Fax: (402) 280-1734
Article first published online: 12 NOV 2003
Copyright © 2003 American Cancer Society
Volume 100, Issue 1, pages 53–64, 1 January 2004
How to Cite
Lynch, H. T., Riley, B. D., Weismann, S., Coronel, S. M., Kinarsky, Y., Lynch, J. F., Shaw, T. G. and Rubinstein, W. S. (2004), Hereditary nonpolyposis colorectal carcinoma (HNPCC) and HNPCC-like families: Problems in diagnosis, surveillance, and management. Cancer, 100: 53–64. doi: 10.1002/cncr.11912
The pedigrees and other figures for this article were composed by Tami Richardson-Nelson, B.G.S.
The contents of this article are solely the responsibility of the authors and do not necessarily represent the official views of the State of Nebraska or the Nebraska Department of Health and Human Services.
- Issue published online: 17 DEC 2003
- Article first published online: 12 NOV 2003
- Manuscript Accepted: 8 OCT 2003
- Manuscript Revised: 22 SEP 2003
- Manuscript Received: 6 AUG 2003
- Nebraska cigarette taxes awarded to Creighton University by the Nebraska Department of Health and Human Services
- National Institutes of Health Grant. Grant Number: #1U01 CA86389
- hereditary cancer;
- colorectal carcinoma (CRC);
- medical genetics;
- genetic counseling;
- Lynch syndrome;
- hereditary nonpolyposis colorectal carcinoma (HNPCC)
To the authors' knowledge, hereditary nonpolyposis colorectal carcinoma (HNPCC) is the most commonly occurring hereditary disorder that predisposes to colorectal carcinoma (CRC), accounting for approximately 2–7% of all CRC cases diagnosed in the U.S each year. Its diagnosis is wholly dependent on a meticulously obtained family history of cancer of all anatomic sites, with particular attention to the pattern of cancer distribution within the family.
The objective of the current study was to illustrate various vexing problems that can deter the diagnosis of HNPCC and, ultimately, its management. This was an observational cohort study. Sixteen HNPCC and HNPCC-like families were selected from a large resource of highly extended HNPCC families. High-risk patients were selected from these HNPCC families. An ascertainment bias was imposed by the lack of a population-based data set. Personal interviews and questionnaires were used for data collection.
There was an array of difficulties highlighted by limitations in compliance, lack of a clinical or molecular basis for an HNPCC diagnosis, ambiguous DNA findings, problems in genetic counseling, failure to meet Amsterdam or Bethesda criteria, small families, lack of medical and pathologic documentation, poor cooperation of family members and/or their physicians, cultural barriers, economic stress, frequent patient fear and anxiety, perception of insurance discrimination, and limited patient and/or physician knowledge regarding hereditary cancer.
The diagnosis and management of HNPCC is predicated on physician knowledge of its phenotypic and genotypic heterogeneity, in concert with the multifaceted problems that impact on patient compliance. Cancer 2004;100:53–64. © 2003 American Cancer Society.
Hereditary nonpolyposis colorectal carcinoma (HNPCC), also known as Lynch syndrome, is the most commonly occurring hereditary syndrome that predisposes to colorectal carcinoma (CRC) (Fig. 1).1 It accounts for approximately 2–7% (2940 and 10,290 cases) of the CRC diagnosed in the U.S. annually.1 Its diagnosis is frequently difficult because, with the exception of cutaneous stigmata in its Muir–Torre syndrome (MTS) variant (which is characterized by sebaceous adenomas, sebaceous carcinomas, multiple keratoacanthomas, and solid tumors consonant with HNPCC2, 3), it lacks phenotypic signs that might facilitate its presymptomatic diagnosis.
Although the literature has increased enormously with regard to diagnostic guidelines for HNPCC (see Table 1 for Amsterdam I and Amsterdam II Criteria, as well as Bethesda guidelines),4–6 information concerning its surveillance, management,1, 7, 8 and molecular genetic testing,9, 10 as well as the ethical and malpractice issues that impact on these concerns,11 the description of barriers to its diagnosis and management, and the compliance of at-risk patients to these recommendations to our knowledge has been lacking.
|Amsterdam I criteriaa|
|• At least 3 relatives with histologically verified colorectal carcinoma:|
|1. One is a first-degree relative of the other 2|
|2. At least 2 successive generations affected|
|3. At least 1 of the relatives with colorectal carcinoma was diagnosed at < 50 years of age|
|4. Familial adenomatous polyposis has been excluded|
|Amsterdam II criteriab|
|• At least 3 relatives with an hereditary nonpolyposis colorectal carcinoma-associated malignancy (colorectal carcinoma, endometrial carcinoma, gastric carcinoma, ovarian carcinoma, carcinoma of the ovary, ureter/renal pelvis, carcinoma of the brain, carcinoma of the small bowel, carcinoma of the hepatobiliary tract, and skin cancer [sebaceous tumors]):|
|1. One is a first-degree relative of the other 2|
|2. At least 2 successive generations affected|
|3. At least 1 of the hereditary nonpolyposis colorectal carcinoma-associated malignancies should be diagnosed at < 50 years of age;|
|4. Familial adenomatous polyposis should be excluded in any colorectal carcinoma cases|
|• Tumors should be verified whenever possible.|
|Bethesda guidelines for testing of colorectal tumors for microsatellite instabilityc|
|1. Individuals with cancer in families that meet the Amsterdam Criteria|
|2. Individuals with two HNPCC-related malignancies, including synchronous and metachronous colorectal carcinomas or associated extracolonic carcinomasd|
|3. Individuals with colorectal carcinoma and a first-degree relative with colorectal carcinoma and/or HNPCC-related extracolonic carcinoma and/or a colorectal adenoma; one of the tumors diagnosed at age < 45 years, and the adenoma diagnosed at age < 40 years|
|4. Individuals with colorectal carcinoma or endometrial carcinoma that was diagnosed at age < 45 years|
|5. Individuals with right-sided colorectal carcinoma with an undifferentiated pattern (solid/cribiform) on histopathology diagnosed at age < 45 yearse|
|6. Individuals with signet ring cell-type colorectal carcinoma that was diagnosed at age < 45 yearsf|
|7. Individuals with adenomas diagnosed at age < 40 years|
The objective of the current study was to present what we believe to be the first report that portrays certain vexing problems that may deter the diagnosis and, ultimately, the management of an HNPCC family.
MATERIALS AND METHODS
This study of HNPCC-prone families was approved by Creighton University's Institutional Review Board. Two additional families were provided by Evanston Northwestern Healthcare Center for Medical Genetics. Each proband had provided genealogy and cancer history information through as many generations as possible. With the permission of the proband, other adult family members were contacted for additional information, and they signed written informed consent forms that enabled us to secure cancer-related medical and pathology documents.7, 12, 13 The data, once compiled, were then cross-referenced for accuracy. Peripheral blood samples were collected from potentially genetically informative consenting relatives, or from HNPCC-affected patients, for molecular genetic testing.14, 15 Consent was obtained for microsatellite instability (MSI) testing of tumor blocks in specific cases.
A subset of 15 HNPCC families was selected from our HNPCC resource of > 400 families based on the presence of factors that could pose barriers to diagnosis, screening, and management. The pedigrees have been encrypted without compromising their scientific accuracy to protect confidentiality.
An HNPCC Family: MLH1 Mutation
Family A (Fig. 2) met the Amsterdam I diagnostic criteria with its 4 generations of colon carcinoma patients (I, II, III, and IV), who ranged in age from 32–74 years (IV-2, IV-1, III-1, II-1, I-1, II-3, and II-4). The MLH1 mutation was identified in the proband (IV-2).
Note that the progenitor (I-1) was diagnosed with cecal carcinoma at age 74 years. Although this is a later age of onset for CRC than typically occurs in HNPCC, it does reflect the variation in age at onset in certain families. It is also interesting to note the rarely occurring carcinoma of the ileum appearing in the proband's brother (IV-1). Lessons learned: this family was presented because of its more typical features and can be used as a “standard” Lynch syndrome example as we discuss pedigrees that demonstrate more atypical variations, which we are referring to as “challenging HNPCC pedigrees.”
Atypical HNPCC: MSH6 Mutation
Family B (Fig. 2) depicts 3 generations (I, II, and III) with HNPCC syndrome malignancies occurring at a relatively late age of onset; the patients ranged in age from 51–74 years (III-2, II-1, II-2, and I-4). It is interesting to note individual II-2, who was diagnosed with synchronous colon and endometrial carcinoma at age 59 years, and in whom the MSH6 mutation was identified.
This family failed to meet the Amsterdam criteria4, 5 because of the late age of cancer onset. Nevertheless, the MSH6 mutation was identified in the maternal lineage of the family. Lessons learned: a so-called atypical or more “benign” presentation with later age of onset of CRC and other HNPCC malignancies, particularly endometrial carcinoma, often characterizes families with the MSH6 mutation.15
HNPCC Diagnosis: Initiated by a Nuclear Cancer Cluster
The proband (IV-7 in Fig. 2) was diagnosed with an AJCC Stage II, moderately differentiated adenocarcinoma of the cecum at age 45 years. His son (V-2) was diagnosed with a Stage II, moderately differentiated adenocarcinoma of the rectum at age 19 years. The proband reported that many of his relatives in this extended family had HNPCC syndrome malignancies. The combination of a right-sided CRC diagnosed prior to age 50 years along with a very early rectal carcinoma diagnosis prompted MSI analysis in individual IV-7, which was positive. A mutation in the MLH1 gene was found in individual IV-7.
The finding of early-onset CRC in a father (IV-7) and son (V-2), particularly in the absence of multiple colonic adenomas in this nuclear cancer cluster, would ordinarily be sufficient to strongly consider an HNPCC diagnosis and to test for MSI in one of their CRC tumor blocks. If positive for MSI, there would be a strong indication to pursue testing of mismatch repair (MMR) genes. If DNA testing remained negative, we would still manage all first-degree relatives of these CRC affected individuals as if they were positive for an MMR gene mutation. Lessons learned: maintain a high level of suspicion for families with some but not all features of Lynch syndrome. Use MSI to help gauge the level of suspicion and guide management if the MMR mutation is not identified.
Atypical HNPCC: MSH2 and MLH1 Mutation Negative
The proband (IV-1 in Fig. 2) remained cancer free at age 50 years. However, he initiated genetic research studies because his daughter (V-2) developed a Stage III, poorly differentiated adenocarcinoma (with focal mucinous carcinoma differentiation) of the ascending colon at age 21 years. The proband's mother (III-3) had died from metastatic carcinoma of the sigmoid colon at age 26 years. The proband's maternal uncle (III-1) reportedly was diagnosed with colon carcinoma at age 46 years. Both MLH1 and MSH2 were fully sequenced on individual V-2 and no mutation was found.
This pedigree (Family D), although having hereditary features, was not consistent with the Amsterdam criteria. Tests for MLH1 and MSH2 mutations were negative. Approximately half of the classic HNPCC cases will be found to be negative for any of the known germline mutations that predispose to HNPCC.1, 7 It also is possible that this family may have an unidentifiable mutation in either MLH1 or MSH2. Approximately 27%14, 16 of families will have a large deletion that may be missed by current genetic testing technologies. The possibility of an MSH6 mutation should be investigated. However, should a cancer-causing mutation fail to be identified, the proband (IV-1 in Fig. 2) and his first-degree and selected second-degree relatives still should be screened and managed as if they were, in fact, potential candidates for manifesting HNPCC malignancies. Lessons learned: maintain a high level of suspicion for families with hereditary features in whom DNA testing is negative and manage accordingly.
Attenuated FAP: A Diagnostic Dilemma That May Be Confused with HNPCC
The proband (III-15 in Fig. 2) was diagnosed with a moderately differentiated periampullary adenocarcinoma at age 53 years. Three of his sisters (III-13, III-12, and III-10) developed CRC at ages 53 years, 65 years, and 70 years, respectively. Their mother (II-3) was diagnosed with 2 separate primary carcinomas of the sigmoid colon and ascending colon at age 85 years. Four first cousins to the proband (III-3, III-5, III-6, and III-8) also developed colon or rectal carcinomas; these individuals ranged in age from 48–61 years at the time of diagnosis. It is interesting to note that 13 members of this family developed nonprofuse adenomatous polyps (V-3, IV-4, IV-6, IV-10, IV-11, III-2, III-5, III-6, III-8, III-10, III-12, III-13, and III-16). Two family members (IV-4 and III-10) had multiple gastric polyps. A mutation was found in the APC gene of the proband (III-15).
This family initially was considered to be a possible atypical late-age onset HNPCC kindred. Although adenomas were prevalent in many family members, the total number in each of these individuals did not reach the level of profuse adenomas as noted in classic familial adenomatous polyposis (FAP). The presence of periampullary carcinoma and multiple gastric polyps was suggestive of FAP. A detailed family history was required and the APC mutation was the sine qua non for the diagnosis of attenuated FAP and, thereby, exclusion of HNPCC. Lessons learned: careful assessment of the number of adenomatous polyps in family members and knowledge of component tumors for hereditary syndromes are required for an accurate diagnosis, which may be confirmed through DNA testing.
Variant of Uncertain Significance: Pathogenic or Polymorphic?
The proband (III-1 in Fig. 2) was unaffected at age 57 years. Her brother (III-2) had had several colonic polyps removed since age 52 years. The proband's mother (II-2) developed a late-stage ovarian carcinoma at age 79 years. A maternal aunt (II-3) was diagnosed with a Stage III endometrial carcinoma and a pathologically unverifiable separate primary tumor of the ovary at age 70 years. The maternal grandmother (I-1) was reported to have a possible CRC at age 46 years (tumor blocks were unavailable for MSI testing). Negative results from BRCA1/2 testing on individual II-2 led to the full sequencing of MLH1/MSH2, in which a variant of uncertain significance was found in MSH2.
Endometrial carcinoma can be a presenting manifestation of HNPCC, and in some studies the penetrance of endometrial carcinoma exceeds that of CRC in female HNPCC mutation carriers.17 The history of an ovarian second primary tumor, in combination with a single individual with CRC diagnosed at age < 50 years, is suggestive of HNPCC. The possible combination of endometrial and ovarian carcinomas as separate primary tumors in the maternal aunt (II-3) was difficult to differentiate; nevertheless, even if certainty is lacking in that differentiation, it can be a barrier to the clinician when making an assessment of the significance of the family history.
To our knowledge, the MSH2 variant, P5R, has not been reported to the mutation database maintained by the International Collaborative Group on HNPCC (unpublished data). It is a missense mutation occurring very early in the coding region, leading to a predicted nonconservative amino acid change (proline to arginine), which raises the question of whether it may lead to altered protein function. However, the amino acid is not evolutionarily conserved and does not comprise a known functional domain. Segregation analysis (using blood or paraffin blocks) could help to confirm coinheritance of the variant in affected individuals but, short of a functional assay for this mutation, its clinical significance is unresolved. Lessons learned: second primary malignancies are a feature of hereditary cancer. Confirmation of hereditary predisposition is complicated by a significant proportion of unclassified variants. Until such variants are defined as deleterious or benign, management should rest on the clinical features observed in the family, not the presence or absence of the variant in an unaffected individual.
Incorrect Screening: Failure to Diagnose and Manage Correctly
The proband (IV-2 in Fig. 3) was aware of her strong family history of CRC, which led her to be concerned about her need for cancer screening. She had undergone what she perceived as an adequate screening procedure, namely flexible sigmoidoscopy, which had been performed periodically since the age of 30 years. However, at age 45 years, 6 months after undergoing flexible sigmoidoscopy, the proband was diagnosed with a Stage III adenocarcinoma at the splenic flexure.
At age 57 years, she was told by her physician that her vaginal bleeding was the result of a “vaginal yeast infection.” However, having been subsequently educated by us regarding the natural history of HNPCC, she sought other medical consultation wherein a Stage I adenocarcinoma of the endometrium was diagnosed. That very same year, two of the proband's offspring (V-1 and V-2) were diagnosed with early-age onset HNPCC syndrome malignancies. An MSH2 mutation was subsequently found in the proband (IV-2).
There were many emerging clues in the family history leading us to consider an HNPCC diagnosis. For example, the proband's mother (III-2) was diagnosed with her first CRC at age 36 years with metachronous CRC diagnosed at age 51 years. After the proband's diagnosis of colon carcinoma, there should have been increased vigilance for endometrial carcinoma given the natural history of HNPCC. This family provides an excellent example of incorrect CRC screening of a patient (the proband, IV-2), namely flexible sigmoidoscopy instead of full colonoscopy, and the absence of screening in her children (V-1 and V-2). Lessons learned: maintain a working knowledge of hereditary features. Colonoscopy is essential for colonic surveillance; flexible sigmoidoscopy is insufficient for early detection, particularly given the predilection for right-sided CRCs. Surveillance should be extended to close relatives.
HNPCC Phenotype: Insufficient Physician Knowledge
The cancer history of Family H (Fig 3) met Amsterdam II criteria.5 Because the proband (IV-8) had been educated by us regarding the natural history of HNPCC, she requested what she perceived to be better gastrointestinal and gynecologic evaluation from her physicians. However, her physicians apparently did not have knowledge of the HNPCC syndrome. Being proactive, the proband switched her health care providers, consulted with a different physician who was knowledgeable about HNPCC, and was diagnosed with a Stage IC, poorly differentiated endometrial carcinoma, endometrioid type, at age 46 years. An MSH2 mutation was identified in her DNA.
The presence of the otherwise rare finding of duodenal carcinoma in the proband's mother (III-3) and grandmother (II-4), coupled with additional syndrome malignancies in Family H (Fig. 3), should have raised physician concern regarding an HNPCC diagnosis. The presence of the syndrome was, in fact, rather glaring given this striking tumor spectrum and its distribution in the family. Lessons learned: to avoid this diagnostic problem, physicians are strongly urged to complete a thorough review of a detailed cancer family history. This can lead to life-saving interventions.
Misinformation: Importance of Patient Education
In Family I (Fig. 3), individual V-1, a 22-year-old male, developed rectal bleeding. Having previously been diagnosed with hemorrhoids, he ignored the bleeding, which stopped 3 days later. Three months later, he mentioned the rectal bleeding to his parents. His father (IV-1), having been diagnosed with colon carcinoma at age 42 years (coupled with a strong family history of Lynch syndrome malignancies) and having been intensively educated by us about the natural history of HNPCC, strongly urged his son to seek physician consultation. Following his father's advice, the son sought consultation and was diagnosed with a Stage II, moderately differentiated adenocarcinoma of the cecum. An MLH1 mutation was later found in the family.
This family provides an example of how intensive education of an HNPCC family can benefit at-risk family members who might otherwise ignore signs and symptoms consistent with HNPCC and attribute same to otherwise “benign” conditions such as hemorrhoids. The son's response to an educated family member's advice to seek medical attention proved to be potentially life saving. Lessons learned: the diagnosis of a hereditary syndrome must be coupled with a well orchestrated educational program for the family18 to maximize the benefit of establishing a diagnosis.
Failure to Diagnose, Screen, and Manage
Family J (Fig. 3) was an extended HNPCC/MTS family. An MLH1 large deletion was found in family members.14 Focusing on 1 branch of the family (Fig. 3), we noted that the proband (III-7) developed endometrial carcinoma at age 44 years. Having attended a family information service (FIS),18 she informed her family physician of her hereditary cancer risk at each medical checkup, stating that five of her siblings (III-3, III-4, III-5, III-8, and III-9) had HNPCC/MTS malignancies, and that she required colonoscopy. However, she stated that her physician would pat her on the back and say, “You are just fine; don't worry!” Letters were sent by us to the proband's physician, explaining her high-risk status for HNPCC and emphasizing the need for annual colonoscopy. Six months later, at age 59 years, during her first colonoscopy, a Stage III adenocarcinoma was found in her cecum.
The proband in Family J (Fig. 3) suffered a delay in implementing targeted colonoscopic screening, and presented with a Stage III CRC. Lessons learned: maintain a working knowledge of component tumors of hereditary syndromes and run a literature search for unusual tumors.
Noncompliance Resulting, in Part, from Fear of Insurance Discrimination
The proband (III-4 in Fig. 4) manifested a Stage I, poorly differentiated adenocarcinoma of the endometrium at age 33 years. Subsequently, a Stage III, well differentiated adenocarcinoma of the cecum was diagnosed at age 42 years. The proband then learned from a report in a lay magazine that early onset endometrial and colon carcinomas were hallmark features of HNPCC. She contacted us, received genetic counseling, and requested genetic testing. An MLH1 mutation then was identified in her DNA.
According to the proband (III-4) her family “is not educated” and did not understand why she consented to mutation testing. To date, in spite of our many admonitions about their need for evaluation, the family has repeatedly failed to follow our recommended screening due, in part, to their fear of insurance discrimination.
As in the proband (III-4 in Fig. 4), when tumors of the colon and endometrium appear in combination, particularly with an early age at onset, one should have a high index of suspicion for HNPCC. Genetic counseling with genetic testing in consenting family members, in concert with highly targeted HNPCC screening of MLH1 mutation carriers, could help to save their lives. Lessons learned: deterrents to screening and gene testing may involve the perception of insurance discrimination; however, patient fear greatly exceeds reality.19–21 Existing legislation information for each U.S. state is available from URL: http:///www.genome.gov/10002338 [accessed 15 July 2003].
Adoption: A Medical and Legal Quagmire
The proband (II-1 in Fig, 4) and her fraternal twin sister (II-2) developed CRC at the ages of 53 years and 55 years, respectively. The proband's daughter (III-3) developed a carcinoma of the transverse colon at age 31 years. The proband (II-1) and her sister (II-2) were adopted into the same family at 2 weeks of age. Knowledge of their biologic family has been virtually absent.
Adoption policies in the past, when this family was ascertained, would not allow the release of medical or genetic information regarding bloodline relatives of the adoptee. Although designed to protect privacy, the rules nevertheless denied Patients II-1 and II-2 the “right to know” their family history, compromising early disease detection and prevention. Lessons learned: although the CRC diagnoses for the proband (II-1) and her fraternal twin sister (II-2) did not meet Amsterdam criteria, their cancer occurrences should be considered highly suspicious when based on the lack of knowledge of the extended family.
Family M (Fig. 4)22 is a family with the MTS variant of HNPCC. The proband (III-1) had been diagnosed with CRC as well as MTS lesions. Four of the proband's 6 siblings (III-2, III-3, III-4, and III-5) also developed CRCs between the ages of 39–51 years. According to the family report, an additional 5 family members (III-7, III-10, IV-3, IV-4, and IV-8) developed CRC before age 43 years. The MSH2 germline mutation was identified in this family.22
However, there appeared to be considerable mistrust among the family membership with respect to our diagnostic, management, and research objectives. The proband (III-1) told us that these negative attitudes toward physicians were common in his family, as were illiteracy and a lack of access to consistent healthcare.
In spite of the discovery of the MSH2 germline mutation, the striking cutaneous stigmata of the MTS, and a history of a large number of family members dying of cancer, the long-term cooperation of these high-risk family members was low. Lessons learned: it is clear that cultural factors, in this case those stemming from a high rate of illiteracy, must be fully reconciled when attempting to assist individuals known to be at inordinately high risk for cancer. An attempt at resolving one cultural sensitivity problem through a communication program for African Americans has been described.23
Sociopsychologic and Religious Fervor
Family N (Fig. 4) was an extended family with classic HNPCC syndrome malignancies manifested through four generations (I–IV). An MLH1 mutation was identified and genetic test results were disclosed in a genetic counseling setting. Individual III-9, after being told she was a carrier of the deleterious mutation, stated she would not follow our screening recommendations claiming, “If God wants me to die, I will die. I won't prevent what God has in store for me. I'm not afraid to die.” Many family members raised other concerns. For example, individual IV-12, also a mutation carrier, stated that he would not follow the screening recommendations because he could not afford to pay for the procedures, being recently divorced and raising a 5-year-old son. His brother (IV-13) refused to receive his result because “the blood draw was painful.” Individual III-13, after being told she was not a mutation carrier, stated she felt intensely guilty (survival guilt) for not being a carrier. Apparently, her niece (IV-18) shunned her after the niece's mother's (III-12) cancer-related death, stating, “My mother died and you didn't.”
Family N (Fig. 4) represents an HNPCC family whose compliance was hindered by religious beliefs and by a variety of psychologic and economic factors that affected their attitudes toward gene testing, surveillance, and management. Lessons learned: an effort should be made to alleviate barriers such as these through compassionate and culturally sensitive genetic counseling (e.g., referring the family for psychologic counseling or working with clergy in whom they are willing to confide.)
Family O (Fig. 4) was a classic HNPCC kindred. After attending an FIS, individual III-6 had been educated regarding the natural history of HNPCC and he understood fully the importance of vigilant screening. However, the economic situation on his farm deterred him from following our screening recommendations. He simply could not afford routine colonoscopies, stating “I have to allocate funds to hold my farm from foreclosure.” At age 58 years, he was diagnosed with a Dukes Stage C, moderately differentiated adenocarcinoma of the sigmoid colon that metastasized to his small bowel at age 60 years, resulting in his death.
This case (Fig. 4) centers around an economic issue of a patient without medical insurance, whose farm was being foreclosed, and who was too proud to ask for charity to defray the expense for colonoscopy. This economic burden likely cost him his life. Lessons learned: physicians must be fully cognizant of patients' economic limitations and seek social worker consultation to help defray medical expenses.
The diagnosis, surveillance, and management of patients at high risk for HNPCC can aid significantly in early cancer detection and improved survival.8 The most important basis for the diagnosis of any form of hereditary CRC is the compilation of a detailed cancer family history (Table 2) with attention to malignancies of all anatomic sites through a minimum of three generations.7, 18, 24–26 The physician's knowledge of the clinicopathologic features that comprise the various hereditary cancer syndromes is essential. From this, a potpourri of cancer control benefits to relatives considered at risk for CRC may become possible; specifically, CRC is potentially one of the most preventable malignancies, through colonoscopy and polypectomy.8
|After developing a detailed family history of cancer for all anatomic sites, look for the following:|
|✓ Early age of onset of hereditary cancer or one or more colonic adenomas, ≥ 15–20 years earlier than sporadic cancer.|
|✓ Specific patterns of multiple primary tumors (e.g., carcinoma of the proximal colon and endometrial carcinoma).|
|✓ Synchronous and metachronous colon carcinomas.|
|✓ Associated carcinoma of the endometrium, ovary, or more rarely occurring small bowel, transitional cell carcinoma of the ureter or renal pelvis.|
|✓ Distinguishing pathology features (i.e., mucoid and signet cell).|
|✓ Identification of MMR gene mutation in MLH1, MSH2, or MSH6.|
Annual full colonoscopy is required at an early age (approximately 20–25 years) in HNPCC (Table 3) because of the early age at onset of CRC (average age of approximately 45 years),1 its predilection to proximal location in the colon,27 and its accelerated carcinogenesis.7, 28, 29 For attenuated familial adenomatous polyposis (AFAP), with its right-sided predominance of adenomas and a later age of CRC onset (average age of approximately 55 years), full colonoscopy is indicated for its surveillance and may be initiated when carriers are aged in their late teens.30
|Once an HNPCC diagnosis is made:|
|✓ Annual full colonoscopy initiated between ages 20–25 years|
|✓ In addition for females, beginning at age 30 years:|
|○ Annual transvaginal ultrasound|
|○ Annual endometrial aspiration|
|○ Annual CA 125|
However useful the Amsterdam criteria4, 5 and Bethesda guidelines6 might be for diagnostic assessment of an HNPCC family, there will be clinical situations involving families, some mentioned in this article, that may be wholly uninformative for the application of these criteria. In these situations, one also may consider so-called “pattern recognition” for the genetic diagnosis of the cancer phenotype in probands and their immediate first-degree and second-degree relatives. Thus, the presence of a striking phenotype such as proximal CRC before age 40 years, metachronous CRC, rarely occurring tumors such as duodenal carcinoma (Family I in Fig. 3), mucoid pathology features with an excess of signet cell features, or the early onset of ≥ 1 colonic adenomas (even in only a single first-degree or second-degree relative) should prompt further investigation of the possibility of HNPCC. MSI testing of the CRC tissue block should be performed because high MSI is the hallmark of HNPCC CRC tumors31 and, if positive, clinicians should proceed with a search for a MMR germline mutation in a genetically informative individual.1, 7, 10
MMR Mutations: Genotypic and Phenotypic Heterogeneity
HNPCC is associated with germline mutations in, or the malfunctioning of postreplicative, MMR genes. Two genes, MLH1 and MSH2, account for nearly 90% of identified MMR mutations, whereas MSH6 will account for approximately 10%. Also observed are mutations in PMS2, MLH3, and EXO1.10
Greater than 400 different pathogenic mutations have been registered in the international database of mutations in HNPCC kindreds (available from URL: http://nfdht.nl [accessed 15 July 2003]).32 The majority truncate the protein and are linked to typical HNPCC. To our knowledge, one of the most widespread recurring mutations reported to date is MLH1 del 616.32 However, approximately 25% of all mutations are nontruncating and many of those are linked to families not fulfilling the Amsterdam I or Amsterdam II criteria for HNPCC.4, 5 The ability to characterize the clinical significance of nontruncating mutations through segregation analysis depends on obtaining the cooperation of families and research laboratories, and the availability of tissue. Most variants remain in an indeterminate status, and the clinical utility of gene testing is thereby hampered. Detection techniques are emerging for genomic rearrangements and large deletions.33 The full clinical utility of MMR gene testing will not be realized until it is incorporated into standard practice.
Variable phenotypic features of HNPCC are associated with specific MMR germline mutations.10, 34 For example, MLH1 is mostly related to the typical form of HNPCC wherein approximately 30% of the mutations are the missense type.17, 35, 36MSH2 also predisposes predominantly to typical HNPCC as well as Muir–Torre syndrome cutaneous stigmata37; however, extracolonic tumors may be more common than in its MLH1 mutation counterpart.34 As shown in the text (Family B in Fig. 1), MSH6 may present as atypical HNPCC. MSI also may be decreased in these MSH6 mutation-positive tumors.15, 38 Alternatively, there may be yet-to-be-identified cancer susceptibility gene mutations that are causally responsible in these atypical HNPCC families.1, 7
In a family in which a specific pathogenic mutation has been identified, testing for that mutation can identify members who are mutation positive and thereby require intensive screening, which can lead to the early detection of cancer. Conversely, relatives who are found to be mutation negative will not be required to undergo the very costly and often inconvenient screening. Major bonuses will be a reduced psychologic and economic burden, and a decreased need for excessive clinical examinations for succeeding generations of the mutation-negative progenitors.39
The significance of heredity in cancer etiology has come of age in all medical specialties. Molecular genetic discoveries not only impact on hereditary cancer syndrome diagnoses, but, moreover, are laying the foundation for gene-targeted therapy.40, 41 Patients are beginning to ask the “right questions” of their physicians pertaining to their family cancer risk, which often are promulgated by the media as well as input from commercial DNA laboratories and pharmaceutical companies. However, the linchpin for “making it work” will center in a significant way on resolving the many barriers to hereditary cancer syndrome diagnosis and management.
- 3The Muir-Torre syndrome: a variant of hereditary nonpolyposis colorectal cancer syndrome. J Tumor Marker Oncol. 1996; 11: 19–31., , .
- 9VogelsteinB, KinzlerKW, editors. The genetic basis of human cancer. New York: McGraw-Hill, 1998.
- 16Molecular analysis of hereditary nonpolyposis colorectal cancer in the United States: high mutation detection rate among clinically selected families and characterization of an American founder genomic deletion of the MSH2 gene. Am J Hum Genet. 2003; 72: 1088–1100., , , et al.
- 22Colorectal cancer and the Muir-Torre syndrome in a Gypsy family: a review. Am J Gastroenterol. 1999; 94: 575–580., , , et al.Direct Link:
- 31A 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. 1998; 58: 5248–5257., , , et al.
- 32International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer. Mutation Database. International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer. Available from URL: http://www.nfdht.nl [accessed April 30, 2003].
- 39Carrier risk status changes resulting from mutation testing in hereditary nonpolyposis colorectal cancer and hereditary breast-ovarian cancer. Am J Hum Genet. In press., , , et al.