CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims priority to U.S. Provisional Application Serial No.

61/908,985 filed November 26, 2013 and U.S. Provisional Application Serial No. 61/909,006 also filed November 26, 2013.

TECHNICAL FIELD OF INVENTION

[0002] The present invention relates to the identification and use of metabolomic markers in maternal, newborn or pediatric serum for the prediction or diagnosis of congenital heart defects (CHD).

BACKGROUND

[0003] Congenital heart defect (CHD) is the most important category of congenital anomalies based both on its frequency, 0.6 to 0.8% of all the births, and cumulative health care costs. In contrast to the routine population pregnancy screening for the detection of less common fetal anomalies such as aneuploidies and neural tube defect, there is no similar screening policies for CHD.

[0004] Reliable detection of CHD is the "holy grail" of prenatal screening. This directly reflects the importance of CHD. Congenital anomaly is the most important cause of infant death in the USA. The prenatal diagnosis of CHD has distinct advantages including the opportunity for early counseling of families-facilitating reproductive choices. Decisions such as transferring of care to an appropriate pregnancy specialist within the same institution or complete transfer to another institution with the appropriate high risk obstetrical, newborn and pediatric expertise often need to be made. Given the high rate of intervention and hospitalization in CHD cases there are significant social and financial implications to affected families. Prenatal diagnosis of CHD has been reported to improve medical costs. Data from the USA also found a greater than tenfold increase in average newborn transportation costs when CHD was diagnosed postnatally compared to prenatal detection. Finally, in some categories of CHD prenatal diagnosis reportedly may improve overall outcome compared to those in which the diagnosis is made after birth.

[0005] There is suggestive evidence that at least in some types of CHD, prenatal diagnosis may improve newborn outcome. The increasing interest in fetal cardiac intervention for such lesions as aortic stenosis and hypoplastic left heart syndrome creates another potentially powerful argument in favor of prenatal diagnosis, at least for cardiac anomalies that are amendable to such approaches.

[0006] An area of concern with respect to the prenatal diagnosis of any congenital anomalies has been concerns about medical selection against affected fetuses. Data from France have however shown that pregnancy termination rates have not increased in proportion to improving prenatal diagnosis of CHD. Indeed termination rates have leveled off and pregnancy termination was exceptional among the more common categories of CHD while at the same time there has been a reduction in early neonatal deaths.

[0007] Ultrasound remains the most widely used prenatal tool for the detection of fetal CHD. While specialist centers that care for high risk patients report high sensitivities for CHD detection the overall performance of prenatal ultrasound in the general population remains substantially below that required for an effective screening test. A recent study in the USA found that slightly less than 40% of CHD cases were detected prenatally in a state-wide obstetric population that had an ultrasound exam at the appropriate gestational age. The overall accuracy of prenatal ultrasound is significantly constrained by its dependence on operator expertise, equipment quality and uncontrollable variables such as fetal position and maternal obesity. The difficulties associated with the of diagnosis of CHD is moreover not limited to the prenatal period as a relatively high percentage of critical CHD fail to be diagnosed in newborns prior to discharge home.

[0008] Metabolomics is a branch of the "omics" sciences in which high through-put techniques are used for the identification and quantification of the small molecules that constitute the metabolome. Metabolites are a very diverse group of molecules including but not limited to amino acids, nucleic acids, lipids, peptides, sugars and organic acids. They represent the substrates and byproducts of the various enzymatic reactions within the cells but also respond to and reflect various physiologic influences, e.g. age and gender, moreover pathologic and environmental influences including, diet, toxins, pharmacologic agents and stress which are important causes and modifiers of disease, significantly influence the metabolome. Based on the latter, metabolomics appears to give a more complete description of cellular phenotype than the genome, transcription or proteome.

[0009] Increasingly metabolomics is being used to develop biomarkers for the detection, screening and monitoring of complex diseases. There is limited prior evidence that CHD may either be caused by or associated with metabolic disturbance in humans. Abnormalities of folate and single carbon metabolism have been linked to the development of CHD. Comprehensive metabolomic analysis for the prediction of fetal, newborn or pediatric CHD has not been previously reported.

SUMMARY

[0010] Prenatal ultrasound remains the only tool currently available for the detection of CHD. Studies that are primarily from referral specialist centers report high diagnostic accuracy with specialized echocardiographic techniques such as Spatio-Temporal Imagery Correlation (STIC) and combined cardiac anomaly detection. Most population studies however paint a considerable less optimistic view of achievable detection rates even among groups with high, i.e. greater than 90% exposure to prenatal ultrasound. The majority, close to 80% of non-chromosomal CHD cases, failed to be diagnosed prenatally in 29 population- based registries in 16 European countries.

[0011] The estimated current screening practices in developed countries detected only 30- 50% of fetal CHD cases. Despite the widely reported low CHD screening performance, few studies have however examined the reasons for such low diagnosis rates. One study systematically reviewed the causes of low CHD detection rate in a 10 year review of a statewide surveillance program in Utah, USA. (Pinto NM, Keenan HT, Minich LL, Puchalksi MD, Heywood M, Botto LD. Barriers to prenatal detection of congenital heart disease: a population based study. Ultrasound Obstet Gynecol 2012; 40:418-25.) CHD prenatal detection rate was only 39% overall. The main factors accounting for failure to diagnose CHD prenatally were location in which the exam was performed, i.e. hospital versus high risk maternal fetal medicine office, the ultrasound interpreter i.e. obstetrician, radiologist or maternal fetal medicine specialist and the absence or presence of extracardiac anomalies. A family history of CHD also increased the detection of cardiac anomalies, likely due to the identification of the patient as being at elevated risk with greater attention to detail on the ultrasound exam. Despite enhanced chances of diagnosis when an MFM performed the ultrasound, in 25% of such cases scanned MFM offices the diagnosis was missed. There was universal availability of ultrasound in the study population. Other factors such as gestational age at the performance of ultrasound, maternal body habitus and fetal lie are known to affect the chances of detecting a fetal cardiac anomaly. None of these limitations would appear to be relevant or significant with maternal biomarkers such as the examples claimed in the present invention. [0012] The inventors have discovered significant disturbance in lipid including phosphatidyl-choline and various sphingo-lipids and choline metabolism in the serum of women carrying CHD fetuses. The present invention relates to the identification and use of metabolomic markers in maternal, newborn or pediatric serum for the prediction and diagnosis of congenital heart defects (CHD).

[0013] In one embodiment, the invention provides a method for assaying metabolite levels in a biologic sample from a patient during pregnancy using nuclear magnetic resonance and mass spectrometry. In some embodiments, the assay is performed during the first trimester of pregnancy. In some embodiments, the assay is performed during the second trimester of pregnancy. In some embodiments, the assay is performed during the third trimester of pregnancy. In some embodiments, the biologic sample is a maternal bodily fluid such as saliva, urine, amniotic fluid, breath condensate, blood or placental tissue. In some embodiments, the biologic sample is blood.

[0014] In some embodiments, the metabolite is an acylcarnitine. In some embodiments, the metabolite is one or more of C3-OH (hydroxypropionylcarnitine), C5.0H (C3DC), CIO, C5:1-DC (glutaconylcarnitine), CI 4:1 -OH (hydroxytetradecenoylcarnitine) and C14:2-OH.

[0015] In one embodiment, the invention further provides measuring the nuchal translucency of the fetus by ultrasound.

[0016] In one embodiment, the invention provides a method for predicting congenital heart defects in a fetus, comprising measuring metabolite levels in a biologic sample from the mother using nuclear magnetic resonance and direct flow injection mass spectrometry. In some embodiments, the measuring is performed during the first trimester of pregnancy. In some embodiments, the assay is performed during the second trimester of pregnancy. In some embodiments, the assay is performed during the third trimester of pregnancy. In some embodiments, the biologic sample is a maternal bodily fluid such as saliva, urine, amniotic fluid, breath condensate or blood. In some embodiments, the biologic sample is blood.

[0017] In some embodiments, the metabolite is an acylcarnitine. In some embodiments, the metabolite is one or more of C3-OH (hydroxypropionylcarnitine), C5.0H (C3DC), CIO, C5:1-DC (glutaconylcarnitine), C14:l-OH (hydroxytetradecenoylcarnitine) and C14:2-OH.

[0018] In one embodiment, the invention further provides measuring the nuchal translucency of the fetus by ultrasound.

[0019] In one embodiment, the invention further provides counseling the mother. In some embodiments, counseling comprises recommending transfer of care to a facility specializing in high-risk pregnancies. In some embodiments, counseling comprises recommending transfer of care to a physician specializing in high-risk pregnancies.

[0020] In one embodiment, the invention provides method for predicting congenital heart defects in a fetus or embryo, comprising measuring metabolite levels in a biologic sample from the mother, wherein the metabolites are more than one of C3-OH

(hydroxypropionylcarnitine), C5.0H (C3DC), CIO, C5:1-DC (glutaconylcarnitine), C14:l- OH (hydroxytetradecenoylcamitine) and C14:2-OH. In some embodiments, the metabolites are C3-OH (hydroxypropionylcarnitine), C5:1-DC (glutaconylcarnitine), and

(hydroxytetradecenoylcamitine). In some embodiments, measuring is achieved by nuclear magnetic resonance and direct flow injection mass spectrometry.

[0021] In one embodiment, mRNA microchip (RT-PCR) analysis is used to measure the expression levels of CHD related genes.

[0022] In one embodiment, the invention provides a method for assaying metabolite levels in a biologic sample from a patient using nuclear magnetic resonance and mass spectrometry. In one embodiment, the patient is a newborn. In one embodiment, the patient is a pediatric patient. In one embodiment, the patient is an embryo i.e. < 8 weeks gestational age.

[0023] In one embodiment, the metabolite is an acylcarnitine. In one embodiment, the metabolite is one or more of C3-OH (hydroxypropionylcarnitine), C5.0H (C3DC), CIO, C5:1-DC (glutaconylcarnitine), C14:l-OH (hydroxytetradecenoylcamitine) and C14:2-OH.

[0024] In one embodiment, the biologic sample is a bodily fluid selected from the group consisting of saliva, urine, breath condensate or blood. In one embodiment, the biologic sample is blood. In one embodiment, the sample is placenta.

[0025] In one embodiment, the invention provides a method for diagnosing congenital heart defects in a patient, comprising measuring metabolite levels in a biologic sample using nuclear magnetic resonance and mass spectrometry. In one embodiment, the patient is a newborn. In one embodiment, the patient is a fetus. In one embodiment, the patient is a pediatric patient.

[0026] In one embodiment, the metabolite is an acylcarnitine. In one embodiment, the metabolite is one or more of C3-OH (hydroxypropionylcarnitine), C5.0H (C3DC), CIO, C5:1-DC (glutaconylcarnitine), C14:l-OH (hydroxytetradecenoylcamitine) and C14:2-OH.

[0027] In one embodiment, the biologic sample is a bodily fluid selected from the group consisting of saliva, urine, breath condensate or blood, including umbilical cord blood. In one embodiment, the biologic sample is blood.

[0028] In one embodiment, the invention provides a method for diagnosing congenital heart defects in a newborn or pediatric patient, comprising measuring metabolite levels in a biologic sample, wherein the metabolites are more than one of C3-OH

(hydroxypropionylcarnitine), C5.0H (C3DC), CIO, C5:1-DC (glutaconylcarnitine), C14:l- OH (hydroxytetradecenoylcamitine) and C14:2-OH. In one embodiment, the metabolites are C3-OH (hydroxypropionylcarnitine), C5:1-DC (glutaconylcarnitine), and

(hydroxytetradecenoylcarnitine).

[0029] In one embodiment, measuring is achieved by nuclear magnetic resonance and direct flow injection mass spectrometry.

BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 depicts a 3-D PLS-DA plot based on DI-MS only.

[0031] Figure 2 depicts a VIP scores Plot, DI-MS Based Metabolites.

[0032] Figure 3 depicts permutation testing for DI-MS only; Permutation Test (n=2000), P-value= 0.0005.

[0033] Figure 4 depicts a 3-D PLS-DA plot using NMR only.

[0034] Figure 5 depicts a ROC comparison of all logistic regression models produced in this study. NT: AUC = 0.753, Metabolites (DI-MS and NMR) + NT: AUC=0.992, Metabolites (DI-MS and NMR): AUC=0.981, Metabolites (NMR) + NT: AUC=0.847, and Metabolites (NMR): AUC=0.749. NT, Nuchal Translucency; Metabolites: DI-MS and NMR, and three metabolites used in the model (Hydroxypropionylcarnitine,

Glutaconylcarnitine, and Hydroxytetradecadienylcarnitine); Metabolites (NMR): Acetone, Ethanol.

DETAILED DESCRIPTION

Definitions

[0035] As used herein, "nuclear magnetic resonance" or "NMR" refers to the absorption of electromagnetic radiation of a specific frequency by an atomic nucleus containing hydrogen (tT) or a proton that is placed in a strong magnetic field, used especially in spectroscopic studies of molecular structure and in medicine to identify and measure molecules involved in metabolism.

[0036] As used herein, "direct flow injection mass spectrometry" or "DI-MS" refers to an analytic technique by which chemical substances are identified by sorting gaseous ions by mass using electric and magnetic fields, wherein sample introduction is by direct insertion with a probe or plate.

[0037] As used herein, "nuchal translucency" or "NT" refers to the measurement of the size of the subcutaneous translucent space behind the neck of the fetus using ultrasound at between 10 and 14 weeks of pregnancy, reflecting the amount of fluid that has accumulated under the skin of the fetus. Nuchal translucency is commonly used as a predictor of chromosome disorders such as, trisomy 18, Turner syndrome and Down syndrome.

[0038] The term "patient," as used herein, means an animal, preferably a human.

[0039] In the first such report the present invention demonstrates the feasibility of use of biomarkers for the prediction of CHD. Abnormal lipid metabolism appears to be a significant feature of CHD pregnancies.

[0040] While abnormality in metabolite levels in the folate pathway such as homocysteine and metabolites related to oxidative stress have been previously reported our study represents the first comprehensive untargeted metabolomics study for the prenatal prediction of CHD. The serum metabolomics profile of a first trimester pregnant women carrying CHD fetus in this study found significant elevation of acylcarnitines. Carnitine (β-hydroxy-y- trimethylammonium butyrate) is a substance that plays a key role in the transfer of fatty acids into the mitochondria for metabolism and energy release. Long chain (multiple carbons) fatty acids bind with carnitine to form acylcarnitines which are transported into the mitochondria for sequential shortening which occurs two carbons at a time. This process is associated with the generation of potential energy stored in ATP. During periods of starvation these fatty acids constitute the main source of energy for skeletal muscle. Approximately 70% of myocardial energy is provided by mitochondrial fatty acid oxidation as described above.

[0041] Abnormality of folate metabolism has been linked to CHD in human and animal studies. Choline is an important nutrient that plays a role in lipid metabolism and in the formation of phosphotidyl choline for cell membrane synthesis. The two major roles of choline are for phospolipid synthesis and to serve as a methyl donor. Choline is oxidized to betaine in the mitochondria and betaine serves as an actual methyl donor which converts homocysteine to methionine. Increased levels of methionine are repeatedly associated with reduced risk of CHD while elevated homocysteine (a metabolic result of decreased methionine) is associated with increased CHD risk. Choline deficiency is also associated with increased lipid accumulation in the liver. There is thus a link between lipid metabolism and single carbon metabolism. Of note, in our data set there was a reduction in carnitine levels in CHD versus normal pregnancies. Providing further potential evidence of a metabolic disturbance in this pathway.

[0042] Disturbances of phosphatidyl choline metabolism is a prominent feature of several cancers including breast cancer. Cancer is a disorder characterized by rapid cell growth, division and apoptosis. Given the critical role of phosphotidylcholine in cell membranes, disturbance in choline metabolism would therefore appear understandable. Organogenesis in the embryonic period has obvious similarities to cancer. It is therefore possible that in CHD abnormalities of tissue re-modeling which affect the rate of cell membrane synthesis and destruction may be manifesting as abnormality of the choline and phosphotidylcholine metabolism.

[0043] Overall, we identified evidence of extensive phosphatidyl-choline and lipid abnormalities in the first trimester serum of pregnant women with CHD fetuses. Some of these metabolic abnormalities such as disturbance of carnitine levels and therefore lipid synthesis could plausibly be tied to aberrations of single carbon metabolism through choline. There is already extensive evidence of an association with altered homocysteine and methionine levels and the development of CHD, while the association with fetal CHD is new.

NMR Metabolomic Analysis

[0044] Prior publications have extensively described the use of the nuclear magnetic resonance (NMR) platform for metabolomic analysis of the serum. (Bahado-Singh RO, et al. Metabolomics and first-trimester prediction of early-onset preeclampsia. J Matem Fetal Neonatal Med 2012; 10: 1840-7; Bahado-Singh RO, et al. Metabolomic analysis for first- trimester Down Syndrome Prediction. Am J Obstet Gyencol 2013; 208; 371.el-8.) In brief the Varian Inova 500 MHz NMR spectrometer was used (International Equipment Treating limited, Vernon Hills, IL). Serum samples were filtered through 3-kd cut off centrifuge filter units (Amicon Micoron YM-3; Sigma-Aldrich, St. Louis, MO) to remove blood proteins. Three hundred and fifty microliters of samples was put into the centrifuge filter device and spun (10,000 rpm for 20 minutes) in order to remove macro molecules such as protein and lipoproteins. If the total volume of sample was <300 μΐ a 50-mmol NaH2P04 buffer (pH7) was added to reach a total sample volume of 300 μΐ. Metabolite concentrations were adjusted for the dilution due to the buffer. Thereafter, 35 μΐ of D ₂ 0 and 15 μΐ of buffer solution (11.667 mmol disodium-2, 2-dimethyl-2-silceptentane-5-sulphonate, 730 mmol imidazole and 0.47% NaN3 in H20) was added to the sample.

[0045] A total of 350 μΐ of sample was transferred to a micro cell NMR tube (Shigemi, Inc., Allison Park, PA). Ή-NMR spectra were collected on a 500-MHz Inova (Varian Inc, Palo Alto, CA) spectrometer with a 5-mm ITCN Z-gradient PFG cold-probe. The singlet produced by the disodium-2,2-dimethyel-2-silcepentane-5-sulphonate methyl groups was used as an internal standard by which to measure the chemical shift. The standard reference substance was set at 0 ppm and used for quantification of metabolites of interest. TheΉ- NMR spectra were analyzed with a Chenomx NMR Suite Professional Software package (Version 7.1 :Chenomx Inc. Edmonton, Alberta, Canada). This permits quantitative and qualitative analysis of the NMR spectrum observed. The observed NMR spectrum was manually fitted to an internal database. Each spectrum was evaluated by at least 2 NMR spectroscopists to rmnimize errors of quantitation and identification. Analysis of the metabolomics data was performed with the MetaboAnalyst web based statistical package.

Combined Direct Flow Injection and LC-MS MS compound identification and quantification

[0046] We applied a targeted quantitative metabolomics approach to analyze the serum samples using a combination of direct injection mass spectrometry (Absolute/Dg™ Kit) with a reverse-phase LC-MS/MS Kit. The Kit is a commercially available assay from

BIOCRATES Life Sciences AG (Austria). This kit, in combination with an ABI 4000 Q-Trap (Applied Biosystems/MDS Sciex) mass spectrometer, can be used for the targeted identification and quantification of up to 180 different endogenous metabolites including amino acids, acylcarnitines, biogenic amines, glycerophospholipids, sphingolipids and sugars. The method used combines the derivatization and extraction of analytes, and the selective mass-spectrometric detection using multiple reaction monitoring (MRM) pairs. Isotope-labeled internal standards and other internal standards are integrated in Kit plate filter for metabolite quantification. The Absolute/Dg kit contains a 96 deep-well plate with a filter plate attached with sealing tape, and reagents and solvents used to prepare the plate assay. First 14 wells in the Kit were used for one blank, three zero samples, seven standards and three quality control samples provided with each Kit. All the serum samples were analyzed with the AbsolutelDQ kit using the protocol described in the AbsolutelDQ user manual. Briefly, serum samples were thawed on ice and were vortexed and centrifuged at 13,000x g. plate and dried in a stream of nitrogen. Subsequently, 20 μΐΐ. of a 5% solution of phenyl- isothiocyanate was added for derivatization. After incubation, the filter spots were dried again using an evaporator. Extraction of the metabolites was then achieved by adding 300 μί, methanol containing 5 mM ammonium acetate. The extracts were obtained by centrifugation into the lower 96-deep well plate, followed by a dilution step with kit MS running solvent. Mass spectrometric analysis was performed on an API4000 Qtrap® tandem mass spectrometry instrument (Applied Biosystems/MDS Analytical Technologies, Foster City, CA) equipped with a solvent delivery system. The samples were delivered to the mass spectrometer by a LC method followed by a direct injection (DI) method. The Biocrates MetlQ software was used to control the entire assay workflow, from sample registration to automated calculation of metabolite concentrations to the export of data into other data analysis programs. A targeted profiling scheme was used to quantitatively screen for known small molecule metabolites using multiple reaction monitoring, neutral loss and precursor ion scans.

[0047J The data were log normalized. Principal component analysis is a multivariate technique and was used to find the two or three most useful metabolites (principal components) for distinguishing the two patient groups of interest. The first principal component is the single most discriminating metabolite while the second most discriminating metabolite is labeled the second principal component and so on.

[0048] Partial least squares discriminant analysis (PLS-DA) rotates around the different principal components to find the optimal combination for discriminating the case from the control group. Permutation analysis uses random re-sampling of case and controls to determine the probability that the observed and control groups is the result of chance. A total of 2000 re-samplings were performed. A p-value which represents the probability of a chance finding is generated. A variable importance in projection (VIP) plot (19) is a visual representation of the significance or importance of the particular metabolite in discriminating the groups of interest. The higher the VTP score, the more useful the metabolite for distinguishing normal from the affected group.

Statistical Analysis

[0049] All data were analyzed using software (Rand MetaboAnalyst). Univariate analysis of continuous data was conducted using Wilcoxon-Mann- Whitney test and categorical data were analyzed using Pearson Chi-Square and Fisher exact tests. Multivariate analyses were conducted using stepwise binary logistic regression with selected features using Lasso algorithm. Independent samples t-test, chi-square and Fisher exact tests were performed. A significance level of PO.05 was used to define statistical significance.

[0050] Three different sets of analyses were performed. Metabolites were analyzed by themselves and also with addition of demographic such as, ethnicity, BMI, parity and an ultrasound marker namely CRL. Finally metabolites with Nuchal translucency (NT) were evaluated. It should be pointed out the Crown Rump Length (CRL) ultrasound measurement of the fetus is the most precise measure of gestational age and therefore used to assess whether first trimester gestational age affected maternal serum level of the metabolites.

[0051] For each metabolite, data normalization is critical to creating a normal or Gaussian distribution. Normalization allows conventional statistical tests to be performed, and it simplifies data interpretation. In this study we used log-transformed metabolite values, as noted previously.

[0052] Metabolite concentrations in CHD versus controls were compared. Stepwise logistic regression analyses were performed with CHD as the dependent variable and metabolites as the independent or determinant variable in order to develop a predictive algorithm for CHD detection. Metabolites with significant correlation with CHD status on univariate analysis were initially entered into the regression model development. Other variables including NT, fetal Crown Rump Length (CRL) and maternal demographics and medical status were combined with metabolite concentrations and run in the regression analyses. Finally, regression analyses including NT and the preceding metabolomic and maternal markers were performed for the prediction of CHD. Based on these analyses, several different regression equations for predicting the individual risk of CHD were developed. Individual risk or probability of CHD was calculated for each patient in the study. A series of paired sensitivity and FPR (1 -specificity) values at different risk thresholds were calculated. A receiver operator characteristic (ROC) curve is plotted with sensitivity values on the Y-axis and the corresponding FPR (1 -specificity) on the X-axis. The area under the ROC curve (AUC) indicates the accuracy of a test for correctly distinguishing a disorder such as CHD cases from normal (control). An AUC=1.0 indicates a perfect discriminating test. The 95% CI and p-values for the AUC curves were calculated. Permutation testing was also performed to determine the probability that the observed AUC obtained was due to chance.

[0053] Using direct injection mass spectrometry with LC-MS/MS (DI-MS) and NMR metabolomic platforms, numerous metabolites were identified in maternal serum that distinguished chromosomally normal versus first trimester CHD cases. The principal metabolite group identified was the acylcarnitines. This chemical group represents intermediates involved in the transport and metabolism of fatty acids in the mitochondria. In addition we demonstrated that the combination of a limited number of metabolites by themselves, e.g. hydroxypropionylcarnitine, glutaconylcarnitine and

hydroxytetradecenoylcamitine appeared to be highly accurate predictors of CHD status. The sensitivity of this combination of metabolites was 92% at a specificity threshold of 93.2%. These values were highly statistically significant.

[0054] When metabolites identified by the NMR platform alone were analyzed, only limited diagnostic accuracy was achieved. The combination of acetone and ethanol had a 67.9% sensitivity at 67.8% specificity. Though modest, these values were again statistically significant.

[0055] Nuchal translucency measurements are an important feature of first trimester ultrasound for aneuploidy risk determination. Several studies have confirmed a modest correlation between NT measurements in the first trimester and the risk of CHD. Higher NT measurements are said to correlate with increased CHD risk even in chromosomally normal fetuses. We therefore also looked at the combination of metabolite markers with NT measurement for the detection of CHD. While it was found to be a statistically significant predictor of CHD by itself, overall, there was no additional benefit of adding NT

measurements to the metabolite markers. There was an approximately 4% increase in sensitivity and specificity when NT measurement was added to the combination of acetone and ethanol, this increase was not statistically significant however.

EXAMPLES

[0056] Study Design: Mass spectrometry and NMR based metabolomic analyses were performed between 11 weeks and 13 weeks 6 days gestation on maternal serum. A total of 27 CHD cases and 59 controls were compared. There were no known or suspected chromosomal or syndromic abnormalities indicated.

[0057] The objectives of the current study are twofold. First, to determine whether there are significant differences in the first-trimester maternal metabolomic profile in pregnancies with a chromosomally normal fetus compared to those affected with a CHD. Secondly, to evaluate metabolite biomarker algorithms that might be useful for the first-trimester prediction of fetal CHD using maternal serum. [0058] Methods: This study is part of an ongoing prospective study for the first-trimester detection and prediction of fetal and maternal disorders. The details of specimen collection has been extensively described. (Bahado-Singh RO, et al. Metabolomics and first-trimester prediction of early-onset preeclampsia. J Matern Fetal Neonatal Med 2012; 10:1840-7;

Bahado-Singh RO, et al. Metabolomic analysis for first-trimester Down Syndrome

Prediction. Am J Obstet Gyencol 2013; 208; 371.el-8.) The patients were prospectively recruited from an average risk population in Britain between 2003-2009. IRB approval was obtained through the institutional review board of King's College Hospital, London, England. Each recruited patient signed a written consent. Crown Rump Length (CRL) was used to estimate gestational age. Routine first-trimester screening for aneuploidy is the current standard of care. Maternal demographic and clinical data was obtained along with serum for PAPP-A and free B-hCG. Nuchal translucency measurement was measured for aneuploidy risk estimation. Karyotype and or newborn exams were performed to assess chromosomal status. CHD status was determined by prenatal imaging and or postnatal imaging and based on physical exam in the normal cases.

[0059] Results: A total of 171 metabolites were evaluated. We identified 118 metabolites that demonstrated significant differences in maternal first-trimester serum in CHD versus normal cases. There was significant disturbance in acylcarnitine levels in CHD pregnancies. Predictive algorithms were developed for CHD detection. High sensitivity (95% CI) = 0.97 (0.92, 1.00) and specificity (95% CI) = 0.89 (0.78, 1.00) for CHD detection was achieved AUC (95% CI) =99.2 (0.973, 1.0).

[0060] There were 27 cases of CHD and 59 normal controls in which metabolomic analysis was able to be completed. Neither CHD or control fetuses had any known or suspected chromosomal or syndromic abnormalities. Table 1 gives the breakdown of the different types of CHD. Table 2 compares maternal pregnancy and other demographic characteristics between study and control groups. There were no significant difference observed. A total of 133 metabolites were identified and analyzed using DI-MS-direct flow injection mass spectrometry metabolomic platform. For NMR analysis a total of 38 metabolites were analyzed.

Table 1: Types of CHD Cases

AVSD/DORV

AVSD/DORV/PA

DORV/PS DORV/PS

DORV/TOF

DORV/TOF

DORV/PA

TGA (3 cases)

TGA - corrected/VSD

TGA/PS

TOF (10 cases)

TOF/MS

TOF/PA (5 cases)

AVSD - Atrioventricular Septal Defect

DORV - Double Outlet Right Ventricle

MS - Mitral Stenosis

PA - Pulmonary Atresia

PS - Pulmonary Valve Stenosis

TGA - Transposition of the Great Artery

TOF - Tetralogy of Fallot

[0061] Table 2:

[0062] Maternal Demographic and medical characteristics: Comparison of CHD and Control Groups

BMI, mean (SD) ^a 24.1 (4.2) 24.4 (3.5) ns

GA-CPvL (weeks), mean (SD) ^a 12.7 (0.7) 12.7 (0.6) ns

CHD: congenital heart defect

ns: not significant

a: Independent sampes t test

b: X ² test

c: Fisher exact test

GA-CRL - gestational age in weeks based on crown rump length

[0063] Using a DI-MS and NMR metabolomic analysis a total of 118 metabolites, were found to have significant concentration differences in maternal serum in CHD versus normal control cases on paired comparisons. The mean (SD) concentrations, T-test statistic and P- values for each metabolites along with fold change and direction of change in CHD cases relative to controls are provided in Table 3 for DI-MS analysis.

[0064] Table 3: Univariate Analysis for DI-MS: CHD vs. Control

C14:2 (Tetradecadienylcarnitine) 0.01(0.01) 0.02(0.01) -1.6 Down 0.0001

C14:2-OH

(Hydroxytetradecadienylcaraitine) 0.01(0.01) 0.01(0) -1.2 Down 0.0039

C16 (Hexadecanoylcamitine) 0.05(0.04) 0.08(0.02) -1.67 Down 0.0000

C16:2 (Hexadecadienylcarnitine) 0.01(0.01) 0.01(0) -1.21 Down 0.0068

CI 8 (Octadecanoylcamitine) 0.03(0.02) 0.04(0.01) -1.44 Down 0.0005

C18:l (Octadecenoylcarnitine) 0.05(0.04) 0.08(0.04) -1.75 Down 0.0000

C18:2 (Octadecadienylcarnitine) 0.02(0.02) 0.03(0.01) -1.54 Down 0.0011 lysoPC a C16:0

(lysoPhosphatidylcholine acyl CI 6:0) 72.19(66.13) 142.06(39.64) -1.97 Down 0.0000 lysoPC a C16:l

(lysoPhosphatidylcholine acyl C16:l) 1.86(1.86) 2.68(1.01) -1.45 Down 0.0155 lysoPC a C17:0

(lysoPhosphatidylcholine acyl CI 7:0) 1.88(2) 2.64(0.81) -1.4 Down 0.0044 lysoPC a CI 8:0

(lysoPhosphatidylcholine acyl CI 8:0) 19.32(17.98) 36.49(11.67) -1.89 Down 0.0001 lysoPC a C18:l

(lysoPhosphatidylcholine acyl CI 8:1) 13.38(12.74) 27.95(9.47) -2.09 Down 0.0000 lysoPC a C18:2

(lysoPhosphatidylcholine acyl C 18:2) 17.58(17.13) 36.34(14.31) -2.07 Down 0.0000 lysoPC a C20:3

(lysoPhosphatidylcholine acyl C20:3) 1.64(1.55) 2.74(1.13) -1.67 Down 0.0011 lysoPC a C20:4

(lysoPhosphatidylcholine acyl C20:4) 4.22(3.72) 7.99(2.26) -1.89 Down 0.0000 lysoPC a C26:0

(lysoPhosphatidylcholine acyl C26:0) 1.04(1.05) 0.73(0.23) 1.42 Up 1.0000 lysoPC a C28:0

(lysoPhosphatidylcholine acyl C28:0) 0.94(1.03) 0.75(0.2) 1.26 Up 0.4399 lysoPC a C28:l

(lysoPhosphatidylcholine acyl C28:l) 0.83(0.79) 0.83(0.2) -1.01 Down 0.4399

PC aa C24:0 (Phosphatidylcholine

diacyl C24:0) 0.24(0.25) 0.16(0.04) 1.48 Up 0.7517

PC aa C26:0 (Phosphatidylcholine

diacyl C26:0) 1.46(0.88) 1.21(0.24) 1.2 Up 0.5766

PC aa C28:l (Phosphatidylcholine

diacyl C28:l) 2.3(1.98) 3.44(1.12) -1.5 Down 0.0168

PC aa C30:0 (Phosphatidylcholine

diacyl C30:0) 3.57(3.29) 6.08(2.74) -1.71 Down 0.0015

PC aa C30:2 (Phosphatidylcholine

diacyl C30:2) 0.27(0.34) 0.59(0.19) -2.17 Down 0 .0000

PC aa C32:0 (Phosphatidylcholine

diacyl C32:0) 11.83(10.25) 21.47(6.56) -1.81 Down 0.0003

PC aa C32:l (Phosphatidylcholine

diacyl C32:l) 12.4(12.89) 23.35(12.85) -1.88 Down 0.0003

PC aa C32:2 (Phosphatidylcholine

diacyl C32:2) 3.4(3.23) 6.59(2.65) -1.94 Down 0.0001

PC aa C32:3 (Phosphatidylcholine

diacyl C32:3) 0.53(0.47) 0.88(0.22) -1.68 Down 0.0014

PC aa C34:l (Phosphatidylcholine 168.65(151.

diacyl C34:l) 58) 316.44(96.26) -1.88 Down 0.0001

PC aa C34:2 (Phosphatidylcholine 284.99(253.

diacyl C34:2) 25) 512.41(109.1 -1.8 Down 0.0002 PC aa C42:5 (Phosphatidylcholine diacyl

C42:5) 0.52(0.47) 0.81(0.26) -1.58 Down 0.0052

PC aa C42:6 (Phosphatidylcholine diacyl

C42:6) 0.93(0.45) 1.15(0.3) -1.23 Down 0.0123

PC ae C30:0 (Phosphatidylcholine acly- alkyl C30:0) 0.55(0.42) 0.58(0.22) -1.07 Down 0.1341

PC ae C30:l (Phosphatidylcholine acly- alkyl C30:l) 0.37(0.35) 0.3(0.12) 1.22 Up 0.9481

PC ae C32:l (Phosphatidylcholine acly- alkyl C32:l) 2.23(1.98) 3.8(1.1) -1.7 Down 0.0009

PC ae C32:2 (Phosphatidylcholine acly- alkyl C32:2) 0.7(0.63) 1.05(0.26) -1.51 Down 0.0085

PC ae C34:0 (Phosphatidylcholine acly- alkyl C34:0) 1.24(1.1) 1.96(0.68) -1.58 Down 0.0045

PC ae C34:l (Phosphatidylcholine acly- alkyl C34:l) 7.95(7.13) 14.62(4.86) -1.84 Down 0.0002

PC ae C34:2 (Phosphatidylcholine acly- alkyl C34:2) 9.03(8.28) 15.62(5.11) -1.73 Down 0.0004

PC ae C34:3 (Phosphatidylcholine acly- alkyl C34:3) 6.56(6.11) 11.14(3.09) -1.7 Down 0.0003

PC ae C36:0 (Phosphatidylcholine acly- alkyl C36:0) 1(1.17) 1.04(0.31) -1.04 Down 0.0195

PC ae C36:l (Phosphatidylcholine acly- alkyl C36:l) 7.06(6.48) 11.33(3.59) -1.6 Down 0.0048

PC ae C36:2 (Phosphatidylcholine acly- alkyl C36:2) 12.11(10.98) 20.84(6.54) -1.72 Down 0.0004

PC ae C36:3 (Phosphatidylcholine acly- alkyl C36:3) 6.38(5.93) 11.03(3.88) -1.73 Down 0.0005

PC ae C36:4 (Phosphatidylcholine acly- alkyl C36:4) 12.7(10.97) 21.9(6.57) -1.72 Down 0.0008

PC ae C36:5 (Phosphatidylcholine acly- alkyl C36:5) 8.03(6.96) 14.27(3.66) -1.78 Down 0.0001

PC ae C38:0 (Phosphatidylcholine acly- alkyl C38:0) 2.41(2.07) 3.92(1.3) -1.63 Down 0.0034

PC ae C38:l (Phosphatidylcholine acly- alkyl C38:l) 0.88(1.24) 0.88(0.41) -1 Down 0.0114

PC ae C38:2 (Phosphatidylcholine acly- alkyl C38:2) 1.95(1.97) 3.01(0.98) -1.55 Down 0.0013

PC ae C38:3 (Phosphatidylcholine acly- alkyl C38:3) 3.9(3.58) 6.05(2.14) -1.55 Down 0.0100

PC ae C38:4 (Phosphatidylcholine acly- alkyl C38:4) 9.65(8.29) 17.69(5.25) -1.83 Down 0.0001

PC ae C38:5 (Phosphatidylcholine acly- alkyl C38:5) 11.95(10.36) 22.83(6.42) -1.91 Down 0.0000

PC ae C38:6 (Phosphatidylcholine acly- alkyl C38:6) 6.04(5.42) 11.04(3.14) -1.83 Down 0.0002

PC ae C40:l (Phosphatidylcholine

acly-alkyl C40:l) 1.09(1.02) 1.76(0.55) -1.62 Down 0.0011

PC ae C40 ^(Phosphatidylcholine

acly-alkyl C40:2) 1.68(1.53) 2.65(0.75) -1.57 Down 0.0062

PC ae C40:3 (Phosphatidylcholine

acly-alkyl C40:3) 1.2(1.17) 1.7(0.52) -1.41 Down 0.0155

PC ae C40:4(Phosphatidylcholine 1.89(1.67) 3.22(1.01) -1.71 Down 0.0017

SM C26:l (Sphingomyeline

C26:l) 0.33(0.29) 0.54(0.14) -1.64 Down 0.0057

3183.4(2827 5506.57(1243

HI (Hexose) .17) .87) -1.73 Down 0.0009

"Non-Parametric testing.

[0065] In Table 4, a similar comparison of metabolite concentrations was performed for only NMR based metabolomics. Significant differences were noted in five metabolites using the NMR platform.

0066] Table 4: Univariate Analysis for NMR : CHD vs Control

Succinate 3.35(1.78) 3.31(1.13) 1.01 Up 0.5506

Threonine 112.57(25.66) 107.31(26.21) 1.05 Up 0.3449

Tyrosine 42.94(11.38) 45.63(19.15) -1.06 Down 0.9296

Valine 147.88(37.36) 143.24(30.09) 1.03 Up 0.5148

-Methylhistidine 26.93(13.26) 23.74(15.16) 1.13 Up 0.0185*

[0067] Using DI-MS data alone, PCA showed clustering of the control from the CHD specimens. This indicates that the two groups could be significantly distinguished based on a limited number (two) of metabolites. The most discriminating metabolite, PCI or principal component 1, accounted for 84.0% of the observed separation between groups. The second metabolite PC2 accounted for 3.5% of the separation. Two-dimensional PLS-DA further confirmed the discrimination of the CHD from control groups that was achieved by these metabolites. In that PLS-DA plot, one metabolite accounted for 83.9% of the discrimination (PCI) while the second metabolite was responsible for 2.7% of the separation between groups. The 3-D PLS-DA plot (figure 1) of the DI-MS identified metabolites similarly demonstrated excellent separation of the two groups using three rather than two

discriminating metabolites. VIP scores plot for the DI-MS based metabolites is shown in Figure 2. The plot shows metabolites with VTP scores >1. The following three metabolites C3-OH, C5.0H(C3DC) and CI 4:1 appeared most predictive for CHD detection with DI-MS analysis.

[0068] Permutation testing using n=2000 re-samplings showed a highly statistically significant p value, <0.005, indicating that the observed separation between groups was unlikely to be due to chance (Figure 3).

[0069] A similar series of analyses were performed using metabolites detected based on the NMR platform. Some clustering of cases relative to controls was observed on PCA analysis; however the separation was not as marked as for the DI-MS analysis. Separation of CHD and controls were observed on 2-D PLS-DA analysis, but were improved when 3 rather than 2 principal components were used. See, 3D plot (figure 4). Permutation testing using 2000 re-samplings for the NMR analysis also yielded a low probability that the observed separation was due to chance, p=0.002. The corresponding VIP plot (figure 5) showed acetone, ethanol, acetate and pyruvate to be the 4 most discriminating metabolites using NMR analysis.

[0070] Using logistic regression analysis the individual probability of a fetus having CHD was calculated using 3 metabolites from the DI-MS based metabolomics results, namely C3- OH, C5:1-DC and C14:2-OH (Table 5). The logistic equations was represented by risk of CHD=f[-42.582-l 2.039 log (C3-OH)+3.194 log (C5:l-DC)-4.050 log (C14:2-OH)], where f represents the logistic function. The table shows the contribution of each of these DI-MS based metabolites to the CHD prediction model. The ROC curve (figure 5) indicates that this metabolite combination was a highly significant predictor of CHD:AUC (95% CI)=0.981 (0.942, 0.999). The sensitivity (95% CI) and specificity (95% CI) of the algorithms were statistically significant: 0.98 (0.95, 1.00) and 0.82 (0.68, 0.96) respectively. Permutation testing for the optimal model was performed using 2000 random samples and indicated a low probability that the diagnostic accuracy represented by the area under the ROC curve was due to chance, p<0.0005. The sensitivities at different false positive thresholds for the algorithm or NMR-based analysis only are shown in Table 6. On a further analysis CRL, ethnicity, BMI or parity did not contribute significantly to CHD prediction using metabolites.

Table 5: Logistic Regression Based Optimal Model for CHD Detection: DI-MS Metabolites Only

Null deviance: 109.32 or 86 degrees of freedom

Residual deviance: 27.67 or 83 degrees of freedom

CHD=f 42.582 - 12.039 log (C3-OH) + 3.194 log (C5:1-DC) - 4.050 log (cl4.2-OH)] represents the equation for the prediction of CHD

Table 6: NMR only prediction of CHD

Metabolites* - acetone and ethanol

P-value represents the permutation testing p-value [0071] We also looked at the performance of the algorithm using the NMR only platform and also these metabolites combined with NT (Table 6). NT contributed only modestly and did not significantly improve performance over NMR only metabolites.

[0072] Using NT measurement only, the following predictive equation for CHD risk estimation was developed: CHD risk=f(-4.821+l .873 NT): where NT was the nuchal translucency measurement. The AUC (95% CI) for this algorithm was 0.753 (0.616, 0.867). Sensitivity (95% CI) was 0.97 (0.92, 1.00) and for specificity this was 0.54 (0.351, 0.72). The positive likelihood ratio was 2.08 with an accuracy of 0.83.

[0073] Analyses were also performed using both NMR and DI-MS metabolites, and NT ultrasound measurement for a CHD prediction. That prediction model was represented by the logistic equation CHD=f[-58.0591+2.1678NT-14.2494 log (C3-0H)+2.9807 log (C5:1-DC) 4.6776 log (C14:2-OH). This combination was a highly statistically significant predictor of CHD with AUC (95% CI): 0.992 (0.973, 1.0). The sensitivity (95% CI) of the model was 0.97 (0.92, 1.00) and the specificity (95% CI) of this model was also statistically significant: 0.89 (0.78, 1.00). The overall accuracy based on combining all metabolites and NT was 0.94 for discriminating the two groups was 0.94 with a false discovery rate of 0.05%. Permutation testing using 2000 resamples confirmed a low probability that the observed area under the ROC curve was due to chance, p<0.0005. This did not however represent a significant improvement over the metabolites only algorithm as shown in table 7.

[0074] Table 7: CHD Prediction based on limited metabolite combinations: DI-MS based Metabolites

NT - nuchal translucency

P-value - permutations testing p-value OTHER EMBODIMENTS

[0075] All publications and patents referred to in this disclosure are incorporated herein by reference to the same extent as if each individual publication or patent application were specifically and individually indicated to be incorporated by reference. Should the meaning of the terms in any of the patents or publications incorporated by reference conflict with the meaning of the terms used in this disclosure, the meaning of the terms in this disclosure are intended to be controlling. Furthermore, the foregoing discussion discloses and describes merely exemplary embodiments of the invention. One skilled in the art will readily recognize from such discussion and from the accompanying drawings and claims, that various changes, modifications and variations can be made therein without departing from the spirit and scope of the invention as defined in the following claims.