Neoplasia and cancer pathogenesis posters
PWE-067 Recessively inherited non-polyposis colorectal cancer: genotype and phenotype
K J Monahan1,2, K Pack2, C Cummings1, H J W Thomas1, I P M Tomlinson2
Introduction: Patients diagnosed before 50 years of age have a likely strong genetic or environmental aetiological factor. There is good evidence from population studies1 that recessive inheritance is common in young colorectal cancer patients.
Methods: A cohort of 133 colorectal cancer patients were diagnosed under the age of 50 years who did not have multiple polyps or a family history suggestive of dominant inheritance. They were identified and recruited from the Bobby Moore Database in the Family Cancer Clinic, St Mark’s Hospital, Harrow. MUTYH was screened for germline mutations. As these patients fulfilled Bethesda criteria they were tested for hereditary non-polyposis colorectal cancer (HNPCC) by microsatellite instability analysis and immunohistochemistry of mismatch repair proteins. Immunohistochemistry was also performed on β-catenin and P53. Loss of heterozygosity of the APC locus at 5q21–22 was tested using a set of microsatellite markers. Sequencing was used to identify somatic mutations in KRAS and BRAF.
Results: Forty-four patients (33%) had cancers proximal to the splenic flexure, 79 (59%) distal and had 11 (8%) synchronous colorectal cancers. Thirty-seven patients (28%) had an affected sibling and 33 (25%) patients had a second-degree relative with cancer at any site. The median age of diagnosis of colorectal cancer was 39 years (range 14–49 years of age). Twenty-six patients (20%) were found to harbour sequence variation in the MUTYH gene but none of these variants were likely to be pathogenic, and there was no difference in the frequency of these compared to a control group of 50 patients. Eighty percent of tumours were found to be microsatellite stable. 20/30 cancers had nuclear localisation of β-catenin and 21/30 had nuclear localisation of P53 antibodies on immunohistochemistry. Loss of heterozygosity of the APC locus at 5q21–22 was present in 14/30 cases. Thus Wnt pathway activation is likely by over half of this group of cancers. Four cancers had BRAF V600E mutations and five had KRAS codon 12 or 13 mutations.
Conclusion: In a cohort of 133 young colorectal cancer patients without multiple polyps, most tumours demonstrated Wnt pathway activation and other somatic changes consistent with the classical adenoma-to-carcinoma sequence. Germline mutations in the colorectal neoplasia predisposition gene MUTYH appear to be rare events in such patients. The majority of recessive inheritance in young patients is probably caused by mutations in unknown predisposition genes.
Colorectal Cancer Development
Colorectal cancer develops via an adenoma to carcinoma sequence with the accumulation of a number of genetic and epigenetic mutations (Figure 1‑3) (Morson 1968; Fearon and Vogelstein 1990). The mutations accumulated vary in hereditary cancer depending on the initiating mutation. In their normal state, tumour suppressor genes inhibit cell proliferation. Growth inhibition is lost when both alleles are inactivated by mutation and/or epigenetic changes, such as promoter methylation which stifles expression of the gene. Tumour suppressor genes broadly conform to Knudson’s classic two-hit hypothesis, where inactivation of both alleles is required for tumour suppressor genes to lose their normal function (Knudson 1971). In contrast, proto-oncogenes act by promoting cell proliferation. Mutation of these genes leads to abnormal oncogenic over-expression or increased activity of the protein.
The adenoma-to-carcinoma sequence for colorectal cancer is probably most commonly initiated by bi-allelic mutation of the APC tumour suppressor gene. APC mutations have been found in microadenomas (Otori et al. 1998), the earliest lesion on the pathway (also called aberrant crypt foci (Roncucci et al. 1991)), and in ~60-80% of early sporadic adenomas and carcinomas (Cottrell et al. 1992; Miyoshi et al. 1992; Nakamura et al. 1992). APC is a key member of the canonical Wnt signalling pathway, and the key mechanism by which mutation of this gene contributes to carcinogenesis is by activation of this pathway. However, further accumulated mutations in additional genes are required for progression of the early lesions to cancer.
Activating mutations of the oncogenes KRAS (Kirsten rat sarcoma viral oncogene homolog) and BRAF (v-raf murine sarcoma viral oncogene homolog B1), both members of the MAPK (mitogen activated protein kinase) signalling pathway, are found in the transition from early to an intermediate lesion in approximately 50% and 10% of cases respectively (Bos et al. 1987; Rajagopalan et al. 2002; Yuen et al. 2002). Mutations of codons 12 and 13 in exon 2 of KRAS tend to occur in 30-60% of colorectal carcinomas (Kressner et al. 1998). The KRASgene product,a 21 kDa protein located at the inner plasma membrane, is involvedin the transduction of mitogenic signals. The Ras protein isactivated transiently as a response to extracellular signalssuch as growth factors, cytokines and hormones that stimulate cell surface receptors (Campbell et al. 1998), and mutations in KRAS constitutively activate the Ras protein.
The substitution mutation of BRAF V600E is present in 4-12% of unselected colorectal tumours, and it is associated with sporadic MSI tumours but not HNPCC tumours (Vandrovcova et al. 2006). MSI tumours outside the context of HNPCC are usually caused by methylation of the MLH1 promoter. Indeed, there is a hypothesis that some tumours might develop through a separate hyperplastic polyp-serrated adenoma pathway (Spring et al. 2006).
Epigenetic changes such as promoter hypo-/hyper-methylation can cause disregulation of expression of many genes important in colorectal cancer (Hitchins et al. 2005; Hitchins et al. 2006). Further progression to late type adenoma is associated with loss of 18q in 50% of large adenomas and 75% of carcinomas (Vogelstein et al. 1988; Fearon et al. 1990). This causes loss of SMAD2 and SMAD4, members of the TGF-ß signalling pathway. Point mutations of these genes have also been identified in colorectal cancer (Eppert et al. 1996; Hahn et al. 1996; Thiagalingam et al. 1996). The adenoma to carcinoma transition appears to be associated with loss of 17p (Fearon et al. 1987; Rodrigues et al. 1990; Akiyama et al. 1998). The 17p locus contains the P53 gene (Baker et al. 1990), the so-called gatekeeper of the cell which has important roles in the regulation of the cell cycle and apoptosis. Loss of heterozygosity of 17p correlates with missense and truncating mutations in P53 (Baker et al. 1989; Baker, Preisinger et al. 1990). Tumour invasion and metastasis are associated with loss of 8p (Hughes et al. 2006), and loss of E-cadherin function, a component of adherens-junctions (Hao et al. 1997; Christofori and Semb 1999).
The progression from adenoma to invasive carcinoma probably takes 10-40 years (Ilyas et al. 1999). However, not all lesions will undergo malignant transformation, the reason for which is unclear. It may be that the necessary mutations do not accumulate because of death, or because of environmental influences such as diet.
Colorectal cancer may be subdivided genetically by the types of mutations which accumulate genome-wide during carcinogenesis. It had been observed nearly a century ago that most cancers were aneuploid, and it has been noted that the degree of aneuploidy in colorectal cancer correlates with the severity of the neoplastic behaviour (Heim and Mitelman 1989). A series of deletions, duplications, and rearrangements occur. Allelic losses appear to be important in the progression from premalignant to malignant neoplasia in the colorectum. This process is called chromosomal instability (CIN) and accounts for approximately 75% of colorectal cancers. Ten to 15% are not CIN but do have smaller mutational events which are caused by loss of DNA mismatch repair (Fishel et al. 1993) and are referred to as MSI tumours. The CpG island mutator phenotype (CIMP) is associated with methylation of promoter regions, CpG rich regions, which causes silencing of genes. Tumours are often both MIN and CIMP, as methylation of mismatch repair gene promoters usually occurs in sporadic MSI tumours (Kane et al. 1997).
The role of genomic instability in causing and promoting tumour growth remains controversial (Lengauer et al. 1998; Tomlinson and Bodmer 1999). Some argue that instability is necessary for tumourigenesis (Loeb 1991), while others take the viewthat Darwinian selection is the driving force. It is becoming clear that many cancers harbour multiple mutations, the great majority of which probably have no significant effect on tumour growth. It may well be that some tumours with an inherited DNA repair defect accumulate more mutations than others.
The hyperplastic polyp and serrated adenoma pathway
The first series of mixed hyperplastic-adenomatous polyps were described in 1990 (Longacre and Fenoglio-Preiser 1990), and have been an increasingly recognised phenomenon. Most hyperplastic polyps have no malignant potential, although there is now some have malignant potential, especially those with serrated architecture (sessile serrated adenomas – SSAs), large hyperplastic polyps, mixed polyps and polyps on the right side of the colon (Torlakovic et al. 2003).
Classification of hyperplastic polyps
A new understanding of the clinical relevance of hyperplastic polyps has emerged over the past decade (Young and Jass 2010). The simple hyperplastic polyp has itself been subclassified into a goblet cell variant and a microvesicular variant, the latter appear to be the precursors for serrated adenomas and thus colorectal cancer. Serrated adenomas usually have serrations low in the crypts which help differentiate them from hyperplastic polyps.
However, serrated polyps also include a broader spectrum of polyp subtypes ranging from these small common lesions to the recently described sessile serrated adenoma (SS
A), which is often large and proximal with abundant mucin secretion, exaggerated serration, and atypical architecture. Rarer serrated polyp subtypes with unequivocal dysplasia include traditional serrated adenoma (SA), which combines the dysplastic features of an adenoma with the architectural features of a hyperplastic polyp and the mixed polyp (MP) in which separate hyperplastic and dysplastic elements are combined within a single polyp (see figure). SSA, SA, and MP are described as “advanced serrated polyps” and comprise ∼5% of all serrated polyps retrieved in colonscopy patients. Importantly, these advanced serrated lesions show frequent BRAF mutation and widespread DNA methylation.
Endoscopic appearances of serrated adenomas
Serrated and hyperplastic polyps present endoscopic features that could help to differentiate them from adenomatous polyps. HPs appear flat and pale and are often covered by a thin film of mucus. They exhibit Kudo type 2 pit pattern typically. As they are not highly vascular they will appear pale compared to surrounding mucosa using narrow-band imaging (NBI). In addition, chromoendoscopy may be helpful in the endoscopic characterisation of these lesions.
Inherited Colorectal Cancer Syndrome?
An inherited hyperplastic polyposis syndrome (HPS) has also been increasingly recognised (Cohen et al. 1981; Sumner et al. 1981). It is now more commonly known as serrated polyposis syndrome. There is no sex predominance and the mean age at diagnosis is around 55 years. HPS has largely been considered a genetic disease, but the pattern of inheritance remains unknown: both autosomal recessive and autosomal dominant patterns have been suggested. Environmental factors could be partially responsible for the phenotypic differences and model the unknown pattern of inheritance. Smoking, being overweight and some drugs have been postulated as potential risk factors of HPs.
In HPS, multiple serrated polyps develop in the colorectum, and approximately 50% of cases present with at least one CRC (Ferrandez et al. 2004; Young and Jass 2006). Boparai et al (2011) have recently described an increased risk of CRC [relative risk (RR) = 5.4] and HPS (RR = 39) in first-degree relatives of probands diagnosed with HPS compared to the general population. Estimates for CRC risk associated with serrated polyposis may range from 7% to 50% and vary with phenotype.
Classification of the Syndrome
In the WHO criteria, Burt and Jass defined HPS as at least five HPs proximal to the sigmoid colon, two of which are > 1 cm diameter, or more than 30 HPs at any site in the large bowel (Burt 2000). Rashid et al, however, used a different classification system, in which HPS was defined as any person with more than 20 HPs, and separate classes were used for patients with large (>1 cm diameter) or multiple (5-10) HPs (Rashid et al. 2000). These differing classification systems reflect a syndrome which may be both genetically and phenotypically heterogeneous, but one which is becoming increasingly recognised.
The serrated pathway to colorectal cancer
Some evidence suggests that some but not all of these tumours develop along a ‘serrated pathway’ separate from the classical adenoma-carcinoma sequence (Sawyer et al. 2002; Spring, Zhao et al. 2006). This serrated pathway involves one group who accumulate BRAF V600E mutations and another separate pathway which involves KRAS mutations(Carvajal-Carmona et al. 2007). In addition the tumours often have methylation of the MLH1 promoter with subsequent microsatellite instability, and other genes such as P16, MGMT, or IGFBP7 may also be epigenetically inactivated. The CIMP phenotype identified by increased levels of methylation in the CpG island marker MINT31(Jass 2005).
HPS (sometimes known as the ‘serrated pathway syndrome’ or ‘serrated pathway syndrome’ (SPS) and sometimes ‘Jass Syndrome’) may, in fact, be a heterogeneous group of conditions leading to sporadic and inherited cases of colorectal neoplasia. There are two alternative clinical criteria for the diagnosis of HPS families (Burt 2000; Rashid, Houlihan et al. 2000). This syndrome is usually associated with somatic mutations in either BRAF or KRAS, but not both together (Carvajal-Carmona, Howarth et al. 2007), providing further evidence of molecular as well as phenotypic heterogeneity. BRAF mutations are associated with low-grade microsatellite instability due to methylation in CpG islands (CIMP)(Young, Jenkins et al. 2007). This may result in loss of expression of DNA repair genes MLH1 and MGMT (O(6)-methylguanine-DNA methyltransferase) in dysplastic mixed polyps from HPS patients, possibly as a result of promoter methylation (Oh et al. 2005).
Linkage analysis in a large family affected with hyperplastic polyposis syndrome demonstrated a maximum parametric LOD score of 2.71 on the short arm of chromosome 8 (8p.21; Monahan et al 2007). Another group have identified genetic linkage to chromosome 2q32.2-q33.3 with a LOD score of 2.07 (Roberts et al 2011 Fam Cancer).
As of this time however, there is no known causative germline mutation responsible for this condition, therefore genetic testing for predisposition is not possible. Because the natural history of HPS is poorly understood, colonoscopic screening guidelines have not been developed. Empirically we recommend 5 yearly colonoscopic screening from the age of the earliest known affected relative, or from 45 years of age.