Statistics from Altmetric.com
Editor—Familial adenomatous polyposis (FAP) is an autosomal dominant condition characterised by the presence of more than 100 adenomatous polyps in the colon and rectum. Polyps generally first appear in the second or third decade of life and are usually most numerous distally. Left untreated, colorectal cancer is virtually inevitable and generally arises in the fourth or fifth decade.1 Adenocarcinoma of the appendix is an uncommon neoplasm and has only rarely been reported in association with FAP.2
The gene responsible for FAP, APC, was initially localised to the long arm of chromosome five (5q) by linkage.3 4 This followed a case report describing carcinomas of the rectum and ascending colon, adenomatous polyposis, mental retardation, and various dysmorphic features in a 42 year old man with a constitutional deletion of 5q.5 Most patients with FAP have normal karyotypes.6 Mental retardation and dysmorphic features are unusual in such people but characterise those rare patients with cytogenetically visible 5q deletions and FAP.5-12 The few reports detailing the clinical findings in patients with submicroscopic deletions ofAPC suggest that such people may be mentally normal.13 14
In this report we describe a patient with adenomatous polyposis, mental retardation, and an apparently balanced translocation t(5;8)(q22;p23.1) causing submicroscopic deletion of APC andMCC.
Clinical data were obtained by review of medical records. In addition, the patient was interviewed and examined by two of the authors (JF and AS) before her death. Cytogenetic studies were performed using standard techniques on a 72 hour peripheral blood culture with GTG banding, as previously reported.6
Slides for fluorescence in situ hybridisation (FISH) were obtained using the cell suspension retained after routine cytogenetic harvest. RNAse treatment, probe and chromosomal denaturation, and hybridisation conditions were as previously described15 with the stringencies adjusted after assessment of the optimal conditions for each probe combination. The biotinylated probes were detected with avidin-FITC (Calbiochem) followed by biotinylated anti-avidin (Vector) and finally avidin-FITC. The digoxigenin labelled probes were detected with mouse anti-digoxigenin followed by sheep anti-mouse Ig-rhodamine conjugated antibody (Boehringer Mannheim). Chromosomes, counterstained with DAPI (Sigma) and visualised by fluorescence microscopy (Zeiss), were captured using a computer image analysis system (Cytovision).
The following probes were used in FISH studies: MCC 40CI (partialMCC cDNA nucleotides 1634-3969), FB70B (partial APC cDNA nucleotides 2877-6452), and APC (full length APC cDNA) each kindly provided by Dr Bert Vogelstein (Johns Hopkins Oncology Center, Baltimore); D5S23 (chromosome 5p probe, Vysis); EGR-1 (chromosome 5q31 probe, Vysis); CHR8B/wcp8 (chromosome 8 library probe, Cambio); D8Z1 (chromosome 8 centromere probe, Oncor).
Genomic DNA was extracted from blood samples using the Instagene purification matrix (Biorad) according to the manufacturer's instructions. Each polymerase chain reaction (PCR) used 50-100 ng genomic DNA, 50 pmol/l of each oligonucleotide primer, 0.2 mmol/l of each dNTP (Pharmacia), 1.25 U Amplitaq DNA polymerase (Perkin Elmer) in Amplitaqreaction buffer (10 mmol/l Tris-HCl, 50 mmol/l KCl, MgCl optimised for each primer pair, pH 8.3) to a final volume of 25 μl. Thirty five amplification cycles were performed using an FTS-320 Thermal Sequencer (Corbett Research).
An RsaI polymorphism in exon 11 ofAPC 16 was analysed by amplification of exon 11 and digestion of the PCR product withRsaI (Boehringer). Reaction products were separated by electrophoresis in a 10% non-denaturing polyacrylamide minigel (Biorad) and visualised by ethidium bromide staining. The 255 bp amplification product yielded digestion fragments of 155 bp and 100 bp. A CA repeat polymorphism within MCC(CAMBC)17 and a CA repeat polymorphism betweenAPC and MCC(LNS-CA)13 18 were labelled by [αS35]dATP incorporation during PCR and characterised by electrophoresis in a 6% denaturing polyacrylamide gel and autoradiography as previously reported.19
The patient (fig 1) was the second child of a 32 year old mother and a 33 year old father. Her older brother was well and mentally normal. She was born prematurely at 29 weeks' gestation after an uncomplicated pregnancy. There were no major problems in the neonatal period but she was slow to speak and did not walk until the age of 2 years. At that time she began to have generalised convulsions. These were only partially controlled by medication and continued until the age of 18. Her performance at school was poor. Her hospital file notes an estimated IQ between 70 and 80 at the age of 10 although the method of assessment was not recorded. Physical examination at this time showed crowded dentition and she later required extensive dental work. The posterior hairline was noted to be low. The head circumference was normal (75th centile) while height was on the 10th centile. The third and fourth toes were short with the fifth toe longer than the fourth. Bilateral genu recurvatum was evident and the patient was mildly ataxic. There were no skin lesions and the fundi were normal. Visual acuity and hearing were unimpaired. An EEG at the age of 4 was abnormal with excessive generalised slow wave activity. A cerebral CT scan at 9 years showed mild ventricular dilatation and localised atrophy in the frontal areas and bordering the interhemispheric commissure. She attended a special school until the age of 17 and was actively involved in sports. Subsequently she worked successfully in a sheltered workshop while living independently with a group of mildly handicapped people.
She presented acutely at 26 years of age with a 24 hour history of right iliac fossa pain. She was febrile and had guarding and rebound tenderness at the site of her pain. A diagnosis of acute appendicitis was made and an appendicectomy performed. The appendix contained a mucin secreting carcinoma arising in a dysplastic villous adenoma and invading through the full thickness of the muscle wall. A right hemicolectomy was performed. The appendiceal stump had foci of adenomatous change but there was no residual carcinoma. Numerous small adenomatous polyps were noted throughout the right colon but the exact number was not recorded. Sigmoidoscopy subsequently showed left sided polyps but these were less numerous than had been found proximally. Eight months after hemicolectomy she presented again with a painful right iliac fossa mass. Laparotomy confirmed local tumour recurrence which could be only partially excised. She was treated with 5-fluorouracil and folinic acid but died approximately 12 months later. A post mortem examination was not performed.
Cytogenetic analysis showed a female karyotype with a translocation involving chromosomes 5 and 8 at breakpoints q22 and p23.1 respectively (fig 2). The translocation was cytogenetically balanced. Both parents had normal karyotypes. The patient's brother refused testing.
FISH studies were performed using several probe combinations: APC, D5S23, and EGR-1; FB70B and D5S23; FB70B and D8Z1; MCC 40CI and CHR8B/wcp8 (fig 3); and MCC 40CI and D8Z1. These studies confirmed the translocation and showed submicroscopic deletion of bothAPC and MCC. TheAPC and MCCprobes hybridised only to the normal chromosome 5 and not to the der(5) or der(8) chromosomes.
To confirm that the translocation identified had resulted in deletion of the APC andMCC genes, intragenic polymorphisms were examined in DNA from the patient and her parents. TheRsaI polymorphism in exon 11 ofAPC was informative in the family. The patient's mother was homozygous for absence of the restriction site whereas her father was homozygous for presence of the restriction site. The patient had only a maternally derived allele, suggesting that she was hemizygous for this region of APC and that the translocation had involved the paternally derived chromosome (fig 4). The MCC CA repeat polymorphism was not informative in this family but the patient amplified only one allele consistent with hemizygosity for MCC. Both parents also shared this allele. Analysis of LNSCA, betweenAPC and MCC, was informative in the family. Again, the patient had inherited a maternal but not a paternal allele. Seven microsatellite polymorphisms from other chromosomes were examined in the three family members. The results did not suggest non-paternity.
There have been 10 previously described cases in eight families of cytogenetically visible deletions encompassing 5q22 in association with adenomatous polyposis of the colon and rectum.5 7-12Varying degrees of mental retardation and a range of dysmorphic features have been described in these patients (table 1). Others have reported patients with similar deletions, mental retardation, and congenital anomalies in whom colonic disease was either not sought or had not yet developed.11 20 21 Submicroscopic deletion of APC has been described in mentally normal subjects without dysmorphic features.13 In the current report, a submicroscopic deletion encompassingAPC and MCC has produced adenomatous polyposis in association with mild retardation and minor dysmorphic features. The mental retardation observed in this case may relate to the greater extent of the submicroscopic deletion. In our patient an area encompassing both APC andMCC was deleted and it is likely that other neighbouring genes have been either deleted or disrupted. In the family described by Mandl et al,13 the deletion spanned a smaller region including theAPC gene from exon 11 to the 3′ untranslated region and the adjacent DP1 gene.MCC was not involved.
We are aware of only one other report of FAP resulting from a chromosome 5 translocation.14 In that report an affected mother and daughter are described with colonic polyposis and a translocation t(5;10)(q22;q25) resulting in disruption of theAPC gene and an intragenic deletion. Neither mental retardation nor dysmorphic features were reported in these two patients.
In the current report, FAP was discovered after the patient presented with acute appendicitis. An appendiceal carcinoma was identified in the resection specimen. Appendiceal carcinoma has occasionally been described in patients with FAP but there is only one other published case of FAP first presenting as acute appendicitis with appendiceal carcinoma. Molecular genetic data were not reported in that case.2 In the current report, appendiceal carcinoma has developed in association with a submicroscopic deletion ofAPC and MCC and polyps were most numerous in the right hemicolon. Similarly, in the report of van der Luijt et al 14of a 5q translocation and polyposis, polyps were noted to be more numerous in the proximal colon of the mother and a caecal adenocarcinoma was found in her resection specimen. The reported distribution of colorectal polyps and carcinomas in patients with cytogenetically visible deletions of APC has been similar and more typical of attenuated adenomatous polyposis coli (AAPC) than of classical FAP. In AAPC, polyps generally develop later and tend to be more dense proximally than distally.22 The reverse is true of classical FAP.1 AAPC has been attributed in some cases to constitutional mutations at the 5′ end ofAPC while in others mutations in exon 9 or at the 3′ end of the gene have been reported.22-25 It has been suggested that the 5′ mutations result in very short protein products that, either through rapid degradation or lack of the necessary oligomerisation domains, do not compete with the residual wild type APC or bind to it, producing a dominant negative effect.14 23-25 Mutations at the 3′ end of APC may also produce null alleles although the explanation for this remains unclear.24Deletions of APC produce a true null allele. In this respect they may be functionally comparable to the 5′ and 3′ APC mutations in AAPC and this may explain some of the phenotypic similarities noted above.
The contribution of MCC deletion to colorectal carcinogenesis in the current case remains unclear. Constitutional deletion of APC is presumably sufficient to explain the adenomatous polyposis in this patient and there are no reported cases of constitutionalMCC mutation or deletion alone producing human colorectal disease. Somatic mutation ofMCC has been reported in colorectal cancers but the role of MCC as an independent colorectal tumour suppressor gene has subsequently been questioned.26 27
We thank the patient and her parents for their cooperation, the patient's surgeon, Dr John Neophyton, and general practitioner, Dr John Cameron, for providing clinical details, Dr Dorothy Painter for reviewing the histopathology, Cathy Abbott for assistance in preparing DNA probes, and Nicole Chia for assistance in the cytogenetic studies.