Dig Dis Sci DOI 10.1007/s10620-014-3234-z
ORIGINAL ARTICLE
Selective Versus Universal Screening for Lynch Syndrome: A Six-Year Clinical Experience Trilokesh D. Kidambi • Amie Blanco • Megan Myers • Peggy Conrad • Kate Loranger Jonathan P. Terdiman
•
Received: 24 April 2014 / Accepted: 26 May 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Lynch syndrome is the most common cause of hereditary colorectal cancer (CRC) and confers increased risk of other cancers. Identification of patients improves morbidity and mortality. Screening tumors for absent mismatch repair (MMR) protein expression by immunohistochemistry (IHC) is a recommended approach. Despite guidelines advocating universal screening, significant variation in clinical practice exists.
T. D. Kidambi (&) Department of Internal Medicine, University of California, San Francisco, 1701 Divisadero, Suite 120, San Francisco, CA 94115, USA e-mail: [email protected] T. D. Kidambi 505 Parnassus Avenue, Room 987, San Francisco, CA 94143, USA A. Blanco M. Myers P. Conrad K. Loranger Gastrointestinal Cancer Prevention Program, University of California, San Francisco, 1701 Divisadero, Suite 120, San Francisco, CA 94115, USA e-mail: [email protected] M. Myers e-mail: [email protected] P. Conrad e-mail: [email protected]
Aims/Methods A retrospective study of two different IHC-based Lynch syndrome screening protocols at an urban, university hospital was performed. Outcomes from a ‘‘selective’’ screening strategy utilized from August 2007– July 2010 on CRC tumors from patients with high-risk features were compared with a ‘‘universal’’ strategy of screening all CRC tumors from July 2010–August 2013. Positively screened patients were referred for genetic counseling and offered germline testing. Results A total of 392 patients with CRC were screened: 107 selectively and 285 universally. The prevalence of Lynch syndrome was 3.1 %, with no difference by strategy. There was a trend (p = 0.06) toward fewer universally screened patients agreeing to genetic counseling compared with those selectively screened. Selective criteria failed to identify one of eight cases of Lynch syndrome from the universal group, though the universal strategy screened 166 additional tumors to find this additional patient. Conclusions Selective screening for Lynch syndrome has similar outcomes as universal screening in terms of identifying Lynch syndrome, despite screening far fewer patients. In addition, fewer eligible patients in our study agreed to undergo genetic counseling and germline testing than in prior studies. These lower rates may better reflect uptake of these services in clinical practice. Keywords Lynch syndrome Cancer screening Immunohistochemistry Hereditary colon cancer
K. Loranger e-mail: [email protected]
Introduction J. P. Terdiman Division of Gastroenterology, University of California, San Francisco, 1701 Divisadero, Suite 120, San Francisco, CA 94115, USA e-mail: [email protected]
Lynch syndrome (hereditary non-polyposis colorectal cancer), an autosomal dominant disease caused by mutations in DNA mismatch repair (MMR) genes MLH1,
123
Dig Dis Sci
MSH2, MSH6, PMS2, or EPCAM, is the most common hereditary cause of colorectal cancer (CRC) and confers an increased risk of extra-colonic cancers including endometrial, ovarian, renal, gastric, pancreatic, skin, and brain [1– 4]. In large studies, Lynch syndrome accounts for 2–7 % of all CRC cases [5–10]. Identifying patients and their family members with Lynch syndrome has important implications as intensive cancer surveillance and prophylactic gynecological surgery have been shown to improve both cancerrelated and all-cause mortality [11–13]. As a result, there is substantial interest in screening for Lynch syndrome. A recent practice guideline provided a detailed description of screening methods for Lynch syndrome [14]. In brief, two major methods have been used and validated for screening: (1) testing the tumor specimen for microsatellite instability (MSI), which is seen in most cases of Lynch syndrome or (2) performing immunohistochemistry (IHC) staining to identify absent MMR protein expression [1, 3–10, 15–18]. Both MSI- and IHC-based screening methods have similar sensitivity that is upwards of 90 % [5, 17, 18], but an IHC-based approach is less expensive, readily available, and has the additional benefit of identifying the potential target gene for the confirmatory, germline genetic testing [9, 17–19]. The Revised Bethesda Guidelines [1] were designed to identify CRC tumors for further MSI testing. Unfortunately, recent studies have shown that the Revised Bethesda Guidelines have a sensitivity of 70–85 % in unselected CRC patients [10, 17, 20], which, while higher than previous estimates of \50 % [3], is still felt to be inadequate for screening purposes. As a result, studies have addressed screening high-risk, younger CRC patients [16, 21] as well as universally screening all CRC patients [5, 6, 8–10, 17, 20] for Lynch syndrome and found that these approaches are feasible. However, studies of universal screening have largely been performed in a research setting [5, 17] or in countries with a national health care system [6, 8–10, 20], which may limit their generalizability to other United States (US)-based practices. Furthermore, the results of decision analysis [19], cost-effectiveness analysis [22], and systematic review guidelines [23, 24] call for universal screening of CRC tumors for Lynch syndrome with either an IHC-based approach or MSI-based approach. Universal screening has not been widely adopted by community hospitals, and there is also no standard protocol for selection of patients to screen utilized among US National Cancer Institute Cancer Centers [25, 26]. At the University of California, San Francisco (UCSF) medical center, screening of CRC tumors for Lynch syndrome with an IHC-based protocol began in 2007. Initially, only CRC patients meeting modified Revised Bethesda Guidelines were ‘‘selectively’’ screened, but beginning in 2010, all CRC patients were universally screened. The
123
aims of this study were to (1) report the 6-year clinical outcomes of Lynch syndrome screening at UCSF and (2) compare the results of the two screening strategies. Our hypothesis was that the selective strategy would have missed a significant number of cases of Lynch syndrome that were identified by universal screening.
Methods A retrospective study examining outcomes of Lynch syndrome screening from August 2007 to August 2013 at the UCSF medical center, a large urban university hospital, was performed. Between August 2007 and July 2010, a ‘‘selective’’ strategy of screening tumors only from patients with surgically resected CRC at UCSF who were under the age of 50 years, between the age of 50–60 years with histology suggestive of MSI (tumor infiltrating lymphocytes, Crohn’s-like lymphocyte reaction, mucinous/signet ring differentiation, or medullary growth pattern), or of any age with prior history of Lynch syndrome-related cancer (small bowel, endometrial, ovarian, gastric, pancreas, biliary tract, urinary tract, sebaceous neoplasms, keratoacanthoma, and brain) or with synchronous or metachronous CRC was utilized. These selective criteria were similar to the Revised Bethesda Guidelines [1] except that family history criteria were not included. Beginning in August 2010, a ‘‘universal’’ strategy of screening all tumors from patients with surgically resected CRC at UCSF for Lynch syndrome, regardless of age, tumor histology, or personal or family cancer history, was adopted. In both strategies, IHC screening was initiated by the pathology department based on agreement between clinicians and the pathologists at our institution. All patients with surgically resected CRC at UCSF between August 2007 and August 2013 were included; CRC tumors referred to our pathology department for Lynch syndrome screening from other hospitals and community practices were excluded from our analysis. Informed consent from patients to screen their tumors for Lynch syndrome was not required to be obtained per institution protocol. CRC tumors were screened by pathologists with IHC staining of four MMR proteins (MLH1, MSH2, MSH6, and PMS2). An IHC screen was considered positive for Lynch syndrome if there was absent IHC staining of any of the four MMR proteins prior to February 2009. Beginning in February 2009, tumors with absent IHC staining of MLH1 were tested for BRAF V600E mutations to exclude sporadic CRC with acquired promoter hypermethylation [9, 17, 19, 22, 23, 27]. Thus, from February 2009 onwards, a patient was considered to have a positive screen for Lynch syndrome if their tumor had absent IHC staining for MSH2, MSH6, or PMS2, or if their tumor had absent IHC
Dig Dis Sci
staining for MLH1 and a negative BRAF V600E mutation result. Abnormal IHC results with accompanying BRAF V600E mutation analysis, when performed, were automatically forwarded to both the attending surgeon as well as the genetic counselors in the Hereditary Gastrointestinal Cancer Prevention Program at UCSF. At the request of the surgeons, the genetic counselors took responsibility for follow-up of the abnormal results. Prior to contacting positively screened patients, the genetic counselors contacted the attending surgeons to make sure they were aware of the abnormal results and to request that they notify their patients of the resulting automatic referral to genetics. The genetics counselors then contacted patients via telephone to invite them for formal genetic counseling. Patients could accept or decline genetic counseling. Patients who accepted genetic counseling were counseled and offered germline testing after informed consent was obtained. In the state of California, most private insurance companies cover both genetic counseling and germline testing; Medicare covers genetic testing but not genetic counseling while Medi-Cal (California’s Medicaid welfare program) covers both genetic testing and counseling but not all labs accept Medicaid insurance. Data were gathered retrospectively from the medical record, which included pathology reports and surgical and genetic clinic notes. Demographic data such as age at diagnosis, gender, personal or family history of cancers in addition to the results of IHC screening, BRAF V600E testing, and Lynch syndrome prevalence were collected. The diagnosis of Lynch syndrome was defined by a positive germline mutation test. Of the positively screened patients, the percentage that agreed to genetic counseling and germline testing was also collected. These outcomes were analyzed according to screening strategy—selective versus universal. Statistical analyses comparing differences according to screening strategy was performed using v2 (categorical variables) or unpaired t tests (continuous variables); Fisher’s exact test was used for categorical variables with outcomes less than five. Results were displayed as either mean (standard deviation) or percentage of either the total number of patients screened by strategy or percentage of positively screened patients by strategy. Statistical significance was defined as p \ 0.05 and analyses were performed using SPSS statistical software (version 21, Chicago, IL).
Results A total of 392 patients with CRC were screened for Lynch syndrome between August 2007 and August 2013, 107 by
the selective strategy and 285 by the universal strategy. A total of 39 patients (9.9 %) had a positive IHC screen, and the overall prevalence of Lynch syndrome in the study was 3.1 %. Of the 285 patients screened by the universal strategy, 119 patients (42 %) would have met the selective screening criteria and 166 would not have been screened. Table 1 displays patient characteristics and the results of screening according to strategy. As expected, patients screened by the selective strategy were, on average, 10 years younger than patients screened by the universal strategy (mean age 49.4 vs. 59.3 years, respectively, p \ 0.01). Neither strategy had a higher rate of IHC positive screening (p = 0.10) or Lynch syndrome prevalence (p = 0.63). There were no differences in the IHC staining patterns of the positively screened patients according to screening strategy. Of the positively screened patients by IHC, there was no difference in age (p = 0.79) by screening strategy. A greater percentage of the positively screened patients by the selective strategy (80 %) had a personal or family history of Lynch syndrome cancers than in the universally screened group (50.4 %). Genetic counselors were successful in contacting patients in both groups (p = 0.25), but there was a trend toward fewer patients in the universally screened group agreeing to receive genetic counseling (p = 0.06). However, there was no difference in the percentage of patients that agreed to have germline testing for Lynch syndrome based on screening strategy after receiving genetic counseling (p = 0.24). Positively screened patients in the universal group who were successfully contacted and refused genetic counseling (N = 8) were older (mean age 61.4 years, SD 11.4 years) than those who accepted counseling (N = 14, mean 49.0 years, SD 15 years), though this difference was not statistically significant (p = 0.06). There were no differences in gender (p = 0.40), family history of cancer (p = 0.86), or personal history of cancer (p = 0.86) between the universally screened patients who agreed to genetic counseling and those who refused. There were differences in the insurance status of positively screened patients in the two screening groups. Of the positively screened patients by the selective strategy, 73.3 % had private/employer-based health insurance, 20 % had Medicare and Medi-Cal, and 6.7 % had unknown insurance. Of the positively screened patients by the universal strategy, 37.5 % had private/employer-based health insurance, 25 % had Medicare, 12.5 % had MediCal, and 25 % had unknown insurance. Furthermore, among the 10 positively screened patients in the universal group who did not agree to genetic counseling, 4 had Medicare, 2 had private insurance, and 4 had unknown insurance.
123
Dig Dis Sci Table 1 Outcomes of Lynch syndrome screening according to screening strategy Selective N = 107
Universal N = 285
p value \0.01
Mean age in years (SD)
49.4 (15.0)
59.3 (13.9)
Male
42 (39 %)
143 (50 %)
0.05
Family or personal history Lynch tumors
39 (36.5 %)
65 (22.8 %)
\0.01
Pathological features suggestive of MSI
43 (40.2 %)
98 (34.4 %)
0.29
Positive IHC screen
15 (14 %)
24 (8.4 %)
0.10
BRAF ? (% total MLH1 absent)
5 (33.3 %)
12 (48 %)
0.36
Lynch syndrome
4 (3.7 %)
8 (2.8 %)
0.63
Characteristics of IHC positively screened patients
N = 15
N = 24
Mean age in yrs (SD)
52.7 (17.3)
54.0 (14.8)
Male
6 (40 %)
10 (41.7 %)
0.92
IHC ? and would meet selective screening criteria
15 (100 %)
14 (58.3 %)
\0.01
Family or personal history Lynch tumors
12 (80 %)
10 (50.4 %)
0.02
0.79
Pathological features suggestive of MSI
7 (46.7 %)
14 (58.3 %)
0.48
MLH1-BRAF (%N)
10 (66.7 %)
13 (54.2 %)
0.44
MSH2 (%N)
1 (6.7 %)
8 (33.3 %)
0.05
MSH6 (%N)
6 (40.0 %)
8 (33.3 %)
0.67
PMS2 (%N)
8 (53.3 %)
14 (58.3 %)
0.76
Contacted by genetic counselors
15 (100 %)
22 (91.7 %)
0.25
Agreed to receive genetic counseling
13 (86.7 %)
14 (58.3 %)
0.06
Underwent genetic testing (% received genetic counseling)
10 (77.0 %)
13 (92.9 %)
0.24
Unless otherwise noted, listed as number of cases (%total cases screened by strategy) or (%IHC positively screened by strategy) MSI microsatellite instability, IHC immunohistochemistry
The characteristics of the 12 identified probands with Lynch syndrome are shown in Table 2. The average age was 44.9 years with a standard deviation 14.9 years. Of the 8 cases of Lynch syndrome detected by universal screening, only one patient (the final patient in Table 2) would not have met selective screening criteria.
Discussion This is the first study to directly compare outcomes of two different Lynch syndrome screening strategies in clinical practice. The primary finding of our study is that a selective strategy of only screening high-risk individuals was similar to a universal screening strategy in terms of identifying CRC patients with Lynch syndrome. Furthermore, the data from this study were generated as part of routine clinical practice at our hospital, so the results likely have broad generalizability to other clinical centers considering instituting Lynch syndrome screening. The 3.1 % prevalence of Lynch syndrome in this study was consistent with the results of the largest pooled screening study [10]. Of note, the prevalence in our study included both universally screened and higher risk, selectively screened patients. This similarity in
123
prevalence to other studies was observed despite the fact that not all participants in our study consented to and provided blood samples for germline testing upon inclusion in the screening arm of the protocol as in prior studies. Therefore, it is possible that the true prevalence of Lynch syndrome in our study was higher than observed since 16 of the 39 patients with a positive IHC screen did not undergo germline testing and were therefore considered as negative in the prevalence calculation. Additionally, IHC screening identified 9.9 % of patients as candidates for germline testing in this study consistent with estimates of 7–16 % in prior universal screening studies [5, 9, 10, 17, 20, 28]. Though the universal screening strategy only identified one case of Lynch syndrome that would have been missed by the selective strategy, many more CRC specimens were screened as part of the former approach. In fact, the universal strategy screened 166 more tumors as compared to the selective strategy to find only one additional Lynch syndrome patient. Our central finding is consistent with data from the largest analysis of universal CRC screening, which aggregated data from two large European studies and a consortium of US and international research-protocol database. The study found that use of the Revised Bethesda guidelines to determine who should be screened for Lynch
Dig Dis Sci Table 2 Characteristics of Lynch syndrome cases Age at diagnosis
Gender
Screening method
Family History of Lynch cancer
Personal history of Lynch cancer
Histology with MSI features
Lynch gene
67
Male
Selective
Yes
Yes
No
MSH2, del ex9 ex 12
58
Male
Selective
No
No
Yes
MSH6
37
Female
Selective
Yes
No
No
MLH1
28
Female
Selective
Yes
Yes
Yes
MLH1
42
Female
Universal
Yes
Yes
Yes
MSH2, IVS6-1G [ C
36
Female
Universal
No
No
Yes
MLH1, 1105dupT
19 64
Male Female
Universal Universal
No Yes
No Yes
Yes Yes
MSH2, del ex 3–16 MLH1
43
Female
Universal
No
Yes
Yes
MSH2, c1321dupA
35
Male
Universal
Yes
No
Yes
MSH2, E786 (2356G [ T)
52
Female
Universal
No
Yes
Yes
MSH2, 1285delC
57
Male
Universal
No
No
No
MLH1, del ex 3-6
syndrome would miss \5 % of Lynch syndrome cases as compared to universal screening, while screening 35 % fewer patients [10]. Though the screening criteria used in our study were slightly different from the aforementioned study, the conclusion in our clinic-based study is the same. Universal screening detects few, if any, additional Lynch syndrome patients, but entails a considerable increase in resource use. Perhaps the most interesting finding in this study was that so few patients, \70 %, who screened positive for Lynch syndrome, agreed to genetic counseling, and therefore, only 59 % actually underwent germline testing. This was far lower than the 90 % previously reported in a populationbased US study [5, 17], but similar to the 71 % of patients in the Cleveland Clinic study that received counseling and 66 % of patients who underwent germline testing [28]. Furthermore, a trend suggesting that selectively screened patients (87 %) were more likely than universally screened patients (58 %) to agree to genetic counseling was observed (p = 0.06). Of note, once patients agreed to and received genetic counseling, 85 % underwent germline testing, and therefore, the barrier appeared to be in convincing patients to undergo genetic counseling. Within the universal group, there were no differences in the limited demographic information collected between the patients who agreed to counseling and those who refused. Given the retrospective nature of this study, we were unable to collect further data to help explain the differences observed, but prior studies have found that insurance, concerns about discrimination, cost, psychological stress, and the perception that germline testing would not benefit the patient or patient’s family all affect willingness to undergo testing [28–30]. The implication of such a low uptake of genetic counseling and germline testing is that the assumptions of prior cost-effectiveness analysis supporting universal screening may need to be re-examined.
For example, the sensitivity analysis for percentage of patients undergoing germline testing in a recent cost-effectiveness analysis [22] utilized a lower estimate of 88 %, which is significantly higher than what was observed in our study. Our results therefore suggest that further large-scale studies of Lynch syndrome screening in non-research-based clinical settings may help inform future cost-effectiveness analyses and guideline recommendations. Several aspects of our screening strategies deserve mention. First, our selective screening strategy used components of the Revised Bethesda Guidelines, but did not include family history, which may have theoretically lowered its sensitivity. Family history data were not used because it was not easily available to our pathologists. In fact, a recent study of patients with metastatic colon cancer found that only 80 % had family history documented on chart review limiting its use when trying to apply Revised Bethesda Guidelines [31]. Second, patients did not undergo informed consent prior to screening of the CRC specimens for absent MMR protein expression by IHC; the ethics of this approach in Lynch syndrome have been discussed previously [32, 33], and surveys of genetics counselors show that informed consent for screening is rarely obtained,\10 % of the time, in clinical practice [25]. Third, a referral to genetic counseling was made automatically by the surgeons when a tumor screened positive and the onus of following up with patients was placed on the genetic counselors, which is consistent with the protocol used at the Cleveland Clinic [28], but different from an Australian study of universal screening [9]. Automatic referrals with genetic counselors taking on the responsibility for follow-up increases the likelihood patients will accept genetic counseling and germline testing [28]. Our study differed from the Cleveland Clinic study in that we report outcomes of different Lynch syndrome screening strategies rather than methods to follow up on abnormal results [28],
123
Dig Dis Sci
and, to our knowledge, this is the first such study to report these outcomes in a ‘‘real-world’’ US-based clinical practice population. This study is timely as many cancer centers in the US struggle with how best to implement Lynch syndrome screening in light of recent guidelines [23, 24] supporting universal screening, and analytic modeling confirms that universal screening can be cost-effective, especially when a large number of relatives of the proband agree to undergo testing themselves [22, 34]. It is well accepted that the identification of additional Lynch syndrome cases can be far reaching as uncovering a patient can potentially identify kindred requiring additional testing and surveillance. However, the results of our study call into question the incremental benefit of universal screening as compared with a more selective, Bethesda-guideline-based approach. The major limitation of our study is the small sample size. However, the differences between the numbers of Lynch syndrome patients identified are still meaningful and interpretable in this study. Additionally, the retrospective nature of the study limited our ability to gather important data on the important subset of patients who refused genetic counseling and germline testing. Furthermore, clinical practice and cancer genetics may have changed during the 6-year period of our study, which was another potential source of bias in this retrospective study. Though the study was set in a singleurban tertiary care center, the fact that the outcomes of IHC screening and Lynch syndrome prevalence were similar to prior larger, multicenter studies suggested that our population was similar to those studied previously. In conclusion, this study confirms that screening for Lynch syndrome is feasible in clinical practice, but requires a multidisciplinary approach involving physicians, pathologists, and genetic counselors. Universal screening did identify a single case of Lynch syndrome that would have been missed by a selective screening approach but at the cost of screening many more patients. There were substantially less patients agreeing to genetic counseling and germline testing in this study than in prior studies, but our finding in this regard may be a more accurate reflection of what occurs in actual clinical practice. In the future, larger, multicenter clinical studies should continue to investigate the optimal screening strategy and identify factors that decrease patients’ willingness to accept genetic counseling and undergo germline testing in clinical practice. Conflict of interest
None.
References 1. Umar A, Boland CR, Terdiman JP, et al. Revised bethesda guidelines for hereditary nonpolyposis colorectal cancer (lynch
123
2.
3. 4.
5.
6.
7.
8.
9.
10. 11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
syndrome) and microsatellite instability. J Natl Cancer Inst. 2004;96:261–268. Vasen HF, Watson P, Mecklin JP, et al. New clinical criteria for hereditary nonpolyposis colorectal cancer (hnpcc, lynch syndrome) proposed by the international collaborative group on hnpcc. Gastroenterology. 1999;116:1453–1456. Lynch HT, de la Chapelle A. Hereditary colorectal cancer. N Engl J Med. 2003;348:919–932. Bonadona V, Bonaiti B, Olschwang S, et al. Cancer risks associated with germline mutations in mlh1, msh2, and msh6 genes in lynch syndrome. JAMA. 2011;305:2304–2310. Hampel H, Frankel WL, Martin E, et al. Screening for the lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med. 2005;352:1851–1860. Aaltonen LA, Salovaara R, Kristo P, et al. Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease. N Engl J Med. 1998;338:1481–1487. Barnetson RA, Tenesa A, Farrington SM, et al. Identification and survival of carriers of mutations in DNA mismatch-repair genes in colon cancer. N Engl J Med. 2006;354:2751–2763. Salovaara R, Loukola A, Kristo P, et al. Population-based molecular detection of hereditary nonpolyposis colorectal cancer. J Clin Oncol. 2000;18:2193–2200. Ward RL, Hicks S, Hawkins NJ. Population-based molecular screening for lynch syndrome: implications for personalized medicine. J Clin Oncol. 2013;31:2554–2562. Moreira L, Balaguer F, Lindor N, et al. Identification of lynch syndrome among patients with colorectal cancer. JAMA. 2012;308:1555–1565. Jarvinen HJ, Aarnio M, Mustonen H, et al. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology. 2000;118:829–834. Schmeler KM, Lynch HT, Chen LM, et al. Prophylactic surgery to reduce the risk of gynecologic cancers in the lynch syndrome. N Engl J Med. 2006;354:261–269. Lindor NM, Petersen GM, Hadley DW, et al. Recommendations for the care of individuals with an inherited predisposition to lynch syndrome: a systematic review. JAMA. 2006;296:1507– 1517. Gibson J, Lacy J, Matloff E, et al. Microsatellite instability testing in colorectal carcinoma: a practical guide. Clin Gastroenterol Hepatol. 2014;12:e171. Aaltonen LA, Peltomaki P, Mecklin JP, et al. Replication errors in benign and malignant tumors from hereditary nonpolyposis colorectal cancer patients. Cancer Res. 1994;54:1645–1648. Terdiman JP, Gum JR Jr, Conrad PG, et al. Efficient detection of hereditary nonpolyposis colorectal cancer gene carriers by screening for tumor microsatellite instability before germline genetic testing. Gastroenterology. 2001;120:21–30. Hampel H, Frankel WL, Martin E, et al. Feasibility of screening for lynch syndrome among patients with colorectal cancer. J Clin Oncol. 2008;26:5783–5788. Terdiman JP. It is time to get serious about diagnosing lynch syndrome (hereditary nonpolyposis colorectal cancer with defective DNA mismatch repair) in the general population. Gastroenterology. 2005;129:741–744. Mvundura M, Grosse SD, Hampel H, et al. The cost-effectiveness of genetic testing strategies for lynch syndrome among newly diagnosed patients with colorectal cancer. Genet Med. 2010;12:93–104. Pinol V, Castells A, Andreu M, et al. Accuracy of revised bethesda guidelines, microsatellite instability, and immunohistochemistry for the identification of patients with hereditary nonpolyposis colorectal cancer. JAMA. 2005;293:1986–1994. Schofield L, Watson N, Grieu F, et al. Population-based detection of lynch syndrome in young colorectal cancer patients using microsatellite instability as the initial test. Int J Cancer. 2009; 124:1097–1102.
Dig Dis Sci 22. Ladabaum U, Wang G, Terdiman J, et al. Strategies to identify the lynch syndrome among patients with colorectal cancer: a cost-effectiveness analysis. Ann Int Med. 2011;155:69–79. 23. Evaluation of Genomic Applications in P, Prevention Working G. Recommendations from the egapp working group: genetic testing strategies in newly diagnosed individuals with colorectal cancer aimed at reducing morbidity and mortality from lynch syndrome in relatives. Genet Med. 2009;11:35–41. 24. Burt RW, Cannon JA, David DS, et al. Colorectal cancer screening. J Natl Compr Cancer Netw JNCCN.. 2013;11:1538–1575. 25. Beamer LC, Grant ML, Espenschied CR, et al. Reflex immunohistochemistry and microsatellite instability testing of colorectal tumors for lynch syndrome among us cancer programs and follow-up of abnormal results. J Clin Oncol. 2012;30:1058–1063. 26. Cohen SA. Current lynch syndrome tumor screening practices: a survey of genetic counselors. J Genet Couns. 2014;23:38–47. 27. Bessa X, Balleste B, Andreu M, et al. A prospective, multicenter, population-based study of braf mutational analysis for lynch syndrome screening. Clin Gastroenterol Hepatol. 2008;6:206–214. 28. Heald B, Plesec T, Liu X, Pai R, et al. Implementation of universal microsatellite instability and immunohistochemistry
29.
30.
31.
32.
33.
34.
screening for diagnosing lynch syndrome in a large academic medical center. J Clin Oncol. 2013;31:1336–1340. Gritz ER, Peterson SK, Vernon SW, et al. Psychological impact of genetic testing for hereditary nonpolyposis colorectal cancer. J Clin Oncol. 2005;23:1902–1910. Sharaf RN, Myer P, Stave CD, et al. Uptake of genetic testing by relatives of lynch syndrome probands: a systematic review. Clin Gastroenterol Hepatol. 2013;11:1093–1100. Cross DS, Rahm AK, Kauffman TL, et al. Underutilization of lynch syndrome screening in a multisite study of patients with colorectal cancer. Genet Med. 2013;15:933–940. Chubak B, Heald B, Sharp RR. Informed consent to microsatellite instability and immunohistochemistry screening for lynch syndrome. Genet Med. 2011;13:356–360. Williams JL, Williams MS. Informed consent and immunohistochemistry screening for lynch syndrome. Genet Med. 2011;13: 848–849. Ladabaum U, Ford JM. Lynch syndrome in patients with colorectal cancer: finding the needle in the haystack. JAMA.. 2012; 308:1581–1583.
123
ORIGINAL ARTICLE
Selective Versus Universal Screening for Lynch Syndrome: A Six-Year Clinical Experience Trilokesh D. Kidambi • Amie Blanco • Megan Myers • Peggy Conrad • Kate Loranger Jonathan P. Terdiman
•
Received: 24 April 2014 / Accepted: 26 May 2014 Ó Springer Science+Business Media New York 2014
Abstract Background Lynch syndrome is the most common cause of hereditary colorectal cancer (CRC) and confers increased risk of other cancers. Identification of patients improves morbidity and mortality. Screening tumors for absent mismatch repair (MMR) protein expression by immunohistochemistry (IHC) is a recommended approach. Despite guidelines advocating universal screening, significant variation in clinical practice exists.
T. D. Kidambi (&) Department of Internal Medicine, University of California, San Francisco, 1701 Divisadero, Suite 120, San Francisco, CA 94115, USA e-mail: [email protected] T. D. Kidambi 505 Parnassus Avenue, Room 987, San Francisco, CA 94143, USA A. Blanco M. Myers P. Conrad K. Loranger Gastrointestinal Cancer Prevention Program, University of California, San Francisco, 1701 Divisadero, Suite 120, San Francisco, CA 94115, USA e-mail: [email protected] M. Myers e-mail: [email protected] P. Conrad e-mail: [email protected]
Aims/Methods A retrospective study of two different IHC-based Lynch syndrome screening protocols at an urban, university hospital was performed. Outcomes from a ‘‘selective’’ screening strategy utilized from August 2007– July 2010 on CRC tumors from patients with high-risk features were compared with a ‘‘universal’’ strategy of screening all CRC tumors from July 2010–August 2013. Positively screened patients were referred for genetic counseling and offered germline testing. Results A total of 392 patients with CRC were screened: 107 selectively and 285 universally. The prevalence of Lynch syndrome was 3.1 %, with no difference by strategy. There was a trend (p = 0.06) toward fewer universally screened patients agreeing to genetic counseling compared with those selectively screened. Selective criteria failed to identify one of eight cases of Lynch syndrome from the universal group, though the universal strategy screened 166 additional tumors to find this additional patient. Conclusions Selective screening for Lynch syndrome has similar outcomes as universal screening in terms of identifying Lynch syndrome, despite screening far fewer patients. In addition, fewer eligible patients in our study agreed to undergo genetic counseling and germline testing than in prior studies. These lower rates may better reflect uptake of these services in clinical practice. Keywords Lynch syndrome Cancer screening Immunohistochemistry Hereditary colon cancer
K. Loranger e-mail: [email protected]
Introduction J. P. Terdiman Division of Gastroenterology, University of California, San Francisco, 1701 Divisadero, Suite 120, San Francisco, CA 94115, USA e-mail: [email protected]
Lynch syndrome (hereditary non-polyposis colorectal cancer), an autosomal dominant disease caused by mutations in DNA mismatch repair (MMR) genes MLH1,
123
Dig Dis Sci
MSH2, MSH6, PMS2, or EPCAM, is the most common hereditary cause of colorectal cancer (CRC) and confers an increased risk of extra-colonic cancers including endometrial, ovarian, renal, gastric, pancreatic, skin, and brain [1– 4]. In large studies, Lynch syndrome accounts for 2–7 % of all CRC cases [5–10]. Identifying patients and their family members with Lynch syndrome has important implications as intensive cancer surveillance and prophylactic gynecological surgery have been shown to improve both cancerrelated and all-cause mortality [11–13]. As a result, there is substantial interest in screening for Lynch syndrome. A recent practice guideline provided a detailed description of screening methods for Lynch syndrome [14]. In brief, two major methods have been used and validated for screening: (1) testing the tumor specimen for microsatellite instability (MSI), which is seen in most cases of Lynch syndrome or (2) performing immunohistochemistry (IHC) staining to identify absent MMR protein expression [1, 3–10, 15–18]. Both MSI- and IHC-based screening methods have similar sensitivity that is upwards of 90 % [5, 17, 18], but an IHC-based approach is less expensive, readily available, and has the additional benefit of identifying the potential target gene for the confirmatory, germline genetic testing [9, 17–19]. The Revised Bethesda Guidelines [1] were designed to identify CRC tumors for further MSI testing. Unfortunately, recent studies have shown that the Revised Bethesda Guidelines have a sensitivity of 70–85 % in unselected CRC patients [10, 17, 20], which, while higher than previous estimates of \50 % [3], is still felt to be inadequate for screening purposes. As a result, studies have addressed screening high-risk, younger CRC patients [16, 21] as well as universally screening all CRC patients [5, 6, 8–10, 17, 20] for Lynch syndrome and found that these approaches are feasible. However, studies of universal screening have largely been performed in a research setting [5, 17] or in countries with a national health care system [6, 8–10, 20], which may limit their generalizability to other United States (US)-based practices. Furthermore, the results of decision analysis [19], cost-effectiveness analysis [22], and systematic review guidelines [23, 24] call for universal screening of CRC tumors for Lynch syndrome with either an IHC-based approach or MSI-based approach. Universal screening has not been widely adopted by community hospitals, and there is also no standard protocol for selection of patients to screen utilized among US National Cancer Institute Cancer Centers [25, 26]. At the University of California, San Francisco (UCSF) medical center, screening of CRC tumors for Lynch syndrome with an IHC-based protocol began in 2007. Initially, only CRC patients meeting modified Revised Bethesda Guidelines were ‘‘selectively’’ screened, but beginning in 2010, all CRC patients were universally screened. The
123
aims of this study were to (1) report the 6-year clinical outcomes of Lynch syndrome screening at UCSF and (2) compare the results of the two screening strategies. Our hypothesis was that the selective strategy would have missed a significant number of cases of Lynch syndrome that were identified by universal screening.
Methods A retrospective study examining outcomes of Lynch syndrome screening from August 2007 to August 2013 at the UCSF medical center, a large urban university hospital, was performed. Between August 2007 and July 2010, a ‘‘selective’’ strategy of screening tumors only from patients with surgically resected CRC at UCSF who were under the age of 50 years, between the age of 50–60 years with histology suggestive of MSI (tumor infiltrating lymphocytes, Crohn’s-like lymphocyte reaction, mucinous/signet ring differentiation, or medullary growth pattern), or of any age with prior history of Lynch syndrome-related cancer (small bowel, endometrial, ovarian, gastric, pancreas, biliary tract, urinary tract, sebaceous neoplasms, keratoacanthoma, and brain) or with synchronous or metachronous CRC was utilized. These selective criteria were similar to the Revised Bethesda Guidelines [1] except that family history criteria were not included. Beginning in August 2010, a ‘‘universal’’ strategy of screening all tumors from patients with surgically resected CRC at UCSF for Lynch syndrome, regardless of age, tumor histology, or personal or family cancer history, was adopted. In both strategies, IHC screening was initiated by the pathology department based on agreement between clinicians and the pathologists at our institution. All patients with surgically resected CRC at UCSF between August 2007 and August 2013 were included; CRC tumors referred to our pathology department for Lynch syndrome screening from other hospitals and community practices were excluded from our analysis. Informed consent from patients to screen their tumors for Lynch syndrome was not required to be obtained per institution protocol. CRC tumors were screened by pathologists with IHC staining of four MMR proteins (MLH1, MSH2, MSH6, and PMS2). An IHC screen was considered positive for Lynch syndrome if there was absent IHC staining of any of the four MMR proteins prior to February 2009. Beginning in February 2009, tumors with absent IHC staining of MLH1 were tested for BRAF V600E mutations to exclude sporadic CRC with acquired promoter hypermethylation [9, 17, 19, 22, 23, 27]. Thus, from February 2009 onwards, a patient was considered to have a positive screen for Lynch syndrome if their tumor had absent IHC staining for MSH2, MSH6, or PMS2, or if their tumor had absent IHC
Dig Dis Sci
staining for MLH1 and a negative BRAF V600E mutation result. Abnormal IHC results with accompanying BRAF V600E mutation analysis, when performed, were automatically forwarded to both the attending surgeon as well as the genetic counselors in the Hereditary Gastrointestinal Cancer Prevention Program at UCSF. At the request of the surgeons, the genetic counselors took responsibility for follow-up of the abnormal results. Prior to contacting positively screened patients, the genetic counselors contacted the attending surgeons to make sure they were aware of the abnormal results and to request that they notify their patients of the resulting automatic referral to genetics. The genetics counselors then contacted patients via telephone to invite them for formal genetic counseling. Patients could accept or decline genetic counseling. Patients who accepted genetic counseling were counseled and offered germline testing after informed consent was obtained. In the state of California, most private insurance companies cover both genetic counseling and germline testing; Medicare covers genetic testing but not genetic counseling while Medi-Cal (California’s Medicaid welfare program) covers both genetic testing and counseling but not all labs accept Medicaid insurance. Data were gathered retrospectively from the medical record, which included pathology reports and surgical and genetic clinic notes. Demographic data such as age at diagnosis, gender, personal or family history of cancers in addition to the results of IHC screening, BRAF V600E testing, and Lynch syndrome prevalence were collected. The diagnosis of Lynch syndrome was defined by a positive germline mutation test. Of the positively screened patients, the percentage that agreed to genetic counseling and germline testing was also collected. These outcomes were analyzed according to screening strategy—selective versus universal. Statistical analyses comparing differences according to screening strategy was performed using v2 (categorical variables) or unpaired t tests (continuous variables); Fisher’s exact test was used for categorical variables with outcomes less than five. Results were displayed as either mean (standard deviation) or percentage of either the total number of patients screened by strategy or percentage of positively screened patients by strategy. Statistical significance was defined as p \ 0.05 and analyses were performed using SPSS statistical software (version 21, Chicago, IL).
Results A total of 392 patients with CRC were screened for Lynch syndrome between August 2007 and August 2013, 107 by
the selective strategy and 285 by the universal strategy. A total of 39 patients (9.9 %) had a positive IHC screen, and the overall prevalence of Lynch syndrome in the study was 3.1 %. Of the 285 patients screened by the universal strategy, 119 patients (42 %) would have met the selective screening criteria and 166 would not have been screened. Table 1 displays patient characteristics and the results of screening according to strategy. As expected, patients screened by the selective strategy were, on average, 10 years younger than patients screened by the universal strategy (mean age 49.4 vs. 59.3 years, respectively, p \ 0.01). Neither strategy had a higher rate of IHC positive screening (p = 0.10) or Lynch syndrome prevalence (p = 0.63). There were no differences in the IHC staining patterns of the positively screened patients according to screening strategy. Of the positively screened patients by IHC, there was no difference in age (p = 0.79) by screening strategy. A greater percentage of the positively screened patients by the selective strategy (80 %) had a personal or family history of Lynch syndrome cancers than in the universally screened group (50.4 %). Genetic counselors were successful in contacting patients in both groups (p = 0.25), but there was a trend toward fewer patients in the universally screened group agreeing to receive genetic counseling (p = 0.06). However, there was no difference in the percentage of patients that agreed to have germline testing for Lynch syndrome based on screening strategy after receiving genetic counseling (p = 0.24). Positively screened patients in the universal group who were successfully contacted and refused genetic counseling (N = 8) were older (mean age 61.4 years, SD 11.4 years) than those who accepted counseling (N = 14, mean 49.0 years, SD 15 years), though this difference was not statistically significant (p = 0.06). There were no differences in gender (p = 0.40), family history of cancer (p = 0.86), or personal history of cancer (p = 0.86) between the universally screened patients who agreed to genetic counseling and those who refused. There were differences in the insurance status of positively screened patients in the two screening groups. Of the positively screened patients by the selective strategy, 73.3 % had private/employer-based health insurance, 20 % had Medicare and Medi-Cal, and 6.7 % had unknown insurance. Of the positively screened patients by the universal strategy, 37.5 % had private/employer-based health insurance, 25 % had Medicare, 12.5 % had MediCal, and 25 % had unknown insurance. Furthermore, among the 10 positively screened patients in the universal group who did not agree to genetic counseling, 4 had Medicare, 2 had private insurance, and 4 had unknown insurance.
123
Dig Dis Sci Table 1 Outcomes of Lynch syndrome screening according to screening strategy Selective N = 107
Universal N = 285
p value \0.01
Mean age in years (SD)
49.4 (15.0)
59.3 (13.9)
Male
42 (39 %)
143 (50 %)
0.05
Family or personal history Lynch tumors
39 (36.5 %)
65 (22.8 %)
\0.01
Pathological features suggestive of MSI
43 (40.2 %)
98 (34.4 %)
0.29
Positive IHC screen
15 (14 %)
24 (8.4 %)
0.10
BRAF ? (% total MLH1 absent)
5 (33.3 %)
12 (48 %)
0.36
Lynch syndrome
4 (3.7 %)
8 (2.8 %)
0.63
Characteristics of IHC positively screened patients
N = 15
N = 24
Mean age in yrs (SD)
52.7 (17.3)
54.0 (14.8)
Male
6 (40 %)
10 (41.7 %)
0.92
IHC ? and would meet selective screening criteria
15 (100 %)
14 (58.3 %)
\0.01
Family or personal history Lynch tumors
12 (80 %)
10 (50.4 %)
0.02
0.79
Pathological features suggestive of MSI
7 (46.7 %)
14 (58.3 %)
0.48
MLH1-BRAF (%N)
10 (66.7 %)
13 (54.2 %)
0.44
MSH2 (%N)
1 (6.7 %)
8 (33.3 %)
0.05
MSH6 (%N)
6 (40.0 %)
8 (33.3 %)
0.67
PMS2 (%N)
8 (53.3 %)
14 (58.3 %)
0.76
Contacted by genetic counselors
15 (100 %)
22 (91.7 %)
0.25
Agreed to receive genetic counseling
13 (86.7 %)
14 (58.3 %)
0.06
Underwent genetic testing (% received genetic counseling)
10 (77.0 %)
13 (92.9 %)
0.24
Unless otherwise noted, listed as number of cases (%total cases screened by strategy) or (%IHC positively screened by strategy) MSI microsatellite instability, IHC immunohistochemistry
The characteristics of the 12 identified probands with Lynch syndrome are shown in Table 2. The average age was 44.9 years with a standard deviation 14.9 years. Of the 8 cases of Lynch syndrome detected by universal screening, only one patient (the final patient in Table 2) would not have met selective screening criteria.
Discussion This is the first study to directly compare outcomes of two different Lynch syndrome screening strategies in clinical practice. The primary finding of our study is that a selective strategy of only screening high-risk individuals was similar to a universal screening strategy in terms of identifying CRC patients with Lynch syndrome. Furthermore, the data from this study were generated as part of routine clinical practice at our hospital, so the results likely have broad generalizability to other clinical centers considering instituting Lynch syndrome screening. The 3.1 % prevalence of Lynch syndrome in this study was consistent with the results of the largest pooled screening study [10]. Of note, the prevalence in our study included both universally screened and higher risk, selectively screened patients. This similarity in
123
prevalence to other studies was observed despite the fact that not all participants in our study consented to and provided blood samples for germline testing upon inclusion in the screening arm of the protocol as in prior studies. Therefore, it is possible that the true prevalence of Lynch syndrome in our study was higher than observed since 16 of the 39 patients with a positive IHC screen did not undergo germline testing and were therefore considered as negative in the prevalence calculation. Additionally, IHC screening identified 9.9 % of patients as candidates for germline testing in this study consistent with estimates of 7–16 % in prior universal screening studies [5, 9, 10, 17, 20, 28]. Though the universal screening strategy only identified one case of Lynch syndrome that would have been missed by the selective strategy, many more CRC specimens were screened as part of the former approach. In fact, the universal strategy screened 166 more tumors as compared to the selective strategy to find only one additional Lynch syndrome patient. Our central finding is consistent with data from the largest analysis of universal CRC screening, which aggregated data from two large European studies and a consortium of US and international research-protocol database. The study found that use of the Revised Bethesda guidelines to determine who should be screened for Lynch
Dig Dis Sci Table 2 Characteristics of Lynch syndrome cases Age at diagnosis
Gender
Screening method
Family History of Lynch cancer
Personal history of Lynch cancer
Histology with MSI features
Lynch gene
67
Male
Selective
Yes
Yes
No
MSH2, del ex9 ex 12
58
Male
Selective
No
No
Yes
MSH6
37
Female
Selective
Yes
No
No
MLH1
28
Female
Selective
Yes
Yes
Yes
MLH1
42
Female
Universal
Yes
Yes
Yes
MSH2, IVS6-1G [ C
36
Female
Universal
No
No
Yes
MLH1, 1105dupT
19 64
Male Female
Universal Universal
No Yes
No Yes
Yes Yes
MSH2, del ex 3–16 MLH1
43
Female
Universal
No
Yes
Yes
MSH2, c1321dupA
35
Male
Universal
Yes
No
Yes
MSH2, E786 (2356G [ T)
52
Female
Universal
No
Yes
Yes
MSH2, 1285delC
57
Male
Universal
No
No
No
MLH1, del ex 3-6
syndrome would miss \5 % of Lynch syndrome cases as compared to universal screening, while screening 35 % fewer patients [10]. Though the screening criteria used in our study were slightly different from the aforementioned study, the conclusion in our clinic-based study is the same. Universal screening detects few, if any, additional Lynch syndrome patients, but entails a considerable increase in resource use. Perhaps the most interesting finding in this study was that so few patients, \70 %, who screened positive for Lynch syndrome, agreed to genetic counseling, and therefore, only 59 % actually underwent germline testing. This was far lower than the 90 % previously reported in a populationbased US study [5, 17], but similar to the 71 % of patients in the Cleveland Clinic study that received counseling and 66 % of patients who underwent germline testing [28]. Furthermore, a trend suggesting that selectively screened patients (87 %) were more likely than universally screened patients (58 %) to agree to genetic counseling was observed (p = 0.06). Of note, once patients agreed to and received genetic counseling, 85 % underwent germline testing, and therefore, the barrier appeared to be in convincing patients to undergo genetic counseling. Within the universal group, there were no differences in the limited demographic information collected between the patients who agreed to counseling and those who refused. Given the retrospective nature of this study, we were unable to collect further data to help explain the differences observed, but prior studies have found that insurance, concerns about discrimination, cost, psychological stress, and the perception that germline testing would not benefit the patient or patient’s family all affect willingness to undergo testing [28–30]. The implication of such a low uptake of genetic counseling and germline testing is that the assumptions of prior cost-effectiveness analysis supporting universal screening may need to be re-examined.
For example, the sensitivity analysis for percentage of patients undergoing germline testing in a recent cost-effectiveness analysis [22] utilized a lower estimate of 88 %, which is significantly higher than what was observed in our study. Our results therefore suggest that further large-scale studies of Lynch syndrome screening in non-research-based clinical settings may help inform future cost-effectiveness analyses and guideline recommendations. Several aspects of our screening strategies deserve mention. First, our selective screening strategy used components of the Revised Bethesda Guidelines, but did not include family history, which may have theoretically lowered its sensitivity. Family history data were not used because it was not easily available to our pathologists. In fact, a recent study of patients with metastatic colon cancer found that only 80 % had family history documented on chart review limiting its use when trying to apply Revised Bethesda Guidelines [31]. Second, patients did not undergo informed consent prior to screening of the CRC specimens for absent MMR protein expression by IHC; the ethics of this approach in Lynch syndrome have been discussed previously [32, 33], and surveys of genetics counselors show that informed consent for screening is rarely obtained,\10 % of the time, in clinical practice [25]. Third, a referral to genetic counseling was made automatically by the surgeons when a tumor screened positive and the onus of following up with patients was placed on the genetic counselors, which is consistent with the protocol used at the Cleveland Clinic [28], but different from an Australian study of universal screening [9]. Automatic referrals with genetic counselors taking on the responsibility for follow-up increases the likelihood patients will accept genetic counseling and germline testing [28]. Our study differed from the Cleveland Clinic study in that we report outcomes of different Lynch syndrome screening strategies rather than methods to follow up on abnormal results [28],
123
Dig Dis Sci
and, to our knowledge, this is the first such study to report these outcomes in a ‘‘real-world’’ US-based clinical practice population. This study is timely as many cancer centers in the US struggle with how best to implement Lynch syndrome screening in light of recent guidelines [23, 24] supporting universal screening, and analytic modeling confirms that universal screening can be cost-effective, especially when a large number of relatives of the proband agree to undergo testing themselves [22, 34]. It is well accepted that the identification of additional Lynch syndrome cases can be far reaching as uncovering a patient can potentially identify kindred requiring additional testing and surveillance. However, the results of our study call into question the incremental benefit of universal screening as compared with a more selective, Bethesda-guideline-based approach. The major limitation of our study is the small sample size. However, the differences between the numbers of Lynch syndrome patients identified are still meaningful and interpretable in this study. Additionally, the retrospective nature of the study limited our ability to gather important data on the important subset of patients who refused genetic counseling and germline testing. Furthermore, clinical practice and cancer genetics may have changed during the 6-year period of our study, which was another potential source of bias in this retrospective study. Though the study was set in a singleurban tertiary care center, the fact that the outcomes of IHC screening and Lynch syndrome prevalence were similar to prior larger, multicenter studies suggested that our population was similar to those studied previously. In conclusion, this study confirms that screening for Lynch syndrome is feasible in clinical practice, but requires a multidisciplinary approach involving physicians, pathologists, and genetic counselors. Universal screening did identify a single case of Lynch syndrome that would have been missed by a selective screening approach but at the cost of screening many more patients. There were substantially less patients agreeing to genetic counseling and germline testing in this study than in prior studies, but our finding in this regard may be a more accurate reflection of what occurs in actual clinical practice. In the future, larger, multicenter clinical studies should continue to investigate the optimal screening strategy and identify factors that decrease patients’ willingness to accept genetic counseling and undergo germline testing in clinical practice. Conflict of interest
None.
References 1. Umar A, Boland CR, Terdiman JP, et al. Revised bethesda guidelines for hereditary nonpolyposis colorectal cancer (lynch
123
2.
3. 4.
5.
6.
7.
8.
9.
10. 11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
syndrome) and microsatellite instability. J Natl Cancer Inst. 2004;96:261–268. Vasen HF, Watson P, Mecklin JP, et al. New clinical criteria for hereditary nonpolyposis colorectal cancer (hnpcc, lynch syndrome) proposed by the international collaborative group on hnpcc. Gastroenterology. 1999;116:1453–1456. Lynch HT, de la Chapelle A. Hereditary colorectal cancer. N Engl J Med. 2003;348:919–932. Bonadona V, Bonaiti B, Olschwang S, et al. Cancer risks associated with germline mutations in mlh1, msh2, and msh6 genes in lynch syndrome. JAMA. 2011;305:2304–2310. Hampel H, Frankel WL, Martin E, et al. Screening for the lynch syndrome (hereditary nonpolyposis colorectal cancer). N Engl J Med. 2005;352:1851–1860. Aaltonen LA, Salovaara R, Kristo P, et al. Incidence of hereditary nonpolyposis colorectal cancer and the feasibility of molecular screening for the disease. N Engl J Med. 1998;338:1481–1487. Barnetson RA, Tenesa A, Farrington SM, et al. Identification and survival of carriers of mutations in DNA mismatch-repair genes in colon cancer. N Engl J Med. 2006;354:2751–2763. Salovaara R, Loukola A, Kristo P, et al. Population-based molecular detection of hereditary nonpolyposis colorectal cancer. J Clin Oncol. 2000;18:2193–2200. Ward RL, Hicks S, Hawkins NJ. Population-based molecular screening for lynch syndrome: implications for personalized medicine. J Clin Oncol. 2013;31:2554–2562. Moreira L, Balaguer F, Lindor N, et al. Identification of lynch syndrome among patients with colorectal cancer. JAMA. 2012;308:1555–1565. Jarvinen HJ, Aarnio M, Mustonen H, et al. Controlled 15-year trial on screening for colorectal cancer in families with hereditary nonpolyposis colorectal cancer. Gastroenterology. 2000;118:829–834. Schmeler KM, Lynch HT, Chen LM, et al. Prophylactic surgery to reduce the risk of gynecologic cancers in the lynch syndrome. N Engl J Med. 2006;354:261–269. Lindor NM, Petersen GM, Hadley DW, et al. Recommendations for the care of individuals with an inherited predisposition to lynch syndrome: a systematic review. JAMA. 2006;296:1507– 1517. Gibson J, Lacy J, Matloff E, et al. Microsatellite instability testing in colorectal carcinoma: a practical guide. Clin Gastroenterol Hepatol. 2014;12:e171. Aaltonen LA, Peltomaki P, Mecklin JP, et al. Replication errors in benign and malignant tumors from hereditary nonpolyposis colorectal cancer patients. Cancer Res. 1994;54:1645–1648. Terdiman JP, Gum JR Jr, Conrad PG, et al. Efficient detection of hereditary nonpolyposis colorectal cancer gene carriers by screening for tumor microsatellite instability before germline genetic testing. Gastroenterology. 2001;120:21–30. Hampel H, Frankel WL, Martin E, et al. Feasibility of screening for lynch syndrome among patients with colorectal cancer. J Clin Oncol. 2008;26:5783–5788. Terdiman JP. It is time to get serious about diagnosing lynch syndrome (hereditary nonpolyposis colorectal cancer with defective DNA mismatch repair) in the general population. Gastroenterology. 2005;129:741–744. Mvundura M, Grosse SD, Hampel H, et al. The cost-effectiveness of genetic testing strategies for lynch syndrome among newly diagnosed patients with colorectal cancer. Genet Med. 2010;12:93–104. Pinol V, Castells A, Andreu M, et al. Accuracy of revised bethesda guidelines, microsatellite instability, and immunohistochemistry for the identification of patients with hereditary nonpolyposis colorectal cancer. JAMA. 2005;293:1986–1994. Schofield L, Watson N, Grieu F, et al. Population-based detection of lynch syndrome in young colorectal cancer patients using microsatellite instability as the initial test. Int J Cancer. 2009; 124:1097–1102.
Dig Dis Sci 22. Ladabaum U, Wang G, Terdiman J, et al. Strategies to identify the lynch syndrome among patients with colorectal cancer: a cost-effectiveness analysis. Ann Int Med. 2011;155:69–79. 23. Evaluation of Genomic Applications in P, Prevention Working G. Recommendations from the egapp working group: genetic testing strategies in newly diagnosed individuals with colorectal cancer aimed at reducing morbidity and mortality from lynch syndrome in relatives. Genet Med. 2009;11:35–41. 24. Burt RW, Cannon JA, David DS, et al. Colorectal cancer screening. J Natl Compr Cancer Netw JNCCN.. 2013;11:1538–1575. 25. Beamer LC, Grant ML, Espenschied CR, et al. Reflex immunohistochemistry and microsatellite instability testing of colorectal tumors for lynch syndrome among us cancer programs and follow-up of abnormal results. J Clin Oncol. 2012;30:1058–1063. 26. Cohen SA. Current lynch syndrome tumor screening practices: a survey of genetic counselors. J Genet Couns. 2014;23:38–47. 27. Bessa X, Balleste B, Andreu M, et al. A prospective, multicenter, population-based study of braf mutational analysis for lynch syndrome screening. Clin Gastroenterol Hepatol. 2008;6:206–214. 28. Heald B, Plesec T, Liu X, Pai R, et al. Implementation of universal microsatellite instability and immunohistochemistry
29.
30.
31.
32.
33.
34.
screening for diagnosing lynch syndrome in a large academic medical center. J Clin Oncol. 2013;31:1336–1340. Gritz ER, Peterson SK, Vernon SW, et al. Psychological impact of genetic testing for hereditary nonpolyposis colorectal cancer. J Clin Oncol. 2005;23:1902–1910. Sharaf RN, Myer P, Stave CD, et al. Uptake of genetic testing by relatives of lynch syndrome probands: a systematic review. Clin Gastroenterol Hepatol. 2013;11:1093–1100. Cross DS, Rahm AK, Kauffman TL, et al. Underutilization of lynch syndrome screening in a multisite study of patients with colorectal cancer. Genet Med. 2013;15:933–940. Chubak B, Heald B, Sharp RR. Informed consent to microsatellite instability and immunohistochemistry screening for lynch syndrome. Genet Med. 2011;13:356–360. Williams JL, Williams MS. Informed consent and immunohistochemistry screening for lynch syndrome. Genet Med. 2011;13: 848–849. Ladabaum U, Ford JM. Lynch syndrome in patients with colorectal cancer: finding the needle in the haystack. JAMA.. 2012; 308:1581–1583.
123
