Maternal serum invasive trophoblast antigen (hyperglycosylated hCG) as a screening marker for down syndrome during the second trimester.
Hyperglycosylated hCG is a unique carbohydrate variant of hCG (6); it is produced by poorly differentiated or invasive trophoblast cells and, hence, is called invasive trophoblast antigen (ITA) (7). An initial 1999 study reported an 80% detection rate (at a 5% false-positive rate) for second-trimester urine ITA measurements (8). A second study in 1999 found a lower detection rate of 45% at 14 weeks of gestation and later (9). Urine ITA was also analyzed as part of the Serum, Urine, and Ultrasound Screening Study (SURUSS) in 2003 and was reported to detect 40% of second-trimester Down syndrome pregnancies (10). In a prospective study in 2004 in which fresh urine samples were used (11), a detection rate of 59% was documented. Although urine is routinely collected during pregnancy, most current screening practice relies on serum samples. The present study, therefore, explores the potential usefulness of maternal serum ITA measurements for Down syndrome screening in the second trimester. We used stored serum samples collected during an earlier observational trial involving an unbiased group of 5345 pregnant women who were about to undergo amniocentesis for reasons other than a positive serum screening test. Chromosome results were available on all of these pregnancies.
Materials and Methods
Second-trimester maternal serum samples from Down syndrome and unaffected pregnancies were collected as part of a previous Maternal and Child Health-sponsored trial (MCJ-237014) between 1990 and 1992 (12). Briefly, 14 prenatal diagnostic centers in California and elsewhere in the United States recruited women already scheduled to have an amniocentesis between 14 and 21 weeks of gestation, usually because of advanced maternal age. None of the women had undergone any serum screening in their current pregnancy. Demographic and pregnancy-related information was collected, including maternal weight, maternal race and diabetic status, and gestational age estimates based on both biparietal diameter and date of last menstrual period. Maternal serum was collected before amniocentesis and shipped by express mail to the study center in Maine, where it was aliquoted and assayed for AFP, uE3, and hCG. Remaining sera were frozen at -20[degrees]C in a frost-free freezer. Subsequently, a case/control set was constructed, with five 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 sample was obtained. In 1996, an aliquot of this case/control set was retrieved for another Maternal and Child Health-sponsored study comparing hCG and the free [beta]-subunit of hCG (MCJ 230802) (13). In 1997, this same sample set was used to study the usefulness of DIA as an additional Down syndrome marker (4).
For the current study, a never-thawed aliquot from this case/control set was available for each of 45 cases and 238 controls. Three of the case samples and seven of the control samples had been exhausted. Sera from the 283 serum samples were sent on dry ice to Quest Diagnostics, Nichols Institute, for measurement of ITA. 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.
Measurements were carried out using an automated immunochemiluminometric assay. The monoclonal antibody (B152) specific for ITA used in this assay has minimal cross-reactivity with hCG and is applicable to many sample types, including serum (14). Singleton samples were assayed. The assay has a calibration range up to 300 /,g/L and a lower limit of detection of 0.2 [micro]g/L. CVs were determined with three controls with ITA concentrations of 1.1, 8.5, and 18.2 [micro]g/L. The intra- and interassay CVs were <3.5% and 7.4%, respectively, for all three controls.
All ITA results were converted to multiple(s) of the median (MoM) and corrected for maternal weight (15). Adjustments to the other analytes and summary population variables have been published (4,13). Correlation coefficients between ITA and serum markers in second-trimester unaffected and Down syndrome pregnancies were derived separately after a logarithmic transformation and exclusion of values outside 3 SD. The Down syndrome screening performance was modeled for selected combinations of serum ITA and other second-trimester serum markers, using the maternal age distribution in the United States for 2000 (16). The modeling methodology is based on overlapping gaussian distributions and has been described previously (11). Continuous variables were compared, after appropriate transformations, by the Mest at a two-tailed significance level of 0.05.
The observed ITA measurements in the case and control samples vs gestational age are shown in Fig. 1. Two sets of medians were computed based on gestational age derived from either last menstrual period or biparietal diameter ultrasound measurements: ultrasound median ITA = [10.sup.389519 - 0.0209419 x days]; and last menstrual period median ITA = [10.sup.3.84743 - 0.0209275 x days]. The weight adjustment equation for ITA is: expected ITA = 68.2934 X 1/weight + 0.5419. The observed ITA performance for detecting Down syndrome was computed after all values were converted to weight-adjusted MoM (Table 1). The observed detection rate of ~40% at a 5% false-positive rate was similar to that found for hCG measurements in serum (13).
[FIGURE 1 OMITTED]
The ITA measurements in Down syndrome and unaffected pregnancies were then fitted to straight lines on a probability plot to compute appropriate population parameters. Fig. 2 displays those data when gestational dating is based on the first day of the last menstrual period. In both unaffected and Down syndrome pregnancies, the ITA measurements fit a gaussian distribution reasonably well after a logarithmic transformation. We obtained similar results when we estimated gestational age using biparietal diameter measurements (not shown). The parameters were derived by fitting the data between the 10th and 90th centiles for the two groups and are shown in Table 2. The median ITA measurements in Down syndrome pregnancies differed considerably by method of dating, but the geometric mean ITA values for last menstrual period and ultrasound dating that were used for modeling were nearly identical (3.07 and 3.15 MoM). The geometric mean is more robust than the median for modeling small data sets.
[FIGURE 2 OMITTED]
The population parameters in Table 2 were used to model univariate ITA performance (Table 1). These should be similar to the observed rates. The modeled detection rates were actually somewhat higher in the present comparison. This is explained by the deviation from the expected (Fig. 2, solid line) in the probability plot at higher centiles of unaffected pregnancies. This deviation occurred because the observed data (Fig. 2, O) were higher than the modeled data (Fig. 2, solid line). Extensive experience with other screening markers indicates that the modeled performance is likely to be the more reliable. Pairwise correlation coefficients between ITA and the other serum measurements are shown in Table 3, and the correlation found between serum ITA and hCG in both unaffected and Down syndrome pregnancies is shown in Fig. 3. The high correlation between ITA and hCG (and ITA and free [beta]-hCG) indicated that including either of these markers with ITA in a Down syndrome risk algorithm would not be useful. Reasonable truncation limits are 0.5-6.0 MoM.
The modeled Down syndrome screening performance is shown in Table 4 and is based on the maternal age distribution for the United States in 2000 (16), along with the ITA parameters from Table 2 and published parameters for AFP, uE3, hCG, and DIA (4,13). The performance estimates for the triple test (AFP, uE3, and hCG) and the quadruple test (triple test plus DIA) differed slightly from those published previously because the distribution of maternal ages in the United States was different in 2000 from that available for the earlier calculations. ITA could be substituted for hCG (or free [beta]-hCG) measurements with no loss in performance for any of the test combinations. At a 5% false-positive rate, 77% of Down syndrome pregnancies were detected by either of the "quadruple tests" when the gestational age was estimated by last menstrual period dating; detection improved to 79% when the dating was based on biparietal diameter ultrasound measurements. Although ITA and hCG had a relatively high correlation coefficient in both affected and unaffected pregnancies (~0.8), adding the two markers together in a screening model with other, more independent markers, still led to materially improved performance.
[FIGURE 3 OMITTED]
Measurements of ITA, a unique carbohydrate variant of hCG, in urine or serum have been reported to be a promising marker for Down syndrome in either the first or second trimester (8-11,17,18). The authors of previous studies using second-trimester serum have reported median ITA concentrations of 3.90 MoM and 5.33 MoM, along with detection rates of 60% (17) and 81% (18), respectively. However, in the first study, the detection rate was based on only 10 samples from affected pregnancies, and in the second study, assessment of the high detection rate was complicated by the associated unexpectedly high detection for hCG measurements of 75%. Our study explores the potential usefulness of ITA measurements in second-trimester maternal serum, both alone and in combination with other commonly used maternal serum markers. Using standard methods of data analysis that have been shown to be robust in the past, we found that ITA performance was essentially equivalent to hCG. For example, among 533 serum ITA measurements collected as part of another study (14), there was no such deviation from expected at high ITA concentrations among the unaffected pregnancies, as found in the current study (Fig. 2). The geometric mean ITA MoM for Down syndrome pregnancies was 3.15, higher than that found for either hCG or free [beta]-hCG measurements. Performance, however, was similar regardless of which of these three markers were used because of a higher population variance for ITA measurements. Our median ITA was somewhat lower than that reported in two previous smaller studies (17,18).
Some laboratories may find that replacing hCG with ITA is more economical or is more easily fit into the laboratory routine. If so, Down syndrome screening performance should not change, providing that the updated population parameters found in our study are used. However, issues such as screening in twin pregnancies or the use of ITA in identifying pregnancies at risk for trisomy 18 should be addressed before ITA is routinely added to the existing triple or quadruple marker panels used to screen for Down syndrome. The highest screening performance for Down syndrome can be obtained by integrating first- and second-trimester serum and ultrasound markers into a single interpretation in the second trimester (19). This integrated test approach can detect 90% of Down syndrome pregnancies at a 3% false-positive rate. The present study indicates that second-trimester measurements of serum ITA might also be substituted for hCG or free [beta]-hCG in such an integrated approach.
We thank Esther Carlton (Clinical Correlations Department, Quest Diagnostics, Nichols Institute) and Julie Lu and Jola Plewnia (Quest Diagnostics, Nichols Institute) for assistance in assaying of the samples for various biomarkers. This study was partially funded by Quest Diagnostics, Nichols Institute, San Juan Capistrano, CA.
(1.) Haddow JE, Palomaki GE, Knight GJ, Williams J, Pulkkinen A, Canick JA, et al. Prenatal screening for Down's syndrome with use of maternal serum markers. N Engl J Med 1992;372:588-93.
(2.) Palomaki GE, Knight GJ, McCarthy JE, Haddow JE, Donhowe JM. Maternal serum screening for Down syndrome in the United States: a 1995 survey. Am J Obstet Gynecol 1997;176:1046-51.
(3.) Wald NJ. Down's syndrome. In: Wald NJ, Leck I, eds. Antenatal and neonatal screening, 2nd ed. Oxford: Oxford University Press, 2000:85-155.
(4.) Haddow JE, Palomaki GE, Knight GJ, Foster DL, Neveux LM. Second trimester screening for Down's syndrome using maternal serum dimeric inhibin A. J Med Screen 1998;5:115-9.
(5.) Wald NJ, Densem J, George L, Muttukrishna S, Knight P. Prenatal screening for Down's syndrome using inhibin-A as a serum marker. Prenat Diagn 1996;16:143-53.
(6.) Elliott MM, Kardana A, Lustbader JW, Cole LA. Carbohydrate and peptide structure of the a- and R-subunits of human chorionic gonadotropin from normal and aberrant pregnancy and choriocarcinoma. Endocrine 1997;7:15-32.
(7.) Kovaleskaya G, Genbacev O, Fisher SJ, Caceres E, O'Connor J. Trophoblast origin of hCG isoforms: cytotrophobasts are the primary source of choriocarcinoma-like hCG. Mol Cell Endocrinol 2002;94:147-55.
(8.) Cole LA, Shahabi S, Oz UA, Bahado-Singh R0, Mahoney MJ. Hyperglycosylated human chorionic gonadotropin (invasive trophoblast antigen) immunoassay: a new basis for gestational Down syndrome screening. Clin Chem 1999;45:2109-19.
(9.) Cuckle HS, Shahabi S, Schmi IK, Jones R, Cole LA. Maternal urine hyperglycosylated hCG in pregnancies with Down syndrome. Prenat Diagn 1999;19:918-20.
(10.) Wald NJ, Rodeck C, Hackshaw AK, Walters J, Chitty L, Mackinson AM. First and second trimester antenatal screening for Down's syndrome: the results of the Serum, Urine and Ultrasound Screening Study (SURUSS). J Med Screen 2003;10:56-104.
(11.) Palomaki GE, Knight GJ, Roberson MM, Cunningham GC, Lee JE, Strom CM, et al. Invasive trophoblast antigen (hyperglycosylated human chorionic gonadotropin) in second trimester maternal urine as a marker of Down syndrome: preliminary results of an observational study on fresh samples. Clin Chem 2004;50:182-9.
(12.) Haddow JE, Palomaki GE, Knight GJ, Cunningham GC, Lustig LS, Boyd PA. Reducing the need for amniocentesis in women 35 years of age or older with serum markers for screening. N Engl J Med 1994;330:1114-8.
(13.) Knight GJ, Palomaki GE, Neveux LM, Fodor KK, Haddow JE. HCG and the free R-subunit as screening tests for Down syndrome. Prenat Diagn 1998;187:235-45.
(14.) Pandian R, Lu J, Ossolinska-Plewnia J. Fully automated chemiluminometric assay for hyperglycosylated human chorionic gonadotropin (invasive trophoblast antigen). Clin Chem 2003;49:808-10.
(15.) Neveux LM, Palomaki GE, Larrivee DA, Knight GJ, Haddow JE. Refinements in managing maternal weight adjustment for interpreting prenatal screening results. Prenat Diagn 1996;16: 1115-9.
(16.) Centers for Disease Control and Prevention. Vital health statistics 2000 natality data set, series 21, number 14 [Database on CD-ROM]. Hyattsville, MD: Department of Health & Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics, April 2002.
(17.) Shahabi S, Oz UA, Bahado-Singh R, Mahoney JM, Omarani A, Baumgarten A, et al. Serum hyperglycosylated hCG: a potential screening test for fetal Down syndrome. Prenat Diagn 1999;19: 488-90.
(18.) Pandian R, Cole LA, Palomaki GE. Second-trimester maternal serum invasive trophoblast antigen: a marker for Down syndrome screening. Clin Chem 2004;50:1433-5.
(19.) Wald NJ, Watt HC, Hackshaw AK. Integrated screening for Down's syndrome on the basis of tests performed during the first and second trimesters. N Engl J Med 1999;341:461-7.
GLENN E. PALOMAKI,  * LOUIS M. NEVEUX,  GEORGE J. KNIGHT,  JAMES E. HADDOW,  and RAJ PANDIAN 
 Foundation for Blood Research, Scarborough, ME.
 Quest Diagnostics, Nichols Institute, San Juan Capistrano, CA.
 Nonstandard abbreviations: hCG, human chorionic gonadotropin; AFP, a-fetoprotein; uE3, unconjugated estriol; DIA, dimeric inhibin-A; ITA, invasive trophoblast antigen; and MoM, multiple(s) of the median.
* Address correspondence to this author at: Foundation for Blood Research, PO Box 190, Scarborough, ME 04070-00190. Fax 207-885-0807; e-mail email@example.com.
Received June 4, 2004; accepted July 26, 2004.
Previously published online at DOI: 10.1373/clinchem.2004.038059
Table 1. Maternal serum ITA: Observed and modeled Down syndrome detection rates at three false-positive rates, stratified by method of gestational dating. Down syndrome detection rate, % LMP (a) dating BPD dating False-positive rate, % Observed Modeled Observed Modeled 1 28 27 22 31 3 28 38 36 42 5 38 45 40 48 (a) LMP, last menstrual period; BPD, biparietal diameter. Table 2. Second-trimester maternal serum ITA: Distribution parameters in Down syndrome and unaffected pregnancies, stratified by method of gestational dating. Median Log(mean) (geometric mean) Log(SD) LMPadating Unaffected 0.0000 1.00 (1.00) 0.3384 Down syndrome 0.4987 3.54 (3.15) 0.4781 BPD dating Unaffected 0.0000 1.00 (1.00) 0.3088 Down syndrome 0.4876 2.94 (3.07) 0.4640 (a) LMP, last menstrual period; BPD, biparietal diameter. Table 3. Second-trimester maternal serum ITA measurements vs five other second-trimester serum markers: Correlation coefficients in Down syndrome and unaffected pregnancies, stratified by method of gestational dating. Correlation coefficient Combination of markers Unaffected Down syndrome LMP (a) dating AFP/ITA -0.013 -0.369 uE3/ITA -0.301 -0.363 hCG/ITA 0.812 0.814 Free [beta]-hCG/ITA 0.800 0.921 DIA/ITA 0.446 0.309 BPD dating AFP/ITA 0.050 -0.214 uE3/ITA -0.153 -0.117 hCG/ITA 0.795 0.780 Free [beta]-hCG/ITA 0.774 0.878 DIA/ITA 0.480 0.226 (a) LMP, last menstrual period; BPD, biparietal diameter. Table 4. Second-trimester maternal serum ITA: Modeled Down syndrome screening performance, alone and in combination with other serum markers at fixed false- positive rates, for two methods of gestational dating. Down syndrome detection rate, % 1% FPR (a) 3% FPR 5% FPR Maternal age and LMP BPD LMP BPD LMP BPD ITA 31 35 46 53 56 58 AFP, ITA 40 45 55 58 62 65 AFP, hCG 44 44 57 59 65 67 AFP, uE3, hCG 45 50 60 65 67 72 AFP, uE3, ITA 44 51 58 65 66 71 AFP, uE3, hCG, ITA 48 55 63 69 70 76 AFP, uE3, hCG, DIA 57 59 71 73 77 79 AFP, uE3, DIA, ITA 57 59 71 73 77 79 AFP, uE3, DIA, hCG, ITA 60 64 74 78 80 83 (a) FPR, false-positive rate; LMP, last menstrual period; BPD, biparietal diameter.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||General Clinical Chemistry|
|Author:||Palomaki, Glenn E.; Neveux, Louis M.; Knight, George J.; Haddow, James E.; Pandian, Raj|
|Date:||Oct 1, 2004|
|Previous Article:||Rapid and simultaneous determination of coproporphyrin and protoporphyrin in feces by derivative matrix isopotential synchronous fluorescence...|
|Next Article:||Kappa free light chains in cerebrospinal fluid as markers of intrathecal immunoglobulin synthesis.|