DOI: https://doi.org/10.21203/rs.3.rs-1924188/v1
Hereditary non-polyposis colorectal cancer (HNPCC) is a form of inheritable colorectal cancer. The condition is usually associated with mutations in DNA mismatch repair genes (MMR) such as MSH2 MLH1 MSH6 PMS2 or other genes. In this report of a family, none of associate gene were involved
The 56-year-old male presented with hematochezia and tenesmus for 5 months. The colonoscopy revealed a rectal tumor at 7 cm above anal verge. Then the tissue biopsy was done. The diagnosis was adenocarcinoma of the rectum of poor differentiation. The immunohistochemistry panel for MMR proteins showed diffusely strong positivity for MLH1 and PMS2 expression and weak positivity for MSH2 and MSH6 but screening for MMR mutations in the index case was negative. Whole exome sequencing was then performed using DNA from 4 family members, the index case, his spouse, their daughter, and their son.
The study identified a mutation on the MET proto-oncogene (c1124A > G) and/or PTCH1 (NM_00264.5: c.4045C > T) as pathogenic variant that explained the inheritance of HNPCC in this family in an autosomal dominant pattern. We concluded that the MET proto-oncogene is a candidate gene for HNPCC.
Approximately 5% of colorectal cancer (CRC) cases have a history of CRC within the same family which suggests a genetic influence on the risk of getting the disease(1). There are 2 main phenotypical patterns of inherited CRC known to date: familial adenomatous polyposis coli (MIM #175100) which is associated with variants in APC, and hereditary non-polyposis colorectal cancer (HNPCC), which is a phenotype associated with variants in various members of the mismatch repair gene (MMR) group(2). HNPCC is a clinical diagnosis defined by clinical criteria such as the Amsterdam Criteria II and the Revised Bethesda Guidelines(3), and HNPCC with molecular pathology in the MMR, either micro-satellite instability and/or deficient MMR protein expression is called Lynch syndrome (MIM #120435)(4). Other syndromes associated with inherited colorectal cancer are MUTYH-associated colorectal cancer, juvenile polyposis syndrome(5), and Peutz-Jeghers syndrome(6).
MMR refers to a group of genes functioning in DNA repair, hence they can be regarded as tumor suppressor genes. More than 80% of patients with Lynch syndrome have mutations in MSH2 and MLH1 while the rest have MSH6 or PMS2(7). In addition to these 4 genes, large deletion of EPCAM which situates upstream of MSH2 has been reported in Lynch syndrome patients(8). HNPCC is inherited in an autosomal dominant pattern and individuals with pathogenic MMR variants are estimated to have a life-time risk of developing CRC of 50–80%. Apart from CRC, patients with Lynch syndrome have a higher risk of developing extracolonic cancers including endometrial cancer and ovarian cancer(1). Apart from Lynch syndrome, more than 50% of patients who meet the criteria of HNPCC but no detectable MMR mutations (9, 10). Mutations in POLD1, POLE, which function in DNA proof reading, have been identified in a number of cases with this condition(11). However, the majority of MMR-negative HNPCC cases have no clear pathogenic genotype to explain the inheritance. In this study, we used next generation sequencing to annotate pathogenic variants that fit the vertical transmission in a family with inherited colorectal cancer. The study found no pathogenic variants in MMR and other putative candidates, but a candidate was found the variant of MET proto-oncogene in familial members with colonic pathology.
The index case was a 56-year-old male who presented with hematochezia and tenesmus for 5 months prior to a surgical consultation in November 2017. The patient had a history of stage-I squamous cell carcinoma at the lower lip which had been removed by a wide excision 6 years before. He also had essential hypertension and dyslipidemia. On colonoscopy, a rectal tumor was seen at 7 cm from the anal verge, from which the rectum had nearly complete obstruction. A tissue biopsy lead to a diagnosis of adenocarcinoma of the rectum of poor differentiation. The immunohistochemistry panel for MMR proteins showed diffusely strong positivity for MLH1 and PMS2 expression and weak positivity for MSH2 and MSH6 (Fig. 1). A computerized tomography study showed no distant metastasis. Magnetic resonance imaging of the rectum revealed an asymmetrical circumferential rectal mass of 1.1 cm thickness and 6.0 cm length. The muscularis propria hyposignal T2W was lost, suggestive of perirectal fat invasion and stage T3N0M0. No colonic polyps were seen. The patient was treated with upfront chemoradiation therapy before undergoing a laparoscopic low anterior resection. Histology of the surgical specimens showed no residual tumor at the rectum and no metastasis in 9 lymph nodes (pT0pN0cM0). As indicated by the histopathology, post-operative chemotherapy (Mayo regimen) was given. On follow-up, the patient was tumor-free at 3 years after the surgery.
Three relatives of the patient also had had colorectal cancer: his mother, his elder brother and his daughter. His mother was diagnosed with CRC at 55 years of age and died from cerebrovascular disease in the same year. The brother was diagnosed with rectal cancer at 55 years and the daughter at 20 years. Histopathology of the rectal cancer in the brother indicated poorly differentiated adenocarcinoma without polyps. The daughter was diagnosed with colonic adenocarcinoma together with ovarian cancer which was treated surgically at another hospital. Unfortunately, we have no histopathological details of the cancer in the daughter or the mother of the patient. Apart from the daughter, the patient also had two others sons, whose colonoscopies were negative. According to the pedigree of this family (Fig. 2), the mode of inheritance was likely autosomal dominant.
Blood specimens were collected from the index case, his brother, his wife, and all 3 of their children. Collection of the biospecimens and the genetic studies were all performed following informed consent. The study was approved by the Human Research Ethics Committee, Faculty of Medicine, Prince of Songkhla University (REC.65-084-10-1). For the genomic studies, DNA was prepared from a peripheral blood sample using a High Pure PCR Template Preparation Kit (Roche, Berlin, Germany). The extracted DNA was qualified and quantified by Bioanalyzer (Agilent Technologies, Santa Clara, California, United States) and Nanodrop (Thermo Scientific, Delaware, United States). Whole exome studies were performed in the index case, the wife, 2 brother and one daughter. Coding regions were captured and enriched using an Agilent Sure Select XT Human All Exon V6 machine. Paired-end exome sequencing was done using an Illumina HiSeq-2000 platform (Illumina, California, United States). The FASTQ samples were qualified by FastQC and quality control was verified using a Trimmomatic machine following the criteria of an average base quality score of more than Q30 and a read length of at least 50 bp. Sequences were mapped with the human reference genome GRCh38 by Burrows-Wheeler Aligner (BWA-MEM) then pre-processed (sort, add read group, mark duplication, base quality score recalibration) with SAMtools, Picard and Genome Analysis Toolkit (GATK), then the mapping statistics were calculated for the aligned sequences, with coverage at 40x of more than 90%. Variants were searched for using a GATK-Haplotypecaller with intermediate genotype (gVCF) mode individually, then representative samples including the index case, his brother, his daughter and his wife were merged together and called for genotype, separately. Variants were filtered out by machine learning using a gaussian mixture model with 99% truth sensitivity for selecting true variants by GATK-Variant Quality Score Recalibration (VQSR). For passed variants, functional annotation was interpreted by SnpEff and Variant Effect Predictor (VEP). Possible pathogenic variants among this family were prioritized, when they were non-synonymous variants found only in the CRC cases (index patient, the daughter and the brother, but not found in the wife), known pathogenic genes reported in colorectal cancer and/or familial cancer, allele calling depth of more than 25%, total read depth of more than 40x, and minor allele frequencies of less than 0.05 in the East Asian population. Supporting genotype-phenotype correlation data of the selected variants used the Human Gene Mutation Database, NCBI: ClinVar and Online Mendelian Inheritance in Man (OMIM). Beginning with well-known MMR genes (MLH1, PMS2, MSH2 and MSH6), we found no remarkable variants in these genes, nor in other genes reported in hereditary colorectal cancer(12). Two pathogenic variants passes the filter criteria: 1 non-synonymous MET variant located on chromosome 7 at position 7:116700208 (c.1124A > G), resulting in a missense mutation (p. Asn375Ser), and a non-synonymous variant in PTCH1 located on chromosome 9 at position 9:95447211 (c.4045C > T), resulting in a missense mutation (p.Pro1349Ser). The heterozygous MET and PTCH1 mutations were confirmed in all CRC cases in this family by direct nucleotide sequencing method (Fig. 3). Both sons’ DNA also were made by direct nucleotide sequencing method. We found only PTCH1 mutation, no MET mutation was detected in both sons.
Familial colorectal cancer is one of the hereditary diseases that are worth thorough investigation. With knowledge of causative genes, identification of an inherited offspring can be done precisely, leading to cancer prevention through early screening, chemoprevention and prophylactic surgery(13). While adenomatous polyposis coli usually has an apparent phenotype in the colon, HNPCC requires clinical criteria to select a CRC patient for further genetic work-up and the 2 most accepted clinical criteria currently used are the Amsterdam II and Revised Bethesda criteria(9). Previously, a diagnosis of Lynch syndrome was usually made when there was evidence of micro-satellite instability or deficiency in MMR expression in a CRC patient with a relevant familial history(14). However, in recent years genetic study techniques have entered the era of next generation sequencing, and pathogenic variants in the MMR or EPCAM have become of more interest. Because the clinical criteria are not sensitive enough to include all CRC cases with genetic risk, the idea of universal screening has been proposed by various institutions (15–17). With universal genome testing, various studies have reported detecting pathogenic variants in MMR in 0.2–2.2% of CRC cases, although 2.6-5.0% fulfilled the Amsterdam II criteria(9, 10, 18). Beginning from CRC who fulfilled Amsterdam criteria, 71% of the cases had micro-satellite instability, and 44% had pathogenic variants in MMR(19). The gap between the incidence by clinical criteria and incidence of MMR variants might be explained by the fact that there might be other germline variants involved in familial CRC and this group of patients is regarded as familial CRC type X syndrome.
Germline mutations in RPS20, SEMA4A, HNRNPA0, and WIF1 have been reported in familial CRC type X(11). In addition, genes involved in other hereditary cancer syndromes such as BRCA1 and BRCA2 which are found in hereditary breast-ovarian cancer syndrome can be found with familial CRC. In our study, no pathogenic or 'likely pathogenic' variants were detected in the MMR gene group or the DNA proof reading group. The novel mutation on PTCH1 found in our index case and his affected family was not in the ClinVar or dbSNP databases. However, its predicted amino acid alteration from proline to serine was at the same codon as rs773298544 (NP_000255.2: p.Pro1349Leu) which has been reported to be associated with Gorlin syndrome in the ClinVar database. However, as protein structure prediction tools suggested the missense mutation is in a 'tolerated' category, the variant is regarded as a 'variant of uncertain significance' (as evaluated in Mar 2019). Although the index case previously had skin cancer, the histological type of squamous cell was not compatible with basal cell carcinoma found as the main phenotype in Gorlin syndrome.
The mutation on the MET proto-oncogene has been annotated before as rs33917957, a non-synonymous polymorphism (Asn375Ser) that has the highest minor allele frequencies at 0.09 and 0.07 in the South Thai Genome database (https://tmrc.psu.ac.th/__genetics/). The MET proto-oncogene encodes a tyrosine kinase receptor c-MET (hepatocyte growth factor receptor) that, when induced with hepatocyte growth factor (HGF), relays a mitogenic signal into the nucleus via PI3K/AKT, Ras/MAPK, Wnt/β-catenin, JAK/STAT, SRC, and other signaling pathways(20). Activating mutations of MET leads to transactivation of growth-related genes by these pathways, especially MAPK. Germline mutations of the MET proto-oncogene were first reported in hereditary renal papillary carcinoma (HRPC)(21). MET mutations have also been detected in thyroid cancer, gastric cancer, lymphoma and colorectal cancer(22–25). While most reported somatic mutations of MET occurred in the intracellular domain, germline mutations also have been reported in the extracellular Sema and PSI domains, although with possible incomplete penetrance(26). The mutation at codon 375 in our patients occurred in the Sema domain of the receptor which functions in dimerization of the protein when stimulated by HGF(27). Our study revealed that 2 descendants of the index case had also inherited the PTCH1 mutations but not inherited the MET mutation and no pathology was detected in their gastrointestinal tract. So, the MET mutation is the main candidate for pathogenic variant in this family. However, these negative findings might be explained by their age and these cases need to be closely followed up if the PTCH1 mutations is the pathogenic variant.
We report a family with hereditary CRC whose phenotype was compatible with HNPCC, but genome sequencing revealed no mutations in the MMR gene group. Bioinformatic analysis showed candidate mutations on the MET and PTCH1 proto-oncogene.
HNPCC: Hereditary non-polyposis colorectal cancer; DNA: Deoxyribonucleic Acid; MMR: Mismatch repair genes; CRC: Colorectal cancer; PCR: Polymerase chain reaction; GATK: Genome Analysis Toolkit; VEP: Variant Quality Score Recalibration; VEP: Variant Effect Predictor; OMIM: Online Mendelian Inheritance in Man; HGF: Hepatocyte growth factor; HRPC: Hereditary renal papillary carcinoma
Ethics approval and consent to participate:
This case report was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. This study protocol was reviewed and approved by Human Research Ethics Committee at Price of Songkhla University, approval number REC.65-084-10-1. Informed written consent to publish this case including details of medical case, pathology images and genomic data were obtained from the patient.
Consent for publication:
Not applicable.
Availability of data and materials:
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests:
The authors declare that they have no competing interests
Funding:
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors
Authors' contributions:
Natthapon Khongcharoen, Jirakrit Saetunga and Wison Laochareonsuk contributed for collecting samples, data analysis. Teeranut Boonpipattanapong contributed for clinical data collection. Kanita Kayasut contributed for pathological data. Natthapon Khongcharoen and Surasak Sangkhathat contributed for study concepts, experimental quality control and paper writing. All authors read and approved the final manuscript.
Acknowledgements:
Not applicable.