Hereditary Nonpolyposis Colorectal Carcinoma

HNPCC is inherited as an autosomal-dominant pattern with almost complete penetrance. It is estimated that from 0.5 to 6.0% of all colorectal carcinomas can be attributed to HNPCC. ^{35 , 50 , 51 , 52}

Based on clinical criteria only, Peltomäki ³⁶ noted the estimated incidence of HNPCC varies between 0.5 and 13% of the total colorectal carcinoma burden. Strictly molecular approaches based on the identification of germline mutation carriers among newly diagnosed colorectal patients whose carcinomas showed MSI have arrived at lower estimates—0.3 to 3% of the total colorectal carcinoma burden.

HNPCC has four major subtypes: (1) Lynch type I (site-specific nonpolyposis colorectal carcinoma); (2) Lynch type II (formerly called the carcinoma family syndrome), in which carcinomas occur in the colon and related organs (endometrium, ovaries, stomach, pancreas, and proximal urinary tract, among others); (3) Muir–Torre syndrome, associated with multiple benign and malignant skin neoplasms, sebaceous gland adenomas, carcinomas, and keratoacanthomas ⁵³ ; and (4) a variant of Turcot’s syndrome (brain neoplasms). HNPCC-related colon carcinomas begin with mutations in the mismatch repair genes and are RER positive.

Clinical criteria were established to confirm the diagnosis of HNPCC at a meeting of the International Collaborative Group on HNPCC in Amsterdam in 1990. ⁵¹ The group agreed that minimum criteria should include (1) at least three relatives with histologically verified colorectal carcinoma, one of whom should be a first-degree relative of the other two (those with FAP should be excluded); (2) at least two successive generations should be affected; and (3) in one of the relatives, colorectal carcinoma should have been diagnosed when the patient was younger than 50 years of age.

These criteria have not proved to be comprehensive; therefore, in 1999, the International Collaborative Group developed revised criteria and these are known as Amsterdam-II criteria. They are essentially the same, except that there should be at least three relatives with HNPCC-associated carcinoma (colorectal, endometrial, small bowel, ureter, or renal pelvis). Even this extension of clinical features fails to identify some families with germline mismatch repair gene mutations, so the Bethesda guidelines for testing of colorectal carcinomas for MSI have been developed. They include the following:

1. 
   

   Individuals with carcinoma in families that meet the Amsterdam criteria.

   
   
2. 
   

   Individuals with two HNPCC-related carcinomas, including synchronous and metachronous colorectal carcinomas or associated extracolonic malignancies (endometrial, ovarian, gastric, hepatobiliary or small bowel carcinoma, or transitional cell carcinoma of the renal pelvis or ureter).

   
   
3. 
   

   Individuals with colorectal carcinoma and a first-degree relative with colorectal carcinoma and/or HNPCC-related extracolonic malignancy and/or a colorectal adenoma; one of the malignancies diagnosed at age younger than 45 years, and the adenoma diagnosed at age younger than 40 years.

   
   
4. 
   

   Individuals with colorectal or endometrial carcinoma diagnosed at age younger than 45 years.

   
   
5. 
   

   Individuals with right-sided colorectal carcinoma with an undifferentiated pattern (solid/cribriform) on histopathology diagnosed at age younger than 45 years. (Note: solid/cribriform defined as poorly differentiated or undifferentiated carcinoma composed of irregular, solid sheets of large eosinophilic cells, and containing small glandlike spaces.)

   
   
6. 
   

   Individuals with signet-ring-cell-type colorectal carcinoma diagnosed at age younger than 45 years. (Note: composed of 50% signet-ring cells.)

   
   
7. 
   

   Individuals with adenomas diagnosed at age younger than 40 years.

In 2002, another workshop was held at the NCI (National Cancer Institute) in Bethesda; the revised guidelines that were created are listed in ▶ Table 21.5. ⁵⁴ It is estimated that 20 to 25% of population-based cases of colorectal carcinoma meet the Bethesda criteria. It is suggested that for all patients who meet these criteria, a search for MSI is indicated. To establish the most effective and efficient strategy for the detection of MSH2/MLH1 gene carriers in HNPCC, Pinol et al ⁵⁵ conducted a prospective multicenter nationwide study (the EPICOLON study) in 20 hospitals in the general community in Spain of 1,222 patients with newly diagnosed colorectal carcinoma. MSI testing and MSH2/MLH1 immunostaining were performed in all patients regardless of age, personal or family history, and carcinoma characteristics. Patients whose carcinoma exhibited MSI and/or lack of protein expression underwent MSH2/MLH1 germline testing. The revised Bethesda guidelines were fulfilled by 23.5% of patients and 7.4% had a mismatch repair deficiency, with the carcinoma exhibiting either MSI or loss of protein expression. Germline testing identified mutations in 0.9% in either MSH2 or MLH1 genes. Strategies based on either MSI testing or immunostaining previous selection of patients according to the revised Bethesda guidelines were the most effective (sensitivity, 81.8 and 81.8%; specificity, 98.0 and 98.2%; and positive predictive value, 27.3 and 29.0%, respectively) to identify MSH2/MLH1 gene carriers. They concluded the revised Bethesda guidelines are the most discriminating set of clinical parameters (OR, 33.3).

In conclusion, MSI and immunohistochemistry analysis using antibodies against MLH1, MSH2, PMS2, and MSH6 appeared to be equally effective for the identification of mutation carriers. Despite its name, polyps are a feature of HNPCC, and a review of the literature revealed a polyp incidence in 8 to 17% of first-degree relatives during colonoscopic screening. ³⁸

Carriers of an MMR defect develop adenomas more frequently than controls. The adenomas identified in carriers are larger, and a significantly higher proportion showed histologic features that are associated with a high risk of malignant degeneration, such as a high degree of dysplasia and the presence of more extensive villous architecture. ⁵⁶ A relatively high proportion of patients develop colorectal carcinoma within 3 years after a clean colonoscopy and this suggests that the adenoma–carcinoma sequence is accelerated and that the progression from adenoma to carcinoma may take fewer than 3 years.

Genetically speaking, HNPCC syndromes are dominantly inherited with nearly 100% penetrance reported by one group of investigators ⁵⁰ but only 70 to 80% (i.e., 20–30% of individuals with a predisposing mutation may never develop carcinoma) by others. ⁵⁷ Asymptomatic gene carriers can pass the causative mutation to their children. ⁵⁷ The disease is heterogeneous. All first-degree relatives of a patient with HNPCC have a 50% risk of carrying one of the deleterious genes. Adenomas in patients carrying HNPCC gene mutations display MSI, suggesting that mismatch repair defects are important early events in colorectal carcinogenesis. Carcinoma formation requires inactivation of both copies of a given mismatch repair gene, one copy by germline mutation and the other by somatic (acquired) mutation. ⁵⁷

Individuals with the Lynch syndrome differ from patients with sporadic colorectal carcinoma in several ways. ³⁵ They show an autosomal-dominant mode of inheritance, ¹ a predominance of proximal colonic carcinoma (72% of first colon carcinomas were located in the right colon, and only 25% were found in the sigmoid colon and rectum), ² an excess of multiple primary colonic carcinomas (18%), ³ an early age of onset (mean, 44 years), ⁴ a significantly improved survival rate compared with family members with distal colonic and rectal carcinomas when compared with right-sided lesions in an American College of Surgeons series (53% vs. 35% 5-year survivals), ⁵ and 24% developed metachronous colon carcinoma with a risk for the development of metachronous lesions in 10 years of 40% based on life-table methods. ⁶ In addition to the above features, Lynch type II syndrome is characterized by a high frequency of other adenocarcinomas and the occasional occurrence of cutaneous manifestations in the form of sebaceous adenomas and carcinomas, epitheliomas, or keratoacanthomas. In a study by Mecklin and Jarvinen, ⁵⁸ of 40 HNPCC families with 315 affected members, a total of 472 malignancies were identified. They included colorectal (63%), endometrial (8%), gastric (6%), biliopancreatic (4%), and uroepithelial (2%) carcinomas.

The RR of these carcinomas ranges from 3 to 25 times that of the general population. ¹⁰ The risk of colorectal carcinoma increases 1.6% per year from age 25 to 75 years in patients with these mutations. ⁵³ In a detailed pedigree analysis of 40 families with HNPCC, Aarnio et al ⁵⁹ identified 414 patients affected with carcinoma. The risk of any metachronous carcinoma reached 90% after treatment of colorectal carcinoma and 70% after endometrial carcinoma; the second malignancy was most often a new colorectal carcinoma or endometrial carcinoma. ⁵⁹ Other sites of carcinoma include breast, pancreas, and possible lymphoma and leukemia. The review by DeFrancisco and Grady ⁴⁸ found the four most common extracolonic carcinomas include (in descending order) endometrial, ovarian, gastric, and transitional cell carcinoma of the uroepithelial tract (bladder, kidney, ureter). Women with HNPCC are at a 10-fold increased risk of endometrial carcinoma, which is usually diagnosed between the ages of 40 and 60 years, that is, 15 years earlier than the general population. The estimated cumulative risk at age 70 years is 40 to 50%. With MSH2 and MLH1 mutations, the risk of colorectal carcinoma or endometrial carcinoma before age 50 years is 20 to 25% compared with 0.2% for the general population. ⁵³ Ovarian carcinoma is less common (incidence approximately 9%). Gastric carcinoma occurs in 5 to 20% of HNPCC families. The RR of gastric carcinoma was 19.3 in MSH2 mutation carriers compared with the general population. Transitional cell carcinoma of the uroepithelial tract occurs in 1% of HNPCC patients. Only the carriers of MSH2 mutations appear to have a significantly increased risk of carcinoma in the urinary tract (RR of 75.3). Overall, the RR of gastric carcinoma, ovarian carcinoma, and carcinoma of the urinary tract has been shown to be higher in patients with mutations in MSH2 as compared to patients with MLH1 mutations. Women with MSH6 mutations appear to be more likely to develop endometrial carcinoma. Other HNPCC-associated extracolonic neoplasms include carcinomas of the small bowel, pancreas, hepatobiliary tree, brain, and skin. The Muir–Torre syndrome, first described in 1967, refers to patients with HNPCC who also develop benign or malignant sebaceous skin neoplasms (sebaceous adenomas, carcinomas, squamous or basal cell), and multiple keratoacanthomas. The syndrome usually arises from mutations in MSH2. The development of glioblastoma multiforme, in association with HNPCC, is called Turcot’s syndrome. The Muir–Torre syndrome is also used to describe central nervous system neoplasms occurring in FAP, although these are usually medulloblastomas instead of glioblastomas. Approximately a third of patients with Turcot’s syndrome have mutations in one of the mismatch repair genes. HNPCC is associated with a 50% risk of a second carcinoma within 15 years of the initial carcinoma diagnosis ⁵³ compared with 5% in the general population. Information regarding the cumulative risk of developing various HNPCC-associated malignancies has been summarized from several reports and tabulated in ▶ Table 21.6. ^{10 , 48 , 53 , 59 , 60}

Plaschke et al ⁶¹ analyzed the involvement and phenotypic manifestations of MSH6 germline mutations in families suspected of HNPCC. Patients were preselected among 706 families by MSI, immunohistochemistry, and/or exclusion of MLH1 or MSH2 mutations and were subjected to MSH6 mutation analysis. Clinical and molecular data of MSH6 mutation families were compared with data from families with MLH1 and MSH2 mutations. They identified 27 families with 24 different pathogenic MSH6 germline mutations, representing 3.8% of the total of the families, and 14.7% of all families with DNA mismatch repair gene mutations. The median age of onset of colorectal carcinoma in putative mutation carriers was 10 years higher for MSH6 (54 years) compared with MLH1 and MSH2 (44 years). Relative to other malignancies, colorectal carcinoma was less frequent in MSH6 families compared with MLH1 and MSH2 families. In contrast, the frequency of non–HNPCC-associated neoplasms was increased. Later age of disease onset and lower incidence of colorectal carcinoma may contribute to a lower proportion of identified MSH6 mutations in families suspected of HNPCC. However, in approximately half of these families, at least one patient developed colorectal or endometrial carcinoma in the fourth decade of life. Therefore, a surveillance program as stringent as that for families with MLH1 or MSH2 mutations is recommended.

Approximately 60% of families that meet the Amsterdam-I criteria (HNPCC) have a hereditary abnormality in a DNA mismatch repair gene. Carcinoma incidence in Amsterdam criteria-I families with mismatch repair gene mutations is reported to be very high, but carcinoma incidence for individuals in Amsterdam criteria-I families with no evidence of a mismatch repair defect is unknown. Lindor et al ⁶² conducted a study to determine if carcinoma risks in Amsterdam criteria-I families with no apparent deficiency in DNA mismatch repair are different from carcinoma risks in Amsterdam criteria-I families with DNA mismatch repair abnormalities. 161 Amsterdam criteria-I pedigrees from multiple population- and clinic-based sources in North America and Germany were identified and grouped into families with (group A) or without (group B) mismatch repair deficiency by testing. A total of 3,422 relatives were included in the analysis. Group A families from both population- and clinic-based series showed increased incidence of HNPCC-related carcinomas. Group B families showed increased incidence only for colorectal carcinoma (standardized incidence ratio [SIR], 2.3) and to a lesser extent than group A (SIR, 6.1). Families who have fulfilled Amsterdam criteria-I but who have no evidence of DNA mismatch repair defect do not share the same carcinoma incidence as families with HNPCC with mismatch-repair deficiency. Relatives in such families have a lower incidence of colorectal carcinoma than those in families with HNPCC (Lynch syndrome) and incidence may not be increased for other carcinomas. These families should not be described or counseled as having HNPCC. To facilitate distinguishing these entities, the designation of “familial colorectal cancer type X” is suggested to describe this type of familial aggregation of colorectal carcinoma.

Jass ⁶³ dispelled any thoughts that the morphogenetic pathway involved flat adenoma or de novo carcinoma in these cases. In a review of 131 carcinomas, none was small and superficial. Residual adenoma (contiguous with carcinoma) was present in 3 of 3 (100%) in situ carcinomas, 8 of 9 (89%) carcinomas involving only submucosa, 4 of 14 (29%) carcinomas limited to the muscle coat, and 13 of 105 (12%) carcinomas extending beyond the muscle coat. Lynch type II syndrome colon carcinomas show a significant increase in the proportion of mucinous, poorly differentiated carcinomas, and Crohn-like lymphoid reaction around the lesion as well as a higher rate of synchronous and metachronous carcinomas. ^{35 , 64}

Other investigators have also studied HNPCC. Sankila et al, ⁶⁵ compared the survival rates of 175 patients with hMLH1-associated HNPCC with those of 14,000 patients with sporadic colorectal carcinoma diagnosed at less than 65 years. The overall 5-year cumulative relative survival rate was 65% for patients with HNPCC and 44% for patients with sporadic colorectal carcinoma.

In a series of 1,042 Japanese patients who underwent resection for colorectal carcinoma, 3.7% were found to have HNPCC. ⁶⁶ The characteristic early age of onset, right-sided predominance, and favorable survival were confirmed. Metachronous colorectal carcinoma developed significantly more often in cases with HNPCC (12.8 vs. 1.8%), and metachronous extracolonic malignancies also developed more often (10.2 vs. 3.5%). In cases with HNPCC, the mean interval between initial operation and the diagnosis of the second malignancy was 61 months (range, 12–153 months). These findings stress the importance of long-term follow-up.

Rodríguez-Bigas et al ⁶⁷ reported the experience from the HNPCC Registry at Roswell Park in Buffalo, New York, which included 301 people in 40 families. In 284 of 301 people, 363 carcinomas were identified. Colorectal carcinomas alone were identified in 64% and in conjunction with extracolonic malignancies in another 11%. Extracolonic malignancies alone were identified in 25%. The median age at diagnosis of colorectal carcinoma was 48 years. Right-sided malignancies predominated (55%), synchronous and metachronous lesions were noted in 33%, and synchronous or metachronous adenomas were documented in 51% of those studied. Generational anticipation was also noted.

Hampel et al ⁶⁸ assessed the frequency of the mismatch repair genes MLH1, MSH2, MSH6, and PMS2, in patients with a new diagnosis of colorectal carcinoma to identify patients with the Lynch syndrome. Genotyping of the carcinoma for MSI was the primary screening method. Among patients whose screening results were positive for MSI, they searched for germline mutations in the MLH1, MSH2, MSH6, and PMS2 genes with the use of immunohistochemical staining for mismatch repair proteins, genomic sequencing, and deletion studies. Family members of carriers of the mutations were counseled and those found to be at risk were offered mutation testing. Of 1,066 patients enrolled in this study, 19.5% had MSI, and 2.2% of these patients had a mutation causing the Lynch syndrome. Among the 23 probands with the Lynch syndrome, 10 were more than 50 years of age and 5 did not meet the Amsterdam criteria or the Bethesda guidelines for the diagnosis of HNPCC. In the families of 21 of the probands, 117 persons at risk were tested, and of these 52 had Lynch syndrome mutations. They concluded that routine molecular screening of patients with colorectal adenocarcinoma for the Lynch syndrome identified mutations in patients and their family members that otherwise would not have been detected. In most centers, it would not seem feasible to test all patients with colorectal carcinoma.

The characteristics of HNPCC mandate the development of surveillance programs. Screening recommendations have varied, but the recommendation by Lynch et al ⁶⁹ for gene carriers has been colonoscopy, beginning at age 20 to 25 years or at least 5 years earlier than the earliest age at which colon carcinoma was diagnosed in a particular kindred with the procedure repeated every other year until the patient is 30 years of age and then annually thereafter. The frequency of colonoscopic examination is justified by the finding that HNPCC adenomas have repair deficient cells with rapidly and relentlessly accumulating mutations that support the clinical concept of “aggressive adenomas” and accelerate the adenoma–carcinoma sequence. The authors strongly believe that germline carriers should be offered prophylactic subtotal colectomy.

The importance of screening colonoscopy was underscored by the report of Sankila et al, ⁶⁵ who compared the colorectal carcinoma death rates of two groups of HNPCC patients—those screened every 3 years and controls who refused screening. The authors’ data indicated that one carcinoma was prevented for every 2.8 polypectomies performed. This is in stark contrast to the National Polyp Study data for the general population, which gave a range of 41 to 119 polypectomies per carcinoma prevented.

The importance of screening is further emphasized by the study that compared screened with unscreened (by choice) “controls” evaluated over a 15-year period with 252 relatives at risk for HNPCC, 119 of whom declined screening. In the screened group, 8 of 133 (6%) developed colorectal carcinoma compared to 19 (16%) in the unscreened group. ⁷⁰ de Vos tot Nederveen Cappel et al ⁷¹ examined the stage of the screening detected carcinomas in relation to the surveillance interval and to assess the risk of developing colorectal carcinoma while on the program in 114 families in the Dutch Family Registry. A total of 35 carcinomas were detected while on the program. With intervals between colorectal carcinoma and the preceding surveillance examination of 2 years or less, carcinomas were at Dukes’ stage A (n = 4), B (n = 11), and C (n = 1). With intervals of more than 2 years, carcinomas were Dukes’ stage A (n = 3), B (n = 10), and C (n = 6). The 10-year cumulative risk of developing colorectal carcinoma was 10.5% in proven mutation carriers, 15.7% after partial colectomy, and 3.4% after subtotal colectomy. There is a substantial risk of developing colorectal carcinoma while on the program. However, all carcinomas but one of subjects who underwent a surveillance examination 2 years or less before detection were at a local stage. They therefore recommend surveillance for HNPCC with an interval of 2 years or less.

For women who are gene carriers, an annual endometrial aspiration curettage should begin at age 30 years. Those patients should be advised of the option of ovarian carcinoma screening with transvaginal ovarian ultrasonography, Doppler color blood flow imagery, and CA-125 evaluation. They should be encouraged to have their children early so they can consider the option of undergoing prophylactic total abdominal hysterectomy and bilateral salpingo-oophorectomy between the ages of 35 and 40 years.

Screening for breast and gastric carcinoma should be initiated earlier in life than might otherwise be considered. If at least one family member is affected, gastroscopy beginning at age 30 to 35 years and every 1 to 2 years is advised. ⁵³ Also, only if at least one family member is affected, ultrasonography of the urinary tract and urine cytology beginning at age 30 to 35 years and every 1 to 2 years is advised. ⁵⁶

Mecklin and Jarvinen ⁷² analyzed 22 Finnish HNPCC families followed up for 7 years. Metachronous neoplasms were diagnosed in 41% of patients treated by segmental resection and 24% of those treated by subtotal colectomy. Extracolonic carcinoma was diagnosed in 30%. The most common malignancy was biliopancreatic carcinoma, which accounted for all five deaths related to carcinoma. The authors concluded that subtotal colectomy is superior to hemicolectomy or segmental resection in patients with HNPCC and that regular follow-up is necessary for surveillance of remaining bowel and extracolonic malignancies.

Itoh et al ⁷³ conducted a study of 130 HNPCC kindreds to determine the risk of death from malignancy in first-degree relatives. The authors found a sevenfold increase in the risk of colon carcinoma in both sexes. In women relatives, the risk of breast carcinoma was increased fivefold, and a lifetime risk of breast carcinoma was 1 in 3.7. From these results, the authors recommended a screening program.

Because clinical premonitory signs such as multiple colonic polyps seen in cases of familial polyposis coli are lacking, the clinician must rely on the clinical findings and a family history typical of HNPCC. Disorders that have been considered in the differential diagnosis of HNPCC have included FAP, attenuated FAP, juvenile polyposis coli, and Peutz–Jeghers syndrome. ³⁵ Once the diagnosis has been made, the physician is faced with the responsibility of informing the patient that other family members are at risk because the genetic implications for carcinoma expression are profound. Identification of individuals at risk may be aided by inspection of the patient’s skin, which may provide clues to the existence of a carcinoma associated with genodermatosis. Criteria providing clues to the diagnosis of Lynch type II syndrome include any patient presenting with an early onset of carcinoma of the colon (particularly of the proximal colon in the absence of multiple polyps), endometrium, or ovaries (particularly when the patient is younger than age 40 years) ¹ ; any patient with multiple primary carcinomas in which the lesions are integral to Lynch type II syndrome (carcinoma of the colon, endometrium, or ovary and other adenocarcinomas) ² ; and any patient who states that one or more first-degree relatives manifested early onset of the carcinomas integral to Lynch type II syndrome. ³

It is important to determine whether an HNPCC syndrome is present because the operative procedure of choice for the index carcinoma is subtotal colectomy as opposed to a more limited resection. ^{35 , 69} In the case of a woman who has completed her family, the resection may be extended to a prophylactic hysterectomy and bilateral salpingo-oophorectomy because of the patient’s inordinately high risk of development of carcinoma of the endometrium and ovaries.

The lifetime risk of endometrial and ovarian carcinomas in HNPCC is up to 60 and 12%, respectively. ⁷⁴ Watson et al ⁷⁵ collected data on 80 ovarian carcinoma patients who are members of HNPCC families, including 31 known mutation carriers, 35 presentive carriers (by colorectal/endometrial carcinomas status), and 14 at-risk family members. Among frankly epithelial cases, most carcinomas were well or moderately differentiated. Ovarian carcinoma in HNPCC differs from ovarian in the general population in several clinically important respects. It occurs at a markedly earlier age (42.7 years). It is more likely to be epithelial (95.6%). If it is a frankly invasive epithelial carcinoma, it is more likely to be well or moderately differentiated (85% were International Federation of Obstetrics and Gynecology [FIGO] stage I or II at diagnosis). HNPCC patients with ovarian carcinoma are more likely to have a synchronous endometrial carcinoma (21.5%).

Familial Colorectal Carcinoma

Ten to fifteen percent of patients with colorectal carcinoma and/or colorectal adenomas have other affected family members, but their family histories do not fit the criteria for either FAP or HNPCC and may not appear to follow a recognizable pattern of inheritance, such as autosomal-dominant inheritance. Such families are categorized as having familial colorectal carcinoma. The presence of colorectal carcinoma in more than one family member may be due to genetic factors, shared environmental risk factors or even to chance. ¹⁰ With a family history, the risk of colorectal carcinoma is increased earlier in life than later such that at age 45 years the annual incidence is more than three times higher than average-risk people, whereas at age 70, risk is not significantly different. The incidence in a 35- to 40-year-old is about the same as that of an average-risk person at age 50 years. A personal history of adenomatous polyps confers a 15 to 20% risk of subsequently developing polyps. A history of adenomatous polyps in a sibling or parent is also associated with increased risk of colorectal carcinoma. Expert recommendations on screening for such persons are similar to those with a positive family history of colorectal carcinoma. Most experts suggest that screening should begin at age 35 to 40 years when the magnitude of risk is comparable to that of a 50-year-old. Because the risk increases with extent of family history, there is room for clinical judgment in favor of even earlier screening, depending on the details of the family history. A common but unproven clinical practice is to initiate colorectal carcinoma screening 10 years before the age of the youngest colorectal carcinoma case in the family.

Evidence demonstrates that patients in families with breast and colon carcinoma (hereditary breast and colon carcinoma) may have a carcinoma family syndrome caused by CHEK2 1100delC mutations that are present in a subset of the families. ⁷⁶ The CHEK2 mutation is incompletely penetrant.

Indications

Genetic testing is discussed when the patient’s family history of carcinoma falls in the autosomal-dominant mode of inheritance. Appropriate situations include families who carry a mutation in APC, hMLH1, hMSH2, hPMS1, hPMS2, or hMSH6 or families fulfilling the clinical FAP or ICG-HNPCC (International Collaborative Group on Hereditary Non-Polyposis Colorectal Cancer) criteria. The decision to undergo genetic testing is a personal one based on informed consent. The first phase in determining whether genetic testing is appropriate is genetic counseling.

Among FAP families, who are candidates for genetic testing? The diagnosis of FAP is made because of the presence of polyps on examination of the presenting patient (proband or index case). In autosomal-dominant inheritance, on average, 50% of the patient’s first-degree relatives (parents, siblings, and offspring) will be at risk. As soon as the FAP diagnosis emerges, either before or after operation, the surgeon should recommend that all first-degree relatives be examined. If a mutation in APC is identified in the patient’s germline DNA, the immediate family and other branches of the family can be encouraged to come forward for genetic counseling and to know their risk status if so desired. If a mutation is not detected, further genetic testing for APC mutations in the family is contraindicated and the clinical decision is to monitor first-degree relatives by colonoscopy at recommended intervals. Depending on the expression of colorectal carcinoma in the proband and the pattern of colorectal carcinoma and other malignancies in the family, genetic testing for HNPCC may be advisable. In every case of FAP, a genetic test to pinpoint the heritable, germline APC mutation is justified, with informed consent of the patient. As far as testing the patient’s family is concerned, a common exception (occurring in 30% of probands) is a patient with FAP whose parents are negative on endoscopic examination and who therefore has a “new mutation” in the APC gene. In this instance, the risk for colorectal carcinoma for siblings falls to that of the general population. Nonetheless, children of the patient will require testing (if a mutation was identified in the proband) if so desired, because they are at 50% risk for FAP. The only obvious example in which genetic testing is not recommended would be in an isolated patient with FAP with no living first-degree relatives.

What patients with suspected HNPCC should be referred for genetic counseling and testing? Because HNPCC is not always obvious based on clinical findings, either preoperatively or during resection, the clinical management of a possible HNPCC patient is complex. A detailed family history, with sites of malignancy verified, is critical to determine whether the patient fits the Amsterdam criteria and other associated features of HNPCC. Analysis of neoplastic material would provide assistance in determining whether to test the proband’s germline DNA (i.e., a blood sample) for HNPCC mutations (after genetic counseling) and refer the first-degree relatives should a mutation be detected. In the instance of a positive family history, it is ideal for testing to be conducted despite MSI phenotype. The clinical algorithm would be much the same as for FAP, with the exception that the multiple genes might need to be analyzed unless tests on pathogenic samples narrow the search. It is important to point out that at present, detection of a confirmed pathogenic mutation occurs in less than 50% of HNPCC families. In the absence of an identified deleterious mutation, genetic testing is inconclusive, which must be explained to the patient by an individual skilled in genetic counseling. Family members can be given general advice for colorectal carcinoma screening of first-degree relatives of patients with this malignancy.

Techniques

The molecular techniques used in testing for germline mutations in hereditary colorectal carcinoma have been described in detail by Rabelo et al ⁷⁸ and the following is a synopsis of that work.

Germline DNA can be tested in three ways:

1. 
   

   With in vitro–synthesized protein (IVSP) assay (also called protein truncation test or in vitro truncation test), specific gene segments are amplified by polymerase chain reaction (PCR) or reversed transcription (RT) PCR, transcribed and translated in vitro, and the protein product is analyzed by gel electrophoresis. If the amplified segment has a mutation that causes the production of a truncated protein, its smaller size will allow it to be more mobile in the gel.

   
   
2. 
   

   Single-stranded conformation analysis (SSCA) is also a mutation detection technique in which specific gene segments are amplified by PCR or RT-PCR, denatured to separate the stands of the DNA product and analyzed by gel electrophoresis. The mutant DNA often migrates differently under certain conditions, allowing the putative localization of mutation within the segment.

   
   
3. 
   

   DNA sequencing is the criterion standard technique for mutation detection, with up to 99% accuracy. It allows the precise identification of the mutation in the DNA sequence by pinpointing any change in the number or identity of bases.

Pathologic specimens are examined in two ways:

1. 
   

   A clue to identify colorectal carcinoma caused by HNPCC is by an associated alteration in the length stability of repetitive tract DNA in the carcinoma of an affected individual. This is known as MSI, also called RER-positive phenotype. The presence of MSI + thereby increases the likelihood that direct testing for defects in HNPCC genes will be informative. There are three outcomes of MSI analysis. ⁵⁶

   
   
   
   
   1. 
      

      If at least two markers tested show instability, the result indicates an increased likelihood that the colorectal carcinoma arose due to an HNPCC-associated mutation (MSI-H).

      
      
   2. 
      

      If no marker shows instability in the absence of a compelling family history, the result is usually interpreted as ruling out HNPCC (MSS stable).

      
      
   3. 
      

      If instability is seen in only one of the markers, the result does not indicate HNPCC. The clinical behavior of MSI-L carcinomas is identical to those with MSS, and does not have the mutational fingerprints of the clinical behavior of MSI-H carcinomas. MSI-L results occur as commonly as do MSI-H results.

   
   
2. 
   

   Immunohistochemistry is a technique for diagnosing a carcinoma that lacks expression of a particular HNPCC gene. It examines either hMLH1 or hMSH2 protein expression in carcinomas by immunostaining using monoclonal antibodies. Formalin-fixed, paraffin-embedded adenocarcinoma tissue is the sample required for testing with a region of normal mucosa adjacent to the carcinoma for comparison. The absence of hMLH1 or hMSH2 immunostaining shows which gene is defective and should be analyzed for the presence of a germline mutation. Studies have shown that the lack of hMLH1 or hMLH2 immunostaining was associated with the presence of MSI+.

Genetic Testing for Familial Adenomatous Polyposis

Presymptomatic genetic diagnosis of FAP in at-risk individuals has been feasible with linkage and direct detection of APC mutations. ¹⁰ If one were to use linkage analysis to identify gene carriers, ancillary family members, including more than one affected individual, would need to be studied. With direct detection, fewer family members’ blood samples are required than for linkage analysis, but the specific mutation must be identified in at least one affected person by DNA mutation analysis or sequencing. Because approximately 96% of the mutations in FAP lead to a truncated protein, it has become routine to use IVSP assay (sometimes called protein truncation test) for mutation detection. When a truncated protein is identified, it is possible to localize the mutation to a specific segment of the gene and then use DNA sequencing to determine the mutated nucleotide(s). The use of IVSP as the sole genetic test in FAP misses approximately 20% of APC mutations. Another screening technique is based on analysis of electrophoretic migration of small segments of the wild-type mutant gene (SSCA). The sequential use of two molecular diagnostic tests has become a common practice: a simple and less expensive screening technique (of high sensitivity and moderate specificity), followed by a definitive test of high sensitivity, usually DNA sequencing.

These mutation search methods, APC protein truncation testing (from lymphocyte RNA), considerably enhance the feasibility of testing at-risk individuals without requiring DNA from multiple affected family members (as linkage requires). In particular, it is useful for testing in small families or in patients with spontaneous or “de novo” mutations (the first occurrence of FAP in a kindred), which may account for as much as one third of incident cases. ¹⁰ Only about 80% of APC mutations can be detected by this method. Therefore, the clinician cannot rule out FAP on the basis of this molecular test alone if other criteria support the FAP diagnosis. When the mutation in a family is known from one or more affected members, one test, usually IVSP, need be performed, as the expression pattern of the mutation is already established for that family. A positive result in such a test is considered a “mutation-positive” result, and the patient can be counseled as recommended. When the mutation in the family is not known, IVSP is performed. In the majority of cases, a mutation will be found in the APC gene that can be further characterized by DNA sequencing. Again, a “mutation-positive” result is obtained leading to the appropriate genetic counseling for the patient. Other options include DNA sequencing and linkage analysis. Even by combining two or more techniques, it is not possible to achieve 100% sensitivity, because the mutations may not be in the coding region of the gene or because a few FAP kindreds do not exhibit linkage to chromosome 5q. In such case, a “no mutation detected” result must not be interpreted as a “negative test” with the very important considerations for counseling the patient.

For at-risk individuals who have been found to be definitively mutation negative by genetic testing, there is no clear consensus on the need for or frequency of colon screening, though it would seem prudent that at least one flexible sigmoidoscopy or colonoscopy examination should be performed in early adulthood (age 18–25 years). ¹⁰

Molecular Genetic Diagnosis of Hereditary Nonpolyposis Colorectal Carcinoma

The HNPCC syndrome is more complex in terms of genetic testing because at least six genes involved in DNA mismatch repair predispose to this syndrome. The two most commonly involved genes are hMSH2 (involved in approximately 45% of cases) and hMLH1 (involved in approximately 49% of cases). Protein truncation is observed in more than 80% of HNPCC cases caused by mutations in hMSH2 and to a lesser degree in hMHL1. Currently, IVSP or SSCA or both are used, followed by DNA sequencing for precise characterization. In the case of a mutation already known in an affected member of the kindred, the same technique used to detect that mutation can be used, and a “mutation-positive” result leads to genetic counseling. If no mutation has been previously described in the kindred, analysis of hMLH1 and hMSH2 in an affected individual by IVSP or SSCA followed by DNA sequencing for confirmation is indicated. If a truncating mutation is found, a causative role can be inferred for the mutation. This “mutation-positive” result allows counseling of the affected individual and further testing of the kindred. If a nontruncating mutation is found, tests such as linkage analysis and functional assays may be required to discriminate between an “inconclusive” and a “mutation-positive” result. It is estimated that approximately 30% of mutations will be missed if IVSP is used as the sole genetic test and that a similar or higher percentage of false negatives may also be expected if SSCA is used instead. Again, even by combining two or more techniques, it is not possible to achieve 100% sensitivity. Other genes may be responsible (locus heterogeneity). A “no mutation detected” result must not be interpreted as a “negative test” result.

Summary

A proposed algorithm for genetic testing of individuals in FAP and HNPCC kindreds is given below. First, use IVSP to locate the mutation, then use DNA sequencing to characterize it precisely in the first affected family member. If the IVSP test is negative, DNA sequencing of the APC gene is warranted. When the mutation is defined for that kindred, the testing of other family members at risk may be accomplished by using IVSP alone. If the mutation in the first affected individual was identified by DNA sequencing, it is logical to continue to use this method for genetic testing of at-risk individuals. Considering the weighty psychological, ethical, and legal implications of the results of genetic tests, the use of two different techniques, one to detect the mutation and the second to identify and confirm it, is recommended in all cases. DNA sequencing, the most sensitive method for mutation detection, should be one of the techniques used. For HNPCC, two techniques are appropriate for initial testing of mutations in these genes, IVSP and SSCA. Screening with IVSP is recommended, but this preference may vary among laboratories. If a mutation is detected by any of these methods, DNA sequencing should be used to identify it precisely in the first affected member of a kindred. If these screening techniques are negative, the sequencing of both hMSH2 and hMLH1 may be warranted. If still negative, in view of the locus heterogeneity seen in HNPCC, the same screening protocol may have to be used for hPMS2, hPMS1, and hMSH6. Again, the use of two different tests in all cases is recommended and DNA sequencing is used and when the pathogenic mutation is defined for that kindred, the testing of other family members at risk may be accomplished with either IVSP or SSCA alone.

When a mutation (a variant gene) is identified by any of the methods, it must be validated as a pathogenic mutation (true-positive result) instead of a polymorphism, a common variant in the protein that does not contribute to the disease phenotype (false positive). When a mutation is not identified (“no mutation detected”), the negative result may be a “true negative” (no predisposing mutations in any relevant genes), or a “false negative” (undetected mutation may exist in known or unknown genes). If the person is affected with colorectal carcinoma, this is either a false negative in a familial colorectal carcinoma syndrome or a sporadic case. This “no mutation detected” result is the one most likely to be misinterpreted, leading to inappropriate genetic counseling. The use of two genetic tests decreases the rate of false-negative results. DNA sequencing is the most sensitive technique for this purpose and its use is advocated whenever a negative result is obtained by IVSP or SSCP for affected members with a strong family history. Even if the APC gene or all five DNA mismatch repair genes are fully sequenced, there will still be a residue of false-negative results caused by mutations in the noncoding regions of those genes, locus heterogeneity, or in still-unidentified genes.

A “negative test” in patients at risk for FAP or HNPCC will rule out these disorders only if a mutation has been identified in an affected family member. If tests to identify a mutation in an affected family member are negative, a “no mutation detected” result should be considered as noninformative or inconclusive (as if no test had been performed). The correct interpretation of a negative result in mutation analysis of hereditary colorectal carcinoma syndromes is mandatory to avoid adverse outcomes, because a false-negative result may lead to lack of appropriate endoscopic surveillance. Molecular genetic testing must be coupled with appropriate genetic counseling.

Interpretation of Genetic Test Results

Impact of Genetic Testing

Individuals need to know the implications and possible impact of genetic test results before testing. An ambiguous test result may be more distressing than a positive test result. Moreover, genetic testing often yields results that are probabilistic. The psychological burden of knowing that one is at high risk for developing a neoplasm may outweigh the possible benefits from intervention. Results of a study by Keller et al ⁸¹ suggest that expressed intention and attitude toward genetic testing do not reliably predict actual uptake of counseling or testing. The actual uptake of genetic testing for HNPCC in a clinical sample of 140 patients fulfilling clinical criteria of HNPCC was 26%. Some 60% of participants experienced pronounced distress related to their potential inheritance of the disorder compared to 35% among nonparticipants. Distress reached a clinically significant level in 28% of participants. Restricted communication within the family was observed frequently. Irrespective of groups, a positive attitude toward obtaining a gene test result predominated. Therefore, the benefits and risks of participating in genetic tests must be described before testing is begun. Undergoing early detection of colorectal carcinoma or prophylactic operation or both can save lives. Surveillance or operation can be targeted to specific sites that are more prone to neoplasia, depending on the identity of the gene. For instance, female members of HNPCC families have a 10-fold increased risk for developing endometrial carcinoma, with an age of onset 10 to 15 years earlier than the same disease in the general population. Options for intervention and early detection and provision of knowledge to other family members are common reasons why individuals seek testing for genes predisposing to carcinoma of the colon and rectum.

Gritz et al ⁸² examined the impact of HNPCC genetic test results on psychological outcomes among carcinoma-affected and carcinoma-unaffected participants up to 1 year after results disclosure. A total of 155 persons completed study measures before HNPCC genetic testing, and 2 weeks and 6 and 12 months after disclosure of test results. Mean scores on all outcome measures remained stable and within normal limits for carcinoma-affected participants, regardless of mutation status. Among unaffected carriers of HNPCC-predisposing mutations, mean depression, state anxiety, and carcinoma worries scores increased from baseline to 2 weeks post disclosure and decreased from 2 weeks to 6 months post disclosure. Among unaffected noncarriers, mean depression and anxiety scores did not differ, but carcinoma worries scores decreased during the same time period. Affected and unaffected carriers had higher mean test-specific distress scores at 2 weeks post disclosure compared with noncarriers in their respective groups; scores decreased for affected carriers and all unaffected participants from 2 weeks to 12 months post disclosure. Classification of participants into high-versus low-distress clusters using mean scores on baseline psychological measures predicted significantly higher or lower follow-up scores, respectively, on depression, state anxiety, quality of life, and test-specific distress measures, regardless of mutation status. Although HNPCC genetic testing does not result in long-term adverse psychological outcomes, unaffected mutation carriers may experience increased distress during the immediate postdisclosure time period. Furthermore, those with higher levels of baseline mood disturbance, lower quality of life, and lower social support may be at risk for both short- and long-term increased distress.

Examining a person’s DNA has the potential for far-reaching consequences beyond the person being tested. Genetic testing can reveal information about relatives with whom we share a common genetic legacy. Despite the possible benefits of testing for susceptibility for malignancy, the uncertainty and limitations of current efforts to lower mortality or morbidity contribute to the psychosocial and ethical complexities. Genetic information must remain confidential. However, breach of confidentiality has been considered permissible if all of the following conditions are met: reasonable effort to encourage disclosure has failed, harm is likely to occur if the information is withheld, the harm is serious and avoidable, and the disease is treatable or preventable. ⁸³ Conflict between issues of privacy and duty to forewarn family members arises because genetic information is, at once, individual and familial. Several key issues may be raised in genetic testing. First, at times genetic test results need to be interpreted in the context of the results of other family members. Second, family members should be independently and autonomously counseled. Third, the decision of each member regarding testing should be followed without coercion by other family members. Fourth, genetic test results have familial implications whether results are disclosed to other family members or not. ⁸⁴

To provide information useful in the education and counseling in individuals considering genetic testing, Lerman et al ⁸⁴ conducted structured interviews with 45 first-degree relatives of colorectal carcinoma patients. Fifty-one percent of respondents indicated that they definitely would want to obtain a genetic test for colon carcinoma susceptibility when it is available. Motivation for genetic testing included the following: to know if more screening tests are needed, to learn if one’s children are at risk, and to be reassured. Barriers to testing included concerns about insurance, test accuracy, and how one’s family would react emotionally. Most participants anticipated that they would become depressed and anxious if they tested positive for a mutation; many would feel guilty and still worry if they tested negative. These preliminary results underscore the importance of the potential risks, benefits, and limitations of genetic testing with particular emphasis on the possibility of adverse psychological effects and implications for health insurance. Ethical issues of revealing or concealing genetic information will be debated long into the future because the potentially enormous consequences must be carefully considered. Hadley et al ⁸⁵ assessed the impact of genetic counseling and testing on the use of endoscopic screening procedures and adherence to recommended endoscopic screening guidelines in 56 symptomatic at-risk individuals in families known to carry an HNPCC mutation. They analyzed data on colonoscopy and flexible sigmoidoscopy screenings collected before genetic counseling and testing and 6 and 12 months postgenetic counseling and testing on 17 mutation-positive and 39 true-negative mutation individuals. Among mutation-negative individuals, use of colonoscopy and flexible sigmoidoscopy decreased significantly between pre- and postgenetic counseling and testing. Among mutation-positive individuals, a nonsignificant increase in use was noted. Age was also associated with use of endoscopic screening after genetic counseling and testing. More mutation-negative individuals strictly adhered to guidelines than did mutation-positive individuals (87 vs. 65%). They concluded genetic counseling and testing for HNPCC significantly influences the use of colonic endoscopy and adherence to recommendations for colon carcinoma screening.

Patients considering genetic testing need to be informed about the potential for genetic discrimination. This is of great concern for asymptomatic family members, because disclosure of a positive mutation result to third parties could be used to limit, raise rates for, or deny access to individual health insurance. However, nondisclosure could nullify a patient’s insurance contract.

A survey of medical directors of United States life insurance companies indicated that familial colon or breast carcinoma constituted sufficient grounds to deny insurance (1 of the 27) or charge higher premiums (6 of the 27), despite not having actuarial data to support underwriting (calculating premiums) guidelines pertaining to these carcinomas.

In the United States, several states have safeguards against the potential misuse of genetic information. The potential for misuse of genetic information has led to federal initiatives in the United States to regulate the use of genetic information. In the context of insurance and employment, it is unclear whether the Americans with Disabilities Act of 1990 (ADA) will provide adequate protection to carriers of carcinoma-susceptibility genes, because the Equal Employment Opportunity Commission does not view a person with genetic predisposition for a disease or a carrier of a gene for a late-onset disorder as having a disability. This implies that these persons are not protected under the ADA. Legislation has been introduced to extend the definition of disability to “genetic or medically identified potential of or predisposition toward a physical or mental impairment that substantially limits a major life activity.” For now, the ADA has been interpreted to offer protection only to those who develop symptoms.

Health Insurance Portability and Accountability Act (Kennedy–Kassebaum bill) came into effect in the summer of 1997. It defined genetic information as part of an individual’s health status. This act was implemented to prohibit employers and insurers from excluding individuals in a group from coverage or charging higher premiums on the basis of health status. Also in 1997, the Genetic Information Nondiscrimination in Health Insurance Act (Slaughter–Snowe bill) called for a ban of the use of “genetic information” in denying or setting rates for group health insurance. In this act, genetic information was defined broadly to include genetic tests as well as information about inherited features. This bill called for a limit on the collection or disclosure of genetic information without the written consent of the individual.

Fourteen states have enacted legislation to provide safeguards and to protect misuse of genetic information by insurers and employers. Other states have followed suit, and amendments are being made to those existing laws in some states, as reviewed by Offit. ⁸⁶

In summary, individuals undergoing genetic testing should understand the advantages and disadvantages of receiving and sharing genetic results. Attempting to protect the individual from genetic discrimination by disseminating results to only a few selected medical staff could impede care. So far, there has been no substantial use of genetic information by health insurers, and laws restricting its use have thus far maintained social fairness in the area of health insurance. When integrated with existing testing protocols for colorectal carcinoma and when applied with appropriate caveats particularly regarding interpretation of negative results, genetic testing can result in improved management of patients and families.