Clinical Chemistry AACC Online Job Center
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Clinical Chemistry 51: 312-320, 2005. First published December 17, 2004; 10.1373/clinchem.2004.042713
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
clinchem.2004.042713v1
51/2/312    most recent
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (16)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Birch, L.
Right arrow Articles by Keer, J. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Birch, L.
Right arrow Articles by Keer, J. T.
Related Collections
Right arrow Molecular Diagnostics and Genetics
(Clinical Chemistry. 2005;51:312-320.)
© 2005 American Association for Clinical Chemistry, Inc.

Molecular Diagnostics and Genetics

Accurate and Robust Quantification of Circulating Fetal and Total DNA in Maternal Plasma from 5 to 41 Weeks of Gestation

Lyndsey Birch1, Claire A. English1, Keelin O’Donoghue2, Olivia Barigye2, Nicholas M. Fisk2 and Jacquie T. Keer1,a

1 BioAnalytical Innovation Team, LGC Ltd., Teddington, United Kingdom.
2 Experimental Fetal Medicine Group, Institute of Reproductive & Developmental Biology, Imperial College London, and Centre for Fetal Care, Queen Charlotte’s & Chelsea Hospital, Hammersmith Campus, London, United Kingdom.

aAddress correspondence to this author at: BioAnalytical Innovation Team, LGC Ltd., Queens Road, Teddington TW11 0LY, United Kingdom. Fax 44-20-8943-2767; e-mail jacquie.keer{at}lgc.co.uk.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Background: Detection of fetal DNA in maternal plasma is achievable at 5 weeks of gestation, but few large-scale studies have reported circulating fetal and maternal DNA across all trimesters.

Methods: Blood samples were collected from 201 women between 5 and 41 weeks of pregnancy. Quantitative PCR was used to assess total and fetal DNA concentrations, and allelic discrimination analysis was investigated as a route to detecting specifically fetal DNA.

Results: Male fetuses were detectable from 5 weeks amenorrhea with increasing fetal DNA concentrations across gestation. The sensitivity of fetal male gender determination in pregnancies with live birth confirmation was 99%, with 100% specificity. Total DNA concentrations did not correlate with gestational age, but appeared slightly higher in the first and third trimesters than in mid-pregnancy. Analysis of short tandem repeats demonstrated that significant improvements in the detection limit are required for specific detection of fetal DNA.

Conclusions: The high sensitivity of PCR-based detection, together with quantification provided by real-time DNA analysis, has clear potential for clinical application in noninvasive prenatal diagnosis. However, accurate quantification using best-fit data analysis, standardization of methods, and performance control indicators are necessary for robust routine noninvasive diagnostics.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
In the 7 years since the demonstration of cell-free fetal DNA in maternal plasma (1), many studies have been undertaken to establish the sensitivity and repeatability of fetal nucleic acid detection at various gestational ages (2). The ultimate goal in this field is to develop reliable noninvasive tests for clinical prenatal diagnosis. Current applications center on detection of common paternally inherited traits such as sex (3) or Rh status (4), but recent reports have described methods to screen for or diagnose diseases in the fetus (5)(6)(7)(8)(9)(10)(11)(12)(13)(14).

Lo et al. first quantified fetal DNA in women tested during both the first and third trimesters of pregnancy, the earliest detection of fetal DNA being at 11 weeks of gestation. Most subsequent investigations have focused on a particular trimester (15)(16), and more recently, fetal DNA has been detected as early as 5 weeks of gestation (17)(18). However, few groups have monitored the concentrations of both fetal and total DNA in maternal plasma across all three trimesters or included the steps taken to ensure accurate quantification.

The way in which detected DNA is quantified has generally not been detailed in these reports, although there are brief references to methodology (15)(19), typically describing the use of serially diluted human genomic DNA as a calibrant for unknown samples. Some studies have provided Pearson correlation coefficients (R2) for calibration curves (15)(19), but few specified the methods used to prepare calibration curves, the number of replicate points in each assay, or how consistency was monitored between assays. Because quantitative analysis by real-time quantitative PCR (Q-PCR)1 is only as accurate as the calibration, the method used for DNA dilution series preparation is central to assay reliability. This is especially important at low concentrations, such as first trimester analyses, where sampling effects relating to the stochastic distribution of the molecules may noticeably affect the accuracy of quantification (20)(21). Here we address the accuracy of quantification and the variation in DNA concentrations throughout gestation.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
patients
Peripheral blood samples (n = 201) were collected from pregnant women (length of gestation, 5–41 weeks) attending the antenatal clinic, in vitro fertilization clinic, gynecology ward, or day assessment unit at Queen Charlotte’s and Chelsea Hospital. All of the women gave informed consent, as approved by the Hammersmith Hospitals Trust Research Ethics Committee (Reg. No. 2001/6194). Where certain, gestational age was based on menstrual dates and was confirmed by ultrasound before week 20 of gestation; where gestational age was not certain, it was confirmed by ultrasound before 15 weeks of gestation. Eight of the samples were from women known to be carrying twins. Blood was collected into EDTA tubes, and all samples were taken before performance of any invasive procedure.

processing of blood samples
EDTA blood was processed at the Institute for Reproductive and Developmental Biology immediately after collection (with a few exceptions, which were stored overnight at 4 °C). In most cases, the exact volume of blood collected was recorded. Plasma was prepared by centrifugation at 1600g for 10 min (Sorvall RC 5C Pluss). The plasma fraction was carefully transferred to a clean tube (volume again recorded) and frozen at –80 °C; it was then sent frozen by courier to LGC, where it was stored at –80 °C until further processing. On the day of analysis, the plasma was thawed and subjected to further centrifugation at 16 000g for 10 min (Jouan MR22i) to pellet any remaining cellular debris. The plasma was then carefully transferred to a clean polypropylene tube. All blood processing was performed in a class II safety cabinet (MDH).

dna extraction from maternal plasma
DNA was prepared from 1.5 mL of plasma by use of a QIAamp Blood Mini Kit (Qiagen) according to the "blood and body fluid protocol". The extraction method was performed as recommended by the manufacturer, except that all reagent volumes were scaled up to accommodate the increased volume of plasma used in each extraction. The sample was loaded on the spin-column by repeated centrifugation, and a vacuum line was used to aspirate the filtrate, minimizing contamination. A final elution volume of 50 µL was used. The extraction efficiency was determined by adding known quantities of male human genomic DNA to nonpregnant female plasma samples and calculating the recovery by Q-PCR. The efficiency ranged from 45% to 75%, with some losses possibly caused by reduced elution volume, leading to diminished yields as stated in the manufacturer’s guidelines.

generation of calibration curves
A human male control DNA (Promega) was used to prepare calibration curves for the chosen amplification targets. A PicoGreen® (Molecular Probes) quantification analysis was performed with a purified DNA calibrator (Cambio), prepared gravimetrically, to determine the concentration of the stock DNA. This value was converted to genome-equivalents/mL by use of the conversion factor of one diploid genome being equivalent to 6.6 pg of DNA. High-concentration (5000 genome-equivalents/µL) DNA stocks were prepared after the initial PicoGreen analysis, aliquoted into small volumes, and stored at 4 °C. For each run, one aliquot of stock DNA was serially diluted with molecular biology grade water (Eppendorf) to construct a nine-point calibration curve in the range 0.5–2500 genome-equivalents/µL.

real-time q-pcr
Real-time Q-PCR was performed with the 5'-nuclease assay. Primers and fluorogenic 5' exonuclease (TaqMan) probes were designed according to the guidelines recommended by Applied Biosystems. A coding region of the male-specific SRY gene was chosen to monitor the presence of fetal DNA, and a coding region of the housekeeping gene GAPDH was used to quantify total DNA. Both targets were detected independently in singleplex reactions to prevent outcompetition of the SRY amplification (data not shown). Reaction conditions for both targets were as follows: 450 nM forward primer, 450 nM reverse primer, 225 nM TaqMan probe, 1x PCR mixture (BioGene), and 5 µL of genomic DNA extract in a final reaction volume of 25 µL. PCR was performed for 55 cycles (uracil N-glycosylase activation at 50 °C for 2 min and denaturation at 95 °C for 10 min, followed by 55 cycles of 95 °C for 5 s and 60 °C for 1 min). The gene-specific primer and probe sequences were as follows:

SRY: Forward primer, 5'-CGATCAGAGGCGCAAGATG-3'; reverse primer, 5'-TGGTATCCCAGCTGCTTGCT-3'; probe, 5'-VIC-TCTAGAGAATCCCAGAATGCGAAACTCAGAGA-TAMRA-3' (where TAMRA is 6-carboxytetramethylrhodamine)

GAPDH: Forward primer, 5'-AGGTTTACATGTTCCAATATGATTCCA-3'; reverse primer, 5'-ATGGGATTTCCATTGATGACAAG-3'; probe, 5'-FAM-CCGTTCTCAGCCTTGACGGTGC-TAMRA-3' (where FAM is 6-carboxyfluorescein)

All probes and primers were obtained from Sigma-Genosys with the exception of the SRY VIC-labeled probe, which was supplied by Applied Biosystems.

Amplification data were collected and analyzed with an ABI Prism 7700 Sequence Detector (Applied Biosystems. The cycle threshold value (Ct) was the measurand in all cases. Each sample was analyzed in triplicate, and multiple negative reaction blanks were included in every analysis for both sample extraction and amplification stages. A calibration curve (triplicate samples) was analyzed on the same reaction plate for each run. Fetal gender of each sample was confirmed phenotypically at birth or by fluorescence in situ hybridization analysis of fetal material. The study was performed blind at LGC; none of the confirmed fetal genders were disclosed by the Institute for Reproductive and Developmental Biology until after the analysis was completed and the results were recorded.

quality control
All reagents for DNA extraction and Q-PCR were prealiquoted into one-shot volumes. Filter tips were used throughout, and in addition 1-mL Reach filter tips (Molecular BioProducts) were used when pipetting from 15-mL tubes. All equipment and work areas were swabbed with fresh 10 mL/L hypochlorite, and DNA extraction was performed in a separated area in a class II hood. A vacuum line was used during the extraction process to minimize generation of aerosols. Ensuring that all lids were tightly closed during centrifugation significantly reduced contamination. During Q-PCR, carryover contamination was prevented by the use of uracil N-glycosylase. Reaction mixtures were prepared in a separated area in a hood (CleneCab; Herolab), and DNA was added to the reaction in a dedicated template addition area.

Both extraction and amplification negative controls were checked for positive signals, which if present would indicate contamination occurring during the process. In the event of contamination, either the extraction and/or the amplification was repeated for the potentially affected batch of samples. The reliability of quantification was assessed by monitoring of the calibration curve for each individual amplification run.

analysis of short tandem repeats
We analyzed five pairs of age-matched samples (gestational ages, 29 weeks/6 days to 41 weeks) from confirmed male and female gender pregnancies as well as artificial mixtures of male and female DNA in ratios of 1:2 to 1:660, respectively, using an ampFLSTR® SGM PlusTM Kit (Applied Biosystems). With the ampFLSTR SGM Plus, we could amplify 11 different loci, but only results for the Amelogenin amplicons are presented here. Amplification was performed according to the manufacturer’s protocol with the following modifications: 5 µL of DNA extract plus 15 µL of 1x Tris-EDTA buffer was used in each reaction to make the volume up to 20 µL. Positive and negative controls were included with every run. Results were analyzed with GeneScan® 3.1.2 and Genotyper® 2.5 software (Applied Biosystems).

statistical analysis
We used Microsoft Excel to perform a factorial ANOVA to test for significant interrun differences in the calibration curve data from all experiments. We used an empirical method to analyze the data, using quasi-Newton exponential weighted regression because this was determined to be the most appropriate model describing the relationship between the target concentration and Ct value for both GAPDH and SRY calibration curves (English et al., Improving the accuracy and reliability of low level DNA analysis using real-time PCR, submitted for publication). We used this model to quantify the amount of each target in the unknown samples, performing quasi-Newton weighted analysis with the software package STATISTICA 6 (StatSoft Ltd). Because the volume of blood taken and plasma prepared was recorded for almost every sample, we were able accurately to calculate the amount of each target gene present in the samples in terms of genome-equivalents/mL of blood, using the following equation:

Where C is the genome-equivalents/mL of blood; Q is the target quantity (genome-equivalents) determined by PCR from the calibration curve; VDNA is the total volume of DNA obtained after extraction (50 µL); VPCR is the volume of DNA used in PCR (5 µL); VEXT is the volume of plasma extracted; Vplasma is the volume of plasma prepared; and Vblood is the volume of blood taken. The mean Ct of triplicate reactions for each sample was calculated and used for quantification.

We analyzed the relationships of SRY and GAPDH concentrations with gestational age and GAPDH concentration with maternal age by linear regression using the least-squares method; we also used a Pearson test to establish any correlation (Microsoft Excel). We used STATISTICA 6 to calculate the 95% prediction intervals for SRY concentrations against gestational age from the linear regression equation. Reference intervals were derived for both circulating fetal and total DNA concentrations in each trimester based on 95% prediction intervals obtained by use of the residual standard deviation from the regression analysis (22)(23).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
anova
The ANOVA results from the calibration curve data demonstrated that for SRY, there was no significant variation between runs (P = 0.46) and the interaction term (concentration with run) was not significant (P = 0.71), which allowed all 12 runs to be pooled. For GAPDH, interrun variability was observed (P <0.01), although overall the mean Ct values between all 12 runs differed by only 1.1 Ct and the interaction term was not significant at the 1% level. We observed high consistency both within and between runs, with error mean square values of 0.25 and 0.35, respectively, and the inherent variability between runs was statistically significant as a result of the high degree of repeatability within experiments. However, the large size of the data set and the observed consistency between runs indicated that the data from all 12 experiments could be pooled to produce a single "composite" calibration curve for each target (Fig. 1 ), thus giving a more accurate estimate of the population mean and variance for each of the calibrators.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 1. Composite calibration curves for SRY (A) and GAPDH (B) constructed from 12 individual curves.

The graph displays the log concentration of target in genome-equivalents against the mean Ct value for each concentration from the 12 individual runs (3 replicates per run). A weighted quasi-Newton exponential regression line has been plotted to model the results.

correlation of fetal dna (sry) copy numbers in maternal plasma with gestational age
The analysis gave a range of quantifications for SRY from 0 to 507.5 genome-equivalents/mL of blood, with a clear difference between positive signals for SRY, which indicated the presence of a male fetus, and the absence of amplifiable SRY, which signified a female fetus. No contamination events were detected, and the cutoff between the two populations was 1 genome-equivalent/mL. Either live birth examination or fluorescence in situ hybridization analysis on invasively sampled fetal material was used to confirm fetal gender. The sensitivity of the SRY assay was 99% in the 183 samples for which confirmation of fetal sex could be obtained. A single SRY false negative, at 7 weeks of gestation, was the male of a dizygotic twin pregnancy. The range of SRY concentrations in each trimester is represented as a box and whisker plot in Fig. 2A , and the log mean value of each set of triplicate results per sample is plotted against gestational age in Fig. 3A . In the male samples, there was a strong positive correlation (R2 = 0.52; y = 0.0051x + 0.6009) between log SRY concentration and gestational age, indicating that >50% of the variation was attributable to this factor.



View larger version (14K):
[in this window]
[in a new window]
  		Figure 2. Box and whisker plots showing the concentrations of SRY (A) and GAPDH (B), in genome-equivalents/mL of blood, by trimester.

The median and 25th and 75th percentiles are represented by the midline and lower and upper limits of the box, respectively, and the 10th and 90th percentiles are denoted by the lower and upper whiskers, respectively. {circ}, outliers; *, extreme outliers.



View larger version (33K):
[in this window]
[in a new window]
  		Figure 3. Concentrations of SRY (A) and GAPDH (B) plotted vs gestational age.

(A), the data points indicate log concentrations in genome-equivalents/mL of blood. •, males; {circ}, females. The linear regression (solid line) is plotted together with the 95% prediction intervals (dotted lines). (B), values are the means of three replicates per sample.

We calculated the 95% prediction intervals for SRY (Fig. 3AUp ); these intervals predict with 95% probability the upper and lower values for an individual sample at a given time-point. We also calculated indicative time-specific reference intervals for both SRY and GAPDH for each trimester, using the 95% prediction intervals derived from the residual standard deviation in the regression analysis; these reference intervals are shown in Table 1 together with the mean and median values for each trimester.


View this table:
[in this window]
[in a new window]
  		Table 1. Time-specific reference intervals for fetal and total DNA concentrations based on SRY and GAPDH detection constructed with 95% prediction intervals around the linear regression line.

effect of gestational age and maternal age on total dna in maternal plasma
A composite calibration curve, including the calibrators from each GAPDH assay, was used to calculate the amount of DNA present in each sample. During this study, one batch of nine samples was reextracted because the extraction negative controls showed low positive signals for GAPDH. In addition, one PCR run of 18 samples was repeated because positive GAPDH signals in the PCR negative controls indicated that contamination had occurred. The repeat results were comparable to the initial data in both cases. Box and whisker plots of the data, shown in Fig. 2BUp , illustrated a noticeably increased proportion of samples with increased GAPDH concentrations in the first and third trimesters compared with mid-pregnancy.

A plot of the log GAPDH concentrations against gestational age (Fig. 3BUp ) showed that there was no significant relationship (R2 = 0.06). Similarly, there was no significant correlation between maternal age and log GAPDH concentrations (R2 = 0.02). The mean and median trimester values are displayed in Table 1Up together with the indicative reference intervals for total DNA concentrations, which were calculated using 95% prediction intervals based on the residual standard deviation from the regression analysis.

positive control for fetal dna
We investigated the potential of using short tandem repeat (STR) analysis to provide positive confirmation of sufficient extracted fetal nucleic acid to be detectable by fluorogenic PCR analysis. STR profiling of plasma samples from known male and female pregnancies, matched by gestational age, and of artificial male/female DNA mixtures was performed as described. Electropherograms of the products of amelogenin-specific PCR of the STR multiplex revealed peaks for the 106-bp amplicon from the X chromosome and a 112-bp amplicon from the Y chromosome, if present (Fig. 4 ). The heights of the fluorescent peaks indicated the amount of PCR product produced and thus of the amount of target material present in the sample. The in-house mixtures showed a detectable male signal only when the male DNA was present in a 1:2 ratio with female DNA. In mixtures containing lower proportions of male target, the 112-bp male amplicon was not detectable. In the profiles from very late pregnancy samples (≥38 weeks), a small minority peak detectable in the male samples at 112 bp was produced from the paternally inherited Y-chromosome amelogenin locus (indicated by arrows in Fig. 4 ). In the later pregnancy samples, both the total amount of SRY present and the ratio of fetal to total DNA in the sample, as assessed by Q-PCR, were increased considerably.



View larger version (18K):
[in this window]
[in a new window]
  		Figure 4. Preliminary STR analysis of artificial male/female DNA mixtures and fetal samples for which the gender was known.

The electropherogram of the gel region containing the X- and Y-chromosome amelogenin PCR products (at 106 and 112 bp, respectively) is shown. The minority fetally derived Y-chromosome peaks at 112 bp are indicated by an arrow. Trace a, 2:1 female/male genomic DNA mixture (5 X:1 Y); trace b, 20:1 female/male genomic DNA mixture (41 X:1 Y); trace c, 200:1 female/male genomic DNA mixture (401 X:1 Y); trace d, 41-week male gender sample (12.6 X:1 Y); trace e, 41-week female gender sample; trace f, allelic ladder, showing even signals from equimolar X and Y targets. RFU, relative fluorescence units.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Our results provide an extensive data set establishing reference values for both fetal and total DNA and variations in DNA concentrations across all three trimesters of pregnancy. Using weighted quasi-Newton exponential analysis of the pooled calibration curves, we calculated the SRY concentration for each unknown sample. The concentration ranges for SRY in each trimester (Fig. 2AUp ) displayed both heteroscedasticity (changing variance of SRY concentration with trimester) and a positive skew in the late third trimester. The midline of each box-plot represents the median of that trimester (values are shown in Table 1Up ). The change in fetal DNA concentrations between the first two trimesters was small, whereas fetal DNA concentrations increased markedly in the third trimester, and the spread of results was much wider as evidenced by the lower median value compared with the mean. We derived indicative SRY reference intervals for each trimester (Table 1Up ); the lower limits for all three trimesters were similar, but there were considerable increases in the upper limits of SRY concentrations in the second and third trimesters.

Performing a log transformation of the data (Fig. 3AUp ) allows the application of normal statistics to heteroscedastic data sets. The increase in SRY concentration with gestational age up to 41 weeks (R2 = 0.52 by linear regression analysis) is clearly demonstrated through the complete coverage across the entirety of gestation presented here, in contrast to more limited studies (15)(16). The broad linear relationship between SRY concentrations and gestational age is not as strong in the very early and late stages of pregnancy. Biological variation in individuals relating to placental size, amounts of DNA shed, fetal cell trafficking, and apoptosis may all contribute to the variation observed (24)(25). The lowest concentration detected was 1.2 genome-equivalents/mL of blood at 37 days and increased to 507.5 genome-equivalents/mL of blood at 291 days, which are within the calculated reference intervals shown in Table 1Up . Toward the very late stages of pregnancy, sharp increases in circulating fetal DNA concentrations can occur as a result of placental breakdown (26), which may explain the very high SRY concentration observed in the single sample taken at 41 weeks of gestation. The earliest detection was at 37 days, which demonstrates a sensitivity equivalent to that described previously (17)(18). Overall the fetal DNA concentrations reported here are in broad agreement with previous studies for which the trimester data can be directly compared (2)(17)(27). Maternal age did not correlate significantly with SRY concentrations, concurring with previous findings (28). We also did not observe a significant relationship between total and fetal DNA concentrations (R2 = 0.16), in contrast to an earlier report (16).

Recent concerns that persistence of fetal cells from previous pregnancies may lead to false positives (29) are not supported by our results. Twenty-seven of the samples in this study group (13.4%) were female fetuses from mothers who had previously carried at least one male fetus. No SRY signal was detected in any of these samples, indicating that no detectable male DNA persisted from the previous pregnancies. Subjecting samples to centrifugation at 16 000g before analysis, as performed in the present study, has been reported to remove any potential cellular material that could contribute to the free-floating DNA signal after extraction (30)(31), explaining the lack of persistence observed.

We used weighted quasi-Newton exponential fit to the pooled calibration curve data to quantify GAPDH in the unknowns. We observed no strong correlation between variations in the concentration of total DNA and gestational or maternal age, or fetal DNA concentrations. However, the highest concentrations appear to occur at the beginning and toward the later stages of pregnancy (Fig. 2BUp and Table 1Up ), although both the highest and lowest GAPDH values were observed in samples taken in the third trimester. Mean and median values for the second trimester showed a decrease in total concentrations and a tighter distribution of results compared with both the first and third trimesters. Differences in total DNA concentrations are primarily determined by the concentration of maternal DNA and may be attributable to inherent variation between individuals, reflecting lower cell shedding or apoptosis in mid-pregnancy (32). This effect has not been reported previously, but earlier studies either have focused on a single trimester or have not been sufficiently large to observe such trends. Technical causes for this observation are unlikely because the GAPDH quantification performed consistently and samples were analyzed in batches of mixed gestational ages, ensuring that the performance of any individual run would not bias the results of one age group or trimester specifically. Other recent studies are in agreement with the present work, including that of Chan et al. (16), who concluded that ß-globin concentrations (measuring total DNA) and gestational age were not significantly correlated.

A potential route to improving the reliability of noninvasive testing is the use of robust and traceable quantitative methods. Ultimately, standardization of techniques will improve comparability among laboratories and increase confidence in analytical results, and other investigators have addressed the importance of standardization in noninvasive diagnosis. A recent interlaboratory comparison to assess noninvasive male fetal detection (33) highlighted the need for optimized protocols, particularly in the extraction process. The importance of sample preparation, including methods for blood processing, has been the subject of several studies (30)(31). Recently, formaldehyde treatment of blood samples at collection has been advocated to reduce the concentrations of circulating maternally derived nucleic acids by reducing cell lysis (34). Extending clear methodologic guidelines to encompass the entire analytical procedure will improve the consistency and reliability of routine diagnostic noninvasive methods.

Development of a universally applicable positive control for fetally derived DNA would allow better use to be made of negative noninvasive diagnostic results, but in the absence of a positive control, such data must be interpreted cautiously (35). Approaches to providing a positive control include use of distinguishable tracer DNA (18) or detection of fetally expressed genes to confirm the presence of fetally derived nucleic acid in plasma extracts (36). Placenta-specific mRNA expression profiles have recently been reported (37)(38); alternatively, a combination of methylation patterns from imprinting and known inherited polymorphisms can be exploited to identify fetal material specifically (39).

To investigate another potential route, we carried out STR analysis on extracted plasma to detect fetal DNA in the maternal samples by identification of paternally inherited alleles. For most loci, either one or two majority peaks should be seen, corresponding to homozygous or heterozygous maternal targets at each locus. When fetal DNA is present in detectable quantities, an additional minority peak corresponding to the paternally inherited fetal allele may also be seen, if distinguishable by size from the two maternally derived targets. This approach could also be used in detection of expanded repeat mutations. In practice, however, we found that the minority fetal DNA was detectable only when present at high concentrations, as determined by use of artificial mixtures of known amounts of female and male genomic DNA (Fig. 4Up ). The 112-bp amelogenin male-specific amplicon was detectable only in the 2:1 mixture of female to male DNA, whereas the female 106-bp amplicon was clearly seen in all samples. Informed by the results from the artificial mixtures, we tested third-trimester plasma samples because earlier samples did not have sufficient fetal DNA to allow detection by STR analysis. The samples analyzed ranged from 29 to 41 weeks of gestation, but only a single weak 112-bp amplicon was detected, in the male 41-week sample shown (Fig. 4Up ).

To exploit the STR approach for noninvasive prenatal analysis, an increase in the sensitivity of the method is required. Potentially this could be achieved either by preferential amplification of fetal DNA or through enrichment strategies during the DNA extraction process (34)(40).

Accurate measurement is underpinned by effective methods for target quantification. There are numerous methods for quantifying DNA, many of which are imprecise. The performance of quantification approaches is ill-defined because of the lack of certified DNA reference materials.

In an attempt to overcome these problems, we used a calibration process traceable to a commercially available quantified human DNA calibration solution (Cambio). This approach has also been reported recently in the quantification of cDNA targets by real-time PCR (41). To minimize interrun variation, we prepared one stock of the human DNA and divided it into aliquots of a high-concentration DNA solution (5000 genome-equivalents/µL), which were stored at 4 °C. All subsequent calibration curves came from a single quantification, and the same analyst was used throughout to reduce variation. The volume used in preparing the dilution series affected the results of the analysis; smaller volumes (20–50 µL) led to greater variability in the Ct values than when larger volumes were used (300 µL to 1 mL), with the greatest effect observed at lower target concentrations. In addition, we observed during initial method development that low-concentration dilutions gave a gradually decreasing signal with length of storage (days) at 4 °C, in agreement with other recent findings (42). To minimize variation through this effect, we made fresh dilution series on the day of each run. The calibration curve for each run was used as an effective performance control indicator, providing an effective means for assessing interrun variation and identifying problems with the assay.

A previous study (43) demonstrated that increasing the number of sample replicates from three to six in a 50-cycle real-time PCR improved the rate of detection of male DNA in maternal plasma from 92.6% to 100%. In the present study, the PCR was run for 55 cycles to increase sensitivity sufficiently to detect fetal material in early first-trimester samples. Because of limitations in the volume of DNA extracted, samples were only run in triplicate for both the GAPDH and SRY assays, but this approach was sufficient to achieve 99% sensitivity of fetal detection. Increasing the cycle number from the 38–40 cycles recommended by the instrument manufacturers improved analytical sensitivity both by detecting positive signals that were measurable only after >40 cycles of amplification and by increasing our confidence in the distinction between low-concentration positive and true negative results.

In conclusion, these data represent the most complete analysis to date of the concentrations of fetal and total DNA across pregnancy, with coverage from 5 to 41 weeks of gestation. The large sample size permitted clear demonstration of the positive correlation between fetal, but not total, DNA concentrations and gestational age. Although STR analysis has potential as a positive control for fetal DNA, as shown in later pregnancy, the current approach is not sensitive enough for such analyses. An empirical approach using weighted quasi-Newton exponential analysis to increase the accuracy of low-level quantification combined with modification of the Q-PCR method to maximize analytical sensitivity increases confidence in the reliability of quantitative trace molecular measurements. Incorporating modifications such as increasing the number of replicate analyses and amplification cycles as well as anchoring quantitative analysis by use of traceable calibrators, where available, has the potential to dramatically increase confidence in measurements of low-concentration nucleic acids and thus remove barriers to routine clinical application.


   Acknowledgments
 
We would like to thank Mahesh Choolani for initiating the collaboration, Malcolm Burns and Steve Ellison for statistical advice, and William Dennes for live birth confirmation data. This work was funded by the National Measurement System Directorate, Department of Trade and Industry, under the Valid Analytical Measurement (VAM) program 2000–2003. Olivia Barigye and Keelin O’Donoghue were supported by grants from the Institute of Obstetrics & Gynaecology Trust.


   Footnotes
 
1 Nonstandard abbreviations: Q-PCR, quantitative PCR; Ct, cycle threshold; and STR, short tandem repeat.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Lo YMD, Corbetta N, Chamberlain PF, Rai V, Sargent IL, Redman CW, et al. Presence of fetal DNA in maternal plasma and serum. Lancet 1997;1997;350:485-487.
  2. Lo YMD, Tein MSC, Lau TK, Haines CJ, Leung TN, Poon PMK, et al. Quantitative analysis of fetal DNA in maternal plasma and serum: implications for non-invasive prenatal diagnosis. Am J Hum Genet 1998;1998;62:768-775.
  3. Costa JM, Benachi A, Gautier E, Jouannic JM, Ernault P, Dumez Y. First-trimester fetal sex determination in maternal serum using real-time PCR. Prenat Diagn 2001;2001;21:1070-1074.
  4. Lo YMD, Hjelm NM, Fidler C, Sargent IL, Murphy MF, Chamberlain PF, et al. Prenatal diagnosis of fetal RHD status by molecular analysis of maternal plasma. N Engl J Med 339;1998;339:1734-1738.
  5. Wald NJ, Morris JK. A new approach to antenatal screening for fragile X syndrome. Prenat Diagn 2003;2003;23:345-351.[CrossRef]
  6. González-González MC, Trujillo MJ, Rodríguez de Alba M, García-Hoyos M, Lorda-Sánchez I, Díaz-Recasens J, et al. Huntington disease—unaffected fetus diagnosed from maternal plasma using QF-PCR. Prenat Diagn 2003;2003;23:232-234.
  7. Vrettou C, Traeger-Synodinos J, Tzetis M, Malamis G, Kanavakis E. Rapid screening of multiple ß-globin gene mutations by real-time PCR on the LightCycler: applications to carrier screening and prenatal diagnosis of thalassemia syndromes. Clin Chem 2003;2003;49:769-776.
  8. Chiu RW, Lau TK, Leung TN, Chui DH, Lo YM. Prenatal exclusion of ß thalassaemia major by examination of maternal plasma. Lancet 2002;2002;28;360:998-1000.
  9. Amicucci P, Gennarelli M, Novelli G, Dallapiccola B. Prenatal diagnosis of myotonic dystrophy using fetal DNA obtained from maternal plasma. Clin Chem 2000;2000;46:301-302.
  10. Gonzalez-Gonzalez MC, Garcia-Hoyos M, Trujillo MJ, Rodriguez de Alda M, Lorda-Sanchez I, Diaz-Recasens J, et al. Prenatal detection of a cystic fibrosis mutation in fetal DNA from maternal plasma. Prenat Diagn 2002;2002;22:946-948.
  11. Nasis O, Thompson S, Hong T, Sherwood M, Radcliffe S, Jackson L, et al. Improvement in sensitivity of allele-specific PCR facilitates reliable non-invasive prenatal detection of cystic fibrosis. Clin Chem 2004;2004;50:694-701.
  12. Farina A, LeShane ES, Lambert-Messerlian GM, Canick JA, Lee T, Neveux LM, et al. Evaluation of cell-free fetal DNA as a second-trimester serum marker of Down syndrome pregnancy. Clin Chem 2003;2003;49:239-242.
  13. Lo YMD, Lau TK, Zhang J, Leung TN, Chang AMZ, Hjelm NM, et al. Increased fetal DNA concentrations in the plasma of pregnant women carrying fetuses with trisomy 21. Clin Chem 1999;1999;45:1747-1751.
  14. Wataganara T, LeShane ES, Farina A, Messerlian GM, Lee T, Canick JA, et al. Maternal serum cell-free fetal DNA levels are increased in cases of trisomy 13 but not trisomy 18. Hum Genet 2003;2003;112:204-208.
  15. Sekizawa A, Kondo T, Iwasaki M, Watanabe A, Jimbo M, Saito H, et al. Accuracy of fetal gender determination by analysis of DNA in maternal plasma. Clin Chem 2001;2001;47:1856-1858.
  16. Chan LYS, Leung TN, Allen Chan KC, Tai HL, Lau TK, Wong EMC, et al. Serial analysis of fetal DNA concentrations in maternal plasma in late pregnancy. Clin Chem 2003;2003;49:678-680.
  17. Honda H, Miharu N, Ohashi Y, Samura O, Kinutani M, Hara T, et al. Fetal determination in early pregnancy through qualitative and quantitative analysis of fetal DNA in maternal serum. Hum Genet 2002;2002;110:75-79.
  18. Guibert J, Benachi A, Grebille AG, Ernault P, Zorn JR, Costa JM. Kinetics of SRY gene appearance in maternal serum: detection by real time PCR in early pregnancy after assisted reproductive technique. Hum Reprod 2003;2003;18:1733-1736.
  19. Wei C, Saller DN, Sutherland JWH. Detection and quantification by homogeneous PCR of cell-free fetal DNA in maternal plasma. Clin Chem 2001;2001;47:336-338.
  20. Kay S, Van den Eede G. The limits of GMO detection. Nat Biotechnol 2001;2001;19:405.
  21. Stenman J, Orpana A. Accuracy in amplification. Nat Biotechnol 2001;2001;19:1011-1012.
  22. Royston P. Constructing time-specific reference ranges. Stat Med 1991;1991;10:675-690.
  23. Miller JN, Miller JC. Statistics and chemometrics for analytical chemistry, 4th ed 2000:288 Prentice Hall . .
  24. Lo YMD, Lau TK, Chan LYS, Leung TN, Chang AMZ. Quantitative analysis of the bi-directional fetomaternal transfer of nucleated cells and plasma DNA. Clin Chem 2000;2000;46:1301-1309.
  25. Kondo T, Sekizawa A, Saito H, Jimbo M, Sugito Y, Okai T. Fate of fetal nucleated erythrocytes circulating in maternal blood: apoptosis is induced by maternal oxygen concentration. Clin Chem 2002;2002;48:1618-1620.
  26. Redman CW, Sargeant IL. Pre-eclampsia, the placenta and the maternal systemic inflammatory response—a review. Placenta 2003;2003;24(Suppl A):S21-S27.[CrossRef]
  27. Cremonesi L, Galbiati S, Foglieni B, Smid M, Gambini D, Ferrari A, et al. Feasibility study for a microchip-based approach for non-invasive prenatal diagnosis of genetic diseases. Ann N Y Acad Sci 2004;2004;1022:105-112.
  28. Farina A, Caramelli E, Concu M, Sekizawa A, Ruggeri R, Bovicelli L, et al. Testing normality of fetal DNA concentration in maternal plasma at 10–12 completed weeks’ gestation: a preliminary approach to a marker for genetic screening. Prenat Diagn 2002;2002;22:148-152.
  29. Invernizzi P, Biondi ML, Battezzati PM, Perego F, Selmi C, Cecchini F, et al. Presence of fetal DNA in maternal plasma decades after pregnancy. Hum Genet 2002;2002;110:587-591.
  30. Chiu RWK, Poon LLM, Lau TK, Leung TN, Wong EMC, Lo YMD. Effects of blood-processing protocols on fetal and total DNA quantification in maternal plasma. Clin Chem 2001;2001;47:1607-1613.
  31. Swinkels DW, Wiegerinck E, Steegers EAP, De Kok JB. Effects of blood-processing protocols on cell-free DNA quantitation in plasma. Clin Chem 2003;2003;49:525-526.
  32. Darmochwal-Kolarz D, Rolinski J, Tabarkiewicz J, Leszczynska-Gorzelak B, Buczkowski J, Wojas K, et al. Myeloid and lymphoid dendritic cells in normal pregnancy and pre-eclampsia. Clin Exp Immunol 2003;2003;132:339-344.
  33. Johnson KL, Dukes KA, Vidaver J, LeShane ES, Ramirez I, Weber WD, et al. Interlaboratory comparison of fetal male DNA detection from common maternal plasma samples by real-time PCR. Clin Chem 2004;2004;50:516-521.
  34. Dhallan R, Au WC, Mattagajasingh S, Emche S, Bayliss P, Damewood M, et al. Methods to increase the percentage of free fetal DNA recovered from the maternal circulation. JAMA 2004;2004;291:1114-1119.
  35. International Blood Group Reference Laboratory. User guide: fetal genotyping. http://www.blooddonor.org.uk/ibgrl/Reference%20Services/RefSer_genotyping.htm..
  36. Ng EKO, Tsui NBY, Lau TK, Leung TN, Chiu RWK, Panesar NS, et al. mRNA of placental origin is readily detectable in maternal plasma. Proc Natl Acad Sci U S A 2003;2003;100:4748-4753.
  37. Page N, Butlin D, Manyonda I, Lowry P. The development of a genetic profile of placental gene expression during the first trimester of pregnancy: a potential tool for identifying novel secreted markers. Fetal Diagn Ther 2000;2000;15:237-245.
  38. Tsui NBY, Chim SSC, Chiu RWK, Lau TK, Ng EKO, Leung TN, et al. Systematic micro-array based identification of placental mRNA in maternal plasma: towards non-invasive prenatal gene expression profiling. J Med Genet 2004;2004;41:461-467.
  39. Poon LLM, Leung TN, Lau TK, Chow KCK, Lo YMD. Differential DNA methylation between fetus and mother as a strategy for detecting fetal DNA in maternal plasma. Clin Chem 2002;2002;48:35-41.
  40. Li Y, Zimmermann B, Rusterholz C, Kang A, Holzgreve W, Hahn S. Size separation of circulatory DNA in maternal plasma permits ready detection of fetal DNA polymorphisms. Clin Chem 2004;2004;50:1002-1011.
  41. Whelan JA, Russell NB, Whelan MA. A method for the absolute quantification of cDNA using real-time PCR. J Immunol Methods 2003;2003;278:261-269.
  42. Teo IA, Choi JW, Morlese J, Taylor G, Shaunak S. LightCycler qPCR optimisation for low copy number target DNA. J Immunol Methods 2002;2002;270:119-133.
  43. Hromadnikova I, Houbova B, Hridelova D, Voslarova S, Kofer J, Komrska V, et al. Replicate real-time PCR testing of DNA in maternal plasma increases the sensitivity of non-invasive fetal sex determination. Prenat Diagn 2003;2003;23:235-238.



The following articles in journals at HighWire Press have cited this article:


Home page
 Hum ReprodHome page
A. Yin, E.H.Y. Ng, X. Zhang, Y. He, J. Wu, and K.Y. Leung
Correlation of maternal plasma total cell-free DNA and fetal DNA levels with short term outcome of first-trimester vaginal bleeding
Hum. Reprod., June 1, 2007; 22(6): 1736 - 1743.
[Abstract] [Full Text] [PDF]

Home page
 Mol Hum ReprodHome page
I. Stanghellini, R. Bertorelli, L. Capone, V. Mazza, C. Neri, A. Percesepe, and A. Forabosco
Quantitation of fetal DNA in maternal serum during the first trimester of pregnancy by the use of a DAZ repetitive probe
Mol. Hum. Reprod., September 1, 2006; 12(9): 587 - 591.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
clinchem.2004.042713v1
51/2/312    most recent
Right arrow Submit an electronic Letter to
the Editor about this paper
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (16)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Birch, L.
Right arrow Articles by Keer, J. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Birch, L.
Right arrow Articles by Keer, J. T.
Related Collections
Right arrow Molecular Diagnostics and Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS