Identification of individuals at increased risk of hereditary cancer allows for the possibility of screening and early cancer detection, possibly resulting in decreased disease-specific mortality, and is the justification for germline genetic testing for specific cancer risk alleles. However, factors of prevalence and age-specific penetrance, effectiveness and invasiveness of screening procedures, and efficacy of early detection influence the potential benefit of such an approach.
For one of the most common hereditary cancer syndromes, Lynch syndrome (LS), also known as hereditary nonpolyposis colorectal cancer (HNPCC), various sets of clinical criteria, combined with pathologic phenotypic characteristics of tumor tissues in probands, have been used to identify individuals at risk in whom it is important to consider germline genetic testing for deleterious mutations in 1 of 4 DNA mismatch repair genes (MLH1, MSH2, MSH6, or PMS2). Lynch syndrome was originally defined by the “Amsterdam” clinical criteria as a history of at least 3 family members with histologically confirmed colorectal cancer (CRC) involving 2 generations with at least 1 person diagnosed before age 50 years.1 Although this approach is fairly specific in identifying families with highly penetrant LS, it is also overly restrictive and does not consider the possibility of later-onset variants of the disease, the implications of extracolonic tumors, or the limitations imposed by small family size.
Many families with known LS do not meet the original Amsterdam criteria, and this approach misses many families because of poor sensitivity. Therefore, it has been suggested, given the availability of an immunohistochemistry-based screening test for mismatch repair that closely mimics phenotypic microsatellite instability, that testing all colorectal (and perhaps endometrial) cancers for loss of proteins associated with mismatch DNA repair deficiency, or all cancers below some age cutoff, may overcome the limitations of selective criteria. However, the effectiveness of such a “universal” approach to screening for LS has not been tested in a population-based manner.
In this issue of JAMA, Moreira and colleagues2 address this question by performing a pooled-data analysis of a large set of population-based patient cohorts from around the world to determine the sensitivity and efficiency of several different strategies for LS screening, including the “universal” approach. Using more than 10 000 CRC samples, the authors found overall that universal tumor testing for mismatch repair deficiency was superior in sensitivity to the Bethesda guidelines, which incorporate personal and family history information, and which have been reported to be more accurate than the Amsterdam criteria,3 or an approach using an age cut-off of 70 years old, but that a hybrid of testing all tumors in individuals 70 years or younger and in older patients who meet Bethesda guidelines provides a reasonable compromise that may result in substantial cost savings.
The study by Moreira et al2 confirms that the prevalence of LS is high enough among patients with CRC, 3.1% in the whole series, that screening should be considered. However, in the EPICOLON cohort, consisting of patients newly diagnosed with CRC in 20 community hospitals in Spain,3 the prevalence was only 0.9% compared with 2.9% to 3.5% in the other 3 cohorts analyzed. The current study does not address possible explanations for this difference. Some patients in the study by Moreira et al were not drawn from a population-based registry but were excluded from analyses on the performance characteristics of screening strategies. Because most germline tests were driven by abnormal tumor testing results and not all patients underwent “gold-standard” (germline) testing, the high sensitivity estimates for screening strategies are probably somewhat optimistic. In addition, the PMS2 gene was not tested in many patients, so some persons with PMS2 mutations were probably missed. In this study, microsatellite instability testing added little to immunohistochemistry, but not all previous studies have found similar results. Selective BRAF mutation testing or promoter methylation testing to identify sporadic CRC were not performed, but the focus of the study was on the sensitivity of screening strategies, not specificity.
These results highlight the limitations of various clinical criteria to identify persons with LS, particularly those with mutations in the MSH6 or PMS2 genes. The study results should remind clinicians that simply asking about a family history of CRC in a first-degree relative will miss the majority of patients with LS: only 43% of patients with LS had such a family history—approximately 50% of those with MLH1 and MSH2 gene mutations and less than 20% of those with MSH6 and PMS2 gene mutations. Furthermore, although the mean age at CRC diagnosis was 48 years in LS patients, only 45% were diagnosed with CRC at 50 years or younger.
The potential for individualized preventive medicine provides the rationale for screening for LS. In deciding whether to establish widespread screening for LS in selected subgroups, the same considerations that govern screening for CRC in the general population could be applied. The target condition must be common enough to justify screening. A long asymptomatic period must allow for effective interventions. The potential benefits must outweigh the risks. The aggregate costs of screening and its consequences must be acceptable.
Moreira et al2 confirm that a variety of strategies can identify a significant number of persons with LS among patients with CRC. The benefits of intensive CRC and adenoma screening and prophylactic hysterectomy-oophorectomy in LS are well established.4 – 7 This raises several questions. What is the balance between the potential benefits and harms of screening? Are the economic costs acceptable? Which screening strategies are preferred?
Several studies have addressed the potential psychological harms of testing for LS. These studies include select populations and are observational. Patients considering genetic testing can develop short-term increases in anxiety, distress, and fear of cancer or death, especially among mutation carriers.8 – 10 Generally, a person’s psychological state returns to baseline after several months. Longer follow-up has demonstrated no major adverse psychological consequences for either mutation carriers or noncarriers after 1 and 3 years.8 – 9 ,11 However, some groups, such as younger men affected by cancer, may be at higher risk for adverse psychological effects.12 Patients may express fear concerning discrimination in employment or health insurance. In the United States, patients may be counseled that such discrimination is illegal under the Genetic Information Nondiscrimination Act.13
In the absence of controlled studies evaluating the long-term consequences of different screening strategies for LS, computerized decision analytic modeling can be used to explore critical questions. Several modeling studies with long-term time horizons have suggested that screening for LS among persons with CRC is likely to be cost-effective.14 – 17 The same may apply to women with endometrial cancer.18 One study that incorporated the potential short-term adverse effects LS testing can have on quality of life19 found that the long-term gains in life expectancy are likely to outweigh any short-term decreases in quality of life, at acceptable costs.20
Key issues raised by the results of Moreira et al2 were explored in a modeling study.17 Moreira et al found a relatively small incremental yield of universal tumor screening vs a highly sensitive selection strategy based on clinical criteria. The modeling study suggested that tumor testing strategies are likely to be costly compared with clinical criteria strategies when both are implemented optimally.17 However, in the model, when the clinical criteria strategies failed to be implemented in as few as 15% of patients, tumor testing strategies became cost-effective. In clinical practice, routine testing of tumors in pathology laboratories may be more feasible than ensuring widespread application of clinical criteria.
Moreira et al2 address the important issue of age at CRC diagnosis as a factor to inform screening strategies. The modeling study17 estimated that immunohistochemistry-based tumor testing in all persons vs only in those 70 years and younger could be considered cost-effective depending on society’s willingness to pay for preventive services. Using the actual age distribution at CRC diagnosis in the study of Moreira et al, instead of the model’s original assumptions, would result in enhanced cost-effectiveness for the “universal” approach.
Lynch syndrome affects families, not only individuals. Patients often identify the potential benefits to their family, especially their children, as a motivating force driving acceptance of genetic testing.21 However, the published uptake rates for genetic testing among relatives at risk have varied from 34% to 52%.22 – 23 The number of relatives unaffected by cancer but at risk for LS who undergo genetic testing is a key determinant of the cost-effectiveness of any screening strategy.16 – 17 Future public health efforts must address this critical factor.
A recent survey of US hospitals reported that routine tumor testing with immunohistochemistry, microsatellite instability, or both is currently performed at 71% of National Cancer Institute comprehensive cancer centers, 36% of American College of Surgeons–accredited community hospital comprehensive cancer programs, but only 15% of community hospital cancer programs.24 Routine tumor testing with immunohistochemistry or microsatellite instability does not require written consent. The authors suggested that this approach to testing may reflect an emerging standard of care. Routine tumor testing programs require mechanisms to track results, contact patients in ways patients will accept, and facilitate consultation with genetics professionals. Those dealing with results must understand complicating factors, including variants of uncertain significance, and must handle difficult cases, such as patients with classic family pedigrees in whom no mutation is found by current methods.
In the not too distant future, advances in genomic sequencing will challenge current genetic testing approaches. Commercial panels of “cancer genes” are already emerging. With the anticipated further reductions in the costs of DNA sequencing, up-front germline testing at the time of CRC diagnosis could become the most cost-effective strategy to screen for LS.17 Such testing would require written informed consent. Population-based genomic profiling could revolutionize the approach to identifying persons with LS.
The majority of patients with CRC do not have LS. But in the haystack of patients with CRC, those with LS are more like large knitting needles than tiny sewing needles—and a systematic search can find them. The investments of effort and resources required for this search can be rewarded by reductions in cancer incidence and mortality that are possible among patients and their unsuspecting relatives.
Context Lynch syndrome is the most common form of hereditary colorectal cancer (CRC) and is caused by germline mutations in DNA mismatch repair (MMR) genes. Identification of gene carriers currently relies on germline analysis in patients with MMR-deficient tumors, but criteria to select individuals in whom tumor MMR testing should be performed are unclear.
Objective To establish a highly sensitive and efficient strategy for the identification of MMR gene mutation carriers among CRC probands.
Design, Setting, and Patients Pooled-data analysis of 4 large cohorts of newly diagnosed CRC probands recruited between 1994 and 2010 (n = 10 206) from the Colon Cancer Family Registry, the EPICOLON project, the Ohio State University, and the University of Helsinki examining personal, tumor-related, and family characteristics, as well as microsatellite instability, tumor MMR immunostaining, and germline MMR mutational status data.
Main Outcome Measures Performance characteristics of selected strategies (Bethesda guidelines, Jerusalem recommendations, and those derived from a bivariate/multivariate analysis of variables associated with Lynch syndrome) were compared with tumor MMR testing of all CRC patients (universal screening).
Results Of 10 206 informative, unrelated CRC probands, 312 (3.1%) were MMR gene mutation carriers. In the population-based cohorts (n = 3671 probands), the universal screening approach (sensitivity, 100%; 95% CI, 99.3%-100%; specificity, 93.0%; 95% CI, 92.0%-93.7%; diagnostic yield, 2.2%; 95% CI, 1.7%-2.7%) was superior to the use of Bethesda guidelines (sensitivity, 87.8%; 95% CI, 78.9%-93.2%; specificity, 97.5%; 95% CI, 96.9%-98.0%; diagnostic yield, 2.0%; 95% CI, 1.5%-2.4%; P < .001), Jerusalem recommendations (sensitivity, 85.4%; 95% CI, 77.1%-93.6%; specificity, 96.7%; 95% CI, 96.0%-97.2%; diagnostic yield, 1.9%; 95% CI, 1.4%-2.3%; P < .001), and a selective strategy based on tumor MMR testing of cases with CRC diagnosed at age 70 years or younger and in older patients fulfilling the Bethesda guidelines (sensitivity, 95.1%; 95% CI, 89.8%-99.0%; specificity, 95.5%; 95% CI, 94.7%-96.1%; diagnostic yield, 2.1%; 95% CI, 1.6%-2.6%; P < .001). This selective strategy missed 4.9% of Lynch syndrome cases but resulted in 34.8% fewer cases requiring tumor MMR testing and 28.6% fewer cases undergoing germline mutational analysis than the universal approach.
Conclusion Universal tumor MMR testing among CRC probands had a greater sensitivity for the identification of Lynch syndrome compared with multiple alternative strategies, although the increase in the diagnostic yield was modest.
- Lynch Syndrome and other non-polyposis inherited cancer syndromes (familyhistorybowelcancer.wordpress.com)
- Preoperative Testing For Lynch Syndrome Yields ‘Huge’ Benefit (familyhistorybowelcancer.wordpress.com)
- Lynch Syndrome (familyhistorybowelcancer.wordpress.com)
- Variants of Unknown Significance (VUS): A new trial for Lynch Syndrome patients (familyhistorybowelcancer.wordpress.com)
- Risks of Primary Extracolonic Cancers Following Colorectal Cancer in Lynch Syndrome (familyhistorybowelcancer.wordpress.com)