Colorectal cancer (CRC) is a rare disease in the pediatric population usually found at an advanced stage with an ominous outcome. This is in contrast to the adult population in which CRC is the third most common cancer for both men and women, with up to 30% thought to have a hereditary basis (1,2). When present in children or adolescents, CRC suggests an underlying genetic predisposition. The most common hereditary CRC susceptibility condition, Lynch syndrome (LS), previously known as hereditary nonpolyposis colorectal cancer (HNPCC), accounts for up to 3% of all CRCs and has traditionally been thought to be a disease affecting only adults (3); however, in some families, CRC is being identified at an earlier age in successive generations of Lynch kindreds, a genetic event known as anticipation, which may be contributing to the early predisposition of CRC in at-risk children and adolescents (4). Pediatric gastroenterologists are being asked to play a greater role in the care and management of children and families affected by LS through referrals of genotype-positive asymptomatic children and adolescents needing surveillance. This review of LS focuses on the molecular genetics, diagnosis, and management of this autosomal dominant disease in the pediatric population and also reviews the recently identified severe biallelic form of LS, known as constitutional mismatch repair deficiency (CMMR-D).
PROTECTION OF THE GENOME
Damage to the host genome is a constant event, and all prokaryotic and eukaryotic cells have elaborate maintenance systems to control and repair DNA damage, thereby preserving genomic integrity. Multiple repair pathways are present that each focus on a specific type of DNA damage and interact with various checkpoint and signal transduction systems to regulate the repair process during replication, transcription, and other mitotic events (5). Single-nucleotide base pair mismatches that are generated during S-phase DNA replication are repaired through the mismatch repair (MMR) system, as illustrated in Figure 1, which identifies and repairs the base pair mismatch while temporarily suspending the cell cycle, thus maintaining DNA fidelity in the dividing cell (6). The MMR system begins with the paired proteins MSH2 and MSH6, which function as a heterodimer to recognize and bind to single-base mismatched pairs or insertion-deletion loops in newly synthesized DNA. A second pair of MMR proteins, MLH1and PMS2, joins the complex and recruits an endonuclease to excise the defective daughter strand, followed by recruitment of a DNA polymerase to generate a new correct complementary daughter strand. Different combinations of MMR proteins can interact with DNA to repair other types of mutations. Inactivation of the MMR system increases the number of spontaneous single base-pair mismatches with point mutations and insertion-deletion loops. These DNA errors result in frameshift and downstream nonsense mutations that when expressed result in truncated, nonfunctional proteins. When MMR function is impaired, genes that regulate cell division can become inactivated through mutation or other mechanisms, leading to a loss of control of cell growth and differentiation, which results in the development of cancer. Inherited germline mutations of 1 of 4 predominant MMR genes are the basis of LS and describe a mutation-prone or mutator phenotype (2). Inactivation of the MMR system is also seen in sporadic human cancers through epigenetic silencing of the MMR genes, particularly hypermethylation of the promoter of MLH1(6). A more in-depth review of the MMR system and its role in human disease is available (7).
The concept of familial CRC was first outlined in 1913 by Warthin (8), who described the family of his seamstress with an inherited predisposition to cancers of the colon, stomach, and endometrium. By 1967, 2 distinct forms of hereditary CRC were appreciated: familial adenomatous polyposis (FAP), manifesting with hundreds to thousands of adenomas in the large intestine, and a second form characterized by relatively few colonic adenomas. The latter group was designated as HNPCC by Henry T. Lynch in 1984. Lacking morphologic or laboratory-defining properties, the first set of clinical criteria for HNPCC, known as the Amsterdam I criteria, was developed in 1991 under the direction of Vasen et al (9) by the International Collaborative Group on Hereditary Nonpolyposis Colorectal Cancer, providing diagnostic criteria as described in Table 1. The International Collaborative Group on Hereditary Nonpolyposis Colorectal Cancer revised the criteria in 1999 (Amsterdam II criteria) to include extracolonic carcinomas of the endometrium, ovary, small bowel, stomach, biliary tract, bladder, ureter, and renal pelvis, thus providing an important assessment tool by which a family can be screened for possible additional cancers (10).
With identification of a genetic etiology, the term LS is now favored over HNPCC and applies to individuals or families who carry an identifiable germline mutation in 1 of the 4 predominant MMR genes (11). It is important to note that approximately half of affected families that meet Amsterdam I criteria will be MMR proficient and lack an identifiable germline MMR defect. This group has a somewhat different phenotype with later age of onset than LS and presently carries the name of familial CRC type X (12).
Although the Amsterdam criteria can help to identify individuals and at-risk families who would benefit from genetic analysis, they were considered too strict and of little value when family history was unavailable. It was noted that the tumors from patients with LS showed distinctive features that could help to identify patients who may benefit from additional molecular testing of their tumors. These histologic features and less-strict personal and family history criteria were formally characterized in 1998 as the Bethesda guidelines with later revisions during a National Cancer Institute workshop in 2004. These guidelines describe the personal and family history criteria and histologic characteristics common to LS tumors that, when present, would warrant further genetic testing (10). The Bethesda guidelines are described in Table 1.
Defective MMR prevents correction of mutations throughout the genome, which affects mono (AAAAAAA) and dinucleotide (CACACACA) repeats known as microsatellites. These conserved regions of up to 20 base pairs in length are found throughout the genome and are assigned specific designations. Mutations within a microsatellite alter the repeat size, variably shortening the allele length in tumor DNA when compared with the normal tissue DNA of the patient. The difference in microsatellite size between tumor and neighboring normal tissue can be documented by polymerase chain reaction amplification, as shown in Figure 2. Microsatellite stability or microsatellite instability (MSI) as originally defined by the Bethesda guidelines involves a panel of 5 microsatellite loci, including 2 mononucleotide (BAT25 and BAT26) and 3 dinucleotide repeats (D2S123, D5S346, and D17S250). A panel of 5 mononucleotide repeats is available, commercially simplifying the assessment for MSI (11,13). MSI in ≥40% of the markers (≥2 of the 5 markers) is the accepted criteria for MSI high category, whereas MSI at only 1 marker is known as microsatellite stability or low (14). Inactivation of the MMR system leads to tumor MSI, which can be demonstrated in approximately 90% of LS-associated CRC tumors. If MSI is present in a tumor, mutational defects in the MMR genes are predicted and thus the patient and the tumor warrant further testing; however, not all MSI cancers will be caused by MMR germline genetic defects. As noted above, MSI is observed in 15% of sporadic CRC because of somatic epigenetic silencing of MLH1 by hypermethylation of the promoter region, which can occur as a result of mutation in BRAF. The BRAF gene product plays a role in regulating the MAP kinase/ERK signaling pathway affecting cell division, differentiation, and secretion (14,15). LS-associated tumors never show mutations in BRAF(16).
Most, but not all, MMR gene mutations result in loss of protein expression in the tumor tissue. Immunohistochemistry (IHC) with antibodies to the commonly affected MMR proteins (MLH1, MSH2, MSH6, and PMS2) can identify loss of protein expression in the tumor when compared with normal surrounding tissue, which then can be used in specific gene mutation analysis, as shown in Figure 3(10,13). There are exceptions in which a mutation will alter protein function and not alter immunogenicity (false-negative). Interpretation of MMR protein IHC-staining patterns is based on an understanding of the cellular heterodimerization of the MMR proteins. Unpaired MMR proteins are unstable and rapidly degraded. Gene defects resulting in loss of MLH1 expression will result in the absence of staining of both MLH1 and its heterodimeric mate PMS2; however, mutation of PMS2 only results in the loss of PMS2 because MLH1 protein interacts with, and is stabilized by, other MMR proteins (10,13). This technique is widely available; that, however, requires expertise and experience for appropriate interpretation. Use of both tumor MSI testing and IHC allows for optimal diagnostic sensitivity for LS.
Many institutions have adopted routine IHC for the 4 MMR proteins for all or a subset of CRC and endometrial tumors, and groups have found varying degrees of MMR protein loss (4%–19% in patients younger than 50, 26%–70% in patients older than 50 years), suggesting that a minority of patients with MMR protein loss did not meet Amsterdam II criteria (17). A recently published study of pooled data from 4 large cohorts of >10,000 patients with CRC who underwent prospective tumor IHC and MMR assessment illustrates some of the limitations of the Amsterdam criteria and Bethesda guidelines in determining which patients with CRC undergo further testing for LS. The authors concluded that universal tumor MMR testing in patients with CRC had the greatest sensitivity in identifying patients with LS (18).
LS IN CHILDREN AND ADOLESCENTS
Although CRC is rare, the pediatric literature has documented the occurrence of it as the most common primary malignant solid tumor of the intestinal tract of childhood (19). One large-center review compared young patients with CRC to similar adult patients with CRC and noted that the young patients had a higher incidence of poorly differentiated tumors (21% vs 8%). These cancers were more advanced at presentation as well, and had a lower rate of survival (23% vs 61%) thought to be the result of lack of recognition (20). A retrospective review from the St Jude Children's Research Hospital noted a 0.9% incidence of epithelial neoplasias of the GI tract, encompassing 99 of 10,986 cancer patients evaluated for malignancy from 1964 to 2003 (21). Systemic tumor analysis allowing for the evaluation of LS was not part of this retrospective report. Several series report that 10% of childhood CRCs have predisposing factors (ulcerative colitis, familial polyposis, Turner syndrome) (22).
The association of LS with CRC in the pediatric population was described in 1949, and there have been numerous recent additional reports (23–25). A more recent study of 13 adolescents studied for CRC noted that 6 of 13 tumors had MSI (26). LS CRC has been identified in children as young as 13 years of age; this occurred in a child who was found to carry a germline heterozygous mutation of MLH1, which was common to other family members (27). Recent analysis of patients with CMMR-D resulting from biallelic MMR mutations has also been described (see section below), which points to the presentation of CRC at a much earlier age in this cohort (28). Because microsatellite and IHC assessment is more routinely applied to tumors in the pediatric population, the incidence of CRC and other cancers attributable to mutations in MMR repair genes will likely increase (29). The clinical features of LS and CMMR-D are described in Table 2.
Family History in Identifying Those at Risk
As implied by the Amsterdam criteria, obtaining an accurate and complete family history is the first step in identifying at-risk individuals. The family history may be incomplete, erroneous, or missing. Small family sizes and early noncancer deaths may also limit the information needed to fit a patient neatly into the Amsterdam criteria. Occasions of incomplete penetrance of a mutation may give the impression of a recessive trait or a skipped generation. Knowledge of extracolonic cancers in the spectrum of LS and detailed history of those cancers are also crucial in the development of a full family pedigree.
LS is one of the most common causes of hereditary cancers, with the risk for CRC approaching 70% and endometrial cancer approaching 40% (30). Surprisingly, up to 20% of individuals who carry an MMR mutation will never develop a cancer in their lifetime (27). Although the average age of onset of CRC occurs much earlier than the general population (mean age 45), multiple reports of adolescent presentation of CRC, especially in families with a history, suggest that individuals in the same family may be affected at a younger age. Epigenetic pressures that may be influencing age of presentation remain unclear, but the primary focus is on the LS family and the identification of genotype-positive affected individuals. Early surveillance-monitoring programs aimed at prevention and early detection can reduce the morbidity and mortality of LS (31). Present guidelines for surveillance recommend that genotype-positive patients begin screen 5 years before the earliest presentation of CRC under the age of 20 to 25 years in the family (32). In families with early CRC, this means that screening should begin in early adolescence. The psychological effect of early testing for at-risk children and adolescents must be carefully considered in this age group, especially for those in whom informed consent becomes a primary issue and a hotly debated topic. Both the American College of Medical Genetics (1995) and the American Society of Clinical Oncology (2003) have recommended delaying genetic testing for adult-onset malignancies unless there is a risk of developing a malignancy in childhood. How the pediatric GI clinician approaches the complex dynamics of genetic testing requires insight into the development and progression of CRC disease and the strategies for surveillance programs. Bartuma et al (33) reported on 9 families with LS, and discussed 3 major family perspectives: becoming a risk family, patterns of communication, and influence on family relations. Notably, family experiences tended to influence risk perception and motivation for genetic testing, especially if an early risk of cancer in children was noted. Genetic counseling in these situations provided families with ways to grasp the complex issues associated with testing children. Communication patterns within a family will influence how information is conveyed between generations and can also stress family relations. Preservation of the family's belief and value system (concern for early diagnosis, psychological effect of the unknown) may outweigh the sensitivity to a child's autonomy for genetic testing. In each instance, close counseling for both the child and parent is necessary to ensure their understanding of the social and clinical implications of the results of testing.
The adenoma to carcinoma progression in LS is thought to be accelerated when compared with sporadic CRC, occurring in 2 to 3 years compared with 8 to 10 years (34). This argues for more careful surveillance and adenoma removal with shorter colonoscopy intervals (1–2 years) and initiation of surveillance colonoscopy at an age 5 years before the age at which the youngest affected family member was diagnosed as having CRC. Special imaging techniques, including chromoendoscopy, narrow band imaging, and magnification colonoscopy, can improve detection, particularly of flat smaller adenomas that tend to develop in the right colon (35,36). Survival rates in closely surveyed and compliant LS populations were shown to approach nearly 98% when compared with 12% of those who declined surveillance, with the CRC incidence declining from 27% to 11% (37). Once a CRC is diagnosed, a colectomy with an ileorectal anastomosis or ileopouch anal anastomosis followed by annual screening of the rectal segment can be an option for patients over continuous annual colonoscopies. Organ-specific surveillance and follow-up strategies especially for affected women (transvaginal ultrasounds with endometrial sampling starting at age 30–35 years) are recommended. Some affected female patients may opt for prophylactic hysterectomy following childbearing because gynecologic surveillance methodologies are not as effective as those for CRC. Gastric and small bowel surveillance needs to be monitored in families with a history of these cancers. Adjuvant chemotherapy following primary resection of CRC follows the common chemotherapy guidelines. A review of surveillance guidelines for LS is described in Table 3(47).
It is important to note that genotype–phenotype correlations in LS should be considered in treatment and surveillance plans. Mutations in MSH2 and MLH1 genes account for up to 70% of defects identified in LS families, whereas mutations in MSH6 and PMS2 are found in 14% and 15%, respectively (20,37). Patients with the more frequent MSH2 and MLH1 mutations have early-onset presentations, whereas patients with MSH6 mutations have a later onset of presentation (38). Individuals with MSH2 mutations appear to have a higher primary risk of developing extracolonic tumors in comparison with patients with MLH1 mutations. This suggests that specific phenotypes need to be closely considered in a surveillance program.
Extraintestinal manifestations in LS are well described in the adult population, with the cumulative lifetime risk of endometrial cancer approaching 40%. This cumulative risk appears to be directly related to age (3.7% at age 40, 42.6% at age 80) and may occur 15 years earlier than in the average population (39). Other digestive (gastric, small bowel, biliary) and urologic cancers are seen in addition to melanomas, acute myeloid leukemias, ovarian cancers, basal cell carcinomas, and squamous cell carcinomas in MMR mutation carriers (40); however, these extracolonic tumors tend to present in the third decade of life or older, leaving the yield for pediatric surveillance unclear. The process of surveying for extracolonic manifestations in pediatric patients with LS needs to be carefully weighed against the invasive nature of the screening tool along with the psychological effects the surveillance could have on the pediatric patient; however, in the unique case when 2 MMR genes are inherited, a different paradigm for surveillance needs to be instituted.
CMMR-D syndrome occurs when an individual inherits 2 MMR gene defects. CMMR-D can occur when a different mutation is inherited from each parent (compound heterozygous) or in LS families with consanguinity in which both alleles of 1 MMR gene carry the same mutation (homozygous). This results in a clinically distinct syndrome from LS characterized by features of neurofibromatosis type 1 (NF1) and a high risk of childhood-onset malignancies including brain, lymphomas, and less commonly leukemia. Children with CMMR-D, both unaffected and with previous hematologic or central nervous system malignancies, frequently go on to develop GI cancers, commonly colorectal or small bowel. This feature of multiple primary cancers at either the same or different sites, known as metachronous malignancy, is common to LS as well. The CMMR-D phenotype most commonly includes café au lait macules. Consanguinity with homozygous mutations has been reported in approximately half of the families with biallelic MMR (bMMR) mutations, with the remainder being compound heterozygotes. Interestingly in kindreds with a CMMR-D-affected individual, expected LS-related cancers are rarely seen and the families often do not meet Amsterdam criteria. A high index of suspicion in cases of consanguinity and or a child with features of NF1 would be lifesaving in this context.
The GI phenotype of CMMR-D includes adenomatous polyps of variable numbers and GI cancers of the small and large bowel. For this reason, the differential diagnosis of patients with either early-onset CRC or adenomatous polyps should be broadened to include not only LS, FAP, attenuated FAP, MutYH-associated polyposis but also CMMR-D syndrome. CRC appears to be the most prevalent GI malignancy in the CMMR-D population (41). The mean age at diagnosis of CRC is 16 years, with a range from 5 to 28 years. Patients have been described with multiple primary CRCs at the same time, ranging from 2 to 22 cancers at diagnosis, a finding known as synchronous malignancy, which is also a shared feature of LS. CRC can be the only cancer present at diagnosis, as was reported in 40% of patients with CMMR-D in a recent study. In contrast to LS, which favors CRC development in the right colon, it is uncertain whether there is a site-specific predilection in this patient population. Early-onset small bowel cancer is thought to be rare; however, patients as young as 11 years of age with small bowel cancer have been found to have underlying CMMR-D mutations. Among the patients with bMMR mutations reported as having small bowel cancer, the mean age at diagnosis of cancer was 20 years (range 11–41 years).
Penetrance of adenomatous polyposis in CMMR-D is unknown; however, prevalence is likely high and may reach 100%. Among CMMR-D patients with CRC, >60% had additional colorectal adenomas identified at diagnosis, with reported numbers ranging from 1 to 50. Clinically, it can be difficult to differentiate polyposis observed in attenuated FAP from CMMR-D, with adenoma distribution involving both the small and large bowel.
EXTRAINTESTINAL CANCERS IN CMMR-D SYNDROME
The original CMMR-D mutation phenotype first described included children with features of NF1, brain tumors, lymphomas, and, less commonly, leukemias; however, with additional observation, adenomatous polyposis and GI cancers were added. The extraintestinal cancers associated with LS have also been identified in CMMR-D, including a 15-year-old with previous glioblastoma multiforme developing metachronous cancers, including CRC, small bowel cancer, and later carcinoma of the renal pelvis and upper ureter (42). Two young women with early endometrial cancer were found to have bMMR mutations (43). The first congenital malformation, agenesis of the corpus callosum, has recently been described with greater frequency in patients with CMMR-D than the general population (44). Because it is difficult to determine the true incidence of CMMR-D mutations, they are likely underrecognized among pediatric patients with cancers such as brain tumors. An international collaboration with a mandate to better understand this condition is presently collecting clinical data and tumor tissue on patients with CMMR-D. Because more people are aware of this cancer syndrome, it will be feasible to determine disease prevalence and a better understanding of the disease phenotype.
Genetics and Molecular Analysis
In contrast to LS, the role of MSI in screening for CMMR-D is controversial because MSI status in CMMR-D may be unreliable in brain tumors and hematological malignancies in which only a minority of tumors exhibit high MSI (see Wimmer and Etzler from the supplemental references for complete discussion, http://links.lww.com/MPG/A262). Interestingly, different MSI genotype is observed in different tumor tissues from patients affected by multiple cancers; MSI status may be tumor specific and not mutation related. Taken together, present knowledge suggests that in contrast to LS, MSI should not be used as the initial screening test for CMMR-D.
Most reports suggest that IHC reveals loss of the mutant protein in both normal and malignant cells, which is concordant with the germline mutation in almost all CMMR-D tumors. The high sensitivity and specificity observed using IHC support the role of this assay as an initial diagnostic tool in tissues.
Biallelic mutations have been reported in all 4 genes (MLH1, MSH2, MSH6, and PMS2); however, the distribution of MMR gene mutations among patients with reported CRC is different from the distribution observed in classic LS. Here, the majority of CMMR-D patients with GI cancers have PMS2 mutations, in contrast to LS patients in whom approximately 70% have mutations in MLH1 or MLH2(45). The penetrance of CRC among PMS2 monoallelic LS carriers is lower than that for other MMR genes, which may partially explain the relative decrease in LS-associated cancers found in the CMMR-D kindreds (46). The reasons for this relative lack of LS cancers in these PMS2 families are still unclear; however, many of the MMR mutations in these families are likely less functionally deleterious than most mutations in typical LS families.
Management and Surveillance
Consensus screening and surveillance guidelines for LS carriers aimed at the adult population have been published and are updated based on systematic review of the literature (47). Owing to the recent identification of this disorder, long-term outcome data for patients with CMMR-D have been lacking. Developing surveillance guidelines that address the wide spectrum of malignancies associated with CMMR-D has also added to the complexity of developing evidence-based guidelines for bMMR surveillance, particularly for the pediatric population. Early recognition and diagnosis of CMMR-D in an individual are important because they allow for genetic counseling and family risk assessment while facilitating colonoscopic CRC surveillance for heterozygous LS family members. Surveillance guidelines, as shown in Table 4, were recently published for patients with CMMR-D based on a 10-year observational experience of 1 kindred (28). Four cancers (2 colon, 1 small bowel, and 1 brain) were identified during surveillance while small and asymptomatic and amenable to resection. Multicenter collaboration and implementation of surveillance recommendations will help to determine the prevalence of extraintestinal cancers as well as genotype–phenotype correlations for CMMR-D.
When identified in children with bMMR mutations, hepatic adenomas should be surgically removed as standard treatment for premalignant lesions and GI cancers identified in the CMMR-D population (48). Differentiating premalignant from malignant masses is important in this population so that these lesions are not misdiagnosed as metastatic disease with resultant change in treatment.
As inherited disorders of DNA repair, LS and CMMR-D represent newly appreciated diseases in the pediatric population. Although rare, they are associated with significant cancer risk, morbidity, and mortality. They remain enigmatic diseases that convey multiorgan cancer risk, yet one-fifth of single-mutation carriers will never develop cancer. These disorders should be considered in the setting of colon adenomas as well as with colon adenocarcinomas. Universal screening of all pediatric and adult newly diagnosed CRC cases should be performed where distinctive tumor histology can suggest an MMR deficiency. Special techniques such as MSI assessment and IHC can be of value; however, special attention must be focused when considering CMMR-D where adjacent noncancerous tissue will also show loss of protein staining. Younger successive generations from LS kindreds have started to develop manifestations years earlier than their parents and grandparents, sometimes developing premalignant and malignant lesions in late childhood or early adolescence. Unlike other polyposis syndromes, spontaneous germline mutation of an MMR gene is unusual, with most mutations being inherited, at times from an unaffected genotype-positive parent (49). With the advent of genetic testing, genotype-positive unaffected children and adolescents are being identified in families at risk for early cancer who request and may benefit from surveillance. It will be important to effectively and efficiently implement surveillance so as to appropriately monitor this population while avoiding unnecessary procedures. Multidisciplinary management involving genetic counselors can be of enormous value in educating the family, identifying other mutation carriers, and helping to enhance compliance with surveillance. In contrast to FAP, which is a disease of cancer initiation, LS is a disorder of cancer progression, in which progression from adenoma to carcinoma can occur in patients in as little as 2 years. For this reason, surveillance intervals in LS are more frequent than for surveillance of adenomas in the general population. Likewise, patients with LS are at great risk for developing multiple cancers at the same time (synchronous malignancy), as well as developing recurrent cancers at the same or other sites later in their lives (metachronous malignancy).
Although it is rare, silencing of both alleles of the same MMR gene or 1 allele of 2 different MMR genes results in a new, particularly aggressive form of LS known as CMMR-D. This ultrahigh risk group will benefit from aggressive cancer surveillance for digestive and other extraintestinal cancers. Compliance with surveillance procedures remains a challenge for most adolescents and young adults with chronic medical conditions. Follow-up education and continued communication between the LS family and the management team that includes a genetic counselor can improve compliance and outcomes. Lifelong medical care is critical for this population and care must be taken to reinforce the need as the pediatric patient transitions to adult care. In this setting, transition to a specialist with expertise in hereditary CRC is advisable.
The authors thank Heather Hampel and C. Richard Boland for their thoughtful review of this manuscript.
1. American Cancer SocietyCancer Facts & Figures 2013. 2013; Atlanta:American Cancer Society, 12.
2. Jemal A, Siegel R, Ward E, et al. Cancer statistics. CA Cancer J Clin
3. Vasen HF, Boland CR. Progress in genetic testing, classification, and identification of Lynch syndrome. JAMA
4. Bozzao C, Lastella P, Stella A. Anticipation in lynch syndrome: where we are where we go. Curr Genomics
5. Martin-López JV, Fishel R. The mechanism of mismatch repair and the functional analysis of mismatch repair defects in Lynch syndrome. Fam Cancer
6. Boland CR, Goel A. Microsatellite instability in colorectal cancer. Gastroenterology
7. Shah SN, Hile SE, Eckert KA. Defective mismatch repair, microsatellite mutation bias and variability in clinical cancer phenotypes. Cancer Res
8. Warthin AS. Heredity with reference to carcinoma as shown by the study of the cases examined in the pathological laboratory of the University of Michigan, 1895–1913. Arch Intern Med
9. 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
10. Umar A, Boland CR, Terdiman JP, et al. Revised Bethesda Guidelines for hereditary nonpolyposis colorectal cancer (Lynch syndrome) and microsatellite instability. J Natl Cancer Inst
11. Boland CR. Evolution of the nomenclature for the hereditary colorectal cancer syndromes. Fam Cancer
12. Lindor NM, Rabe K, Petersen GM, et al. Lower cancer incidence in Amsterdam-I criteria families without mismatch repair deficiency: familial colorectal cancer type X. JAMA
13. Geiersbach KB, Samowitz WS. Microsatellite instability and colon cancer. Arch Pathol Lab Med
14. Boland CR, Thibodeau SN, Hamilton SR, et al. A National Cancer Institute Workshop on microsatellite instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res
15. Bouzourene H, Hutter P, Losi L, et al. Selection of patients with germline MHL1 mutated Lynch syndrome by determination of MLH1 methylation and Braf mutations. Fam Cancer
16. Loughry MB, Waring PM, Tan A, et al. Incoporation of somatic BRAF mutation testing into an algorithm for investigation of hereditary non-poluposis colorectal cancer, 2007. Fam Cancer
17. Van Lier MGF, Leenen CHM, Wagner A, et al. Yield of routine molecular analyses in colorectal cancer patients, 70 years detect underlying Lynch syndrome. J Pathol
18. Moreira L, Balaguer F, Lindor N, et al. Identification of Lynch syndrome among patients with colorectal cancer. JAMA
19. Mayo CW, Pagtalunan RJG. Malignancy of colon and rectum in patients under 30 years of age. Surgery
20. Behbehani A, Sakwa M, Ehrlichman R, et al. Colorectal carcinoma in patients under age 40. Ann Surg
21. Hill DA, Furman WL, Billups CA, et al. Colorectal carcinoma in childhood and adolescence: a clinicopathologic review. J Clin Oncol
22. Karnak I, Ciftci AO, Senoacak ME, et al. Colorectal carcinoma in children. J Pediatr Surg
23. Bacon HE. Anus, Rectum, Sigmoid Colon. 3rd ed, vol 21949; Philadelphia:JB Lippincott, 603.
24. Niessen RC, Berends MJW, Wu Y, et al. Identification of mismatch repair gene mutations in young patients with colorectal cancer and in patients with multiple tumours associated with hereditary non-polyposis colorectal cancer. Gut
25. Bodas A, Pérez-Segura P, Maluenda C, et al. Lynch syndrome in a 15-year-old boy. Eur J Pediatr
26. Datta RV, LaQuaglia MP, Paty PB, et al. Genetic and phenotypic correlates of colorectal cancer in young patients. N Engl J Med
27. Huang SC, Lavine JC, Boland PS, et al. Germline characterization of early-aged onset of hereditary non-polyposis colorectal cancer. J Pediatr
28. Durno CA, Aronson M, Tabori U, et al. Oncologic surveillance for subjects with biallelic mismatch repair gene mutations: 10 year follow-up of a kindred. Pediatr Blood Cancer
29. Gonzalez RS, Shulman SC, Katzenstein HM, et al. Colorectal adenocarcinoma: a pediatric case review with a focus on mismatch repair gene mutations and E-cadherin expression. Pediatr Dev Pathol
30. Stoffel E, Mkherhee B, Raymon VM, et al. Calculation of risk of colorectal and endometrial cancer among patients with Lynch syndrome. Gastroenterology
31. Jong AE, Hendriks YM, Kleibeuker JH, et al. Decrease in mortality in Lynch syndrome families because of surveillance. Gastroenterology
33. Bartuma K, Nilbert M, Carlsson C. Family perspectives in Lynch Syndrome becoming a family at risk, patterns of communication, and influence on relations. Hered Cancer Clin Pract
34. Eledstein DL, Axilbund J, Baxter M, et al. Rapid development of colorectal neoplasia in patients with Lynch Syndrome. Clin Gastroenterol Hepatol
35. Stoffel EM, Turgeon DK, Stockwell DH, et al. Missed adenomas during colonoscopic surveillance in individuals with Lynch syndrome (hereditary nonpolyposis colorectal cancer). Cancer Prev Res (Phila)
36. Hüneburg R, Lammert F, Rabe C, et al. Chromocolonoscopy detects more adenomas than white light colonoscopy or narrow band imaging colonoscopy in hereditary nonpolyposis colorectal cancer screening. Endoscopy
37. Palomaki GE, McClain MR, Melillo S, et al. EGAPP supplementary evidence review: DNA testing strategies aimed at reducing morbidity and mortality from Lynch syndrome. Genet Med
38. Hendricks YM, de Jong AE, Morreau H, et al. Diagnostic approach and management of Lynch syndrome (hereditary nonpolyposis colorectal carcinoma): a guide for clinicians. CA Cancer J Clin
39. Broaddus RR, Lynch HT, Chen LM, et al. Pathologic features of endometrial carcinoma associated with HNPCC: a comparison with sporadic endometrial carcinoma. Cancer
40. Jasperson KW, Vu TM, Schwab AL, et al. Evaluating Lynch syndrome in very early onset colorectal cancer probands without apparent polyposis. Fam Cancer
41. Durno CA, Holter S, Sherman PM, et al. The gastrointestinal phenotype of germline biallelic mismatch repair gene mutations. Am J Gastroenterol
42. Krüger S, Kinzel M, Walldorf C, et al. Homozygous PMS2 germline mutations in two families with early-onset haematological malignancy, brain tumours, HNPCC-associated tumours, and signs of neurofibromatosis type 1. Eur J Hum Genet
43. Plaschke J, Linnebacher M, Kloor M, et al. Compound heterozygosity for two MSH6 mutations in a patient with early onset of HNPCC-associated cancers, but without hematological malignancy and brain tumor. Eur J Hum Genet
44. Baas AF, Gabbett M, Rimac M, et al. Agenesis of the corpus callosum and gray matter heterotopia in three patients with constitutional mismatch repair deficiency syndrome. Eur J Hum Genet
45. Bonadona V, Bonaiti B, Olschwang S, et al. Cancer risks associated with germline mutations in MLH1, MSH2,
genes in Lynch syndrome. JAMA
46. Senter L, Clendenning M, Sotamaa K, et al. The clinical phenotype of Lynch syndrome due to germ-line PMS2 mutations. Gastroenterology
47. Vasen HFA, Blanco I, Aktan-Collan K, et al. Revised guidelines for the clinical management of Lynch syndrome (HNPCC): recommendations by a group of European experts. Gut
48. Holter S, Pollett A, Zogopoulos G, et al. Hepatic adenomas caused by somatic HNF1A mutations in children with biallelic mismatch repair gene mutations. Gastroenterology
49. Win KA, Jenkins MA, Buchanan DD, et al. Determining the frequency of de novo germline mutations in DNA mismatch repair genes. J Med Genet
50. Lynch HT, Lynch JF, Shaw TG. Hereditary gastrointestinal cancer syndromes. Gastrointest Cancer Res 2011; 4 (Suppl 1):S9–S17.