Implications of polygenic risk for personalised colorectal cancer screening

BACKGROUND

We modelled the utility of applying a personalised screening approach for colorectal cancer (CRC) when compared with standard age-based screening. In this personalised screening approach, eligibility is determined by absolute risk which is calculated from age and polygenic risk score (PRS), where the PRS is relative risk attributable to common genetic variation. In contrast, eligibility in age-based screening is determined only by age.

DESIGN

We calculated absolute risks of CRC from UK population age structure, incidence and mortality rate data, and a PRS distribution which we derived for the 37 known CRC susceptibility variants. We compared the number of CRC cases potentially detectable by personalised and age-based screening. Using Genome-Wide Complex Trait Analysis to calculate the heritability attributable to common variation, we repeated the analysis assuming all common CRC risk variants were known.

RESULTS

Based on the known CRC variants, individuals with a PRS in the top 1% have a 2.9-fold increased CRC risk over the population median. Compared with age-based screening (aged 60: 10-year absolute risk 1.96% in men, 1.19% in women, as per the UK NHS National Bowel Screening Programme), personalised screening of individuals aged 55-69 at the same risk would lead to 16% fewer men and 17% fewer women being eligible for screening with 10% and 8%, respectively, fewer screen-detected cases. If all susceptibility variants were known, individuals with a PRS in the top 1% would have an estimated 7.7-fold increased risk. Personalised screening would then result in 26% fewer men and women being eligible for screening with 7% and 5% fewer screen-detected cases.

CONCLUSION

Personalised screening using PRS has the potential to optimise population screening for CRC and to define those likely to maximally benefit from chemoprevention. There are however significant technical and operational details to be addressed before any such programme is introduced.

Replication study of SNP associations for colorectal cancer in Hong Kong Chinese

BACKGROUND

Recent genome-wide association studies of colorectal cancer (CRC) have identified common single-nucleotide polymorphisms (SNPs) mapping to 10 independent loci that confer modest increased risk. These studies have been conducted in European populations and it is unclear whether these observations generalise to populations with different ethnicities and rates of CRC.

METHODS

An association study was performed on 892 CRC cases and 890 controls recruited from the Hong Kong Chinese population, genotyping 32 SNPs, which were either associated with CRC in previous studies or are in close proximity to previously reported risk SNPs.

RESULTS

Twelve of the SNPs showed evidence of an association. The strongest associations were provided by rs10795668 on 10p14, rs4779584 on 15q14 and rs12953717 on 18q21.2. There was significant linear association between CRC risk and the number of independent risk variants possessed by an individual (P=2.29 × 10(-5)).

CONCLUSION

These results indicate that some previously reported SNP associations also impact on CRC risk in the Chinese population. Possible reasons for failure of replication for some loci include inadequate study power, differences in allele frequency, linkage disequilibrium structure or effect size between populations. Our results suggest that many associations for CRC are likely to generalise across populations.

A large-scale meta-analysis to refine colorectal cancer risk estimates associated with MUTYH variants

Background:

Defective DNA repair has a causal role in hereditary colorectal cancer (CRC). Defects in the base excision repair gene MUTYH are responsible for MUTYH-associated polyposis and CRC predisposition as an autosomal recessive trait. Numerous reports have suggested MUTYH mono-allelic variants to be low penetrance risk alleles. We report a large collaborative meta-analysis to assess and refine CRC risk estimates associated with bi-allelic and mono-allelic MUTYH variants and investigate age and sex influence on risk.

Methods:

MUTYH genotype data were included from 20 565 cases and 15 524 controls. Three logistic regression models were tested: a crude model; adjusted for age and sex; adjusted for age, sex and study.

Results:

All three models produced very similar results. MUTYH bi-allelic carriers demonstrated a 28-fold increase in risk (95% confidence interval (CI): 6.95–115). Significant bi-allelic effects were also observed for G396D and Y179C/G396D compound heterozygotes and a marginal mono-allelic effect for variant Y179C (odds ratio (OR)=1.34; 95% CI: 1.00–1.80). A pooled meta-analysis of all published and unpublished datasets submitted showed bi-allelic effects for MUTYH, G396D and Y179C (OR=10.8, 95% CI: 5.02–23.2; OR=6.47, 95% CI: 2.33–18.0; OR=3.35, 95% CI: 1.14–9.89) and marginal mono-allelic effect for variants MUTYH (OR=1.16, 95% CI: 1.00–1.34) and Y179C alone (OR=1.34, 95% CI: 1.01–1.77).

Conclusions:

Overall, this large study refines estimates of disease risk associated with mono-allelic and bi-allelic MUTYH carriers.

Familial adenomatous polyposis and the small bowel: a loco-regional review and current management strategies

Small-bowel tumours are an important cause of morbidity and death in patients with familial adenomatous polyposis. Intensive endoscopic surveillance is now standard in the long-term management of this condition. Thus, lesions occurring throughout the small bowel are increasingly noted by oesophagogastroduodenoscopy and flexible pouchoscopy. Some occur commonly de novo (in stomach, duodenum and ampulla), while others may occur following surgery (polyps of the ileostomy, ileoanal pouch, or small bowel above an anastomosis). These differ widely in incidence, natural history and management. This review provides a regional overview of these lesions, in terms of current research findings and management protocols.

Mutation analysis of three genes encoding novel LKB1-interacting proteins, BRG1, STRADalpha, and MO25alpha, in Peutz-Jeghers syndrome

Mutations in LKB1 lead to Peutz–Jeghers syndrome (PJS). However, only a subset of PJS patients harbours LKB1 mutations. We performed a mutation analysis of three genes encoding novel LKB1-interacting proteins, BRG1, STRADα, and MO25α, in 28 LKB1-negative PJS patients. No disease-causing mutations were detected in the studied genes in PJS patients from different European populations.

Apparent Mendelian inheritance of breast and colorectal cancer: chance, genetic heterogeneity or a new gene?

It is not uncommon for cancer geneticists to be referred families with apparently Mendelian co-inheritance of breast and bowel cancer. Such families present a particular problem as regards the intensity of their screening for these diseases and the utility of genetic testing. Many 'breast-colon' cancer families probably result from chance clustering of two common cancers. Other 'breast-colon' cancer families may result from known cancer syndromes, such as hereditary breast-ovarian cancer or hereditary non-polyposis colon cancer, either by conferring a high risk of one cancer type and a slightly increased risk of the other, or through a predisposition to one of the two cancers and chance occurrence of the other. Anecdotally, however, many geneticists wonder about the existence of a distinct 'breast-colon cancer syndrome', since some families present good a priori evidence of genetic disease and yet cannot readily be accounted for by known genes or chance. The identification of unknown 'breast-colon cancer' genes is likely to be difficult, relying primarily on candidate gene analysis, including loci separately implicated in breast or colorectal cancer, or in other multiple cancer syndromes. Studies such as those on APC I1307K and CHEK2 1100delC may suggest the way forward for the identification of 'breast-colon cancer' genes.

Variability in the severity of colonic disease in familial adenomatous polyposis results from differences in tumour initiation rather than progression and depends relatively little on patient age

INTRODUCTION

As large scale genetic analysis becomes increasingly efficient, attention is turning to problems arising from inaccurate measurement of the phenotype. We have investigated the underlying basis of variation in disease severity in the large intestine of familial adenomatous polyposis (FAP) patients. The development of objective and reproducible measures may have future use in genetic studies, such as analysis of modifier genes.

METHODS

We examined the ratio of adenomas to crypts from microscopic slides taken from all parts of the colon of 44 resected FAP specimens. These findings were compared with a carefully reported macroscopic polyp count. Age dependency of adenoma counts (in the period around colectomy) was also analysed.

RESULTS

The adenoma:crypt ratio strongly correlated with reported macroscopic polyp count (r=0.82, p<0.001) with no significant residual variation. Polyp density measured using the adenoma: crypt ratio did not vary significantly within an individual colon. Apparent visible variation in polyp density within any colon was not found at the microscopic level. There was no detectable age related increase in macroscopic adenoma count between siblings over the age range at which colectomies were performed.

DISCUSSION

The severity of colonic polyposis in FAP can be determined accurately by counting the adenoma:crypt ratio in sections derived from stored tissue blocks. Variation between patients-dependent on APC genotype and, probably, modifier genes-is manifest at both the microscopic and macroscopic levels. Thus variation in disease severity is more likely to result from different rates of tumour initiation than from differences in progression of microadenomas to macroscopic tumours. The absence of a detectable relationship between adenoma number and age (over the range studied) suggests that most tumours may be initiated relatively early in the patient's life, perhaps at a time of particular susceptibility.

Comprehensive analysis of SMAD4 mutations and protein expression in juvenile polyposis: evidence for a distinct genetic pathway and polyp morphology in SMAD4 mutation carriers

Juvenile polyposis syndrome (JPS; OMIM 174900) is a rare disorder which is characterized by the presence of hamartomatous polyps throughout the gastrointestinal tract and an increased risk of gastrointestinal malignancy. Mutations of the SMAD4 gene on chromosome 18q21.1 have been shown to cause a subset of JPS cases, with estimates ranging from 20% to >50%. Characterization of the genes that cause the remainder of JPS cases relies on the certainty that SMAD4 is not the causative gene. We have undertaken a comprehensive analysis of germline SMAD4 mutations in a cohort of JPS patients to define the spectrum of mutations that cause JPS. We have analyzed a series of polyps from these patients for SMAD4 protein expression. We have also performed a blinded assessment of polyp material to look for morphological differences between polyps from patients with and without a germline SMAD4 mutation. The results indicate that almost all germline SMAD4 mutations are readily detectable by screening genomic DNA using polymerase chain reaction-based methods; SMAD4 can be excluded as the causative gene in the majority of our JPS cohort. Loss of SMAD4 expression occurs in most polyps from SMAD4 mutation carriers, even those with missense germline mutations. SMAD4 loss in polyps is, however, not a feature of cases that are not caused by SMAD4 mutations, indicating that these polyps develop along a SMAD4-independent pathway. The morphology of polyps from SMAD4 mutation carriers is subtly different from other JPS polyps, notably including a more prominent epithelial component in the former.

SMAD4 mutations in colorectal cancer probably occur before chromosomal instability, but after divergence of the microsatellite instability pathway

Loss of chromosome 18q21 is well documented in colorectal cancer, and it has been suggested that this loss targets the DCC, DPC4/SMAD4, and SMAD2 genes. Recently, the importance of SMAD4, a downstream regulator in the TGF-beta signaling pathway, in colorectal cancer has been highlighted, although the frequency of SMAD4 mutations appears much lower than that of 18q21 loss. We set out to investigate allele loss, mutations, protein expression, and cytogenetics of chromosome 18 copy number in a collection of 44 colorectal cancer cell lines of known status with respect to microsatellite instability (MSI). Fourteen of thirty-two MSI(-) lines showed loss of SMAD4 protein expression; usually, one allele was lost and the other was mutated in one of a number of ways, including deletions of various sizes, splice site changes, and missense and nonsense point mutations (although no frameshifts). Of the 18 MSI(-) cancers with retained SMAD4 expression, four harbored missense mutations in the 3' part of the gene and showed allele loss. The remaining 14 MSI(-) lines had no detectable SMAD4 mutation, but all showed allele loss at SMAD4 and/or DCC. SMAD4 mutations can therefore account for about 50-60% of the 18q21 allele loss in colorectal cancer. No MSI(+) cancer showed loss of SMAD4 protein or SMAD4 mutation, and very few had allelic loss at SMAD4 or DCC, although many of these MSI(+) lines did carry TGFBIIR changes. Although SMAD4 mutations have been associated with late-stage or metastatic disease, our combined molecular and cytogenetic data best fit a model in which SMAD4 mutations occur before colorectal cancers become aneuploid/polyploid, but after the MSI(+) and MSI(-) pathways diverge. Thus, MSI(+) cancers may diverge first, followed by CIN(+) (chromosomal instability) cancers, leaving other cancers to follow a CIN(-)MSI(-) pathway.

Molecular pathology of solid tumours: some practical suggestions for translating research into clinical practice

"Molecular pathology" can be broadly defined as the use of genetic data, in addition to the standard pathological parameters, to optimise diagnosis and to indicate treatment and prognosis. The benefit to be gained from the exploitation of molecular techniques to provide additional information to aid patient management is potentially vast. Currently, molecular pathology is rarely used in clinical practice, although it is anticipated that it will eventually become a part of routine practice. However, incorporating molecular techniques into routine practice will not be straightforward because there are several issues to be resolved. Following on from a symposium held at the Royal College of Pathologists to discuss some of these issues, the establishment of a committee of molecular pathology is proposed to plan and coordinate the introduction of molecular pathology into routine clinical practice.

Polymorphisms and colorectal tumor risk

BACKGROUND & AIMS

Increasingly, studies of the relationship between common genetic variants and colorectal tumor risk are being proposed. To assess the evidence that any of these confers a risk, a systematic review and meta-analysis of published studies was undertaken.

METHODS

Fifty studies of the effect of common alleles of 13 genes on risk were identified. To clarify the impact of individual polymorphisms on risk, pooled analyses were performed.

RESULTS

Of the 50 studies identified, significant associations were seen in 16, but only 3 were reported in more than one study. Pooling studies, significant associations were only seen for 3 of the polymorphisms: adenomatosis polyposis coli (APC)-I1307K (odds ratio [OR] = 1.58, 95% confidence interval [CI]: 1.21-2.07); Harvey ras-1 variable number tandem repeat polymorphism (HRAS1-VNTR; OR = 2.50, 95% CI: 1.54-4.05); and methylenetetrahydrofolate reductase (MTHFR)(Val/Val) (OR = 0.76, 95% CI: 0.62-0.92). For tumor protein 53 (TP53), N-acetyl transferase 1 (NAT1), NAT2, glutathione-S transferase Mu (GSTM1), glutathione-S transferase Theta (GSTT1), and glutathione-S transferase Pi (GSTP1) polymorphisms, the best estimates are sufficient to exclude a 1.7-fold increase in risk of colorectal cancer.

CONCLUSIONS

APC-I1307K, HRAS1-VNTR, and MTHFR variants represent the strongest candidates for low penetrance susceptibility alleles identified to date. Although their genotypic risks are modest, their high frequency in the population implies that they may well have considerable impact on colorectal cancer incidence. Determining precise risk estimates associated with other variants and gene-gene and gene-environment interactions will be contingent on further studies with sample sizes larger than typically used to date.

CDX2 mutations do not account for juvenile polyposis or Peutz-Jeghers syndrome and occur infrequently in sporadic colorectal cancers

Peutz-Jeghers syndrome (PJS) and juvenile polyposis (JPS) are both characterized by the presence of hamartomatous polyps and increased risk of malignancy in the gastrointestinal tract. Mutations of the LKB1 and SMAD4 genes have been shown recently to cause a number of PJS and JPS cases respectively, but there remains considerable uncharacterized genetic heterogeneity in these syndromes, particularly JPS. The mouse homologue of CDX2 has been shown to give rise to a phenotype which includes hamartomatous-like polyps in the colon and is therefore a good candidate for JPS and PJS cases which are not accounted for by the SMAD4 and LKB1 genes. By analogy with SMAD4, CDX2 is also a candidate for somatic mutation in sporadic colorectal cancer. We have screened 37 JPS families/cases without known SMAD4 mutations, 10 Peutz-Jeghers cases without known LKB1 mutations and 49 sporadic colorectal cancers for mutations in CDX2. Although polymorphic variants and rare variants of unlikely significance were detected, no pathogenic CDX2 mutations were found in any case of JPS or PJS, or in any of the sporadic cancers.

Two hits revisited again

INTRODUCTION AND METHODS

Since the concept of the "two hit hypothesis" was introduced over 20 years ago, a wealth of genetic data has accumulated on the mutations found at tumour suppressor loci. Perhaps surprisingly, these data conceal large gaps in our knowledge which genetic and functional studies are beginning to uncover. The "two hit hypothesis" must be updated to take account of this new information.

RESULTS AND DISCUSSION

Here, we discuss both the results of recent studies and some of the questions that they highlight. In particular, how valid are conclusions from inherited Mendelian syndromes when applied to sporadic cancers? Why is allelic loss so common and how does it occur? Are the "two hits" random or interdependent? Is abolition of protein function always optimal for tumorigenesis? Can "third hits" occur and, if so, why? How can mismatch repair deficiency and the methylator phenotype be incorporated into the "two hit" hypothesis? We suggest that the "two hit hypothesis" is not fixed but is evolving as our knowledge expands.

Mutations in normal breast tissue and breast tumours

The accumulation of mutations is a feature of all normal cells. The probability of any individual gene in any cell acquiring a mutation is, however, low. Cancer is therefore a rare disease in comparison with the number of susceptible cells. Mutations in normal tissue are stochastic, vary widely among cells and are therefore difficult to detect using standard methods because each change is so rare. If, however, a tissue such as the breast undergoes considerable clonal expansion, particularly if relatively late in life, normal tissue may have accumulated many thousands of detectable mutations. Since breast cancers are clonal and have almost certainly undergone many more cell divisions than normal cells, each tumour may have many millions of mutations, most of which are entirely innocent and some of which have accumulated in the cell of origin prior to tumorigenesis. Despite some claims to the contrary, even at normal mutation rates, clonal expansion within a tumour is quite sufficient to account for the mutations of five or six genes that are generally supposed necessary for carcinogenesis to occur. Hypermutability does, however, contribute to the pathogenesis of many cancers and, although evidence is indirect in breast cancer, may take forms such as karyotypic instability via centrosome amplification.

The adenomatous polyposis coli (APC) tumour suppressor--genetics, function and disease

Mutations in the adenomatous polyposis coli (APC) gene are the basis of familial adenomatous polyposis and the majority of sporadic colorectal cancer. APC is expressed in a wide variety of tissues, interacts with the cytoskeleton, is involved in regulating levels of beta-catenin and, most recently, has been shown to bind DNA, suggesting that it may possess a nuclear role. The mutation spectrum implicated in tumorigenesis and its correlation with disease phenotype is well characterized and has contributed to our understanding of important functional domains in APC. Despite these advances, APC continues to provide a fertile subject of research for both colorectal tumorigenesis and cancer in general.

Defects in mismatch repair occur after APC mutations in the pathogenesis of sporadic colorectal tumours

The roles of the intrinsic mutation rate and genomic instability in tumorigenesis are currently controversial. In most colorectal tumours, it is generally supposed that the first mutations occur at the adenomatous polyposis coli (APC) locus; APC mutations are thought to provide cells with a selective advantage but have no known effect on the mutation rate. It has also been suggested that genomic instability is the initiating event in colorectal tumorigenesis and, if this is true, mutations of DNA mismatch repair (MMR) genes (or at similar loci) are the most likely candidates. If defective MMR precedes APC mutations, the APC mutations of colon tumours with defective MMR and hence replication errors (RER+) should differ from those of RER- tumours, in at least three specific ways: (1) a higher frequency of allele loss at APC in RER- tumours; (2) more frameshift than nonsense mutations in RER+ tumours; and (3) APC mutations in simple repeat sequences [(N)n, (N1N2)n, or (N1N2N3)n] in RER+ tumours. We found no evidence that sporadic RER+ and RER- colon cancers (including cell lines) differ in any of these three ways. Although it remains possible that MMR is abnormal in tumours from HNPCC families before APC mutations occur, it is likely that in sporadic colon tumours, APC mutations, rather than genomic instability, are the initiating events in tumorigenesis.

Analysis of genetic and phenotypic heterogeneity in juvenile polyposis

BACKGROUND

Juvenile polyposis syndrome (JPS) is characterised by gastrointestinal (GI) hamartomatous polyposis and an increased risk of GI malignancy. Juvenile polyps also occur in the Cowden (CS), Bannayan-Ruvalcaba-Riley (BRRS) and Gorlin (GS) syndromes. Diagnosing JPS can be problematic because it relies on exclusion of CS, BRRS, and GS. Germline mutations in the PTCH, PTEN and DPC4 (SMAD4) genes can cause GS, CS/BRRS, and JPS, respectively.

AIMS

To examine the contribution of mutations in PTCH, PTEN, and DPC4 (SMAD4) to JPS.

METHODS

Forty seven individuals from 15 families and nine apparently sporadic cases with JPS were screened for germline mutations in DPC4, PTEN, and PTCH.

RESULTS

No patient had a mutation in PTEN or PTCH. Five different germline mutations were detected in DPC4; three of these were deletions, one a single base substitution creating a stop codon, and one a missense change. None of these patients had distinguishing clinical features.

CONCLUSIONS

Mutations in PTEN and PTCH are unlikely to cause juvenile polyposis in the absence of clinical features indicative of CS, BRRS, or GS. A proportion of JPS patients harbour DPC4 mutations (21% in this study) but there remains uncharacterized genetic heterogeneity in JPS.

Identification of a novel mRNA species of the LKB1/STK11 Peutz-Jeghers serine/threonine kinase

Germline mutations in the LKB1/STK11 serine/threonine kinase cause Peutz-Jeghers syndrome and this gene is also mutated at a moderate frequency in a wide variety of sporadic tumours. The translated region of LKB1/STK11 (1302bp) codes for a serine/threonine kinase of otherwise unknown function. We report a novel LKB1/STK11 mRNA species which is found at variable levels in all tissues examined. The novel mRNA, which we believe may be an unusual splice variant, consists of a 444bp in-frame deletion of exons 5-7 and part of exon 8. This deletion removes a large part of the kinase domain and comparison with other LKB1/STK11 mutations shows that kinase function is undoubtedly abolished. The role of the novel mRNA species remains unclear, but it retains a putative cAMP-dependent kinase phosphorylation site and may play some regulatory role.

APC mutations in sporadic colorectal tumors: A mutational "hotspot" and interdependence of the "two hits"

Although APC mutations occur at a high frequency in colorectal cancers, few studies have performed a comprehensive analysis by screening the whole gene for mutations and assessing allelic loss. APC seems to act as a tumor-suppressor gene in a "nonclassical" fashion: data from familial adenomatous polyposis (FAP) show that the site of the germ-line mutation determines the type of "second hit" in FAP tumors, and simple protein inactivation is selected weakly, if at all. In this study, we screened the entire coding region of APC for mutations and assessed allelic loss in a set of 41 colorectal cancer cell lines. Of 41 cancers, 32 (83%) showed evidence of APC mutation and/or allelic loss. We identified several APC mutations and found a "hotspot" for somatic mutation in sporadic colorectal tumors at codon 1,554. Our results suggest that APC mutations occur in the great majority of colorectal cancers, the exceptions almost all being RER+ tumors, which may substitute for altered APC function by mutations in beta-catenin and/or at other loci. When combined with previously published data, our results show that there is interdependence of the "two hits" at APC in sporadic colorectal tumors as well as in FAP. APC mutations in the "mutation cluster region," especially those close to codon 1,300, are associated with allelic loss, whereas tumors with mutations outside this region tend to harbor truncating mutations. The causes of this phenomenon are probably selection for retained N-terminal and lost C-terminal APC functions, effects on beta-catenin levels, and APC protein stability.

The androgen receptor exon 1 trinucleotide repeat does not act as a modifier of the age of presentation in breast cancer

The CAG repeat in exon 1 of the androgen receptor (AR) genes has been postulated as both a susceptibility allele and phenotypic modifier in BRCA1-associated breast cancers. We have analysed this repeat in a set of 178 breast cancer cases who have been selected only for age of presentation at 65 years or less. No effect of repeat length on age of presentation was found and there was no association between repeat length and family history. In combination with the data from other workers, our findings suggest that the androgen receptor repeat does not act as a modifier gene or susceptibility locus outside the context of the hereditary breast/ovarian cancer syndrome.

Detecting low penetrance genes in cancer: the way ahead

The search for the genes responsible for many complex genetic diseases is well under way and has already been successful in some cases. The study of cancer as a complex genetic disease has lagged behind other conditions, largely because of particular problems that are associated with malignant disease. Cancer also, however, presents specific opportunities for gene identification, which are not found in many other diseases. While the methods of genetic mapping and gene cloning used for other complex diseases will be applied to cancer, these must almost certainly be complemented by other methods, such as the study of somatic mutations, cancer associated phenotypes, and modifier genes for Mendelian cancers. Here, we review the strategies available for identifying cancer predisposition genes of low and moderate penetrance.

Optimising methods for determining RER status in colorectal cancers

Approximately 13% of colorectal cancers display microsatellite instability (MSI), a form of replication error repair. Colorectal cancers developing in individuals with constitutional defects in the mismatch repair (MMR) genes hMLH1, hMSH2, hPMS1 and hPMS2 consistently show evidence of this phenomenon. Since MSI is indicative of MMR deficiency, testing colorectal cancers for MSI provides a method of refining the identification of carriers of germline MMR mutations. To assess which microsatellites represent the best reporters of replication error (RER) status we have examined 116 early onset colorectal cancers for MSI. MSI was assessed using eight dinucleotide- and two mononucleotide-repeat fluorescently labelled polymerase chain reaction (PCR) markers. The two mononucleotide repeat markers (BAT25 and BAT26) were highly sensitive and typing of either represents an efficient strategy for defining RER status of colorectal cancers and obviates the requirement of typing numerous microsatellite markers.

Genetic pathways in colorectal and other cancers

Cells from cancers show aberrant behaviour such as unrestrained growth, invasion into adjacent tissue and metastasis. All these features of cancer cell behaviour can be explained in terms of genetic changes and the functional impact of these changes. In this review, colorectal cancer (CRC) is examined as a classical example of multistep carcinogenesis. First there is an overview which shows that cancers develop by a process of somatic evolution. This gives rise to preferred genetic pathways of tumorigenesis. The factors which may influence the development and ultimate choice of genetic pathways are then examined. Next, CRC is studied as a specific disease and the putative genetic pathways are described. The mutations that comprise these pathways and the possible functional sequelae of these are explored. The review concludes with a look at those avenues which may further elucidate the natural history of CRC and lead to improved therapy.

Tumor burden and clonality in multiple intestinal neoplasia mouse/normal mouse aggregation chimeras

Aggregation chimeras were formed between C57BL/6 mice heterozygous for the Apc(min) (Min) mutation and wild-type SWR mice, that differ in their Pla2g2a status, a modifier of Apc(min), and also in their resistance to intestinal polyp formation. Variation in the dolichos biflorus agglutinin-staining patterns of the intestines of these mouse strains was used to determine the chimeric composition of the intestine in individual mice and to examine the clonal composition of adenomas. Macroscopic adenoma numbers in chimeric mice were compared with the expected adenoma numbers based on the percentage of C57BL/6J-Apc(min/+) epithelium in individual mice. These results unexpectedly show that there was no apparent inhibitory effect of the SWR-derived (Pla2g2a wild-type) tissue on adenoma formation in the C57BL/6J-Apc(min/+) epithelium. This suggests that the main genetic modifiers of the Min phenotype act at a cellular or crypt-restricted level with no discernable systemic effect. All adenomas were seen to contain C57BL/6J-Apc(min/+)-derived epithelium, confirming that the germ-line mutation of the mApc gene is necessary to initiate tumorigenesis in this model system, and that the mApc gene acts in a cell autonomous fashion.

Mechanisms of inactivation of mismatch repair genes in human colorectal cancer cell lines: the predominant role of hMLH1

Fifteen to twenty-five percent of sporadic colorectal carcinomas are replication error (RER) positive. Because the frequency of mutations in the mismatch repair genes (hMLH1 and hMSH2) is low in these tumors, we have investigated the role of mutational inactivation, methylation of the promoter region, and loss of heterozygosity (LOH) as a possible explanation for the mutator phenotype of RER+ colorectal cancer cell lines. Genomic DNA was extracted from a panel of 49 human colorectal cancer cell lines. The RER status was determined by amplification of BAT-26. All exons of hMLH1 and hMSH2 were amplified with the PCR and screened by using single-strand conformational polymorphism and direct sequencing. The methylation status was ascertained by methylation-specific PCR after bisulfite modification of DNA. Western blotting for hMLH1 was performed on methylated cell lines before and after the addition of the demethylating agent 5-azacytidine. LOH was sought by GENESCAN analysis of amplified CA repeat markers and indirectly by determining the number of homozygotes in the cell lines and human random controls. Twelve cell lines from ten tumors (24%) were RER+. Hypermethylation of the hMLH1 promoter occurred in five of ten (50%) RER+ tumors, whereas three of thirty-two (6%) RER tumors showed partial methylation. None of the fully methylated cell lines expressed hMLH1, although all reexpressed hMLH1 after treatment with 5-azacytidine. There was no LOH in the RER+ tumors in either hMLH1 or hMSH2. Our results suggest that mutations of hMLH1 together with hypermethylation of the promoter region, but not LOH, are the cause of the mutator phenotype in the majority (70%) of RER+ tumors.

Mutations in Bcl10 are very rare in colorectal cancer

Bcl10 is a recently identified gene reported to be involved commonly in human malignancy (Willis et al (1999) Cell 96: 1-20). To investigate whether it is frequently mutated in colorectal cancer we have analysed a series of 132 colorectal cancers and eight colorectal cancer cell lines for mutations in Bcl10. One feature of the Bcl10 gene is that it harbours two polyadenine tracts. These repeating elements in genes can be prone to a high rate of mutation if there is defective mismatch repair. To examine the possibility that Bcl10 may be preferentially mutated in mismatch repair-deficient cancers, 49 of the tumours and cell lines were known to be replication error (RER)-positive and, of these, ten were from individuals harbouring germline mutations in hMLH1 or hMSH2. No pathogenic mutations were detected in the tumours and only one mutation was found in the colorectal cancer cell lines. These results indicate that Bcl10 is unlikely to be involved in the pathways of colorectal carcinogenesis.

Loss of heterozygosity of methylenetetrahydrofolate reductase in colon carcinomas

Folate derivatives are essential for DNA synthesis and methylation. A large proportion of the Caucasian population is heterozygous for a common substitution, 677C-->T (alanine-->valine), in methylenetetrahydrofolate reductase (MTHFR), an enzyme of folate interconversion. Homozygous mutant individuals, approximately 10-15% of North Americans, have been reported to have a reduced risk of colorectal cancer. We examined lymphocyte and tumor tissue DNA from colorectal carcinoma patients from two different populations to assess loss of heterozygosity (LOH) of MTHFR. We observed LOH in approximately 16% of colorectal tumors; in 8 of the 11 tumors with LOH, the mutant valine allele was lost. Additional studies are required to determine if preferential loss of the mutant allele is a common finding that could contribute to colorectal tumorigenesis.

Germline mutations of the LKB1 (STK11) gene in Peutz-Jeghers patients

Germline mutations of the LKB1 (STK11) serine/threonine kinase gene (chromosome 19p13.3) cause Peutz-Jeghers syndrome, which is characterised by hamartomas of the gastrointestinal tract and typical pigmentation. Peutz-Jeghers syndrome carries an overall risk of cancer that may be up to 20 times that of the general population. Here, we report the results of a screen for germline LKB1 mutations by DNA sequencing in 12 Peutz-Jeghers patients (three sporadic and nine familial cases). Mutations were found in seven (58%) cases, in exons 1, 2, 4, 6, and 9. Five of these mutations, two of which are identical, are predicted to lead to a truncated protein (three frameshifts, two nonsense changes). A further mutation is an in frame deletion of 6 bp, resulting in a deletion of lysine and asparagine; the second of these amino acids is conserved between species. The seventh mutation is a missense change in exon 2, converting lysine to arginine, affecting non-conserved amino acids and of uncertain functional significance. Despite the fact that Peutz-Jeghers syndrome is usually an early onset disease with characteristic clinical features, predictive and diagnostic testing for LKB1 mutations will be useful for selected patients in both familial and non-familial contexts.

Allele loss and mutation screen at the Peutz-Jeghers (LKB1) locus (19p13.3) in sporadic ovarian tumours

Germline mutations in the LKB1 (STK11) gene (chromosome sub-band 19p13.3) cause characteristic hamartomas and pigmentation to develop in patients with Peutz-Jeghers syndrome. Peutz-Jeghers syndrome carries an overall risk of cancer that may be up to 20 times that of the general population and Peutz-Jeghers patients are at increased risk of benign and malignant ovarian tumours, particularly granulosa cell tumours. Loss of heterozygosity (allele loss, LOH) has been reported in about 50% of ovarian cancers on 19p13.3. LKB1 is therefore a candidate tumour suppressor gene for sporadic ovarian tumours. We found allele loss at the marker D19S886 (19p13.3) in 12 of 49 (24%) sporadic ovarian adenocarcinomas. Using SSCP analysis, we screened ten ovarian cancers with LOH, 35 other ovarian cancers and 12 granulosa cell tumours of the ovary for somatic mutations in LKB1. No variants were detected in any of the adenocarcinomas. Two mutations were detected in one of the granulosa cell tumours: a mis-sense mutation affecting the putative 'start' codon (ATG --> ACG, M1T); and a silent change in exon 7 (CTT --> CTA, leucine). Like BRCA1 and BRCA2, therefore, it appears that LKB1 mutations can cause ovarian tumours when present in the germline, but occur rarely in the soma. The allele loss on 19p13.3 in ovarian cancers almost certainly targets a different gene from LKB1.

Genetic pathways in colorectal and other cancers

Cells from cancers show aberrant behaviour such as unrestrained growth, invasion into adjacent tissue and metastasis. All these features of cancer cell behaviour can be explained in terms of genetic changes and the functional impact of these changes. In this review, colorectal cancer (CRC) is examined as a classical example of multistep carcinogenesis. First there is an overview which shows that cancers develop by a process of somatic evolution. This gives rise to preferred genetic pathways of tumorigenesis. The factors which may influence the development and ultimate choice of genetic pathways are then examined. Next, CRC is studied as a specific disease and the putative genetic pathways are described. The mutations that comprise these pathways and the possible functional sequelae of these are explored. The review concludes with a look at those avenues which may further elucidate the natural history of CRC and lead to improved therapy.

Mutations and impaired function of LKB1 in familial and non-familial Peutz-Jeghers syndrome and a sporadic testicular cancer

Germline mutations in LKB1 have been reported to underlie familial Peutz-Jeghers syndrome (PJS) with intestinal hamartomatous polyps and an elevated risk of various neoplasms. To investigate the prevalence of LKB1 germline mutations in PJS more generally, we studied samples from 33 unrelated PJS patients including eight non-familial sporadic patients, 20 familial patients and five patients with unknown family history. Nineteen germline mutations were identified, 12 (60%) in familial and four (50%) in sporadic cases. LKB1 mutations were not detected in 14 (42%) patients, indicating that the existence of additional minor PJS loci cannot be excluded. LKB1 is predicted to encode a serine/threonine kinase. To demonstrate the putative Lkb1 kinase function and to study the consequences of LKB1 mutations in PJS and sporadic tumors, we have analyzed the kinase activity of wild-type and mutant Lkb1 proteins. Interestingly, while most of the small deletions or missense mutations resulted in loss-of-function alleles, one missense mutation (G163D) previously identified in a sporadic testicular tumor demonstrated severely impaired but detectable kinase activity.

Germline PTEN mutations in Cowden syndrome-like families

Cowden syndrome (CS) or multiple hamartoma syndrome (MIM 158350) is an autosomal dominant disorder with an increased risk for breast and thyroid carcinoma. The diagnosis of CS, as operationally defined by the International Cowden Consortium, is made when a patient, or family, has a combination of pathognomonic major and/or minor criteria. The CS gene has recently been identified as PTEN, which maps at 10q23.3 and encodes a dual specificity phosphatase. PTEN appears to function as a tumour suppressor in CS, with between 13-80% of CS families harbouring germline nonsense, missense, and frameshift mutations predicted to disrupt normal PTEN function. To date, only a small number of tumour suppressor genes, including BRCA1, BRCA2, and p53, have been associated with familial breast or breast/ovarian cancer families. Given the involvement of PTEN in CS, we postulated that PTEN was a likely candidate to play a role in families with a "CS-like" phenotype, but not classical CS. To answer these questions, we gathered a series of patients from families who had features reminiscent of CS but did not meet the Consortium Criteria. Using a combination of denaturing gradient gel electrophoresis (DGGE), temporal temperature gel electrophoresis (TTGE), and sequence analysis, we screened 64 unrelated CS-like subjects for germline mutations in PTEN. A single male with follicular thyroid carcinoma from one of these 64 (2%) CS-like families harboured a germline point mutation, c.209T-->C. This mutation occurred at the last nucleotide of exon 3 and within a region homologous to the cytoskeletal proteins tensin and auxilin. We conclude that germline PTEN mutations play a relatively minor role in CS-like families. In addition, our data would suggest that, for the most part, the strict International Cowden Consortium operational diagnostic criteria for CS are quite robust and should remain in place.

The APC variants I1307K and E1317Q are associated with colorectal tumors, but not always with a family history

Classical familial adenomatous polyposis (FAP) is a high-penetrance autosomal dominant disease that predisposes to hundreds or thousands of colorectal adenomas and carcinoma and that results from truncating mutations in the APC gene. A variant of FAP is attenuated adenomatous polyposis coli, which results from germ-line mutations in the 5' and 3' regions of the APC gene. Attenuated adenomatous polyposis coli patients have "multiple" colorectal adenomas (typically fewer than 100) without the florid phenotype of classical FAP. Another group of patients with multiple adenomas has no mutations in the APC gene, and their phenotype probably results from variation at a locus, or loci, elsewhere in the genome. Recently, however, a missense variant of APC (I1307K) was described that confers an increased risk of colorectal tumors, including multiple adenomas, in Ashkenazim. We have studied a set of 164 patients with multiple colorectal adenomas and/or carcinoma and analyzed codons 1263-1377 (exon 15G) of the APC gene for germ-line variants. Three patients with the I1307K allele were detected, each of Ashkenazi descent. Four patients had a germ-line E1317Q missense variant of APC that was not present in controls; one of these individuals had an unusually large number of metaplastic polyps of the colorectum. There is increasing evidence that there exist germ-line variants of the APC gene that predispose to the development of multiple colorectal adenomas and carcinoma, but without the florid phenotype of classical FAP, and possibly with importance for colorectal cancer risk in the general population.

Loss of Bcl-2 expression correlates with tumour recurrence in colorectal cancer

AIMS

To investigate the association between immunohistochemical expression of Bcl-2 and p53 in colorectal cancer and tumour recurrence following surgery.

METHODS

Sixty six cases of Dukes' B colorectal carcinoma were studied. All tumours were moderately differentiated and were shown to be histologically clear of the resection margins. Immunohistochemistry was performed on formalin fixed paraffin wax embedded tissue using monoclonal antibodies for p53 and Bcl-2. The Bcl-2 staining was assessed separately for relative intensity of staining and percentage of positive tumour cells and given a final score which combined the two factors. The p53 staining was assessed on number of positive tumour cells only. The patterns of immunostaining of those cases in which there had been tumour recurrence were compared with those cases in which there was no tumour recurrence (controls).

RESULTS

A statistically significant inverse association was found between Bcl-2 score and tumour recurrence (median Bcl-2 score of 6 (interquartile range (IQR) 2-9) in patients with recurrent disease; median Bcl-2 score of 8 (IQR 6-10) in those without recurrence; p = 0.03). When examined separately, both the intensity of expression and percentage of positive tumour cells were significantly associated with tumour recurrence (p = 0.04 in each case). There was no association between p53 staining and tumour recurrence.

CONCLUSION

Results suggest that, when controlled for differentiation, Bcl-2 expression is a prognostic marker and may be useful as an adjunctive test in clinical decision making.

Carney complex, Peutz-Jeghers syndrome, Cowden disease, and Bannayan-Zonana syndrome share cutaneous and endocrine manifestations, but not genetic loci

Carney complex (CC), Peutz-Jeghers syndrome (PJS), Cowden disease (CD), and Bannayan-Zonana syndrome (BZS) share clinical features, such as mucocutaneous lentigines and multiple tumors (thyroid, breast, ovarian, and testicular neoplasms), and autosomal dominant inheritance. A genetic locus has been identified for CC on chromosome 2 (2p16), and the genes for PJS, CD, and BZS were recently identified; genetic heterogeneity appears likely in both CC and PJS. The genes for PJS and CD/BZS, STK11/LKB1 and PTEN, respectively, may act as tumor suppressors, because loss of heterozygosity (LOH) of the PJS and CD/BZS loci has been demonstrated in tumors excised from patients with these disorders. We studied 2 families with CC in whom the disease could not be shown to segregate with polymorphic markers from the 2p16 locus. Their members presented with lesions frequently seen in PJS and the other lentiginosis syndromes. We also tested 16 tumors and cell lines established from patients with CC for LOH involving the PJS and CD/BZS loci. DNA was extracted from peripheral blood, tumor cell lines, and tissues and subjected to PCR amplification with primers from microsatellite sequences flanking the STK11/LKB1 and PTEN genes on 19p13 and 10q23, respectively, and a putative PJS locus on 19q13. All loci were excluded as candidates in both families with LOD scores less than 2 and/or by haplotype analysis. LOH for these loci was not present in any of the tumors that were histologically identical to those seen in PJS. The overall rate of LOH for the PJS and CD/BZS loci in tumors from patients with CC was less than 10%. We conclude that despite substantial clinical overlap among CC, PJS, CD, and BZS, LOH for the STK11 and PTEN loci is an infrequent event in CC-related tumors. Linkage analysis excluded the PJS and CD/BZS loci on chromosomes 19 (19p13 and 19q13) and 10 (10q23) from harboring the gene defect(s) responsible for the phenotype in these 2 families.

Mutations in the SMAD4/DPC4 gene in juvenile polyposis

Familial juvenile polyposis is an autosomal dominant disease characterized by a predisposition to hamartomatous polyps and gastrointestinal cancer. Here it is shown that a subset of juvenile polyposis families carry germ line mutations in the gene SMAD4 (also known as DPC4), located on chromosome 18q21.1, that encodes a critical cytoplasmic mediator in the transforming growth factor-beta signaling pathway. The mutant SMAD4 proteins are predicted to be truncated at the carboxyl-terminus and lack sequences required for normal function. These results confirm an important role for SMAD4 in the development of gastrointestinal tumors.

Modifier genes in humans: strategies for identification

A number of genetic disorders exhibit inter- and intra-familial variability. Understanding the factors that control the expression of disease genes should provide insight into the fundamental disease processes and will have implications for counselling patients. Different mechanisms can account for this variability, including environmental factors, genotype-phenotype correlations and imprinting. There is also evidence that, in a number of genetic diseases, gene expression is under the control of modifier loci. In cases where the biological basis of the genetic disease is understood, any genes involved in the pathogenic process represent candidate modifier genes which can easily be evaluated. Alternatively, modifiers can be identified through approaches such as mouse models. Since modifier genes will generally be common and because of confounding environmental influences, linkage analyses in humans will generally be based upon affected or discordant sib pairs. Discordant sib pairs represent an attractive option for linkage studies, because recurrence rates are high and the reduced survival characteristics associated with severe phenotypes will make the likelihood of obtaining clinical material from two living cases difficult. Furthermore, the use of discordant siblings will select for those siblings which possess sufficient dissimilarity at the modifier locus to overcome any shared environmental influence.

Genetic prognostic markers in colorectal cancer

The contribution of molecular genetics to colorectal cancer has been restricted largely to relatively rare inherited tumours and to the detection of germline mutations predisposing to these cancers. However, much is now also known about somatic events leading to colorectal cancer. A number of studies has been undertaken examining possible relations between genetic features and prognostic indices. While many of these studies are small and inconclusive, it is clear that a number of different pathways exist for the development of this cancer and some molecular characteristics correlate with clinicopathological features. With the advent of methods for the rapid genotyping of large numbers of colorectal cancers, it should be possible to evaluate fully the clinical usefulness of colorectal cancer genotypes through multivariate analyses.

Peutz-Jeghers syndrome

Peutz-Jeghers syndrome (PJS, MIM 175,2000) is a disease of autosomal dominant inheritance that is characterised by hamartomatous gastrointestinal polyps and mucocutaneous pigmentation. In addition to problems such as intussusception, PJS predisposes to cancers of several sites. The unusual combination of clinical features makes the identification of the defect underlying PJS particularly interesting. Recently, the PJS gene has been mapped to chromosome 19p13.

Exclusion of PTEN and 10q22-24 as the susceptibility locus for juvenile polyposis syndrome

Juvenile polyposis syndrome (JPS; MIM 174900) is an autosomal dominant condition with incomplete penetrance characterized by hamartomatous polyps of the gastrointestinal tract and a risk of gastrointestinal cancer. Gastrointestinal hamartomatous polyps are also present in Cowden syndrome (CS; MIM 158350) and Bannayan-Zonana syndrome (BZS; also called Ruvalcaba-Myhre-Smith syndrome; MIM 153480). The susceptibility locus for both CS and BZS has recently been identified as the novel tumor suppressor gene PTEN, encoding a dual specificity phosphatase, located at 10q23.3. A putative JPS locus, JP1, which most likely functions as a tumor suppressor, had previously been mapped to 10q22-24 in both familial and sporadic juvenile polyps. Given the shared clinical features of gastrointestinal hamartomatous polyps among the three syndromes and the coincident mapping of JP1 to the region of PTEN, we sought to determine whether JPS was allelic to CS and BZS by mutation analysis of PTEN and linkage approaches. Microsatellite markers spanning the CS/BZS locus (D10S219, D10S551, D10S579, and D10S541) were used to compute multipoint lod scores in eight informative families with JPS. Lod scores of < -2.0 were generated for the entire region, thus excluding PTEN and any genes within the flanking 20-cM interval as candidate loci for familial JPS under our statistical models. In addition, analysis of PTEN using a combination of denaturing gradient gel electrophoresis and direct sequencing was unable to identify a germline mutation in 14 families with JPS and 11 sporadic cases. Therefore, at least a proportion of JPS cases are not caused by germline PTEN alteration or by an alternative locus at 10q22-24.

Beta-catenin mutations in cell lines established from human colorectal cancers

beta-catenin has functions as both an adhesion and a signaling molecule. Disruption of these functions through mutations of the beta-catenin gene (CTNNB1) may be important in the development of colorectal tumors. We examined the entire coding sequence of beta-catenin by reverse transcriptase-PCR (RT-PCR) and direct sequencing of 23 human colorectal cancer cell lines from 21 patients. In two cell lines, there was apparent instability of the beta-catenin mRNA. Five different mutations (26%) were found in the remaining 21cell lines (from 19 patients). A three-base deletion (codon 45) was identified in the cell line HCT 116, whereas cell lines SW 48, HCA 46, CACO 2, and Colo 201 each contained single-base missense mutations (codons 33, 183, 245, and 287, respectively). All 23 cell lines had full-length beta-catenin protein that was detectable by Western blotting and that coprecipitated with E-cadherin. In three of the cell lines with CTNNB1 mutations, complexes of beta-catenin with alpha-catenin and APC were detectable. In SW48 and HCA 46, however, we did not detect complexes of beta-catenin protein with alpha-catenin and APC, respectively. These results show that selection of CTNNB1 mutations occurs in up to 26% of colorectal cancers from which cell lines are derived. In these cases, mutation selection is probably for altered beta-catenin function, which may significantly alter intracellular signaling and intercellular adhesion and may serve as a complement to APC mutations in the early stages of tumorigenesis.

The interactions of APC, E-cadherin and beta-catenin in tumour development and progression

Much progress has been made in identifying genes mutated during the development of colorectal carcinoma. Mutation of the APC gene in particular appears to be fundamental for colorectal tumour initiation. In contrast, loss of expression of E-cadherin appears to be a late event, which may be important in the development of invasion. Recent clarification of the function of APC, however, has shown that it exists in equilibrium with beta-catenin and E-cadherin. This review discusses the function of these molecules, their interactions, and how APC mutations may alter the equilibrium with beta-catenin and E-cadherin. It is argued that these changes cause aberrant architectural development of tissue, which results in loss of growth control. It is this escape from growth control, rather than acquisition of cell-autonomous growth, which results in the initial development of adenomas. The role of the E-cadherin-catenin unit in colorectal tumour invasion is discussed and the evidence is reviewed for the involvement of APC and E-cadherin in tumours arising from non-intestinal epithelia.

The interactions of APC, E-cadherin and beta-catenin in tumour development and progression

Much progress has been made in identifying genes mutated during the development of colorectal carcinoma. Mutation of the APC gene in particular appears to be fundamental for colorectal tumour initiation. In contrast, loss of expression of E-cadherin appears to be a late event, which may be important in the development of invasion. Recent clarification of the function of APC, however, has shown that it exists in equilibrium with beta-catenin and E-cadherin. This review discusses the function of these molecules, their interactions, and how APC mutations may alter the equilibrium with beta-catenin and E-cadherin. It is argued that these changes cause aberrant architectural development of tissue, which results in loss of growth control. It is this escape from growth control, rather than acquisition of cell-autonomous growth, which results in the initial development of adenomas. The role of the E-cadherin-catenin unit in colorectal tumour invasion is discussed and the evidence is reviewed for the involvement of APC and E-cadherin in tumours arising from non-intestinal epithelia.

Allele loss in colorectal cancer at the Cowden disease/juvenile polyposis locus on 10q

The genes that are mutated in inherited cancer syndromes are often involved in the pathogenesis of sporadic cancers of the types that characterize those syndromes. In colorectal cancer such loci include the familial adenomatous polyposis (APC) gene and the hereditary nonpolyposis colorectal cancer (DNA mismatch repair) genes. Juvenile hamartomatous polyposis syndromes, which include Juvenile Polyposis and Cowden disease, also predispose to colorectal cancer. The gene for Cowden disease has recently been localized to chromosome 10q22-q23, and a juvenile polyposis locus, JP1, has been reported as mapping to the same location. We have studied up to 70 cases of sporadic colorectal cancer for allele loss at markers predominantly on the long arm of chromosome 10, including loci flanking the putative Cowden Disease/JP1 locus. Frequencies of allele loss of about 35% were found close to this locus, whereas low frequencies of allele loss were found elsewhere on 10q. Mutations at the putative Cowden Disease/JP1 locus may therefore be important in sporadic colorectal cancer and fine mapping of allele loss on 10q in sporadic colon cancers may help to refine the position of this gene.

Game-theory models of interactions between tumour cells

The population of cells that comprises a tumour may consist of genetically different individuals. Often, such polymorphisms result from the expansion of a new, advantageous clone. The hypothesis is presented that sometimes tumour cells may adopt genetically-determined strategies to boost their own replication at the expense of other cells in the tumour. Simple game-theory models have been used to study this hypothesis, taking as an example the hypothetical advantage gained by tumour cells which produce a cytotoxin to harm other tumour cells. The models show that genotypes which cause cells to produce cytotoxic substances can spread through the tumour cell population, as can genotypes for resistance to the cytotoxin; in other circumstances, stable polymorphisms between these strategies can occur. The path of the tumour cell population to internal or external equilibrium is often complex, with large fluctuations in genotype frequencies. Flexible strategies are usually superior to fixed strategies. As in populations of whole organisms, 'social' interactions between tumour cells can act in favour of the individual cell at the expense of the tumour as a whole: strategies that retard the growth of the tumour can be selected and tumour regression is theoretically possible. Many mutations in tumours, especially in large or late-stage lesions, may play a role in relations between tumour cells rather than providing those cells with a simple replicative advantage.

Linkage analysis of candidate regions for coeliac disease genes

A strong HLA association is seen in coeliac disease [specifically to the DQ(alpha1*0501,beta1*0201 heterodimer], but this cannot entirely account for the increased risk seen in relatives of affected cases. One or more genes at HLA-unlinked loci also predispose to coeliac disease and are probably stronger determinants of disease susceptibility than HLA. A recent study has proposed a number of candidate regions on chromosomes 6p23 (distinct from HLA), 6p12, 3q27, 5q33.3, 7q31.3, 11p11, 15q26, 19p13.3, 19q13.1, 19q13.4 and 22cen for the location of a non-HLA linked susceptibility gene. We have examined these regions in 28 coeliac disease families by linkage analysis. There was excess sharing of chromosome 6p markers, but no support for a predisposition locus telomeric to HLA. No significant evidence in favour of linkage to coeliac disease was obtained for chromosomes 3q27, 5q33.3, 7q31.3, 11p11, 19p13.3, 19q13.1, 19q13.4 or 22cen. There was, however, excess sharing close to D15S642. The maximum non-parametric linkage score was 1.99 (P = 0.03). Although the evidence for linkage of coeliac disease to chromosome 15q26 is not strong, the well established association between coeliac disease and insulin dependent diabetes mellitus, together with the mapping of an IDDM susceptibility locus (IDDM3) to chromosome 15q26, provide indirect support for this as a candidate locus conferring susceptibility to coeliac disease in some families.

Frequency of germline hereditary non-polyposis colorectal cancer gene mutations in patients with multiple or early onset colorectal adenomas

BACKGROUND

The hereditary non-polyposis colorectal cancer (HNPCC) syndrome is caused by germline mutations in mismatch repair genes and predisposes individuals to cancers of the colon and other specific sites. On theoretical grounds, it is expected that patients with HNPCC also develop more colorectal adenomas than the general population. In essence, if the mutation rate is raised owing to mutations at a mismatch repair locus, more mutations are expected at loci such as APC (adenomatous polyposis coli) and more adenomas will start to grow. Not all data support this expectation, however.

AIM

To search for germline mutations at HNPCC loci in patients with multiple adenomas.

SUBJECTS

Twenty five patients (without known APC mutations) who have developed colorectal adenomas in unusually large numbers and, in some cases, at an early age.

METHODS

Germline APC mutations were excluded using the protein truction test for exon 15 and mismatch chemical cleavage analysis for remaining exons. Germline HNPCC mutations were detected by using PCR and single strand conformation polymorphism analysis.

RESULTS

Just one germline HNPCC mutation was found (4% of cases) and this was of uncertain functional effect.

CONCLUSIONS

In general, germline HNPCC mutations are not responsible for the phenotype of patients with multiple colonic adenomas. It is possible that patients with HNPCC tend to develop adenomas more frequently and earlier than the general population, but that this effect is relatively subtle. These results suggest that individuals with colorectal adenomas only should not strictly be classified as "affected" in HNPCC families (although they should certainly not be classified as "unaffected" either and may warrant intensive screening). In the absence of a personal or strong family history of colorectal cancer, it is probably not worthwhile performing diagnostic or predictive genetic testing for HNPCC mutations in subjects with colorectal adenomas. Although undetected APC mutations may be responsible for some of the phenotypes in this sample, as yet uncharacterised adenoma predisposing genes probably exist.

The interactions of APC, E-cadherin and beta-catenin in tumour development and progression

Much progress has been made in identifying genes mutated during the development of colorectal carcinoma. Mutation of the APC gene in particular appears to be fundamental for colorectal tumour initiation. In contrast, loss of expression of E-cadherin appears to be a late event, which may be important in the development of invasion. Recent clarification of the function of APC, however, has shown that it exists in equilibrium with beta-catenin and E-cadherin. This review discusses the function of these molecules, their interactions, and how APC mutations may alter the equilibrium with beta-catenin and E-cadherin. It is argued that these changes cause aberrant architectural development of tissue, which results in loss of growth control. It is this escape from growth control, rather than acquisition of cell-autonomous growth, which results in the initial development of adenomas. The role of the E-cadherin-catenin unit in colorectal tumour invasion is discussed and the evidence is reviewed for the involvement of APC and E-cadherin in tumours arising from non-intestinal epithelia.