- Email: [email protected]
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia JET BLIEK, BA-AS, SASKIA MAAS, MD, MARIEL ALDERS, PHD, JOHANNES H. M. MERKS, MD, PHD,
AND
MARCEL MANNENS, PHD
Objective To investigate whether epigenotyping of patients with isolated hemihyperplasia (IH) can, analogous to genetic screening of patients with Beckwith-Wiedemann syndrome, be used for the prediction of tumor risk and tumor type of individual patients. Study design Methylation analysis of H19 and KCNQ1OT1 of 73 patients. Questionnaires were sent to referring clinicians. Results In 75% of the clinically confirmed patients with IH no epigenetic defect was detected. Paternal uniparental disomy was found in 15%, demethylation of KCNQ1OT1 in only 6%, and hypermethylation of H19 in 3% of isolated hemihyperplasia cases. Ten percent of the patients with IH had development of a childhood tumor associated with paternal uniparental disomy (2/8) or no methylation defect (2/30). No genetic defect was detected in 10 of 14 additional patients with cancer with IH. In these latter patients, a methylation defect of H19 was seen 3 times and a paternal uniparental disomy once. The female-to-male ratio was 6:1. Conclusions Aberrant methylation of the 11p15 region is not common in patients with IH and can at present not be used for tumor risk determination. (J Pediatr 2008;153:95-100) emihyperplasia, also called hemihypertrophy (HH), is a congenital abnormality characterized by asymmetric growth of the limbs, trunk, face, or the entire body. HH has a heterogeneous cause: in its isolated form it is called idiopathic or isolated hemihyperplasia (IH [MIM 235000]),1 but HH is often associated with various syndromes such as BeckwithWiedemann syndrome (BWS [MIM130650]),1 Klippel-Trenaunay-Weber syndrome (KTWS [MIM149000]),1 Silver-Russell syndrome (SRS [MIM 180860]), or Proteus syndrome (MIM 176920).1 The incidence of IH has been reported to be approximately 1:13,000 to 1:86,000 live births.2 A genetic cause of IH was first described by Martin et al3 in 2005. In 8/27 (30%) patients with IH, abnormal methylation profiles were found in 2 genes (KCNQ1OT1 and H19) in 2 distinct imprinted regions on chromosome 11p15. Genes that are subject to imprinting are expressed from 1 chromosome only. The gene on the homologous chromosome is silenced by methylation of CpG nucleotides. KCNQ1OT1 and H19 are both antisense transcripts; they do not code for a protein but regulate expression of other genes nearby. One of the genes regulated by these transcripts is IGF2, a growth factor expressed only in the embryo. Mice overexpressing IGF2 have a large birth weight.4 KCNQ1OT1 and H19 are imprinted in the opposite direction: KCNQ1OT1 is methylated on the maternal allele and expressed from the paternal allele, whereas H19 is methylated on the paternal allele and expressed from the maternal allele. Patients with both IH and BWS have an increased risk for the development of From the Department of Clinical Genetics (J.B., S.M., M.A., M.M.), the Department of childhood tumors compared with the incidence in the general population (0.17%). In Paediatrics (S.M.), and the Department of patients with IH, a total of 10 tumors developed in 9 patients (5.9%), 4 unilateral Wilms Paediatric Oncology, Emma Children’s Hospital (J.M.), Academic Medical Centre, tumors (WT), 2 bilateral WTs, 2 adrenal cell carcinomas, 1 hepatoblastoma, and 1 Amsterdam, the Netherlands. leiomyosarcoma(i). All tumors developed before the age of 5 years. Submitted for publication Mar 31, 2007; 5 Patients with BWS have an increased tumor risk of 7.5% to 10%. In a review of 278 last revision received Oct 26, 2007; ac6 cepted Dec 10, 2007. patients with BWS, we demonstrated that the different genetic subgroups differ in their Reprint requests: Jet Bliek BA-AS, Acarelative tumor risk and the type of tumor that develops.
H
BWS BWS⫹HH BWS-HH HH IH
Beckwith Wiedemann syndrome BWS patients with HH BWS patients without HH Hemihypertrophy Isolated hemihypertrophy
KTWS pUPD SRS WT
Klippel-Trenaunay-Weber syndrome Paternal uniparental disomy Silver Russell syndrome Wilms tumor
demic Medical Centre, Department of Clinical Genetics, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands. E-mail: j. [email protected]. 0022-3476/$ - see front matter Copyright © 2008 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2007.12.022
95
Patients with BWS with a methylation defect of H19 have a very high risk for development of a childhood tumor, and these are uniquely Wilms tumors. Patients with a methylation defect of KCNQ1OT1 have a low tumor risk, and Wilms tumors never develop in these patients. Patients with paternal uniparental disomy (pUPD) and without detectable defect on chromosome 11p15 have an intermediate tumor risk; both Wilms’ tumors and other childhood tumor are found in these patients. Epigenotyping can therefore be used to estimate the tumor risk and predict the tumor type of individual patients with BWS. The question arises whether epigenotyping can concordantly be used for the delineation of the tumor risk of individual patients with IH. Can patients with IH be considered as patients with BWS with a mild phenotype and is it possible to determine the tumor risk of individual patients with IH on the same epigenetic basis as patients with BWS? IH might, however, be a distinct phenotype with a distinct distribution of tumor risk and tumor type over the different epigenetic subgroups. At last, IH might (in part) be caused by completely different, hitherto unknown, genetic defects. To answer these questions, we collected clinical and epigenetic data on 74 patients referred to our DNA diagnostic laboratory for methylation screening.
METHODS Patients This study included individuals with IH who were referred for clinical assessment or molecular studies by pediatricians or clinical geneticists in the Netherlands. The study was approved by the medical ethics committee of our hospital. Either all patients were examined by a clinical geneticist, or a checklist with clinical findings possibly related to IH was completed by referring pediatricians and clinical geneticists for each patient. Blood samples from all patients were sent to our laboratory for molecular analysis. Two independent clinical geneticists with expertise in evaluating children with syndromes reviewed all records. Classification of IH was independent from the results of genetic analysis. When possible, updates were collected on the development of childhood tumors. None of these patients were born after in vitro fertilization procedures. To compare the epigenetic distribution of patients with BWS and IH, we used methylation data of 110 patients with BWS who were referred to our laboratory for diagnostic testing. All patients met the clinical criteria for BWS. The patients were grouped according to the presence of HH, BWS⫹HH (BWS patients with HH), and BWS⫺HH (BWS patients without HH). In the study on malformation syndromes found among patients with cancer performed by Merks et al,7 14 patients presented with HH. Methylation analysis was performed in our laboratory on 13 patients. Three patients (HT1, HT7, and HT9 in Appendix 2; available at www.jpeds.com) were referred to our laboratory for diagnostic testing before the start of their study. One additional patient was included after 96
Bliek et al
the publication of the study (HT 14). In total we present methylation data on 14 patients with cancer with IH. Detailed clinical data of these patients are presented here, in combination with results of epigenetic profiling.
Sample Analysis Methylation analysis was performed with standard Southern blotting, and pUPD was analyzed with CA repeat markers as described previously.8 We compared the distribution of patients over the 3 epigenetic profiles (pUPD, hypermethylation of H19, and demethylation of KCNQ1OT1) and the female-to-male ratio by use of the 2 test (P ⬍ .05 was considered statistically significant).
RESULTS Clinical data were available for 73 patients with HH that were referred to our hospital for diagnostic testing. Detailed clinical data are listed in Appendix 1 (available at www.jpeds.com) and summarized in Table I. Forty-seven patients were classified as patients with IH. In 16 of these patients, a maximum of 1 finding typical for BWS was present. Birth weight was increased in 8 patients. Nephromegaly was present in 4 cases. In all cases the enlarged kidney was present in the hyperplastic side of the body. One patient presented with nephroblastomatosis at the age of 1 year. Three patients had nevus flammeus (6%). One patient was diagnosed with transient neonatal hypoglycemia. Three patients showed anomalies in addition to HH, indicating the presence of a syndrome other than BWS. Six patients showed hypomethylation of H19. The hyperplasia was reclassified as hypoplasia as part of the SRS. These patients were described elsewhere.9 Fourteen patients presented more than 1 additional finding typical for BWS but did not meet the criteria described by DeBaun and Tucker.10 Three patients displayed other unexplained anomalies; they were classified as HH⫹ patients. Of the control population of 110 patients with BWS, 54 patients had HH (BWS⫹HH), and 56 patients showed no HH (BWS⫺HH). The epigenetic profiles of the 47 patients with IH in this study are summarized in Table I. Thirty-five patients (74%) showed no epigenetic defect on chromosome 11p15. In the group with epigenetic alterations, pUPD 11p15 was the most frequently observed defect. In our study of 41 patients with an age of more than 5 years, 4 patients (10%) had a tumor develop. Detailed clinical data of 14 patients with cancer with IH described by Merks et al7 are presented in Appendix 2 (available at www.jpeds.com) and summarized in Table II. One patient exhibits 1 minor feature associated with BWS, inguinal hernia, but did not meet the BWS criteria. One patient was mentally retarded. No specific association, apart from IH, between a phenotypic abnormality and tumor development could be detected. We established the methylation status of these patients (Table I). Most tumors were found in patients with no methylation defect. The Journal of Pediatrics • July 2008
Table I. Overview of epigenetic data and clinical classification of 74 patients with HH and 14 patients with cancer with IH Patients HH⫹ SRS BWS⫺ Other syndrome IH (percentage) Tumors in this study* Control BWS population (percentage) Epigenetic distribution of cancer patients with hemihyperplasia No of cancer patients Tumor type in cancer patients with IH
Total
pUPD (H191 and KCNQ1OT12)
BWSIC1 (H191)
BWSIC2 (KCNQ1OT12)
No methylation defect
0
0
3
3 0 1 (2%) (0/1)
3 6 14 3 47 4/41 (10%)
2 0 8 (17%) 2 ⫻ WT (2/8)
0 6 H192 1 0 3 (6%) (0/2)
110
29 (26%)
11 (10%)
51 (46%)
3 (21%) 2 ⫻ WT 1 ⫻ RMS
0 (0%) 0
8 3 35 (74%) 1 ⫻ RMS 1 ⫻ NB (2/30) 19 (17%)
14 14 14
1 (7%) 1 ⫻ bWT
7 1 1 1
10 (72%) ⫻ WT ⫻ RMS ⫻ NB ⫻ sacrococcygeal teratoma
BWSIC1, Beckwith Wiedemann Imprinting Centre 1; BWSIC2, Beckwith Wiedemann Imprinting Centre 2; HH⫹, hemihyperplasia with other, not BWS related, features; BWS⫺, IH patients with one or more additional signs of BWS but not meeting the clinical criteria; bWT, bilateral Wilms tumor. *Data available on 41 patients older than 5 years.
Of the patients with IH in this study, the female-to-male ratio was 1.14:1 (25 females, 22 males). This is not significantly different from the normal population (P ⫽ .66). In contrast, in the group of patients with cancer with IH there is a significantly (P ⫽ .008) increased female-to-male ratio of 6:1 (12 females, 2 males). In the 1073 patients with cancer in the study by Merks et al,7 the female-to-male ratio was 0.88:1. The side of the hyperplasia was known in 40 patients with IH who were referred to our hospital for epigenetic screening. There was an equal ratio between right and left sidedness (20:20). Of the patients with cancer with IH, the side where the tumor developed and the hyperplastic side were known in all cases. In this patient group, the sidedness of the HH is slightly in favor of the left side (8:5). One might expect the tumor to develop on the HH side of the body. This is not the case; of the 9 unilateral WTs, 4 developed ipsilateral and 4 developed contralateral (one patient with WT showed HH of the right breast and left leg). A degree of hyperplasia is difficult to establish. The difference between normal and hyperplastic body regions varies during life, and no data were available on differences in
circumference or weight of the affected versus the unaffected tissues. An indication of the extent of overgrowth is the number of body parts involved. Of 41 patients with IH, details were available on the body parts involved; in 17 patients only one extremity was hyperplastic, and in 24 patients more than 1 body region was involved. This did not differ significantly (P ⫽ .68) from the patients with cancer with IH, whereas in 6 patients only 1 extremity was involved, and in 8 patients more than 1 body region was affected.
DISCUSSION HH can be considered a mosaic form of overgrowth and might be caused by mosaic genetic defects. Indeed, pUPD is always found in a mosaic form, and enlarged organs consist of a larger proportion of disomic cells. Methylation patterns, for diagnostic purposes, are established in blood lymphocytes and not in the hyperplastic tissues. Shuman et al11 postulate that in a part of the unexplained IH cases, pUPD is present only in hyperplastic tissue and therefore not detected in blood by routine diagnostic testing. Recently, methylation patterns in
Table II. Female-to-male distribution and sidedness of hemihyperplasia and tumor Patients
F:M
WT tumor*
L>R
R>L†
1 extremity
>1 extremity
IH Patients for diagnostic testing Cancer patients with IH
25:22 12:2
– 4 ipsilateral 4 contralateral
20 8
20 5
17 7
24 8
L, left; R, right; F, female; M, male. *Two patients had development of a hepatoblastoma, 1 patient showed contralateral hyperplasia, 1 had development of a bilateral tumor, and 2 patients had development of a midline tumor. †One patient showed contralateral hyperplasia.
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
97
Table III. Overview of epigenetic data of patients with IH and patients with cancer with IH (included data from literature)3,11 Total IH Patients for diagnostic testing IH patients in this study IH patients (combined with literature) IH Patients with a methylation defect only IH patients (combined with literature) Control BWS⫹ HH patients Control BWS⫺HH patients Tumor risk of IH patients Tumors in IH patients in this study Tumors in IH patients (combined with literature) Epigenotype and tumor type of cancer patients with IH Cancer patients in this study 14 1 ⫻ WT Cancer patients (combined with literature)
pUPD (H191 and KCNQ1OT12)
BWSIC1 (H191)
BWSIC2 (KCNQ1OT12)
No methylation defect
47 125
8 (17%) 19 (15%)
3 (6%) 4 (3%)
1 (2%) 8 (6%)
31
19 (61%)
4 (13%)
8 (25%)
—
45 46
18 (40%) 11 (24%)
7 (16%) 4 (9%)
20 (44%) 31 (67%)
— —
4/41 (10%) 14/119 (12%)
14 29
35 (74%) 94 (75%)
2 ⫻ WT (2/8) 1 ⫻ RMS, 1 ⫻ 6/19 (32%)
(0/2)
(0/1)
NBL (2/30)
0/3 (0%)
0/8 (0%)
9/89 (10%)
1 ⫻ WT 1 ⫻ RMS 3 (10%)
2 ⫻ WT 1 ⫻ RMS 4 (14%)
0
8 ⫻ WT 1 ⫻ NBL 22 (75%)
0
BWSIC1, Beckwith Wiedemann Imprinting Centre 1; BWSIC2, Beckwith Wiedemann Imprinting Centre 2; BWT, bilateral Wilms tumor.
different tissues of a patient with BWS were studied by Itoh et al.12 They showed that the proportion of cells with pUPD correlates with organ enlargement in BWS. However, pUPD could also be detected in a lower percentage in blood lymphocytes and would therefore also have been diagnosed by routine testing. In the study by Itoh et al,13 the degree of disomic cells was high (40% to 50%). When pUPD is present in only a small percentage of cells, it might have been missed in lymphocytes. Analogous to pUPD, isolated methylation defects of H19, or KCNQ1OT1 found in patients with IH are never complete. The defect is present in a mosaic form. It is possible that the proportion of aberrantly methylated cells is higher in enlarged organs. It is difficult to obtain different tissues from children with hyperplasia, and, until recently, the detection of small aberrations in methylation patterns still required large amounts of DNA. Another possible explanation for absence of methylation defects in most patients with IH might be that HH is associated with a variety of syndromes and that patients with IH present a mild form of any of these syndromes. The phenotype of BWS is very variable, and only patients with multiple features meet the criteria for BWS. Consequently, part of the patients with IH might be regarded as having a mild form of BWS. In these patients, a distribution of the methylation defects similar to BWS would be expected. Furthermore, in 20% of patients with BWS, the genetic defect is unknown. Because the distribution of epigenetic changes in patients with IH and BWS is quite different, the mechanism 98
Bliek et al
causing BWS in these latter patients might also be responsible for IH in a larger part of the patients with IH. However, HH is also found in other syndromes, such as Klippel-Trenaunay-Weber syndrome (KTWS) and Proteus syndrome, and IH might also represent mild forms of these syndromes. The exact genetic defect of KTWS is unknown, but mutations in PTEN have been described in some patients with Proteus syndrome. Thirteen of 14 patients with cancer with IH in this study have been screened for mutations in PTEN, but no mutations were found. Six patients with IH presenting with prenatal overgrowth have been screened for GPC3 mutations; no mutations were found (data not shown). Among patients with BWS, demethylation of KCNQ1OT1 is the most common genetic aberration found. This is not the case among patients with IH. Methylation profiles of patients with IH and BWS with (BWS⫹HH) and without IH (BWS⫺HH) are compared in Table III. The most common genetic defect found among patients with IH is pUPD. It was found in 61% of patients with a genetic defect. In contrast, the most common genetic defect found among patients with BWS is demethylation of KCNQ1OT1, which is found in 44% of patients with BWS with HH and in 67% of patients with BWS without HH. There was no significant difference (P ⫽ .214) between the genetic distribution of patients with IH and patients with BWS and HH. The population with IH, however, differed significantly from the population with BWS-HH (P ⫽ .002) because of overrepresentation of patients with pUPD in the group with IH The Journal of Pediatrics • July 2008
Figure. A, Epigenotype and tumor risk of 278 patients with BWS. B, Epigenotype and tumor risk of 125 patients with IH.
and a shift toward KCNQ1OT1 demethylation among the patients with BWS-HH. This confirms previous reports13,14 that pUPD is associated with hemihyperplasia. Analogous to BWS, patients with IH have an increased risk for development of childhood tumors. In 2005 we reviewed epigenetic and tumor data in 278 patients with BWS described in the literature.6 Data are summarized in the Figure, A. The patient populations in the reviewed articles all show a bias because of overrepresentation of patients with cancer for research purposes. The overall tumor risk in the review is 26%, compared with the estimated risk of 7.5% to 10% for patients with BWS. The tumor risk of patients with BWS with KCNQ1OT demethylation is relatively low. WT has never been detected in this group. Patients with pUPD represent 20% of patients with BWS and have a tumor risk of 36%; both WT and other childhood tumors are found. Patients showing hypermethylation of H19 have the highest risk for development of cancer. They represent 10% of the patients with BWS and have a tumor risk of 50%; only WT is found among these patients. The remaining 20% have no detectable methylation defect but still have an increased tumor risk of 22%; both WT and other childhood tumors are found.
On the basis of these data, it is possible to delineate tumor risk for the different epigenetic subgroups. Scott et al15 propose preventive screening of all patients with a tumor risk higher than 5%. Patients with isolated demethylation of KCNQ1OT1, the largest epigenetic subgroup representing 50% of patients with BWS, are therefore exempted from preventive screening. Data on tumor risk in different epigenetic subgroups in patients with IH are less abundantly available. We reviewed the literature and combined our data with the data presented by Martin et al3 and Shuman et al11 on a total of 125 patients with IH in a first attempt to delineate the tumor risk in patients with IH. In our study, 41/47 patients with IH were more than 5 years old; of these, 4 (10%) had a tumor develop. This is comparable to the risk that is found in the study by Hoyme et al.2 Two patients with a tumor showed pUPD (2/8); they had development of a WT. Two other patients with a tumor did not show a methylation defect (2/30); one had a rhabdomyosarcoma develop and one a neuroblastoma. No tumors were found in the patients with hypermethylation of H19 (0/2) and demethylation of KCNQ1OT1 (0/1). In the study by Martin et al,3 methylation data were presented on 27 patients with IH. One patient had a tumor develop. This patient showed no methylation defect, but the tumor type is unknown (Martin RA, personal communication, 2005). Shuman et al11 presented data on 51 patients with IH, of which 10 (20%) had a tumor develop. Four patients with a pUPD (4/8) had development of a tumor (3 hepatoblastomas, 1 WT). Six patients with no detectable defect (6/40) had a tumor develop (3 WT, 1 NB, 1 adrenal cell carcinomas). There were no patients with hypermethylation of H19, and no tumors were found among patients with demethylation of KCNQ1OT1 (0/3). Data from our study and the literature are combined and shown in the Figure, B. Including the study by Niemitz et al,16 data are available on 29 patients with cancer with IH. Comparable with the general IH cohort, most showed no methylation defects on chromosome 11p15. Four of 29 patients showed hypermethylation of H19. This demonstrates that patients with this methylation profile can develop tumors, although in the general IH cohort none of the IH patients with this profile developed a tumor. Demethylation of KCNQ1OT1 was not found among patients with cancer, but this defect is found in only a minority of patients with IH (6% [8/125]). A remarkable finding is the female overrepresentation in our cancer cohort (12:2), for which we have no explanation. We would like to acknowledge K. v.d. Lip and S. de Leng for excellent technical assistance, J. M. Cobben, MD, PhD, for evaluation of the clinical records of the patients, and Prof. N. J. Leschot and Prof. A. Westerveld for critically reviewing this manuscript.
REFERENCES 1. OMIM Online Mendelian Inheritance in Man, John Hopkins University, available at http://ncbi.nlm.nih.gov. Accessed February 12, 2008.
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
99
2. Hoyme HE, Seaver LH, Jones KL, Procopio F, Crooks W, Feingold M. Isolated hemihyperplasia (hemihypertrophy): report of a prospective multicenter study of the incidence of neoplasia and review. Am J Med Genet 1998;79:274-8. 3. Martin RA, Grange DK, Zehnbauer B, DeBaun MR. LIT1 and H19 methylation defects in isolated hemihyperplasia. Am J Med Genet A 2005;134:129-31. 4. DeChiara TM, Efstratiadis A, Robertson EJ. A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature 1990;345:78-80. 5. Wiedemann HR. Tumours and hemihypertrophy associated with WiedemannBeckwith syndrome. 141 ed. 1983. p. 129. 6. Bliek J, Gicquel C, Maas S, Gaston V, Le BY, Mannens M. Epigenotyping as a tool for the prediction of tumor risk and tumor type in patients with BeckwithWiedemann syndrome (BWS). J Pediatr 2004;145:796-9. 7. Merks JH, Caron HN, Hennekam RC. High incidence of malformation syndromes in a series of 1,073 children with cancer. Am J Med Genet A 2005;134:132-43. 8. Bliek J, Maas SM, Ruijter JM, Hennekam RC, Alders M, Westerveld A, et al. Increased tumour risk for BWS patients correlates with aberrant H19 and not KCNQ1OT1 methylation: occurrence of KCNQ1OT1 hypomethylation in familial cases of BWS. Hum Mol Genet 2001;10:467-76. 9. Bliek J, Terhal P, van den Bogaard MJ, Maas S, Hamel B, Salieb-Beugelaar G, et al. Hypomethylation of the H19 gene causes not only Silver-Russell syndrome (SRS) but also isolated asymmetry or an SRS-like phenotype. Am J Hum Genet 2006;78:604-14. 10. DeBaun MR, Tucker MA. Risk of cancer during the first four years of life in
100
Bliek et al
children from The Beckwith-Wiedemann Syndrome Registry. J Pediatr 1998;132(Pt 1):398-400. 11. Shuman C, Smith AC, Steele L, Ray PN, Clericuzio C, Zackai E, et al. Constitutional UPD for chromosome 11p15 in individuals with isolated hemihyperplasia is associated with high tumor risk and occurs following assisted reproductive technologies. Am J Med Genet A 2006;140:1497-503. 12. Itoh N, Becroft DM, Reeve AE, Morison IM. Proportion of cells with paternal 11p15 uniparental disomy correlates with organ enlargement in Wiedemann-beckwith syndrome. Am J Med Genet 2000;92:111-6. 13. Slatter RE, Elliott M, Welham K, Carrera M, Schofield PN, Barton DE, et al. Mosaic uniparental disomy in Beckwith-Wiedemann syndrome. J Med Genet 1994; 31:749-53. 14. Henry I, Puech A, Riesewijk A, Ahnine L, Mannens M, Beldjord C, et al. Somatic mosaicism for partial paternal isodisomy in Wiedemann-Beckwith syndrome: a post-fertilization event. Eur J Hum Genet 1993;1:19-29. 15. Scott RH, Walker L, Olsen OE, Levitt G, Kenney I, Maher E, et al. Surveillance for Wilms tumour in at-risk children: pragmatic recommendations for best practice. Arch Dis Child 2006;91:995-9. 16. Niemitz EL, Feinberg AP, Brandenburg SA, Grundy PE, DeBaun MR. Children with idiopathic hemihypertrophy and Beckwith-Wiedemann syndrome have different constitutional epigenotypes associated with Wilms tumor. Am J Hum Genet 2005; 77:887-91.
The Journal of Pediatrics • July 2008
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
Appendix I. Epigenetic and clinical details of 73 patients with HH
Patient
Sex
Group
M.I. KCNQ1OT1
M.I. H19
Tumor
Birth weight >P97
Macroglossia
Abdominal wall defects
Pits and creases
Hypoglycemia
Nevus flammeus
Nephromegaly
Patients with hemihyperplasia IH1 F 1
0.34
0.66
⫺
⫺
⫺
⫺
⫺
⫹
⫺
⫺
IH2
F
1
0.68
0.23
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH31
F
1
0.31
0.69
⫺
⫹
⫺
⫺
⫺
⫺
⫺
⫺
IH42
M
1
0.34
0.60
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH5*
F
1
0.39
0.57
bWT
U
⫺
⫺
⫺
⫺
⫺
⫺
IH6
M
1
0.3
⫺
⫺
⫺
⫺
⫺
⫺
U
⫺
⫹
IH7*
F
1
0.35
nd
bWT
U
⫺
⫺
⫺
⫺
⫺
⫺
IH8 IH9
F F
1 2
0.15 0.51
0.79 0.60
⫺
⫹ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
U ⫺
U ⫺
⫺ ⫹
IH10 IH113
M M
2 2
0.49 0.47
0.66 0.73
⫺ ⫺
⫺ u
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫹
IH12
M
3
0.39
0.5
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH13
M
4
0.47
0.53
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
Other features
Hypertrophy Right arm 1,5 cm⬎ left; right leg 1 cm⬎left Right arm and leg⬎ left, becoming less pronounced Right arm incl. hand; right leg incl. foot; thorax and abdomen: right side thicker Complete left body side larger Breasts and leg right⬎left Left arm and leg circumference 1-2 cm larger, left foot 1 shoe size bigger, no difference in length of leg Breast, arm and leg left⬎right
100.e1
Right lower leg, right foot⬎ left Leg left⬎right Both lower legs and feet have a pasty skin, feet very broad, looks like Proteus. Short dig I, II, IV and V Arms right⬎left, minimal difference Left leg, arm, thorax
100.e2
Appendix I. Continued
Bliek et al
Tumor
Macroglossia
Abdominal wall defects
Pits and creases
Hypoglycemia
Nevus flammeus
Nephromegaly
The Journal of Pediatrics • July 2008
Sex
Group
IH14 IH15 IH16* IIH17
M F F F
4 4 4 4
0.50 0.51 0.46 0.46
0.48 0.47 0.46 0.48
⫺ ⫺ NBL ⫺
⫺ ⫹ U ⫹
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
IH18
M
4
0.50
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫹
IH19 IH20 IH21 IH22
F M F F
4 4 4 4
0.47 0.47 0.56 0.55
0.50 0.50 0.51 0.48
⫺ ⫺ ⫺ ⫺
⫺ u U U
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
IH23
F
4
0.48
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH24 IH25
M M
4 4
0.51 0.46
0.52 0.45
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
IH26
M
4
0.48
0.48
⫺
U
⫺
⫺
⫺
⫺
⫺
⫺
IH27 IH28
M F
4 4
0.45 0.49
0.55 0.55
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
IH29
F
4
0.51
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH30 IH31
F M
4 4
0.49 0.48
0.51 0.50
⫺ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ U
⫺ ⫺
⫺ ⫺
IH32 IH33
V M
4 4
0.48 0.49
0.46 0.49
⫺ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ U
⫺ ⫺
⫺ ⫺
IH34*
F
4
0.44
0.54
RMS
U
⫺
⫺
⫺
⫺
⫺
⫺
Patient
M.I. H19
Birth weight >P97
M.I. KCNQ1OT1
Other features
Hypertrophy Right arm and leg
Left leg ⬎right, difference 0.5 to 1 cm at different places. Leg circumference right⬎left arm circumference right ⬎left Leg right⬎left Arm left⬎right Leg left⬎right Lower leg and foot, asymmetrical face right⬎left Left upper leg shorter and thinner, left foot shorter, asymmetrical face Leg left⬎right Face, arm and leg right⬎left Face, arm and leg right⬎left Leg right⬎left Arm, leg and foot left⬎right Lower arm, hand, leg and foot left⬎right Arm and hand right⬎left Leg right⬎left Arm and hand right⬎left Ear, arm and leg left⬎right
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
Appendix I. Continued Birth Abdominal Pits M.I. M.I. weight wall and Nevus Patient Sex Group KCNQ1OT1 H19 Tumor >P97 Macroglossia defects creases Hypoglycemia flammeus Nephromegaly IH35
F
4
0.47
0.49
⫺
U
⫺
⫺
⫺
⫺
⫺
⫺
IH36
F
4
0.51
0.52
⫺
⫹
⫺
⫺
⫺
⫺
⫺
⫺
IH37
M
4
0.48
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH38 IH39
F F
4 4
0.47 0.51
0.41 0.53
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫹ ⫺
⫺ ⫺
IH40
V
4
0.47
0.53
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫹
IH41 IH42
M M
4 4
0.50 0.50
0.51 0.5
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
IH43
F
4
0.500
0.50
⫺
U
⫺
⫺
⫺
⫺
⫺
⫺
IH44
M
4
0.48
0.50
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH45
M
4
0.47
0.47
⫺
⫺
⫺
⫺
⫺
U
⫹
⫺
IH46 IH47
M M
4 4
0.46 0.53
0.53 0.57
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
Patients with hemihyperplasia with additional major features IH⫹1 M 3 0.39 0.51 ⫺
IH⫹2
M
4
0.46
0.46
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH⫹3
M
4
0.49
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
Patients with hemihyperplasia with 1 or more BWS features BWS⫺1 F 1 0.22 0.78 ⫺ ⫺
⫺
⫹
⫺
⫺
⫺
⫺
Other features
Hypertrophy Left leg 4 cm longer (at 2 y) Left arm, leg, face ⬎ right Arm and leg right⬎left Right leg Face, arm, leg, hand and foot Face and thorax right⬎left Left leg ⬎ right Whole body left⬎right Face, arm and leg left⬎right Face, arm and leg right⬎left Leg and foot left⬎right Lower leg and foot circumference left⬎right
Cyanosis, advanced Right leg⬎ left skeletal age, prominent occiput, partial hair albinism, depigmentation/ hyperpigmentation skin of the neck. Syndactyly of dig 3-4 Body side right ⬎ left hand and dig left 1-2-3 left feet Mental retardation, Left arm⬎right prominent occiput Scoliosis
Left body side except face
100.e3
100.e4
Appendix I. Continued
Bliek et al
Macroglossia
Abdominal wall defects
Pits and creases
Hypoglycemia
Nevus flammeus
Nephromegaly
The Journal of Pediatrics • July 2008
Sex
Group
BWS⫺2
M
1
0.38
0.62
⫺
⫺
⫺
⫺
⫺
⫺
⫹
⫹
BWS⫺3 BWS⫺4
M F
2 3
0.47 0.38
0.65 0.51
⫺ ⫺
⫺ ⫺
⫺ ⫹
⫹ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫹
⫺ ⫺
BWS⫺5 BWS⫺6
F M
3 3
0.20 0.00
0.50 0.48
⫺ ⫺
⫹ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
⫹ U
⫹ ⫺
⫺ ⫺
BWS⫺7 BWS⫺8 BWS⫺9 BWS⫺10
F F M F
4 4 4 4
0.46 0.48 0.52 0.45
0.49 0.52 0.53 0.50
⫺ ⫺ ⫺ ⫺
⫹ ⫺ ⫺ ⫺
⫹ ⫺ ⫹ ⫹
⫺ ⫹ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ U ⫺ U
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
BWS⫺11 BWS⫺12
F M
4 4
0.51 0.45
0.49 0.48
⫺ ⫺
⫺ ⫺
⫹ ⫹
⫺ ⫺
⫺ ⫺
U ⫺
⫺ ⫺
⫺ ⫺
BWS⫺13
F
4
0.49
0.48
WT
⫺
⫺
⫺
⫹
⫺
⫹
⫺
BWS⫺14
F
4
0.49
0.49
⫺
⫺
⫺
⫹
⫺
⫺
⫺
⫺
Patient
M.I. H19
Birth weight Tumor >P97
M.I. KCNQ1OT1
Hypertrophy Other features Right upper arm and leg; right leg 2 cm longer; right kidney; right foot⬎left Vision problems, advanced skeletal age, underdeveloped maxilla, chubby face, prognathism Thin upper lip
Face, arm and leg length left⬎right
Seizures Seizures Feeding problems, apnea, polycythemia Plagiocephaly
Face, leg length, thorax, arm right⬎left Mammae and slightly in the face Leg: right⬎left
HH⫹, hemihyperplasia with other, not BWS related, features; BWS⫺, IH patients with one or more additional signs of BWS but not meeting the clinical criteria; M, male; F, female; MI, methylation index; bWT, bilateral Wilms tumor; U, unknown; nd, not done; ⫹, present; ⫺, absent; ⬎P97, Birth weight above 97 percentile. 1. nephrocalcinosis 2. 1st year nephroblastoma 3. hepatomegaly but HH on right side of the body *Also in study by Merks et al.7
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
Appendix II. Epigenetic and clinical detail of 14 patients with cancer with IH
100.e5
Sex
Group
M.I. KCNQ1OT1
M.I. H19
Tumor
HT1
F
1
0.35
U
BWT
B
R⬎L
HT2
F
2
0.52
0.6
WT
R
R⬎L
HT3
M
2
0.46
0.63
RMS
L
L⬎R
HT4
F
2
0.49
0.6
WT
L
R⬎L
HT5
F
4
0.39
0.57
WT
L
R breast ⬎L, L leg ⬎ R
HT6
M
4
0.50
0.49
WT
L
L⬎R
HT7
F
4
0.44
0.54
RMS
L
L⬎R
HT8
F
4
0.49
0.51
WT
L
L⬎R
HT9
F
4
0.46
0.46
NBL
Median
R⬎L
Obesity, Convex nose
HT10
F
4
0.47
0.48
WT
R
L⬎R
Upward slant, Concave nose, Broad flat nasal bridge, Multiple nevi, Multiple café au lait spots
Number
Sidedness tumor
Sidedness IH
Beckwith features
Hernia inguinalis
Phenotype
Hypertrophy
Plagiocephaly, Asymmetric pupils, Concave nose, Wine mark Retrognathia, Scoliosis, Hyperlaxity joints
Hemihypertrophy right breast, Upper and lower limb Hemihypertrophy right upper and lower limb Left lower limb 2 cm thicker and 0.5 cm longer the right. L Hemihypertrophy right ear, mamma, upper and lower limb Asymmetric breasts, Hemihypertrophy right lower limb Hemihypertrophy left pectoral muscle, Hemihypertrophy left upper limb
Overhanging nasal tip, Thin upper lip, Caries, Sandal gap Fetal pads, Clinodactyly, Hypermobility large joints Diastema Clinodactyly, Multiple small nevi Deeply set eyes, Upward slant, Thin nasal bridge, Thin upper lip, Camptodactyly, Sandal gap, Café au lait (2 ⫻ 1 cm) Upward slant, Broad and deviated nasal tip, Hyperlaxity, Multiple small nevi Sparsely implanted hair, Ptosis
Hemihypertrophy left ear, upper and lower limb Hemihypertrophy left upper and lower limb Asymmetric mammae, Hemihypertrophy right upper limb, Asymmetric face, Slight hypotrophy right iris Asymmetric face, Hemihypertrophy left upper and lower limbs
100.e6
Appendix II. Continued
Bliek et al
Sex
Group
M.I. KCNQ1OT1
M.I. H19
HT11
F
4
0.50
0.51
WT
R
L⬎R
HT12
F
4
0.50
0.50
WT
R
L⬎R
HT13
F
4
0.49
0.53
WT
L
L⬎R
HT14
F
4
0.51
0.46
Sacrococcygeal teratoma
Median
R⬎L
Number
Tumor
Sidedness tumor
Sidedness IH
HT, Hemihypertrophy and tumor; bWT, bilateral Wilms tumor; L, left; R, right; b, bilateral; F, female; M, male; MI, methylation index.
Beckwith features
Phenotype Low anterior hairline, Upward slant, Epicanthal folds, Almond shaped eyes, Hemihypertrophy left lower limb, Café au lait spot (2.5 ⫻ 4 cm) Clinodactyly, Fetal pads, Multiple nevi (⬎0.5 cm) Blepharophimosis, Full lower lip, Overfolded helix, Multiple nevi Mental retardation, Cor vitium, Thorax vsd
Hypertrophy Hemihypertrophy left lower limb
Hemihypertrophy left lower limb Mamma hyperplasia, Hemihypertrophy lower limb Hemihypertrophy right upper and lower limb
The Journal of Pediatrics • July 2008
AND
MARCEL MANNENS, PHD
Objective To investigate whether epigenotyping of patients with isolated hemihyperplasia (IH) can, analogous to genetic screening of patients with Beckwith-Wiedemann syndrome, be used for the prediction of tumor risk and tumor type of individual patients. Study design Methylation analysis of H19 and KCNQ1OT1 of 73 patients. Questionnaires were sent to referring clinicians. Results In 75% of the clinically confirmed patients with IH no epigenetic defect was detected. Paternal uniparental disomy was found in 15%, demethylation of KCNQ1OT1 in only 6%, and hypermethylation of H19 in 3% of isolated hemihyperplasia cases. Ten percent of the patients with IH had development of a childhood tumor associated with paternal uniparental disomy (2/8) or no methylation defect (2/30). No genetic defect was detected in 10 of 14 additional patients with cancer with IH. In these latter patients, a methylation defect of H19 was seen 3 times and a paternal uniparental disomy once. The female-to-male ratio was 6:1. Conclusions Aberrant methylation of the 11p15 region is not common in patients with IH and can at present not be used for tumor risk determination. (J Pediatr 2008;153:95-100) emihyperplasia, also called hemihypertrophy (HH), is a congenital abnormality characterized by asymmetric growth of the limbs, trunk, face, or the entire body. HH has a heterogeneous cause: in its isolated form it is called idiopathic or isolated hemihyperplasia (IH [MIM 235000]),1 but HH is often associated with various syndromes such as BeckwithWiedemann syndrome (BWS [MIM130650]),1 Klippel-Trenaunay-Weber syndrome (KTWS [MIM149000]),1 Silver-Russell syndrome (SRS [MIM 180860]), or Proteus syndrome (MIM 176920).1 The incidence of IH has been reported to be approximately 1:13,000 to 1:86,000 live births.2 A genetic cause of IH was first described by Martin et al3 in 2005. In 8/27 (30%) patients with IH, abnormal methylation profiles were found in 2 genes (KCNQ1OT1 and H19) in 2 distinct imprinted regions on chromosome 11p15. Genes that are subject to imprinting are expressed from 1 chromosome only. The gene on the homologous chromosome is silenced by methylation of CpG nucleotides. KCNQ1OT1 and H19 are both antisense transcripts; they do not code for a protein but regulate expression of other genes nearby. One of the genes regulated by these transcripts is IGF2, a growth factor expressed only in the embryo. Mice overexpressing IGF2 have a large birth weight.4 KCNQ1OT1 and H19 are imprinted in the opposite direction: KCNQ1OT1 is methylated on the maternal allele and expressed from the paternal allele, whereas H19 is methylated on the paternal allele and expressed from the maternal allele. Patients with both IH and BWS have an increased risk for the development of From the Department of Clinical Genetics (J.B., S.M., M.A., M.M.), the Department of childhood tumors compared with the incidence in the general population (0.17%). In Paediatrics (S.M.), and the Department of patients with IH, a total of 10 tumors developed in 9 patients (5.9%), 4 unilateral Wilms Paediatric Oncology, Emma Children’s Hospital (J.M.), Academic Medical Centre, tumors (WT), 2 bilateral WTs, 2 adrenal cell carcinomas, 1 hepatoblastoma, and 1 Amsterdam, the Netherlands. leiomyosarcoma(i). All tumors developed before the age of 5 years. Submitted for publication Mar 31, 2007; 5 Patients with BWS have an increased tumor risk of 7.5% to 10%. In a review of 278 last revision received Oct 26, 2007; ac6 cepted Dec 10, 2007. patients with BWS, we demonstrated that the different genetic subgroups differ in their Reprint requests: Jet Bliek BA-AS, Acarelative tumor risk and the type of tumor that develops.
H
BWS BWS⫹HH BWS-HH HH IH
Beckwith Wiedemann syndrome BWS patients with HH BWS patients without HH Hemihypertrophy Isolated hemihypertrophy
KTWS pUPD SRS WT
Klippel-Trenaunay-Weber syndrome Paternal uniparental disomy Silver Russell syndrome Wilms tumor
demic Medical Centre, Department of Clinical Genetics, Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands. E-mail: j. [email protected]. 0022-3476/$ - see front matter Copyright © 2008 Mosby Inc. All rights reserved. 10.1016/j.jpeds.2007.12.022
95
Patients with BWS with a methylation defect of H19 have a very high risk for development of a childhood tumor, and these are uniquely Wilms tumors. Patients with a methylation defect of KCNQ1OT1 have a low tumor risk, and Wilms tumors never develop in these patients. Patients with paternal uniparental disomy (pUPD) and without detectable defect on chromosome 11p15 have an intermediate tumor risk; both Wilms’ tumors and other childhood tumor are found in these patients. Epigenotyping can therefore be used to estimate the tumor risk and predict the tumor type of individual patients with BWS. The question arises whether epigenotyping can concordantly be used for the delineation of the tumor risk of individual patients with IH. Can patients with IH be considered as patients with BWS with a mild phenotype and is it possible to determine the tumor risk of individual patients with IH on the same epigenetic basis as patients with BWS? IH might, however, be a distinct phenotype with a distinct distribution of tumor risk and tumor type over the different epigenetic subgroups. At last, IH might (in part) be caused by completely different, hitherto unknown, genetic defects. To answer these questions, we collected clinical and epigenetic data on 74 patients referred to our DNA diagnostic laboratory for methylation screening.
METHODS Patients This study included individuals with IH who were referred for clinical assessment or molecular studies by pediatricians or clinical geneticists in the Netherlands. The study was approved by the medical ethics committee of our hospital. Either all patients were examined by a clinical geneticist, or a checklist with clinical findings possibly related to IH was completed by referring pediatricians and clinical geneticists for each patient. Blood samples from all patients were sent to our laboratory for molecular analysis. Two independent clinical geneticists with expertise in evaluating children with syndromes reviewed all records. Classification of IH was independent from the results of genetic analysis. When possible, updates were collected on the development of childhood tumors. None of these patients were born after in vitro fertilization procedures. To compare the epigenetic distribution of patients with BWS and IH, we used methylation data of 110 patients with BWS who were referred to our laboratory for diagnostic testing. All patients met the clinical criteria for BWS. The patients were grouped according to the presence of HH, BWS⫹HH (BWS patients with HH), and BWS⫺HH (BWS patients without HH). In the study on malformation syndromes found among patients with cancer performed by Merks et al,7 14 patients presented with HH. Methylation analysis was performed in our laboratory on 13 patients. Three patients (HT1, HT7, and HT9 in Appendix 2; available at www.jpeds.com) were referred to our laboratory for diagnostic testing before the start of their study. One additional patient was included after 96
Bliek et al
the publication of the study (HT 14). In total we present methylation data on 14 patients with cancer with IH. Detailed clinical data of these patients are presented here, in combination with results of epigenetic profiling.
Sample Analysis Methylation analysis was performed with standard Southern blotting, and pUPD was analyzed with CA repeat markers as described previously.8 We compared the distribution of patients over the 3 epigenetic profiles (pUPD, hypermethylation of H19, and demethylation of KCNQ1OT1) and the female-to-male ratio by use of the 2 test (P ⬍ .05 was considered statistically significant).
RESULTS Clinical data were available for 73 patients with HH that were referred to our hospital for diagnostic testing. Detailed clinical data are listed in Appendix 1 (available at www.jpeds.com) and summarized in Table I. Forty-seven patients were classified as patients with IH. In 16 of these patients, a maximum of 1 finding typical for BWS was present. Birth weight was increased in 8 patients. Nephromegaly was present in 4 cases. In all cases the enlarged kidney was present in the hyperplastic side of the body. One patient presented with nephroblastomatosis at the age of 1 year. Three patients had nevus flammeus (6%). One patient was diagnosed with transient neonatal hypoglycemia. Three patients showed anomalies in addition to HH, indicating the presence of a syndrome other than BWS. Six patients showed hypomethylation of H19. The hyperplasia was reclassified as hypoplasia as part of the SRS. These patients were described elsewhere.9 Fourteen patients presented more than 1 additional finding typical for BWS but did not meet the criteria described by DeBaun and Tucker.10 Three patients displayed other unexplained anomalies; they were classified as HH⫹ patients. Of the control population of 110 patients with BWS, 54 patients had HH (BWS⫹HH), and 56 patients showed no HH (BWS⫺HH). The epigenetic profiles of the 47 patients with IH in this study are summarized in Table I. Thirty-five patients (74%) showed no epigenetic defect on chromosome 11p15. In the group with epigenetic alterations, pUPD 11p15 was the most frequently observed defect. In our study of 41 patients with an age of more than 5 years, 4 patients (10%) had a tumor develop. Detailed clinical data of 14 patients with cancer with IH described by Merks et al7 are presented in Appendix 2 (available at www.jpeds.com) and summarized in Table II. One patient exhibits 1 minor feature associated with BWS, inguinal hernia, but did not meet the BWS criteria. One patient was mentally retarded. No specific association, apart from IH, between a phenotypic abnormality and tumor development could be detected. We established the methylation status of these patients (Table I). Most tumors were found in patients with no methylation defect. The Journal of Pediatrics • July 2008
Table I. Overview of epigenetic data and clinical classification of 74 patients with HH and 14 patients with cancer with IH Patients HH⫹ SRS BWS⫺ Other syndrome IH (percentage) Tumors in this study* Control BWS population (percentage) Epigenetic distribution of cancer patients with hemihyperplasia No of cancer patients Tumor type in cancer patients with IH
Total
pUPD (H191 and KCNQ1OT12)
BWSIC1 (H191)
BWSIC2 (KCNQ1OT12)
No methylation defect
0
0
3
3 0 1 (2%) (0/1)
3 6 14 3 47 4/41 (10%)
2 0 8 (17%) 2 ⫻ WT (2/8)
0 6 H192 1 0 3 (6%) (0/2)
110
29 (26%)
11 (10%)
51 (46%)
3 (21%) 2 ⫻ WT 1 ⫻ RMS
0 (0%) 0
8 3 35 (74%) 1 ⫻ RMS 1 ⫻ NB (2/30) 19 (17%)
14 14 14
1 (7%) 1 ⫻ bWT
7 1 1 1
10 (72%) ⫻ WT ⫻ RMS ⫻ NB ⫻ sacrococcygeal teratoma
BWSIC1, Beckwith Wiedemann Imprinting Centre 1; BWSIC2, Beckwith Wiedemann Imprinting Centre 2; HH⫹, hemihyperplasia with other, not BWS related, features; BWS⫺, IH patients with one or more additional signs of BWS but not meeting the clinical criteria; bWT, bilateral Wilms tumor. *Data available on 41 patients older than 5 years.
Of the patients with IH in this study, the female-to-male ratio was 1.14:1 (25 females, 22 males). This is not significantly different from the normal population (P ⫽ .66). In contrast, in the group of patients with cancer with IH there is a significantly (P ⫽ .008) increased female-to-male ratio of 6:1 (12 females, 2 males). In the 1073 patients with cancer in the study by Merks et al,7 the female-to-male ratio was 0.88:1. The side of the hyperplasia was known in 40 patients with IH who were referred to our hospital for epigenetic screening. There was an equal ratio between right and left sidedness (20:20). Of the patients with cancer with IH, the side where the tumor developed and the hyperplastic side were known in all cases. In this patient group, the sidedness of the HH is slightly in favor of the left side (8:5). One might expect the tumor to develop on the HH side of the body. This is not the case; of the 9 unilateral WTs, 4 developed ipsilateral and 4 developed contralateral (one patient with WT showed HH of the right breast and left leg). A degree of hyperplasia is difficult to establish. The difference between normal and hyperplastic body regions varies during life, and no data were available on differences in
circumference or weight of the affected versus the unaffected tissues. An indication of the extent of overgrowth is the number of body parts involved. Of 41 patients with IH, details were available on the body parts involved; in 17 patients only one extremity was hyperplastic, and in 24 patients more than 1 body region was involved. This did not differ significantly (P ⫽ .68) from the patients with cancer with IH, whereas in 6 patients only 1 extremity was involved, and in 8 patients more than 1 body region was affected.
DISCUSSION HH can be considered a mosaic form of overgrowth and might be caused by mosaic genetic defects. Indeed, pUPD is always found in a mosaic form, and enlarged organs consist of a larger proportion of disomic cells. Methylation patterns, for diagnostic purposes, are established in blood lymphocytes and not in the hyperplastic tissues. Shuman et al11 postulate that in a part of the unexplained IH cases, pUPD is present only in hyperplastic tissue and therefore not detected in blood by routine diagnostic testing. Recently, methylation patterns in
Table II. Female-to-male distribution and sidedness of hemihyperplasia and tumor Patients
F:M
WT tumor*
L>R
R>L†
1 extremity
>1 extremity
IH Patients for diagnostic testing Cancer patients with IH
25:22 12:2
– 4 ipsilateral 4 contralateral
20 8
20 5
17 7
24 8
L, left; R, right; F, female; M, male. *Two patients had development of a hepatoblastoma, 1 patient showed contralateral hyperplasia, 1 had development of a bilateral tumor, and 2 patients had development of a midline tumor. †One patient showed contralateral hyperplasia.
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
97
Table III. Overview of epigenetic data of patients with IH and patients with cancer with IH (included data from literature)3,11 Total IH Patients for diagnostic testing IH patients in this study IH patients (combined with literature) IH Patients with a methylation defect only IH patients (combined with literature) Control BWS⫹ HH patients Control BWS⫺HH patients Tumor risk of IH patients Tumors in IH patients in this study Tumors in IH patients (combined with literature) Epigenotype and tumor type of cancer patients with IH Cancer patients in this study 14 1 ⫻ WT Cancer patients (combined with literature)
pUPD (H191 and KCNQ1OT12)
BWSIC1 (H191)
BWSIC2 (KCNQ1OT12)
No methylation defect
47 125
8 (17%) 19 (15%)
3 (6%) 4 (3%)
1 (2%) 8 (6%)
31
19 (61%)
4 (13%)
8 (25%)
—
45 46
18 (40%) 11 (24%)
7 (16%) 4 (9%)
20 (44%) 31 (67%)
— —
4/41 (10%) 14/119 (12%)
14 29
35 (74%) 94 (75%)
2 ⫻ WT (2/8) 1 ⫻ RMS, 1 ⫻ 6/19 (32%)
(0/2)
(0/1)
NBL (2/30)
0/3 (0%)
0/8 (0%)
9/89 (10%)
1 ⫻ WT 1 ⫻ RMS 3 (10%)
2 ⫻ WT 1 ⫻ RMS 4 (14%)
0
8 ⫻ WT 1 ⫻ NBL 22 (75%)
0
BWSIC1, Beckwith Wiedemann Imprinting Centre 1; BWSIC2, Beckwith Wiedemann Imprinting Centre 2; BWT, bilateral Wilms tumor.
different tissues of a patient with BWS were studied by Itoh et al.12 They showed that the proportion of cells with pUPD correlates with organ enlargement in BWS. However, pUPD could also be detected in a lower percentage in blood lymphocytes and would therefore also have been diagnosed by routine testing. In the study by Itoh et al,13 the degree of disomic cells was high (40% to 50%). When pUPD is present in only a small percentage of cells, it might have been missed in lymphocytes. Analogous to pUPD, isolated methylation defects of H19, or KCNQ1OT1 found in patients with IH are never complete. The defect is present in a mosaic form. It is possible that the proportion of aberrantly methylated cells is higher in enlarged organs. It is difficult to obtain different tissues from children with hyperplasia, and, until recently, the detection of small aberrations in methylation patterns still required large amounts of DNA. Another possible explanation for absence of methylation defects in most patients with IH might be that HH is associated with a variety of syndromes and that patients with IH present a mild form of any of these syndromes. The phenotype of BWS is very variable, and only patients with multiple features meet the criteria for BWS. Consequently, part of the patients with IH might be regarded as having a mild form of BWS. In these patients, a distribution of the methylation defects similar to BWS would be expected. Furthermore, in 20% of patients with BWS, the genetic defect is unknown. Because the distribution of epigenetic changes in patients with IH and BWS is quite different, the mechanism 98
Bliek et al
causing BWS in these latter patients might also be responsible for IH in a larger part of the patients with IH. However, HH is also found in other syndromes, such as Klippel-Trenaunay-Weber syndrome (KTWS) and Proteus syndrome, and IH might also represent mild forms of these syndromes. The exact genetic defect of KTWS is unknown, but mutations in PTEN have been described in some patients with Proteus syndrome. Thirteen of 14 patients with cancer with IH in this study have been screened for mutations in PTEN, but no mutations were found. Six patients with IH presenting with prenatal overgrowth have been screened for GPC3 mutations; no mutations were found (data not shown). Among patients with BWS, demethylation of KCNQ1OT1 is the most common genetic aberration found. This is not the case among patients with IH. Methylation profiles of patients with IH and BWS with (BWS⫹HH) and without IH (BWS⫺HH) are compared in Table III. The most common genetic defect found among patients with IH is pUPD. It was found in 61% of patients with a genetic defect. In contrast, the most common genetic defect found among patients with BWS is demethylation of KCNQ1OT1, which is found in 44% of patients with BWS with HH and in 67% of patients with BWS without HH. There was no significant difference (P ⫽ .214) between the genetic distribution of patients with IH and patients with BWS and HH. The population with IH, however, differed significantly from the population with BWS-HH (P ⫽ .002) because of overrepresentation of patients with pUPD in the group with IH The Journal of Pediatrics • July 2008
Figure. A, Epigenotype and tumor risk of 278 patients with BWS. B, Epigenotype and tumor risk of 125 patients with IH.
and a shift toward KCNQ1OT1 demethylation among the patients with BWS-HH. This confirms previous reports13,14 that pUPD is associated with hemihyperplasia. Analogous to BWS, patients with IH have an increased risk for development of childhood tumors. In 2005 we reviewed epigenetic and tumor data in 278 patients with BWS described in the literature.6 Data are summarized in the Figure, A. The patient populations in the reviewed articles all show a bias because of overrepresentation of patients with cancer for research purposes. The overall tumor risk in the review is 26%, compared with the estimated risk of 7.5% to 10% for patients with BWS. The tumor risk of patients with BWS with KCNQ1OT demethylation is relatively low. WT has never been detected in this group. Patients with pUPD represent 20% of patients with BWS and have a tumor risk of 36%; both WT and other childhood tumors are found. Patients showing hypermethylation of H19 have the highest risk for development of cancer. They represent 10% of the patients with BWS and have a tumor risk of 50%; only WT is found among these patients. The remaining 20% have no detectable methylation defect but still have an increased tumor risk of 22%; both WT and other childhood tumors are found.
On the basis of these data, it is possible to delineate tumor risk for the different epigenetic subgroups. Scott et al15 propose preventive screening of all patients with a tumor risk higher than 5%. Patients with isolated demethylation of KCNQ1OT1, the largest epigenetic subgroup representing 50% of patients with BWS, are therefore exempted from preventive screening. Data on tumor risk in different epigenetic subgroups in patients with IH are less abundantly available. We reviewed the literature and combined our data with the data presented by Martin et al3 and Shuman et al11 on a total of 125 patients with IH in a first attempt to delineate the tumor risk in patients with IH. In our study, 41/47 patients with IH were more than 5 years old; of these, 4 (10%) had a tumor develop. This is comparable to the risk that is found in the study by Hoyme et al.2 Two patients with a tumor showed pUPD (2/8); they had development of a WT. Two other patients with a tumor did not show a methylation defect (2/30); one had a rhabdomyosarcoma develop and one a neuroblastoma. No tumors were found in the patients with hypermethylation of H19 (0/2) and demethylation of KCNQ1OT1 (0/1). In the study by Martin et al,3 methylation data were presented on 27 patients with IH. One patient had a tumor develop. This patient showed no methylation defect, but the tumor type is unknown (Martin RA, personal communication, 2005). Shuman et al11 presented data on 51 patients with IH, of which 10 (20%) had a tumor develop. Four patients with a pUPD (4/8) had development of a tumor (3 hepatoblastomas, 1 WT). Six patients with no detectable defect (6/40) had a tumor develop (3 WT, 1 NB, 1 adrenal cell carcinomas). There were no patients with hypermethylation of H19, and no tumors were found among patients with demethylation of KCNQ1OT1 (0/3). Data from our study and the literature are combined and shown in the Figure, B. Including the study by Niemitz et al,16 data are available on 29 patients with cancer with IH. Comparable with the general IH cohort, most showed no methylation defects on chromosome 11p15. Four of 29 patients showed hypermethylation of H19. This demonstrates that patients with this methylation profile can develop tumors, although in the general IH cohort none of the IH patients with this profile developed a tumor. Demethylation of KCNQ1OT1 was not found among patients with cancer, but this defect is found in only a minority of patients with IH (6% [8/125]). A remarkable finding is the female overrepresentation in our cancer cohort (12:2), for which we have no explanation. We would like to acknowledge K. v.d. Lip and S. de Leng for excellent technical assistance, J. M. Cobben, MD, PhD, for evaluation of the clinical records of the patients, and Prof. N. J. Leschot and Prof. A. Westerveld for critically reviewing this manuscript.
REFERENCES 1. OMIM Online Mendelian Inheritance in Man, John Hopkins University, available at http://ncbi.nlm.nih.gov. Accessed February 12, 2008.
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
99
2. Hoyme HE, Seaver LH, Jones KL, Procopio F, Crooks W, Feingold M. Isolated hemihyperplasia (hemihypertrophy): report of a prospective multicenter study of the incidence of neoplasia and review. Am J Med Genet 1998;79:274-8. 3. Martin RA, Grange DK, Zehnbauer B, DeBaun MR. LIT1 and H19 methylation defects in isolated hemihyperplasia. Am J Med Genet A 2005;134:129-31. 4. DeChiara TM, Efstratiadis A, Robertson EJ. A growth-deficiency phenotype in heterozygous mice carrying an insulin-like growth factor II gene disrupted by targeting. Nature 1990;345:78-80. 5. Wiedemann HR. Tumours and hemihypertrophy associated with WiedemannBeckwith syndrome. 141 ed. 1983. p. 129. 6. Bliek J, Gicquel C, Maas S, Gaston V, Le BY, Mannens M. Epigenotyping as a tool for the prediction of tumor risk and tumor type in patients with BeckwithWiedemann syndrome (BWS). J Pediatr 2004;145:796-9. 7. Merks JH, Caron HN, Hennekam RC. High incidence of malformation syndromes in a series of 1,073 children with cancer. Am J Med Genet A 2005;134:132-43. 8. Bliek J, Maas SM, Ruijter JM, Hennekam RC, Alders M, Westerveld A, et al. Increased tumour risk for BWS patients correlates with aberrant H19 and not KCNQ1OT1 methylation: occurrence of KCNQ1OT1 hypomethylation in familial cases of BWS. Hum Mol Genet 2001;10:467-76. 9. Bliek J, Terhal P, van den Bogaard MJ, Maas S, Hamel B, Salieb-Beugelaar G, et al. Hypomethylation of the H19 gene causes not only Silver-Russell syndrome (SRS) but also isolated asymmetry or an SRS-like phenotype. Am J Hum Genet 2006;78:604-14. 10. DeBaun MR, Tucker MA. Risk of cancer during the first four years of life in
100
Bliek et al
children from The Beckwith-Wiedemann Syndrome Registry. J Pediatr 1998;132(Pt 1):398-400. 11. Shuman C, Smith AC, Steele L, Ray PN, Clericuzio C, Zackai E, et al. Constitutional UPD for chromosome 11p15 in individuals with isolated hemihyperplasia is associated with high tumor risk and occurs following assisted reproductive technologies. Am J Med Genet A 2006;140:1497-503. 12. Itoh N, Becroft DM, Reeve AE, Morison IM. Proportion of cells with paternal 11p15 uniparental disomy correlates with organ enlargement in Wiedemann-beckwith syndrome. Am J Med Genet 2000;92:111-6. 13. Slatter RE, Elliott M, Welham K, Carrera M, Schofield PN, Barton DE, et al. Mosaic uniparental disomy in Beckwith-Wiedemann syndrome. J Med Genet 1994; 31:749-53. 14. Henry I, Puech A, Riesewijk A, Ahnine L, Mannens M, Beldjord C, et al. Somatic mosaicism for partial paternal isodisomy in Wiedemann-Beckwith syndrome: a post-fertilization event. Eur J Hum Genet 1993;1:19-29. 15. Scott RH, Walker L, Olsen OE, Levitt G, Kenney I, Maher E, et al. Surveillance for Wilms tumour in at-risk children: pragmatic recommendations for best practice. Arch Dis Child 2006;91:995-9. 16. Niemitz EL, Feinberg AP, Brandenburg SA, Grundy PE, DeBaun MR. Children with idiopathic hemihypertrophy and Beckwith-Wiedemann syndrome have different constitutional epigenotypes associated with Wilms tumor. Am J Hum Genet 2005; 77:887-91.
The Journal of Pediatrics • July 2008
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
Appendix I. Epigenetic and clinical details of 73 patients with HH
Patient
Sex
Group
M.I. KCNQ1OT1
M.I. H19
Tumor
Birth weight >P97
Macroglossia
Abdominal wall defects
Pits and creases
Hypoglycemia
Nevus flammeus
Nephromegaly
Patients with hemihyperplasia IH1 F 1
0.34
0.66
⫺
⫺
⫺
⫺
⫺
⫹
⫺
⫺
IH2
F
1
0.68
0.23
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH31
F
1
0.31
0.69
⫺
⫹
⫺
⫺
⫺
⫺
⫺
⫺
IH42
M
1
0.34
0.60
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH5*
F
1
0.39
0.57
bWT
U
⫺
⫺
⫺
⫺
⫺
⫺
IH6
M
1
0.3
⫺
⫺
⫺
⫺
⫺
⫺
U
⫺
⫹
IH7*
F
1
0.35
nd
bWT
U
⫺
⫺
⫺
⫺
⫺
⫺
IH8 IH9
F F
1 2
0.15 0.51
0.79 0.60
⫺
⫹ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
U ⫺
U ⫺
⫺ ⫹
IH10 IH113
M M
2 2
0.49 0.47
0.66 0.73
⫺ ⫺
⫺ u
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫹
IH12
M
3
0.39
0.5
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH13
M
4
0.47
0.53
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
Other features
Hypertrophy Right arm 1,5 cm⬎ left; right leg 1 cm⬎left Right arm and leg⬎ left, becoming less pronounced Right arm incl. hand; right leg incl. foot; thorax and abdomen: right side thicker Complete left body side larger Breasts and leg right⬎left Left arm and leg circumference 1-2 cm larger, left foot 1 shoe size bigger, no difference in length of leg Breast, arm and leg left⬎right
100.e1
Right lower leg, right foot⬎ left Leg left⬎right Both lower legs and feet have a pasty skin, feet very broad, looks like Proteus. Short dig I, II, IV and V Arms right⬎left, minimal difference Left leg, arm, thorax
100.e2
Appendix I. Continued
Bliek et al
Tumor
Macroglossia
Abdominal wall defects
Pits and creases
Hypoglycemia
Nevus flammeus
Nephromegaly
The Journal of Pediatrics • July 2008
Sex
Group
IH14 IH15 IH16* IIH17
M F F F
4 4 4 4
0.50 0.51 0.46 0.46
0.48 0.47 0.46 0.48
⫺ ⫺ NBL ⫺
⫺ ⫹ U ⫹
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
IH18
M
4
0.50
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫹
IH19 IH20 IH21 IH22
F M F F
4 4 4 4
0.47 0.47 0.56 0.55
0.50 0.50 0.51 0.48
⫺ ⫺ ⫺ ⫺
⫺ u U U
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
IH23
F
4
0.48
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH24 IH25
M M
4 4
0.51 0.46
0.52 0.45
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
IH26
M
4
0.48
0.48
⫺
U
⫺
⫺
⫺
⫺
⫺
⫺
IH27 IH28
M F
4 4
0.45 0.49
0.55 0.55
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
IH29
F
4
0.51
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH30 IH31
F M
4 4
0.49 0.48
0.51 0.50
⫺ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ U
⫺ ⫺
⫺ ⫺
IH32 IH33
V M
4 4
0.48 0.49
0.46 0.49
⫺ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ U
⫺ ⫺
⫺ ⫺
IH34*
F
4
0.44
0.54
RMS
U
⫺
⫺
⫺
⫺
⫺
⫺
Patient
M.I. H19
Birth weight >P97
M.I. KCNQ1OT1
Other features
Hypertrophy Right arm and leg
Left leg ⬎right, difference 0.5 to 1 cm at different places. Leg circumference right⬎left arm circumference right ⬎left Leg right⬎left Arm left⬎right Leg left⬎right Lower leg and foot, asymmetrical face right⬎left Left upper leg shorter and thinner, left foot shorter, asymmetrical face Leg left⬎right Face, arm and leg right⬎left Face, arm and leg right⬎left Leg right⬎left Arm, leg and foot left⬎right Lower arm, hand, leg and foot left⬎right Arm and hand right⬎left Leg right⬎left Arm and hand right⬎left Ear, arm and leg left⬎right
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
Appendix I. Continued Birth Abdominal Pits M.I. M.I. weight wall and Nevus Patient Sex Group KCNQ1OT1 H19 Tumor >P97 Macroglossia defects creases Hypoglycemia flammeus Nephromegaly IH35
F
4
0.47
0.49
⫺
U
⫺
⫺
⫺
⫺
⫺
⫺
IH36
F
4
0.51
0.52
⫺
⫹
⫺
⫺
⫺
⫺
⫺
⫺
IH37
M
4
0.48
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH38 IH39
F F
4 4
0.47 0.51
0.41 0.53
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫹ ⫺
⫺ ⫺
IH40
V
4
0.47
0.53
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫹
IH41 IH42
M M
4 4
0.50 0.50
0.51 0.5
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
IH43
F
4
0.500
0.50
⫺
U
⫺
⫺
⫺
⫺
⫺
⫺
IH44
M
4
0.48
0.50
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH45
M
4
0.47
0.47
⫺
⫺
⫺
⫺
⫺
U
⫹
⫺
IH46 IH47
M M
4 4
0.46 0.53
0.53 0.57
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
Patients with hemihyperplasia with additional major features IH⫹1 M 3 0.39 0.51 ⫺
IH⫹2
M
4
0.46
0.46
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
IH⫹3
M
4
0.49
0.51
⫺
⫺
⫺
⫺
⫺
⫺
⫺
⫺
Patients with hemihyperplasia with 1 or more BWS features BWS⫺1 F 1 0.22 0.78 ⫺ ⫺
⫺
⫹
⫺
⫺
⫺
⫺
Other features
Hypertrophy Left leg 4 cm longer (at 2 y) Left arm, leg, face ⬎ right Arm and leg right⬎left Right leg Face, arm, leg, hand and foot Face and thorax right⬎left Left leg ⬎ right Whole body left⬎right Face, arm and leg left⬎right Face, arm and leg right⬎left Leg and foot left⬎right Lower leg and foot circumference left⬎right
Cyanosis, advanced Right leg⬎ left skeletal age, prominent occiput, partial hair albinism, depigmentation/ hyperpigmentation skin of the neck. Syndactyly of dig 3-4 Body side right ⬎ left hand and dig left 1-2-3 left feet Mental retardation, Left arm⬎right prominent occiput Scoliosis
Left body side except face
100.e3
100.e4
Appendix I. Continued
Bliek et al
Macroglossia
Abdominal wall defects
Pits and creases
Hypoglycemia
Nevus flammeus
Nephromegaly
The Journal of Pediatrics • July 2008
Sex
Group
BWS⫺2
M
1
0.38
0.62
⫺
⫺
⫺
⫺
⫺
⫺
⫹
⫹
BWS⫺3 BWS⫺4
M F
2 3
0.47 0.38
0.65 0.51
⫺ ⫺
⫺ ⫺
⫺ ⫹
⫹ ⫺
⫺ ⫺
⫺ ⫺
⫺ ⫹
⫺ ⫺
BWS⫺5 BWS⫺6
F M
3 3
0.20 0.00
0.50 0.48
⫺ ⫺
⫹ ⫺
⫺ ⫹
⫺ ⫺
⫺ ⫺
⫹ U
⫹ ⫺
⫺ ⫺
BWS⫺7 BWS⫺8 BWS⫺9 BWS⫺10
F F M F
4 4 4 4
0.46 0.48 0.52 0.45
0.49 0.52 0.53 0.50
⫺ ⫺ ⫺ ⫺
⫹ ⫺ ⫺ ⫺
⫹ ⫺ ⫹ ⫹
⫺ ⫹ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
⫺ U ⫺ U
⫺ ⫺ ⫺ ⫺
⫺ ⫺ ⫺ ⫺
BWS⫺11 BWS⫺12
F M
4 4
0.51 0.45
0.49 0.48
⫺ ⫺
⫺ ⫺
⫹ ⫹
⫺ ⫺
⫺ ⫺
U ⫺
⫺ ⫺
⫺ ⫺
BWS⫺13
F
4
0.49
0.48
WT
⫺
⫺
⫺
⫹
⫺
⫹
⫺
BWS⫺14
F
4
0.49
0.49
⫺
⫺
⫺
⫹
⫺
⫺
⫺
⫺
Patient
M.I. H19
Birth weight Tumor >P97
M.I. KCNQ1OT1
Hypertrophy Other features Right upper arm and leg; right leg 2 cm longer; right kidney; right foot⬎left Vision problems, advanced skeletal age, underdeveloped maxilla, chubby face, prognathism Thin upper lip
Face, arm and leg length left⬎right
Seizures Seizures Feeding problems, apnea, polycythemia Plagiocephaly
Face, leg length, thorax, arm right⬎left Mammae and slightly in the face Leg: right⬎left
HH⫹, hemihyperplasia with other, not BWS related, features; BWS⫺, IH patients with one or more additional signs of BWS but not meeting the clinical criteria; M, male; F, female; MI, methylation index; bWT, bilateral Wilms tumor; U, unknown; nd, not done; ⫹, present; ⫺, absent; ⬎P97, Birth weight above 97 percentile. 1. nephrocalcinosis 2. 1st year nephroblastoma 3. hepatomegaly but HH on right side of the body *Also in study by Merks et al.7
Epigenotype, Phenotype, and Tumors in Patients with Isolated Hemihyperplasia
Appendix II. Epigenetic and clinical detail of 14 patients with cancer with IH
100.e5
Sex
Group
M.I. KCNQ1OT1
M.I. H19
Tumor
HT1
F
1
0.35
U
BWT
B
R⬎L
HT2
F
2
0.52
0.6
WT
R
R⬎L
HT3
M
2
0.46
0.63
RMS
L
L⬎R
HT4
F
2
0.49
0.6
WT
L
R⬎L
HT5
F
4
0.39
0.57
WT
L
R breast ⬎L, L leg ⬎ R
HT6
M
4
0.50
0.49
WT
L
L⬎R
HT7
F
4
0.44
0.54
RMS
L
L⬎R
HT8
F
4
0.49
0.51
WT
L
L⬎R
HT9
F
4
0.46
0.46
NBL
Median
R⬎L
Obesity, Convex nose
HT10
F
4
0.47
0.48
WT
R
L⬎R
Upward slant, Concave nose, Broad flat nasal bridge, Multiple nevi, Multiple café au lait spots
Number
Sidedness tumor
Sidedness IH
Beckwith features
Hernia inguinalis
Phenotype
Hypertrophy
Plagiocephaly, Asymmetric pupils, Concave nose, Wine mark Retrognathia, Scoliosis, Hyperlaxity joints
Hemihypertrophy right breast, Upper and lower limb Hemihypertrophy right upper and lower limb Left lower limb 2 cm thicker and 0.5 cm longer the right. L Hemihypertrophy right ear, mamma, upper and lower limb Asymmetric breasts, Hemihypertrophy right lower limb Hemihypertrophy left pectoral muscle, Hemihypertrophy left upper limb
Overhanging nasal tip, Thin upper lip, Caries, Sandal gap Fetal pads, Clinodactyly, Hypermobility large joints Diastema Clinodactyly, Multiple small nevi Deeply set eyes, Upward slant, Thin nasal bridge, Thin upper lip, Camptodactyly, Sandal gap, Café au lait (2 ⫻ 1 cm) Upward slant, Broad and deviated nasal tip, Hyperlaxity, Multiple small nevi Sparsely implanted hair, Ptosis
Hemihypertrophy left ear, upper and lower limb Hemihypertrophy left upper and lower limb Asymmetric mammae, Hemihypertrophy right upper limb, Asymmetric face, Slight hypotrophy right iris Asymmetric face, Hemihypertrophy left upper and lower limbs
100.e6
Appendix II. Continued
Bliek et al
Sex
Group
M.I. KCNQ1OT1
M.I. H19
HT11
F
4
0.50
0.51
WT
R
L⬎R
HT12
F
4
0.50
0.50
WT
R
L⬎R
HT13
F
4
0.49
0.53
WT
L
L⬎R
HT14
F
4
0.51
0.46
Sacrococcygeal teratoma
Median
R⬎L
Number
Tumor
Sidedness tumor
Sidedness IH
HT, Hemihypertrophy and tumor; bWT, bilateral Wilms tumor; L, left; R, right; b, bilateral; F, female; M, male; MI, methylation index.
Beckwith features
Phenotype Low anterior hairline, Upward slant, Epicanthal folds, Almond shaped eyes, Hemihypertrophy left lower limb, Café au lait spot (2.5 ⫻ 4 cm) Clinodactyly, Fetal pads, Multiple nevi (⬎0.5 cm) Blepharophimosis, Full lower lip, Overfolded helix, Multiple nevi Mental retardation, Cor vitium, Thorax vsd
Hypertrophy Hemihypertrophy left lower limb
Hemihypertrophy left lower limb Mamma hyperplasia, Hemihypertrophy lower limb Hemihypertrophy right upper and lower limb
The Journal of Pediatrics • July 2008