Apc-Associated Polyposis Conditions

Clinical characteristics.

APC-associated polyposis conditions include: familial adenomatous polyposis (FAP), attenuated FAP, and gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS).

• FAP is a colon cancer predisposition syndrome in which hundreds to thousands of adenomatous colonic polyps develop, beginning, on average, at age 16 years (range 7-36 years). By age 35 years, 95% of individuals with FAP have polyps; without colectomy, colon cancer is inevitable. The mean age of colon cancer diagnosis in untreated individuals is 39 years (range 34-43 years). Extracolonic manifestations are variably present and include: polyps of the gastric fundus and duodenum, osteomas, dental anomalies, congenital hypertrophy of the retinal pigment epithelium (CHRPE), soft tissue tumors, desmoid tumors, and associated cancers.
• Attenuated FAP is characterized by multiple colonic polyps (average of 30), more proximally located polyps, and a diagnosis of colon cancer at a later age than in FAP. Certain extracolonic manifestations, such as gastric and duodenal polyps or cancers, are variably present in attenuated FAP; risk management may be substantially different between FAP and attenuated FAP.
• GAPPS is characterized by gastric fundic gland polyposis, increased risk of gastric cancer, and limited colonic involvement in most individuals reported.

Diagnosis/testing.

The diagnosis of an APC-associated polyposis condition is now established by molecular genetic testing. Typically, it is suspected in an individual with suggestive personal and/or family history features and confirmed by identification of a heterozygous germline pathogenic variant in APC.

Management.

Treatment of manifestations: There is an absolute indication for colectomy when colorectal cancer is diagnosed or suspected, or when there are significant symptoms (bleeding/obstruction). Relative indications for colectomy include presence of multiple adenomas larger than 6 mm that cannot be reasonably removed endoscopically, a significant increase in adenoma number between surveillance exams, presence of adenomas with high-grade dysplasia, or inability to adequately survey the colon (e.g., due to innumerous diminutive adenomas or limited access to/compliance with colonoscopy). Several studies have shown that nonsteroidal anti-inflammatory drugs (NSAIDs) have caused regression of adenomas in FAP and decreased the polyp burden, though there are currently no FDA-approved chemopreventive agents for FAP. Endoscopic or surgical removal of duodenal adenomas is considered if polyps exhibit villous change or severe dysplasia, exceed one centimeter in diameter, or cause symptoms. Osteomas may be removed for cosmetic reasons. Desmoid tumors may be surgically excised or treated with NSAIDs, anti-estrogens, cytotoxic chemotherapy, and/or radiation.

Surveillance: Colorectal screening beginning at age ten to 12 years for FAP and in late teens for attenuated FAP; esophagogastroduodenoscopy by age 20-30 years or prior to colon surgery; regular physical examinations including thyroid palpation, neurologic examination, and abdominal examination (for desmoids); consider annual thyroid ultrasound imaging for increased risk of thyroid cancer. Consider screening for hepatoblastoma by liver ultrasound and measurement of serum alpha-fetoprotein concentration (until age 5 years). The efficacy of screening for gastric cancer or prophylactic gastrectomy is currently unknown for individuals with GAPPS.

Evaluation of relatives at risk: Molecular genetic testing for early identification of at-risk family members improves diagnostic certainty and reduces the need for costly screening procedures in those at-risk family members who have not inherited the pathogenic variant.

Genetic counseling.

APC-associated polyposis conditions are inherited in an autosomal dominant manner. Approximately 75%-80% of individuals with an APC-associated polyposis condition have an affected parent. Offspring of an affected individual are at a 50% risk of inheriting the pathogenic variant in APC. Prenatal testing and preimplantation genetic diagnosis are possible if a pathogenic variant has been identified in an affected family member.

Suggestive Findings

The National Comprehensive Cancer Network (NCCN) has published an algorithm for consideration of the diagnosis of both FAP and attenuated FAP [Provenzale et al 2016]. These guidelines include recommendations for genetic testing of APC. Consensus guidelines specific for GAPPS are not yet available.

According to the NCCN guidelines, an APC-associated polyposis condition should be suspected in individuals with any of the following clinical features:

• Multiple colorectal adenomatous polyps (at least 10-20 cumulative colorectal adenomatous polyps)
• Family history of multiple colorectal adenomatous polyps (>10 in a single individual, or fewer if >1 relative has multiple polyps, especially if diagnosed at a young age) and/or extracolonic features mentioned below. The total number of polyps, age of onset, number of affected relatives, and presence or absence of extracolonic findings may influence if referral for genetic counseling and/or testing is necessary based on family history.
• Hepatoblastoma
• Multifocal/bilateral congenital hypertrophy of the retinal pigment epithelium (CHRPE)
• Desmoid tumor
• Cribriform-morular variant of papillary thyroid cancer

Additional features suggestive of an APC-associated polyposis condition. The presence of the following additional features may influence the decision to offer APC genetic testing: early-onset colorectal cancer with few to no adenomatous polyps, dental abnormalities (e.g., supernumerary teeth), osteomas, odontomas, epidermoid cysts, duodenal adenomas and cancer, gastric fundic gland polyposis, gastric cancer, pancreatic cancer, small bowel carcinoma, and/or medulloblastoma.

Establishing the Diagnosis

The diagnosis of an APC-associated polyposis condition is typically considered in an individual with characteristic clinical findings and is established by identification of a heterozygous germline pathogenic variant in APC (see Table 1).

Note: A variety of terms have been used to describe individuals with an APC-associated polyposis condition; FAP, attenuated FAP, Gardner syndrome, Turcot syndrome, or the recently described gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS). The clinical features associated with these phenotypes are included here; however, all are now genetically defined as caused by pathogenic variants in APC. Terms such as Gardner syndrome and Turcot syndrome are of historical interest and should not be used any longer as both are now known to be part of the FAP spectrum. GAPPS was only recently described and the current phenotype is not yet well understood. Management guidelines differ between FAP and attenuated FAP. A consensus has not been reached regarding the precise clinical diagnostic criteria that distinguish FAP from attenuated FAP.

The diagnosis of familial adenomatous polyposis (FAP) is considered in an individual with a heterozygous germline pathogenic variant in APC by molecular genetic testing (see Table 1) and ONE of the following:

• At least 100 colorectal adenomatous polyps (individuals at younger ages or those with colectomies may have fewer than 100 colorectal adenomatous polyps).
  Note: (1) The presence of 100 or more colorectal polyps is not specific to FAP; genetic testing of APC may help distinguish FAP from other colonic polyposis conditions (see Differential Diagnosis).
• Multiple but fewer than 100 adenomatous polyps and a relative with confirmed FAP

The diagnosis of attenuated FAP is considered in an individual with a heterozygous germline pathogenic variant in APC by molecular genetic testing (see Table 1) and:

• A relative with confirmed attenuated FAP; AND/OR
• Fewer than 100 colorectal adenomatous polyps; OR
• More than 100 colorectal adenomatous polyps at an advanced age (>40 years).

The diagnosis of gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS) is considered in an individual with the following [Worthley et al 2012]:

• Gastric polyps restricted to the body and fundus
• More than 100 polyps in the proximal stomach or more than 30 polyps in a first-degree relative of an individual with GAPPS
• Predominantly fundic gland polyps (FGPs); some having regions of dysplasia (or a family member with either dysplastic FGPs or gastric adenocarcinoma).
• An autosomal dominant pattern of inheritance
• No evidence of colorectal or duodenal polyposis

Note: (1) Other causes of gastric polyposis should be excluded, including proton pump inhibitor use (repeat endoscopy off therapy should be performed in individuals on proton pump inhibitors) and other heritable causes. (2) The above diagnostic criteria were originally proposed prior to the identification of disease-associated variants in promoter 1B of APC as the cause of GAPPS. Given this new information, disease-associated variants in promoter 1B of APC should be included in the diagnostic criteria for GAPPS, but these diagnostic criteria are rapidly evolving as more information becomes available. For example, colonic polyposis has now been seen in a select number of individuals reported to have GAPPS [Li et al 2016]

Clinical Description

APC-associated polyposis conditions include familial adenomatous polyposis (FAP), attenuated FAP, and gastric adenocarcinoma and proximal polyposis of the stomach (GAPPS).

Genotype-Phenotype Correlations

Although variation occurs among individuals and among and within families with identical APC pathogenic variants [Giardiello et al 1994, Friedl et al 2001], much effort has gone into making genotype-phenotype correlations. Some have suggested basing management strategies on these associations [Vasen et al 1996], whereas others feel that therapeutic decisions should not be based on genotype [Friedl et al 2001].

While not in routine use at present, the following correlations may become important in management decisions in the future (see Table 3 for reference sequences for pathogenic variants discussed in this section):

• The most frequent APC pathogenic variant is located at codon 1309 (c.3927_3931delAAAGA) [Friedl & Aretz 2005]. Pathogenic variants at this codon lead to a high number of colonic adenomas at an early age [Friedl et al 2001, Bertario et al 2003].
• The average age of onset in individuals with colonic symptoms [Friedl et al 2001] varied by pathogenic variant location:
  • At codon 1309: age 20 years
  • Between codon 168 and 1580 (excluding 1309): age 30 years
  • 5' of codon 168 and 3' of codon 1580: age 52 years
• Profuse polyposis (an average of 5000 polyps) has been reported with pathogenic variants in codons 1250-1464 [Nagase et al 1992].
• Attenuated FAP is associated with the following:
  • Pathogenic variants (usually truncating variants) in the 5' part of the gene (codons 1-177) [Sieber et al 2006], exon 9 [van der Luijt et al 1995, Soravia et al 1998, Sieber et al 2006], and the distal 3' end of the gene [Friedl et al 1996, van der Luijt et al 1996, Walon et al 1997, Sieber et al 2006]
  • Interstitial deletions of chromosome 5q22 that include APC [Pilarski et al 1999]
  • Partial and whole-gene deletions [Nielsen et al 2007a]
  • Somatic mosaicism for APC pathogenic variants that are generally associated with FAP [Friedl & Aretz 2005, Aretz et al 2007, Hes et al 2008]
• A fourfold increased risk for duodenal adenomas was found in individuals with pathogenic variants between codons 976 and 1067 in one study of Italian individuals with FAP [Bertario et al 2003].
• Prominent extracolonic manifestations often correlate (though not completely) with more distal APC pathogenic variants. A retrospective study of 190 individuals with FAP that evaluated nine extracolonic manifestations (desmoid tumors, osteomas, epidermoid cysts, duodenal adenomas, gastric polyps, hepatoblastoma, dental anomalies, periampullary cancers, and brain tumors) [Wallis et al 1999] revealed that:
  • Individuals with pathogenic variants in codons 1395-1493 have significantly higher rates of desmoid tumors, osteomas, and epidermoid cysts than those with pathogenic variants in codons 177-452;
  • Individuals with pathogenic variants in codons 1395-1493 have significantly higher rates of desmoid tumors and osteomas than those with pathogenic variants in codons 457-1309;
  • No individuals with pathogenic variants in codons 177-452 developed osteomas or periampullary cancers;
  • Only individuals with pathogenic variants in codons 457-1309 developed hepatoblastoma and/or brain tumors.
• Desmoid tumors also have shown the following correlations:
  • After reviewing combined data on 2098 individuals with FAP, Sinha et al [2011] found that APC pathogenic variants 3’ to codon 1399 were associated with desmoid tumor development with an odds ratio of 4.37.
  • A study of 269 individuals with identifiable APC pathogenic variants found desmoid tumors in 20% of individuals with pathogenic variants 5' to codon 1444, 49% of individuals with pathogenic variants 3' to codon 1444, and 61% of individuals with pathogenic variants in codons 1445-1580 [Friedl et al 2001].
  • Several families with severe desmoid tumors with pathogenic variants at the extreme 3' end of the gene have been reported [Eccles et al 1996, Scott et al 1996, Couture et al 2000].
  • Nieuwenhuis & Vasen [2007] revealed a consistent association of desmoid tumors with pathogenic variants distal to codon 1444.
  • The data regarding desmoid tumors and genotype associations is controversial. Church et al [2015] recently evaluated the association between incidence and severity of desmoid disease in FAP and genotype. Although a higher incidence was found with pathogenic variants 3’ of codon 1399, the authors concluded that desmoid disease can develop regardless of the APC mutation site [Church et al 2015]. Additionally, pathogenic variants in this region are more often symptomatic and lethal [Church et al 2015].
• CHRPE is associated with:
  • Pathogenic variants between codons 148 to 2043 [Burger et al 2011];
  • APC whole-gene deletions [Aretz et al 2005].
• In individuals with thyroid cancer and FAP:
  • In 24 individuals, the majority of pathogenic variants identified were 5' to codon 1220 [Cetta et al 2000];
  • Nine of 12 individuals had APC pathogenic variants identified proximal to the mutation cluster region (codons 1286-1513) [Truta et al 2003].
• A review of the literature up to August 2006 and a report of additional affected individuals by Nielsen et al [2007a] revealed 89 submicroscopic APC deletions (42 partial and 47 whole-gene deletions). Most partial and whole APC deletions are associated with 100-2000 colonic adenomas, although attenuated FAP has been seen [Nielsen et al 2007a]. Extracolonic findings were seen in 36% of individuals, with no significant differences in those with partial- vs. whole-gene deletions [Nielsen et al 2007a]. One whole-gene deletion and one partial-gene deletion were found in 94 individuals with FAP [Smith et al 2016]. In six other reports of individuals with whole- or partial-gene deletions, all were classified as having a phenotype consistent with FAP as opposed to attenuated FAP [Quadri et al 2015].

Penetrance

In FAP, the penetrance of colonic adenomatous polyposis and colon cancer is virtually 100% in untreated individuals.

In attenuated FAP, the penetrance of colonic polyposis is less well understood, although the estimate of colon cancer risk by age 80 years is approximately 70% [Neklason et al 2008].

In GAPPS, the penetrance of gastrointestinal polyps and cancer is currently unknown.

See Clinical Description for discussion of penetrance of other intestinal and extraintestinal manifestations in APC-associated polyposis conditions.

Nomenclature

FAP is also known as familial polyposis coli. A term used historically for FAP is adenomatous polyposis coli (i.e., APC); APC now refers to the relevant gene. FAP is often referred to as classic FAP when more than 100 colorectal polyps are present. Classic FAP and FAP may be used interchangeably.

Gardner syndrome is the association of colonic adenomatous polyposis of FAP with osteomas and soft tissue tumors (epidermoid cysts, fibromas, desmoid tumors) [Gardner & Richards 1953]. When these findings are prominent, many clinicians continue to use the term Gardner syndrome. However, they can occur in any individual with FAP, whether or not other extraintestinal findings are present. Gardner syndrome was once thought to be a distinct clinical entity; however, it is now known that pathogenic variants in APC give rise to both FAP and Gardner syndrome. Other manifestations of FAP (e.g., upper gastrointestinal polyposis) are also found in Gardner syndrome. Some correlation exists between extraintestinal growths and pathogenic variant location in APC. See Genotype-Phenotype Correlations.

Turcot syndrome is the association of colonic polyposis or colorectal cancer and central nervous system (CNS) tumors. Historically, Turcot syndrome was used to describe individuals with either FAP or Lynch syndrome (see Differential Diagnosis) who also had CNS tumors. The molecular basis of Turcot syndrome is either an APC pathogenic variant associated with FAP or a pathogenic variant in one of the genes associated with Lynch syndrome [Hamilton et al 1995]. The CNS tumors in individuals with APC pathogenic variants are typically medulloblastoma, whereas those with Lynch syndrome are usually glioblastoma multiforme. Like Gardner syndrome, Turcot syndrome was once thought to be a distinct clinical entity; however, it is now assumed that all individuals with FAP are at increased risk for brain tumors, albeit a relatively low lifetime risk. Thus, Turcot syndrome is a historical term of little clinical utility.

Attenuated FAP (also referred to as attenuated adenomatous polyposis coli) appears to be the same as the "hereditary flat adenoma syndrome" [Lynch et al 1992].

Prevalence

Estimates of the prevalence of FAP vary from 1:6,850 to 1:31,250 live births (2.29 to 3.2 cases per 100,000 individuals) [Bussey 1975, Järvinen 1992, Bisgaard et al 1994, Bülow et al 1996, Björk et al 1999, Iwama et al 2004, Scheuner et al 2010]. The frequency is fairly constant throughout the world, with men and women being affected equally.

Attenuated FAP is likely underdiagnosed, given the lower number of colonic polyps and lower risk for colorectal cancer compared to FAP [Neklason et al 2008].

The prevalence of GAPPS is currently unknown.

APC-associated polyposis conditions historically accounted for about 0.5% of all colorectal cancers; this figure is declining as more at-risk family members undergo successful treatment following early polyp detection and prophylactic colectomy.

Hereditary Disorders to Consider in the Differential Diagnosis

MUTYH-associated polyposis (MAP). The colonic phenotype of MAP can be similar to attenuated FAP but is inherited in an autosomal recessive manner. Germline biallelic pathogenic variants in MUTYH predispose individuals to multiple adenoma or polyposis coli. If an APC pathogenic variant is not identified in an individual with colonic polyposis, molecular genetic testing of MUTYH should be considered [Sieber et al 2003].

Biallelic MUTYH pathogenic variants have been found in a few individuals diagnosed with colorectal cancer at age 50 years or younger who have had few or no polyps [Wang et al 2004]. The frequency of duodenal polyposis is between 4% and 25% among individuals with biallelic MUTYH pathogenic variants; extraintestinal findings are also noted on occasion [Aretz et al 2006].

In one study of individuals with polyposis without an identified APC pathogenic variant, the detection rate of MUTYH pathogenic variants varied by the colonic severity [Aretz et al 2006]; biallelic MUTYH pathogenic variants were found in:

• Forty (18%) of 227 individuals diagnosed with ten to 100 polyps after age 25 years or more than 100 polyps after age 45 years;
• Seven (27%) of 26 individuals with more than 100 polyps diagnosed between ages 35 and 45 years;