|
|
||||||||
General Clinical Chemistry |
1 Foundation for Blood Research, Scarborough, ME.
2 Women and Infants Hospital, Providence, RI.
3 Quest Diagnostics Nichols Institute, San Juan Capistrano, CA.
aAddress correspondence to this author at: Division of Medical Screening, Department of Pathology, Women and Infants Hospital, 101 Dudley St., Providence, RI 02905. Fax 207-657-7887; e-mail gpalomaki{at}wihri.org.
| Abstract |
|---|
|
|
|---|
Methods: As part of an earlier observational study, serum samples from 54 Down syndrome and 276 matched unaffected pregnancies were collected between 9 and 15 weeks of gestation. Samples had been aliquoted and stored at 20 °C for 8 years. ITA was measured and converted to weight-adjusted multiples of the median (MoM). The distributions of other first-trimester markers are from a single published study.
Results: Median ITA MoM in Down syndrome pregnancies increase as gestational age increases (2.02 MoM at 11 and 2.44 MoM at 13 completed weeks). At 75% detection, maternal age in combination with ITA and PAPP-A measurements have an 8.0% false-positive rate, slightly lower than the 8.8% found for the free ß and PAPP-A combination; adding NT measurements reduces false positives for the 2 combinations to 2.0% and 1.8%, respectively.
Conclusion: Serum ITA appears to be a useful first-trimester Down syndrome marker that could replace free ß measurements while maintaining performance.
| Introduction |
|---|
|
|
|---|
-fetoprotein, unconjugated estriol, and human chorionic gonadotropin (hCG)
1
measurements. When the pregnancy is dated by ultrasound, triple testing can detect up to 71% of Down syndrome pregnancies at a 5% false-positive rate (1). Some programs have added measurements of dimeric inhibin-A (DIA) to create a quadruple test and have increased detection to as high as 82% at the same false-positive rate (2). A new standard of care may be emerging because Down syndrome screening is possible in the first trimester of pregnancy by use of maternal age in combination with maternal serum pregnancy-associated plasma protein-A (PAPP-A), the free ß subunit of human chorionic gonadotropin (free ß), and ultrasound measurement of nuchal translucency thickness (NT) (3). At a 5% false-positive rate, this combination can detect up to 84% of Down syndrome pregnancies (4), but the actual impact on birth prevalence might be lower because increased NT measurements are known to be associated with an increased rate of fetal loss in Down syndrome (and unaffected) pregnancies (5). Rather than choosing either first- or second-trimester screening, it has been proposed that selected results from both trimesters be combined into a single "integrated" risk that is provided in the second trimester (6). The current study focuses on the possible role that a newly proposed marker, invasive trophoblast antigen (ITA), might play in a first-trimester screening test for Down syndrome. ITA is essentially equivalent to hyperglycosylated human chorionic gonadotropin (hCG). To date, studies have found that ITA is increased in Down syndrome pregnancies: in both maternal urine and serum in the second trimester, and in maternal urine in the first trimester (7)(8)(9)(10)(11)(12)(13)(14). There are 3 published reports of the Down syndrome screening performance of urine ITA in the first trimester, and all report detection rates at a 5% false-positive rate. One study, using the manual assay, found a 38% detection rate among 8 cases (8); the second study, using the automated assay, found a 24% detection rate among 17 cases (14); and the third found that the detection rate in 73 cases varied from 16% to 43% at 11 to 13 weeks of gestation (13). The strengths of the current study are that the samples were collected as part of an observational study in the first trimester (15), ITA was measured in maternal serum with an automated assay (16), and a large number of first-trimester Down syndrome pregnancies were included.
| Materials and Methods |
|---|
|
|
|---|
-fetoprotein, unconjugated estriol, hCG, free ß, and PAPP-A. Remaining sera were aliquoted into 1-mL cryovials and frozen at 20 °C. Subsequently, a casecontrol set was constructed, with 5 samples from pregnancies with normal karyotypes matched for each Down syndrome case, as follows: length of freezer storage, maternal age, gestational age, race, and site where the sample was obtained. For the current study, a never-thawed aliquot from this casecontrol set was available for 54 Down syndrome cases and 276 controls. Aliquots of the 330 serum samples were sent on dry ice to Quest Diagnostics Nichols Institute for ITA measurement. Samples were quick-thawed in a 22 °C waterbath because this handling has been shown appropriate for maintaining ITA reactivity in urine (7). The assays were completed without knowledge of whether the sample was from a case or control, and the ITA results were reported to the Foundation for Blood Research for statistical analysis.
ita measurements
ITA was analyzed by an automated immunochemiluminometric assay. The monoclonal antibody (B152) specific for ITA used in this assay has minimal cross-reactivity with hCG and its free ß-subunit, and it is applicable to many sample types, including serum (16). Each sample was assayed in singleton. The assay has a calibration range of
300 µg/L and a detection limit of 0.2 µg/L. CVs were determined by use of 3 controls with ITA concentrations of 1.1, 8.5, and 18.2 µg/L. The intra- and interassay CVs were <3.5% and <7.4%, respectively, for all 3 controls.
data analysis
All ITA results were converted to multiples of the median (MoM) and corrected for maternal weight (17). Adjustments to the other analytes measured in the same samples, along with summary population characteristics, have been published (18). The standard deviation of ITA in Down syndrome pregnancies was adjusted to take into account the varying mean concentrations by gestation week. Correlation coefficients between ITA and other serum markers in first-trimester unaffected and Down syndrome pregnancies were derived separately after logarithmic transformation and exclusion of values outside 3 SD. Correlations between ITA and NT measurements in Down syndrome and unaffected pregnancies were assumed to be 0.
The Down syndrome screening performance was modeled for selected combinations of serum ITA and other first-trimester serum markers, with and without NT measurements. The modeling methodology is based on overlapping gaussian distributions and has been described previously (9)(18). The maternal age distribution in the United States for 2000 (19) was used as the baseline population. The required values for all other first-trimester markers have been taken from a published study (13).
| Results |
|---|
|
|
|---|
29% per week.
|
Fig. 2
shows the observed median ITA MoM in Down syndrome pregnancies by completed week of gestation. ITA measurements are more predictive of Down syndrome (i.e., higher) as gestational age increases. The increasing values fit a linear regression well. The predicted median MoM for weeks 1013 were 1.81, 2.02, 2.23, and 2.44, respectively. Before 12 weeks of gestation, 5 of 25 cases (20%) exceeded the 95th centile (3.19 MoM). At 12 weeks of gestation and later, 10 of 29 cases (34%) had increased ITA measurements, with the latter proportion being significantly higher (P = 0.03, Fisher exact test).
|
Fig. 3
shows a probability plot of the ITA MoM in the 54 Down syndrome and 276 unaffected pregnancies. Both groups fit a gaussian distribution well after logarithmic transformation. As expected, the median MoM in unaffected pregnancies was 1.00 (logarithmic mean of 0.000), with a corresponding logarithmic SD of 0.3064. Among the Down syndrome pregnancies, the observed logarithmic SD was 0.2393. After taking into account the variation in mean MoM values shown in Fig. 2
, the revised estimate of the SD was 0.2326. This revised estimate was used for modeling screening performance. Correlation coefficients between ITA and PAPP-A, free ß, hCG, and DIA in unaffected and Down syndrome pregnancies were 0.137, 0.637, 0.766, and 0.617 and 0.186, 0.680, 0.695, and 0.528, respectively. Reasonable truncation limits for ITA measurements are between 0.6 and 5.8 MoM.
|
Fig. 4
shows a scatterplot of ITA MoM values (logarithmic x axis) vs free ß MoM values (logarithmic y axis) for both unaffected and Down syndrome pregnancies. Of the 14 Down syndrome pregnancies with ITA measurements exceeding the 95th centile, 10 also exceeded the 95th centile for free ß. This, along with the relatively high correlation, indicates that once one of these markers is included in a protocol, the addition of the second is unlikely to be of much value. A similar finding was observed when ITA and hCG measurements were compared (data not shown).
|
Table 1
shows the results of modeling first-trimester Down syndrome screening using ITA measurements in combination with maternal age and other first-trimester markers. Down syndrome detection rates are shown for 3 different false-positive rates, separately, for 11, 12, and 13 weeks of gestation. The last 3 columns show the 3 pooled detection rates, assuming that 25% of the pregnancies were screened at 11 and 13 weeks of gestation, with the remaining 50% of pregnancies screened at 12 weeks of gestation. When we considered a combination of maternal age, PAPP-A, and 1 of the 3 hCG-related analytes (free ß, hCG, or ITA), both free ß and ITA were equivalent, with a pooled detection rate of
67% at a 5% false-positive rate. When NT measurements were included, performance was again similar for the 2 combinations, with an 84% pooled detection rate. When DIA measurements were also added, the detection rate increased only slightly for both combinations: to 87% or 88%. Substituting hCG measurements for free ß or ITA measurements gave similar performance (84% without and 87% with DIA measurements), as long as NT measurements were included. The addition of DIA measurements produced only a modest improvement. Combining both ITA and free ß measurements together again produced only modest improvement in performance, when NT measurements were included.
|
Table 2
is constructed in a similar manner but provides the false-positive rates at 3 selected Down syndrome detection rates. Overall, combinations with free ß measurements had false-positive rates similar to those found when ITA was used instead. Table 3
shows Down syndrome detection and false-positive rates at 4 first-trimester Down syndrome risk cutoffs (1:100, 1:150, 1:200, and 1:250) that roughly correspond to first-trimester risks in 38-, 36-, 35-, and 34-year-old women. In general, combinations using free ß measurements had slightly lower detection and false-positive rates than the corresponding combinations with ITA measurements. For example, at a risk cutoff of 1:200, the combination of maternal age, PAPP-A, free ß, and NT had a detection rate of 83% at a false-positive rate of 4.7%. Substituting ITA measurements for free ß leads to a slight increase in the detection rate, to 84%, with a concomitant increase in the false-positive rate, to 5.2%. Overall, measurements of hCG performed nearly as well as ITA. This was not so surprising when the combinations included NT measurements, but the similar performance when only serum markers were used is not consistent with much of the literature in the first trimester, which suggests that hCG measurements are not quite as predictive (20)(21). Our estimates for these 2 markers (Tables 13
) were, as expected, nearly identical to those in the Serum, Urine, and Ultrasound Screening Study (SURUSS) [Table 44 in Ref. (13)].
|
|
| Discussion |
|---|
|
|
|---|
One area of concern is that the logarithmic SD for ITA measurements in Down syndrome pregnancies (0.2326) is much tighter than in unaffected pregnancies (0.3064). In our previous study of ITA in second-trimester serum samples, the SD in unaffected pregnancies was similar to that found in the current study (0.3088) (12), suggesting that this estimate is reliable. The SD of ITA measurements in second-trimester Down syndrome pregnancies, however, was broader (0.4640), suggesting that one of these estimates may not be reliable. It is possible that the very high ITA measurements found in first-trimester Down syndrome pregnancies are near the upper limit of 300 µg/L for the current assay. This study relied on results from undiluted samples. On the other hand, a recent literature review summarizing published reports for DIA, PAPP-A, hCG, free ß, and NT measurements (Palomaki et al., Maternal serum dimeric inhibin-A and other markers of Down syndrome in the late first trimester of pregnancies, submitted for publication), the consensus pooled SD in Down syndrome pregnancies is always smaller than the corresponding values in unaffected pregnancies. The results from individual studies reporting hCG and free ß differed, however, from the other 3 analytes in that nearly one-half of them reported smaller SDs among the Down syndrome pregnancies. This finding thus may not be so unexpected, but additional studies are needed to confirm the reported first-trimester serum ITA population values.
In a wide variety of first-trimester Down syndrome screening test combinations and cutoffs, the performance of serum ITA and free ß was essentially equivalent. If the strength of the associations between ITA measurements and Down syndrome found in the current study are confirmed, the decision to use free ß or ITA measurements will most likely depend on considerations other than screening performance. These concerns may include cost, ease of assay performance (automation), availability of quality reagents, and licensing issues. Although hCG measurements alone are not as good a marker of Down syndrome as either free ß or ITA measurements, when combined with maternal age, PAPP-A, and NT measurements, the difference in overall performance is small.
| Acknowledgments |
|---|
| Footnotes |
|---|
| References |
|---|
|
|
|---|
The following articles in journals at HighWire Press have cited this article:
![]() |
S.F. de Medeiros and R.J. Norman Human choriogonadotrophin protein core and sugar branches heterogeneity: basic and clinical insights Hum. Reprod. Update, January 1, 2009; 15(1): 69 - 95. [Abstract] [Full Text] [PDF] |
||||
![]() |
U.-H. Stenman, A. Tiitinen, H. Alfthan, and L. Valmu The classification, functions and clinical use of different isoforms of HCG Hum. Reprod. Update, November 1, 2006; 12(6): 769 - 784. [Abstract] [Full Text] [PDF] |
||||
![]() |
T. W. Hallahan, D. A. Krantz, V. J. Macri, and T. J. Shine First-Trimester Down Syndrome Screening Clin. Chem., January 1, 2006; 52(1): 160 - 161. [Full Text] [PDF] |
||||
![]() |
G. E. Palomaki, G. J. Knight, L. M. Neveux, J. E. Haddow, and R. Pandian First-Trimester Down Syndrome Screening: Reply Clin. Chem., January 1, 2006; 52(1): 161 - 161. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |