specifications. The predicted concentration of each analyte obtained by the ELISA was then compared with the relative concentration obtained by MS from the same samples. Person's R correlation value was then generated for comparison. Table 10 provides epitope and clonal information for the kits tested for analytes IBP4_HUMAN and SHBG_HUMAN. Table 11 shows that not all ELISA kits are MS-related, even for proteins in which a correlation exists. See, for example: IBP4, CHL1, ANGT, PAPP1.

[0366] 120 previously frozen serum samples with known outcomes from the PAPR study were selected for comparison between ELISA and MS assays. These samples had a gestational age (GABD) between 119 and 180 days at the time of blood collection. Samples were not excluded due to maternal BMI. ELISA was performed on commercially available IBP4 (AL-126, ANSCH Labs Webster, Texas) and SHBG (DSHBG0B, R&D Systems Minneapolis, Minnesota) kits. Assays were performed according to the manufacturer's specifications. Internal standards were used for inter-plate normalization. Scores were calculated from ELISA concentration values ​​based on LN([IBP4] / [SHBG]) and by MS based on LN(IBP4RR / SHBGRR), where RR refers to the relative area of ​​the endogenous peptide to the SIS peptide peak. Scores from the two methods were compared in case vs. control separations (p-values ​​were derived from an unpaired t-test assuming equal standard deviations) (Figure 31).

[0367] Fifty-seven previously frozen serum samples with known outcomes from the PAPR study (19 sPTB cases and 38 full-term controls) were selected for comparison between ELISA and MS assays. These samples had a gestational age (GABD) between 133 and 148 days at the time of blood collection. ELISA was performed on commercially available IBP4 (AL-126, ANSCH Labs Webster, Texas) and SHBG (DSHBG0B, R&D Systems Minneapolis, Minnesota) kits. Assays were performed according to the manufacturer's specifications. Samples run on different plates were normalized using internal standards. Scores were calculated from ELISA concentration values ​​based on LN([IBP4] / [SHBG]) and from MS based on LN(IBP4RR / SHBGRR), where RR refers to the relative area of ​​the endogenous peptide to the SIS peptide peak. Immunoassay performance was then determined by the area under the receiving working curve (AUC) and compared with the AUC from MS sources of the same sample groups (Figure 32). After applying BMI segmentation (BMI>22≤37) to the samples, resulting in a total of 34 samples (13 sPTB cases and 21 full-term controls), the AUC values ​​were also determined (Figure 33).

[0368] Sixty previously frozen serum samples with known outcomes from the PAPR study were analyzed by ELISA and MS. These samples predicted gestational age at draw (GABD) between 133 and 146 days. Correlation analyses were performed on samples with all BMIs (Figure 34, right inset) or subgroups of samples with BMIs >22 or ≤37 (Figure 34, left inset). ELISA was performed on commercially available IBP4 (AL-126, ANSCH Labs Webster, Texas) and SHBG (DSHBG0B, R&D Systems Minneapolis, Minnesota) kits. Assays were performed according to the manufacturer's specifications. Internal standards were used for inter-plate normalization. Scores were calculated by MS from ELISA concentration values ​​based on LN([IBP4] / [SHBG]) and LN(IBP4RR / SHBGRR), where RR refers to the relative peak area of ​​the endogenous peptide to the SIS peptide. Scores from the two methods were compared by correlation and case-to-control separation (p-values ​​were derived from an unpaired t-test assuming equal standard deviations). Table 12 shows the ELISA kits for IBP4 and SHBG demonstrating spTB separation relative to control (univariate).

[0369] Comparison of SHBG measurements by mass spectrometry and clinical analyzer

[0370] Thirty-five samples from individual subjects and mixtures of pregnant and non-pregnant serum were analyzed simultaneously at Sera Prognostics and two independent reference laboratories, ARUP Laboratories and Intermountain Laboratory Services. Aliquots were transported frozen to each laboratory, and transport was coordinated so that testing would begin on the same day at all three laboratories. ARUP used a Roche cobas e602 analyzer, and Intermountain used an Abbott Architect CMIA, both semi-automated immunoassay tools. Sera Prognostics employed a unique proteomics analysis approach that included immune removal of samples, enzymatic digestion, and analysis on an Agilent 6490 mass spectrometer. ARUP and IHC results are reported in nmol / L, while Sera used the relative ratio (RR) of heavy and light peptide substitutes. Data from ARUP and Intermountain were compared to each other to determine accuracy. Figure 39 Linearity and accuracy are well matched across a wide range of results, with a linearity slope of 1.032 and r. 2 The value was 0.990. Then, the data from each reference laboratory were compared with Sera's RR and linear regression plots. Figure 37 and38 The data shows good comparison with Sera's results, where ARUP's r... 2 The value is 0.937, and the r of Intermountain is... 2 The value is 0.934.

[0371] Example 4. PreTRM TM SNPs, insertions, deletions, and structural variants within IBP4 and SHBG peptides

[0372] This example demonstrates PreTRM. TM Known SNPs, insertions and deletions (indels), and structural variants within IBP4 and SHBG peptides.

[0373] Tables 13 and 14 describe PreTRM in detail. TM Known SNPs, insertions, deletions (indels), and structural variants within the IBP4 and SHBG peptides. This information is derived from the Single Nucleotide Polymorphism Database (dbSNP) Build 146. The single missense variation (G>C) A179P (dbSNP id: rs115336700) in SHBG has the highest overall allele frequency of 0.0048. Although this allele frequency is low, it is significantly higher in some subpopulations studied in the 1000 Genomes Project. These groups (allele frequencies) are: Americans of African ancestry from SW, USA (0.0492); African Caribbean from Barbados (0.0313); Yoruba from Ibadan, Nigeria (0.0278); Luhia from Webuye, Kenya (0.0101); Esan from Nigeria (0.0101); Colombians from Medellín, Colombia (0.0053); and Gambians from the western region of Gambia (0.0044). All other subgroups studied showed no change at this nucleotide position. The header includes cluster number (dbSNP rs number), heterozygosity - average heterozygosity, validation - validation method (or blank if no validation is available), MAF - minor allele frequency, function - functional characteristics of the polymorphism, dbSNP alleles - allele nucleotide identity, protein residues - residues produced by the alleles, codon position - position in the codon, NP_001031.2 amino acid position - amino acid position in the reference sequence NP_001031.2, and NM_001040.2 mRNA position - nucleotide position in the reference sequence NM_001040.2.

[0374] Example 5. IBP4 / SHBG reversed and amplified the diagnostic signal of spTB and reduced analytical variability.

[0375] This example demonstrates the amplification and reduction of variability in diagnostic signals obtained using the IBP4 / SHBG reversal strategy.

[0376] The levels of IBP4 and SHBG, determined by MS, were shown in the specified gestational age range for sPTB cases and full-term controls, respectively. Figure 44 and Figure 45 The curves were generated by mean smoothing of the relative ratios of the peptides (endogenous peptide peak area relative to the corresponding SIS peak area). For IBP4 and SHBG, the signal from cases to controls corresponded to a maximum difference of approximately 10%. When the score calculated as ln(IBP4RR / SHBGRR) was plotted, the signal amplification was evident (maximum difference of approximately 20%). Figure 46 These data demonstrate the amplification of diagnostic signals obtained using the IBP4 / SHBG reversal strategy.

[0377] Because each protein underwent the same analytical and pre-analytical processing steps, the ratio of the two protein levels could reduce variability. To test the effect of variability, CVs were determined for the RR of individual proteins (IBP4 and SHBG) and the ratio of IBP4 RR / SHBG RR in a pooled control serum sample from a pregnant donor (pHGS). Biologically indistinguishable pooled control samples were analyzed across multiple batches and over several days. Reversed variability was smaller than variability associated with individual proteins. Figure 48 )

[0378] To investigate whether reversal formation generally amplifies diagnostic signals, we examined the ROC performance (AUC) of high-performance reversals (AUC > 0.6) formed by ratios of multiple proteins. The top inset of Figure 47 shows the range of AUC values ​​using data from samples collected between 19 / 0 and 21 / 6 weeks of pregnancy (sPTB cases vs. term-pregnancy controls). Nearby box plots show the range of ROC performance for individual upregulated and downregulated proteins used to form the relevant reversals. Similarly, the p-values ​​for reversals derived from the Wilcoxon test (sPTB cases vs. term-pregnancy controls) were more significant than the p-values ​​for the corresponding individual proteins (Figure 47, bottom).

[0379] To investigate whether the formation of reversal more generally reduces variability, we examined the analytical variability of 72 different reversal values ​​(i.e., ratios of relative peak areas) relative to the analytical variability of individual proteins containing reversal in a pooled control serum sample from a pregnant donor (pHGS). Biologically indistinguishable pooled control samples were analyzed across multiple batches and over several days. The reversal variability was smaller than the variability associated with individual proteins. Figure 49 ).

[0380] The universality of reversal strategies is used to reduce analytical variability.

[0381] Figure 48 CVs calculated for pHGS samples (mixed pregnancy samples) analyzed in the laboratory over several batches, days, and with several instruments are reported. Since CVs were calculated using pHGS samples lacking biological differences, they correspond to a measure of analytical variability introduced in laboratory sample processing. The analytical variability associated with the ratio of 72 reversals is lower than the analytical variability of the relative peak areas of the individual upregulated and downregulated proteins used to form reversals. Figure 49 .

[0382] Example 6. Medically Specified PTB Analysis

[0383] This embodiment confirms that the classifier is sensitive to medically indicated PTB components based on conditions such as preeclampsia or gestational diabetes.

[0384] PreTRM was developed TM It was validated as a predictor of spontaneous PTB. In the United States, approximately 75% of all PTB is spontaneous, while the remainder is medically indicated PTB due to maternal or fetal complications (e.g., preeclampsia, intrauterine growth restriction, infection). Forty-one medically indicated PTB samples from the PAPR biobank were analyzed in the laboratory, and PreTRM was calculated. TM Score. PreTRM was compared with other indicators in subjects labeled as having medically defined preeclampsia. TM Scores. Subjects who delivered prematurely due to preeclampsia, as medically defined, had significantly higher scores than other subjects. Figure 50 ).

[0385] Figure 52 A heatmap showing the intensity of reversal with diabetes annotations is displayed. Red arrows indicate diabetic subjects. The sample is listed at the bottom, with PTB cases on the right side of the screen and full-term deliveries on the left. The diabetic patients are clustered on the right, demonstrating the potential to differentiate the reversal of gestational diabetes and thus potentially enabling the development of diagnostic tests based on biomarkers to predict gestational diabetes.

[0386] Example 7. Other transformations and peptides

[0387] Table 16 shows comparative IBP4 peptide and transformation MS data. Four different relabeled peptides (R*+10 Daltons) illustrate the various transformations of IBP4 that can be monitored to quantify IBP4 and their relative strengths. Those skilled in the art can potentially select any of these peptides or transformations, or other peptides or transformations not illustrated, to quantify IBP4.

[0388] Table 17 shows comparative IBP4 peptide and transformation MS data. IBP4 trypsin-hydrolyzed peptides derived from recombinant protein were analyzed by MRM-MS to identify candidate alternative peptides and their transformations. Those skilled in the art can potentially select any of these peptides or transformations, or other peptides or transformations not illustrated, to quantify IBP4. IBP4 was identified in the RBM (above), and then synthetic peptides were ordered to establish the assay.

[0389] Table 18 shows comparative SHBG peptide and transformation MS data. SHBG trypsin-hydrolyzed peptides derived from recombinant protein or mixed pregnant serum were analyzed by MRM-MS to identify candidate alternative peptides and their transformations. Those skilled in the art can potentially select any of these peptides or transformations, or other peptides or transformations not illustrated, to quantify SHBG. Isotype-specific peptides identified in serum are also shown.

[0390] Table 19 shows the altered serum levels of proteins in PTB samples during GA at 17–25 weeks. *Other proteins in PTB are limited to GA at 19–21 weeks. LC-MS (MRM) determination and analysis of 148 proteins from multiple pathways were performed on serum samples from 312 women (104 sPTB cases, 208 full-term controls) at GA 17–25 weeks. MRM peak area data were analyzed by hierarchical clustering, t-tests, and relationships with GA. After analysis filtering, 25 proteins showed significant differences (p<0.05) between sPTB and full-term subjects (Table 1). Throughout the GA range, 14 proteins were elevated and 3 were lower in sPTB samples. Other proteins were found to be dynamically regulated within sub-intervals of GA. For example, in GA 19–21 weeks, 7 other proteins were elevated and 1 was decreased in sPTB.

[0391] Table 20 lists 44 proteins that met the analytical filter criteria and were upregulated or downregulated in spTB relative to the full-term control.

[0392] Example 8. Understanding the mechanism of serum proteomic biomarkers indicating spontaneous preterm birth

[0393] This example demonstrates that biomarkers exhibit significantly different performance during GA as specific protein expression dynamically changes throughout pregnancy. Differentially expressed proteins play roles in steroid metabolism, placental development, immune tolerance, angiogenesis, and pregnancy maintenance. Figures 55, 57-59. The differences in these proteomic profiles seen in sPTB reflect developmental shifts in the damaged intraventricular space of the fetal / placental region during the second trimester.

[0394] Briefly, the research objective described in this embodiment is to understand the physiological basis of biomarkers associated with spontaneous preterm birth (sPTB) prediction.

[0395] Research Design

[0396] Pathways such as inflammation, infection, and bleeding have been involved in the etiology of preterm birth. However, little is known about which proteins in the blood are measurable and when they are destroyed during pregnancy. To answer these questions, we established LC-MS (MRM) assays for 148 proteins from multiple pathways and analyzed serum samples from 312 women (104 cases of spTB and 208 full-term controls) at gestational age (GA) 17–25 weeks.

[0397] Briefly, high-abundance proteins were removed from serum samples, and stable, heavily isotopically labeled standard (SIS) peptides were added to almost all proteins after trypsin hydrolysis. SIS peptides were used for normalization by generating response ratios, where the peak area (i.e., conversion) of the peptide fragment ion measured in serum was divided by the peak area of ​​the corresponding SIS conversion. The response ratios of the MRM peak area data were analyzed using hierarchical clustering, t-tests, and relationships with GA.

[0398] like Figure 53 As shown, multiple peptides correlate well with the same proteins. Discontinuous branches (grouped by color) correspond to identifiable functional classes, such as acute-phase proteins, defatted proteins, and known pregnancy-specific proteins. Important protein complexes in reproductive biology, such as PAPP1:PRG2, INHBE:INHBC, and IGF2:IBP3:ALS, are clearly identified. These quality assessments and highlighted relationships validate the highly multi-channel MRM-MS assays described in this application for exploring pregnancy biology and identifying analytes indicating spTB.

[0399] Figure 54 shows differentially expressed proteins that play a role in extracellular matrix interactions. TENX activates potential TGF-β and localizes to the fetal and maternal matrix at the transition point of cytotrophoblast differentiation. Alcaraz, L. et al., 2014 J. Cell Biol. 205(3) 409–428; Damsky, C. et al., 1992 J. Clin. Invest. 89(1) 210–222. Decreased serum TENX levels in sPTB indicate vascular defects or reduced TGF-β activity in the placenta. NCAM1 (CD56) is highly expressed on neurons and natural killer cells. NCAM1 is also expressed by intravascular trophoblasts, but is reduced or absent in PE placentas. Red-Horse, K. et al., 2004 J. Clin. Invest. 114: 744–754. Inverted serum NCAM1 levels in sPTB cases may reflect poor spiral artery remodeling and / or defective immune regulation. CHL1 is homologous to NCAM1 and directs integrin-mediated cell migration. BGH3 (TGFBI) is a cell adhesion molecule expressed in vascular endothelial cells and inhibits angiogenesis through a specific interaction with αv / β3 integrins. Son, HN. et al., 2013 Biochimica et Biophysica Acta 1833(10)2378–2388. Elevated TGFBI in sPTB cases may indicate reduced placental angiogenesis.

[0400] Figure 55 shows a kinetic diagram of differentially expressed proteins that play a role in the IGF-2 pathway, exhibiting maximum separation at 18 weeks. In early pregnancy, IGF2 stimulates proliferation, differentiation, and endometrial invasion of extravillous trophoblasts. IBP4 binds to and regulates the bioavailability of IGF2 at the maternal-fetal junction. Elevated IBP4 and decreased IGF2 in the first trimester are associated with IUGR and SGA, respectively. Qiu, Q. et al., 2012 J.Clin.Endocrino.l Metab.97(8):E1429-39; Demetriou, C. et al., 2014 PLOS 9(1):e85454. PAPP1 is a placenta-specific protease that cleaves IBP4 and releases active IGF2. Low serum PAPP1 levels in early pregnancy are associated with IUGR, PE, and PTB. Huynh, L. et al., 2014 Canadian Family Physician 60(10)899-903. PRG2 (proMBP) is expressed in the placenta and covalently binds to and inactivates PAPP1. The PRG2:PAPP1 inactivation complex circulates in maternal serum. Huynh, L et al., 2014 Canadian Family Physician 60(10) 899-903. The disrupted pathway regulation is consistent with the impaired IGF2 activity in sPTB cases that may lead to abnormal placental formation. Figure 56A This diagram illustrates the dynamic regulation and bioavailability of the aforementioned proteins during sPTB.

[0401] Figure 56B This diagram illustrates intracellular signaling preferentially activated by insulin bound to IR-B and by insulin bound to either IR-A or IGF1R. Belfiore and Malaguarnera, Endocrine-Related Cancer (2011) 18R125–R147. Activation of IR-A and IGF1R via insulin and IGF leads to growth dominance and proliferation signaling via phosphorylation of IRS1 / 2 and Shc proteins. Shc activation leads to recruitment of the Grb2 / Sos complex, and subsequent activation of Ras / Raf / MEK1 and Erk1 / 2. This latter kinase translocates to the nucleus and induces transcription of several genes involved in cell proliferation and survival. Phosphorylation of IRS1 / 2 induces activation of the PI3K / PDK1 / AKT pathway. In addition to its role in metabolic effects, AKT also leads to activation of effector factors involved in the control of apoptosis and survival (BAD, Mdm2, FKHR, NFkB, and JNK) as well as protein synthesis and cell growth (mTOR).

[0402] Figure 57 shows a kinetic diagram of differentially expressed proteins that play a role in metabolic hormone homeostasis. Sex hormone-binding globulin (SHBG) is a placental protein that increases during pregnancy and determines the bioavailability and metabolism of sex steroid hormones. Decreased SHBG levels lead to higher levels of free androgens and estrogens. Free androgens can be converted to estrogens via placental aromatase activity. Progesterone opposing activity of estrogens promotes pregnancy / delivery. Thyroxine-binding globulin (THBG) is estrogen-induced and increases ~2.5-fold in mid-pregnancy. Elevated serum THBG levels in sPTB cases can lead to decreased free thyroid hormones. Hypothyroidism during pregnancy is associated with an increased risk of miscarriage and preterm birth. Stagnaro-Green A. and Pearce E. 2012 Nat. Rev. Endocrinol. 8(11):650-8. In mid-pregnancy, estrogen increases angiotensinogen ~3-fold to stimulate a ~40% increase in plasma volume. Upregulation of ANGT can lead to gestational hypertension, a condition associated with an increased risk of spTB.

[0403] Figure 58 shows the kinetics of differentially expressed proteins that play a role in angiogenesis. TIE1 is an inhibitory co-receptor to the TIE2 angiogenesis receptor, which blocks the function of Ang-2 to stimulate angiogenesis. Seegar, T. et al., 2010 Mol. Cell. 37(5): 643–655. Pigment epithelial cell-derived factor (PEDF) is an anti-angiogenic factor expressed in the placenta, which stimulates the cleavage and inactivation of VEGFR-1 by γ-secretase. Cathepsin D (CATD) cleaves prolactin to produce angioinhibitory angiogenesis. Elevated serum CATD and angioinhibitory are associated with preeclampsia. Nakajima, R et al., 2015 Hypertension Research 38, 899-901. The leucine-rich α-2-glycoprotein (LRG1 / A2GL) promotes TGF-β signaling by binding to the co-receptor endoglin. TGF-β activates endothelial cell mitosis and angiogenesis via the Smad1 / 5 / 8 signaling pathway. Wang, X et al., 2013 Nature 499(7458). PSG3 induces anti-inflammatory cytokines from monocytes and macrophages and stimulates angiogenesis by binding to TGF-β. Low PSG levels are associated with IUGR. Moore, T., and Dveksler, G. 2014 Int. J. Dev. Biol. 58:273-280. ENPP2 (autotaxin) is an extracellular enzyme with lysophospholipase D activity that produces lysophosphatidic acid (LPA). LPA acts on placental receptors to stimulate angiogenesis and chemotaxis in NK cells and monocytes. Autotaxin levels are decreased in PIH and early-onset PE cases. Chen, SU et al., 2010 Endocrinology 151(1):369–379.

[0404] Figure 59 shows the kinetics of differentially expressed proteins that play a role in innate immunity. LBP delivers bacterial LPS to Toll-like receptor-4 via its co-receptor CD14 to elicit an inflammatory response in the innate immune pathway. Fetoglobulin-A (α-2-HS-glycoprotein) is a carrier protein of fatty acids in the blood, and the FetA-FA complex can bind to and activate the TLR4 receptor. Pal, D et al., 2012 Nature Med. 18(8):1279-85.

[0405] Figure 60 shows a kinetic diagram of differentially expressed proteins that play a role in blood clotting.

[0406] Figure 61 shows the kinetics of differentially expressed serum / secreted proteins.

[0407] Figure 62 shows the kinetics of differentially expressed PSGs / IBPs.

[0408] Figure 63 shows the kinetics of differentially expressed ECM / cell surface proteins.

[0409] Figure 64 shows the kinetics of differentially expressed complement / acute phase protein-1.

[0410] Figure 65 shows the kinetics of differentially expressed complement / acute phase protein-2.

[0411] Figure 66 shows the kinetics of differentially expressed complement / acute phase protein-3.

[0412] Figure 67 shows the kinetics of differentially expressed complement / acute phase protein-4.

[0413] Example 9. Kinetic Analysis of SDT4 / SV4

[0414] This embodiment provides kinetic analysis of all the analytes as illustrated at the beginning of Example 1, with data from 17 weeks 0 days to 28 weeks 6 days.

[0415] For Figures 68-85, the mean relative ratio of each peptide conversion was plotted using the R ggplot2 package for GABD, with the mean smoothing function applied to the mean (window = + / - 10 days). The figures show separate plots of cases versus controls using two different gestational age cutoffs at birth (<370 / 7 vs>=37 0 / 7 weeks and <35 0 / 7 vs>=35 0 / 7 weeks). The figure titles display the protein abbreviation, underline, and peptide sequence. Analyte sequences can be adjusted for the title to fit the figure.

[0416] The kinetic analyses illustrated in this article serve several purposes. These analyses indicate whether and in what direction analyte levels change during pregnancy, whether the changes differ between cases and controls, and reveal diagnostic differences related to gestational age. In some cases, the diagnostic signal lies within a narrow gestational age range and increases or decreases over time. The shape of the kinetic plot also provides visual guidance for selecting proteins that pair well during reversal.

[0417] Analytes showing significant separation of cases from controls in the early window, such as sample collection between 18 and 20 weeks of gestation, include (e.g.) AFAM, B2M, CATD, CAH1, C1QB, C1S, F13A, GELS, FETUA, HEMO, LBP, PEDF, PEPD, PLMN, PRG2, SHBG, TENX, THRB, and VCAM1. Analytes showing significant separation of cases from controls in the late window, such as sample collection between 26 and 28 weeks of gestation, include (e.g.) ITIH4, HEP2, IBP3, IGF2, KNG1, PSG11, PZP, VASN, and VTDB. Using a cutoff value of less than 35 0 / 7 weeks vs. greater than or equal to 35 0 / 7 weeks, relative to a cutoff value of less than 37 0 / 7 weeks vs. greater than or equal to 37 0 / 7 weeks, the separation of cases relative to controls was improved, as seen with respect to the following analytes, including (e.g.) AFAM, APOH, CAH1, CATD, CD14, CLUS, CRIS3, F13B, IBP6, ITIH4, LYAM1, PGRP2, PRDX, PSG2, PTGDS, SHBG, and SPRL1. It has been found that using a smaller gestational cutoff value at birth resulted in improved separation of various inflammatory and immune regulatory molecules. Those skilled in the art will understand that any analyte shown in the figures that exhibited significant separation between cases and controls for a given time window is a candidate, either as a single biomarker or as part of a group of biomarkers for use in the reversal pairs of the present invention.

[0418] Finally, according to the method of the present invention, for cases lacking differences relative to controls, but which show changes in analyte intensity during pregnancy, kinetic plots of analytes are useful in the pregnancy clock. These analytes are also referred to herein as "clock proteins," which can be used to determine the date of pregnancy in the absence of other date-determining methods (e.g., date of last menstrual period, date determined by ultrasound) or in combination with other date-determining methods. Table 60 provides a list of clock proteins useful in the pregnancy clock of the present invention.

[0419] Example 10. Discovery and Analysis of SPTB Cases

[0420] This embodiment describes an analysis of all previously analyzed sPTB cases as described in previous embodiments, their matched controls (two per case), and two new controls. This analysis, as described in this embodiment, expands the commercial blood sampling window beyond 19 and 20 weeks, generating additional data for predicting sPTB <35 weeks based on a large sample size from all previous embodiments, leading to the discovery of new analytes and reversals, defining molecular clock proteins, clarifying risk thresholds, and forming accurate validation claims for future clinical studies.

[0421] Sample processing methods

[0422] Establish standard procedures for conducting clinical studies of proteomics evaluation for controlling the risk of preterm birth (PAPR). These procedures also specify that samples and clinical information can be used to study other pregnancy complications. Samples are obtained from women at 11 Internal Review Board (IRB) approved sites throughout the United States. Serum and plasma samples, along with relevant information on patient demographics, past medical and pregnancy history, current pregnancy, and concurrent drug treatments, are obtained after informed consent is obtained. Postpartum, data related to maternal and infant conditions and complications are collected. Serum and plasma samples are processed according to procedures requiring standardized cryogenic centrifugation, aliquoting of samples into 0.5 mL 2-D barcode vials, and subsequent freezing at -80°C.

[0423] After delivery, preterm birth cases were reviewed individually to determine whether they were spontaneous or medically induced preterm births. Only spontaneously preterm birth cases were used for this analysis. For the discovery of preterm birth biomarkers, LC-MS data were generated for 413 samples (82 spTB cases and 331 term controls) between gestational ages of 17 0 / 7 and 21 6 / 7 weeks, and each preterm sample was matched with four term controls by gestational age at blood collection. Except for one day, each gestational age day between 17 0 / 7 and 21 6 / 7 weeks included at least one spTB case (and a matched term control). Four term controls whose blood was drawn on that day were selected. One term control whose laboratory analysis failed was not reanalyzed in this study.

[0424] Subsequently, the high-abundance proteins in the serum samples were removed using the Human 14 Multiple Affinity Removal System (MARS-14), which removed the 14 most abundant proteins. Equal volumes of each clinical sample or mixture of two quality control serums were diluted with column buffer for experimental replicates and filtered to remove precipitates. Filtered samples were removed using a MARS-14 column (4.6 × 100 mm, Agilent Technologies). Samples were cooled to 4°C in an autosampler, the removal column was run at room temperature, and the collected fractions were held at 4°C until further analysis. Unbound fractions were collected for further analysis.

[0425] The removed serum sample was reduced with dithiothreitol, alkylated with iodoacetamide, and then hydrolyzed with trypsin. After trypsin hydrolysis, a mixture of stable isotope standards at a concentration approximately equal to that of the surrogate peptide analyte was added to the sample. The samples with added SIS were mixed and aliquoted into two equal volumes. Each aliquot was stored at -80°C until ready to continue the procedure. One frozen aliquot from each sample was removed from the -80°C storage, thawed, and then desalted on a C18 solid-phase extraction plate (Empore, 3M). The eluted peptides were lyophilized to dryness. The lyophilized sample was redissolved in a reconstitution solution (IS Recon) containing an internal standard for monitoring only the quality of the LC-MS step.

[0426] Fully processed samples were analyzed using dynamic multiple reaction monitoring (dMRM). Peptides were separated using an Agilent 1290 UPLC at a flow rate of 0.4 mL / min on a 2.1 × 100 mm Poroshell EC-C18 column with 2.7 μm particle size, and eluted using an acetonitrile gradient to an Agilent 6490 triple quadrupole mass spectrometer with an electrospray source operating in cation mode. dMRM measurements determined 442 transformations corresponding to 119 peptides and 77 proteins, which played a diagnostic and mass role. Peak integration was performed using MassHunter quantitative analysis software (Agilent Technologies). The ratio of the peak area of ​​the surrogate peptide analyte to the peak area of ​​the corresponding SIS is reported.

[0427] Table 21 shows a summary of the transformations, SIS transformations, and ISRecon standards for proteins, peptides, and serum analytes measured in the dMRM method. MARS-14-removed proteins identified the analytes targeted by the MARS-14 immunoremoval column and were measured for quality control purposes. Quantitative transitions were used for relative response ratios, and qualitative transitions were used for quality control purposes. An asterisk (*) indicates a name change. CSH indicates that the peptide corresponds to both CSH1 and CSH2. Because the peptide is conserved in several HLA type I isotypes, HLAG is now referred to as HLACI. LYAM3 is now referred to as LYAM1 because although the peptide sequence is present in each, it is derived from LYAM1 only by trypsin cleavage. Because the peptide is specific for both SOM2 and CSH, SOM2 is now referred to as SOM2.CSH.

[0428] Significant proteins and reverse selection

[0429] For each analyte, in each of the two-week and three-week overlapping windows, with and without BMI limits and using two SPTB definitions (37 / 37 and 35 / 35), the fold change indicating whether the mean of the SPTB case sample was higher or lower than the mean of the TERM control sample was calculated. Tables 22 and 23 show the protein / conversion AUROC for the two-week gestational age windows with an overlap of one week (e.g., 119–132 refers to 119–132 days of gestation, which corresponds to 17 and 18 weeks of gestation). Performance in each two-week window was reported for two different case vs. control cutoff values ​​(<37 0 / 7 vs>=37 0 / 7, <35 0 / 7 vs>=35 0 / 7) and with and without (rBMI) BMI division. Tables 24 and 25 show the protein / conversion AUROC for overlapping two-week three-week gestational windows (expressed in days, e.g., "119-139" refers to days 119-139 of gestation, which corresponds to weeks 17, 18, and 19 of gestation). Performance within each three-week window was reported for two different case vs. control cutoff values ​​(<37 0 / 7 vs>=37 0 / 7, <35 0 / 7 vs>=35 0 / 7) and for BMI division using (rBMI) and not using (aBMI).

[0430] Figures 86–95 show kinetic plots of multiple peptide transitions in cases relative to controls using a cutoff of gestational age at birth of <37 0 / 7 vs> = 37 0 / 7 weeks. Figures 96–105 show kinetic plots of multiple peptide transitions in cases relative to controls using a cutoff of gestational age at birth of <35 0 / 7 vs> = 35 0 / 7 weeks. Briefly, the mean relative ratio of each peptide transition was plotted against GABD using the R ggplot2 package with the mean smoothing function (window = + / - 10 days). The figures show separate plots of cases relative to controls using two different cutoffs of gestational age at birth (<37 0 / 7 vs> = 37 0 / 7 weeks and <35 0 / 7 vs> = 35 0 / 7 weeks). The figure titles display the protein abbreviation, underline, and peptide sequence. Analyte sequences can be adjusted for the title to fit the figure.

[0431] Based on the fold change indicating whether the SPTB case sample mean is greater than or less than the TERM control sample mean, each analyte is labeled as upregulated or downregulated for each combination (i.e., overlapping 2- or 3-week windows, BMI limitation, and SPTB definition), and an analyte is designated as an overall upregulated analyte if it has a majority of the combinations labeled as upregulated, and vice versa. This is shown in Table 26.

[0432] Based on these upregulation and downregulation allocations (55 upregulations and 30 downregulations), reversals were generated by dividing the relative ratio of each upregulating analyte by the relative ratio of the downregulating analyte and taking the natural logarithm of the result. This resulted in 1650 reversals (55 × 30 = 1650). For each reversal, the area under the ROC curve (AUCROC) representing the separation of SPTB and TERM was calculated, along with a p-value indicating whether the AUCROC value was significantly different from AUCROC = 0.5 (i.e., no significant separation of SPTB and TERM). For different conditions (e.g., pregnancy window, with and without BMI restrictions, and two SPTB cutoff values), the performance of each reversal with AUCROC > 0.6 and p-value < 0.05 is listed in a table. Tables 27–42 show the reversal classification performance at 17 and 18 weeks of gestation. Tables 47–58 show the reversal classification performance at 17, 18, and 19 weeks of gestation. Tables 43–46 show the reversal classification performance at 17–21 weeks of gestation. Table 59 shows other reversals with potential significance.

[0433] The improved performance of predictors formed by more than one reversal (17–21 weeks) is also shown. Briefly, combinations of reversals that provide strong predictive performance at early (e.g., 17–19 weeks) or late (e.g., 19–21 weeks) gestational ages are evaluated, and the performance of predictors formed by multiple reversal combinations (SumLog) is assessed across the entire blood draw range. This is shown in Table 61. Predictor scores are derived from the sum of individual reversal log values ​​(SumLog), but those skilled in the art may choose other models (e.g., logistic regression). This multiple reversal approach is also considered for application to reversal combinations specific to preterm premature rupture of membranes (PPROM) versus non-PPROM preterm labor (PTL), fetal sex, and conception. It is also considered that predictors may contain indicator variables that take into account knowledge of blood draw date, fetal sex, or conception, selecting the reversal subgroup to be used.

[0434] Figure 110 The relationship between predictor score (lnIBP4 / SHBG) and morbidity-modified relative risk of spTB (positive predictive value) is shown using a cutoff of <37 0 / 7 weeks versus >=37 0 / 7 weeks of pregnancy. Samples were drawn with a BMI >22 and <=37 between 19 1 / 7 weeks and 20 6 / 7 weeks. As the predictor score increased, the relative risk increased from a background rate of 7.3% (the mean population risk of spTB in singleton pregnancies) to approximately 50%. Screening positivity rate curves were added for all score thresholds. Confidence intervals (grey shading) were calculated assuming a binomial distribution of observations and approximating the error distribution with a normal distribution. The sample distribution divided by classifier scores is shown in a bar chart according to the color scheme in the legend.

[0435] Figure 111 The relationship between predictor score (lnIBP4 / SHBG) and morbidity-modified relative risk of spTB (positive predictive value) is shown using a cutoff of <35 0 / 7 weeks versus >=35 0 / 7 weeks of pregnancy. Samples were drawn between 19 1 / 7 weeks and 20 6 / 7 weeks. As the predictor score increased, the relative risk increased from a background rate of 4.4% (the average population risk of spTB (<35) in singleton pregnancies) to approximately 50%. Screening positivity rate curves were added for all score thresholds. Confidence intervals (grey shading) were calculated assuming a binomial distribution of observations and approximating the error distribution with a normal distribution. The sample distribution divided by classifier scores is shown in a bar chart according to the color scheme in the legend.

[0436] Clinical observation: spTB, PPROM, and PTL

[0437] Reversal performance (GABD weeks 17–21) was independently evaluated for the two distinct phenotypes of sPTB, PPROM, and PTL. PPROM occurred more frequently in the early stages and was associated with infection or inflammation. PTL could occur later and was generally considered a less severe phenotype. For PPROM, more significant and better-performing reversals were observed, and these reversals were composed of proteins known to be involved in inflammation and infection. The choice of reversal was considered to establish independent assays for PPROM and PTL, or to maximize overall performance through a combination of more than one type of reversal among single predictors. In the analyses shown in Tables 61–64, an AUC > 0.65 and p < 0.05 were required for either PPROM or PTL.

[0438] Table 61 shows the cutoff values ​​for case vs. control with <37 0 / 7 vs ≥37 0 / 7 weeks of gestation, for PPROM and PTL, respectively, without BMI classification, for reversing AUROC from 17 0 / 7 to 21 6 / 7 weeks of gestation. Table 62 shows the cutoff values ​​for case vs. control with <37 0 / 7 vs >=37 0 / 7 weeks of gestation, for PPROM and PTL, respectively, with BMI classification (>22<=37). Table 63 shows the cutoff values ​​for case vs. control with <35 0 / 7 vs >=35 0 / 7 weeks of gestation, for PPROM and PTL, respectively, without BMI classification, for reversing AUROC from 17 0 / 7 to 21 6 / 7 weeks of gestation. Table 64 shows the cutoff values ​​for case vs. control at <35 0 / 7 vs>=35 0 / 7 weeks of gestation, with BMI division (>22<=37), for PPROM and PTL, respectively, for reversal of AUROC from 17 0 / 7 to 21 6 / 7 weeks of gestation.

[0439] For GABD at 19-20 weeks, the best-performing analytes for PTL and PPROM were identified, and some reversals were constructed from the best-performing analytes. IBP4 was found to be a good analyte for both PTL and PPROM, thus enabling its general application to sPTB. Table 76 lists the case-vs-control cutoff values ​​for 19 1 / 7 to 20 6 / 7 weeks of gestation for PTL without BMI classification, using <37 0 / 7vs> = 37 0 / 7 weeks of gestation. Table 77 lists the case-vs-control cutoff values ​​for 19 1 / 7 to 20 6 / 7 weeks of gestation for PPROM without BMI classification, using <37 0 / 7vs> = 37 0 / 7 weeks of gestation. Figure 108 illustrates reversals with good performance at 19-20 weeks of gestation in PTL. Figure 109 illustrates reversals with good performance at 19-20 weeks of gestation in PPROM.

[0440] Clinical observation: primiparous and multiparous women

[0441] For the two distinct phenotypes of sPTB, primiparous, and multiparous women, reversal performance was further evaluated independently (17–21 weeks). In Tables 65–68, the best-performing reversals (17–21 weeks) were shown for primiparous and multiparous subjects, respectively. Primiparous women were most in need of testing to predict PTB probability because they lacked a pregnancy history for physicians to determine / estimate risk. These results enabled independent testing for both groups, or combining the best-performing reversals in a single classifier to predict risk for both. In the analyses shown in Tables 65–68, AUC > 0.65 and p < 0.05 were required for primiparous or multiparous women.

[0442] Table 65 shows the case-control cutoff values ​​for <37 0 / 7 vs> = 37 0 / 7 weeks of gestation, without BMI segmentation, for primiparous and multiparous women, at 17 0 / 7 to 21 6 / 7 weeks of gestation. Table 66 shows the case-control cutoff values ​​for <37 0 / 7 vs> = 37 0 / 7 weeks of gestation, with BMI segmentation (>22 <= 37), for primiparous and multiparous women, at 17 0 / 7 to 21 6 / 7 weeks of gestation. Table 67 shows the case-control cutoff values ​​for <35 0 / 7 vs> = 35 0 / 7 weeks of gestation, without BMI segmentation, for primiparous and multiparous women, at 17 0 / 7 to 21 6 / 7 weeks of gestation. Table 68 shows the cutoff values ​​for case vs. control using <35 0 / 7 vs>=35 0 / 7 weeks of gestation, with BMI classification (>22<=37), for primiparous and multiparous women, for reversal AUROC from 170 / 7 to 21 6 / 7 weeks of gestation.

[0443] Clinical observation: fetal sex

[0444] Reversal performance was independently evaluated in subjects carrying male vs. female fetuses (17–21 weeks). Some reversals were found to have fetal sex-specific indicative performance. Figure 106 shows fetal sex-specific differences in IBP4 and SHBG analytes and scores (IBP4 / SHBG). IBP4 was significantly higher in subjects carrying male fetuses. For gestational ages of 19–21 weeks without BMI classification, reversal performance remained comparable (Figure 106). Additionally, in the PAPR clinical trial, male fetuses were found to have an increased risk of spTB (p=0.0002 with an odds ratio of 1.6). Finally, fetal sex can be incorporated into predictors (e.g., reversal value plus fetal sex). In the analyses shown in Tables 69–72, AUC > 0.69 and p < 0.05 were required for both male and female fetuses.

[0445] Table 69 shows the cutoff values ​​for case vs. control at <37 0 / 7 vs> = 37 0 / 7 weeks of gestation, for reversing AUROC at 17 0 / 7 to 21 6 / 7 weeks of gestation, solely by fetal sex, without BMI segmentation. Table 70 shows the cutoff values ​​for case vs. control at <37 0 / 7 vs> = 37 0 / 7 weeks of gestation, for reversing AUROC at 17 0 / 7 to 21 6 / 7 weeks of gestation, solely by fetal sex, with BMI segmentation (>22 <= 37). Table 71 shows the cutoff values ​​for case vs. control at <35 0 / 7 vs> = 35 0 / 7 weeks of gestation, for reversing AUROC at 17 0 / 7 to 21 6 / 7 weeks of gestation, solely by fetal sex, without BMI segmentation. Table 72 shows the cutoff values ​​for case vs. control at <35 0 / 7 vs>=35 0 / 7 weeks of gestation, using BMI classification (>22<=37), for reversal of AUROC at 17 0 / 7 to 21 6 / 7 weeks of gestation solely by fetal sex.

[0446] Example 11. Correlation between mass spectrometry and immunoassay data

[0447] This example demonstrates the implementation of an immunoassay using the MSD platform (e.g., for IBP4 and SHBG, MSD data are correlated with commercial ELISA data and MS data).

[0448] Material

[0449] The following antibodies were used: sex hormone-binding protein (Biospacific catalog numbers #s 6002-100051 and 6001-100050; R&D Systems catalog numbers #s MAB2656 and AF2656), and IGFBP-4 (Ansh catalog numbers #s AB-308-AI039 and AB-308-AI042). SHBG proteins from Origene (catalog number #TP328307), Biospacific (catalog number #J65200), NIBSC (number: 95 / 560), and R&D Systems (obtained only as part of the ELISA SHBG kit) were tested as calibrators. Recombinant human IGFBP-4 (Ansh, catalog number #AG-308-AI050) was used as a calibrator.

[0450] Produces a single U-PLEX-conjugated antibody solution

[0451] For a final volume ≥200 μL, dilute each biotinylated antibody to 10 μg / mL in Diluent 100. Then, add 300 μL of the designated U-PLEX linker to each biotinylated antibody. (Use a different linker for each biotinylated antibody). Vortex the sample and incubate at room temperature for 30 minutes. Add 200 μL of stop solution to each tube. Vortex the tubes and incubate at room temperature for 30 minutes.

[0452] Preparation of multi-channel coating solution

[0453] Combine each U-PLEX-conjugated antibody solution (600 μL) into one tube and vortex mix. When combining fewer than 10 antibodies, use a stop solution to bring the solution volume to 6 mL to achieve a final 1× concentration. Note that in these experiments, only a single antibody was present in each well.

[0454] Coating U-PLEX board.

[0455] Add 50 μL of multichannel coating solution to each well. Seal the plate with the adhesive plate seal and incubate at room temperature for 1 hour with shaking at approximately 700 rpm or overnight at 2–8 °C. After washing three times with at least 150 μL of 1×MSD ​​washing buffer, the plate is ready for use.

[0456] Sample Analysis

[0457] Add 50 μl of sample or standard to each well. Seal the plate and incubate at room temperature for 1 hour with shaking at approximately 700 rpm. Then, wash the plate three times with at least 150 μL of 1×MSD ​​wash buffer*. Add 50 μL of detection antibody solution to each well. Seal the plate and incubate at room temperature for 1 hour with shaking at approximately 700 rpm. Wash the plate three times with at least 150 μL of 1×MSD ​​wash buffer. After adding 150 μL of 2× read buffer to each well, immediately read the plate on the MSD instrument.

[0458] SHBG antibody and marker screening

[0459] In both capture-detector orientations, all antibodies were tested for all paired combinations. Capture antibodies were prepared at 10 μg / mL, coupled to U-PLEX adapters, and coated onto U-PLEX plates. SHBG R&D Systems calibrators were diluted in Diluent 43 to generate a 7-point standard curve (with assay diluent as a blank). Samples were diluted in Diluent 43 and tested in assays as follows: Sera SHBG “High” and “Low” samples: 100- and 500-fold dilutions, and Sera Pregnant mixtures: 100-, 200-, 400-, and 800-fold dilutions. Detection antibodies were tested at 1 μg / mL in Diluent 3. The standard curves and binding to native analytes in serum were evaluated. The optimal analyte pairs were then tested using NIBSC and Biospacific calibrators diluted as described above.

[0460] IGFBP-4 antibody and marker screening.

[0461] Two antibodies were tested in two capture-detector orientations. The capture antibody was prepared at 10 μg / mL, conjugated to a U-PLEX adapter, and coated onto a U-PLEX plate. The IGFBP-4 calibrator was diluted in Diluent 12 to generate a 7-point standard curve (with the assay diluent as a blank). Samples were diluted in Diluent 12 and tested in the assay as follows: Sera IGFBP-4 “high” and “low” samples: 5-fold, Sera Pregnant mixture: 2-fold dilutions from 2- to 64-fold, and two single-human serum samples (MSD samples): 2-, 4-, 8-, and 16-fold. The detection antibody was tested at 1 μg / mL in Diluent 12. The standard curve and binding to the native analyte in serum were evaluated.

[0462] SHBG and IGFBP-4 tests were performed on 60 Sera samples.

[0463] Antibody pair 12 was selected to measure SHBG in 60 plasma samples from Sera, repeated twice. For IGFBP-4, pair 2 was selected. For SHBG and IGFBP-4, plasma samples were diluted 1:1000 and 1:20, respectively. The results of the MSD ELISA were compared with those of a commercial ELISA kit and MS-MRM data.

[0464] result:

[0465] SHBG antibody screening

[0466] Only antibody pair 1 (R&D mono capture, poly detection) provided a strong signal with the Origene calibrator, indicating that this calibrator could represent the endogenous SHBG analyte subset. Therefore, other calibrators were tested in subsequent studies to identify those that worked with all pairs. Nevertheless, all antibody pairs recognized the natural analytes in mixed samples of Sera high, low, and pregnancy. R&Dpoly AF2656 and Biospacific mono 6001-100050 provided similar performance. Sample dilutions of pairs 2, 3, and 12 showed approximately linear titrations (Table 73). The performance of the four best antibody pairs was then tested with three other calibrators. Among the three calibrators, the four best pairs achieved good calibrator curves (Table 74). The differences in signal may be partly due to differences in the specified concentrations.

[0467] The inset plot at the bottom shows that the NIBSC or Biospacific signal differs relative to the R&D calibrator depending on the antibody pair. Pairs 3 and 10 (the same antibody with capture-detection orientation reversal) have similar spectra. For both NIBSC and Biospacific, pair 2 provides a lower signal (same capture compared to pair 3). Pair 12 provides a higher signal for Biospacific, and for the NIBSC standard, it provides more than 3 times the signal.

[0468] IGFBP-4 antibody screening

[0469] Compared to pair 1, the antibody pair 2 standard curve provided 4–6 times higher specificity calibrator signal and background (Table 75). For most of the dilutions tested, the signal for serum samples was within the linear range; the pregnancy mixture was close to background at 32- and 64-fold dilutions. Pair 2 samples provided ~12-fold higher signal, resulting in a 2–4-fold difference in quantification. For both pairs, the signal CV was typically <5%.

[0470] Measurement of SHBG and IGFBP-4 in 60 serum samples.

[0471] For SHBG, after 1000-fold dilution, the sample fell within the range of calibrator standards 1-3. The median measured concentration was 58.4 μg / mL. The CV for repeated measurements was low, with a median CV of 2.4%. The median measured concentration for IGFBP-4 was 234 ng / mL, and the median CV between repeated samples was 2.2%. As shown in Figure 107, good correlations between the two proteins were observed in the MSD assay compared to commercial ELISA kits and MS-MRM assays.

[0472] Based on the foregoing description, it is evident that changes and variations can be made to the invention described herein to suit various uses and conditions. These embodiments are also within the scope of the following claims.

[0473] In this document, the enumeration of elements in any definition of a variable includes the definition of the variable as any single element or combination (or sub-combination) of the listed elements. Similarly, the enumeration of implementations includes implementations as any single implementation or in combination with any other implementation or parts thereof.

[0474] All patents and patent disclosures mentioned in this specification are incorporated herein by reference to the same extent that each individual patent and patent disclosure is specifically and individually indicated and incorporated herein by reference.

[0475]

[0476]

[0477]

[0478]

[0479]

[0480]

[0481]

[0482] Table 4

[0483]

[0484] Table 5

[0485]

[0486]

[0487]

[0488]

[0489]

[0490]

[0491]

[0492]

[0493] Table 11: Indicate whether the kit is related to or not to the MS data.

[0494]

[0495] Table 12: Display of spTB vs control IBP4 and SHBG ELISA kits (univariate)

[0496]

[0497]

[0498]

[0499]

[0500]

[0501]

[0502]

[0503]

[0504]

[0505]

[0506]

[0507]

[0508]

[0509]

[0510]

[0511]

[0512]

[0513]

[0514]

[0515]

[0516]

[0517]

[0518]

[0519]

[0520]

[0521]

[0522]

[0523]

[0524]

[0525]

[0526]

[0527]

[0528]

[0529]

[0530]

[0531]

[0532]

[0533]

[0534]

[0535]

[0536]

[0537]

[0538]

[0539]

[0540]

[0541]

[0542]

[0543]

[0544]

[0545]

[0546] Table 19: Proteins with altered serum levels in PTB samples between 17 and 25 weeks of GA.

[0547] protein change Functional categories THRB Upward Coagulation / Acute Phase Reaction VTNC Upward Cell adhesion / acute phase response HEMO Upward Ferrous protoporphyrin transport / acute phase reaction FETUA Upward Inflammation / Acute Phase Response LBP Upward Innate immunity / acute phase reaction IBP4 Upward Growth factor regulation CD14 Upward Innate immunity HABP2 Upward Cell adhesion / migration INHBC Upward Growth factor regulation CFAB Upward Complement / Acute Phase Response ICAM1 Upward Cell adhesion / migration IC1 Upward Complement / Acute Phase Response APOH Upward Coagulation / Autoimmune B2MG Upward MHC / Immune C1S Increase* complement APOE Increase* Cholesterol metabolism APOC3 Increase* Triglyceride metabolism PEDF Increase* Angiogenesis CATD Increase* ECM remodeling / cell migration INHBE Increase* Growth factor regulation IBP6 Increase* Growth factor regulation PRG2 Lower Growth factor regulation SHBG Lower Inflammation / steroid metabolism GELS Lower Actin binding / acute phase response PSG4 Lower* Growth factor regulation

[0548] *Other proteins in PTB are limited to GA at weeks 19-21.

[0549] Table 20: The analysis filters were used to identify 44 proteins that were upregulated or downregulated in spTB vs. full-term control.

[0550]

[0551]

[0552] *Peptide substitutes for HLA-G are not unique for this protein.

[0553] Table 21

[0554]

[0555]

[0556]

[0557]

[0558]

[0559]

[0560]

[0561]

[0562]

[0563]

[0564]

[0565] * indicates a name change. CSH represents the peptide corresponding to both CSH1 and CSH2. HLAG is now called HLACI, as this peptide is conserved in some type I HLA isotypes. LYAM3 is now called LYAM1 because, although the peptide sequence is present in each, it is derived from LYAM1 only by trypsin cleavage. SOM2 is now called SOM2.CSH because the peptide is specific for both SOM2 and CSH.

[0566]

[0567]

[0568]

[0569]

[0570]

[0571]

[0572]

[0573]

[0574]

[0575]

[0576]

[0577]

[0578]

[0579]

[0580]

[0581] Table 26: Proteins / conversions used for reversing upregulation and downregulation

[0582]

[0583]

[0584]

[0585]

[0586]

[0587]

[0588]

[0589]

[0590]

[0591]

[0592]

[0593]

[0594]

[0595]

[0596]

[0597]

[0598]

[0599]

[0600]

[0601]

[0602]

[0603]

[0604]

[0605]

[0606]

[0607]

[0608]

[0609]

[0610]

[0611]

[0612]

[0613]

[0614]

[0615]

[0616]

[0617]

[0618]

[0619]

[0620]

[0621]

[0622]

[0623]

[0624]

[0625]

[0626]

[0627]

[0628]

[0629]

[0630]

[0631]

[0632]

[0633]

[0634]

[0635]

[0636]

[0637]

[0638]

[0639]

[0640]

[0641]

[0642]

[0643]

[0644]

[0645]

[0646]

[0647]

[0648]

[0649]

[0650]

[0651]

[0652]

[0653]

[0654]

[0655]

[0656]

[0657]

[0658]

[0659]

[0660]

[0661]

[0662]

[0663]

[0664]

[0665]

[0666]

[0667]

[0668]

[0669]

[0670]

[0671]

[0672]

[0673]

[0674]

[0675]

[0676]

[0677]

[0678]

[0679]

[0680]

[0681]

[0682]

[0683]

[0684]

[0685]

[0686]

[0687]

[0688]

[0689]

[0690]

[0691]

[0692]

[0693]

[0694]

[0695]

[0696]

[0697]

[0698]

[0699]

[0700]

[0701]

[0702]

[0703]

[0704]

[0705]

[0706]

[0707]

[0708]

[0709]

[0710]

[0711]

[0712]

[0713]

[0714]

[0715]

[0716]

[0717]

[0718]

[0719]

[0720]

[0721]

[0722]

[0723]

[0724]

[0725]

[0726]

[0727]

[0728]

[0729]

[0730]

[0731]

[0732]

[0733]

[0734]

[0735]

[0736]

[0737]

[0738]

[0739]

[0740]

[0741]

[0742]

[0743]

[0744]

[0745]

[0746]

[0747]

[0748]

[0749]

[0750]

[0751]

[0752]

[0753]

[0754]

[0755]

[0756]

[0757]

[0758]

[0759]

[0760]

[0761]

[0762]

[0763]

[0764]

[0765]

[0766]

[0767]

[0768]

[0769]

[0770] Table 59: Other reversals

[0771]

[0772]

[0773]

[0774]

[0775] Table 60: Clock Proteins

[0776]

[0777]

[0778]

[0779]

[0780]

[0781]

[0782]

[0783]

[0784]

[0785]

[0786]

[0787]

[0788]

[0789]

[0790]

[0791]

[0792]

[0793]

[0794]

[0795]

[0796]

[0797]

[0798]

[0799]

[0800]

[0801]

[0802]

[0803]

[0804]

[0805]

[0806]

[0807]

[0808]

[0809]

[0810]

[0811]

[0812]

[0813]

[0814]

[0815]

[0816]

[0817]

[0818]

[0819]

[0820]

[0821]

[0822]

[0823]

[0824]

[0825]

[0826]

[0827]

[0828]

[0829]

[0830]

[0831]

[0832]

[0833]

[0834]

[0835]

[0836]

[0837]

[0838]

[0839]

[0840]

[0841]

[0842]

[0843]

[0844]

[0845]

[0846]

[0847]

[0848]

[0849]

[0850]

[0851]

[0852] Table 74: Using other calibrators, the best-performing SHBG antibody was...

[0853]

[0854] Table 75: IGFBP-4 antibody screening

[0855]

[0856] Table 76: Transformation classification performance, weeks 19-20.

[0857] Using the case vs. control cutoff value of <37 0 / 7vs> = 37 0 / 7 weeks, without BMI classification, the conversion AUROC for PTL from 19 1 / 7 to 20 6 / 7 weeks of gestation.

[0858]

[0859] Table 77: Transformation classification performance, weeks 19-20.

[0860] Using the case vs. control cutoff value of <37 0 / 7vs> = 37 0 / 7 weeks, without BMI classification, the converted AUROC for PPROM at 19 1 / 7 to 20 6 / 7 weeks of gestation.

[0861]

[0862] sequence list <110> Sera Forecasting Company <120> Biomarkers used to predict preterm birth <130> 13271-018-228 <140> PCT / US2016 / 038198 <141> 2016-06-17 <150> 62 / 290,796 <151> 2016-02-03 <150> 62 / 387,420 <151> 2015-12-24 <150> 62 / 182,349 <151> 2015-06-19 <160> 161 <170> PatentIn version 3.5 <210> 1 <211> 8 <212> PRT <213> Homo sapiens <400> 1 Leu Pro Gly Gly Leu Glu Pro Lys 1 5 <210> 2 <211> 10 <212> PRT <213> Homo sapiens <400> 2 Gln Cys His Pro Ala Leu Asp Gly Gln Arg 1 5 10 <210> 3 <211> 15 <212> PRT <213> Homo sapiens <400> 3 Thr His Glu Asp Leu Tyr Ile Ile Pro Ile Pro Asn Cys Asp Arg 1 5 10 15 <210> 4 <211> 17 <212> PRT <213> Homo sapiens <400> 4 Glu Asp Ala Arg Pro Val Pro Gln Gly Ser Cys Gln Ser Glu Leu His 1 5 10 15 Arg <210> 5 <211> 12 <212> PRT <213> Homo sapiens <400> 5 Cys Arg Pro Pro Val Gly Cys Glu Glu Leu Val Arg 1 5 10 <210> 6 <211> 6 <212> PRT <213> Homo sapiens <400> 6 Val Asn Gly Ala Pro Arg 1 5 <210> 7 <211> 7 <212> PRT <213> Homo sapiens <400> 7 Leu Ala Ala Ser Gln Ser Arg 1 5 <210> 8 <211> 7 <212> PRT <213> Homo sapiens <400> 8 Asn Gly Asn Phe His Pro Lys 1 5 <210> 9 <211> 6 <212> PRT <213> Homo sapiens <400> 9 Cys Trp Cys Val Asp Arg 1 5 <210> 10 <211> 13 <212> PRT <213> Homo sapiens <400> 10 Gly Glu Leu Asp Cys His Gln Leu Ala Asp Ser Phe Arg 1 5 10 <210> 11 <211> 8 <212> PRT <213> Homo sapiens <400> 11 Thr Ser Ser Ser Phe Glu Val Arg 1 5 <210> 12 <211> 16 <212> PRT <213> Homo sapiens <400> 12 Thr Trp Asp Pro Glu Gly Val Ile Phe Tyr Gly Asp Thr Asn Pro Lys 1 5 10 15 <210> 13 <211> 9 <212> PRT <213> Homo sapiens <400> 13 Asp Asp Trp Phe Met Leu Gly Leu Arg 1 5 <210> 14 <211> 22 <212> PRT <213> Homo sapiens <400> 14 Asp Gly Arg Pro Glu Ile Gln Leu His Asn His Trp Ala Gln Leu Thr 1 5 10 15 Val Gly Ala Gly Pro Arg 20 <210> 15 <211> 7 <212> PRT <213> Homo sapiens <400> 15 Trp His Gln Val Glu Val Lys 1 5 <210> 16 <211> 17 <212> PRT <213> Homo sapiens <400> 16 Met Glu Gly Asp Ser Val Leu Leu Glu Val Asp Gly Glu Glu Val Leu 1 5 10 15 Arg <210> 17 <211> 9 <212> PRT <213> Homo sapiens <400> 17 Gln Val Ser Gly Pro Leu Thr Ser Lys 1 5 <210> 18 <211> 14 <212> PRT <213> Homo sapiens <400> 18 Ile Ala Leu Gly Gly Leu Leu Phe Pro Ala Ser Asn Leu Arg 1 5 10 <210> 19 <211> 12 <212> PRT <213> Homo sapiens <400> 19 Leu Pro Leu Val Pro Ala Leu Asp Gly Cys Leu Arg 1 5 10 <210> 20 <211> 6 <212> PRT <213> Homo sapiens <400> 20 Asp Ser Trp Leu Asp Lys 1 5 <210> 21 <211> 13 <212> PRT <213> Homo sapiens <400> 21 Gln Ala Glu Ile Ser Ala Ser Ala Pro Thr Ser Leu Arg 1 5 10 <210> 22 <211> 23 <212> PRT <213> Homo sapiens <400> 22 Ser Cys Asp Val Glu Ser Asn Pro Gly Ile Phe Leu Pro Pro Gly Thr 1 5 10 15 Gln Ala Glu Phe Asn Leu Arg 20 <210> 23 <211> 19 <212> PRT <213> Homo sapiens <400> 23 Asp Ile Pro Gln Pro His Ala Glu Pro Trp Ala Phe Ser Leu Asp Leu 1 5 10 15 Gly Leu Lys <210> 24 <211> 24 <212> PRT <213> Homo sapiens <400> 24 Val Val Leu Ser Ser Gly Ser Gly Pro Gly Leu Asp Leu Pro Leu Val 1 5 10 15 Leu Gly Leu Pro Leu Gln Leu Lys 20 <210> 25 <211> 8 <212> PRT <213> Homo sapiens <400> 25 Val Val Leu Ser Gln Gly Ser Lys 1 5 <210> 26 <211> 9 <212> PRT <213> Homo sapiens <400> 26 Leu Asp Val Asp Gln Ala Leu Asn Arg 1 5 <210> 27 <211> 22 <212> PRT <213> Homo sapiens <400> 27 Ala Leu Ala Leu Pro Pro Leu Gly Leu Ala Pro Leu Leu Asn Leu Trp 1 5 10 15 Ala Lys Pro Gln Gly Arg 20 <210> 28 <211> 21 <212> PRT <213> Homo sapiens <400> 28 Ser His Glu Ile Trp Thr His Ser Cys Pro Gln Ser Pro Gly Asn Gly 1 5 10 15 Thr Asp Ala Ser His 20 <210> 29 <211> 14 <212> PRT <213> Homo sapiens <400> 29 Leu Pro Ala Glu Ile Ser Ala Ser Ala Pro Thr Ser Leu Arg 1 5 10 <210> 30 <211> 7 <212> PRT <213> Homo sapiens <400> 30 Thr Leu Pro Pro Leu Phe Ala 1 5 <210> 31 <211> 19 <212> PRT <213> Homo sapiens <400> 31 Gly Glu Asp Ser Ser Thr Ser Phe Cys Leu Asn Gly Leu Trp Ala Gln 1 5 10 15 Gly Gln Arg <210> 32 <211> 15 <212> PRT <213> Homo sapiens <400> 32 Asn Trp Gly Leu Ser Val Tyr Ala Asp Lys Pro Glu Thr Thr Lys 1 5 10 15 <210> 33 <211> 10 <212> PRT <213> Homo sapiens <400> 33 Leu Ser Ile Thr Gly Thr Tyr Asp Leu Lys 1 5 10 <210> 34 <211> 10 <212> PRT <213> Homo sapiens <400> 34 Asp Leu Leu Leu Pro Gln Pro Asp Leu Arg 1 5 10 <210> 35 <211> 6 <212> PRT <213> Homo sapiens <400> 35 Leu Gln Val Leu Gly Lys 1 5 <210> 36 <211> 19 <212> PRT <213> Homo sapiens <400> 36 Leu His Thr Glu Ala Gln Ile Gln Glu Glu Gly Thr Val Val Glu Leu 1 5 10 15 Thr Gly Arg <210> 37 <211> 10 <212> PRT <213> Homo sapiens <400> 37 Asp Ala Asp Pro Asp Thr Phe Phe Ala Lys 1 5 10 <210> 38 <211> 7 <212> PRT <213> Homo sapiens <400> 38 His Phe Gln Asn Leu Gly Lys 1 5 <210> 39 <211> 7 <212> PRT <213> Homo sapiens <400> 39 Leu Val Thr Asp Leu Thr Lys 1 5 <210> 40 <211> 13 <212> PRT <213> Homo sapiens <400> 40 Ile Arg Pro His Thr Phe Thr Gly Leu Ser Gly Leu Arg 1 5 10 <210> 41 <211> 8 <212> PRT <213> Homo sapiens <400> 41 Leu Glu Tyr Leu Leu Leu Ser Arg 1 5 <210> 42 <211> 12 <212> PRT <213> Homo sapiens <400> 42 Asp Pro Thr Phe Ile Pro Ala Pro Ile Gln Ala Lys 1 5 10 <210> 43 <211> 13 <212> PRT <213> Homo sapiens <400> 43 Ser Leu Asp Phe Thr Glu Leu Asp Val Ala Ala Glu Lys 1 5 10 <210> 44 <211> 9 <212> PRT <213> Homo sapiens <400> 44 Ala Lys Pro Ala Leu Glu Asp Leu Arg 1 5 <210> 45 <211> 9 <212> PRT <213> Homo sapiens <400> 45 Ser Pro Glu Leu Gln Ala Glu Ala Lys 1 5 <210> 46 <211> 7 <212> PRT <213> Homo sapiens <400> 46 Asp Tyr Trp Ser Thr Val Lys 1 5 <210> 47 <211> 11 <212> PRT <213> Homo sapiens <400> 47 Gly Trp Val Thr Asp Gly Phe Ser Ser Leu Lys 1 5 10 <210> 48 <211> 9 <212> PRT <213> Homo sapiens <400> 48 Ala Thr Val Val Tyr Gln Gly Glu Arg 1 5 <210> 49 <211> 9 <212> PRT <213> Homo sapiens <400> 49 Glu His Ser Ser Leu Ala Phe Trp Lys 1 5 <210> 50 <211> 10 <212> PRT <213> Homo sapiens <400> 50 Val Glu His Ser Asp Leu Ser Phe Ser Lys 1 5 10 <210> 51 <211> 10 <212> PRT <213> Homo sapiens <400> 51 Val Asn His Val Thr Leu Ser Gln Pro Lys 1 5 10 <210> 52 <211> 12 <212> PRT <213> Homo sapiens <400> 52 Leu Thr Leu Leu Ala Pro Leu Asn Ser Val Phe Lys 1 5 10 <210> 53 <211> 7 <212> PRT <213> Homo sapiens <400> 53 Val Leu Thr Asp Glu Leu Lys 1 5 <210> 54 <211> 8 <212> PRT <213> Homo sapiens <400> 54 Ile Asn Pro Ala Ser Leu Asp Lys 1 5 <210> 55 <211> 14 <212> PRT <213> Homo sapiens <400> 55 Val Pro Gly Leu Tyr Tyr Phe Thr Tyr His Ala Ser Ser Arg 1 5 10 <210> 56 <211> 9 <212> PRT <213> Homo sapiens <400> 56 Gly Gly Pro Phe Ser Asp Ser Tyr Arg 1 5 <210> 57 <211> 8 <212> PRT <213> Homo sapiens <400> 57 Val Gly Phe Ala Glu Ala Ala Arg 1 5 <210> 58 <211> 10 <212> PRT <213> Homo sapiens <400> 58 Val Ser Thr Leu Pro Ala Ile Thr Leu Lys 1 5 10 <210> 59 <211> 15 <212> PRT <213> Homo sapiens <400> 59 Glu Ala Leu Ile Gln Phe Leu Glu Gln Val His Gln Gly Ile Lys 1 5 10 15 <210> 60 <211> 10 <212> PRT <213> Homo sapiens <400> 60 Asn Asn Ala Asn Gly Val Asp Leu Asn Arg 1 5 10 <210> 61 <211> 18 <212> PRT <213> Homo sapiens <400> 61 Leu Thr Val Gly Ala Ala Gln Val Pro Ala Gln Leu Leu Val Gly Ala 1 5 10 15 Leu Arg <210> 62 <211> 15 <212> PRT <213> Homo sapiens <400> 62 Ser Trp Leu Ala Glu Leu Gln Gln Trp Leu Lys Pro Gly Leu Lys 1 5 10 15 <210> 63 <211> 14 <212> PRT <213> Homo sapiens <400> 63 Val Ser Glu Ala Asp Ser Ser Asn Ala Asp Trp Val Thr Lys 1 5 10 <210> 64 <211> 11 <212> PRT <213> Homo sapiens <400> 64 Tyr Gly Leu Val Thr Tyr Ala Thr Tyr Pro Lys 1 5 10 <210> 65 <211> 10 <212> PRT <213> Homo sapiens <400> 65 Thr Ala Val Thr Ala Asn Leu Asp Ile Arg 1 5 10 <210> 66 <211> 9 <212> PRT <213> Homo sapiens <400> 66 Val Ile Ala Val Asn Glu Val Gly Arg 1 5 <210> 67 <211> 12 <212> PRT <213> Homo sapiens <400> 67 Ala Ser Ser Ile Ile Asp Glu Leu Phe Gln Asp Arg 1 5 10 <210> 68 <211> 17 <212> PRT <213> Homo sapiens <400> 68 Leu Phe Asp Ser Asp Pro Ile Thr Val Thr Val Pro Val Glu Val Ser 1 5 10 15 Arg <210> 69 <211> 10 <212> PRT <213> Homo sapiens <400> 69 Ile His Trp Glu Ser Ala Ser Leu Leu Arg 1 5 10 <210> 70 <211> 11 <212> PRT <213> Homo sapiens <400> 70 Thr Leu Leu Pro Val Ser Lys Pro Glu Ile Arg 1 5 10 <210> 71 <211> 7 <212> PRT <213> Homo sapiens <400> 71 Val Phe Gln Phe Leu Glu Lys 1 5 <210> 72 <211> 16 <212> PRT <213> Homo sapiens <400> 72 Ala Leu Asn His Leu Pro Leu Glu Tyr Asn Ser Ala Leu Tyr Ser Arg 1 5 10 15 <210> 73 <211> 11 <212> PRT <213> Homo sapiens <400> 73 Ser Glu Tyr Gly Ala Ala Leu Ala Trp Glu Lys 1 5 10 <210> 74 <211> 7 <212> PRT <213> Homo sapiens <400> 74 Ser Leu Leu Gln Pro Asn Lys 1 5 <210> 75 <211> 25 <212> PRT <213> Homo sapiens <400> 75 Tyr His Phe Glu Ala Leu Ala Asp Thr Gly Ile Ser Ser Glu Phe Tyr 1 5 10 15 Asp Asn Ala Asn Asp Leu Leu Ser Lys 20 25 <210> 76 <211> 8 <212> PRT <213> Homo sapiens <400> 76 Gln Ala Leu Glu Glu Phe Gln Lys 1 5 <210> 77 <211> 8 <212> PRT <213> Homo sapiens <400> 77 Ser Gly Phe Ser Phe Gly Phe Lys 1 5 <210> 78 <211> 7 <212> PRT <213> Homo sapiens <400> 78 Ala Val Ser Pro Pro Ala Arg 1 5 <210> 79 <211> 9 <212> PRT <213> Homo sapiens <400> 79 Tyr Glu Asp Leu Tyr Ser Asn Cys Lys 1 5 <210> 80 <211> 19 <212> PRT <213> Homo sapiens <400> 80 Ala His Gln Leu Ala Ile Asp Thr Tyr Gln Glu Phe Glu Glu Thr Tyr 1 5 10 15 Ile Pro Lys <210> 81 <211> 14 <212> PRT <213> Homo sapiens <400> 81 Ile Ser Leu Leu Leu Ile Glu Ser Trp Leu Glu Pro Val Arg 1 5 10 <210> 82 <211> 23 <212> PRT <213> Homo sapiens <400> 82 Thr Glu Phe Leu Ser Asn Tyr Leu Thr Asn Val Asp Asp Ile Thr Leu 1 5 10 15 Val Pro Gly Thr Leu Gly Arg 20 <210> 83 <211> 10 <212> PRT <213> Homo sapiens <400> 83 Thr Tyr Leu His Thr Tyr Glu Ser Glu Ile 1 5 10 <210> 84 <211> 16 <212> PRT <213> Homo sapiens <400> 84 Gly Asp Thr Tyr Pro Ala Glu Leu Tyr Ile Thr Gly Ser Ile Leu Arg 1 5 10 15 <210> 85 <211> 9 <212> PRT <213> Homo sapiens <400> 85 Ile Ala Gln Tyr Tyr Tyr Thr Phe Lys 1 5 <210> 86 <211> 10 <212> PRT <213> Homo sapiens <400> 86 Thr Gly Tyr Tyr Phe Asp Gly Ile Ser Arg 1 5 10 <210> 87 <211> 8 <212> PRT <213> Homo sapiens <400> 87 Ile Pro Ser Asn Pro Ser His Arg 1 5 <210> 88 <211> 7 <212> PRT <213> Homo sapiens <400> 88 Phe Ser Val Val Tyr Ala Lys 1 5 <210> 89 <211> 10 <212> PRT <213> Homo sapiens <400> 89 His Thr Leu Asn Gln Ile Asp Glu Val Lys 1 5 10 <210> 90 <211> 15 <212> PRT <213> Homo sapiens <400> 90 Glu Ser Ser Ser His His Pro Gly Ile Ala Glu Phe Pro Ser Arg 1 5 10 15 <210> 91 <211> 13 <212> PRT <213> Homo sapiens <400> 91 Gln Gly Phe Gly Asn Val Ala Thr Asn Thr Asp Gly Lys 1 5 10 <210> 92 <211> 6 <212> PRT <213> Homo sapiens <400> 92 Phe Leu Asn Trp Ile Lys 1 5 <210> 93 <211> 11 <212> PRT <213> Homo sapiens <400> 93 Asn Phe Pro Ser Pro Val Asp Ala Ala Phe Arg 1 5 10 <210> 94 <211> 16 <212> PRT <213> Homo sapiens <400> 94 Ser Gly Ala Gln Ala Thr Trp Thr Glu Leu Pro Trp Pro His Glu Lys 1 5 10 15 <210> 95 <211> 13 <212> PRT <213> Homo sapiens <400> 95 Trp Ala Ala Val Val Val Pro Ser Gly Glu Glu Gln Arg 1 5 10 <210> 96 <211> 13 <212> PRT <213> Homo sapiens <400> 96 Thr Glu Gly Asp Gly Val Tyr Thr Leu Asn Asn Glu Lys 1 5 10 <210> 97 <211> 7 <212> PRT <213> Homo sapiens <400> 97 Val Val Glu Ser Leu Ala Lys 1 5 <210> 98 <211> 9 <212> PRT <213> Homo sapiens <400> 98 Leu Ile Gln Gly Ala Pro Thr Ile Arg 1 5 <210> 99 <211> 8 <212> PRT <213> Homo sapiens <400> 99 Phe Leu Asn Val Leu Ser Pro Arg 1 5 <210> 100 <211> 11 <212> PRT <213> Homo sapiens <400> 100 Tyr Gly Gln Pro Leu Pro Gly Tyr Thr Thr Lys 1 5 10 <210> 101 <211> 13 <212> PRT <213> Homo sapiens <400> 101 Gly Ala Gln Thr Leu Tyr Val Pro Asn Cys Asp His Arg 1 5 10 <210> 102 <211> 15 <212> PRT <213> Homo sapiens <400> 102 His Leu Asp Ser Val Leu Gln Gln Leu Gln Thr Glu Val Tyr Arg 1 5 10 15 <210> 103 <211> 9 <212> PRT <213> Homo sapiens <400> 103 Gly Ile Val Glu Glu Cys Cys Phe Arg 1 5 <210> 104 <211> 16 <212> PRT <213> Homo sapiens <400> 104 Ser Cys Asp Leu Ala Leu Leu Glu Thr Tyr Cys Ala Thr Pro Ala Lys 1 5 10 15 <210> 105 <211> 8 <212> PRT <213> Homo sapiens <400> 105 Ala Leu Pro Ala Pro Ile Glu Lys 1 5 <210> 106 <211> 7 <212> PRT <213> Homo sapiens <400> 106 Gly Phe Pro Ser Val Leu Arg 1 5 <210> 107 <211> 9 <212> PRT <213> Homo sapiens <400> 107 Leu Asp Phe His Phe Ser Ser Asp Arg 1 5 <210> 108 <211> 7 <212> PRT <213> Homo sapiens <400> 108 Ala Ser Ser Ile Leu Ala Thr 1 5 <210> 109 <211> 12 <212> PRT <213> Homo sapiens <400> 109 Glu Leu Trp Phe Ser Asp Asp Pro Asp Val Thr Lys 1 5 10 <210> 110 <211> 11 <212> PRT <213> Homo sapiens <400> 110 Asn Val Asp Gln Ser Leu Leu Glu Leu His Lys 1 5 10 <210> 111 <211> 7 <212> PRT <213> Homo sapiens <400> 111 Ala Leu Asp Leu Ser Leu Lys 1 5 <210> 112 <211> 8 <212> PRT <213> Homo sapiens <400> 112 Ile Leu Asp Asp Leu Ser Pro Arg 1 5 <210> 113 <211> 17 <212> PRT <213> Homo sapiens <400> 113 Asn Pro Leu Val Trp Val His Ala Ser Pro Glu His Val Val Val Thr 1 5 10 15 Arg <210> 114 <211> 19 <212> PRT <213> Homo sapiens <400> 114 Gln Leu Gly Leu Pro Gly Pro Pro Asp Val Pro Asp His Ala Ala Tyr 1 5 10 15 His Pro Phe <210> 115 <211> 8 <212> PRT <213> Homo sapiens <400> 115 Val Arg Pro Gln Gln Leu Val Lys 1 5 <210> 116 <211> 19 <212> PRT <213> Homo sapiens <400> 116 Asp Ile Pro Thr Asn Ser Pro Glu Leu Glu Glu Thr Leu Thr His Thr 1 5 10 15 Ile Thr Lys <210> 117 <211> 9 <212> PRT <213> Homo sapiens <400> 117 Gln Val Val Ala Gly Leu Asn Phe Arg 1 5 <210> 118 <211> 9 <212> PRT <213> Homo sapiens <400> 118 Ile Thr Gly Phe Leu Lys Pro Gly Lys 1 5 <210> 119 <211> 11 <212> PRT <213> Homo sapiens <400> 119 Ile Thr Leu Pro Asp Phe Thr Gly Asp Leu Arg 1 5 10 <210> 120 <211> 8 <212> PRT <213> Homo sapiens <400> 120 Ser Tyr Tyr Trp Ile Gly Ile Arg 1 5 <210> 121 <211> 12 <212> PRT <213> Homo sapiens <400> 121 Gly Leu Gly Glu Ile Ser Ala Ala Ser Glu Phe Lys 1 5 10 <210> 122 <211> 10 <212> PRT <213> Homo sapiens <400> 122 Asp Ile Pro His Trp Leu Asn Pro Thr Arg 1 5 10 <210> 123 <211> 13 <212> PRT <213> Homo sapiens <400> 123 Leu Asp Gly Ser Thr His Leu Asn Ile Phe Phe Ala Lys 1 5 10 <210> 124 <211> 12 <212> PRT <213> Homo sapiens <400> 124 Leu Gln Ser Leu Phe Asp Ser Pro Asp Phe Ser Lys 1 5 10 <210> 125 <211> 10 <212> PRT <213> Homo sapiens <400> 125 Thr Val Gln Ala Val Leu Thr Val Pro Lys 1 5 10 <210> 126 <211> 14 <212> PRT <213> Homo sapiens <400> 126 Ala Gly Leu Leu Arg Pro Asp Tyr Ala Leu Leu Gly His Arg 1 5 10 <210> 127 <211> 20 <212> PRT <213> Homo sapiens <400> 127 Asp Gly Ser Pro Asp Val Thr Thr Ala Asp Ile Gly Ala Asn Thr Pro 1 5 10 15 Asp Ala Thr Lys 20 <210> 128 <211> 8 <212> PRT <213> Homo sapiens <400> 128 Gly Leu Phe Ile Ile Asp Gly Lys 1 5 <210> 129 <211> 14 <212> PRT <213> Homo sapiens <400> 129 Trp Asn Phe Ala Tyr Trp Ala Ala His Gln Pro Trp Ser Arg 1 5 10 <210> 130 <211> 25 <212> PRT <213> Homo sapiens <400> 130 Asp Leu Tyr His Tyr Ile Thr Ser Tyr Val Val Asp Gly Glu Ile Ile 1 5 10 15 Ile Tyr Gly Pro Ala Tyr Ser Gly Arg 20 25 <210> 131 <211> 7 <212> PRT <213> Homo sapiens <400> 131 Phe Gln Leu Pro Gly Gln Lys 1 5 <210> 132 <211> 9 <212> PRT <213> Homo sapiens <400> 132 Leu Phe Ile Pro Gln Ile Thr Pro Lys 1 5 <210> 133 <211> 9 <212> PRT <213> Homo sapiens <400> 133 Ile His Pro Ser Tyr Thr Asn Tyr Arg 1 5 <210> 134 <211> 16 <212> PRT <213> Homo sapiens <400> 134 Val Ser Ala Pro Ser Gly Thr Gly His Leu Pro Gly Leu Asn Pro Leu 1 5 10 15 <210> 135 <211> 20 <212> PRT <213> Homo sapiens <400> 135 Asp Val Leu Leu Leu Val His Asn Leu Pro Gln Asn Leu Pro Gly Tyr 1 5 10 15 Phe Trp Tyr Lys 20 <210> 136 <211> 8 <212> PRT <213> Homo sapiens <400> 136 Leu Phe Ile Pro Gln Ile Thr Arg 1 5 <210> 137 <211> 7 <212> PRT <213> Homo sapiens <400> 137 Gly Pro Gly Glu Asp Phe Arg 1 5 <210> 138 <211> 9 <212> PRT <213> Homo sapiens <400> 138 Asn Tyr Gly Leu Leu Tyr Cys Phe Arg 1 5 <210> 139 <211> 8 <212> PRT <213> Homo sapiens <400> 139 Ser Val Glu Gly Ser Cys Gly Phe 1 5 <210> 140 <211> 11 <212> PRT <213> Homo sapiens <400> 140 Val Leu Thr His Ser Glu Leu Ala Pro Leu Arg 1 5 10 <210> 141 <211> 16 <212> PRT <213> Homo sapiens <400> 141 Leu Asn Trp Glu Ala Pro Pro Gly Ala Phe Asp Ser Phe Leu Leu Arg 1 5 10 15 <210> 142 <211> 15 <212> PRT <213> Homo sapiens <400> 142 Leu Ser Gln Leu Ser Val Thr Asp Val Thr Thr Ser Ser Leu Arg 1 5 10 15 <210> 143 <211> 8 <212> PRT <213> Homo sapiens <400> 143 Ala Val Leu His Ile Gly Glu Lys 1 5 <210> 144 <211> 20 <212> PRT <213> Homo sapiens <400> 144 Val Ser Trp Ser Leu Pro Leu Val Pro Gly Pro Leu Val Gly Asp Gly 1 5 10 15 Phe Leu Leu Arg 20 <210> 145 <211> 8 <212> PRT <213> Homo sapiens <400> 145 Tyr Leu Gly Glu Glu Tyr Val Lys 1 5 <210> 146 <211> 6 <212> PRT <213> Homo s...

Claims

1. Use of capture reagents that specifically bind to IBP4 and capture reagents that specifically bind to SHBG in the preparation of kits for determining the probability of preterm birth in biological samples from human pregnant women, wherein IBP4 comprises the amino acid sequence QCHPALDGQR and wherein SHBG comprises the amino acid sequence IALGGLLFPASNLR.

2. The use as described in claim 1, wherein the determination is performed by: a) Detecting the presence of IBP4 and SHBG in the biological sample by contacting the biological sample with the capture reagent that specifically binds IBP4 and the capture reagent that specifically binds SHBG; and b) Detect the binding between IBP4 and the capture reagent that specifically binds to IBP4, and between SHBG and the capture reagent that specifically binds to SHBG.

3. Use of the isolated biomarker pair consisting of IBP4 / SHBG in the preparation of a kit for determining the probability of preterm birth in biological samples from human pregnant women, wherein IBP4 contains the amino acid sequence QCHPALDGQR and wherein SHBG contains the amino acid sequence IALGGLLFPASNLR.

4. The use as claimed in claim 3, wherein the determination is performed by detecting the presence of IBP4 and SHBG in the biological sample, the detection comprising a proteomics workflow including mass spectrometry quantification of the sample.

5. The use as described in any one of claims 1-4, wherein the biological sample is selected from the group consisting of whole blood, plasma and serum.

6. The use as described in claim 5, wherein the biological sample is serum.

7. The use as claimed in claim 5, wherein the sample was obtained between 19 and 21 weeks of gestation.

8. The use as described in claim 4, wherein the proteomics workflow includes quantifying stable isotope-labeled peptides.

9. The use as described in claim 2 or 4 further includes measurable features for detecting one or more risk indicators.

10. The use as claimed in claim 9, wherein the risk refers to the group consisting of free body mass index, pregnancy and fetal sex.

11. The use as described in claim 10, wherein the risk indicator is body mass index.

12. The use as claimed in claim 11, wherein the body mass index of the human pregnant woman is greater than 22 and less than or equal to 37 kg / m². 2 .

13. The use as described in claim 4, wherein the mass spectrometry is selected from the group consisting of surface-enhanced laser desorption / ionization time-of-flight mass spectrometry; electrospray ionization 3D or linear 2D ion trap mass spectrometry; electrospray ionization quadrupole orthogonal time-of-flight mass spectrometry; electrospray ionization Fourier transform mass spectrometry system; silicon desorption / ionization; secondary ion mass spectrometry; ion migration spectroscopy and inductively coupled plasma mass spectrometry.

14. The use as described in claim 13, wherein the mass spectrometry is matrix-assisted laser desorption / ionization time-of-flight mass spectrometry.

15. The use as claimed in claim 13, wherein the mass spectrometry is a matrix-assisted laser desorption / ionization time-of-flight source followed by decay.

16. The use as described in claim 13, wherein the mass spectrometry is matrix-assisted laser desorption / ionization time-of-flight / time-of-flight.

17. The use as claimed in claim 13, wherein the mass spectrometry is electrospray ionization mass spectrometry.

18. The use as claimed in claim 13, wherein the mass spectrometry is electrospray ionization-mass spectrometry / mass spectrometry.

19. The use as claimed in claim 13, wherein the mass spectrometry is electrospray ionization mass spectrometry / (mass spectrometry)n, where n is an integer greater than zero.

20. The use as claimed in claim 13, wherein the mass spectrometry is electrospray ionization triple quadrupole mass spectrometry.

21. The use as described in claim 13, wherein the mass spectrometry is atmospheric pressure chemical ionization mass spectrometry / mass spectrometry.

22. The use as described in claim 13, wherein the mass spectrometry is atmospheric pressure chemical ionization mass spectrometry.

23. The use as described in claim 13, wherein the mass spectrometry is atmospheric pressure photoionization-mass spectrometry.

24. The use as described in claim 13, wherein the mass spectrometry is atmospheric pressure photoionization mass spectrometry / mass spectrometry.

25. The use as claimed in claim 13, wherein the mass spectrometry is atmospheric pressure photoionization mass spectrometry.

26. The use as claimed in claim 4, wherein the mass spectrometry comprises immunoprecipitation mass spectrometry.

27. The use as claimed in claim 4, wherein the mass spectrometry comprises liquid chromatography-mass spectrometry.

28. The use as claimed in claim 4, wherein the mass spectrometry includes multiple reaction monitoring or selected reaction monitoring.

29. Use of biomarker pair IBP4 and SHBG in the preparation of a kit for determining the probability of preterm birth in a human pregnant woman based on the reversal value of the biomarker pair in a biological sample from a human pregnant woman, wherein the IBP4 biomarker comprises the amino acid sequence QCHPALDGQR, and wherein the SHBG biomarker comprises the amino acid sequence IALGGLLFPASNLR.

30. The use as claimed in claim 29, wherein the presence of a reversal value change between the human pregnant woman and the full-term control indicates the probability of preterm birth in the human pregnant woman.

31. The use as claimed in claim 29, wherein the determination comprises measuring the alternative peptide of the biomarker in a biological sample obtained from the human pregnant woman.

32. The use as described in claim 29, wherein the probability is represented as a risk score.

33. The use as claimed in claim 29, wherein the biological sample is selected from the group consisting of whole blood, plasma and serum.

34. The use as described in claim 29, wherein the biological sample is serum.

35. The use as claimed in claim 33, wherein the sample was obtained between 19 and 22 weeks of gestation.

36. The use as claimed in claim 31, wherein the measurement includes mass spectrometry.

37. The use as claimed in claim 31, wherein the measurement includes determination using a capture reagent.

38. The use as described in claim 37, wherein the capture agent is selected from the group consisting of antibodies, antibody fragments, nucleic acid-based protein binding agents, small molecules or variants thereof.

39. The use as described in claim 37, wherein the assay is selected from the group consisting of enzyme immunoassay, enzyme-linked immunosorbent assay, and radioimmunoassay.

40. The use as described in claim 29 further includes measurable features for detecting one or more risk indicators.

41. The use as described in claim 40, wherein the risk refers to the group consisting of free body mass index, pregnancy and fetal sex.

42. The use as described in claim 41, wherein the risk indicator is body mass index.

43. The use as claimed in claim 29, wherein the use further includes predicting gestational age at birth prior to determining the probability of preterm birth.

44. The use as described in claim 42, wherein the body mass index of the human pregnant woman is greater than 22 and less than or equal to 37 kg / m². 2 .

45. The use as claimed in claim 29, wherein the reversal value is based on the ratio of IBP4 to SHBG: IBP4 / SHBG, to determine the probability of preterm birth in the human pregnant woman, wherein a higher rate in the human pregnant woman than in the full-term control indicates an increased risk of preterm birth.

46. ​​The use as claimed in claim 36, wherein the mass spectrometer is selected from the group consisting of: surface-enhanced laser desorption / ionization time-of-flight mass spectrometry (SELDI-TOF); ESI 3D or linear (2D) ion trap MS; ESI quadrupole orthogonal TOF (Q-TOF); ESI Fourier transform MS system; silicon-on-silicon desorption / ionization (DIOS); secondary ion mass spectrometry (SIMS); ion migration spectroscopy (IMS); and inductively coupled plasma mass spectrometry (ICP-MS).

47. The use as described in claim 46, wherein the mass spectrometry is matrix-assisted laser desorption / ionization time-of-flight (MALDI-TOF) MS.

48. The use as claimed in claim 46, wherein the mass spectrometry is MALDI-TOF post-source decay (PSD).

49. The use as claimed in claim 46, wherein the mass spectrometer is MALDI-TOF / TOF.

50. The use as claimed in claim 46, wherein the mass spectrometry is electrospray ionization mass spectrometry (ESI-MS).

51. The use as described in claim 46, wherein the mass spectrometry is ESI-MS / MS.

52. The use as claimed in claim 46, wherein the mass spectrometry is ESI-MS / (MS)n, where n is an integer greater than zero.

53. The use as described in claim 46, wherein the mass spectrometry is an ESI triple quadruple MS.

54. The use as described in claim 46, wherein the mass spectrometry is atmospheric pressure chemical ionization mass spectrometry (APCI-MS).

55. The use as described in claim 46, wherein the mass spectrometry is APCI-MS / MS.

56. The use as claimed in claim 46, wherein the mass spectrometer is APCI-(MS)n, where n is an integer greater than 0.

57. The use as described in claim 46, wherein the mass spectrometer is an atmospheric pressure photoionization mass spectrometer (APPI-MS).

58. The use as described in claim 46, wherein the mass spectrometry is APPI-MS / MS.

59. The use as claimed in claim 46, wherein the mass spectrometry is APPI-(MS)n, where n is an integer greater than 0.

60. The use as claimed in claim 36, wherein the mass spectrometry comprises immunoprecipitation-mass spectrometry.

61. The use as claimed in claim 36, wherein the mass spectrometry comprises liquid chromatography-mass spectrometry.

62. The use as claimed in claim 36, wherein the mass spectrometry includes multiple reaction monitoring or selected reaction monitoring.

Images