Congenital Gastric Teratoma in Wiedemann- Beckwith Syndrome
Tzipora C. Falik-Borenstein, Julie R. Korenberg, Irene Davos, Lawrence D. Platt, Stephen Gans, Barbara Goodman, Rhona Schreck, and John M. Graham, Jr.
Ahmanson Pediatric Center, Medical Genetics-Birth Defects Center (T.C.F .- B.,J.R.K.,B.G.,R.S.,J.M.G.), Departments of Pathology (I.D.) and Obstetrics and Gynecology (L.D.P.) USC School of Medicine, and Department of Pediatric Surgery, Cedars-Sinai Medical Center, UCLA School of Medicine (S.G.) Los Angeles, California
Wiedemann-Beckwith syndrome (WBS) may be associated with abdominal tumors, includ- ing Wilms tumor, adrenocortical carcinoma, hepatoblastoma, gonadoblastoma, rhabdo- myosarcoma, and neuroblastoma. We report on a newborn infant with WBS and a congeni- tal teratoma of the stomach. This is the sole report of any teratoma being associated with WBS and also the first report of a tumor pre- sent at birth and visible prenatally in WBS. At birth this infant boy had the diagnostic find- ings of WBS with macroglossia, ear lobule creases and pits, nevus flammeus, and omphalocele, and an abdominal mass. Abnor- malities were detected prenatally when ultra- sound examination showed placental over- growth, polyhydramnios, omphalocele, and posterior abdominal calcifications. Resection of the mass and partial gastrectomy were per- formed at age 10 days; histologic study showed an immature grade-II teratoma con- taining a mixture of mature and immature tis- sues from all germ layers. Results of cyto- genetic studies of blood and teratoma were normal (46,XY). This congenital gastric ter- atoma in a newborn boy with classical WBS may represent either a tumor or an included twin. We discuss its implications for the asso- ciation of WBS with neoplasia and monozygo- tic (MZ) twinning, review various neoplasias associated with WBS, and consider patho- genetic mechanisms.
KEY WORDS: gastric teratoma, twin neo- plasm, Wilms tumor
INTRODUCTION
Wiedemann-Beckwith syndrome (WBS) is one of the most common congenital genetic overgrowth syn- dromes, affecting more than 1 in 13,700 births and re- sulting in an increased risk for neoplasia [Sotelo-Avila and Gooch, 1976]. This condition was first described in 1963 by Beckwith [Beckwith, 1963], who reported on 3 cases of a syndrome of macroglossia, omphalocele, cyto- megaly of the adrenal cortex, renal medullary dyspla- sia, and hyperplastic visceromegaly. In 1964 Wied- emann described the same syndrome in 3 sibs and reported additional findings of macrosomia and neona- tal hypoglycemia. Later, other investigators reported additional associated anomalies, including characteris- tic ear lobe creases and pits, facial nevus flammeus, and hemihypertrophy [Pettanati et al., 1986]. There is wide variability in the clinical and pathologic abnormalities of WBS. Intelligence is usually normal. In severe cases, risk of death can be as high as 21%, usually resulting from congestive heart failure, severe malformations as- sociated with the syndrome, or extreme prematurity due to fetal overgrowth and polyhydramnios [Cohen, 1990]. Many previous cases have appeared to be sporadic but autosomal dominant inheritance with incomplete penetrance and variable expressivity has been sug- gested by Niikawa et al. [Niikawa et al., 1986].
When WBS occurs on a familial basis, the inheritance pattern appears unusual in that transmission often oc- curs through an unaffected mother [Niikawa et al., 1986; Lubinsky et al., 1974; Aleck and Hadro, 1989] and monozygotic twins have been discordant with one twin affected and the other not [Bose et al., 1985; Benke, 1978; Berry et al., 1978; Olney et al., 1988; Litz et al., 1988; Graham, 1989]. WBS appears to result from ge- netic changes on chromosome 11p [Koufos et al., 1985] which is homologous to a mouse imprinted area [Searle et al., 1989]. The unusual inheritance pattern in this disorder may result from genomic imprinting-the dif- ferential expression of genes depending on whether they come from a father or a mother [Solter, 1988].
WBS is associated with a significant risk (5-10%) of neoplasia [Sotelo-Avila and Gooch, 1976; Cohen, 1990]. These patients are predisposed to the development of
Received for publication January 25, 1990; revision received March 20, 1990.
Address reprint requests to John M. Graham, M.D., Sc.D., Direc- tor of Clinical Genetics and Dysmorphology, Cedars-Sinai Medical Center, 444 South San Vicente Boulevard, #1001, Los Angeles, CA 90048.
embryonal tumors such as Wilms tumor, rhabdomyo- sarcoma, hepatoblostoma, adrenocortical carcinoma, gonadoblastoma, and neuroblastoma [Sotelo-Avila and Gooch, 1976]. Early diagnosis of this striking condition is important, both to avoid the complications of hypo- glycemia, and to monitor for neoplasia. Overgrowth in WBS has been associated with primary cellular hyper- plasia [Cohen, 1990] and tissue-specific increased pre- natal mitotic activity. Some tumors in WBS have been shown to undergo reduction to homozygosity for genes on 11p [Koufos et al., 1985], and increased prenatal mitotic activity may predispose cells toward a muta- tional event, and lead to the development of an embryo- nal tumor specific to that tissue [Koufos et al., 1985]. The development of a gastric teratoma in the case de- scribed herein may have been due to a similar loss of heterozygosity. Alternatively, the occurrence of this gastric teratoma might represent the abnormal develop- ment of a monozygotic (MZ) twin. Literature review suggests a possible excess of MZ twinning associated with WBS [Bose et al., 1985; Benke, 1978; Berry et al., 1978; Olney et al., 1988; Litz et al., 1988; Graham, 1989]. The differential gene expression in one of mono- zygotic twins with WBS may also be related to the phe- nomenon of imprinting. This case causes us to consider various pathogenetic factors in WBS, which may be af- fected by imprinting and predispose toward overgrowth, tumor development, and possibly MZ twinning.
CLINICAL REPORT
V.U. was born at term to a 38-year-old G6P4→5, AB1 Hispanic mother and a 43-year-old father. A high mater- nal serum AFP was detected at 17 weeks. Ultrasound examination performed at 18 weeks documented an um- bilical abdominal wall defect, placental overgrowth, and polyhydramnios (Fig. 1A) Follow-up examination showed posterior abdominal wall calcifications in the fetus, and on amniocentesis normal chromosomes (46,XY) and amniotic fluid AFP were found. There was no history of exposure to teratogens during pregnancy. The boy was born at 39 weeks gestation by C-section with birth weight of 4,065 g (>90th centile), length was 52.5 cm (>75th centile), and head circumference was at the 25th centile.
On examination he had a prominent nevus flammeus over the central forehead and eyelids, macroglossia with a protruding tongue, linear fissures in the lobule of the external ear, and 3 mm semi-circular pits on the poste- rior rim of the helix, bilaterally (Fig. 1). The heart sounds were normal. There was a firm mass in the left upper quadrant and a 3 x 3 cm omphalocele. The limbs were normally formed and of equal size bilaterally, but extreme hyperextensibility of the joints was noted. The diagnosis of WBS was made based on the characteristic macrosomia, macroglossia, omphalocele, and ear creases. Parents did not have any signs of WBS, and the family history was unremarkable.
After delivery, the baby had respiratory difficulties and was intubated. At age 5 hours, surgery was per- formed to correct the omphalocele, and a large abdomi- nal mass was palpated behind the stomach. Post-surgi- cal recovery was complicated by hypoglycemia and
hypocalcemia. Magnetic resonance imaging studies of the abdomen showed a posterior abdominal mass with calcifications, and cystic and solid components. The mass was not attached to the kidney, and there was no evidence for viceromegaly. At age 10 days, a gastric tumor was removed. The tumor was localized in the stomach, obliterating its outlet. It was attached also to the pancreas and to the inferior vena cava. The tumor was resected, and a partial gastrectomy with end-to-end anastomosis was performed.
The tumor was semicystic. It weighed 210 g and meas- ured 9.5 x 8 x 7 cm. It enclosed a short conical segment of stomach, which was compressed and distally obliter- ated by the mural mass. There was replacement of the gastric wall by the tumor. Histologically, the tumor con- tained a variety of tissues derived from all 3 germ layers. Mature tissues predominated and were intimately but haphazardly admixed with either partially mature or immature embryonic tissues (Fig. 2). The immature or partially mature components composed over 10% of the total sampled tumor specimen (an estimated 20-40% of the tumor was examined) and consisted of neural, mes- enchymal, and epithelial elements. The tumor was con- sidered to be an immature but benign teratoma with a favorable prognosis. No adrenal, renal, or pancreatic elements were recognized. Results of prometaphase ban- ded cytogenetic studies of blood and the teratoma were identical and normal (46,XY).
At 4 months the patient developed seizures related to hypoglycemic episodes 1-3 hours after feeding. There was no demonstrable hypoglycemia after a 24-hour fast. These findings were compatible with the diagnosis of post-gastrectomy dumping syndrome and were not con- sidered to be related to the patient’s WBS. Use of a carbohydrate-free formula with addition of uncooked corn starch and frequent feeding prevented further epi- sodes. Results of follow-up magnetic resonance imaging of abdomen and chest at age 4 months were normal.
DISCUSSION
Our patient with WBS and a gastric teratoma has an unusual tumor, not previously reported in association with WBS. A teratoma is a true tumor containing tis- sues derived from all germinal layers. Gastric ter- atomas are rare, accounting for less than 1% of all child- hood teratomas. Approximately 60 cases have been reported, and all were histologically benign [Gonzalez- Crussi, 1982]. A gastric teratoma usually appears as a large, partially calcified intra-abdominal mass; in our patient such calcifications were demonstrated pre- natally. The treatment is total excision of the tumor, with excellent prognosis.
Patients with WBS have a substantially increased risk for development of neoplasia with 5-10% of pa- tients developing embryonal tumors [Sotelo-Avila and Gooch, 1976; Pettanati et al., 1986; Haas et al., 1986; Little et al., 1988]. It is important to emphasize that reported cases of WBS with neoplasia have an ascertain- ment bias that favors reporting cases with tumors, and the true frequency of neoplasia might actually be lower. Approximately 40% of neoplasias are associated with hemi-hypertrophy; thus, patients with WBS and hemi-
F
5
A
B
C
D
hypertrophy (13%) have a higher risk of developing tu- mors than those WBS cases without asymmetric over- growth [Haas et al., 1986].
Overgrowth conditions such as WBS, hemi-hypertro- phy and Sotos sequence may all be associated with var- ious tumors of the kidney, liver and adrenal [Cohen, 1990]. In WBS, nephroblastoma is the most commonly reported tumor, followed by adrenocortical carcinoma and hepatoblastoma [Pettanati et al., 1986; Cohen, 1990]. Other neoplasms, such as gonadoblastoma, rhab- domyosarcoma and neuroblastoma also have been re- ported. Most congenital tumors associated with prena- tal overgrowth syndromes are embryonic tumors. They are considered to reflect a developmental disturbance during organogenesis that causes arrest at an early stage in the normal differentiation process [Koufos et al., 1985]. The embryonal tissue continues to proliferate inappropriately (especially in some overgrowth syn- dromes) and produces a mass of immature tissue which is seen as the embryonal tumor.
Loss of heterozygosity has been established as a com- mon mechanism for the development of embryonal tu- mors both in sporadic tumors and in those associated
with WBS [Koufos et al., 1985]. Data obtained from molecular analysis of Wilms tumors, hepatoblastomas and rhabdomyosarcomas show that they may share this common pathogenetic mechanism of losing DNA markers on chromosome 11p. These chromosome changes may unmask recessive mutant alleles at the Wilms tumor (WAGR) locus or another locus on the short arm of 11 and allow its phenotypic expression by elim- ination of the wild type homologue [Koufos et al., 1985]. When this event occurs during the development of kid- ney, muscle, or liver, it may apparently result in the formation of an embryonal tumor. The immature differ- entiation state in such tumors implies that the gene product of the locus that acquired homozygosity is nec- essary for normal progression of differentiation [Koufos et al., 1985; Cohen, 1981]. These losses of heterozygosity have recently been shown to include the genes for WAGR, C-Ha-Ras-1 proto-oncogene, calcitonin on 11p13-15, and insulin on 11p15-15.1 in patients with WBS who developed tumors [Little et al., 1988; Koufos et al., 1985; Ping et al., 1989]. These findings support the hypothesis that somatic loss of heterozygosity for a locus on 11p is involved in the pathogenetic mechanism for
the development of several embryonal tumors in WBS, and that there may be another locus distinct from WAGR which affects tumor predisposition in WBS.
The gastric teratoma described here is of particular interest. It might have risen by a similar loss of hetero- zygosity, but there may be alternative mechanisms for its formation. Unlike other congenital tumors, ter- atomas may originate either from the fetus itself or from an incompletely formed twin. Teratomas may represent complex mixtures of both well-differentiated tissues and fetal structures. Several previous reports suggest that these complex teratomas may actually represent fetus- in-fetu of a monozygotic (MZ) twin [Gonzalez-Crussi, 1982; Heifetz et al., 1988]. Studying the infant and the teratoma for homozygosity of loci on 11p might help to test this hypothesis. If the infant and the teratoma were both found to be heterozygotic for these loci, it would support our hypothesis that the teratoma resulted from a MZ twinning event. If the teratoma was found to be homozygotic for these loci, while the infant is hetero- zygotic, it would suggest that the teratoma reflects the tumorogenic tendency in WBS. If the infant is proven to be homozygotic for loci on 11p, this case would be unin- formative and not testable in this way. In this context, it
is interesting that our review of the literature appears to support the association of WBS with MZ twinning. Al- though it might result from bias of ascertainment, WBS is more frequently reported in monozygotic twins [Bose et al., 1985; Benke, 1978; Berry et al., 1978; Olney et al., 1988; Litz et al., 1988; Graham, 1989] than in dizygotic twins [Piussan et al., 1980; Koufos et al., 1985; Beck- with, 1969].
In all cases, the MZ twins with WBS were discordant with regard to the expression of the condition. These reports suggest that the expression of phenotype in this disorder may reflect factors which result in differing effects for MZ twins. One possibility is that a fetal or placental growth factor is differentially perfused in such MZ twins. For this reason, it is most important to an- alyze placental tissue and the vascular anastomoses associated with twinning in this disorder. Another pos- sible explanation is that there is a difference in growth factor responsiveness in twins (similar to that seen in hemihypertrophy). For example, somatic mutations or differential imprinting of genes related to growth factor receptors might result in localized overgrowth. On the other hand, mutations occurring earlier, when lat- erality is being established, might result in hemi-
hypertrophy. A somatic mutation or change in imprint- ing at the time of separation of inner cell masses in the developing MZ twins might also give rise to discordance in MZ twins.
We suggest that the gastric teratoma in this patient might have been an abnormal MZ twin, a fetus-in-fetu, reflecting the high relative incidence of MZ twinning, as opposed to the tumorogenic tendency of WBS.
ACKNOWLEDGMENTS
This work was supported by the UCLA Intercampus Training Program, grant 08243 from the National Insti- tute of Health and by SHARE’s Cedars-Sinai Child Dis- ability Center. We are indebted to Sheilah Levin for her secretarial expertise and to Dr. J. Bruce Beckwith of the Children’s Hospital, Denver, Colorado, for reviewing the contents of this manuscript.
REFERENCES
Aleck KA, Hadro TA (1989): Dominant inheritance of Wiedemann- Beckwith syndrome: Further evidence for transmission of “unstable premutation” through carrier women. Am J Med Genet 33:155-160.
Beckwith JB (1969): Macroglossia, omphalocele, adrenal cytomegaly, gigantism and hyperplastic viceromegaly. BD:OAS 5:188-196.
Beckwith JB (1963): Extreme cytomegaly of the adrenal fetal cortex, omphalocele, hyperplasia of kidneys and pancreas and Leydig cell hyperplasia: Another syndrome? Western Society for Pediatric Re- search, Los Angeles, CA, November 11, 1963.
Bencke PJ (1978): Familial Beckwith-Wiedemann syndrome. Pre- sented at the annual Birth Defects Conference, San Francisco, Cali- fornia, June 11-12, 1978.
Berry AC, Belton EM, Chantler C (1980): Monozygotic twins discordant for Beckwith-Wiedemann syndrome and the implications for ge- netic counseling. J Med Genet 17:136-138.
Bose B, Wilkie RA, Madlom M, Forsyth JS, Faed MJW (1985): Wied- emann-Beckwith syndrome in one of monozygotic twins. Arch Dis Child 60:1191-1192.
Cohen MM Jr (1990): A comprehensive and critical assessment of overgrowth and overgrowth syndromes. In: Harris H, Hirschhorn K (eds): “Advances in Human Genetics,” Vol. 18 New York: Plenum Press, in press.
Cohen MM Jr (1981): “The Child with Multiple Birth Defects.” New York: Raven Press, pp 107-111 and 120-138.
Gonzalez-Crussi F (1982): Extragonadal teratomas. Armed Forces In- stitute of Pathology, Washington, DC, pp 20-24.
Graham JM Jr (1989): Personal communication. (3 unreported cases of discordant monozygotic twins).
Haas OA, Zoubek A, Grumayer ER, Gadner H (1986): Constitutional insertion deletion of 11p11 and pericentric inversion of chromosome
9 in a patient with Beckwith-Wiedemann syndrome and hepato- blastoma. Cancer Genet Cytogenet 23:95-104.
Heifetz SA, Alrabeeah A, Brown BJ, Lau H (1988): Fetus in fetu: A fetiform teratoma. Pediatr Pathol 8:215-226.
Kosseff AL, Herrmann J, Gilbert EF, Viseskul C, Lubinsky M, Opitz JM. (1976): Studies of malformation syndromes of man XXIX: The Wiedemann-Beckwith syndrome: Clinical, genetic and patho- genetic studies of 12 cases. Eur J Pediatr 123:139-166.
Koufos A, Grundy P, Margan K, Aleck KD, Hadlo T, Lampkin C, Kalbakji A, Cavnee WK (1989): Familial Wiedemann-Beckwith syndrome and a second Wilm’s tumor locus both map to 11p15.5. Am J Hum Genet 44:711-719.
Koufos A, Hansen MF, Copeland NG, Jenkins NA, Lampkin C, Cavenee WK (1985): Loss of heterozygosity in three embryonal tumors suggests a common pathogenetic mechanism. Nature 316:330-334.
Little MH, Thousan D, Hayward MNK, Smith PL (1988): Loss of alleles on the short arm of chromosome 11 in a hepatoblastoma from a child with Beckwith-Wiedemann syndrome. Hum Genet 79:186-189.
Litz CE, Taylor KA, Qiu JS, Pescovitz OH, deMartinville B (1988): Absence of detectable chromosomal and molecular abnormalities in monozygotic twins discordant for the Wiedemann-Beckwith syn- drome. Am J Med Genet 30:821-833.
Lubinsky M, Herrmann J, Kosseff AL, Opitz JM (1974): Autosomal- dominant sex-dependent transmission of the Wiedemann-Beckwith syndrome. Lancet 1:932.
Niikawa N, Ishikiriyama S, Takahashi S, Inagawa A, Tonoki H, Ohto Y, Hase N, Kameit T, Kajii T (1986): The Wiedemann-Beckwith syndrome: Pedigree studies on five families with evidence for auto- somal dominant inheritance with variable expressivity. Am J Med Genet 24:41-55.
Olney AH, Beuhler BA, Waziri M (1988): Wiedemann-Beckwith syn- drome in apparently discordant monozygotic twins. Am J Med Ge- net 29:491-499.
Pettanati NJ, Haines JL, Higgins RR, Wappner RS, Palmer CG, Weaver DD (1986): Wiedemann-Beckwith syndrome: Presentation of clinical and cytogenetic data on 22 new cases and review of the literature. Hum Genet 74:143-154.
Ping AJ, Reeve AE, Law DJ, Young MD, Boehnke M, Feinber AP (1989): Genetic linkage of Beckwith-Wiedemann syndrome to 11p15. Am J Hum Genet 44:720-723.
Piussan C, Risbourg B, Lenaerts C, Del Valle ZN, Gontier MF, Vitse M (1980): Syndrome de Wiedemann et Beckwith: Une nouvelle obser- vation familiale. J Génét Hum 28:281-291.
Searle AG, Peters J, Lyon MF, Hall JG, Evans EP, Edwards JH, Buckle VJ (1989): Chromosome maps of man and mouse. Ann Hum Genet IV:53:89-140.
Solter D (1988): Differential imprinting and expression of maternal and paternal genomes. Annu Rev Genet 22:127-146.
Sotelo-Avila C, Gooch WMN (1976): Neoplasms associated with Beck- with-Wiedemann syndrome. Perspect Pediatr Pathol 3:255-263.
Wiedemann HR (1964): Complex malformative on familial avec hernie ombilicale et macroglossie-un ‘syndrome nouveau’? J Génét Hum 13:223-232.