FAP Genetic Testing
The policy for FAP genetic testing is based on a 1998 Blue Cross Blue Shield Association (BCBSA) Technology Evaluation Center (TEC) Assessment, which offered the following conclusions:
- Genetic testing for familial adenomatous polyposis (FAP) may improve health outcomes by identifying which currently unaffected at-risk family members require intense surveillance or prophylactic colectomy.
- At-risk subjects are considered to be those with greater than 10 adenomatous polyps; or close relatives of patients with clinically diagnosed FAP or of patients with an identified APC mutation.
- The optimal testing strategy is to define the specific genetic mutation in an affected family member and then test the unaffected family members to see if they have inherited the same mutation.
The additional policy information on attenuated FAP (AFAP) and on MUTYH-associated polyposis (MAP) diagnostic criteria and genetic testing is based on information from GeneReviews (Burt et al., 2008) and from several publications (Kastrinos, Syngal, 2007; Lefevre et al., 2009; Avezzu et al., 2008; Balaguer et al., 2007) that build on prior, cited research. In addition, GeneReviews summarizes clinical FAP genotype-phenotype correlations that could be used to determine different patient management strategies. The authors of the review conclude, however, that there is not yet agreement about using such correlations to direct management choices.
The National Comprehensive Cancer Network (NCCN) recommends APC genetic testing in a proband (index case), if possible, to confirm a diagnosis of FAP and allow for mutation-specific testing in other family members.
Lynch Syndrome Genetic Testing
The policy for Lynch syndrome is based on an evidence report published by the Agency for Healthcare Research and Quality (AHRQ) (Bonis et al.), a supplemental assessment to that report contracted by the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group (Palomaki et al., 2009), and an EGAPP recommendation for genetic testing in colorectal cancer (Group EoGAiPaPEW, 2009). Based on the AHRQ report and supplemental assessment, the EGAPP recommendation came to the following conclusions regarding genetic testing for MMR mutations in patients already diagnosed with colorectal cancer:
- Family history, while important information to elicit and consider in each case, has poor sensitivity and specificity as a screening test to determine who should be considered for MMR mutation testing and should not be used as a sole determinant or screening test.
- MSI [microsatellite instability] and IHC [immunohistochemical] screening tests for MMR mutations have similar sensitivity and specificity. MSI screening has a sensitivity of about 89% for MLH1 and MSH2 and 77% for MSH6 and a specificity of about 90% for all. It is likely that, using high-quality MSI testing methods, these parameters can be improved. IHC screening has a sensitivity for MLH1, MSH2, and MSH6 of about 83% and a specificity of about 90% for all.
- Optional BRAF testing can be used to reduce the number of patients, who are negative for MLH1 expression by IHC, needing MLH1 gene sequencing, thus improving efficiency without reducing sensitivity for MMR mutations.
- A chain of indirect evidence can be constructed for the clinical utility of testing all patients with colorectal cancer for MMR mutations.
- The chain of indirect evidence from well-designed experimental nonrandomized studies (as noted below) is adequate to demonstrate the clinical utility of testing unaffected (without cancer) first- and second-degree relatives of patients with Lynch syndrome who have a known MMR mutation.
- Seven studies examined how counseling affected testing and surveillance choices among unaffected family members of Lynch syndrome patients. About half of relatives received counseling, and 95% of these chose MMR gene mutation testing. Among those positive for MMR gene mutations, uptake of colonoscopic surveillance beginning at age 20–25 years was high at 53–100%.
- One long-term, nonrandomized controlled study and one cohort study of Lynch syndrome family members found significant reductions in colorectal cancer among those who followed recommended colonic surveillance vs. those who did not.
- Surveillance, prevention for other Lynch syndrome cancers (for detail, refer to last outline bullet).
- The chain of evidence from descriptive studies and expert opinion (as noted below) is inadequate (inconclusive) to demonstrate the clinical utility of testing the probands with Lynch syndrome (i.e., cancer index patient).
- Subtotal colectomy is recommended as an alternative to segmental resection, but has not been shown superior in follow-up studies.
- Although a small body of evidence suggests that MSI-positive tumors are resistant to 5-fluorouracil and more sensitive to irinotecan than MSI-negative tumors, no alteration in therapy according to MSI status has yet been recommended.
- Surveillance, prevention for other Lynch syndrome cancers:
i. While invasive and not actively recommended, women may choose hysterectomy with salpingo-oophorectomy to prevent gynecologic cancer. In one retrospective study, women who chose this option had no gynecologic cancer over 10 years, whereas about one-third of women who did not have surgery developed endometrial cancer, and 5.5% developed ovarian cancer
ii. In one study, surveillance endometrial biopsy detected endometrial cancer and potentially precancerous conditions at earlier stages in those with Lynch syndrome, but results were not statistically significant, and a survival benefit has yet to be shown. Transvaginal ultrasound (TVUS) is not a highly effective surveillance mechanism for endometrial cancer in patients with Lynch syndrome; however, TVUS in conjunction with endometrial biopsy has been recommended for surveillance.
iii. Gastroduodenoscopy for gastric cancer surveillance and urine cytology for urinary tract cancer surveillance are recommended based on expert opinion only, in the absence of adequate supportive evidence.
Based on an indirect chain of evidence with adequate evidence of benefit to unaffected family members found to have Lynch syndrome, the EGAPP working group recommended testing all patients with colorectal cancer for MMR gene mutations.
In addition to DNA mismatch repair (MMR) gene mutation testing, evidence now supports testing for EPCAM deletions in particular cases where all MMR gene mutation testing is negative, but tumor MSH2 IHC indicates lack of expression, and tumor MSI testing shows a high level of instability. EPCAM is found just upstream, in a transcriptional sense, of MSH2. Deletions of EPCAM that encompass the last two exons of the EPCAM gene including the polyadenylation signal that normally ends transcription of DNA into messenger RNA result in transcriptional ‘read-through’ and subsequent hypermethylation of the nearby and downstream MSH2 promoter. This hypermethylation prevents normal MSH2 protein expression and leads to Lynch syndrome in a fashion similar to Lynch cases in which an MSH2 mutation prevents MSH2 gene expression. Several studies have characterized such EPCAM deletions, established their correlation with the presence of EPCAM-MSH2 fusion messenger RNAs (apparently non-functional) and with the presence of MSH2 promoter hypermethylation, and, most importantly, have shown the co-segregation of these EPCAM mutations with Lynch-like disease in families (Niessen et al., 2009; Kloor et al., 2011; Kuiper et al., 2011; Kovacs et al., 2009; Ligtenberg et al., 2009; Rumilla et al., 2011). Because studies differ slightly in how patients were selected, prevalence of these EPCAM mutations is difficult to estimate but may be in the range of 20-40% of patients/families who meet Lynch syndrome criteria, do not have a MMR mutation, but have MSI-high tumor tissue. Kempers et al. reported that carriers of an EPCAM deletion had a 75% (95% confidence interval [CI], 65–85) cumulative risk of colorectal cancer by age 70 years, not significantly different from that of carriers of an MSH2 deletion (77%, 64–90); mean age at diagnosis was 43 years. However, the cumulative risk of endometrial cancer was low at 12% (95% CI, 0–27) by age 70, compared to carriers of a mutation in MSH2 (51% [95% CI, 33–69], p=0.0006) (Kempers et al., 2011).
Although MMR gene sequencing of all patients is the most sensitive strategy, it is highly inefficient and cost-ineffective and not recommended. Rather, a screening strategy of MSI or IHC testing (with or without optional BRAF testing) is recommended and retains a relatively high sensitivity. Some evidence suggests that IHC requires particular training and experience (Overbeek et al., 2008). Although a particular strategy was not recommended by the EGAPP Working Group, several are potentially effective; efficiency and cost-effectiveness may depend upon local factors.
Previous recommendations have used family history as an initial screen to determine who should proceed further to MMR laboratory testing. Recent studies have shown that limiting laboratory testing to patients who met even the more sensitive Revised Bethesda criteria (i.e., compared to the Amsterdam II criteria) would miss as much as 28% of Lynch syndrome cases (Hampel et al., 2008; Canard et al., 2011). Family history is important for counseling families, but based on this and similar evidence, is not recommended as an initial screening tool to make decisions about testing patients who already have colorectal cancer. However, the Amsterdam II or Revised Bethesda criteria may be used in identifying those without colorectal cancer who might be tested.
Limiting testing for Lynch Syndrome on the basis of age (e.g., test only patients younger than age 50 years) is also not recommended. For example, Hampel et al. found that among 18 Lynch syndrome patients discovered among 500 unselected colorectal cancer patients, only 8 (44%) patients were diagnosed at age younger than 50 years. Similarly, Canard et al. reported that restricting screening to patients younger than 50 years would have missed about half of patients eventually found to have Lynch Syndrome. Another group screened colorectal cancer patients who were younger than age 60 and identified 98 likely (MSI positive, BRAF negative) Lynch syndrome cases; of these, 47% were between ages 50 and 60 (Schofield et al., 2009). A large study of Lynch syndrome family studies found that the cumulative risk of colorectal cancer in MMR mutation carriers was only 13% (95% CI, 9-19) by age 50, but 35% (95% CI, 25-49%) by age 70 (Bonadona et al., 2011). For MSH6 mutation carriers, however, colorectal cancer risks do not appear to increase until after age 60.
The estimated risk of stomach cancer in a large study of Lynch syndrome families was 6% (95% CI, 0.2-17%) for carriers of MLH1 mutations and warrants further study to address the utility of gastric surveillance (Bonadona et al., 2011).
As the EGAPP recommendations noted, the evidence to date is limited to clearly support benefit from genetic testing to the index patient with colorectal cancer if found to have Lynch syndrome. However, professional societies have reviewed the evidence and concluded that genetic testing likely has direct benefits for at least some patients with colorectal cancer and Lynch syndrome on the basis of differing recommendations for post-surgical surveillance, and for those who choose prophylactic surgical treatment instead of surveillance.
In the absence of preventive surgery, heightened surveillance is recommended. The NCCN guidelines for colon cancer and for colorectal cancer screening recommend post-surgical colonoscopy at one year and, if normal, again in 2-3 years, then every 3-5 years based on findings. However, for Lynch syndrome patients, colonoscopy is recommended every 1-2 years throughout life based on the high likelihood of cancer for patients diagnosed with Lynch syndrome prior to a cancer diagnosis, and on the high likelihood of a second primary cancer in those diagnosed with Lynch syndrome based on a first cancer diagnosis (deVos et al., 2002). If the patient is not a candidate for routine surveillance, subtotal colectomy may be considered (NCCN, 2011).
Early documentation of the natural history of colorectal cancer in highly selected families with a strong history of hereditary colorectal cancer indicated risks of synchronous and metachronous cancers as high as 18% and 24%, respectively, in patients who already had colorectal cancer (Fitzgibbons et al., 1987). As a result, in 1996, the Cancer Genetic Studies Consortium, a temporary NIH (National Institutes of Health)-appointed body, recommended that if colorectal cancer is diagnosed in patients with an identified mutation or a strong family history, a subtotal colectomy with ileorectal anastomosis (IRA) should be considered in preference to segmental resection (Burke et al.,1997). Although the average risk of a second primary is now estimated to be somewhat lower overall (see Description) in patients with Lynch syndrome and colorectal cancer, effective prevention measures remain imperative. One study suggested that subtotal colectomy with IRA markedly reduced the incidence of second surgery for metachronous cancer from 28% to 6% but could not rule out the impact of surveillance (Van Dalen et al., 2003). A mathematical model comparing total colectomy and IRA to hemicolectomy resulted in increased life expectancies of 2.3, 1, and 0.3 years for ages 27, 47, and 67, respectively; for Duke’s A, life expectancies for the same ages are 3.4, 1.5, and 0.4, respectively (de Vos et al., 2003). Based on this work, the joint American Society of Clinical Oncology (ASCO) and Society of Surgical Oncology (SSO) review of risk-reducing surgery in hereditary cancers recommends offering both options to the patient with Lynch syndrome and colorectal cancer, especially those who are younger (Guillem et al., 2006). This ASCO/SSO review also recommends offering Lynch syndrome patients with an index rectal cancer the options of total proctocolectomy with ileal pouch anal anastomosis or anterior proctosigmoidectomy with primary reconstruction. The rationale for total proctocolectomy is the 17% to 45% rate of metachronous colon cancer in the remaining colon after an index rectal cancer in Lynch syndrome patients.
The ASCO/SSO review recommends offering prophylactic total abdominal hysterectomy to female patients with colorectal cancer who have completed childbearing or to women undergoing abdominal surgery for other conditions, especially when there is a family history of endometrial cancer. This recommendation is based on the high rate of endometrial cancer in mutation-positive individuals and the lack of efficacy of screening. A recent study estimated the risk of endometrial cancer in mutation carriers at 34% (95% CI, 17-60%) by age 70, and of ovarian cancer 8% (95% CI, 2-39%) by age 70 (Bonadona et al., 2011). Risks do not appear to appreciably increase until after age 40. When surgery is chosen, oophorectomy should also be performed because of the risk of ovarian cancer in Lynch syndrome. As already noted, in one retrospective study, women who chose this option had no gynecologic cancer over 10 years whereas about one-third of women who did not have surgery developed endometrial cancer, and 5.5% developed ovarian cancer (Schmeler et al., 2006). In another retrospective cohort study, hysterectomy improved survival among female colon cancer survivors with Lynch syndrome (Obermair et al., 2010). This study also estimated that for every 100 women diagnosed with Lynch syndrome-associated colorectal cancer, about 23 will be diagnosed with endometrial cancer within 10 years absent a hysterectomy. Recent data on mutation-specific risks suggests that prophylactic gynecological surgery benefits for carriers of MSH6 mutations may offer less obvious benefits compared to harms as lifetime risk of endometrial cancer is lower than for carriers of MLH1 or MSH2 mutations, and lifetime risk of ovarian cancer is similar to the risk for the general population (Bonadona et al., 2011). An alternative to prophylactic surgery is surveillance for endometrial cancer using transvaginal ultrasound (TVUS) and endometrial biopsy. Evidence indicates that such surveillance significantly reduces the risk of interval cancers, but no evidence as yet indicates surveillance reduces mortality due to endometrial cancer (Auranen et al., 2011) Surveillance in Lynch syndrome populations for ovarian cancer has not yet been demonstrated successful at improving survival (Auranen et al., 2011).
The European Society for Medical Oncology (ESMO) published clinical practice guidelines for familial colorectal cancer risk in 2010 (Balmana et al., 2010). These guidelines addressed Lynch Syndrome, familial adenomatous polyposis, and MAP. No specific recommendations were made regarding how to initially identify Lynch syndrome cases; several methods, including clinical criteria and universal screening of all colorectal cancer cases, were mentioned. Other ESMO recommendations are consistent with the information in this policy.
NCCN guideline for colorectal cancer screening notes that screening of all colorectal and endometrial cancers for Lynch syndrome mutations has been implemented at some centers and does not recommend for or against this practice. The guideline does not specifically mention EPCAM deletion testing but does indicate that individuals with loss of MSH2 and/or MSH6 protein expression by IHC, regardless of germline MMR mutation status, should be followed as though they have Lynch syndrome. These guidelines also address FAP, AFAP, and MAP, consistent with the information in this policy.
Results of testing for the APC mutation in individuals with a family history of FAP, or a known APC mutation in the family, lead to changes in surveillance and prophylactic treatment. For patients with a positive result, enhanced surveillance and/or prophylactic treatment will reduce the future incidence of colon cancer and improve health outcomes. Therefore APC testing may be medically necessary for patients with a family history of FAP or a known APC mutation in the family. A related familial polyposis syndrome, MAP syndrome, is associated with mutations in the MUTYH gene. Testing for this genetic mutation may be medically necessary when the differential diagnosis includes both FAP and MAP, since distinguishing between the two leads to different management strategies. In some cases, Lynch syndrome may be part of the same differential diagnosis, depending on presentation.
A substantial portion of patients with colorectal cancer will be found to have Lynch syndrome, which is associated with mutations in the MMR gene. A positive genetic test for the MMR mutation can lead to enhanced surveillance, changes in recommendations about treatment options, and possible prophylactic treatment for other Lynch syndrome malignancies., Therefore, testing for Lynch syndrome may be medically necessary in patients with newly diagnosed colorectal cancer and in patients at high risk for Lynch syndrome, defined by meeting the clinical criteria such as Amsterdam II or Revised Bethesda. The EPCAM mutation is less common than MMR mutations as a cause of Lynch syndrome, and should be part of the diagnostic testing for Lynch syndrome in patients who are negative for all MMR mutations but who screen positive for MSI and lack MSH2 IHC evidence of protein expression.
Disclaimer for coding information on Medical Policies
Procedure and diagnosis codes on Medical Policy documents are included only as a general reference tool for each policy. They may not be all-inclusive.
The presence or absence of procedure, service, supply, device or diagnosis codes in a Medical Policy document has no relevance for determination of benefit coverage for members or reimbursement for providers. Only the written coverage position in a medical policy should be used for such determinations.
Benefit coverage determinations based on written Medical Policy coverage positions must include review of the member’s benefit contract or Summary Plan Description (SPD) for defined coverage vs. non-coverage, benefit exclusions, and benefit limitations such as dollar or duration caps.