1611120
© 1990 S. Karger AG, Basel 0379-8305/90/0154-0200$2.75/0
Dev Pharmacol Ther 1990;15:200-210
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome (Steroid 21-Hydroxylase Deficiency) Maria I. New Department of Pediatrics, New York Hospital, New York, N.Y., USA
Key Words. Congenital adrenal hyperplasia ■ 17-Hydroxyprogesterone • HLA genotyping • Chorionic villus sampling • Dexamethasone administration Abstract. Prenatal treatment of pregnancies at risk for congenital adrenal hyperplasia due to 21-hydroxylase deficiency was carried out in conjunction with chorionic villus sampling (CVS) in the first trimester for analysis of restriction fragment length polymorphisms. Dexa methasone administration to the pregnant woman was initiated at a mean gestational age of 7 weeks (range 4-10 weeks) before testing to determine whether the fetus was affected with 21-hydroxylase deficiency, and CVS was performed at a gestational age of 8-10 weeks. Two affected female fetuses were identified by molecular genetic techniques among this group. The duration of unnecessary prenatal dexamethasone treatment for unaffected or male fetuses was substantially reduced in the CVS group compared with a cohort of 8 prenatally treated pregnancies where amniocentesis was performed in the early second trimester. No major morbidities were observed in the treated pregnancies. Postnatal confirmation of CVS diagnosis was obtained in all cases where DNA from an affected sibling was available for comparative analysis with the DNA from chorionic villus tissue. The external genitalia of the affected females who were treated prenatally appeared normal. Based on these data we con clude that the benefit:risk ratio is favorable for prenatal administration of dexamethasone in pregnancies at risk for 21-hydroxylase deficiency. Treatment should be initiated during the first trimester in conjunction with diagnosis by CVS/molecular genetic techniques. Long term postnatal surveillance is recommended for all offspring of dexamethasone-treated preg nancies.
may be preventable by prenatal administration of dexamethasone to the mother, which suppresses the abnormal secretion of androgens by the fetal adrenal gland. Results from follow-up and survey studies suggest that it is Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
Females with the severe, classic form of 21-hydroxylase deficiency are born with ambiguous genitalia and often require multiple reconstructive surgical procedures to gain normal reproductive function. This problem
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
Background: Diagnosis and Management It had long been known that excretion in the urine of the steroid metabolite pregnane triol was the salient biochemical feature of the adrenogenital syndrome [1], classical congenital adrenal hyperplasia (CAH). The underlying abnormality is a deficiency of the enzyme steroid 21-hydroxylase, and this is the enzyme defect exhibited by the greatest number of CAH patients. (The other adrenal enzyme defects, each with characteristic ab normalities of sex differentiation [2], have since been exhaustively described [3, 4] but together they comprise only the remaining 5% of cases. Classical CAH owing to 21hydroxylase deficiency is the most common basis of female pseudohermaphroditism.
In virilizing CAH, affected females are born with ambiguous external genitalia. They are at risk for wrong sex assignment and missed diagnosis. Both sexes undergo precocious advance in growth and progres sive virilization. Three out of four cases also exhibit salt wasting and fluid loss, often de veloping acute adrenal insufficiency. Before the era of hormonal replacement, many of these patients would die from such episodes. Steroid replacement was shown first in 1950 to correct the endocrine imbalance and pro vide essential hormonal function [5, 6]. Problems of gender confusion in affected females may develop from improper sex as signment or uncertain sex assignment and rearing [7-10], Prenatal androgen effects on gender remain controversial.
Prenatal Diagnosis In 1965, Jeffcoate et al. [11] successfully diagnosed classical 21-hydroxylase defi ciency in a fetus on the basis of raised amniotic fluid concentrations of 17-ketosteroids and pregnanetriol near term compared with normal values. By 1976, prenatal diag nosis of CAH by hormonal assay of amniotic fluid, maternal urine or both had been at tempted by a number of groups [reviewed in ref. 12], Linkage between 21-hydroxylase and HLA was reported by Dupont et al. [13] and Levine et al. [14], Since then, HLA geno typing has been used in family studies as a genetic marker for the enzyme defect in family members. Hormonal Measurement. The most spe cific serum hormonal determinant for 21hydroxylase deficiency is elevated 17ahydroxyprogesterone (17-OHP). Amniotic fluid 17-OHP is always high when the fetus Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
important to begin therapy no later than the 7th week of gestation, and that the optimal dexamethasone dose is 20 pg/kg/day divided in two or three doses. In only 1 of 4 pregnancies in question will the fetus be affected, and one half of these will be males who do not need prenatal treat ment for prevention of virilization. Thus, only one eighth of fetuses require treatment. Prenatal diagnosis of 21-hydroxylase defi ciency, utilizing the new molecular probes for human leukocyte antigens (HLA) and CYP21 probes combined with hormonal studies and serological HLA genotyping, is now possible early in gestation with tissue obtained by chorionic villus sampling (CVS). This is of great advantage in the prenatal management of potential cases in pregnancies of families at risk because the earlier diagnosis afforded by CVS permits sooner termination of ste roid therapy, unnecessary when the fetus is a genetic male or is unaffected.
201
is affected with salt-wasting 21-hydroxylase deficiency; if the disease takes the simple vir ilizing form, 17-OHP levels may be only ambiguously elevated in the amniotic fluid [15-19]. The amniotic fluid A4-androstenedione concentration may be assayed as an adjunct to 17-OHP measurement [18], Am niotic fluid testosterone levels have also been measured but may be within the nor mal male range in the case of an affected male [18, 20], An important factor in prena tal hormonal evaluation is that normal am niotic fluid values for testosterone, proges terone and 17-OHP differ according to fetal sex, and may change significantly with gesta tional age [21], HLA Typing. The gene loci for the ste roid 21-hydroxylase enzyme (a microsomal membrane-bound cytochrome P450, cyt P450c21) are within the close linkage group HLA (once HL-A for antigen-A), the human major histocompatibility complex, or Mhc. Elements of HLA exhibit a high degree of polymorphism, making them of particular utility as markers. Transmission of 21-hy droxylase deficiency genes can thus be fol lowed in a pedigree through the cosegregat ing HLA antigens, when these are known. HLA genotyping is ideally done and com plete pedigree taken for any family manifest ing 21-hydroxylase deficiency and wishing more offspring. HLA types are needed at minimum for the parents (who should test as carriers on adrenocorticotropic hormone (ACTH) testing [22] - unless themselves af fected [23], and one affected child. The fetal HLA genotype is determined from standard serotyping of a fetal cell cul ture or molecular probe analysis of fetal DNA, which may also derive from cell cul ture or can be amplified directly from fetal cell DNA by the polymerase chain reaction
New
technique (PCR). Fetal cell cultures are grown up from viable fetal cells obtained by one of two invasive procedures: free cells, mostly fibroblasts, from a quantity of am niotic fluid drawn on amniocentesis, or more recently, villus cells from the chorion frondosum obtained by needle aspiration in transabdominal or transcervical CVS. Prenatal diagnosis of 21-hydroxylase de ficiency makes use of HLA associations [2426]. The fetal HLA type is compared with that of the family index case (an affected liv ing sib) and accordingly, as it is HLA identi cal or shares one or neither haplotype, the fetus is predicted to be affected, to be a car rier, or to be unaffected and not a carrier. Confounding factors in presumptive diagno sis of 21 -hydroxylase deficiency by HLA typ ing are (1) intra-HLA recombination events, (2) homozygosity at the HLA-B locus in either parent and (3) antigen sharing be tween the parents. When HLA typing is by serological techniques, typing of the HLA antigens called class II (HLA-DR, DQ and DP) requires a more specialized assay [27], Unless the HLA-DR antigen is typed, recom bination between the HLA-B and DR loci, which may include the 21-hydroxylase locus, segregating it away from its primary marker HLA-B, will go unidentified. If HLA typing is by DNA analysis, molecular probes used include many for class II, whereas uncer tainty arising from homozygosity or antigen sharing requires the use of probes for serum complement isotypes C4A and C4B, adja cent to the 21-hydroxylase locus, or probes for the 21-hydroxylase gene itself, CYP21, and pseudogene CYP21P. The CYP21 probes are not used routinely because there are as yet few mutations known to affect restriction enzyme sites within these genes. Instead, HLA class I and class II probes are Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
202
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
203
The synthetic steroid dexamethasone (9afluoro- 16a-methyl -1 lß, 17a,21 -trihydroxypregna-l,4-diene-3,20-dione), a substituted A1- or 1,2-dehydrocortisol, has potent gluco corticoid action and a long plasma half-life. It is not bound by corticosteroid-binding globulin (CBG) and crosses the placenta without inactivation. Theoretically, dexamethasone adminis tered to the mother from the onset of fetal adrenal activity at 6-7 weeks of gestation will suppress the fetal adrenals, reducing DNA Analysis/CYS androgen production in an affected female CVS, normally done at 8-11 weeks of ges and preventing labial fusion, formation of a tation, permits evaluation of the fetus at risk urogenital sinus, and clitoromegaly. Where in the first trimester. Fetal DNA is geno- as an earlier study found the fetal pituitarytyped using molecular probes (HLA, C4 and adrenal axis not to suppress after acute ad 21-hydroxylase), and HLA by serology may ministration of dexamethasone at mid-term still be done for confirmation or to direct the [31], prenatal treatment of fetuses affected choice of informative DNA probes. Karyo with or at risk for 21-hydroxylase defi typing or Y-chromosome probes are used to ciency has recently been employed with sex the fetus. Amniotic fluid in the aspirate success [reviewed in ref. 32]. In 1989, sum has been assayed experimentally for eleva marizing treatment and outcome for 43 tion of 17-OHP levels. Viable cells are plated pregnancies at risk, Forest et al. [26] re and cultures grown up for sufficient DNA to ported a decisive reduction in masculinizabe obtained for analysis (10 days to 2 weeks tion following treatment with dexametha time). PCR-amplification of fetal cell DNA sone in 6 affected female fetuses. Although directly from the CVS sample is very rapid treated cases are few, oral dexamethasone but may generate inauthentic sequences 0.5 mg twice daily before the 10th week brought about complete suppression of fetal [30], Amniocentesis has proven useful in pre adrenocortical hormones as measured at natal diagnosis of steroid 21-hydroxylase de amniocentesis. In a group reporting from ficiency, but CVS is preferred when treat Germany on 3 pregnancies at risk for CAH ment is involved. The advantage of CVS treated with dexamethasone [33, 34], the over amniocentesis is the shorter time to protocol included suspension of therapy for diagnosis and therefore sooner termination of 5 days immediately before amniocentesis. unnecessary steroid treatment. Masculiniza- High amniotic fluid 17-OHP levels were tion of external genital processes begins by measured, and it was postulated that they the 6-7th week of gestation. CVS in the 8- could have been caused by increased meta 10th week provides the earliest information bolic clearance of dexamethasone, or inade as to whether prenatal steroid treatment is quate dosage. The babies were born with necessary (fig. 1). ambiguous genitalia. Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
used [28], Panels of synthetic oligonucleo tide probes will screen for already known mutations. Amniotic fluid 17-OHP may in some cir cumstances add to the diagnostic evaluation when there is a question concerning the HLA types, and in others corroborate the HLA prediction, and hormonal measurement is still advocated [29].
204
New
Fig. 1. Relation between differentiation of gonads and anatomic differentiation of human male and female embryos.
Dexamethasone treatment may be insti tuted early in all pregnancies at risk and maintained until diagnosis by CVS and with
out ill effect for the fetus. The current recom mendation is oral dexaméthasone to the mother upon confirmation of pregnancy at a starting dose of 20 pg/kg/day (to a maximum of 1.5 mg) divided equally in 3 doses. To Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
Current Recommendations
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
treated with dexamethasone. Of these 8, there were 2 affected fetuses - 1 male and 1 female - who were born with normal genita lia. Three female fetuses who were affected and were not treated, had ambiguous genita lia. In the CVS group of 18 pregnancies (7 males and 11 females), two females were diagnosed to be affected with classical 21hydroxylase deficiency. These females were treated prenatally with dexamethasone and were born with no labial fusion. One had a normal clitoris and the other very mild clitoromegaly, which did not require clitoroplasty. This compared favorably with the sig nificant genital ambiguity of the untreated older affected sisters. Accurate prenatal diagnosis of 21 -hydrox ylase deficiency minimizes unnecessary treatment of males and unaffected females. A problem to be resolved is the risk to the fetus and mother of the possible procedures. Amniocentesis has a 0.5% mortality risk to the fetus. In experienced hands, CVS has no greater risk than amniocentesis. Dexametha sone has no known teratological effects.
Fetal Adrenal Function and Safety of Steroids The good result obtained in affected fe male infants born to mothers administered dexamethasone from early pregnancy to term without interruption gives empiric proof of the suppressibility of the fetal adre nal by this modality of treatment. Before this time, there has not been even a single in stance of a genetic female with 21-hydroxy lase deficiency CAH born with completely normal genitalia, and the virilization suf fered by the untreated developing fetus may extend to an external genital appearance inDownloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
date, no fetus of a mother treated with dexamethasone in low doses has been found to have any congenital malformations. A specific concern in connection with corticosteroid use and fetal development is cleft palate formation. This sequela, with placental degeneration and fetal death has been observed in rodents exposed in utero to high-dose glucocorticoids [35, 36]. An older survey of reports in the literature identified this malformation in 2 out of 260 exposures, in both cases the steroid given in large dose before the 14th week [37], and low-dose corticoids were felt to present little risk. Long term follow-up of prenatal treatment in France finds all children (affected and unaf fected fetuses) normal [26, 38] after 8.5 years. In some cases treatment was interrupted 5-7 days before amniocentesis, in order for the diagnosis to be confirmed on the basis of amniotic fluid hormonal values. Increased genital ambiguity in affected females has been attributed to the marked rise in fetal adrenal output at this time. Although it is still advocated in some protocols [39], the diagnostic advantage of this test is out weighed by the apparent increased risk of masculinization in an affected female, and we do not recommend it. In a recent study from our group, 49 fetuses were diagnosed by amniocentesis and CVS [32]. DNA analysis utilizing HLA ge netic probes as well as 21-hydroxylase ge netic probes were utilized on the DNA ex tracted from the CVS. An algorithm for diag nosis and treatment is shown in figure 2. Treatment was begun at 7 weeks of gestation before the diagnosis was known. Out of 49 fetuses studied, 27 proved to be males and 22 proved to be females. In the amniocente sis group of 31 women, 8 consented to be
205
206
New
3-4 weeks
(7-10 weeks) STOP
dexamethasone
STOP
2-S days later
1 -2 weeks later
dexamethasone (15-20 weeks}
2-5 days later
1 2 weeks later
Fig. 2. Procedure for management of pregnancies at risk for congenital adrenal hyperplasia due to 21-hydroxylase (21-OH) deficiency. hCG = Human chorionic gonadotropin.
The rise throughout gestation of the hormone estriol, which derives from fetal adrenal dehydroepiandrosterone-sulfate via fetal hepatic 16a-hydroxylation and placental aromatization, and thus expresses fetoplaDownloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
distinguishable from normal male. By now there are a number of successfully treated cases who at birth showed decisively reduced virilization or even to have normal female external genital appearance.
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
normal or CAH-affected fetal adrenal func tion. The possibility of zone-specific or tran sient sensitivity to other trophic factor(s) [49] notwithstanding, there is abundant evi dence in favor of ACTH as the primary adre nocortical control factor in the developing fetus [42, 50-53], Thus, although ACTH and cAMP-independent pathways for fetal adre nal stimulation may exist, it seems implausi ble either that (a) reduced virilization in infants born of treated pregnancies is inci dental, or that (b) dexamethasone acts to a significant degree through any mechanism other than inhibition of ACTH release. The safety of the gestational use of gluco corticoids has been reasonably well estab lished. Shepard’s Catalog of Teratogenic Agents [54] categorizes corticosteroids as ‘not highly dangerous’ in human pregnancy, with a relative risk for congenital malforma tions from prenatal exposure in human pregnancy of less than 1, i.e. not significant [55], In the experience of the author’s group and again in the large series reported from France [38], low-dose dexamethasone ther apy has no teratogenic potential. Neverthe less, it seems prudent to reduce the duration of unnecessary treatment, making CVS a more desirable diagnostic measure than am niocentesis. As experience with these proce dures in prenatal diagnosis and treatment increases, a better assessment of the risks and benefits of each will be obtained. A new case of prenatal treatment of a fetus with 21-hydroxylase deficiency has recently been reported. Dexamethasone was given from the third week of gestation until term in a dose of 0.5 mg every 12 h. The female in fant was born with a normal clitoris but mild posterior labial fusion. The authors recommend, based on this experience, that Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
cental steroidogenic activity [40], has been used as an index of fetal well-being (exagger ated rise may signal a multiple gestation). In France, Forest et al. [38] collected data on pregnancies at risk for 21-hydroxylase defi ciency CAH (but undiagnosed - thus repre senting both affected and unaffected fetuses) being managed on comparable prenatal treatment protocols (dexamethasone 1 or 1.5mg/day from 7-10 weeks to term). Es triol levels (measured as concentration in amniotic fluid or maternal serum, or amount secreted in 24-h urine) were significantly lower in treated than in age-matched normal fetuses. Adrenal development and regulation in utero is a complex issue [41, 42], The effect of ACTFI and dexamethasone on fetoplacen tal estrogen production following their ad ministration directly into the fetus or in the amniotic cavity has been measured [43], but to our knwledge the threshold for in vivo ACTH stimulation of the fetal adrenal has not been determined. It has been suggested that the fetal adre nal may be regulated by, or at least be responsive to other factors. Autopsy and as sessment of adrenal tissue in anencephaly have revealed partial evolution of steroido genic capacity in some cases [ 18, 44, 45], and some recent observations in vitro support the idea of non-ACTH-dependent or noncAMP-mediated control [46, 47], The em bryologie deficit in anencephalic fetuses, however, is not uniform or absolute and there may be remnants of endocrine feed back systems (in vivo data from at least one anencephalic has clearly shown measurable and stimulable ACTH levels [48]). Different cases, therefore, may not always be compara ble, nor must in vitro data from tissues of anencephalics be valid for understanding
207
New
the dose of dexamethasone be given three times daily [56], Infants treated prenatally from the French cohort (both unaffected and affected) have been followed carefully and normal growth and cognition up to a period of 8.5 years have been reported [38]. It is impor tant that the growth and development of infants exposed to dexamethasone prena tally should be followed closely for signs of unidentified side effects emerging in the long term, involving, for instance, changes in in tellectual or motoric function or somatic growth. To date, however, prenatal dexa methasone treatment given in proper dosage appears to be safe and effective prenatal treatment for CAH.
References 1 Butler, G.C.; Marrian, G.F.: The isolation of preg nane-3, 17,20-triol from the urine of women show ing the adrenogenital syndrome. J. biol. Chem. 119: 565-572 (1937). 2 Bongiovanni, A.M.; Eberlein, W.R.; Goldman, A.S.; New, M.I.: Disorders of adrenal steroid bio genesis. Recent Prog. Horm. Res. 23: 375-449 (1967). 3 New, M.I.; White, P.C.; Pang, S.; Speiser, P.W.: Dupont, B.: The congenital adrenal hyerplasias; in Scriver, The metabolic basis of inherited disease; 6th ed., pp. 1881-1917 (McGraw-Hill, New York 1989). 4 White, P.C.; New, M.I.; Dupont, B.: Congenital adrenal hyperplasia. New Engl. J. Med. 316: 1519-1524 and 1580-1586 (1987). 5 Wilkins, L.; Lewis, R.A.; Klein, R.; Rosemberg, E.: The suppression of androgen secretion by cor tisone in a case of congenital adrenal hyperplasia. Bull. Johns Hopkins Hosp. 56: 249-252 (1950). 6 Bartter, F.C.; Albright, F.; Forbes, A.P.; Leaf, A.; Dempsey, E.; Carroll, E.: The effects of adreno corticotropic hormone and cortisone in the adre nogenital syndrome associated with congenital adrenal hyperplasia. An attempt to explain and
7
8
9
10
11
12
13
14
15
16
correct its disordered hormonal pattern. J. clin. Invest. 30: 237-251 (1951). Klingensmith, G.J.; Garcia, S.C.; Jones, H.W., Jr.; Migeon, C.J.; Blizzard, R.M.: Glucocorticoid treatment of girls with congenital adrenal hyper plasia: effects on height, sexual maturation, and fertility. J. Pediatr. 90: 996-1004 (1977). Mulaikal, R.M.; Migeon, C.J.; Rock, J.A.: Fertil ity rates in female patients with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. New Engl. J. Med. 316: 178-182 (1987). Thibaud, E.; Rappaport, R.; Salomon Bernard, Y.; Paniel, B.J.; Dero, M.; Lepillier, P.; Philippe, F.; Pellerin, D.: Fertilité normale chez 2 femmes traitées pour hyperplasie congénitale des surréna les (FICS) avec syndrome de perte de sel (SPS) et ambiguïté génitale Prader III et IV (abstract); in Bertrand et al.: Sexual differentiation: basic and clinical aspects. Proc. Inti. Symp. Montpellier 1989. Federman, D.D.: Psychosexual adjustment in congenital adrenal hyperplasia (editorial). New Engl. J. Med. 316: 209 (1987). Jeffcoate, T.N.A.; Fliegner, J.R.H.; Russell, S.H.; Davis, J.C.; Wade, A.P.: Diagnosis of the adreno genital syndrome before birth. Lancet ii: 553-555 (1965). New, M.I.: Prenatal diagnosis of congenital adre nal hyperplasia; in New, Fiser, Diabetes and other endocrine disorders during pregnancy and in the newborn. Prog. clin. biol. Res., voi. 10, pp. 205219 (Alan R. Liss, New York 1976). Dupont, B.; Oberfield, S.E.; Smithwick, E.M.; Lee, T.D.; Levine, L.S.: Close genetic linkage be tween HLA and congenital adrenal hyperplasia (21-hydroxylase deficiency). Lancet ii: 1309-1311 (1977). Levine, L.S.; Zachmann, M.; New, M.I.; Prader, A.; Pollack, M.S.; O’Neill, G.J.; Yang, S.-Y.; Ober field, S.E.; Dupont, B.: Genetic mapping of the 21-hydroxylase deficiency gene within the HLA linkage group. New Engl. J. Med. 299: 911-912 (1978). Frasier, S.D.; Thorneycroft, I.H.; Weiss, B.A.; Horton, R.: Elevated amniotic fluid concentration of 17a-hydroxyprogesterone in congenital adrenal hyperplasia. J. Pediatr. 86: 310-312 (1975). Nagamani, M.; McDonough, P.G.; Ellegood, J.O.; Mahesh, V.B.: Maternal and amniotic fluid 17hydroxyprogesterone levels during pregnancy: Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
208
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
26 Forest, M.G.; Betuel, H.; Couillin, P.; Boue, A.: Prenatal diagnosis of congenital adrenal hyperpla sia (CAPI) due to 21-hydroxylase deficiency by steroid analysis in the amniotic fluid of mid-preg nancy: comparison with HLA typing in 17 preg nancies at risk for CAH. Prenat. Diagn. 1: 197— 207 (1981). 27 Maurer, D.H.; Pollack, M.S.: The use of gamma interferon to increase HLA antigen expression on cultured amniotic cells used for the prenatal diag nosis of 21-hydroxylase deficiency; in New, Con genital adrenal hyperplasia. Ann. N.Y. Acad. Sci., vol. 458, pp. 148-155 (1985). 28 Mornet, E.; Boue, J.; Raux-Demay, M.; Couillin, P.; Oury, J.E.; Dumez, Y.; Dausset, J.; Cohen, D.; Bouc, A.: First trimester prenatal diagnosis of 21hydroxylase deficiency by linkage analysis to HLA-DNA probes and by 17-hydroxyprogesterone determination. Hum. Genet. 73: 358-364 (1986). 29 Pang, S.; Pollack, M.S.; Loo, M.; Green, O.; Nussbaum, R.; Clayton, G. ; Dupont, B.; New, M.I.: Pitfalls of prenatal diagnosis of 21-hydroxylase deficiency congenital adrenal hyperplasia. J. clin. Endocr. Metab. 61: 89-97 (1985). 30 Ennis, P.D.; Zemmour, J.; Salter, R.D.; Parham, P.: Rapid cloning of HLA-A,B cDNA by using the polymerase chain reaction: frequency and nature of errors produced in amplification. Proc. natn. Acad. Sci. USA 87: 2833-2837 (1990). 31 Charnvises, S.; Fencl, M. de M.; Osathanondh, R.; Zhu, M.-G.; Underwood, R.; Tulchinsky, D.: Ad renal steroids in maternal and cord blood after dexamethasone administration in midterm. J. din. Endocr. Mctab. 61: 1220-1222 (1985). 32 Speiser, P.W.; Laforgia, N.; Kato, K.; Pareira, J.; Khan, R.: Yang, S.-Y.; Whorwood, C; White, P.C.; Elias, S; Schriock, E.; Schriock, E. [sic], Simpson, J.L.; Taslimi, M.; Najjar, J.; May, S.; Mills, G.; Crawford, C; New, M.I.: First trimester prenatal treatment and molecular genetic diagno sis of congenital adrenal hyperplasia (21-hydroxy lase deficiency). J. din. Endocr. Metab. 70: 838— 848 (1990). 33 Knorr, D.; Bidlingmaier, F.; Dörr, H.G.; Kuhnle, U.: Prenatal treatment of a girl with 21-hydroxylase deficiency (CAH) (abstract No. 117). Pro gram and Abstracts, 2nd Joint Mtg. Lawson Wil kins Pediatr. Endocr. Soc. and Eur. Soc. Pediatr. Endocr., Baltimore 1985. Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
diagnosis of congenital adrenal hyperplasia. Am. J. Obstet. Gynec. 130: 791-794 (1978). 17 Hughes, I.A.; Laurence, K.M.: Antenatal diagno sis of congenital adrenal hyperplasia. Lancet ii: 79 (1979). 18 Pang, S.; Levine, L.S.; Cederqvist, L.L.; Fuentes, M.; Riccardi, V.M.; Holcombe, J.H.; Nitowsky, H.M.; Sachs, G.; Anderson, C.E.; Duchon, M.A.; Owens, R.; Merkatz, L; New, M.L: Amniotic fluid concentrations of A5 and A4 steroids in fetuses with congenital adrenal hyperplasia due to 21hydroxylase deficiency and in anencephalic fe tuses. J. din. Endocr. Metab. 51: 223-229 (1980). 19 Hughes, I.A.; Laurence, K.M.: Prenatal diagnosis of congenital adrenal hyperplasia due to 21-hy droxylase deficiency: amniotic fluid steroid analy sis. Prenat. Diagn. 2: 97-102 (1982). 20 Frasier. S.D.; Weiss, B.A.; Horton, R.: Amniotic fluid testosterone: implications for the prenatal diagnosis of congenital adrenal hyperplasia. J. Pe diatr. 84: 738-741 (1974). 21 Warne, G.L.; Faiman, C.; Reyes, F.L; Winter, J.S.D.: Studies on human sexual development. V. Concentrations of testosterone, 17-hydroxyprogesterone and progesterone in human amniotic fluid throughout gestation. J. din. Endocr. Metab. 44: 934-938 (1977). 22 New, M.L; Lorenzen, F.; Lerner, A.J.; Kohn, B.; Oberfield, S.E.; Pollack, M.S.; Dupont, B.; Stoner, E.; Levy, D.J.; Pang, S.; Levine, L.S.: Genotyping steroid 21-hydroxylase deficiency: hormonal ref erence data. J. din. Endocr. Metab. 57: 320-326 (1983). 23 New, M.L; Mercado, A.B.; Wood, E.; Buegeleisen, M.; Karaviti, L.P.; Speiser, P.W.: Prenatal diagno sis of 21-hydroxylase deficiency when the parent is the patient (abstract). Proc. 72nd Annu. Mtg. Endocr. Soc. (Atlanta). Endocrinology 389: (suppl.), pp. 1458 (1990). 24 Marcus, E.S.; Holcombe, J.H.; Tulchinsky, D.; Rich, R.R.; Riccardi, V.M.: Prenatal diagnosis of congenital adrenal hyperplasia. Am. J. med. Gen et. 4: 201-204 (1979). 25 Rosenmann, A.; Schumert, Z.; Theodor, R.; Co hen, T.; Brautbar, C.: Amniotic 17a-hydroxyprogesterone and HLA typing for the prenatal diagno sis of 2la-hydroxylase deficiency - congenital ad renal hyperplasia. Am. J. med. Genet. 6: 295-300 (1980).
209
210
47
48
49 50
51
52
53
54
55
56
hormone regulation of mRNAs for insulin-like growth factor II and the cholesterol side-chaincleavage enzyme, P450scc, in human steroido genic tissues. Proc. natn. Acad. Sci. USA 84: 1590-1594 (1987). Rainey, W.E.; Mason, J.I.; Cochet, C.; Carr, B.R.: Protein kinase-C in the human fetal adrenal gland. J. din. Endocr. Metab. 67: 908-914 (1988). Allen, J.P.; Greer, M.A.; McGilvra, R.; Castro, A.; Fisher, D.A.: Endocrine function in an anencé phalie infant. J. din. Endocr. Metab. 38: 94-98 (1974). Seron-Ferré, M.; Jaffe, R.B.: The fetal adrenal gland. Ann. Rev. Physiol. 43: 141-162 (1981). Carr, B.R.; Ohashi, M.; Parker, C.R., Jr.; Simp son, E.R.: The role of cyclic adenosine 3',5'monophosphate in cholesterol metabolism and steroidogenesis by the human fetal adrenal gland. J. din. Endocr. Metab. 52: 1124-1128 (1981). Carr, B.R.; Ohashi, M.; Simpson, E.R.: Low-den sity lipoprotein binding and de novo synthesis of cholesterol in the neocortex and fetal zones of the human fetal adrenal gland. Endocrinology 110: 1994-1998 (1982). Carr, B.R.; Milburn, J., Jr., Wright, E.E.; Simp son, E.R.: Adenylate cyclase activity in neocortex and fetal zone membrane fractions of the human fetal adrenal gland. J. din. Endocr. Metab. 60: 718-722 (1985). Carr, B.R.; Rainey, W.E.; Mason, J.I.: The role of calcium in steroidogenesis in fetal zone cells of the human fetal adrenal gland. J. din. Endocr. Metab. 63: 913-917 (1986). Shepard, T.H.: Catalog of teratogenic agents; 5th ed. (Johns Hopkins University Press, Baltimore (1986). Heinonen, O.P.; Slone, D.; Shapiro, S.: Birth de fects and drugs in pregnancy (Publishing Sciences Group, Littleton (1977). Loeuille, G.A.; David, M.; Forest, M.G.: Prenatal treatment of congenital adrenal hyperplasia: re port of a new case. Eur. J. Pediatr. 149: 237-240 (1990). Maria I. New Professor and Chairman Department of Pediatrics Room N-236 The New York Hospital-Cornell Medical Center 525 East 68th Street New York, NY 10021 (USA) Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
34 Dörr, H.G.; Sippell, W.G.; Haack, D.; Bidlingmaier, F.; Knorr, D.: Pitfalls of prenatal treatment of congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency (abstract No. 26). Pro gram and Abstracts, 25th Annu. Mtg. Eur. Soc. Pediatr. Endocr., Zürich 1986. 35 Goldman, A.S.; Shapior, B.H.; Katsumata, M.: Human foetal palatal corticoid receptors and tera togens for cleft palate. Nature 272: 464-466 (1978). 36 Goldman, A.S.: Biochemical mechanism of gluco corticoid- and phenytoin-induced cleft palate. Curr. Top. Devi Biol. 19: 217-239 (1984). 37 Bingiovanni, A.M.; McPadden, A.J.: Steroids dur ing pregnancy and possible fetal consequences. Fert. Steril. 11: 181-186 (1960). 38 Forest, M.G.; Bétuel, H.; David, M.: Prenatal treatment in congenital adrenal hyperplasia due to 21-hydroxylase deficiency: update 88 of the French multicentric study. Endocr. Res. 15: 277301 (1989). 39 Knorr, D.; Bidlingmaier, F.; Dörr, H.G.; Holler, W.; Albert, E.D.; Scholz, S.: Congenital adrenal hyperplasia - Diagnosis, treatment and monitor ing; in Imura et al., Progress in endocrinology. Proc. 8th Inti. Congr. Endocrinol., Kyoto 1988, pp. 165-170 (Elsevier, Amsterdam 1988). 40 Ryan, K.J.: Metabolism of C-16-oxygenated ste roids by human placenta: the formation of estriol. J. biol. Chem. 234: 2006-2013 (1959). 41 Johannisson, E.: The foetal adrenal cortex in the human: its ultrastructure at different stages of de velopment and in different functional states. Acta endocrinol. 58: suppl. 130, pp 1-107 (1968). 42 Winter, J.S.D.: The adrenal cortex in the fetus and neonate; in Anderson, Winter, Adrenal cortex, pp. 32-56 (Butterworths, London 1985). 43 Arai, K.; Kuwabara, Y.; Okinaga, S.: The effect of adrenocorticotropic hormone and dexamethasone, administered to the fetus in utero, upon fetal and maternal estrogens. Am. J. Obstet. Gynec. 113: 316-322 (1972). 44 Benirschke, K.: Adrenals in anencephaly and hy drocephaly. Obstet. Gynec. 8: 412-425 (1956). 45 John, M.E.; Simpson, E.R.; Carr, B.R.; Magness, R.R.; Rosenfeld, C.R.; Waterman, M.R.; Mason, J.I.: Ontogeny of adrenal steroid hydroxylases: evidence for cAMP-independent gene expression. Mol. cell, endocrinol. 50: 263-268 (1987). 46 Voutilainen, R.; Miller, W.L.: Coordinate tropic
New
© 1990 S. Karger AG, Basel 0379-8305/90/0154-0200$2.75/0
Dev Pharmacol Ther 1990;15:200-210
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome (Steroid 21-Hydroxylase Deficiency) Maria I. New Department of Pediatrics, New York Hospital, New York, N.Y., USA
Key Words. Congenital adrenal hyperplasia ■ 17-Hydroxyprogesterone • HLA genotyping • Chorionic villus sampling • Dexamethasone administration Abstract. Prenatal treatment of pregnancies at risk for congenital adrenal hyperplasia due to 21-hydroxylase deficiency was carried out in conjunction with chorionic villus sampling (CVS) in the first trimester for analysis of restriction fragment length polymorphisms. Dexa methasone administration to the pregnant woman was initiated at a mean gestational age of 7 weeks (range 4-10 weeks) before testing to determine whether the fetus was affected with 21-hydroxylase deficiency, and CVS was performed at a gestational age of 8-10 weeks. Two affected female fetuses were identified by molecular genetic techniques among this group. The duration of unnecessary prenatal dexamethasone treatment for unaffected or male fetuses was substantially reduced in the CVS group compared with a cohort of 8 prenatally treated pregnancies where amniocentesis was performed in the early second trimester. No major morbidities were observed in the treated pregnancies. Postnatal confirmation of CVS diagnosis was obtained in all cases where DNA from an affected sibling was available for comparative analysis with the DNA from chorionic villus tissue. The external genitalia of the affected females who were treated prenatally appeared normal. Based on these data we con clude that the benefit:risk ratio is favorable for prenatal administration of dexamethasone in pregnancies at risk for 21-hydroxylase deficiency. Treatment should be initiated during the first trimester in conjunction with diagnosis by CVS/molecular genetic techniques. Long term postnatal surveillance is recommended for all offspring of dexamethasone-treated preg nancies.
may be preventable by prenatal administration of dexamethasone to the mother, which suppresses the abnormal secretion of androgens by the fetal adrenal gland. Results from follow-up and survey studies suggest that it is Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
Females with the severe, classic form of 21-hydroxylase deficiency are born with ambiguous genitalia and often require multiple reconstructive surgical procedures to gain normal reproductive function. This problem
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
Background: Diagnosis and Management It had long been known that excretion in the urine of the steroid metabolite pregnane triol was the salient biochemical feature of the adrenogenital syndrome [1], classical congenital adrenal hyperplasia (CAH). The underlying abnormality is a deficiency of the enzyme steroid 21-hydroxylase, and this is the enzyme defect exhibited by the greatest number of CAH patients. (The other adrenal enzyme defects, each with characteristic ab normalities of sex differentiation [2], have since been exhaustively described [3, 4] but together they comprise only the remaining 5% of cases. Classical CAH owing to 21hydroxylase deficiency is the most common basis of female pseudohermaphroditism.
In virilizing CAH, affected females are born with ambiguous external genitalia. They are at risk for wrong sex assignment and missed diagnosis. Both sexes undergo precocious advance in growth and progres sive virilization. Three out of four cases also exhibit salt wasting and fluid loss, often de veloping acute adrenal insufficiency. Before the era of hormonal replacement, many of these patients would die from such episodes. Steroid replacement was shown first in 1950 to correct the endocrine imbalance and pro vide essential hormonal function [5, 6]. Problems of gender confusion in affected females may develop from improper sex as signment or uncertain sex assignment and rearing [7-10], Prenatal androgen effects on gender remain controversial.
Prenatal Diagnosis In 1965, Jeffcoate et al. [11] successfully diagnosed classical 21-hydroxylase defi ciency in a fetus on the basis of raised amniotic fluid concentrations of 17-ketosteroids and pregnanetriol near term compared with normal values. By 1976, prenatal diag nosis of CAH by hormonal assay of amniotic fluid, maternal urine or both had been at tempted by a number of groups [reviewed in ref. 12], Linkage between 21-hydroxylase and HLA was reported by Dupont et al. [13] and Levine et al. [14], Since then, HLA geno typing has been used in family studies as a genetic marker for the enzyme defect in family members. Hormonal Measurement. The most spe cific serum hormonal determinant for 21hydroxylase deficiency is elevated 17ahydroxyprogesterone (17-OHP). Amniotic fluid 17-OHP is always high when the fetus Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
important to begin therapy no later than the 7th week of gestation, and that the optimal dexamethasone dose is 20 pg/kg/day divided in two or three doses. In only 1 of 4 pregnancies in question will the fetus be affected, and one half of these will be males who do not need prenatal treat ment for prevention of virilization. Thus, only one eighth of fetuses require treatment. Prenatal diagnosis of 21-hydroxylase defi ciency, utilizing the new molecular probes for human leukocyte antigens (HLA) and CYP21 probes combined with hormonal studies and serological HLA genotyping, is now possible early in gestation with tissue obtained by chorionic villus sampling (CVS). This is of great advantage in the prenatal management of potential cases in pregnancies of families at risk because the earlier diagnosis afforded by CVS permits sooner termination of ste roid therapy, unnecessary when the fetus is a genetic male or is unaffected.
201
is affected with salt-wasting 21-hydroxylase deficiency; if the disease takes the simple vir ilizing form, 17-OHP levels may be only ambiguously elevated in the amniotic fluid [15-19]. The amniotic fluid A4-androstenedione concentration may be assayed as an adjunct to 17-OHP measurement [18], Am niotic fluid testosterone levels have also been measured but may be within the nor mal male range in the case of an affected male [18, 20], An important factor in prena tal hormonal evaluation is that normal am niotic fluid values for testosterone, proges terone and 17-OHP differ according to fetal sex, and may change significantly with gesta tional age [21], HLA Typing. The gene loci for the ste roid 21-hydroxylase enzyme (a microsomal membrane-bound cytochrome P450, cyt P450c21) are within the close linkage group HLA (once HL-A for antigen-A), the human major histocompatibility complex, or Mhc. Elements of HLA exhibit a high degree of polymorphism, making them of particular utility as markers. Transmission of 21-hy droxylase deficiency genes can thus be fol lowed in a pedigree through the cosegregat ing HLA antigens, when these are known. HLA genotyping is ideally done and com plete pedigree taken for any family manifest ing 21-hydroxylase deficiency and wishing more offspring. HLA types are needed at minimum for the parents (who should test as carriers on adrenocorticotropic hormone (ACTH) testing [22] - unless themselves af fected [23], and one affected child. The fetal HLA genotype is determined from standard serotyping of a fetal cell cul ture or molecular probe analysis of fetal DNA, which may also derive from cell cul ture or can be amplified directly from fetal cell DNA by the polymerase chain reaction
New
technique (PCR). Fetal cell cultures are grown up from viable fetal cells obtained by one of two invasive procedures: free cells, mostly fibroblasts, from a quantity of am niotic fluid drawn on amniocentesis, or more recently, villus cells from the chorion frondosum obtained by needle aspiration in transabdominal or transcervical CVS. Prenatal diagnosis of 21-hydroxylase de ficiency makes use of HLA associations [2426]. The fetal HLA type is compared with that of the family index case (an affected liv ing sib) and accordingly, as it is HLA identi cal or shares one or neither haplotype, the fetus is predicted to be affected, to be a car rier, or to be unaffected and not a carrier. Confounding factors in presumptive diagno sis of 21 -hydroxylase deficiency by HLA typ ing are (1) intra-HLA recombination events, (2) homozygosity at the HLA-B locus in either parent and (3) antigen sharing be tween the parents. When HLA typing is by serological techniques, typing of the HLA antigens called class II (HLA-DR, DQ and DP) requires a more specialized assay [27], Unless the HLA-DR antigen is typed, recom bination between the HLA-B and DR loci, which may include the 21-hydroxylase locus, segregating it away from its primary marker HLA-B, will go unidentified. If HLA typing is by DNA analysis, molecular probes used include many for class II, whereas uncer tainty arising from homozygosity or antigen sharing requires the use of probes for serum complement isotypes C4A and C4B, adja cent to the 21-hydroxylase locus, or probes for the 21-hydroxylase gene itself, CYP21, and pseudogene CYP21P. The CYP21 probes are not used routinely because there are as yet few mutations known to affect restriction enzyme sites within these genes. Instead, HLA class I and class II probes are Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
202
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
203
The synthetic steroid dexamethasone (9afluoro- 16a-methyl -1 lß, 17a,21 -trihydroxypregna-l,4-diene-3,20-dione), a substituted A1- or 1,2-dehydrocortisol, has potent gluco corticoid action and a long plasma half-life. It is not bound by corticosteroid-binding globulin (CBG) and crosses the placenta without inactivation. Theoretically, dexamethasone adminis tered to the mother from the onset of fetal adrenal activity at 6-7 weeks of gestation will suppress the fetal adrenals, reducing DNA Analysis/CYS androgen production in an affected female CVS, normally done at 8-11 weeks of ges and preventing labial fusion, formation of a tation, permits evaluation of the fetus at risk urogenital sinus, and clitoromegaly. Where in the first trimester. Fetal DNA is geno- as an earlier study found the fetal pituitarytyped using molecular probes (HLA, C4 and adrenal axis not to suppress after acute ad 21-hydroxylase), and HLA by serology may ministration of dexamethasone at mid-term still be done for confirmation or to direct the [31], prenatal treatment of fetuses affected choice of informative DNA probes. Karyo with or at risk for 21-hydroxylase defi typing or Y-chromosome probes are used to ciency has recently been employed with sex the fetus. Amniotic fluid in the aspirate success [reviewed in ref. 32]. In 1989, sum has been assayed experimentally for eleva marizing treatment and outcome for 43 tion of 17-OHP levels. Viable cells are plated pregnancies at risk, Forest et al. [26] re and cultures grown up for sufficient DNA to ported a decisive reduction in masculinizabe obtained for analysis (10 days to 2 weeks tion following treatment with dexametha time). PCR-amplification of fetal cell DNA sone in 6 affected female fetuses. Although directly from the CVS sample is very rapid treated cases are few, oral dexamethasone but may generate inauthentic sequences 0.5 mg twice daily before the 10th week brought about complete suppression of fetal [30], Amniocentesis has proven useful in pre adrenocortical hormones as measured at natal diagnosis of steroid 21-hydroxylase de amniocentesis. In a group reporting from ficiency, but CVS is preferred when treat Germany on 3 pregnancies at risk for CAH ment is involved. The advantage of CVS treated with dexamethasone [33, 34], the over amniocentesis is the shorter time to protocol included suspension of therapy for diagnosis and therefore sooner termination of 5 days immediately before amniocentesis. unnecessary steroid treatment. Masculiniza- High amniotic fluid 17-OHP levels were tion of external genital processes begins by measured, and it was postulated that they the 6-7th week of gestation. CVS in the 8- could have been caused by increased meta 10th week provides the earliest information bolic clearance of dexamethasone, or inade as to whether prenatal steroid treatment is quate dosage. The babies were born with necessary (fig. 1). ambiguous genitalia. Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
used [28], Panels of synthetic oligonucleo tide probes will screen for already known mutations. Amniotic fluid 17-OHP may in some cir cumstances add to the diagnostic evaluation when there is a question concerning the HLA types, and in others corroborate the HLA prediction, and hormonal measurement is still advocated [29].
204
New
Fig. 1. Relation between differentiation of gonads and anatomic differentiation of human male and female embryos.
Dexamethasone treatment may be insti tuted early in all pregnancies at risk and maintained until diagnosis by CVS and with
out ill effect for the fetus. The current recom mendation is oral dexaméthasone to the mother upon confirmation of pregnancy at a starting dose of 20 pg/kg/day (to a maximum of 1.5 mg) divided equally in 3 doses. To Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
Current Recommendations
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
treated with dexamethasone. Of these 8, there were 2 affected fetuses - 1 male and 1 female - who were born with normal genita lia. Three female fetuses who were affected and were not treated, had ambiguous genita lia. In the CVS group of 18 pregnancies (7 males and 11 females), two females were diagnosed to be affected with classical 21hydroxylase deficiency. These females were treated prenatally with dexamethasone and were born with no labial fusion. One had a normal clitoris and the other very mild clitoromegaly, which did not require clitoroplasty. This compared favorably with the sig nificant genital ambiguity of the untreated older affected sisters. Accurate prenatal diagnosis of 21 -hydrox ylase deficiency minimizes unnecessary treatment of males and unaffected females. A problem to be resolved is the risk to the fetus and mother of the possible procedures. Amniocentesis has a 0.5% mortality risk to the fetus. In experienced hands, CVS has no greater risk than amniocentesis. Dexametha sone has no known teratological effects.
Fetal Adrenal Function and Safety of Steroids The good result obtained in affected fe male infants born to mothers administered dexamethasone from early pregnancy to term without interruption gives empiric proof of the suppressibility of the fetal adre nal by this modality of treatment. Before this time, there has not been even a single in stance of a genetic female with 21-hydroxy lase deficiency CAH born with completely normal genitalia, and the virilization suf fered by the untreated developing fetus may extend to an external genital appearance inDownloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
date, no fetus of a mother treated with dexamethasone in low doses has been found to have any congenital malformations. A specific concern in connection with corticosteroid use and fetal development is cleft palate formation. This sequela, with placental degeneration and fetal death has been observed in rodents exposed in utero to high-dose glucocorticoids [35, 36]. An older survey of reports in the literature identified this malformation in 2 out of 260 exposures, in both cases the steroid given in large dose before the 14th week [37], and low-dose corticoids were felt to present little risk. Long term follow-up of prenatal treatment in France finds all children (affected and unaf fected fetuses) normal [26, 38] after 8.5 years. In some cases treatment was interrupted 5-7 days before amniocentesis, in order for the diagnosis to be confirmed on the basis of amniotic fluid hormonal values. Increased genital ambiguity in affected females has been attributed to the marked rise in fetal adrenal output at this time. Although it is still advocated in some protocols [39], the diagnostic advantage of this test is out weighed by the apparent increased risk of masculinization in an affected female, and we do not recommend it. In a recent study from our group, 49 fetuses were diagnosed by amniocentesis and CVS [32]. DNA analysis utilizing HLA ge netic probes as well as 21-hydroxylase ge netic probes were utilized on the DNA ex tracted from the CVS. An algorithm for diag nosis and treatment is shown in figure 2. Treatment was begun at 7 weeks of gestation before the diagnosis was known. Out of 49 fetuses studied, 27 proved to be males and 22 proved to be females. In the amniocente sis group of 31 women, 8 consented to be
205
206
New
3-4 weeks
(7-10 weeks) STOP
dexamethasone
STOP
2-S days later
1 -2 weeks later
dexamethasone (15-20 weeks}
2-5 days later
1 2 weeks later
Fig. 2. Procedure for management of pregnancies at risk for congenital adrenal hyperplasia due to 21-hydroxylase (21-OH) deficiency. hCG = Human chorionic gonadotropin.
The rise throughout gestation of the hormone estriol, which derives from fetal adrenal dehydroepiandrosterone-sulfate via fetal hepatic 16a-hydroxylation and placental aromatization, and thus expresses fetoplaDownloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
distinguishable from normal male. By now there are a number of successfully treated cases who at birth showed decisively reduced virilization or even to have normal female external genital appearance.
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
normal or CAH-affected fetal adrenal func tion. The possibility of zone-specific or tran sient sensitivity to other trophic factor(s) [49] notwithstanding, there is abundant evi dence in favor of ACTH as the primary adre nocortical control factor in the developing fetus [42, 50-53], Thus, although ACTH and cAMP-independent pathways for fetal adre nal stimulation may exist, it seems implausi ble either that (a) reduced virilization in infants born of treated pregnancies is inci dental, or that (b) dexamethasone acts to a significant degree through any mechanism other than inhibition of ACTH release. The safety of the gestational use of gluco corticoids has been reasonably well estab lished. Shepard’s Catalog of Teratogenic Agents [54] categorizes corticosteroids as ‘not highly dangerous’ in human pregnancy, with a relative risk for congenital malforma tions from prenatal exposure in human pregnancy of less than 1, i.e. not significant [55], In the experience of the author’s group and again in the large series reported from France [38], low-dose dexamethasone ther apy has no teratogenic potential. Neverthe less, it seems prudent to reduce the duration of unnecessary treatment, making CVS a more desirable diagnostic measure than am niocentesis. As experience with these proce dures in prenatal diagnosis and treatment increases, a better assessment of the risks and benefits of each will be obtained. A new case of prenatal treatment of a fetus with 21-hydroxylase deficiency has recently been reported. Dexamethasone was given from the third week of gestation until term in a dose of 0.5 mg every 12 h. The female in fant was born with a normal clitoris but mild posterior labial fusion. The authors recommend, based on this experience, that Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
cental steroidogenic activity [40], has been used as an index of fetal well-being (exagger ated rise may signal a multiple gestation). In France, Forest et al. [38] collected data on pregnancies at risk for 21-hydroxylase defi ciency CAH (but undiagnosed - thus repre senting both affected and unaffected fetuses) being managed on comparable prenatal treatment protocols (dexamethasone 1 or 1.5mg/day from 7-10 weeks to term). Es triol levels (measured as concentration in amniotic fluid or maternal serum, or amount secreted in 24-h urine) were significantly lower in treated than in age-matched normal fetuses. Adrenal development and regulation in utero is a complex issue [41, 42], The effect of ACTFI and dexamethasone on fetoplacen tal estrogen production following their ad ministration directly into the fetus or in the amniotic cavity has been measured [43], but to our knwledge the threshold for in vivo ACTH stimulation of the fetal adrenal has not been determined. It has been suggested that the fetal adre nal may be regulated by, or at least be responsive to other factors. Autopsy and as sessment of adrenal tissue in anencephaly have revealed partial evolution of steroido genic capacity in some cases [ 18, 44, 45], and some recent observations in vitro support the idea of non-ACTH-dependent or noncAMP-mediated control [46, 47], The em bryologie deficit in anencephalic fetuses, however, is not uniform or absolute and there may be remnants of endocrine feed back systems (in vivo data from at least one anencephalic has clearly shown measurable and stimulable ACTH levels [48]). Different cases, therefore, may not always be compara ble, nor must in vitro data from tissues of anencephalics be valid for understanding
207
New
the dose of dexamethasone be given three times daily [56], Infants treated prenatally from the French cohort (both unaffected and affected) have been followed carefully and normal growth and cognition up to a period of 8.5 years have been reported [38]. It is impor tant that the growth and development of infants exposed to dexamethasone prena tally should be followed closely for signs of unidentified side effects emerging in the long term, involving, for instance, changes in in tellectual or motoric function or somatic growth. To date, however, prenatal dexa methasone treatment given in proper dosage appears to be safe and effective prenatal treatment for CAH.
References 1 Butler, G.C.; Marrian, G.F.: The isolation of preg nane-3, 17,20-triol from the urine of women show ing the adrenogenital syndrome. J. biol. Chem. 119: 565-572 (1937). 2 Bongiovanni, A.M.; Eberlein, W.R.; Goldman, A.S.; New, M.I.: Disorders of adrenal steroid bio genesis. Recent Prog. Horm. Res. 23: 375-449 (1967). 3 New, M.I.; White, P.C.; Pang, S.; Speiser, P.W.: Dupont, B.: The congenital adrenal hyerplasias; in Scriver, The metabolic basis of inherited disease; 6th ed., pp. 1881-1917 (McGraw-Hill, New York 1989). 4 White, P.C.; New, M.I.; Dupont, B.: Congenital adrenal hyperplasia. New Engl. J. Med. 316: 1519-1524 and 1580-1586 (1987). 5 Wilkins, L.; Lewis, R.A.; Klein, R.; Rosemberg, E.: The suppression of androgen secretion by cor tisone in a case of congenital adrenal hyperplasia. Bull. Johns Hopkins Hosp. 56: 249-252 (1950). 6 Bartter, F.C.; Albright, F.; Forbes, A.P.; Leaf, A.; Dempsey, E.; Carroll, E.: The effects of adreno corticotropic hormone and cortisone in the adre nogenital syndrome associated with congenital adrenal hyperplasia. An attempt to explain and
7
8
9
10
11
12
13
14
15
16
correct its disordered hormonal pattern. J. clin. Invest. 30: 237-251 (1951). Klingensmith, G.J.; Garcia, S.C.; Jones, H.W., Jr.; Migeon, C.J.; Blizzard, R.M.: Glucocorticoid treatment of girls with congenital adrenal hyper plasia: effects on height, sexual maturation, and fertility. J. Pediatr. 90: 996-1004 (1977). Mulaikal, R.M.; Migeon, C.J.; Rock, J.A.: Fertil ity rates in female patients with congenital adrenal hyperplasia due to 21-hydroxylase deficiency. New Engl. J. Med. 316: 178-182 (1987). Thibaud, E.; Rappaport, R.; Salomon Bernard, Y.; Paniel, B.J.; Dero, M.; Lepillier, P.; Philippe, F.; Pellerin, D.: Fertilité normale chez 2 femmes traitées pour hyperplasie congénitale des surréna les (FICS) avec syndrome de perte de sel (SPS) et ambiguïté génitale Prader III et IV (abstract); in Bertrand et al.: Sexual differentiation: basic and clinical aspects. Proc. Inti. Symp. Montpellier 1989. Federman, D.D.: Psychosexual adjustment in congenital adrenal hyperplasia (editorial). New Engl. J. Med. 316: 209 (1987). Jeffcoate, T.N.A.; Fliegner, J.R.H.; Russell, S.H.; Davis, J.C.; Wade, A.P.: Diagnosis of the adreno genital syndrome before birth. Lancet ii: 553-555 (1965). New, M.I.: Prenatal diagnosis of congenital adre nal hyperplasia; in New, Fiser, Diabetes and other endocrine disorders during pregnancy and in the newborn. Prog. clin. biol. Res., voi. 10, pp. 205219 (Alan R. Liss, New York 1976). Dupont, B.; Oberfield, S.E.; Smithwick, E.M.; Lee, T.D.; Levine, L.S.: Close genetic linkage be tween HLA and congenital adrenal hyperplasia (21-hydroxylase deficiency). Lancet ii: 1309-1311 (1977). Levine, L.S.; Zachmann, M.; New, M.I.; Prader, A.; Pollack, M.S.; O’Neill, G.J.; Yang, S.-Y.; Ober field, S.E.; Dupont, B.: Genetic mapping of the 21-hydroxylase deficiency gene within the HLA linkage group. New Engl. J. Med. 299: 911-912 (1978). Frasier, S.D.; Thorneycroft, I.H.; Weiss, B.A.; Horton, R.: Elevated amniotic fluid concentration of 17a-hydroxyprogesterone in congenital adrenal hyperplasia. J. Pediatr. 86: 310-312 (1975). Nagamani, M.; McDonough, P.G.; Ellegood, J.O.; Mahesh, V.B.: Maternal and amniotic fluid 17hydroxyprogesterone levels during pregnancy: Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
208
Prenatal Diagnosis and Treatment of Adrenogenital Syndrome
26 Forest, M.G.; Betuel, H.; Couillin, P.; Boue, A.: Prenatal diagnosis of congenital adrenal hyperpla sia (CAPI) due to 21-hydroxylase deficiency by steroid analysis in the amniotic fluid of mid-preg nancy: comparison with HLA typing in 17 preg nancies at risk for CAH. Prenat. Diagn. 1: 197— 207 (1981). 27 Maurer, D.H.; Pollack, M.S.: The use of gamma interferon to increase HLA antigen expression on cultured amniotic cells used for the prenatal diag nosis of 21-hydroxylase deficiency; in New, Con genital adrenal hyperplasia. Ann. N.Y. Acad. Sci., vol. 458, pp. 148-155 (1985). 28 Mornet, E.; Boue, J.; Raux-Demay, M.; Couillin, P.; Oury, J.E.; Dumez, Y.; Dausset, J.; Cohen, D.; Bouc, A.: First trimester prenatal diagnosis of 21hydroxylase deficiency by linkage analysis to HLA-DNA probes and by 17-hydroxyprogesterone determination. Hum. Genet. 73: 358-364 (1986). 29 Pang, S.; Pollack, M.S.; Loo, M.; Green, O.; Nussbaum, R.; Clayton, G. ; Dupont, B.; New, M.I.: Pitfalls of prenatal diagnosis of 21-hydroxylase deficiency congenital adrenal hyperplasia. J. clin. Endocr. Metab. 61: 89-97 (1985). 30 Ennis, P.D.; Zemmour, J.; Salter, R.D.; Parham, P.: Rapid cloning of HLA-A,B cDNA by using the polymerase chain reaction: frequency and nature of errors produced in amplification. Proc. natn. Acad. Sci. USA 87: 2833-2837 (1990). 31 Charnvises, S.; Fencl, M. de M.; Osathanondh, R.; Zhu, M.-G.; Underwood, R.; Tulchinsky, D.: Ad renal steroids in maternal and cord blood after dexamethasone administration in midterm. J. din. Endocr. Mctab. 61: 1220-1222 (1985). 32 Speiser, P.W.; Laforgia, N.; Kato, K.; Pareira, J.; Khan, R.: Yang, S.-Y.; Whorwood, C; White, P.C.; Elias, S; Schriock, E.; Schriock, E. [sic], Simpson, J.L.; Taslimi, M.; Najjar, J.; May, S.; Mills, G.; Crawford, C; New, M.I.: First trimester prenatal treatment and molecular genetic diagno sis of congenital adrenal hyperplasia (21-hydroxy lase deficiency). J. din. Endocr. Metab. 70: 838— 848 (1990). 33 Knorr, D.; Bidlingmaier, F.; Dörr, H.G.; Kuhnle, U.: Prenatal treatment of a girl with 21-hydroxylase deficiency (CAH) (abstract No. 117). Pro gram and Abstracts, 2nd Joint Mtg. Lawson Wil kins Pediatr. Endocr. Soc. and Eur. Soc. Pediatr. Endocr., Baltimore 1985. Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
diagnosis of congenital adrenal hyperplasia. Am. J. Obstet. Gynec. 130: 791-794 (1978). 17 Hughes, I.A.; Laurence, K.M.: Antenatal diagno sis of congenital adrenal hyperplasia. Lancet ii: 79 (1979). 18 Pang, S.; Levine, L.S.; Cederqvist, L.L.; Fuentes, M.; Riccardi, V.M.; Holcombe, J.H.; Nitowsky, H.M.; Sachs, G.; Anderson, C.E.; Duchon, M.A.; Owens, R.; Merkatz, L; New, M.L: Amniotic fluid concentrations of A5 and A4 steroids in fetuses with congenital adrenal hyperplasia due to 21hydroxylase deficiency and in anencephalic fe tuses. J. din. Endocr. Metab. 51: 223-229 (1980). 19 Hughes, I.A.; Laurence, K.M.: Prenatal diagnosis of congenital adrenal hyperplasia due to 21-hy droxylase deficiency: amniotic fluid steroid analy sis. Prenat. Diagn. 2: 97-102 (1982). 20 Frasier. S.D.; Weiss, B.A.; Horton, R.: Amniotic fluid testosterone: implications for the prenatal diagnosis of congenital adrenal hyperplasia. J. Pe diatr. 84: 738-741 (1974). 21 Warne, G.L.; Faiman, C.; Reyes, F.L; Winter, J.S.D.: Studies on human sexual development. V. Concentrations of testosterone, 17-hydroxyprogesterone and progesterone in human amniotic fluid throughout gestation. J. din. Endocr. Metab. 44: 934-938 (1977). 22 New, M.L; Lorenzen, F.; Lerner, A.J.; Kohn, B.; Oberfield, S.E.; Pollack, M.S.; Dupont, B.; Stoner, E.; Levy, D.J.; Pang, S.; Levine, L.S.: Genotyping steroid 21-hydroxylase deficiency: hormonal ref erence data. J. din. Endocr. Metab. 57: 320-326 (1983). 23 New, M.L; Mercado, A.B.; Wood, E.; Buegeleisen, M.; Karaviti, L.P.; Speiser, P.W.: Prenatal diagno sis of 21-hydroxylase deficiency when the parent is the patient (abstract). Proc. 72nd Annu. Mtg. Endocr. Soc. (Atlanta). Endocrinology 389: (suppl.), pp. 1458 (1990). 24 Marcus, E.S.; Holcombe, J.H.; Tulchinsky, D.; Rich, R.R.; Riccardi, V.M.: Prenatal diagnosis of congenital adrenal hyperplasia. Am. J. med. Gen et. 4: 201-204 (1979). 25 Rosenmann, A.; Schumert, Z.; Theodor, R.; Co hen, T.; Brautbar, C.: Amniotic 17a-hydroxyprogesterone and HLA typing for the prenatal diagno sis of 2la-hydroxylase deficiency - congenital ad renal hyperplasia. Am. J. med. Genet. 6: 295-300 (1980).
209
210
47
48
49 50
51
52
53
54
55
56
hormone regulation of mRNAs for insulin-like growth factor II and the cholesterol side-chaincleavage enzyme, P450scc, in human steroido genic tissues. Proc. natn. Acad. Sci. USA 84: 1590-1594 (1987). Rainey, W.E.; Mason, J.I.; Cochet, C.; Carr, B.R.: Protein kinase-C in the human fetal adrenal gland. J. din. Endocr. Metab. 67: 908-914 (1988). Allen, J.P.; Greer, M.A.; McGilvra, R.; Castro, A.; Fisher, D.A.: Endocrine function in an anencé phalie infant. J. din. Endocr. Metab. 38: 94-98 (1974). Seron-Ferré, M.; Jaffe, R.B.: The fetal adrenal gland. Ann. Rev. Physiol. 43: 141-162 (1981). Carr, B.R.; Ohashi, M.; Parker, C.R., Jr.; Simp son, E.R.: The role of cyclic adenosine 3',5'monophosphate in cholesterol metabolism and steroidogenesis by the human fetal adrenal gland. J. din. Endocr. Metab. 52: 1124-1128 (1981). Carr, B.R.; Ohashi, M.; Simpson, E.R.: Low-den sity lipoprotein binding and de novo synthesis of cholesterol in the neocortex and fetal zones of the human fetal adrenal gland. Endocrinology 110: 1994-1998 (1982). Carr, B.R.; Milburn, J., Jr., Wright, E.E.; Simp son, E.R.: Adenylate cyclase activity in neocortex and fetal zone membrane fractions of the human fetal adrenal gland. J. din. Endocr. Metab. 60: 718-722 (1985). Carr, B.R.; Rainey, W.E.; Mason, J.I.: The role of calcium in steroidogenesis in fetal zone cells of the human fetal adrenal gland. J. din. Endocr. Metab. 63: 913-917 (1986). Shepard, T.H.: Catalog of teratogenic agents; 5th ed. (Johns Hopkins University Press, Baltimore (1986). Heinonen, O.P.; Slone, D.; Shapiro, S.: Birth de fects and drugs in pregnancy (Publishing Sciences Group, Littleton (1977). Loeuille, G.A.; David, M.; Forest, M.G.: Prenatal treatment of congenital adrenal hyperplasia: re port of a new case. Eur. J. Pediatr. 149: 237-240 (1990). Maria I. New Professor and Chairman Department of Pediatrics Room N-236 The New York Hospital-Cornell Medical Center 525 East 68th Street New York, NY 10021 (USA) Downloaded by: University of Exeter 144.173.6.94 - 5/5/2020 4:58:11 AM
34 Dörr, H.G.; Sippell, W.G.; Haack, D.; Bidlingmaier, F.; Knorr, D.: Pitfalls of prenatal treatment of congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency (abstract No. 26). Pro gram and Abstracts, 25th Annu. Mtg. Eur. Soc. Pediatr. Endocr., Zürich 1986. 35 Goldman, A.S.; Shapior, B.H.; Katsumata, M.: Human foetal palatal corticoid receptors and tera togens for cleft palate. Nature 272: 464-466 (1978). 36 Goldman, A.S.: Biochemical mechanism of gluco corticoid- and phenytoin-induced cleft palate. Curr. Top. Devi Biol. 19: 217-239 (1984). 37 Bingiovanni, A.M.; McPadden, A.J.: Steroids dur ing pregnancy and possible fetal consequences. Fert. Steril. 11: 181-186 (1960). 38 Forest, M.G.; Bétuel, H.; David, M.: Prenatal treatment in congenital adrenal hyperplasia due to 21-hydroxylase deficiency: update 88 of the French multicentric study. Endocr. Res. 15: 277301 (1989). 39 Knorr, D.; Bidlingmaier, F.; Dörr, H.G.; Holler, W.; Albert, E.D.; Scholz, S.: Congenital adrenal hyperplasia - Diagnosis, treatment and monitor ing; in Imura et al., Progress in endocrinology. Proc. 8th Inti. Congr. Endocrinol., Kyoto 1988, pp. 165-170 (Elsevier, Amsterdam 1988). 40 Ryan, K.J.: Metabolism of C-16-oxygenated ste roids by human placenta: the formation of estriol. J. biol. Chem. 234: 2006-2013 (1959). 41 Johannisson, E.: The foetal adrenal cortex in the human: its ultrastructure at different stages of de velopment and in different functional states. Acta endocrinol. 58: suppl. 130, pp 1-107 (1968). 42 Winter, J.S.D.: The adrenal cortex in the fetus and neonate; in Anderson, Winter, Adrenal cortex, pp. 32-56 (Butterworths, London 1985). 43 Arai, K.; Kuwabara, Y.; Okinaga, S.: The effect of adrenocorticotropic hormone and dexamethasone, administered to the fetus in utero, upon fetal and maternal estrogens. Am. J. Obstet. Gynec. 113: 316-322 (1972). 44 Benirschke, K.: Adrenals in anencephaly and hy drocephaly. Obstet. Gynec. 8: 412-425 (1956). 45 John, M.E.; Simpson, E.R.; Carr, B.R.; Magness, R.R.; Rosenfeld, C.R.; Waterman, M.R.; Mason, J.I.: Ontogeny of adrenal steroid hydroxylases: evidence for cAMP-independent gene expression. Mol. cell, endocrinol. 50: 263-268 (1987). 46 Voutilainen, R.; Miller, W.L.: Coordinate tropic
New
![[Prenatal diagnosis and therapy of adrenogenital syndrome with 21-hydroxylase deficiency].](/assets/img/hold.png)
