American Journal of Medical Genetics 36480-483 (1990)
Abnormal Maternal Serum Levels of Human Chorionic Gonadotropin Free Subunits in Trisomy 18 Mehmet Ozturk, Aubrey Milunsky, Bruno Brambati, E.S. Sachs, Susan L. Miller, and Jack R. Wands Molecular Hepatology Laboratory, MGH Cancer Center, Massachusetts General Hospital, Charleston, and Department of Medicine, Harvard Medical School (M.O., J.R.W.), and Center for Human Genetics, Boston University School of Medicine (A.M., S.L.M.), Boston; First Institute of Obstetrics and Gynecology, University of Milan, Italy (B.B.); Department of Clinical Genetics, Erasmus University, Rotterdam, The Netherlands (E.S.S.)
We have measured maternal serum levels of free Q and p subunits of human chdionic gonadotropin between 8 and 12 weeks of gestation in 704 women at increased risk for trisomy. This group was studied because of advanced maternal age or a previous birth with chromosomal abnormality. All sera had been collected prior to chorion villus biopsy for prenatal diagnosis. Serum levels of free Q and phCG were determined by specific monoclonal antibody-based immunoradiometric assays. Analysis of chorionic tissue showed that in 38 of 704 (5.4%) pregnancies the fetus had a chromosome abnormality.There were 8 fetuses with trisomy 18(1.1%)and 9 (1.3%)with trisomy 21. In all pregnancies carrying a trisomy 18 fetus, we observed either high levels of free ahCG or low levels of free phCG or both. More importantly, the calculated ratio of free PhCGhhCG was less than 0.25 multiples of the median (MOM) in 6 of 8 (75%) trisomy 18 cases. Only 21 of 666 mothers (3.2%) carrying a normal fetus had a ratio less than 0.25 MOM (P < 0.0001). There was no difference between this ratio in trisomy 21 and normal pregnancy. Thus, when adjusted for gestational age, a low free phCG/ahCG ratio in maternal serum indicates a pregnancy at high risk [RR = 72 (95%CI 32,16211 for trisomy 18.
INTRODU~CTION Highly sensitive and specific monoclonal immunoradiometric assays have been developed for measurement of free a and p subunits of human chorionic gonadotropin (ahCG and phCG) in serum [Ozturk et al., 19871. Our previous investigations have demonstrated that serum levels of free a and p hCG were closely regulated throughout normal pregnancy. For example, we have observed that free phCG levels were higher than ahCG between 5 and 8 weeks of pregnancy. Similar concentrations of both free subunits were found between 9 and 12 weeks of gestation followed by a reversal such that higher levels of free ahCG were measured at term [Cole et al., 1984; Ozturk e t al., 1987, 1988al. We [Ozturk et al., 1987, 1988a], and others [Cole et al., 19841 suggested that the progressive changes in the relative amount of free subunits in serum were due in part to the continuous differentiation of trophoblastic cells. In support of this hypothesis was the finding that higher than normal levels of free phCG [Khazaeli et al., 1986; Ozturk et al., 1987,1988b;Hay, 19881were noted to accompany low levels of free ahCG in serum of women with gestational trophoblastic diseases such as hydatidiform mole and choriocarcinoma [Ozturk et al., 1987, 1988133. Extensive histopathologic and cytogenetic changes have suggested that trophoblastic differentiation is abnormal i n molar placenta and choriocarcinoma tissues; such changes have been accompanied by specific chromosome abnormalities [Szulman and Surti, 1978; KEY WORDS: free aihCG, free PhCG, chroJacobs e t al., 19821. Indeed, abnormal development of mosomal abnormality the placenta is one of the most common pathologic changes in the autosomal trisomies, but little is known regarding specific morphological changes [Hassold and Jacobs, 1984; Bou6 et al., 19851.Bogart et al. [19871has recently described elevated levels of maternal serum hCG in fetal Down syndrome. Based on these observations, we have measured free hCG subunit levels in the sera of mothers bearing fetuses with chromosome abnormalities and compared our results to those for women of Ekceived €or publication August 22,1989; revision received Dea similar age with a normal pregnancy outcome. To our cember 1, 1989. Address reprint requests to Mehmet Ozturk, PhD, Molecular surprise, we found a strikingly low phCG/ahCG Hepatology Laboratory, MGH Cancer Center, MGH East, 7th (p/ahCG) ratio in trisomy 18 but not in trisomy 21 or other chromosome abnormalities. Floor, 149 13th Street, Charlestown, MA 02129.
0 1990 Wiley-Liss, Inc.
Chorionic Gonadotropin in Trisomy 18 METHODS Patients Pregnancies at increased risk for chromosomal abnormalities were studied between 8 and 12 weeks of gestation. Two groups were studied: 1)The largest population was prospectively evaluated in Milan, Italy, and consisted of 491 women planning to undergo chorion villus sampling (CVS) because of advanced maternal age or previous birth of a child with a chromosome or other development defect. Fetal age was established precisely by ultrasound examination and serum samples were obtained between 8 and 12 weeks of gestation and shortly before CVS. Sera were stored at - 20°C prior to analysis. A detailed clinical description of this population was previously reported [Brambati et al., 1987; Milunsky et al., 19881.2)The second patient population, from Rotterdam, was also investigated prospectively. Two hundred and thirteen women were studied at 9-11 weeks of gestation. Similar to the first study, fetal age was established by ultrasonography and maternal sera were obtained prior to CVS biopsy. All measurements of free ahCG and phCG subunit serum levels were performed under code without knowledge of the clinical outcome and cytogenetic results.
481
free a and phCG was determined from a standard curve and statistical analysis was performed as described previously [Ozturk et al., 1987; Milunsky et al., 19881.
RESULTS The Milan population comprised 491 women of whom 25 (5.1%)had CVS biopsies diagnostic of a fetus with an abnormal karyotype. Chromosome analysis of CVS tissue from the Rotterdam population showed that 13out of 213 (6.1%)carried an aneuploid fetus. Of the 666 combined normal pregnancies, the median maternal serum free ahCG levels were found to be 17.5 ng/ml a t 8 weeks, 25.5 ng/ml at 10 weeks, and 32 ng/ml at 12 weeks. Median free phCG levels were found to be 79.5 ng/ml at 8 weeks, 65.5 ng/ml a t 10 weeks, and 60.5 ng/ml at 12 weeks respectively (Table I). Table I1 depicts the absolute maternal serum free subunit concentrations in 38 pregnancies associated with fetal aneuploidy. The plahCG ratios are calculated and multiples of the median (MOM)values are determined by comparison to normal pregnancies adjusted to the same gestational age. As shown in Figure 1, this ratio in normal pregnancy was 4.9 at 8 weeks and progressively decreased with advancing gestational age to a level of 1.65 at 12 weeks. Figure 1 compares the p/ahCG ratios in trisomy 18 to those measured in normal pregnancies. The MOMin trisomy 18 was less than 0.5 in all cases. More importantly, the free p/ahCG ratio, which was less than 0.25 in only 21 out of 666 normal pregnancies (3.2%),was less than 0.25 MOMin 6 out of 8 (75.0%)trisomy 18. This ratio was found to be significantly lower than the population carrying a normal fetus matched for the same gestational age (P < 0.0001). Thus, when adjusted for gestational age, a low p/ahCG ratio in maternal serum indicates a pregnancy at high risk [RR = 72 (95% CI 32, 162)l for trisomy 18. No consistent changes in a or p free subunit ratios were observed in trisomy 21 or other chromosome abnormalities. This indicates a high degree of specificity of our findings for trisomy 18 (Table 11).Indeed, this low ratio appeared associated with trisomy 18 since only 2 of 30 pregnancies (6.7%) with other chromosome abnormalities had such a low ratio and when considered as a group did not significantly differ from the ratio found in normal pregnancy.
Immunoradiometric Assays The specific and sensitive immunoradiometric assays used in this study for the detection of free ahCG and phCG were described previously [Ozturk et al., 19871. The general assay design employswell-definedmonoclonal antibodies that recognize specific epitopes on the free subunits. By incubation with antibody-coated beads, the free subunit is first captured on the solid phase support. Following a washing step, the beads are incubated with a second '251-labeled antibody which binds to the captured a and phCG. The number of counts bound per minute is directly proportional to the amount of subunits present in serum. The free ahCG assay uses a free a subunit-specific capture antibody (AHT2O) on the solid phase support [Ozturk et al., 1987;Bidart et al., 19881. The radiolabeled indicator antibody (HT13) has been shown to react with a distinct but different epitope on the free a subunit. The free ahCG assay has a lower limit of sensitivity of 0.20 ng/ml and does not detect native hCG or any other glycoproteinhormones [Ozturk DISCUSSION et al., 19871. Similarly, the free phCG assay uses a free The aim of this study was to determine whether maPhCG-specific antibody (FBT11)on the solid phase support [Ozturk et al., 1987; Bidart et al., 19871.The radio- ternal first trimester serum levels of hCG free subunits labeled indicator antibody (FBT10) binds to a separate are abnormal in women carrying a fetus with a chromoepitope on the phCG molecule. This assay detects free phCG a t a lower level of 0.20 nglml and does not react with intact hCG [Ozturk et al., 19873. Detailed studies of TABLE I. Medians of hCG Free Subunit Serum the assay design, antibody specificity, and assay perforConcentrations in Normal Pregnancies mance (intra-assay and inter-assay variations) have been previously published [Ozturk et al., l987,1988a,bI. Gestational age Free ahCG Free phCG Data Analysis The ratio of free p/ahCG was calculated by dividing phCG by ahCG concentrations expressed as nglml following simultaneous measurements of free subunit levels on the same serum sample. The absolute level of
(weeks)
n
(ngiml)
(nglml)
8 9 10 11 12
14 197 360 75 20
17.5 21.0 25.5 23.0 32.0
79.5 75.0 65.5 58.0 60.5
482
Ozturk et al. TABLE 11. Serum Levels of hCG Free Subunits With Chromosome Abnormalities
LMP Fetal karyotype (weeks) Trisomv 18 47,XY, + 18 9 47,XY, + 18 9 47,XY. + 18 9 47;xy; + 18 9 47.xx. + 18 10 47;XXj + i 8 8 47,xx, + 18 10 47,XX, + 18 10 Trisomy 21 and other chromosome defects 47,XY, + 21 10 47,xx, + 21 10 47,XY, + 21 9 47,xx, + 21 9 47,XY, + 21 10 47,xx, + 21 12 47,XY, + 21 11 47,XY, + 21 11 11 47,XY,+21 10 47,XX, + 13 47;XY; + 13 11 47,xx, + 22 10 47,XXY 9 48;XX, + 18,+ 11 10 47,XX,+21/48,XX,+ 11,+21 10 46,XXde113,t(13;y) 10 8 46,X,inv(Y),- 14,t(13;14) 45,X/46,XX 10 45,X/46,XX 10 46;XYl45,X 10 46,XX,t(1;2-8) 10 47.XY. + m 10 46;XY; + t(4;8) 10 10 46.XXI47.XX. + 7 10 92:xxy Y’ 10 47,XY + 22 46,XY,t(6;11) 10 10 46,XW45x 46,XY,t(5;12) 10 46,XY,lqh+ 10
some defect. Our study indicates that serum levels of free ahCG were high, whereas those of free phCG were low in trisomy 18,thus giving rise to an abnormally low ratio offree phCG to ahCG. One possible explanation for our observation is the abnormal placental development often observed in trisomy 18.This phenomenon has also been proposed for the abnormal levels of a-fetoprotein and human chorionic gonadotropin found in Down syndrome pregnancies [Wald et al., 19881. Human chorionic gonadotropin and its subunits are synthesized by the trophoblastic tissues of the placenta [Hussa, 19811. It has been previously demonstrated that serum levels of hCG free subunits were abnormal in gestational trophoblastic diseases [Ozturk et al., 1987, 198813; Khazaeli et al., 1986; Hay, 19881. However, diseases such as hydatidiform mole and choriocarcinoma are associated with increased levels of free PhCG and therefore an increased ratio of free phCG to ahCG has been observed [Ozturk et al., 1987, 1988bl. As reported by Cole et a1 [1988]low free phCG serum levels appear to be associated with increased risk of abortion in pregnan-
Free ahCG (nglml)
Free PhCG (nglml)
PhCGIahCG ratio (MOM)
19 28 62 64 85 34 57 49
37 17 62 55 29 37 43 8
1.95 (0.46) 0.61 (0.14) 1.00 (0.24) 0.86 (0.20) 0.34 (0.13) 1.09 (0.22) 0.75 (0.28) 0.16 (0.06)
14 11 17 17 33 86 43 67 35 5 34 14 28 8 4 6 9 47 20 23 30 15 13 25 17 10 39 16 31 42
114 161 120 88 232 46 94 44 221 41 23 130 82 34 55 55 77 164 62 67 46 17 144 27 19 61 11 112 53 129
8.14 (2.99) 14.64 (5.38) 7.06 (1.68) 5.18 (1.23) 7.03 (2.58) 0.53 (0.32) 2.19 (0.83) 0.66 (0.25) 6.31 (2.40) 8.20 (3.01) 0.68 (0.26) 9.29 (3.42) 2.93 (0.70) 4.25 (1.56) 13.75 (5.06) 9.17 (3.37) 8.56 (5.19) 3.49 (1.28) 3.10 (1.14) 2.91 (1.07) 1.53 (0.56) 1.13 (0.41) 11.08 (4.07) 1.08 (0.40) 1.12 (0.41) 6.10 (2.24) 0.28 (0.10) 7.00 (2.57) 1.71 (0.63) 3.07 (1.13)
cies produced by in vitro fertilization. It is also well established that about 50% of spontaneous abortions in early pregnancy are associated with chromosome abnormalities [Hassold and Jacobs, 1984; Bou6 et al., 19851. Thus, we conclude that some chromosome abnormalities are associated with abnormal free subunit hCG serum levels. This phenomenon is probably due in part to abnormal development of fetal trophoblastic tissue. As shown in the present investigation, it appears to be much more common in trisomy 18 than trisomy 21 and other chromosome abnormalities. Although the histopathology of molar pregnancies has been studied extensively [Szulman and Surti, 19781, little is known regarding specificmorphological changes in placentas derived from a mother carrying a trisomy fetus. It will be of interest to establish if trisomy 18 is associated with specifichistopathological changes in the trophoblastic tissue. Such studies should also lead to a better understanding of the mechanisms of human chorionic gonadotropin regulation during normal and abnormal gestation.
Chorionic Gonadotropin in Trisomy 18
-
-
Median
\
4
7
8
9
10
11
12
13
4
WEEKS A F T€R L MP Fig. 1. The ratio offree piahCG in maternal serum of mothers with a trisomy 18 fetus (n = 81, as compared to values obtained from normal pregnancies (n = 666) matched for age.
ACKNOWLEDGMENTS The authors gratefully acknowledge the assistance of Mrs. Kristin E. Cambria-Shaw in preparation of this manuscript. This work was supported in part by grants CA-35711, AA-02666, and HD-20469 from the National Institutes of Health and Association pour la Recherche sur le Cancer, Villejuif, France. J.R.W. is the recipient of a Research Career Scientist Development Award, AA-00048. REFERENCES Bidart JM, Troalen F, Bousfield GR, Birken S, Bellet DH (1988): Antigenic determinants on human choriogonadotropin a-subunit.J Biol Chem 263:10364-10369.
483
Bidart JM, Troalen F, Salesse R, Bousfield GR, Bohuon CJ, Bellet DH (1987):Topographic antigenic determinantsrecognized by monoclonal antibodies on human choriogonadotropin psubunit. J Biol Chem 262:8551-8556. Bogart MH, Pandan MR, Jones DW (1987): Abnormal maternal serum chorionic gonadotropin levels in pregnancies with fetal chromosome abnormalities. Prenat Diagn 7523-630. Bou6 A, Bou6 J, Gropp A (1985): Cytogenetics of pregnancy wastage. Adv Hum Gen 14:l-57. Brambati B, Oldrini EF, Lonzani A (1987): Chorionic villus sampling: An analysis of the obstetric experience of 1000 cases. Prenat Diagn 7:157-169. Cole LA, Restrepo-Candello H, Lavy G, Decherney A (1988): hCG free p-subunit, a n early marker of outcome of in uztro fertilization clinical pregnancies. J Clin Endocrinol Metab 58:1328-1329. Cole LC, Kroll TD, Ruddon RW, Hussa Ro (19841: Differential occurrence of free beta and free alpha subunits of human chorionic gonadotropin (hCG) in pregnancy sera. J Clin Endocrinol Metab 58:1200-1202. Hassold TJ, Jacobs PA (1984): Trisomy in man. Annu Rev Genet 18:69-97. Hay DL (1988): Histological origins of discordant chorionic gonadotropin secretion in malignancy. J Clin Endocrinol Metab 66:557-564. Hussa RO (1981): Human chorionic gonadotropin-a clinical marker: Review of its biosynthesis. Ligand Rev 3:l-43. Jacobs PA, Wilson CM, Sprenkle JA, Rosenshein NB, Migeon BR (1982): Mechanism of origin of complete hydatidiform moles. Nature 286:714-716. Khazaeli MB, Hedayat MM, Hatch KD, Soong SJ, Shingleton HM, Boots LR, Lubuglio AF (1986): Radioimmunoassay of free betasubunit of human chorionic gonadotropin as a prognostic test for persistent trophoblastic disease in molar pregnancy. Am J Obstet Gynecol 26: 208-2 12. Milunsky A, Wands JR, Brambati B, Bonacchi I, Currie K (1988):Firsttrimester maternal serum a-fetoprotein screening for chromosome defects. Am J Obstet Gynecol 159:1209-1213. Ozturk M, Bellet D, Manil L, Hennen G, Frydman R, Wands J (1987): Physiologic studies of hCG, alpha hCG and beta hCG as measured by specific monoclonal immunoradiometric assays. Endocrinology 120549-558. Ozturk M, Berkowitz R, Goldstein D, Bellet D, Wands J R (1988b): Differential production of human chorionic gonadotropin and free subunits in gestational trophoblastic disease. Am J Obstet Gynecol 158:193- 198. Ozturk M, Brown N, Milunsky A, Wands J (1988a): Physiologcal studies of human chorionic gonadotropin and free subunits in the amniotic fluid compartment compared to those in maternal serum. J Clin Endocrinol Metab 67:1117-1119. Szulman AE, Surti U (1978): The syndromes of hydatidiform mole I. Cytogenetic and morphologic correlations. Am J Obstet Gynecol 131:665-671. Wald NJ, Cuckle HS, Densem JW, Nanchahal K, Royston P, Chard T, Haddow JE, Knight GJ, Palomake GE, Canick JA (1988): Maternal serum screening for Down’s syndrome in early pregnancy Br Med J 297:883-885.
Abnormal Maternal Serum Levels of Human Chorionic Gonadotropin Free Subunits in Trisomy 18 Mehmet Ozturk, Aubrey Milunsky, Bruno Brambati, E.S. Sachs, Susan L. Miller, and Jack R. Wands Molecular Hepatology Laboratory, MGH Cancer Center, Massachusetts General Hospital, Charleston, and Department of Medicine, Harvard Medical School (M.O., J.R.W.), and Center for Human Genetics, Boston University School of Medicine (A.M., S.L.M.), Boston; First Institute of Obstetrics and Gynecology, University of Milan, Italy (B.B.); Department of Clinical Genetics, Erasmus University, Rotterdam, The Netherlands (E.S.S.)
We have measured maternal serum levels of free Q and p subunits of human chdionic gonadotropin between 8 and 12 weeks of gestation in 704 women at increased risk for trisomy. This group was studied because of advanced maternal age or a previous birth with chromosomal abnormality. All sera had been collected prior to chorion villus biopsy for prenatal diagnosis. Serum levels of free Q and phCG were determined by specific monoclonal antibody-based immunoradiometric assays. Analysis of chorionic tissue showed that in 38 of 704 (5.4%) pregnancies the fetus had a chromosome abnormality.There were 8 fetuses with trisomy 18(1.1%)and 9 (1.3%)with trisomy 21. In all pregnancies carrying a trisomy 18 fetus, we observed either high levels of free ahCG or low levels of free phCG or both. More importantly, the calculated ratio of free PhCGhhCG was less than 0.25 multiples of the median (MOM) in 6 of 8 (75%) trisomy 18 cases. Only 21 of 666 mothers (3.2%) carrying a normal fetus had a ratio less than 0.25 MOM (P < 0.0001). There was no difference between this ratio in trisomy 21 and normal pregnancy. Thus, when adjusted for gestational age, a low free phCG/ahCG ratio in maternal serum indicates a pregnancy at high risk [RR = 72 (95%CI 32,16211 for trisomy 18.
INTRODU~CTION Highly sensitive and specific monoclonal immunoradiometric assays have been developed for measurement of free a and p subunits of human chorionic gonadotropin (ahCG and phCG) in serum [Ozturk et al., 19871. Our previous investigations have demonstrated that serum levels of free a and p hCG were closely regulated throughout normal pregnancy. For example, we have observed that free phCG levels were higher than ahCG between 5 and 8 weeks of pregnancy. Similar concentrations of both free subunits were found between 9 and 12 weeks of gestation followed by a reversal such that higher levels of free ahCG were measured at term [Cole et al., 1984; Ozturk e t al., 1987, 1988al. We [Ozturk et al., 1987, 1988a], and others [Cole et al., 19841 suggested that the progressive changes in the relative amount of free subunits in serum were due in part to the continuous differentiation of trophoblastic cells. In support of this hypothesis was the finding that higher than normal levels of free phCG [Khazaeli et al., 1986; Ozturk et al., 1987,1988b;Hay, 19881were noted to accompany low levels of free ahCG in serum of women with gestational trophoblastic diseases such as hydatidiform mole and choriocarcinoma [Ozturk et al., 1987, 1988133. Extensive histopathologic and cytogenetic changes have suggested that trophoblastic differentiation is abnormal i n molar placenta and choriocarcinoma tissues; such changes have been accompanied by specific chromosome abnormalities [Szulman and Surti, 1978; KEY WORDS: free aihCG, free PhCG, chroJacobs e t al., 19821. Indeed, abnormal development of mosomal abnormality the placenta is one of the most common pathologic changes in the autosomal trisomies, but little is known regarding specific morphological changes [Hassold and Jacobs, 1984; Bou6 et al., 19851.Bogart et al. [19871has recently described elevated levels of maternal serum hCG in fetal Down syndrome. Based on these observations, we have measured free hCG subunit levels in the sera of mothers bearing fetuses with chromosome abnormalities and compared our results to those for women of Ekceived €or publication August 22,1989; revision received Dea similar age with a normal pregnancy outcome. To our cember 1, 1989. Address reprint requests to Mehmet Ozturk, PhD, Molecular surprise, we found a strikingly low phCG/ahCG Hepatology Laboratory, MGH Cancer Center, MGH East, 7th (p/ahCG) ratio in trisomy 18 but not in trisomy 21 or other chromosome abnormalities. Floor, 149 13th Street, Charlestown, MA 02129.
0 1990 Wiley-Liss, Inc.
Chorionic Gonadotropin in Trisomy 18 METHODS Patients Pregnancies at increased risk for chromosomal abnormalities were studied between 8 and 12 weeks of gestation. Two groups were studied: 1)The largest population was prospectively evaluated in Milan, Italy, and consisted of 491 women planning to undergo chorion villus sampling (CVS) because of advanced maternal age or previous birth of a child with a chromosome or other development defect. Fetal age was established precisely by ultrasound examination and serum samples were obtained between 8 and 12 weeks of gestation and shortly before CVS. Sera were stored at - 20°C prior to analysis. A detailed clinical description of this population was previously reported [Brambati et al., 1987; Milunsky et al., 19881.2)The second patient population, from Rotterdam, was also investigated prospectively. Two hundred and thirteen women were studied at 9-11 weeks of gestation. Similar to the first study, fetal age was established by ultrasonography and maternal sera were obtained prior to CVS biopsy. All measurements of free ahCG and phCG subunit serum levels were performed under code without knowledge of the clinical outcome and cytogenetic results.
481
free a and phCG was determined from a standard curve and statistical analysis was performed as described previously [Ozturk et al., 1987; Milunsky et al., 19881.
RESULTS The Milan population comprised 491 women of whom 25 (5.1%)had CVS biopsies diagnostic of a fetus with an abnormal karyotype. Chromosome analysis of CVS tissue from the Rotterdam population showed that 13out of 213 (6.1%)carried an aneuploid fetus. Of the 666 combined normal pregnancies, the median maternal serum free ahCG levels were found to be 17.5 ng/ml a t 8 weeks, 25.5 ng/ml at 10 weeks, and 32 ng/ml at 12 weeks. Median free phCG levels were found to be 79.5 ng/ml at 8 weeks, 65.5 ng/ml a t 10 weeks, and 60.5 ng/ml at 12 weeks respectively (Table I). Table I1 depicts the absolute maternal serum free subunit concentrations in 38 pregnancies associated with fetal aneuploidy. The plahCG ratios are calculated and multiples of the median (MOM)values are determined by comparison to normal pregnancies adjusted to the same gestational age. As shown in Figure 1, this ratio in normal pregnancy was 4.9 at 8 weeks and progressively decreased with advancing gestational age to a level of 1.65 at 12 weeks. Figure 1 compares the p/ahCG ratios in trisomy 18 to those measured in normal pregnancies. The MOMin trisomy 18 was less than 0.5 in all cases. More importantly, the free p/ahCG ratio, which was less than 0.25 in only 21 out of 666 normal pregnancies (3.2%),was less than 0.25 MOMin 6 out of 8 (75.0%)trisomy 18. This ratio was found to be significantly lower than the population carrying a normal fetus matched for the same gestational age (P < 0.0001). Thus, when adjusted for gestational age, a low p/ahCG ratio in maternal serum indicates a pregnancy at high risk [RR = 72 (95% CI 32, 162)l for trisomy 18. No consistent changes in a or p free subunit ratios were observed in trisomy 21 or other chromosome abnormalities. This indicates a high degree of specificity of our findings for trisomy 18 (Table 11).Indeed, this low ratio appeared associated with trisomy 18 since only 2 of 30 pregnancies (6.7%) with other chromosome abnormalities had such a low ratio and when considered as a group did not significantly differ from the ratio found in normal pregnancy.
Immunoradiometric Assays The specific and sensitive immunoradiometric assays used in this study for the detection of free ahCG and phCG were described previously [Ozturk et al., 19871. The general assay design employswell-definedmonoclonal antibodies that recognize specific epitopes on the free subunits. By incubation with antibody-coated beads, the free subunit is first captured on the solid phase support. Following a washing step, the beads are incubated with a second '251-labeled antibody which binds to the captured a and phCG. The number of counts bound per minute is directly proportional to the amount of subunits present in serum. The free ahCG assay uses a free a subunit-specific capture antibody (AHT2O) on the solid phase support [Ozturk et al., 1987;Bidart et al., 19881. The radiolabeled indicator antibody (HT13) has been shown to react with a distinct but different epitope on the free a subunit. The free ahCG assay has a lower limit of sensitivity of 0.20 ng/ml and does not detect native hCG or any other glycoproteinhormones [Ozturk DISCUSSION et al., 19871. Similarly, the free phCG assay uses a free The aim of this study was to determine whether maPhCG-specific antibody (FBT11)on the solid phase support [Ozturk et al., 1987; Bidart et al., 19871.The radio- ternal first trimester serum levels of hCG free subunits labeled indicator antibody (FBT10) binds to a separate are abnormal in women carrying a fetus with a chromoepitope on the phCG molecule. This assay detects free phCG a t a lower level of 0.20 nglml and does not react with intact hCG [Ozturk et al., 19873. Detailed studies of TABLE I. Medians of hCG Free Subunit Serum the assay design, antibody specificity, and assay perforConcentrations in Normal Pregnancies mance (intra-assay and inter-assay variations) have been previously published [Ozturk et al., l987,1988a,bI. Gestational age Free ahCG Free phCG Data Analysis The ratio of free p/ahCG was calculated by dividing phCG by ahCG concentrations expressed as nglml following simultaneous measurements of free subunit levels on the same serum sample. The absolute level of
(weeks)
n
(ngiml)
(nglml)
8 9 10 11 12
14 197 360 75 20
17.5 21.0 25.5 23.0 32.0
79.5 75.0 65.5 58.0 60.5
482
Ozturk et al. TABLE 11. Serum Levels of hCG Free Subunits With Chromosome Abnormalities
LMP Fetal karyotype (weeks) Trisomv 18 47,XY, + 18 9 47,XY, + 18 9 47,XY. + 18 9 47;xy; + 18 9 47.xx. + 18 10 47;XXj + i 8 8 47,xx, + 18 10 47,XX, + 18 10 Trisomy 21 and other chromosome defects 47,XY, + 21 10 47,xx, + 21 10 47,XY, + 21 9 47,xx, + 21 9 47,XY, + 21 10 47,xx, + 21 12 47,XY, + 21 11 47,XY, + 21 11 11 47,XY,+21 10 47,XX, + 13 47;XY; + 13 11 47,xx, + 22 10 47,XXY 9 48;XX, + 18,+ 11 10 47,XX,+21/48,XX,+ 11,+21 10 46,XXde113,t(13;y) 10 8 46,X,inv(Y),- 14,t(13;14) 45,X/46,XX 10 45,X/46,XX 10 46;XYl45,X 10 46,XX,t(1;2-8) 10 47.XY. + m 10 46;XY; + t(4;8) 10 10 46.XXI47.XX. + 7 10 92:xxy Y’ 10 47,XY + 22 46,XY,t(6;11) 10 10 46,XW45x 46,XY,t(5;12) 10 46,XY,lqh+ 10
some defect. Our study indicates that serum levels of free ahCG were high, whereas those of free phCG were low in trisomy 18,thus giving rise to an abnormally low ratio offree phCG to ahCG. One possible explanation for our observation is the abnormal placental development often observed in trisomy 18.This phenomenon has also been proposed for the abnormal levels of a-fetoprotein and human chorionic gonadotropin found in Down syndrome pregnancies [Wald et al., 19881. Human chorionic gonadotropin and its subunits are synthesized by the trophoblastic tissues of the placenta [Hussa, 19811. It has been previously demonstrated that serum levels of hCG free subunits were abnormal in gestational trophoblastic diseases [Ozturk et al., 1987, 198813; Khazaeli et al., 1986; Hay, 19881. However, diseases such as hydatidiform mole and choriocarcinoma are associated with increased levels of free PhCG and therefore an increased ratio of free phCG to ahCG has been observed [Ozturk et al., 1987, 1988bl. As reported by Cole et a1 [1988]low free phCG serum levels appear to be associated with increased risk of abortion in pregnan-
Free ahCG (nglml)
Free PhCG (nglml)
PhCGIahCG ratio (MOM)
19 28 62 64 85 34 57 49
37 17 62 55 29 37 43 8
1.95 (0.46) 0.61 (0.14) 1.00 (0.24) 0.86 (0.20) 0.34 (0.13) 1.09 (0.22) 0.75 (0.28) 0.16 (0.06)
14 11 17 17 33 86 43 67 35 5 34 14 28 8 4 6 9 47 20 23 30 15 13 25 17 10 39 16 31 42
114 161 120 88 232 46 94 44 221 41 23 130 82 34 55 55 77 164 62 67 46 17 144 27 19 61 11 112 53 129
8.14 (2.99) 14.64 (5.38) 7.06 (1.68) 5.18 (1.23) 7.03 (2.58) 0.53 (0.32) 2.19 (0.83) 0.66 (0.25) 6.31 (2.40) 8.20 (3.01) 0.68 (0.26) 9.29 (3.42) 2.93 (0.70) 4.25 (1.56) 13.75 (5.06) 9.17 (3.37) 8.56 (5.19) 3.49 (1.28) 3.10 (1.14) 2.91 (1.07) 1.53 (0.56) 1.13 (0.41) 11.08 (4.07) 1.08 (0.40) 1.12 (0.41) 6.10 (2.24) 0.28 (0.10) 7.00 (2.57) 1.71 (0.63) 3.07 (1.13)
cies produced by in vitro fertilization. It is also well established that about 50% of spontaneous abortions in early pregnancy are associated with chromosome abnormalities [Hassold and Jacobs, 1984; Bou6 et al., 19851. Thus, we conclude that some chromosome abnormalities are associated with abnormal free subunit hCG serum levels. This phenomenon is probably due in part to abnormal development of fetal trophoblastic tissue. As shown in the present investigation, it appears to be much more common in trisomy 18 than trisomy 21 and other chromosome abnormalities. Although the histopathology of molar pregnancies has been studied extensively [Szulman and Surti, 19781, little is known regarding specificmorphological changes in placentas derived from a mother carrying a trisomy fetus. It will be of interest to establish if trisomy 18 is associated with specifichistopathological changes in the trophoblastic tissue. Such studies should also lead to a better understanding of the mechanisms of human chorionic gonadotropin regulation during normal and abnormal gestation.
Chorionic Gonadotropin in Trisomy 18
-
-
Median
\
4
7
8
9
10
11
12
13
4
WEEKS A F T€R L MP Fig. 1. The ratio offree piahCG in maternal serum of mothers with a trisomy 18 fetus (n = 81, as compared to values obtained from normal pregnancies (n = 666) matched for age.
ACKNOWLEDGMENTS The authors gratefully acknowledge the assistance of Mrs. Kristin E. Cambria-Shaw in preparation of this manuscript. This work was supported in part by grants CA-35711, AA-02666, and HD-20469 from the National Institutes of Health and Association pour la Recherche sur le Cancer, Villejuif, France. J.R.W. is the recipient of a Research Career Scientist Development Award, AA-00048. REFERENCES Bidart JM, Troalen F, Bousfield GR, Birken S, Bellet DH (1988): Antigenic determinants on human choriogonadotropin a-subunit.J Biol Chem 263:10364-10369.
483
Bidart JM, Troalen F, Salesse R, Bousfield GR, Bohuon CJ, Bellet DH (1987):Topographic antigenic determinantsrecognized by monoclonal antibodies on human choriogonadotropin psubunit. J Biol Chem 262:8551-8556. Bogart MH, Pandan MR, Jones DW (1987): Abnormal maternal serum chorionic gonadotropin levels in pregnancies with fetal chromosome abnormalities. Prenat Diagn 7523-630. Bou6 A, Bou6 J, Gropp A (1985): Cytogenetics of pregnancy wastage. Adv Hum Gen 14:l-57. Brambati B, Oldrini EF, Lonzani A (1987): Chorionic villus sampling: An analysis of the obstetric experience of 1000 cases. Prenat Diagn 7:157-169. Cole LA, Restrepo-Candello H, Lavy G, Decherney A (1988): hCG free p-subunit, a n early marker of outcome of in uztro fertilization clinical pregnancies. J Clin Endocrinol Metab 58:1328-1329. Cole LC, Kroll TD, Ruddon RW, Hussa Ro (19841: Differential occurrence of free beta and free alpha subunits of human chorionic gonadotropin (hCG) in pregnancy sera. J Clin Endocrinol Metab 58:1200-1202. Hassold TJ, Jacobs PA (1984): Trisomy in man. Annu Rev Genet 18:69-97. Hay DL (1988): Histological origins of discordant chorionic gonadotropin secretion in malignancy. J Clin Endocrinol Metab 66:557-564. Hussa RO (1981): Human chorionic gonadotropin-a clinical marker: Review of its biosynthesis. Ligand Rev 3:l-43. Jacobs PA, Wilson CM, Sprenkle JA, Rosenshein NB, Migeon BR (1982): Mechanism of origin of complete hydatidiform moles. Nature 286:714-716. Khazaeli MB, Hedayat MM, Hatch KD, Soong SJ, Shingleton HM, Boots LR, Lubuglio AF (1986): Radioimmunoassay of free betasubunit of human chorionic gonadotropin as a prognostic test for persistent trophoblastic disease in molar pregnancy. Am J Obstet Gynecol 26: 208-2 12. Milunsky A, Wands JR, Brambati B, Bonacchi I, Currie K (1988):Firsttrimester maternal serum a-fetoprotein screening for chromosome defects. Am J Obstet Gynecol 159:1209-1213. Ozturk M, Bellet D, Manil L, Hennen G, Frydman R, Wands J (1987): Physiologic studies of hCG, alpha hCG and beta hCG as measured by specific monoclonal immunoradiometric assays. Endocrinology 120549-558. Ozturk M, Berkowitz R, Goldstein D, Bellet D, Wands J R (1988b): Differential production of human chorionic gonadotropin and free subunits in gestational trophoblastic disease. Am J Obstet Gynecol 158:193- 198. Ozturk M, Brown N, Milunsky A, Wands J (1988a): Physiologcal studies of human chorionic gonadotropin and free subunits in the amniotic fluid compartment compared to those in maternal serum. J Clin Endocrinol Metab 67:1117-1119. Szulman AE, Surti U (1978): The syndromes of hydatidiform mole I. Cytogenetic and morphologic correlations. Am J Obstet Gynecol 131:665-671. Wald NJ, Cuckle HS, Densem JW, Nanchahal K, Royston P, Chard T, Haddow JE, Knight GJ, Palomake GE, Canick JA (1988): Maternal serum screening for Down’s syndrome in early pregnancy Br Med J 297:883-885.
