Familial Adenomatous Polyposis with Atypical Clinical Morphology and Genetic Variants

Abstract

Familial adenomatous polyposis (FAP) is caused by pathogenic variants of the APC gene on the long arm of chromosome 5. An analysis showed an association between germline APC gene variants and clinical signs of FAP; however, attenuated FAP has also been reported in cases with pathogenic variants. In contrast, a phenotype of FAP with no APC germline pathogenic variant and with few signs has been reported. We herein report a 16-year-old girl in whom the presence of multiple large bowel cancers from a young age and several small bowel cancers reflected a carcinogenic tendency higher than that typical for FAP.

Introduction

Familial adenomatous polyposis (FAP) is a genetic disorder characterized by multiple adenomatous polyps in the large intestine. Several cases of FAP are caused by pathogenic variants of the APC gene on the long arm of chromosome 5 (1,2). There have been reports of a relationship between germline APC gene variants and clinical signs of FAP (3-9). Attenuated FAP with mild adenomatous polyposis in the presence of a pathogenic variant of the APC gene has also been reported (7).

In contrast, a case of a FAP phenotype with no pathogenic variant of the germline APC gene was reported with a few signs, such as fundic gland polyps, duodenal adenoma, and osteoma (10). There has been another report of fewer than 100 colorectal adenomas with poor prognoses (11). Furthermore, some studies have found that dense adenomatous polyposis and duodenal adenoma are infrequent in this phenotype (8,12).

In the case described in the present report, the risk of carcinogenesis tended to be higher than that for typical FAP; however, the patient did not have a pathogenic variant of the APC gene based on a routine search, although an additional search revealed a variant of the lesion. Patients previously reported to have no pathogenic variant of the APC gene were considered to have similar risks, and the present case is reported with this consideration.

Case Report

A 16-year-old girl with mild intellectual disability was found to have anemia on a preoperative examination for plastic surgery for a subcutaneous mass (epidermoid cyst) on the right back and buttocks. She was subsequently followed as an outpatient at the pediatrics department while receiving oral iron treatment. She was referred to our department because of progressive anemia and a positive fecal occult blood test result. Regarding her family history, while her parents were divorced and her paternal family history unknown, there was no other family history of colon cancer or colon polyps. The patient may thus have been the proband.

Total colonoscopy showed ＞100 polyps of the nondense type distributed throughout the colorectum. Polypectomy was first performed for polyps larger than 2 cm, and 20 large polyps were resected. A histological examination showed that the polyps comprised 1 adenocarcinoma and 19 tubular adenomas.

Esophagogastroduodenoscopy (EGD) revealed multiple fundic gland and duodenal polyps (Fig. 1a). The pathological findings of the duodenal polyps were suggestive of malignancy. No malignant findings were observed in the gastric polyps. She tested negative for Helicobacter pylori.

Figure 1.

Diagnostic examinations of the patient. (a) EGD revealed multiple fundic gland polyps. No malignant findings were observed in the gastric polyps. The serum antibody test for Helicobacter pylori was negative. (b) CT revealed intussusception due to the development of a 4-cm sigmoid polyp. (c) The upright abdominal radiograph showed the advanced part of the incarceration. (d) Endoscopic resection was performed after endoscopic reduction. The pathological diagnosis was adenocarcinoma. (e) Balloon enteroscopy showed 22 polyps. Endoscopic resection for each polyp was performed. Three of the polyps were well-differentiated adenocarcinomas, and the rest were high-grade adenomas. (f, g) Duodenal papillectomy was performed for a 2-cm duodenal papillary tumor, and it was found to be a carcinoma in adenoma. EGD: esophagogastroduodenoscopy, CT: computed tomography

CT revealed intussusception due to a 4-cm sigmoid colon polyp 6 months later (Fig. 1b-d). Endoscopic resection was performed after endoscopic reduction of the incarceration. A pathological examination revealed adenocarcinoma. At the same time, 3 colon polyps ≥3 cm in size were resected and pathologically diagnosed as tubular adenomas.

Balloon enteroscopy was performed. The entire small intestine was observed through the oral and trans-anal routes using balloon endoscopy. In the jejunum and ileum, 22 polyps were subsequently resected; 3 were well-differentiated adenocarcinomas, and 19 were high-grade adenomas (Fig. 1e). No polyps were observed at the terminal ileum.

Duodenal papillectomy was performed on a 2-cm duodenal papillary tumor (Fig. 1f, g), and it was found to be a carcinoma in adenoma. Ultimately, ＞100 colonic polyps and 30 small intestinal polyps were resected within the year.

The patient provided her informed consent to take part in the multi-institutional clinical genomics study, “Diagnosis and Research on Familial, Young-onset and Hereditary Cancers” (protocol no. 2013-303, approved by the research ethics committee at the National Cancer Center). Her DNA was extracted from peripheral blood leukocytes using the FlexiGene DNA Kit (Qiagen, Hilden, Germany) and analyzed using the Agilent SureSelect target enrichment system (Agilent, Santa Clara, USA), a multigene panel based on hybrid capture sequencing. A multigene panel, NOP_FC ver 2.0, was designed to sequence exons and some intronic regions of 147 genes, which were selected by the authors for known or possible relationships with hereditary predispositions to cancers (Supplementary material). Sequencing was performed using a NextSeq500 (Illumina, San Diego, USA), and the data were analyzed and annotated using GATK Best Practice and GermlineCNVCaller (Broad Institute, Cambridge, USA). The Integrative Genome Viewer (IGV; Broad Institute) was used to visualize the alignment of the sequence reads to the reference sequence. Furthermore, multiplex ligation-dependent probe amplification (MLPA) analyses were performed for the APC (P043-E1) and MUTYH (P378-D1) genes according to the manufacturer's protocol (MRC-Holland, Amsterdam, the Netherlands).

The multigene panel did not detect any pathogenic or likely pathogenic single nucleotide variants (SNVs) or short indels in the genes related to colorectal polyposis syndromes. In addition, the MLPA did not detect any copy number abnormalities in the APC or MUTYH genes. However, GermlineCNVCaller predicted a 1,380-bp deletion spanning from the 3' end of exon 4 to intron 4 of the APC gene (NM_000038.6, NC_000005.10). In line with the results of GermlineCNVCaller, an inversion was identified on IGV from the middle of APC exon 4 to the repetitive long terminal repeat (LTR) sequence located at 51 Mb downstream of the gene (Fig. 2a). To validate the inversion, five “forward” polymerase chain reaction (PCR) primers were designed around the LTR sequence at the 3' end of the inverted region (Fig. 2a, upper right), and PCR with the APC intron 3 primer showed amplified fragments in the case but not in three control samples for the PCR1 and PCR2 primers (Fig. 2b). Direct sequencing of the PCR products by PCR1 or PCR2 confirmed the junction between exon 4 and the LTR sequence (Fig. 2c).

Figure 2.

Structural variant identified in the APC gene. A schematic presentation of an inversion detected first by the alignment of the sequence reads of a multigene panel, NOP_FC, by IGV (Integrative Genome Viewer, Broad Institute, Cambridge USA). (a) PCR1 to PCR5 are PCR primers for validations by agarose gel electrophoresis (b) and Sanger sequencing (c) of the PCR products spanning the junction between APC exon 4 and the LTR sequence located 51 Mb downstream of the APC gene. PCR: polymerase chain reaction, LTR: long terminal repeat

In previous reports, other than pathogenic variants of the APC gene, gene mutations that exhibit strong phenotypes in the gastrointestinal tract include polymerase proofreading-associated polyposis (PPAP) and constitutional mismatch repair deficiency (CMMRD). However, it is possible to point out these factors in the multifactor panel test; hence, the cause of the present case is believed to be a pathogenic variant in the APC gene.

Discussion

Pathogenic variants of the APC gene have been associated with FAP patterns. The frequency of pathogenic variants in the APC gene in patients with FAP is 60-80%. APC gene-pathogenic variant-positive FAP has been reported to cause colonic adenomatous polyposis, duodenal papillary adenoma, osteoma, retinal pigment epithelial hyperplasia, duodenal adenoma or cancer, small intestine adenoma or cancer, and hepatoblastoma (3-9).

Gastrointestinal signs in the present case included multiple cancers in the large intestine from a young age and several cancers in the small intestine. This represents a significantly stronger carcinogenic tendency than that of typical FAP. Nongastrointestinal signs included epidermal cysts and mild intellectual disability. Intellectual disability is rarely observed in typical FAP; hence, the present case is considered atypical FAP.

The phenotype of patients with APC gene-pathogenic variant-negative FAP is characterized by a few signs, such as fundic gland polyps, duodenal adenoma, and osteoma (10). In addition, the number of colorectal adenomas is ＜100, and the prognosis is poor (11). Dense adenomatous polyposis and duodenal adenoma have been reported to be less frequent in APC pathogenic variant-negative cases than APC pathogenic variant (8,12). Several cases without pathogenic variants of the APC gene have a mild phenotype, and no consistent trend has been reported (4,13,14).

In addition, malignant tumors (pancreatic cancer, gastric cancer, rectal carcinoid, etc.) that are rarely observed in FAP may occur in patients with APC pathogenic variant-negative FAP (15,16). These cases may include Lynch syndrome and MUTYH-associated polyposis, as well as other conditions related to splicing abnormalities of the APC gene and rare presentations, as in the present case. CMMRD, which is characterized by biallelic congenital pathogenic variants of the MMR gene, is also possible.

In a report of small intestinal lesions associated with FAP, 26.2% of 149 Japanese people with FAP had high-grade adenoma of the small intestine, and 5.4% had cancer (17). The incidence of small intestinal cancer is considered low for typical cases of FAP; however, the present case is considered atypical because multiple cancers were detected.

A 51-Mb inversion was identified as a cause of the pathogenic disruption of the APC gene at exon 4 in this case. The other end of the inversion at the genome coordinate chr5:164,099,476 does not seem to harbor a distinct protein-coding gene, and the impact of the inversion on the specific phenotype of the case other than FAP is not known. A further analysis, such as long-read sequencing, may be warranted to explore any pathogenic variations of the inverted sequence.

APC mutations at codons 1251-1580 and/or high Spigelman scores have been reported (18). However, endoscopic resections of 22 jejunal polyps in the small intestine were performed, and 3 were identified as well-differentiated adenocarcinomas, with the rest being high-grade adenomas. The Spigelman score for the present case was relatively high at 12, and the patient was diagnosed with stage IV disease. For stage IV high-grade atypical adenomas or high-grade adenomas, an evaluation of surgical indications or surveillance by a specialist every 6-12 months is recommended. As in the present case, cancerous transformation was observed in 7-36% of stage IV patients (19,20); thus, pancreaticoduodenectomy (PD), pyrorus-preserving pancreaticoduodenectomy (PPPD), or pancreas-sparing duodenectomy (PSD) should be considered. Whether endoscopic surveillance should be continued or surgical resection should be considered as a treatment policy for atypical FAP cases with multiple cancer lesions at a young age is controversial.

Endoscopic surveillance risks the late detection and diagnosis of cancer; however, aggressive intestinal resection also carries risks of its own, such as short bowel syndrome and adhesion and intestinal obstruction due to multiple surgeries. Even if the treatment policy is decided in consultation with the patient, once the tumor is removed, it cannot be reversed. There are no prospective studies on the validity of surveillance and treatment using these staging systems, and this is a topic for future research. Therefore, we believe that the treatment policy should be decided in consultation with the patient. Once the intestinal tract is resected, it cannot be undone; in the present case, we believe that careful follow-up with thorough endoscopic examination may avoid surgery.

Conclusion

We encountered a patient with a severe phenotype of familial adenomatous polyposis in which the APC gene-pathogenic variant was difficult to detect.

Written informed consent was obtained from the patient for the publication of their data.

The authors state that they have no Conflict of Interest (COI).

Financial Support

This research was supported in part by the Japan Agency for Medical Research and Development (AMED) under grant number JP22ck0106554.

Yoriaki Komeda and Hideki Ishikawa contributed equally to this work.

Supplementary Material

Supplementary Table

The list of the 147 genes analyzed by the multi-gene panel, NOP_FC ver 2.0, in this study.