A plasma metabolite signature and kit for predicting spontaneous preterm birth

Abstract

The present invention belongs to the field of medical prevention, and specifically relates to a plasma metabolite characteristic and a kit for predicting spontaneous premature birth. Among them, spontaneous premature birth is divided into premature birth with intact membranes (sPTB) and premature birth after premature rupture of membranes (pPROM). The plasma metabolite characteristic includes ten mass-to-charge ratios, including m / z 357.24963, m / z 289.25259, m / z 250.18016, m / z 279.15909, and m / z 184.07332. The present invention detects maternal blood samples, clarifies the changes in pPROM and sPTB metabolites, and uses metabolites to predict in the second trimester, providing methods and markers for the prediction of pPROM and sPTB. The present invention can accurately predict spontaneous premature birth of two different etiologies, pPROM and sPTB, with high efficiency, and has guiding significance for subsequent preventive interventions.

Description

Technical Field

[0001] The invention belongs to the field of medical prevention, and in particular relates to a plasma metabolite characteristic and a kit for predicting spontaneous premature birth. Background Art

[0002] Preterm birth refers to delivery before 37 weeks of gestation. The latest data shows that the global preterm birth rate is 9.9% of live births, which is the main cause of neonatal morbidity and mortality worldwide. my country's 2024 Clinical Prevention and Treatment Guidelines for Premature Birth point out that premature birth is mainly divided into two categories: therapeutic premature birth and spontaneous premature birth, the latter of which can be further divided into premature birth with intact membranes (sPTB) and premature birth after premature rupture of membranes (pPROM).

[0003] According to the guidelines for premature birth, there is currently no effective way to predict premature birth. Studies have shown that:

[0004] (1) Although risk factors such as a history of previous miscarriage or premature birth and a mid-pregnancy cervical length ≤ 25 mm have been recommended by multiple guidelines for predicting premature birth, their effectiveness is low and the measures are limited;

[0005] (2) Currently, there are many studies on biochemical molecular markers such as fetal fibronectin, phosphorylated insulin-like growth factor binding protein 1, PAMG-1, and inflammatory cytokines, which are still controversial in predicting premature birth and are still under further exploration and practice;

[0006] (3) Studies on the use of ultrasound to measure cervical length and cervical angle to predict premature birth have not yet yielded consistent results, and the predictive value needs further exploration;

[0007] (4) Other researchers have tried to use combined indicators to predict preterm birth. For example, fetal fibronectin combined with cervical length assessment can effectively improve the prediction of preterm birth, but the evidence is not sufficient and further research is needed.

[0008] (5) Researchers are constantly exploring the path to predicting premature birth and have discovered that some lipid biomarkers and differences in vaginal flora levels are highly correlated with premature birth. Other factors such as anxiety, pain, fatigue, and happiness are also associated with an increased risk of premature birth.

[0009] In general, the prediction of preterm birth is still in the exploratory stage, and no exact biomarkers have been found in the clinic to predict the risk of preterm birth, especially the difference between sPTB and pPROM in spontaneous preterm birth. During pregnancy, a large number of physiological changes and metabolic adaptations occur every week. Metabolomics is a method that can comprehensively analyze metabolites in organisms and provide new perspectives for the early diagnosis of diseases. In 2020, Mads Melbye et al. established a metabolic clock that not only highly matches the prediction of gestational age by early pregnancy ultrasound (the clinical gold standard), but also restores individual pregnancy changes that are not detected by ultrasound but may affect the timing of delivery. It is reported that there are not many studies on the changes in blood metabolomics in patients with premature birth using ultra-high performance liquid chromatography-mass spectrometry (UPLC-MS). One study measured the changes in metabolites in the serum of premature and normal pregnant women after 24 weeks of pregnancy, but did not distinguish between sPTB and pPROM. There is also a study that showed that in cases of preterm birth, elevated lipid levels and changes in key endpoint metabolites were observed in maternal serum (between 29 weeks + 0 days and 36 weeks + 5 days of gestation). Therefore, the detection of metabolome during pregnancy is likely to become a way to predict preterm birth, but to date, there is no technology that uses plasma samples of maternal spontaneous preterm birth in mid-pregnancy for non-targeted metabolomics research and predicts preterm birth, especially the separate prediction of two diseases with different possible causes of spontaneous preterm birth, sPTB and pPROM.

[0010] Therefore, based on this, the technical solution of the present invention is proposed. Summary of the invention

[0011] In order to solve the problems existing in the prior art, the present invention provides a plasma metabolite feature for predicting spontaneous preterm birth (including sPTB and pPROM), wherein the plasma metabolite feature comprises the following mass-to-charge ratios: m / z 357.24963, m / z 289.25259, m / z 502.29281, m / z 478.29282, m / z 319.19039, m / z 250.18016, m / z279.15909, m / z 184.07332, m / z 290.15981, m / z 155.14304.

[0012] Based on the same technical concept, another embodiment of the present invention is to provide a kit for predicting spontaneous preterm birth, wherein the kit includes a plasma metabolite characteristic for predicting spontaneous preterm birth, and the plasma metabolite characteristic comprises the following mass-to-charge ratios: m / z 357.24963, m / z 289.25259, m / z 502.29281, m / z 478.29282, m / z319.19039, m / z 250.18016, m / z 279.15909, m / z 184.07332, m / z 290.15981, and m / z155.14304.

[0013] Preferably, the kit further comprises extracting solution 1, extracting solution 2 and extracting solution 3; wherein:

[0014] The extract 1 is a methanol-water solution;

[0015] And / or, the extract 2 is an aqueous formic acid solution and an ammonium formate aqueous solution;

[0016] And / or, the extract 3 is acetonitrile.

[0017] Preferably, the volume concentration of methanol in the methanol aqueous solution is 80%;

[0018] and / or, in the formic acid aqueous solution, the volume concentration of formic acid is 0.1%;

[0019] And / or, in the ammonium formate aqueous solution, the concentration of ammonium formate is 2 mmoL / L.

[0020] Preferably, the test sample of the kit is a blood sample of a pregnant woman. Metabolites in blood are relatively stable, while other metabolites, such as metabolites in urine, are easily affected by changes in the environment, diet, etc.; vaginal secretions are difficult to sample, prone to ascending infection, and pregnant women are unwilling to cooperate. As a result, the test results are more error-prone, and the probability of false negative or false positive results is higher.

[0021] Preferably, the blood sample is a peripheral blood sample collected from a pregnant woman at 14-20 weeks of pregnancy.

[0022] In practice, metabolites are chemical entities transformed during metabolism, providing a functional readout of cellular biochemistry. As intermediate or end products of metabolic reactions, their levels are affected by a variety of factors, and many metabolites are often products or substrates of enzyme reactions, providing more biologically relevant information. Metabolomics has the sensitivity to detect subtle changes in biological pathways, quantitatively measuring thousands of metabolites from minimal amounts of biological material and performing analysis.

[0023] The expansive applications of the present invention are as follows:

[0024] 1. Detection and analysis of plasma metabolites to deeply explore disease mechanisms

[0025] The plasma metabolome evaluates hundreds of small molecules through non-targeted metabolomics, linking cellular pathways with biological mechanisms, and reshaping our understanding of the evolution and progression of diseases. These small molecules may carry key reaction information, allowing us to clearly and intuitively understand the metabolic process of disease occurrence and development. With the establishment of the human metabolome database, in the future we can analyze and compare the changes in key metabolites, deduce and support more accurate hypotheses on disease mechanisms, and open up new directions for disease diagnosis and treatment. The specific applications of this technology include but are not limited to:

[0026] ① Research on metabolic mechanisms of pregnancy-related diseases;

[0027] ②Exploration of therapeutic intervention targets for pregnancy-related diseases;

[0028] ③Study on drug toxicity during pregnancy;

[0029] ④Prediction of the risk of pregnancy-related diseases and stratified management and prevention;

[0030] ⑤ Research on the mechanism of labor initiation and prediction of labor trajectory.

[0031] 2. Plasma metabolomics analysis for the prevention and treatment of pregnancy-related diseases

[0032] Metabolomics is increasingly being used in pregnancy-related diseases such as gestational diabetes and eclampsia. Studies have found that lipid metabolites change significantly in pregnant women with gestational diabetes, and previously unknown metabolites associated with gestational diabetes with strong discrimination have been discovered. New discoveries in metabolomics may bring new ideas for disease diagnosis and treatment, solve key problems, build a more complete diagnosis and treatment system, and improve the prognosis of mothers and children. At the same time, by analyzing plasma metabolomics to discover key differential metabolites, comparing their changes in normal pregnancy and pregnancy-related diseases, disease prediction is performed to facilitate early intervention. The specific applications of this technology include but are not limited to:

[0033] ① Exploration of the pathogenesis of pregnancy-related diseases such as gestational diabetes;

[0034] ②Study on maternal endocrine changes during pregnancy;

[0035] ③Nutrition research during pregnancy;

[0036] ④ Early prediction of pregnancy-related diseases such as premature birth and fetal growth restriction;

[0037] ⑤Study on fetal growth and development changes in twin pregnancy;

[0038] ⑥ Research on the pathogenesis of neonatal diseases;

[0039] ⑦Study on metabolic clocks in normal and disease states;

[0040] When using plasma metabolites to study the pathogenesis of a disease, it is necessary to ensure that the plasma source is sufficient, meets the research standards, and is not contaminated by other metabolites. It should be stored under unified conditions before formal testing to avoid repeated freezing and thawing to ensure that the subsequent metabolome analysis is accurate and reliable.

[0041] 3. Plasma metabolomics analysis provides new directions for drug development

[0042] Plasma metabolomics analysis technology provides an opportunity to deeply analyze the changes in plasma components at the metabolite level. Combined with drug metabolism kinetics research and evaluation, it can provide many references and new ideas for drug development. Further consider the relevant effects of medication during pregnancy and improve drug safety performance through tracking detection and toxicological evaluation. The specific applications of this technology include but are not limited to:

[0043] ① In-depth evaluation of drug treatment mechanisms and metabolic pathways during pregnancy, and prediction of drug effects;

[0044] ②Physiological monitoring, drug safety assessment and new drug development;

[0045] ③ Identify the side effects of pregnancy-related drugs and conduct drug treatment tests;

[0046] ④ Track the metabolic pathway of drugs in pregnant women through changes in metabolites;

[0047] ⑤ Reveal the potential mechanism of drug toxicity and make drug toxicity prediction;

[0048] ⑥ Promote precision medicine and personalized medication.

[0049] The beneficial effects of the present invention are:

[0050] The present invention uses liquid chromatography-mass spectrometry to perform non-targeted metabolomics detection on maternal mid-pregnancy (14-20 weeks) blood samples, analyzes the changes in metabolites in different types of spontaneous preterm birth, clarifies the changes in metabolites in different types of premature birth, and uses metabolites to predict different types of spontaneous preterm birth in mid-pregnancy, providing methods and markers for the prediction of sPTB and pPROM at an earlier time. Our invention can accurately predict sPTB and pPROM, two types of spontaneous preterm birth with different etiologies, with high efficiency, and has stronger guiding significance for subsequent preventive interventions.

[0051] The present invention screened out 10 differential metabolites that are significantly correlated with the two types of premature birth, sPTB and pPROM, from nearly a thousand metabolites in pregnant women's plasma. Compared with other disclosed invention patents, the 10 metabolites are more intuitive and clear, and only the changes in the 10 key metabolites need to be focused on. This can effectively save detection time, improve detection efficiency, and save costs. And prediction through metabolites is different from the previous exploration of premature birth prediction from the perspective of inflammatory markers, and it has opened up a new way to solve the industry-recognized problem of spontaneous premature birth prediction.

[0052] The preferred embodiment of the present invention also has the following beneficial effects:

[0053] The present invention extracts relevant plasma metabolites from the peripheral blood of pregnant women for detection. Compared with the scheme disclosed in the prior art, it not only adopts a non-invasive detection method to predict the risk of sPTB and pPROM, but also only takes maternal peripheral blood, which guarantees the safety of pregnant women and fetuses to the greatest extent. The plasma metabolites also have strong authenticity and stability, and are not easily affected by external factors such as environment and diet. While fully carrying molecular information, the operation is simple and easy. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0055] Figure 1 Random forest screened out the top 5 metabolites in terms of importance in pPROM vs. control group (which can reflect the weight of the metabolites).

[0056] Figure 2 Random forest screened out the top 5 metabolites in terms of importance in sPTB vs. control group (which can reflect the weight of the metabolites).

[0057] Figure 3 It is a box plot of the top 5 differential metabolite abundances in pPROM vs. control group.

[0058] Figure 4 It is a box plot of the top 5 differential metabolite abundances in sPTB vs. control group.

[0059] Figure 5 This is the data result diagram of the pPROM training set.

[0060] Figure 6 This is the data result diagram of the pPROM validation set.

[0061] Figure 7 This is the data result diagram of the sPTB training set.

[0062] Figure 8 This is the data result diagram of the sPTB validation set. DETAILED DESCRIPTION

[0063] To make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be described in detail below. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other implementation methods obtained by ordinary technicians in this field without creative work belong to the scope of protection of the present invention.

[0064] Example:

[0065] The materials and equipment used in this embodiment are shown in Table 1:

[0066]

[0067] 1. Sample collection:

[0068] The subjects were women aged between 18 and 45 years at the time of delivery, including two types of preterm birth: premature birth after premature rupture of membranes (20 cases), premature birth with intact membranes (20 cases), and full-term birth (30 cases). Peripheral blood samples were collected between 14 and 20 weeks of gestation. All subjects were from Tianjin Central Obstetrics and Gynecology Hospital.

[0069] (II) Sample processing:

[0070] (1) Centrifuge the freshly collected whole blood sample at 1500g and 4°C for 10 min, carefully collect the supernatant to obtain the plasma sample, take 100 μL of the sample and place it in an EP tube, and add 400 μL of 80% methanol aqueous solution (i.e., extract 1);

[0071] (2) Vortex and place on ice for 5 min, centrifuge at 15,000 g and 4 °C for 20 min, take a certain amount of supernatant and dilute it with mass spectrometry grade water to a methanol content of 53%, centrifuge at 15,000 g and 4 °C for 20 min, and collect the supernatant;

[0072] (3) Take an equal volume of sample from each sample and mix them as QC (quality control) samples. Set one QC sample every 6-10 experimental samples in the sample queue to verify the consistency and stability of the machine;

[0073] (4) Using the LC-MS method, the treated plasma samples were loaded onto a high performance liquid chromatography-mass spectrometer in a random order for continuous analysis to obtain the corresponding plasma metabolite information.

[0074] (III) Chromatographic separation of samples:

[0075] (1) First, adjust the chromatographic parameters: set mobile phase A to extract 2 (containing 0.1 vol.% formic acid and 2 mmol / L ammonium formate), set mobile phase B to extract 3 (i.e., acetonitrile); set gradient elution conditions: 0-1 min, 95% A; 1-5 min, 95%-40% A; 5-8 min, 40%-0% A; 8-11 min, 0% A; 11-14 min, 0%-40% A; 11-15 min, 40%-95% A; 15-18 min, 95% A;

[0076] (2) Design the sample queue: Take an equal volume of sample from each experimental sample and mix it as the QC sample. Set a QC sample every 6 experimental samples in the sample queue to monitor and evaluate the stability of the system and the reliability of the experimental data.

[0077] (3) Sample loading: The processed plasma samples are loaded onto the machine in a random order and analyzed continuously.

[0078] (IV) Sample compound information detected by mass spectrometry:

[0079] (1) Setting mass spectrometry parameters: The electrospray ion source adopts positive and negative ion scanning mode, and the electrospray voltage is set to 2.8 kV; the sheath gas flow rate is 35 arb, the auxiliary gas flow rate is 10 arb, and the capillary temperature is 320 °C;

[0080] (2) Data scanning: first-level full scan, resolution 70000, scanning range 70~1050 (m / z); second-level data-dependent scan, resolution 17500, stepped NCE values ​​20, 40 and 60V;

[0081] (3) The parent and daughter ion fragments are separated by mass through the mass analyzer of the mass spectrometer, and the mass spectrum information of the metabolites is obtained through the detector;

[0082] (4) Export the offline file, perform parameter screening (retention time, mass-to-charge ratio, etc.) for each metabolite, set the retention time deviation to 0.2 min and the mass deviation to 5 ppm, perform peak alignment on different samples, and quantify the peak area;

[0083] (5) Compare the spectra of the molecular ion peaks and fragment ions with those in the database, standardize the original quantitative results, and obtain the identification and relative content quantitative results of the metabolites.

[0084] 5. Predicting preterm birth through detected metabolite information

[0085] In this example, a total of 745 metabolites were identified from 70 plasma samples, and the identified metabolites were annotated using the KEGG database, HMDB database, and LIPIDMaps database. Finally, the metabolite features for predicting preterm birth were screened and determined through differential metabolite functional analysis and random forest screening. According to the identification results of differential metabolites, the present invention divided the training set and the validation set, and used random forest to screen the top 5 metabolites in importance in the two types of spontaneous preterm birth in the training set ( Figure 1-Figure 2 ), based on the screening results, the kit was designed to detect 10 plasma metabolites including 2 types of spontaneous preterm birth, of which ①-⑤ were used to predict pPROM, and ⑥-⑩ were used to predict sPTB. These 10 metabolites and their mass-to-charge ratio characteristics are:

[0086] ①Polo-like kinase (PLK): m / z 357.24963;

[0087] ② All-trans-13,14-dihydroretinol: m / z 289.25259;

[0088] ③1-Hydroxy-2-eicosanoate (LysoPE (0:0 / 20:4 (8Z, 11Z, 14Z, 17Z))): m / z 502.29281;

[0089] ④ Lysophosphatidylethanolamine (18:2) (LPE 18:2): m / z 478.29282;

[0090] ⑤ Coenzyme Q2: m / z 319.19039;

[0091] ⑥N-Desmethyltramadol: m / z 250.18016;

[0092] ⑦Octyl hydrogen phthalate: m / z 279.15909;

[0093] ⑧Phosphocholine: m / z 184.07332;

[0094] ⑨3-Methylglutarylcarnitine: m / z 290.15981;

[0095] ⑩Eucalyptol: m / z 155.14304.

[0096] It should be emphasized that in Figure 1 and Figure 2 In the figure, the horizontal axis "Mean Decrease in Accuracy" indicates the degree to which the accuracy of random forest prediction decreases when the value of a variable is changed to a random number. The larger the value, the more important the variable is.

[0097] The horizontal axis "Mean Decrease Gini" means: calculate the impact of each variable on the heterogeneity of the observed values ​​at each node of the classification tree, so as to compare the importance of the variable. The larger the value, the greater the importance of the variable.

[0098] therefore, Figure 1 and Figure 2 The weight of each metabolite can be clearly expressed.

[0099] 6. Results and analysis

[0100] In order to further demonstrate the differences in metabolites between the two types of spontaneous preterm birth, we compared the differential metabolites of the two types of spontaneous preterm birth with those of the normal pregnancy group (pPROM vs. control group; sPTB vs. control group), and drew box plots to show the abundance characteristics of individual differential metabolites. It can be seen from the figure that the 10 metabolites in the premature birth group have a consistent abundance difference with the normal pregnancy group (such as Figure 3~Figure 4 as shown).

[0101] 7. Model prediction

[0102] In this embodiment, the samples were randomly divided into a training set and a validation set in a ratio of 7:3, without duplication (30 cases in the control group, 20 cases in pPROM, and 20 cases in sPTB in the original data). A random forest was set to generate 500 decision trees, a random seed was set, and metabolites with ROC>0.95 were input for training to obtain the top 5 metabolites (i.e., ①-⑩ above) in terms of importance (MeanDecreaseAccuracy), and the relative content values ​​of these 5 metabolites were re-input into the random forest model for training to obtain the final model.

[0103] Note: Random Forest is an ensemble learning algorithm, which belongs to the bagging type. It combines multiple weak classifiers, and the final result is voted or averaged, so that the overall model has high accuracy and generalization performance. It can achieve good results, mainly due to "random" and "forest", one makes it resistant to overfitting, and the other makes it more accurate.

[0104] 8. Application of prediction models

[0105] The samples to be tested were processed according to the above method, and then the samples were separated by the above chromatography, and the sample compound information was detected by mass spectrometry to obtain the relative content values ​​of the metabolites. The relative content values ​​of the top 5 metabolites related to the above pPROM and sPTB diseases were screened out (pPROM: ①-⑤; sPTB: ⑥-⑩), and they were respectively substituted into the corresponding random forest prediction model to calculate the classification probability value. The default threshold is 0.5, and those greater than 0.5 are considered to be pPROM or sPTB.

[0106] Note: QC samples are mixed by taking equal volumes of samples to be tested. During testing, QC samples should be set at least at the first, last and approximately middle positions of the samples.

[0107] 9. Validation of prediction model

[0108] After the validation set test, the positive predictive value of the pPROM group was 0.857, and the positive predictive value of the sPTB group was 1. As shown in Table 2, the model can predict two types of spontaneous preterm birth. Figure 5 As shown, the data results of the pPROM validation set are as follows Figure 6 As shown; the data results of the sPTB training set are as follows Figure 7 As shown, the data results of the sPTB validation set are as follows Figure 8 shown.

[0109] In this example, 10 plasma differential metabolites were screened using a random forest model, and the classification probability values ​​were calculated based on the random forest model to assess the risk of pPROM or sPTB in pregnant women. Based on the prediction results, more targeted preventive measures can be taken to avoid risks in a timely manner and efficiently ensure the safety of mothers and children.

[0110]

[0111] In summary, the embodiments of the present invention can successfully distinguish two types of premature births and normal pregnancy through the characteristics of plasma metabolites, and then accurately predict the risk of premature birth, and take preventive measures in a timely manner according to the cause and symptoms.

[0112] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A plasma metabolite signature for predicting spontaneous premature birth, wherein spontaneous premature birth is divided into premature birth with intact membranes and premature birth after premature rupture of membranes, characterized in that: The plasma metabolite signature for predicting preterm birth with intact membranes contained the following mass-to-charge ratios: m / z 250.18016, m / z 279.15909, m / z 184.07332, m / z 290.15981, m / z 155.14304; the plasma metabolite signature for predicting preterm birth after premature rupture of membranes contained the following mass-to-charge ratios: m / z 357.24963, m / z289.25259, m / z 502.29281, m / z 478.29282, m / z 319.19039; Preterm birth is premature birth between 14 and 20 weeks of gestation.

2. A kit for predicting spontaneous premature birth, characterized in that: The kit includes a plasma metabolite signature for predicting spontaneous preterm birth, wherein the plasma metabolite signature includes a plasma metabolite signature for predicting preterm birth with intact fetal membranes and a plasma metabolite signature for predicting preterm birth after premature rupture of fetal membranes; wherein: The plasma metabolite signature for predicting preterm birth with intact membranes contained the following mass-to-charge ratios: m / z 250.18016, m / z279.15909, m / z 184.07332, m / z 290.15981, m / z 155.14304; The plasma metabolite signature for predicting preterm birth after premature rupture of membranes contained the following mass-to-charge ratios: m / z 357.24963, m / z289.25259, m / z 502.29281, m / z 478.29282, m / z 319.19039.

3. The kit for predicting spontaneous premature birth according to claim 2, characterized in that: The kit further comprises extracting solution 1, extracting solution 2 and extracting solution 3; wherein: The extract 1 is a methanol-water solution; And / or, the extract 2 is an aqueous formic acid solution and an ammonium formate aqueous solution; And / or, the extract 3 is acetonitrile.

4. The kit for predicting spontaneous premature birth according to claim 3, characterized in that: In the methanol aqueous solution, the volume concentration of methanol is 80%; and / or, in the formic acid aqueous solution, the volume concentration of formic acid is 0.1%; And / or, in the ammonium formate aqueous solution, the concentration of ammonium formate is 2 mmoL / L.

5. The kit for predicting spontaneous premature birth according to claim 2, characterized in that: The test sample of the kit is a blood sample of a pregnant woman.

6. The kit for predicting spontaneous premature birth according to claim 5, characterized in that: The blood samples are peripheral blood samples collected from pregnant women at 14 to 20 weeks of pregnancy.

Images