A Lack of Neuroblastoma in Down Syndrome: a Study from 11 European Countries1

(CANCER RESEARCH 58. 448-452. February 1. I998| A Lack of Neuroblastoma in Down Syndrome: A Study from 11 European Countries1

Daniel SatgÃ©,2Annie J. Sasco, Niels L. T. Carlsen, Charles A. Stiller, HervÃ©Rubie, Barbara Hero, Bruno de Bernardi, Jan de Kraker, Carole Coze, Per Kogner, Froydis Langmark, Friederike G. A. J. Hakvoort-Cammel, Daniel Beck,3 Nicolas von der Weid, Sheila Parkes, Olivier Hartmann, Robert J. J. Lippens, Willem A. Kamps, and Daniele Sommelet

Laboratory of Pathologv. Centre Hospitalier. 19000 Tulle. Frunce Â¡D.S.]: Epideiniologv for Cancer Prevention, 1ARC, 69372 Lyon, France Â¡A.J. S.j; Pediatrie Hematology Oncolngy. Odense University Hospital, DK-5000 Odense C, Denmark Â¡N.L T. C.Â¡; Childhood Cancer Research Group, Department of Pediatrics. Oxford OX6HJ. United Kingdom {C.A.S.f: Pediatrie Hemalologv Oneologv. MÃ©decineInfantile G, Centre Hospitalier Universitaie Purpan, 31059 Toulouse, France Â¡H.R.f: Pediatrie Hematologv Oncnlngy, Klinik fÃ¼rKinderheilkunde, D-5IXM) Cologne. Germany Â¡B.H.Â¡:Department of Hemalology Oncology. Giannina Gaslini Children's Hospital, 1614K Genova. Italy IB. d. B.I: PÃ©diatrieHetnatologv Oncologi: Emma Kinder Ziekenhuis. Academic Medical Center. 1I05AZ Amsterdam, the Netherlands [J. d. K.I: Pediatrie Oncology. University Hospital Lit Tiimme. 133N5 Marseille, France /C. C/; Childhood Cancer Research Unit. Department of Pediatrics. Karolinska Hospital. S-17176 Stockholm. Sweden Â¡P.K.j: The Cancer Registry of Nonvav. Institute for Epidemiological Research Mtmtebello. 0310 Oslo. Nonvav IF. L.j: Pediatrie Hetnatologv Oneologv: Sophia Children's Hospital, 60 3015 (ij Rotterdam, rite Netherlands Â¡F.G. A. J. H-C.j: Pediatrie Hemalology Onetilogy, Centre Hospitalier Universitaire. Vaudois, 1011 Lausanne, Switzerland Â¡D.B.Â¡;Pediatrie HeniaÃoli>gvOncologi', Inselspital, 3010 Hern. Switzerland Â¡N.v. d.W.Â¡; West Middlands Regional Children's Tumour Research Group, Birmingham Children's Hospital. Birmingham B16 RET, United Kingdom Â¡S.P.j: Pediatrie Hetnatology Oncology. Institut Gustave Roussy, Rue Camille Desmoulins, 94805 Villejuif, France Â¡O.H.]: Pediatrie Hemuiology Oncology, University Hospital Nijmegen. 6.500 HB Nijmegen. the Netherlands Â¡R.J.J.Lj: Pediairic Hematology Oncology. Beatrix Children's Hospital. 9700 RB Groningen, the Netherlands Â¡W.A. K.J: and Pediatrie Hematology Oncology, University Hospital, 54511 Nancy, France Â¡D.SJ

ABSTRACT neuroblastoma is the most frequent solid cancer found in children before the age of 5 years, accounting for 6 to 10% of all childhood An epidemiolÃ³gica! investigation in 11 European countries comprising cancers (7), we suspected that this finding could be the indication of u total childhood population of 54.1 million children and using 8 separate a lower incidence of this neoplasm in this constitutional chromosomal data sources was conducted to evaluate the occurrence of neurohlastoma anomaly. We investigated this hypothesis by studying a series of more in Down syndrome IDS). No cases of US were detected among 6724 infants and children with neurohlastoma, although more than five were expected. than 6000 cases of neuroblastoma compiled in 11 European countries. This highly significant result (P = 11.11045according to the Poisson test) is consistent with data in the literature, which contains only two poorly detailed cases in epidemiolÃ³gica! studies and one ganglioneuroma in a DS MATERIALS AND METHODS mosaic patient. Like other tumors, such as leukemias, testicular germ cell tumors and lymphomas are in excess in US patients; the lack of neuro- Cases were identified in 11 European countries. 6 of them through pediatrie blastomas does not reflect a general decreased incidence of cancer but cancer registries: the British National Registry on Childhood Tumors for Great rather a specific underrepresentation of this precise tumor. S-100 h pro Britain; the German Childhood Tumor Registry for Germany; and the Nordic tein, the gene for which maps to the long arm of chromosome 21, (a) is Childhood Cancer Registry for Finland. Iceland. Norway, and Sweden. The overproduced in DS patients, (/>) produces growth inhibition and differ data from Denmark came from the Danish Cancer Registry. In Germany, the entiation of neural cells in vitro, (c) is abundant in good-prognosis neuro- National Childhood Cancer Registry began in 1980, and coverage was less hlastomas, and u/> has been shown to induce growth inhibition and complete in the early years (8). These German data include only the former differentiation and cell death in several human and murine neuroblas- West Germany and not East Germany before reunification. In the Netherlands, toma cell lines and could be responsible for this variation. Additional cases came from the four regional pediatrie oncology centers that cover the epidemiolÃ³gica! and experimental studies are warranted to confirm our whole geographic area. The Swiss data about incidence of pediatrie cancer interpretation of these data. were provided through the network of the Swiss Pediatrie Oncology Group, which includes all (n = 8) Swiss institutions involved in the management of childhood cancer in this country. In Switzerland, this coverage is considered to INTRODUCTION be close to 100%. In France and Italy, cases were gathered from centralized sources that hold data on children with neuroblastoma who are seen in France by the members of the SociÃ©tÃ©FranÃ§aised'Oncologie PÃ©diatriqueand in Italy An association between congenital anomalies and childhood can cers is now clearly demonstrated: thus, a better knowledge of this link by the members of the Cooperative Group for Neuroblastoma. These two data sources are considered to cover 90c/c of the whole population. To gather as may help identify genes involved in cancer development ( 1). Among genetic diseases, there are several, such as DS,4 neurofibromatosis many cases as possible, we included the longest period available for each type 1, Beckwith-Wiedemann syndrome, and tuberous sclerosis, registry or centralized data source varying from 5 to 48 years. Although the upper limit for age varied according to each registry, inclusion of the age range which are associated with an excess of various neoplasms (2). As far between 14 and 18 years does not lead to any substantial problem, given the as we know, a decreased incidence of a specific neoplasm in a fact that neuroblastomas occur mainly before 10 years of age (7). Eligible cases particular genetic disease has never been reported. An updated review were those diagnosed as histologically proven peripheral neuroblastomas, of the literature on solid tumors in DS patients (3) yielded only two including, tor some registries, in situ neuroblastomas and central nervous cases of neuroblastoma (4, 5) and one ganglioneuroma (6). Because system neuroblastomas. In situ neuroblastomas are minute, incidentally en countered adrenal neuroblastomas usually found during autopsies of infants

Received 8/28/97; accepted 12/2/97. (9). In five national registries, systematic recording of the conditions associ The costs of publication of (his article were defrayed in part by the payment of page ated with each case is provided, permitting the detection of DS cases. In the charges. This article must therefore be hereby marked advertisement in accordance with others, systematic registration was not indicated, but major malformations 18 U.S.C. Section 1734 solely to indicate this fact. 1This study was supported by u grant from Ligue pour la Lutte contre le Cancer. were mentioned, and it was possible to indicate the presence or absence of DS Division CorrÃ©zienne. The Childhood Cancer Research Group was supported by the in children. Moreover, cancer treatment for a DS patient raises difficult Department of Health, the Scottish Home and Health Department, and the Cancer therapeutic problems due to enhanced drug toxicity (well documented for Research Campaign. leukemias; Ref. 10). which would have attracted the attention of clinicians and - To whom requests for reprints should be addressed, at Laboratory of Pathology. alerted them to the condition. For comparison purposes, we checked the Centre Hospitalier. 19000 Tulle. Trance. Phone: (33) 555-29-79-13; Fax: (33) 555-29- 79-31. presence of three other conditions clinically associated with neuroblastomas; 1Deceased. neurofibromatosis type 1. Hirschsprung's disease, and Beckwith-Wiedemann 4 The abbreviation used is: DS. Down syndrome. syndrome (11. 12). 448

We evaluated the DS population based on the rate at birth as indicated by in every 5,000 infants born alive births and its diagnosis is made the European Registration of Congenital Anomalies (13). which gives the mostly during the 1st year of life (18). the only case observed in this crude rates during the years 1990-1994 in the 25 registries at around 10.4 DS study does not support the idea of an excess of neuroblastoma in this per 10,000 live births (or one for 961.5 births). Because prenatal screening has condition. Taken together, these results indicate that in this population improved during the last decade (14). we think that this value underestimates of 0-18 years of age. infants and children with diseases known to be the DS population at birth for the first years of the study. Life expectancy is at risk for this nervous system neoplasm were not missed. lower in individuals with DS compared to the general population, with only 76.6% of individuals still alive at age 15. This allowed us to calculate a minimum estimate of the proportion of the population ages 0-14 years with DISCUSSION DS in the study regions (15). This study failed to detect a single case of DS among 6724 infants We used the Poisson statistic test to compare the number of cases of neuroblastoma observed in the DS population with the number expected based and children with neuroblastoma, whereas according to our estimation on general population rates. of the DS population, more than 5 cases were expected. This result is highly significant but has to be discussed regarding possible biases, RESULTS particularly the possibility of undeclared DS cases. The eventuality of unrecognized DS among neuroblastoma patients is unlikely, because Table 1 shows that no case of DS was found among 6724 cases of this syndrome is clearly visible and clinically identifiable (19) with an neuroblastoma. Taking the rate of occurrence of DS as 1 in 961.5 live easy cytogenetic verification. More subtle and rare mulformative births and the survival rate to age 15 as 76.6%, at least 5.40 cases of conditions and syndromes are recorded (12) in some registries. A DS would be expected among 6724 cases of neuroblastoma. The second possibility is the underdiagnosis and undertreatment of a Poisson probability of observing no case is 0.0045. This estimation is malignancy such as neuroblastoma in this genetic disorder. This is conservative, taking into account the fact that most neuroblastoma also unlikely: the well-known excess of leukemia argues against a cases occur before 10 years of age and that 78.4% of DS children general underdiagnosis of malignant conditions in DS patients. Fur survive that age. By allowing in our population older subjects at a very thermore, clinicians will be alerted because of the difficulties of low risk of neuroblastoma. we may underestimate the expected figure, clinical and complementary examinations due to the mental impair thereby diluting slightly the observed effect. ment of these patients (20), as well as treatment problems due to Because the incidence of neurofibromatosis type 1 is estimated at 1 enhanced drug toxicity in the case of leukemias ( 10) and solid tumors in 3,000 individuals (16), and because some cases are sometimes not (21). For these reasons, it is unlikely that a DS case would be diagnosed before 15 years of age, the 19 cases observed in the study overlooked by a team or a clinician. A third possibility is the lack of show an important excess of neuroblastomas in this condition, as registration of recognized DS cases. This does not apply to the 5386 expected. Similarly, because the incidence of Beckwith-Wiedemann cases included in the registries in which registration of associated syndrome, which is usually diagnosed in the first months of life, is malformations is systematic. This is also not likely for the other 1338 estimated to be 1 in 13,700 (17) the four cases observed in this study cases (all cases from West Midlands. United Kingdom, before 1970; show an important excess of neuroblastomas in this condition, as fewer than half of the French cases; and all cases from the Nether expected. Because Hirschsprung's disease occurs in approximately 1 lands, Finland, Iceland, Norway, and Sweden), because major prob-

Table 1 Cases of DS compared to cases of neurofibromatosis type I. Beckwith-Widemann syndrome, and Hindupnuig'l disease in a European series of neun>blu.\lnmus

range for inclusion of registration of CountryDenmarkFinlandFranceGermanyGreatcovered1943-19911985-19941990-1in theregistry0-140-140-150-150-150-140-140-140-140-150-180-180-140-140-140-140-16Numberneuroblastomas34194Â«Of171*781398I94y100310834or/724813486152214*6724SystematicassociatedconditionsYesYeÂ»YeÂ»NoNoYeÂ»YeiNoYeÂ»YeÂ«NoNoNoNoYeÂ»YesYeÂ»NF"10200161)05010200119/2.2"BW*00000010000(102(I014/0.5"HD'0(11)000100000000001/1.3"DSrf0000(10(100000000000/5.4

995'1987-1992"1992-1994"1979-19951971-19941957-1970'1985-19941979-19941972-1994'1975-1995*1982-1995'1970-1995'"1985-19941985-19941971-1995Age

BritainIcelandItalyNetherlandsNorwaySwedenSwitzerlandTotalPeriod

" Cases of neurofibromatosis type I. h Cases of Beckwith-Wiedemann syndrome. ' Cases of Hirschsprung's disease. rfCases of DS. f Locali/ed disease and metastatic disease under 1 year of age. in siiti neuroblastomas included. * Metastatic disease after 1 year of age. '' Central nervous system neuroblastomas included. ' West Middlands. J Amsterdam area. * Nijmegen area. Groningen area. '" Rotterdam area. " Cases of neurofibromatosis type I. Beckwilh-Wiedemann syndrome, and Hirschsprung's disease expected (age not adjusted). " Cases of DS expected (age adjusted). 449

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1998 American Association for Cancer Research. DOWN SYNDROME AND NEUROBLASTOMA lems with diagnosis and treatment were mentioned. In addition. Table 39). A risk has been suggested in the offspring of mothers at the 1 includes 19 cases of neurofibromatosis type 1. 1 of Hirschsprung's extremes (young or old) of their childbearing age (34) and after disease, and 4 of Beckwith-Wiedemann syndrome, indicating that exposure to sex hormones (39). Because early or late childbearing age congenital abnormalities were not being missed. A fourth possibility is a well-known risk factor for trisomy 21 (19) and because exposure would be missing a neuroblastoma in a DS child who died of some to sex hormones is a suspected risk factor for trisomy 21 (40), we other cause. This is probably also unlikely to happen, but, given the would expect a higher incidence of neuroblastoma in individuals with incidence of in xiiu neuroblastomas in some autopsy series of infants DS compared to the general population. Given the lack of strong with congenital anomalies (9). it does need to be considered. There environmental or other types of external confounding factors and fore, we believe that the results strongly suggest a decreased incidence given the major repercussions of the supplementary chromosome on of neuroblastomas in the DS population. the DS phenotype, we consider that a genetic mechanism is probably These results are in agreement with a previous search of the also involved in the rarity of neuroblastomas in individuals with DS. literature. There is only one ganglioneuroma published as a case The DS phenotype is the result of an imbalance due to an excess of report in a mosaic DS patient (6) and two neuroblastomas included in genetic material on the supernumerary chromosome 21. A parental two series of childhood tumors associated with malformations (4, 5). imprinting mechanism is theoretically possible, but it has not been These last cases are very poorly documented; one is found in an observed, at least postnatally. for chromosome 21 (41). The dysregu- epidemiolÃ³gica! series at a time when the various kinds of childhood lation of one or more genes situated on other chromosomes has been small round cell tumors were not well differentiated (4), and it is not seen in an animal model of DS (42). This implies that it is not possible possible to completely exclude a nonneuroblastoma case registered as to exclude, a priori, the role of a gene that is not situated on a neuroblastoma. Finally, after a careful manual and computer search, chromosome 21 in certain aspects of DS. However, the most probable we could not find a single well-documented neuroblastoma in a mechanism is an overproduction through gene dosage effect of pro nonmosaic DS subject. Two series of 1632 neuroblastomas (22, 23) teins, the genes for which map to chromosome 21 (43). We propose and a compilation of malformations associated with neuroblastomas that chromosome 21 may have a protective effect against neuroblas (24) contained no trisomy 21 cases. Similarly, no case of neuroblas toma by inhibiting the onset and/or the growth of the tumor. Because toma was found among 135 DS children with cancer in the British central issues in molecular cancer research include to find out whether Registry of Childhood Tumors (2), whereas this cancer accounts for particular genetic alterations can provide information that is of diag 6-10% of childhood neoplasms in the Caucasian population. Our data nosis and pronostic value (44), we think that this observation is are also interesting, because this is the first large study in a European important. It is interesting to note that three recent studies on neuro region with populations as different as those of Italy and the Nordic blastoma genomic imbalances using comparative genomic hybridiza countries. In situ neuroblastoma, the incidence of which is estimated tion showed that gain of chromosomal material is much more often to be 40-50 times higher than florid neuroblastoma (11), has been found in neuroblastoma cells than are chromosome losses (45-47). observed with various malformations (9, 25) including trisomy 13 All of these investigations found that gain of chromosome 17 material (26) and trisomy 18 (27). To our knowledge, it has never been is the most common aberration detected in 63-75% of investigated reported with trisomy 21, which is. however, much more common. cases. Moreover, extra material from chromosome 21 was not found Finally, besides trisomy 13 and 18. neuroblastoma has been observed in any investigated tumor in these independent studies, whereas the with various exceptional constitutional chromosomal anomalies (28, phenomenon of additional chromosomal material could be detected 29). On the other hand, it is well accepted that DS individuals have a for all other chromosomes in several tumors. 20-30-fold excess risk of leukemias. particularly of the M7 type (30). On the distal part of the long arm of chromosome 21 maps the They may possibly also have more lymphomas (31. 32) and testicular S-ÃŒOOÃŸgene (48), which fills many conditions for being implicated, germ cell tumors (21. 33) than the general population. Thus, the at least in part, in the lack of neuroblastomas in individuals with DS. supposed decreased incidence of neuroblastoma does not reflect a A previous work (49) discusses the following arguments more exten general decrease of cancers in individuals with DS. but rather suggests sively: S-100 b protein, the dimeric product of the gene, is secreted a possible specific phenomenon concerning this particular peripheral principally in glial cells and nonneural cells of the peripheral nervous nervous tumor. system (50) and induces differentiation of neural cells in vitro (51). A modification of a neoplasm incidence rate may result from S-100 b protein is overproduced in blood plasma (52) and nervous a genetic impairment, from environmental factors, or from a com tissues (53) in patients with DS. A comparative study of 44 DS bination of the two (34). Several genetic disorders have been de patients and 28 controls showed a clear excess of serum S-100 b scribed in association with neuroblastomas: neurofibromatosis type 1, protein (160 pg/ml) compared to controls (76 pg/ml; P < 0.001; Ref. Beckwith-Wiedemann syndrome. Hirschsprung's disease. Turner syn 52). An immunohistochemical study suggested that the amount of drome. Duchenne's muscular dystrophy, and cystic fibrosis. Further S-100 protein in astrocytes of DS brains was even greater than the more, homozygous deletion of neurofibromatosis I gene has recently 1.5-fold expected (53). This protein is abundant in the stroma of been observed in a primary human neuroblastoma (35). The first two well-differentiated neuroblastomas with good prognosis and is rare or conditions are most likely to represent risk factors, but at the present absent in undifterentiated neuroblastomas of very poor outcome (54). time, no definitive relationship between a higher risk of neuroblas Protein S-100 b induces inhibition of growth and death of several toma and a genetic disease has been proven (11, 12). Genetic factors human and murine neuroblastoma cell lines in vitro (49, 55-57). have been evoked to explain some ethnic variations in the incidence Several observations indicate an underdevelopment of the periph of neuroblastomas (36, 37). More recently, it has been proposed that eral nervous system in individuals with DS, with a reduced number of the underlying risk could well be the same in all populations and that neurons in esophageal plexus ganglions (58) and tooth agenesis, variations in recorded incidence simply reflect variations in the pro which is probably related to impaired mandibular innervation (59). portions of tumors that are diagnosed and registered (37). We are not Furthermore, autopsy studies of children with DS disclosed fairly aware of any work suggesting a lower risk of developing neuroblas hypoplastic changes of the adrenal medulla, which is considered as a toma in a particular genetic disease (38). sympathetic ganglion (60-62). It is tempting to speculate that this Several gestational factors have been considered either in isolated inhibition of growth of the nervous tissue could also result from a case reports or in analytical epidemiological studies (11, 12, 23, 34, moderate but chronic overproduction of S-100 b protein, which in- 450

Downloaded from cancerres.aacrjournals.org on September 25, 2021. © 1998 American Association for Cancer Research. DOWN SYNDROME AND NEUROBLASTOMA hibits neural cell growth and enhances differentiation, in prenatal (53, 7. Stiller. C. A., and Parkin. D. M. International variations in the incidence of neuro 63) and postnatal life (52). A unifying model would be that in blastoma. Im. J. Cancer. 52: 538-543. 1992. 8. Kaatsch. P., Haap, G., and Michaelis. J. Childhood malignancies in Germany: individuals with DS. the overproduction of S-100 b protein through a methods and results in nationwide registry. Eur. J. Cancer. 3IA: 993-999. 1995. gene dosage effect could produce both a deleterious effect on periph 9. Beckwith. J. B.. and Perrin. H. V. In situ neuroblustomas: a contribution to ine natural history of neural crest tumors. Am. J. Pathol.. 4Ã•: 1089-1104, 1963. eral nervous tissue and a beneficial effect, inhibiting the onset and/or 10. Ragab, A. H.. Abdel-Mageed. A., Shuster, J. J.. Frankel. L. S., Pullen. J.. Van Eys, J., the growth of neuroblastomas. Sullivan. M. P.. Boyett. J.. Borowit/. M.. and Crist, W. M. Clinical characteristics and Because the level of differentiation in neuroblastoma is of great treatment outcome of children with acute lymphoblastic leukemia and Down's syndrome. Cancer (Phila.). 67: 1057-1063. 1991. importance for the response to treatment (64), we think that a 11. McWilliams. N. B. Neuroblastoma in infancy, neonatal neuroblastoma and associated protein inducing neuroblastoma cells and inhibition of growth, conditions. In: C. Pocheldy (ed.), Neuroblastoma: Tumor Biology and Therapy, pp. differentiation, and death in vitro is a good candidate to explain the 230-232. Boca Raton: CRC Press, 1990. 12. Carlsen, N. L. T. Neuroblastomas in Denmark 1943-19X0. Acta Paediatr. Suppl.. 403: rarity of this tumor in a disease in which it is overproduced. 1-27, 1994. However, because cancer is a multistep process, we do not exclude 13. EUROCAT (European Registration of Congenital Anomalies). Down syndrome. In: the possibility that other factors and possibly other genes on 15 Years Surveillance of Congenital Anomalies in Europe: 1980-1994. pp. 120-149. Brussels: EUROCAT Working Group. 1997. chromosome 21 act together to prevent the development of this 14. Mutton. D. E., Alberman, E., Ide. R.. and Bobrow. M. Results of first year ( 1989) of neuroectodermal tumor in people with DS. For instance, several a national register of Down's syndrome in England and Wales. Br. Med. J., 303: 1295-1297, 1991. genes of the IFN system map to the long arm of chromosome 21, 15. Baird. P. A., and Sadovnick. A. D. Life tables for Down syndrome. Hum. Genet., 82: whereas IFN has extensive and quite diverse effects on nervous 291-292, 1989. tissues (65). If S-100 b protein is effective in decreasing the 16. Jones, K. L. Neurofibromatosis syndrome. In: Smith's Recognizable Patterns of Human Malformation, pp. 452-453. Philadelphia: W. B. Saunders. 1988. incidence of neuroblastomas in individuals with DS, it could be 17. Schneid. H., Vazquez, M. P.. Vacher, C.. Gourmelen. M.. Cabro!. S., and Le Bouc. very interesting to see, at least in some patients with sporadic Y. The Beckwith-Wiedemann syndrome phenolype and the risk of cancer. Med. neuroblastoma, whether there was a quantitative or qualitative Pediatr. Oncol.. 28: 411-415. 1997. 18. Rescorla, F. J.. Morrison, A. M.. Engles. D., West, K. W., and Grosfeld. J. !.. anomaly of this protein or its biochemical pathways. Also, we Hirschsprung's disease: evaluation of mortality and long-term function in 260 cases. wonder whether S-100 b protein could not be used as a differen Arch. Surg.. 127: 934-942. 1992. 19. A Report from the International Clearinghouse for Birth Defects Monitoring Systems: tiation inducer to treat neuroblastomas and even to prevent the Down's Syndrome. In: Congenital Malformations Worldwide, pp. 157-162. Amster onset of neuroblastomas in conditions at risk, when they will be dam: Elsevier, 1991. known. 20. SatgÃ©,D.. Le Toumeau. A.. Verger. J. P.. Lefort, S., Geneix, A., Malet. P., Diebold. J.. and Vekemans, M. Down syndrome and cenlroblaslic lymphoma. Palhol. Res. We conclude that this large study clearly demonstrates the lack of Pracl., 192: 1266-1269, 1996. neuroblastoma in people with DS. This is in full agreement with the 21. Dexeus, F. H.. Logothetis. C. J.. Chong. C.. Sella. A., and Ogden. S. Genetic literature research we conducted. To our knowledge, this is the first abnormalities in men with germ cell tumors. J. Urol., 140: 80-84. 1988. 22. Miller. R. W., Fraumeni, J. F.. and Hill. J. A. Neuroblastoma: epidemiologie approach time that a genetic disease has been linked with the lack of a precise to its origin. Am. J. Dis. Child.. 115: 253-261. 1968. tumor. This model may be the paradigm of a particular approach of 23. Negus, J. P., Smithson. W. A.. Gunderson. P.. King. F. L.. Singher. L. J., and Robison. L. L. Prenatal and perinatal risk factors for neuroblastoma. Cancer (Phila.), the decreased incidence of some neoplasms in genetic diseases. It can fil: 2202-2206, 1988. be applied to other tumors that are underrepresented in those with DS 24. Sy, W. M., and Edmonson, J. H. The developmental defects associated with neuro and to other genetic diseases. More specifically, our results are in blastoma: enologie implications. Cancer (Phila.), 22: 234-238. 1968. 25. Shanklin. D. R.. and Sotelo-Avila. C. In \itu tumors in fetuses, newborns and young favor of a protective effect of chromosome 21 against neuroblastoma infants. Biol. Neonal., 14: 286-316. 1969. development, as supported by recent genetic studies using compara 26. Feingold, M., Gheradi, G. J.. and Simons, C. Familial neuroblasioma and trisomy 13. tive genomic hybridization. However, at the present time, to confirm Am. J. Dis. Child., 121: 451, 1971. 27. Robinson. M. G.. and McCorquodale. M. M. Trisomy 18 and neurogenic neoplasia. these data and our interpretation, more epidemiological and experi J. Pediatr. 99: 428-429. 1981. mental studies are needed, such as genetic transfer involving whole 28. Fryns. J. P.. Moerman. P.. Kleczkowska, A., and Van den Berghe. H. Neuroblasioma chromosome 21, chromosomal regions, or expression vectors for and partial I5q duplication in the fetus. Ann. Genet.. 37: 42-43, 1994. 29. Qureshi. F.. Jacques. S. M.. Johnson. M. P.. Reichler. A., and Evans. M. I. Micro specific genes, including the gene S-100 ÃŸ. scopic neuroblastoma in a fetus with a tie /irmi unbalanced translocalion 3:10. Am. J. Med. Genet., 53: 24-28, 1994. 30. Power, L. W.. and Register. M. K. Down syndrome and acute leukemia: epidemio ACKNOWLEDGMENTS logical and genetic relationships. Lab. Med.. 22: 630-636, 1991. 31. Barber. R., and Spiers, P. Oxford survey of childhood cancers: progress report II. We gratefully acknowledge Beatrice Bezamat. Nathalie Villeyras. and Su- Mon. Bull. Minist. Health Lond.. 23: 46-52, 1964. 32. Scholl. T., Stein. Z., and Hansen. H. Leukemia and other cancers, anomalies and zan Haver for secretarial assistance. We thank every contributor not included infections as causes of death in Down's syndrome in the United Slates during 1976. in the author list who kindly provided additional information concerning cases Dev. Med. Child. Neurol.. 24: 817-829, 1982. recorded and members of the different registries for use of the data on 33. SatgÃ©,D.,Sasco, A. J., CurÃ©,H..Leduc. B.. Sommelet. D., and Vekemans. M. J. An excess of teslicular germ cell tumors in Down's syndrome: three case reports and a neuroblastoma. review of the literature. Cancer (Phila.), 80: 929-935, 1997. 34. Carlsen. N. L. T. Neuroblasiomas presenting in the first year of life: epidemiological REFERENCES differences from those presenting al older ages. Cancer Delecl. Prev., 20: 251-261, 1996. 1. Narod. S. A.. Hawkins. M. M. Robertson, C. M.. and Stiller. C. A. Congenital 35. Martinsson. T., SjÃ¶berg.R-M.. Hedborg. F.. and Rogner. P. Hormwygous deletion of anomalies and childhood cancer in Great Britain. Am. J. Hum. Genet.. 60: 474-485. the neurofibromatosis-1 gene in the tumor of a patient with neuroblastoma. Cancer 1997. Genet. Cytogenet., 95: 183-189, 1997. 2. Narod. S. A.. Stiller, C. A., and Lenoir. G. M. An estimate of the heritable fraction 36. Miller. R. W. Rarity of neuroblastoma in East Africa. Lancet. 335: 659-660. 1990. of childhood cancer. Br. J. Cancer. 63: 993-999. 1991. 37. Stiller, C. A., and Parkin, D. M. Geographic and ethnic variations in the incidence of 3. Satge. D., Sommelet. D-, Geneix. A.. Malet. P., and Vekemans. M. Tumoral Profile childhood cancer. Br. Med. Bull.. 52: 682-703. 1996. in Down syndrome: a step towards the understanding of the consequences of aneu- 38. Brodeur. G. M., and Fong, C. T. Molecular biology and genetics of human neuro ploidy and the development of cancer. Am. J. Hum. Genet.. 55 (Suppl.l: A303. 1994. blastoma. Cancer Genet. Cytogenet.. 41: 153-174. 1989. 4. Miller. R. W. Childhood cancer and congenital defects: a study of LI. S. death 39. Mandel. M., Toren. A.. Rechavi, G., Dor. J.. Ben-Bassat, I., and Neumann. Y. certificates during the period 1960-1966. Pediatr. Res.. 3: 389-397. 1969. Hormonal treatment in pregnancy: a possible risk factor for neuroblastoma. Med. 5. Foulkes, W. D.. Buu, P. N.. Filiatrault. D.. Ledere. J-M.. and Narod. S. A. Excess of Pediatr. Oncol.. 23: 133-135. 1994. congenital abnormalities in French-Canadian children with neuroblastoma: A case 40. Janerich. D. T.. and Bracken. M. B. Epidemiology of trisomy 21: a review and series study from Montreal. Med. Pediat. Oncol.. 29: 272-279. 1997. theoretical analysis. J. Chron. Dis., 39: 1079-1093, 1986. 6. Hosoi. H., Sawada. T.. Nakajima. T.. Kuroda. H., Saida. T.. Sugimoto. T., Tokiwa. 41. Sloll. C., Alembik, Y.. Doit. B.. and Feingold, J. No evidence for genomic imprinting K., and Ogita, S. A case of mosaic Down's syndrome concomitant with ganglioneu- in liveborn Down syndrome patients. Ann. Genet.. 3K: 13-18, 1995. roma. J. Pediatr. Surg., 24: 210-211, 1989. 42. Holtzman, D. M.. Bayney. R. M.. Li. Y.. Khosrovi. H.. Berger. C. N., Epstein, C. J., 451

and Mobley. W. C. Dysregulation of gene expression in mouse trisomy 16. an animal White. C. L.. and Araos. C. Brain interleukin I and S-IOO immunoreactivity are model of Down syndrome. EMBO J.. //. 619-627, 1992. elevated in Down syndrome and Alzheimer disease. Proc. Nati. Acad. Sci. USA, 86: 43. Epstein. C. J. Specificity wr.v/i.v nonspecificity in the palhogenesis of aneuploid 7611-7615. 1989. phenotypes. Am. J. Med. Genet.. 29: 161-165. 1988. 54. Nagoshi. M.. Tsuneyoshi. M.. and Enjoji. M. S-KK) positive undifferentiated neuro- 44. Schwab. M. Is there a neuroblastoma ami-oncogene? In: A. E. Evans. G. J. D'Angio. hlastomas with a special reference to the tumor stroma related to favorable prognosis. A. G. Knudson. Jr., and R. Seeger (eds.). Advances in Neuroblastoma Research. Vol. Pathol. Res. Pract.. Ititi: 273-283. 1992. 3. pp. 1-9. New York: Wiley-Liss. 1991. 55. Kligman. D.. and Hsieh. L. S. Neunte extension factor induces rapid morphological 45. Planta?. D.. Mohapatra. G.. Maithay. K. K.. Pellarin. M.. Seeger. R. C., and Feuer differentiation of mouse neuroblastoma cells in defined medium. Dev. Brain Res.. .Ã•.Ã•: stein. B. Ci. Gain of chromosome 17 is the most frequent abnormality detected in 296-300. 1987. neuroblastoma by comparative genomic hybridisation. Am. J. Pathol.. 150: 81-89, 56. Selinfreund. R. H.. Barger. S. W.. Pledger, W. J.. and Van Eldik. L. J. Neurotrophic 1997. protein S100 ÃŸstimulates glial cell proliferation. Proc. Nati. Acad. Sci. USA, 88: 46. Lastowska. M.. Nacheva, E., McGuckin. A.. Curtis. A.. Grace. C.. Pearson, A., and 3554-3558. 1991. Bown. N. Comparative genomic hybridi/ation study of primary neuroblastoma tu 57. Fano. G., Mariggio, M. A., Angelella, P., Nicoletti, !.. AmÃ³nica, A., FÃ¼lle,S., and mors. Genes Chromosomes Cancer, /#: 162-169. 1997. Calissano. P. The S-KK) protein causes an increase of intracellular calcium and death 47. Brinkschmidt. C.. Christiansen. H.. Terpe, H. J.. Simon. R.. Boecker. W.. Lampert. F.. of PC 12 cells. Neuroscience. S3: 919-925, 1993. and Stoerkel. S. Comparative genomic hybridi/.alion (CGH) analysis of neuroblasto- 58. Nakazato. Y.. and Landing. B. H. Reduced number of neurons in esophageal plexus mas: an important methodological approach in paediatric tumour pathology. ganglia in Down syndrome: additional evidence for reduced cell number as a basic J. Pathol.. INI: 394-400, 1997. feature of the disorder. Pedialr. Pathol.. 5. 55-63. 1986. 48. Allore. R.. O'Hanlon, D.. Price, R.. Neilson. K.. Willard. H. F., Cox, D. R.. Marks. 59. RÃ¼ssel.B. G.. Kjaer. I. Tooth agenesis in Down syndrome. Am. J. Med. Genet.. 55: A., and Dunn. R. J. Gene encoding the ÃŸsubunit ufSIOO protein is on chromosome 466-471. 1995. 21: implications for Down syndrome. Science (Washington DC). 23V: 1311-1313, 60. Lange. P. Beitrage zur pathologischen Anatomie des Mongolismus. Monatsschr. 1988. Kinderheilkde.. 5: 233-243. 1906. 49. SalgÃ©.D.A decreased incidence of neuroblastomas in Down syndrome and overpro 61. Gordon. M. B. Morphological changes in the endocrine glands in mongolian idiocy duction of S-KK) b protein: an hypothesis. Med. Hypotheses. 46: 393-399. 1996. with report of two cases. Endocrinology, 14: 1-6, 1930. 50. Donato. R. Perspectives in S-HK) protein biology. Cell Calcium. 12: 713-726. 1991. 62. Benda. C. E. The adrenals in mongolism. In: The Child with Mongolism. pp. 51. Winningham-Major. F.. Slaecker. J. L.. Barger. S. W., Coats, S., and Van Eldik, L. J. 136-139. New York: GruÃ±e& Stratton, Inc.. I960. Neurile extension and neuronal survival activities of recombinant SI(K) fi proteins that 63. Mito. T.. and Becker. L. E. Developmental changes of S-KM) protein and glial differ in the content and position of cysteine residues. J. Cell. Biol.. KM: 3063-3071, fibrillary acidic protein in the brain in Down syndrome. Exp. Neurol.. 120: 170-176, 1989. 1993. 52. Kalo, K., Suzuki, F.. Kurobe, N.. Okajima. K.. Ogasawara, N.. Nagaya. M., and 64. Israel. M. A. Disordered differentiation as a target for novel approaches to the Yamanaka. T. Enhancement of S-KX) ÃŸprotein in blood of patients with Down's treatment of neuroblastoma. Cancer (Phila.). 71: 3310-3313, 1993. syndrome. J. Mol. Neurosci.. 2: 109-113. 1990. 65. Maroun. L. E. Interferon action and chromosome 21 trisomy (Down syndrome) 15 53. Griffin. W. S. T.. Stanley. L. C.. Ling. C.. White. L. MacLeod. V.. Perrot. L. J.. years later. J. Theor. Biol.. 181: 41-46. 1996.

452

Daniel Satgé, Annie J. Sasco, Niels L. T. Carlsen, et al.

Cancer Res 1998;58:448-452.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/58/3/448

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/58/3/448. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.