Familial adenomatous polyposis (FAP) is a rare condition, with a prevalence of approximately 1:10,000, and is inherited as an autosomal dominant trait with high penetrance. There is also a relatively high spontaneous mutation rate, which may account for as many as 15% to 20% of cases (1). This may be a relative overestimate, given the possibility of undiscovered nonpaternity and the difficulties inherent in obtaining an accurate family history. Furthermore, both in classic forms of FAP but even more commonly in attenuated forms (<100 colorectal adenomas), biallelic mutations in the base excision repair gene MYH have been discovered recently, which are associated with an autosomal recessive inheritance and would be missed by conventional screening (2).
The gene responsible for FAP, adenomatous polyposis coli (APC) is located on chromosome 5q21 and is a tumor suppressor gene, playing a key role in regulating levels of β-catenin. In patients with FAP, multiple adenomas typically develop throughout the large bowel; this may begin to occur in adolescence. By the 5th decade, colorectal carcinoma is almost inevitable if colectomy is not performed; duodenal carcinoma may also develop. Other extracolonic manifestations have been described in patients with an APC mutation, including hepatoblastoma (HB) and desmoid tumors in children (3,4).
A recent study has suggested that even in the absence of a family history of FAP, approximately 10% of children with a diagnosis of HB may have a germline APC mutation (5). HB is the most common primary tumor in childhood. Treatment involves a combination of preoperative and postoperative chemotherapy and surgical management, which includes partial hepatectomy or liver transplantation for tumors that are unresectable (6). With this combined approach, the overall 5-year survival rate has dramatically improved and now exceeds 80% (7). In light of the improving prognosis for children with HB, we aimed to assess the frequency and type of APC mutation in sporadic HB, to investigate whether routine mutation screening should be a part of our evaluation at the time of diagnosis.
MATERIALS AND METHODS
King's College Hospital is a supraregional referral center for the assessment and surgical treatment of children with HB. Resection of the tumor or transplantation, including the living related donor option with preoperative and postoperative chemotherapy according to the pan-European protocol, is the mainstay of treatment (6,7).
A retrospective review of our database of children with HB was performed in a search for patients who had received a diagnosis of HB between 1991 and 2004. All of the children included in this study had a radiologically and histologically confirmed diagnosis of HB. Blood samples, banked with parental informed consent, were collected where available on patients presenting with HB during the specified study period. Ethical permission was received for the study from the King's College Hospital Research Ethics Committee.
DNA was extracted by use of a commercially available kit (DNA Microextraction Kit, Stratagene Inc, USA). The DNA was then examined in duplicate for germline APC gene mutations and compared with 4 known APC mutation-positive and 4 APC mutation-negative samples by use of the Light Scanner (Idaho Technology, USA). The Light Scanner is a plate-read, high-throughput, automated mutation detection scanner that identifies mutant samples on the basis of differences in fluorescence of a dye bound to the DNA sample in a simple polymerase chain reaction. Any alteration in DNA sequence alters the denaturation point of the DNA molecule when heated, which can be detected and compared with other samples (8). Any abnormal samples are therefore identified quickly and can be sequenced, with significant savings in time and cost. This was particularly relevant to this study, inasmuch as the APC gene is one of the largest genes in the human genome (2843 codons) and the quantity of DNA was limited by the small sample volumes taken from these children.
Blood samples were available from 29 patients (18 male) with sporadic HB, with a median age at HB diagnosis of 22 months (range 6–119 months). There was no family history of FAP or early onset colorectal carcinoma. All of the children had an elevated serum α-fetoprotein at presentation.
Of the patients sequenced, 4 (13%) were found to have the same point mutation (transition of purine G>A) within codon 1493, in region 9 of the mutation cluster region in exon 15 of the APC gene (codons 1286–1513) (Fig. 1). This nucleotide transition results in a different codon at this locus. However, owing to redundancy within the human genetic code (whereby 64 possible codons code for only 20 major amino acids), this G>A transition does not result in a change in the amino acid coded for (threonine –T), and hence there is no alteration in the APC protein product. Therefore, no functional or phenotypic difference would be caused, and these 4 HB samples represent synonymous (neutral) mutations, conferring no higher or lesser HB risk on the affected individuals.
Analysis on the publicly available human genome browser (www.genomebrowser.org) confirmed this to be nonpathogenic and previously described. Screening of the remainder of the APC gene failed to reveal any additional mutations.
Our study did not find any evidence of germline APC mutations in patients presenting with sporadic HB to a tertiary center in the United Kingdom. Hence, our results do not support the routine screening of patients with sporadic HB for germline APC mutations, unless there is a strong clinical suspicion of FAP, such as a suggestive family history of early-onset colorectal cancer or the other tumors described in association with FAP. Consequently, in the absence of these factors there seems not to be a role for routine colonoscopy (which would not be performed until the teenage years in the absence of symptoms, inasmuch as adenomas are rarely found in early childhood).
Where there is a positive family history of FAP in a child with HB, ophthalmological assessment in a search for congenital hypertrophy of retinal pigment epithelium (CHRPE) can be useful, particularly in the 20% of families in which no germline mutations can be identified (9). If CHRPE is present, it has almost 100% positive predictive value for APC mutations. However, it is important to recognize that the absence of CHRPE cannot be assumed, particularly in non-FAP kindreds, to have excluded APC mutations (10). We did not prospectively assess for the presence of CHRPE in our study population.
It is important that awareness of the association between HB and FAP among clinicians is reinforced so that relevant questions regarding a pertinent family history can be posed in children presenting with HB. This will obviously have potential implications not just for the child with a diagnosis of HB but also the rest of the family. Diagnosis of an APC mutation will necessitate genetic counseling and screening of the index patient's family and potentially lifelong screening in other affected family members. Furthermore, identification of a specific mutation may inform the clinician of the likelihood of extracolonic disease because there seems to be a relation between mutation site and extracolonic manifestations of FAP (11).
Hepatoblastoma is the most common primary liver tumor in the pediatric age range and characteristically occurs in the absence of chronic liver disease. Our patients with HB showed the standard male overrepresentation (62%) and typical young age at diagnosis that is seen in most series of HB (12). Nonetheless, the results of our mutation analysis seem to be different from those that have been described in a seemingly comparable study from the German group; however, the difference between the APC detection rates in the 2 studies is not statistically significant.
Previous studies have suggested variable mutation rates (5,13–15). Some authors have assessed somatic mutations in tumor blocks and in germline mutations (13). This is of course useful in advancing our understanding of the etiopathogenesis of HB and is of significant interest academically. This procedure was not performed in our study, however, and should not influence screening policies in a search for germline mutations. Other studies may not have included those mutations that do not affect β-catenin expression or function (14).
It could be speculated that because of the large wider population from which our children are referred, and also the fact that we are a tertiary pediatric liver unit without a specific research interest in colorectal malignancy molecular genetics, our study population represents an unequivocally sporadic HB group. Some “new mutation” patients with FAP would be expected to present with HB, but our series suggests that this is less common than has been reported. Nonetheless, further larger collaborative studies are required to clarify the situation before routine screening is to become mandatory. Even in the absence of germline APC mutations, further searches for evidence of somatic mutations in the tumor itself, and a search for possible candidate genes for study, including somatic β-catenin mutations and AXIN2 gene mutations, would help our understanding of the etiopathogenesis of the tumor.
1. Rustin RB, Jagelman DG, McGannon E, et al
. Spontaneous mutation in familial adenomatous polyposis
. Dis Colon Rectum 1990; 33:52–55.
2. Russell AM, Zhang J, Luz J, et al
. Prevalence of MYH germline mutations in Swiss APC mutation-negative polyposis patients. Int J Cancer 2006; 118:1937–1940.
3. Hirschman BA, Pollock BH, Tomlinson GE. The spectrum of APC mutations in children with hepatoblastoma
from familial adenomatous polyposis
kindreds. J Pediatr 2005; 147:263–266.
4. Clark SK, Pack K, Pritchard J, et al
. Familial adenomatous polyposis
presenting with childhood desmoids. Lancet 1997; 349:471–472.
5. Aretz S, Koch A, Uhlhaas S, et al
. Should children at risk for familial adenomatous polyposis
be screened for hepatoblastoma
and children with apparently sporadic hepatoblastoma
be screened for APC germline mutations? Pediatr Blood Cancer 2006; 47:811–818.
6. Avila LF, Luis AL, Hernandez F, et al
. Liver transplantation for malignant tumours in children. Eur J Pediatr Surg 2006; 16:411–414.
7. Czauderna P, Otte JB, Aronson DC, et al
. Guidelines for surgical treatment of hepatoblastoma
in the modern era: recommendations from the Childhood Liver Tumour Strategy Group of the International Society of Paediatric Oncology (SIOPEL). Eur J Cancer 2005; 41:1031–1036.
8. Chou LS, Lyon E, Wittwer CT. A comparison of high-resolution melting analysis with denaturing high performance liquid chromatography for mutation scanning: cystic fibrosis transmembrane conductance regulator gene as a model. Am J Clin Pathol 2005; 124:330–338.
9. Ruhswurm I, Zehetmayer M, Dejaco C, et al
. Ophthalmic and genetic screening in pedigrees with familial adenomatous polyposis
. Am J Ophthalmol 1998; 125:680–686.
10. Bertario L, Bandello F, Rossetti C, et al
. Congenital hypertrophy of retinal pigment epithelium (CHRPE) as a marker for familial adenomatous polyposis
(FAP). Eur J Cancer Prev 1993; 2:69–75.
11. Bertario L, Russo A, Sala P, et al
. Multiple approaches to the exploration of genotype-phenotype correlation in familial adenomatous polyposis
. J Clin Oncol 2003; 21:1698–1707.
12. Ang JP, Heath JA, Donath S, et al
. Treatment outcomes for hepatoblastoma
: an institution's experience over two decades. Pediatr Surg Int 2007; 23:103–109.
13. Koch A, Denkhaus D, Albrecht S, et al
. Childhood hepatoblastomas frequently carry a mutation degradation targeting box of the beta-catenin gene. Cancer Res 1999; 59:269–273.
14. Oda H, Imai Y, Nakatsuru Y, et al
. Somatic mutations of the APC gene in sporadic hepatoblastomas. Cancer Res 1996; 56:3320–3323.
15. Cetta F, Mazzarella L, Bon G, et al
. Genetic alterations in hepatoblastoma
and hepatocellular carcinoma associated with familial adenomatous polyposis
. Med Pediatr Oncol 2003; 41:496–497.