• Tumor susceptibility genes
• Multistep carcinogenesis
• Somatic mutation
• Microsatellite instability
• Variable expressivity and incomplete penetrance
Major Phenotypic Features
• Age at onset: Middle adulthood
• Colorectal cancer
• Multiple primary cancers
History and Physical Findings
P.P., a 38-year-old banker and mother of three children, was referred to the cancer genetics clinic by her physician for counseling regarding her family history of cancer. Her father, brother, nephew, niece, and paternal uncle all developed colorectal cancer while her paternal grandmother had been diagnosed in her 40’s with uterine cancer. P.P. did not have a history of medical or surgical problems. The findings from her physical examination were normal. The geneticist explained to P.P. that her family history was suggestive of Lynch syndrome (also known as hereditary nonpolyposis colon cancer, HNPCC), and that the most efficient and effective way to determine the genetic cause of Lynch syndrome in her family was through molecular testing of a living affected family member. After some discussion with her niece, the only surviving affected family member, P.P. and her niece returned to the clinic for testing. Testing of an archived tumor sample from the niece's resected colon identified microsatellite instability (MSI); subsequent sequencing of DNA from a blood sample obtained from the niece revealed a germline mutation in MLH1. P.P. did not carry the mutation; therefore the geneticist counseled that her risk and her children's risk for development of cancer were similar to that of the general population. Her unaffected brother was found to carry the mutation and continued to have an annual screening colonoscopy.
Disease Etiology and Incidence
At least 50% of individuals in Western populations develop a colorectal tumor by the age of 70 years, and approximately 10% of these individuals eventually develop colorectal cancer. Lynch syndrome (MIM 120435) is a genetically heterogeneous autosomal dominant cancer predisposition syndrome that is often caused by mutations in DNA mismatch repair genes. Lynch syndrome has a prevalence of 2 to 5 per 1000 and accounts for approximately 3% to 8% of colorectal cancer.
In most colorectal cancers, including in familial adenomatous polyposis, the tumor karyotype becomes progressively more aneuploid (see Chapter 15). Approximately 15% of colorectal cancers do not have such chromosomal instability but have insertion or deletion mutations in repetitive sequences (MSI). Microsatellite instability occurs in 85% to 90% of Lynch syndrome tumors. Consistent with this observation, approximately 70% of Lynch syndrome families with carcinomas exhibiting MSI have germline mutations in one of four DNA mismatch repair genes: MSH2, MSH6, MLH1, or PMS2.
DNA mismatch repair reduces DNA replication errors by 1000-fold. Errors of DNA synthesis cause mispairing and deform the DNA double helix. A complex of mismatch repair proteins recruits other enzymes to excise the segment of newly synthesized mismatched DNA and resynthesize it.
As is typical for tumor suppressor genes, both alleles of a DNA mismatch repair gene must lose function to cause MSI. This somatic loss of function can occur by loss of heterozygosity, intragenic mutation, or promoter hypermethylation.
In Lynch syndrome, a number of microsatellite loci mutate during the progression from adenoma to carcinoma. Inactivation of genes containing microsatellite sequences could play key roles in tumor progression. For example, MSI induces frameshift mutations in the transforming growth factor receptor II gene (TGFBR2). Mutations within TGFBR2 cause the loss of TGFβRII expression, which reduces the ability of TGFβ to inhibit the growth of colonic epithelial cells. TGFBR2 mutations occur in early Lynch syndrome lesions and may contribute to the growth of adenomas. Lynch syndrome also results from epigenetic silencing of MSH2 caused by deletion of 3′ exons of EPCAM and intergenic regions directly upstream of MSH2.
MLH1 and MSH2 germline mutations account for approximately 90% of mutations in Lynch syndrome families. MSH6 mutations account for an additional 7% to 10%, whereas PMS2 mutations are found in fewer than 5% of cases.
Phenotype and Natural History
Although patients with Lynch syndrome develop polyps similar in number to those of the general population, they develop them at younger ages. Their median age at diagnosis with a colorectal adenocarcinoma is younger than 50 years, that is, 10 to 15 years younger than the general population (Fig. C-29). Patients with Lynch syndrome and a defined MLH1 or MSH2 germline mutation have an 80% lifetime risk for development of colorectal cancer; the penetrance of MSH6 and PMS2 mutations is much lower. Sixty percent to 70% of adenomas and carcinomas in Lynch syndrome occur between the splenic flexure and ileocecal junction. By way of contrast, most sporadic colorectal cancers (and cancer in familial adenomatous polyposis) (Case 15) occur in the descending colon and sigmoid. Carcinomas in Lynch syndrome are less likely to have chromosome instability and aneuploidy, and behave less aggressively than sporadic colon cancer. For this reason, patients with Lynch syndrome have a better age- and stage-adjusted prognosis than do patients with familial adenomatous polyposis or colorectal tumors with chromosome instability.
FIGURE C-29 Family segregating an MLH1 mutation. Note the frequent occurrence of colorectal cancer as well as other Lynch syndrome-associated cancers, such as endometrial cancer, pancreatic cancer, and ovarian cancer. Note that one family member had cancers of the colorectum and endometrium and that another had sporadic colon cancer (tested negative for family mutation). The consultand is indicated by an arrow. Shaded symbols indicate a diagnosis of cancer. Ages are shown directly below each symbol. A plus sign identifies carriers of the MLH1 mutation, and a minus signidentifies noncarriers. Cancer diagnoses are followed by the age at diagnosis. CRC, Colorectal cancer; endo, endometrial cancer; lung, lung cancer; ovary, ovarian cancer; panc, pancreatic cancer. See Sources & Acknowledgments.
In addition to colorectal cancer, Lynch-associated cancers include cancer of the stomach, small bowel, pancreas, kidney, endometrium, and ovaries; cancers of the lung and breast are not associated (see Fig. C-29). Patients with Lynch syndrome and a defined germline mutation have a more than 90% lifetime risk for development of colorectal cancer, one of these associated cancers, or both.
Lynch syndrome patients do not have distinguishing physical features. The minimal criteria for considering Lynch syndrome are the occurrence of colorectal cancer or another Lynch syndrome-associated tumor in three members of a family, at least two of whom are first-degree relatives, across two or more generations, and the development of colorectal cancer in at least one affected individual before the age of 50 years. In patients without a family history but with early-onset colorectal cancer, genetic testing for Lynch syndrome is ideally performed in a stepwise manner: evaluation of tumor tissue for MSI through molecular MSI testing and/or immunohistochemistry of the four mismatch repair proteins. The presence of MSI in the tumor alone is not sufficient to diagnosis Lynch syndrome because 10% to 15% of sporadic colorectal cancers exhibit MSI due to somatic methylation of the MLH1 promoter. Immunohistochemistry testing helps identify the mismatch repair gene that most likely harbors a germline mutation.
Early recognition of Lynch syndrome is necessary for effective intervention; surveillance colonoscopy of the proximal colon beginning at the age of 25 years increases life span expectancy by 13.5 years, and prophylactic surgical removal of the colon at the age of 25 years increases life span expectancy by more than 15 years. Surveillance endometrial biopsies and abdominal ultrasound scans for at-risk women have not proved to be effective preventive measures for the uterine or ovarian cancer seen in this condition. In families with known germline mutations, identification of the DNA mismatch repair gene mutation can focus surveillance on those patients carrying the mutation, but in Lynch syndrome families without an identified germline mutation, the absence of a mutation does not negate the need for frequent surveillance.
The empirical Western general population risk for the development of colorectal cancer is 5% to 6%. This risk is markedly modified by family history. Patients with a first-degree relative with colorectal cancer have a 1.7 relative risk; this relative risk increases to 2.75 if two or more first-degree relatives had colorectal cancer. If an affected first-degree relative developed colorectal cancer before 44 years of age, the relative risk increases to more than 5.
In contrast, a patient with a DNA mismatch repair gene germline mutation has a 50% risk for having a child carrying a germline mutation. Each child carrying such a mutation has a lifetime cancer risk of up to 90%, assuming the 80% penetrance of MLH1 or MSH2 mutations is responsible for a cancer risk over and above the background risk in the general population for colon cancer and the other cancers of the types associated with Lynch syndrome. Prenatal diagnosis is highly controversial and not routine but is theoretically possible if the germline mutation has been identified in the parent. Because of incomplete penetrance and variation in expressivity, the severity and onset of Lynch syndrome and the occurrence of associated cancers cannot be predicted.
Questions for Small Group Discussion
1. Compare the mechanisms of tumorigenesis in disorders of nucleotide excision repair, chromosomal instability, and microsatellite instability.
2. How should a patient with a family history of Lynch syndrome be counseled if testing for DNA mismatch repair gene mutations is positive? Negative?
3. Discuss the ethics of testing of minors for Lynch syndrome.
Brenner H, Kloor M, Pox CP. Colorectal cancer. Lancet. 2014;383:1490–1502.
Kohlmann W, Gruber SB. Lynch syndrome. Pagon RA, Bird TD, Dolan CR, et al. GeneReview. 2004 [Updated 2014]; Available from:] http://www.ncbi.nlm.nih.gov/books/NBK1211/.
Matloff J, Lucas A, Polydorides AD, et al. Molecular tumor testing for Lynch syndrome in patients with colorectal cancer. J Natl Compr Canc Netw. 2013;11:1380–1385.