Genetic Testing for Lynch Syndrome, an Inherited Cancer of the Bowel, Endometrium, and Ovary

Abstract

Genetic screening for the mismatch repair genes found in patients with Lynch syndrome leads to improvements in health outcomes among carriers and members of their family. Clinicians now have a simple and easily employed means of determining if an individual carries the genetic mutations found with Lynch syndrome. This article reviews the background and incidence of Lynch syndrome and presents screening criteria, including the use of Web-based algorithms to estimate the likelihood of an individual having inherited Lynch mutations. Comprehensive management plans based on individual risk and family history plus appropriate preventive measures are recommended. Primary care providers including obstetrician-gynecologists are encouraged to adopt a proactive, evidence-based approach to address patients and their relatives with Lynch syndrome.

In the United States, over 140,000 new cases of colorectal and 47,000 new cases of endometrial cancer are diagnosed each year. Lynch syndrome is the most common hereditary cause of colorectal and endometrial cancer in women. It accounts for 2% to 4% of all colorectal cancers and 2% to 5% of endometrial cancers in women.¹

Lynch syndrome, originally termed hereditary nonpolyposis colorectal cancer (HNPCC), is an autosomal dominant multicancer disorder. In women, Lynch syndrome is associated with more frequent and earlier than expected onset of colorectal, endometrial, ovarian, and other cancers.² First characterized in 1966 by Henry T. Lynch, the syndrome is caused by genetic defects in one or more DNA mismatch repair (MMR) genes, including MLH1, MSH2, MSH6, and PMS2.³ These MMR genes identify and correct mismatched DNA base pairs.⁴ Specific MMR gene defects are associated with specific clinical presentations, types of cancer, and age of onset. For example, the majority (up to 80%) of colorectal cancers associated with Lynch syndrome are the result of mutations in either MLH1 or MSH2. However, these mutations are not uniform, and are made up of a broad spectrum of truncating, frame shift, and missense mutations. Lynch syndrome-associated cancers have a high incidence of synchronous (different cancers occurring at the same time) and metachronous (different cancers occurring at separate times) lesions. Other cancers found more frequently in patients with Lynch syndrome include stomach, small intestine, pancreas, prostate, liver, kidney, urinary tract, brain, and skin. In addition, germline MMR gene mutations linked to Lynch syndrome account for 10% to 15% of nonserous epithelial ovarian cancers.⁵

Background

In the era of personalized medicine, clinicians need to increase their understanding of genetic susceptibility to cancer, determine when to screen unaffected at-risk individuals, and manage ongoing surveillance for cost-effective improvement in health outcomes. Commercially available testing for MLH1 and MSH2 was introduced in the 1990s.⁶ The testing is performed on DNA isolated from peripheral blood mononuclear cells obtained by blood draw or from an oral rinse. Today, in addition to MLH1 and MSH2, mutations in the genes MSH6, PMS2, and EPCAM are also included in testing of individuals suspected to have Lynch syndrome. Testing is also available for all of the MMR gene mutations associated with Lynch syndrome.¹

Risk of Cancer

Individuals from the general population have a 2% lifetime risk of developing colorectal cancer, whereas the risk for patients with Lynch syndrome is over 80%.¹ In families with Lynch syndrome, colorectal cancers commonly onset at a younger age and progress more rapidly; 25% of affected family members are found to have colorectal disease by age 50 compared with only 0.2% in the population overall. Adenomatous polyps commonly take 10 or more years to progress to cancer for patients in the general population, but for those with Lynch syndrome the transition can occur in just 30 months. Women with Lynch syndrome have up to a 71% lifetime risk of developing endometrial cancer; for women in the general population, the lifetime risk is only 1.5%.¹ Lynch syndrome patients are also at increased risk for having a second primary cancer: up to 50% within 15 years as compared with the general population, where the risk of having a second cancer is only around 5%. For half of women with Lynch syndrome and both endometrial and colon cancer, the endometrial cancer presented first. With Lynch syndrome, the risk of a woman having ovarian cancer is 10% to 15% (compared with < 1% in the general population); 62% of cases were diagnosed before age 50 and MSH2/MSH6 mutations accounted for 82% of these ovarian tumors.⁵

Who Should Be Tested?

Because of the risk of early-onset colorectal and endometrial cancer, women should be tested for the most common genetic defects found in Lynch syndrome if they have a primary relative (parent, sibling, or child) with Lynch syndrome or colorectal cancer or other malignancy associated with Lynch syndrome, especially if the proband’s cancer developed before age 50.

One approach to determine who needs to be screened for Lynch syndrome is simply to test all surgical specimens from patients newly diagnosed with colorectal or endometrial cancer for the presence of characteristic defects in DNA MMR gene expression. Tumors containing MMR gene mutations associated with Lynch syndrome often have a characteristic histologic appearance exhibiting microsatellite instability (MSI), which is a surrogate marker for DNA mismatch repair gene dysfunction. In addition, immunohistochemistry can be used to show the presence or absence of the protein products of mismatch repair genes.⁷ MSI, the accumulation of errors in short repetitive DNA sequences, is found in 90% of patients with MLH1 and MLH2 Lynch syndrome-related colorectal cancer but is seen less frequently in patients with MSH6 mutations. Families with MSH6 mutations are slightly different. They are more likely to have endometrial than colorectal cancer, and are found to exhibit onset of disease at a later age.⁸ Currently, the combined use of immunohistochemistry and MIS profiling is the most advanced method available to identify candidates for genetic testing for Lynch syndrome.⁹ Unfortunately, 10% to 15% of sporadic cases of colon cancer also exhibit MSI or abnormal immunohistochemistry test results of mismatch repair genes, so that only a portion of persons with abnormal results on tumor testing will actually have Lynch syndrome. In addition, most false-positive results arise from an acquired methylation of the MLH1 gene in tumor tissue. Direct testing for this promoter methylation (or by testing for a specific BRAF mutation) demonstrates that the change is acquired and the MLH1 dysfunction is not inherited. This added step may decrease by half the number of patients in whom genetic testing is indicated by positive MSI or immunohistochemistry tumor results.⁹

Making the Diagnosis

Lynch syndrome should be suspected on the basis of a strong family history of colorectal cancer, endometrial cancer, or any of the other types of cancers associated with the syndrome (Table 1, Figure 1). The finding of one or more MMR gene mutations confirms the diagnosis of Lynch syndrome. Having confirmed the diagnosis in a patient or proband, at-risk family members can then be offered testing for the presence of the particular gene mutation found in the index case. Testing for a specific mismatch repair gene mutation is more expedient and cost effective than screening an at-risk individual for all the genetic possibilities associated with the syndrome.

TABLE 1.

Criteria to Identify High-Risk Candidates for Molecular Genetic Testing for Lynch Syndrome

Early onset colorectal cancer (< age 50 years)

Early onset endometrial cancer (< age 50 years)

Previously identified Lynch syndrome mutation within the family

Two or more Lynch syndrome cancersa

In the same individual or

Among close relatives

^aLynch syndrome cancers include: colorectal, endometrial, gastric, ovarian, ureter/renal pelvis, biliary tract, small bowel, pancreas, brain, and sebaceous adenoma.

Figure 1.

Family history questionnaire for common hereditary cancer syndromes.

Over the past few years, repeated attempts have been made to establish clinical criteria to aid in deciding who is at risk and should be tested for Lynch syndrome. The oldest of these, the Amsterdam criteria (Table 2), were established in 1991 to facilitate identification of families with Lynch syndrome and to make diagnosis more uniform. Shortcomings of the original Amsterdam criteria include the realization that multiple generations were not always affected because of small family size, variable penetrance, or incomplete family history.⁹ The Amsterdam II criteria (Table 3) and Modified Amsterdam criteria (Table 4) were therefore developed to address these concerns and include the extracolonic cancers associated with Lynch syndrome.

TABLE 2.

Amsterdam Criteria (1991)

Three or more family members with a confirmed diagnosis of colorectal cancer, one of whom is a first-degree (parent, sibling, child) relative of the other two

Two successive affected generations

One or more colon cancers diagnosed < age 50 years

Familial adenomatous polyposis (FAP) has been excluded.

Data from Vasen HF et al. Dis Colon Rectum. 1991;34:424–425.

TABLE 3.

Amsterdam Criteria II (1998)

Three or more family members with Lynch syndrome-related cancers, one of whom is a first-degree relative of the other two

Two successive affected generations

One or more of the Lynch syndrome-related cancers diagnosed < age 50 years

Familial adenomatous polyposis (FAP) has been excluded.

Data from Vasen HF et al. Gastroenterology. 1999;116:1453–1456.

TABLe 4.

Modified amsterdam Criteria

In very small families, two cases of colorectal cancer in first-degree relatives spanning at least two generations

In families with two first-degree relatives with colorectal cancer, a third relative with an unusual early-onset cancer or endometrial cancer

Data from Umar A et al. Nat Rev Cancer. 2004;4:153–158.

However, the Amsterdam criteria lack sensitivity. About 35% of patients meeting the Amsterdam criteria do not have a DNA mismatch-repair gene mutation and 50% of families with the syndrome fail to meet the Amsterdam criteria. ⁹ The Bethesda Guidelines (Table 5) were developed specifically to identify individuals and additional family members who should undergo surgical specimen tumor testing for the genetic MMR gene mutations found in Lynch syndrome.⁹

TABLE 5.

Bethesda Criteria (2002)

Colorectal cancer diagnosed in a patient < age 50 years

Presence of synchronous or metachronous Lynch syndrome-associated tumors, regardless of age

Colorectal cancer with MSI-H histology diagnosed in a patient < age 60 years

Colorectal cancer diagnosed in a patient with one or more first-degree relatives with a Lynch syndrome-related cancer, with one of the cancers being diagnosed at < age 50 years

Colorectal cancer diagnosed in a patient with two or more first- or seconddegree relatives with Lynch syndrome-related cancers regardless of age

MSI-H, high microsatellite instability.

Data from Umar A et al. J Natl Cancer Inst. 2004;96:261–268.

Computed Algorithm Methods

An entirely different approach to assessing risk for Lynch syndrome uses Web-based algorithms to examine the types of cancer in a family, the age at occurrence, and the relationships of family members with cancer to estimate the likelihood of an individual having Lynch syndrome. Several computerized algorithms are available online for this purpose, including MMRpredict, Leiden, MMRpro, and PREMM.^1,2,7,9 These algorithms can be used to determine whether MSI or immunohistochemical testing should be performed on available tumor tissue (colon, endometrium, and ovary) or whether the risk is sufficient to proceed directly to genetic testing. Modeling studies consider relevant information to determine the optimal approach for testing an individual who is concerned about the possibility of having one or more of the genetic defects associated with Lynch syndrome.

Using Markov modeling, Ladabaum and colleagues⁷ determined that immunohistochemical and MSI testing of all colorectal tumor specimens followed by germline genetic testing of family members in cases where the proband demonstrated having a Lynch syndrome mutation was both medically appropriate and cost effective. Costeffectiveness improved with the number of relatives tested in situations where a disease-causing mutation was found in the index case.⁷ A similar argument can be made for using the same approach and testing all specimens of endometrial cancer for Lynch syndrome-related mutations.

The more common clinical presentation is that of an unaffected woman with a family history suspicious for Lynch syndrome who wants to know her risk for cancer. Practitioners facing this question may want to use one of the modeling algorithms to assess the patient’s threshold risk for having Lynch syndrome. Dinh and colleagues¹ applied the PREMM model^1,2,7 to determine the risk for the syndrome in individuals regardless of whether they had a history of colorectal or endometrial cancer. They found that primary screening with genetic testing starting at age 25 to 35 years in patients with a 5% or higher risk for having a Lynch syndrome mutation was both medically appropriate and cost effective. A person’s risk for Lynch syndrome is quickly and easily calculated by using the PREMM model^1,2,7 or any of the other models previously mentioned. The demonstrated success of this approach is important not only in view of the frequency of the presentation of unaffected persons with high familial risk, but also because tissue from relatives of such persons may be difficult or impossible to obtain. This primary screening approach is distinct from current guidelines, all of which recommend clinically informed genetic testing in relatives of family members who have already developed malignancies.¹

How Should a Woman Known to Carry a Lynch Syndrome Genetic Mutation Be Managed?

With their significantly higher risk for developing colorectal, endometrial, ovarian, and other cancers, women known to carry one of the genetic mutations associated with Lynch syndrome should plan on more aggressive screening protocols and consider following a comprehensive management plan based on their individual circumstances and family history. One possible approach is outlined next (Figure 2).

Figure 2.

Recommended management plan for patients with Lynch syndrome. CA-125, cancer antigen 125; TAH-BSO, total abdominal hysterectomy-bilateral salpingo-oophorectomy.

Treatment

In patients with Lynch syndrome, surgical excision of the tumor and often of the affected organ remains the frontline of therapy. Timing of surgical intervention depends on the patient, her reproductive desires, her willingness to accept future risk, and her preference given the various surgical interventions at her disposal.

Yang and colleagues¹⁰ used a decision analytic model to investigate the cost-effectiveness of prophylactic surgery (total abdominal hysterectomy-bilateral salpingo-oophorectomy; TAH-BSO) at age 30 years versus surveillance, annual gynecologic screening, and annual gynecologic examination in women with Lynch syndrome and their lifetime risk of developing endometrial and ovarian cancer. The authors determined that risk-reducing surgery (TAH-BSO), from a societal health care cost perspective, is the least expensive option and leads to the highest number of quality adjusted life-years (QALY) gained. They recommend that women with Lynch syndrome should consider undergoing a prophylactic TAH-BSO as early as possible, particularly after completion of childbearing. In their study, the decision on colorectal surveillance and prevention was separate and independent from the discussion of gynecologic cancer risks.¹⁰

Strategies and Evidence for Screening

The decision to screen either a particular population or a specific patient for genetic disease involves weighing the benefits against the costs. In the case of Lynch syndrome, the most important benefits are the reduction in the risk of death and the number of life-years gained. Costs include monetary as well as the toll of the screening regime itself (such as pain, inconvenience, and anxiety). The ratio of benefit to cost varies significantly with the woman’s age.

Ladabaum and colleagues⁷ maintain that screening for Lynch syndrome up to age 70 years is reasonable, and screening all persons with colorectal or endometrial cancer (regardless of age) may be desirable, depending on society’s willingness and ability to support the expense.

For individuals, widespread colorectal tumor testing to identify families with Lynch syndrome yields substantial clinical benefits at acceptable costs.⁷ Such strategies provide greater benefits to younger relatives without cancer than to older probands who present with colorectal cancer, and may have greater benefits to women than to men. The greater benefits for women were attributed to prevention of gynecologic cancers and to women having a longer life expectancy than men.⁷

For society, the number of relatives tested per proband is critically important. In order to be cost effective, three to four first-order relatives per proband need to be tested. In this setting, the cost could be as little as $36,000 per life-year gained.⁷ Costs of less than $50,000 per life-year gained are generally considered to be cost effective.

Clinicians should encourage patients known to carry genetic mutations associated with Lynch syndrome to follow enhanced surveillance guidelines. Clinicians should also promote genetic testing for Lynch syndrome-associated genetic mutations in the family members of probands with Lynch syndrome. Family members should undergo genetic testing because:

• A negative test result provides valuable information to direct future medical management.

• Medical management is different from treatment based on family history alone.

• Options such as increased surveillance, chemoprevention, and prophylactic surgery have been proven to greatly reduce the risk of developing cancer in patients with Lynch syndrome.

Is There a Place for Universal Screening?

Distinct from current guidelines that recommend clinically informed genetic testing in individuals who have already developed malignancies is the possibility of using a primary screening approach wherein those who might wish to know their genetic status and risk for carrying one of the Lynch syndrome mutations could be tested. From a cost-effectiveness perspective, although universal screening offers the greatest benefit in clinical outcomes, it does so at the least attractive, highest cost. However, if the threshold for genetic testing is set at 5% relative risk and testing is begun between age 25 and 35 years, the cost-effectiveness value falls below the benchmark of $50,000 per QALY gained.7 Adoption of any proposed screening strategy carries a dual obligation for individuals to consult with their physicians and for physicians to implement the strategy. Universal testing or testing on demand without first assessing the patient’s risk for having Lynch syndrome is neither cost effective nor medically indicated at this time. In the future, as the expense of testing for hereditary diseases becomes less costly, the indications for testing for Lynch syndrome and other hereditary conditions may become more cost effective and more widely utilized.

Summary

Determining who carries the genetic mutations associated with Lynch syndrome is important so that affected individuals and their relatives can take appropriate preventive measures. Several screening strategies exist (Table 6). Testing all colorectal and endometrial cancer specimens for genetic mutations associated with Lynch syndrome is cost effective and helps to identify relatives who might be at risk and would benefit from genetic testing.

TABLE 6.

Recommendations for Screening

Two strategies for screening exist: Use clinical data with or without online risk assessment algorithms to calculate an individual’s risk for Lynch syndrome. If the risk is determined to be 5% or greater, proceed to genetic testing. This approach has been shown to be cost effective, especially in situations where a family member is known to have had a cancer suspicious for Lynch syndrome but tumor tissue from that relative is not available for testing.8 Test all surgical specimens of newly diagnosed colorectal or endometrial cancer for mutations associated with Lynch syndrome. If the tumor is positive, test healthy first-degree relatives. An important reason to test the tumor first is that a negative result makes the existence of Lynch syndrome very unlikely and further testing of family members becomes unnecessary.

Whereas discussions about genetic risk and genetic counseling are typically handled in specialized clinics, a growing awareness of the importance of genetics and hereditary disease in clinical practice mandates that all clinicians have greater knowledge and understanding of the relevant principles and clinical applications of genetic testing. Primary care providers and obstetrician-gynecologists should utilize genetic testing when clinically indicated. In this way, primary care providers may join with oncologists and surgical specialists in the fundamental roles of prevention, surveillance, and management of patients with Lynch syndrome.¹

Primary genetic screening for mismatch repair genes found in Lynch syndrome patients leads to improvements in health outcomes among carriers and affected family members and is cost effective if their risk exceeds a 5% lifetime probability of developing cancer.⁶ This finding supports the concept that genetic screening of unaffected at-risk individuals, when conducted in association with appropriate risk assessment and followed by surveillance for colorectal and endometrial cancer, can improve health outcomes in a cost-effective manner. Furthermore, this strategy offers an evidence-based justification for a shift in the clinical approach to Lynch syndrome from one that is reactive to one that is proactive.

In the future, with greater appreciation of the clinical significance of specific gene mutations in hereditary disease syndromes, clinicians should be able to offer more individualized options to carriers of the mismatch repair genes MLH1, MSH2, MSH6, PMS2, and EPCAM found in Lynch syndrome.^6,8 The role of environmental factors in the development of malignancies in patients with Lynch syndrome requires further study.⁶

Main Points.

• Diagnosis and management of individuals with, and at risk for, Lynch syndrome should move from reactive to proactive.

• In the future, as knowledge of clinical significance expands, specific gene mutations will become increasingly important relative to screening for disease and management of disease potential.

• Genetic counseling will become a major component of primary care as more is learned about inherited conditions. Testing for genetic disease will be a large part of a primary care practice.