Cowden syndrome (CS) is a very rare autosomal dominant familial hamartoma/neoplasia syndrome [59]. In CS, the most common involvement sites are ectoderm and endoderm layers although all three germ cell layers are affected by hamartomas, and almost all patients have pathognomonic dermomucosal lesions (trichilemmomas, acral keratoses, oral papillomas, mucosal lesions) [59, 60, 67]. Breast lesions (fibroadenomas, fibrocystic disease, and adenocarcinoma) are more common in women, and cases of breast carcinoma were reported in men as well [72]. The risk for breast carcinoma in women is 30–50 %, of which 25 % are bilateral [59]. Thyroid abnormalities may include multinodular goiter, adenoma, and rarely carcinoma [67, 73]. The risk of carcinoma is increased by 10 % compared to normal population [59]. Increased risk for macrocephaly, cerebellar gangliocytoma, genitourinary malformations and endometrial carcinoma, as well as renal cell carcinoma and melanoma has been reported [60, 67].

In CS, germline mutations in tumor suppressor gene PTEN (phosphatase and tensin homologue, deleted on chromosome 10) located in chromosome 10q22-23 were reported [59].

The cystic, dilated mucin-filled crypts within lamina propriety in hamartomatous polyps of CS cannot be histologically differentiated from that of juvenile polyps [61]. Hamartomatous polyps may occur in stomach, small intestines, and/or colon. Glycogenic acanthosis of the esophagus is very common [59]. In addition, colonic lipoma, fibrolipoma, fibroma, ganglioneuroma, and adenoma have been reported [67].

Although the risk of CRC in CS remains unknown, other organs with a risk of malignancy should be evaluated. It is recommended that women should undergo mammography and transvaginal ultrasound and endometrial biopsy beginning at age 30, and both genders should be examined by thyroid ultrasound, renal ultrasound, and colonoscopy after 40 years of age [60].

In 1990 Cohen suggested that Ruvalcaba–Myhre–Smith syndrome, Bannayan–Myhre–Smith syndrome, and Riley–Smith syndrome are phenotypically and genetically similar, and unified them as BRRS [74]. Marsh et al. identified an association between PTEN gene and BRRS [75]. It is indicated that CS and BRRS have mutation of the same PTEN gene, even within the same exons; however, the phenotypic difference between these two syndromes can be explained by the hypothesis that there is a distinct expression in the PTEN gene [76].

Clinically, macrocephaly, hemangiomas of varying size, lipomatosis, genital pigmentations, hamartomatous intestinal polyps, and lipid myopathy may occur in BRRS [60, 67, 73]. The polyps in GIT and those of JP are morphologically similar, and if they occur, extraintestinal evidence is required for diagnosis of CS/BRRS [61]. One or more extraintestinal manifestation with or without polyps or a familial history of CS/BRRS can be considered as the main diagnostic criterion, although it is uncertain [61]. Patients with BRRS should undergo the same surveillance programs as patients with CS [60].

It is classified in the group of PTEN hamartoma syndrome [60]. There is a disproportionate growth of body systems, which may affect skeletal, skin, and central nervous systems [60, 77]. In addition to PTEN mutations, somatic activation in AKT1 gene in PS was also reported [78]. Periodic ophthalmologic evaluations and brain MRI are recommended for follow-up of these patients [77].

Majority of colorectal cancers (CRCs) are known to develop from ordinary adenomas through chromosomal instability molecular pathway [90]. Nonetheless, approximately 15 % of CRCs are characterized with high-frequency microsatellite instability (MSI-H) caused by defective deoxyribonucleic acid (DNA) mismatch repair (MMR) [90, 91]. MSI was first described in relation to Lynch syndrome (hereditary nonpolyposis colorectal cancer) that results from mutations in MMR genes [90]. Hypermethylation of cytosine-phosphorothioate-guanine (CpG) islands of MLH1, promoter of methylator phenotype (CIMP) results in hypermethylation of MLH1 MMR gene, leading to defective MMR, and eventually to MSI [90]. Methylation of the promoter regions in CpG islands of DNA repair genes (MLH1 and methyl guanine methyl transferase- MGMT) and suppressor genes (p16, p14, THSBS1) is also considered to be the main mechanism in progression to sporadic MSI-H carcinomas [90, 92, 93]. Thus, CIMP seen in almost all cases with sporadic MSI and those who are microsatellite stable (MSS) in CRC [90]. CIMP has a strong association with point mutations in BRAF [94]. Mutations in BRAF gene result in activation of the protein kinase pathway that promotes cell proliferation and survival [90]. About 40 % of CRCs with MSI and MLH1 hypermethylation carry oncogene mutations in BRAF [90].

Recently, with the identification of serrated neoplasia pathway, it has been agreed that some types of hyperplastic polyps may also evolve to malignancy [91, 96]. Thus, it was proposed that through this pathway hyperplastic or hyperplastic-like polyps (serrated polyps) could progress to adenocarcinoma of dysplastic sessile serrated adenoma/polyp (DSSA/P) and MSI-H [94–99]. Histological classification of serrated polyps is based on comparison of molecular features of the main polyp subtypes [90]. It is believed that many CRCs with CIMP phenotype develop via the serrated neoplasia pathway [100]. This pathway activates the mitogen-activated protein kinase (MAPK) pathway as a result of BRAF mutation in the colorectal mucosa, leading to increased expression of p16INK4a and secretion of insulin-like growth factor binding protein 7 (IGFBP7). p16INK4a or IGFBP7 methylation leads to development of SSA/P [99, 101]. A definite relationship was found between the serrated polyps, particularly SSA/P and MSI and the sporadic CRCs which promote CIMP [102]. CIMP was defined as a specific marker for development of MVHP and advanced serrated polyp/adenoma [90, 103]. CIMP often occurs in SSA/P located in proximal colon and polyps with an SSA/P histology at the margins of MSI CRCs [86, 104]. hMLH1 methylation occurs at a later stage in serrated polyps, leading to dysplasia [94]. Furthermore, advanced age, proximal location, poor differentiation, mucinous histology, high BRAF mutation, and low p53 mutation rates have also been found to contribute to the development of CIMP-H CRCs [93].

Genetically, CIMP-H CRCs can be divided into two groups: those with MSI and BRAF mutations and the second with MSS and KRAS mutations [93]. BRAF mutation and negatively correlated KRAS mutations are also reported to be associated with development of colorectal adenocarcinomas [91]. BRAF V600E mutation occurs at an early stage of pathway, and often found in MVHP and SSA/P with high level CIMP [86, 94]. BRAF and KRAS mutations are early molecular alterations in serrated lesions. No BRAF mutation is observed in ordinary adenomas [105]. Neither BRAF mutations nor CIMP have been found in CRCs with germline MMR mutations that result in Lynch syndrome [94] These findings highlight that SSA/Ps are precursors of sporadic MSI CRCs.

A second alternative pathway of carcinoma development is associated with KRAS mutations in CIMP-low and microsatellite-stable cancers [93–95, 99]. The precursor lesion of this pathway is GPHP/TSA [90, 99, 101]. Recent reports have indicated that KRAS mutation in normal mucosa activates Wnt signal pathway, leading to TSA [101]. KRAS mutations are common in rectal and polypoid TSAs, but rare in SSA/Ps [106]. In this presumed serrated pathway, inhibition of repair gene MGMT DNA by promoter hypermethylation in development of carcinogenesis has been thought to associate with KRAS mutation and CIMP-low status [94]. Thus, despite increased methylation in TSA, it is absolutely not methylation of MLH1, and it is not associated with CIMP-H MSI carcinomas [88]. Such carcinomas are reported to have poorer prognosis [95].

Two serrated neoplasia pathways segregate to KRAS or BRAF mutations based on lesion type, either polypoid or flat, respectively [90, 96]. There are data available suggesting that about 30 % of colon adenocarcinomas is derived from the serrated neoplasia pathway [101]. These pathways are summarized in Fig. 11.4 [90, 95, 96, 101].

Nowadays, various factors involved in the tumorigenesis of CRC have been investigated. For example, estrogen receptor alpha (Erα), being expressed in normal colorectal mucosa, and secreted frizzled related protein (SFRP)1 genes as well as CpG island methylation are suggested to be associated with tumorigenesis [107]. ERα and SFRP1 expressions were reported likely to be associated with adenoma, and hyperplastic polyp/serrated adenoma; however, study failed to show any association [107].

HPs account for 10–15 % of all colon polyps, and the most common (80–90 %) of all serrated polyps [1, 96]. While “traditional hyperplastic polyps” were regarded as a distinct entity, now they are considered as a subgroup of serrated polyps [1, 90]. These polyps can be found throughout the entire colon, but they are more common particularly in distal colon and rectum [83]. The incidence increases with age [108]. It occurs in asymptomatic individuals older than 50 years, and more common in women [109]. Although lifestyle, alcohol consumption, nutritional factors, low folate intake, and obesity are mostly associated with adenoma, they are also reported to be associated with hyperplastic polyps [109]. A study by Wallance et al. reported that CpG island methylation was associated with folate levels, and showed that individuals with low folate intake had an increased methylation in normal colon mucosa, leading to carcinoma [107]. As these lesions are usually asymptomatic, and they are frequently detected incidentally during colonoscopy. In some cases, these lesions cannot be differentiated endoscopically from adenomas, and thus they require biopsy and cauterization [1, 83].

Macroscopically, the majority of hyperplastic polyps measure less than 0.5 cm, and are mostly found in rectum and distal colon [110]. The hyperplastic polyps greater than 2 cm have a small risk of dysplasia and malignant transformation [83].

Microscopically, all types are characterized with a nondysplastic appearance, displaying an irregular sawtooth or luminal serrated contour on upper half of the crypts mostly, with a tubular architecture at the base [97, 98]. These polyps show a simple tubular architecture (decrescendo pattern) with a combination of goblet and mucous cells, or only goblet cells without any branching and budding or a crypt dilatation [110]. Tubules have a perpendicular and symmetrical pattern, without any horizontal elongation or branching at the base [98]. In HPs, there is a regularly thickened basal membrane beneath the surface epithelium that is not common in SSAs [111]. Proliferation area is confined to the lower part of the basally located crypt, and wider than normal mucosa at the crypt as detected by Ki67, immunohistochemically [112, 113]. In hyperplastic polyps there is a strong expression of MUC2, and possible expression of MUC5AC and MUC6 [114]. Recent studies have shown AMACR- and p16-positive at the basal portion of the crypts in hyperplastic polyps of the colon, and AMACR was also found to be positive in the colorectal carcinomas of this region [115, 116].

A study by Torlakovic et al. [83] classified left-sided hyperplastic polyps into three groups based on their morphological growth pattern, and lack of proliferation or maturation: microvesicular cell-type, goblet cell-type, and mucin-poor-type. Microvesicular hyperplastic polyp (MVHP) is the most common one, and considered to represent a “typical” hyperplastic polyp by many pathologists [83]. It frequently involves the left colon [87, 117]. Compared to a normal mucosa, it is characterized by abundancy of vesicular mucin/vesicular cells and scanty number of goblet cells (Fig. 11.5). These lesions are more common in the left colon, and characterized by expanded proliferative regions covering more than basal half of crypts and pronounced luminal serration [83]. Nuclear atypia is variable, mild nuclear atypia is seen in most of the polyps, but severe atypia is rare [83]. Dystrophic goblet cells and cells with round vesicular nuclei or with prominent nucleolus are usually rare, and if present, they are located in the lower part of the crypt. Most importantly, these lesions display surface maturation, while traditional serrated adenomas have no characteristic eosinophilic cytoplasm certainly. There may be focally concentrated hypermucinous cells in the superficial part. However, the cells have a mature appearance with small nuclei, but no hyperchromasia or atypia. The deep regenerative part of the microvesicular hyperplastic polyps appear hyperchromatic compared to the normal cells of colonic crypts, which may be confused with adenoma. However, hyperchromatic regenerative part has a regular maturation towards surface. Muscularis mucosa shows thickening and extension into the lamina propria [83].

Rarely, large microvesicular hyperplastic polyp may have a slight structural distortion or slight cryptic dilatation. In this case, some researchers believe that microvesicular hyperplastic polyp is a precursor of the sessile serrated polyps. Although rare, an increase in mitosis and in Kulchitsky cells (eosinophilic neuroendocrine cells) may occur [83]. Neuroendocrine cells with clear cytoplasm can be observed between the proliferative and nonproliferative zones in the mid-one-third of the crypts [83]. Immunohistochemistry has shown CK20-positivity in the epithelial surface, which occupies a larger area than normal mucosa, and Ki67-positivity at the base of the crypts [106]. BRAF mutation and DNA methylation abnormalities were common in cases with microvesicular hyperplastic polyp [86, 90, 91, 101, 110, 117]. BRAF mutation was identified in 76 % of cases as a specific marker, and DNA methylation (CIMP) was less common than serrated adenomas in 47 % of cases. MGMT loss was found in a small percent (26 %), KRAS in 13 %, and MLH1 in 39 % [91]. A study reported that specific antibody VE1 which allows detection of BRAF V600E mutation immunohistochemically was expressed specifically in the serrated areas of microvesicular hyperplastic polyps, and staining was stronger in serrated polyps [117].

Goblet cell hyperplastic polyps (GCHP) represent the second most common type of hyperplastic polyps, and they are usually small sessile lesions, smaller than 0.5 cm. They are more common in left colon [87, 98]. These polyps show elongated crypts with no microvesicular mucin but rich in goblet cells. Normally they are sessile, however, mostly confined to the surface and upper portions of the crypts, with very few luminal serrations [83]. Nuclear atypia and mitosis are usually not seen [83]. There is no or minimal nuclear stratification [83]. They show typical morphological findings of HP including extension of mucosa, thickened basal membrane and muscularis mucosa, and increased Kulchitsky cells [83]. Clear cell neuroendocrine cells are found very rarely [83]. In GCHP, typically DNA methylation (CIMP) is rare, and TP53 mutations are often negative, and there is no or little APC mutation [105]. In these polyps, KRAS mutation is dominant (42 %), while BRAF mutation is 21 % [87, 110].

Mucin-poor hyperplastic polyps are the least common type [83, 110]. Some authors believe that they are derived from microvesicular hyperplastic polyps resulting from inflammation, mucin depletion, and irritation due to injury and prolapse [98]. These lesions have a nonpedunculated, regenerative and nonstratifying appearance, consisting of small cells that have narrow cytoplasm and more hyperchromatic nuclei than MVHP [98]. They show a micropapillary architecture, increased clear cell neuroendocrine cells, mitosis, decreased Kulchitsky and goblet cells, but increased irregular, dystrophic goblet cells with prominent nuclear atypia [83]. Inflammation in the lamina propria is typical. KRAS mutation can be found in mucin-poor hyperplastic polyps [117] In a study by Kim et al. [93] KRAS, BRAF mutations were observed in 25 % and CIMP in 75 % respectively.

Hyperplastic polyps should be differentiated from TSA, SSA/P and prolapse-type inflammatory polyps.

Conventional adenomas show hyperchromatic, pseudostratified, cigar-like atypical nuclei. Adenomas show misplaced glands as well as atypical features such as many mitoses, even single-cell necrosis [88].

TSAs are easily differentiated from HPs, with pronounced eosinophilic cytoplasm of the cells throughout the crypt, their larger and villous structure, serration at the basal of the crypts, and presence of dilatation and branching [98, 110, 118, 119].

HPs might need to be differentiated from SSA/Ps due to presence of focal serrated areas, particularly in MVHPs [98, 119]. Unlike HPs, SSA/Ps show elongated crypts, marked serration throughout the crypts, reversed “T” or “L” shaped elongated, horizontal and branching crypts at the bottom, increased dystrophic goblet cells, focal nuclear stratification, and mitosis on the upper portion of the crypts [98, 118, 119].

Prolapse-type inflammatory polyps are frequently inflamed or even ulcerated, and they show characteristic fibromuscular hyperplasia in the lamina propria and markedly regenerating epithelium with superficial ischemic findings [24–26].

Until recent times, definitive treatment has not been considered for small hyperplastic polyps, and they were removed by biopsy since they are not differentiated from adenomas during endoscopy [1, 83]. However, there is epidemiological and morphological evidence showing that HPs, particularly microvesicular type, rarely progress to carcinoma due to their natural course and biologic characteristics [86, 120]. In fact, there is increasing evidence that the risk of progression to cancer should be determined depending on the number, size, and location of “hyperplastic” polyps rather than the specific morphologic subtypes [121]. It was reported that the risk for colorectal carcinoma increased in HPs that were larger than 1 cm, multiple, and located in the proximal colon in the serrated pathway of carcinogenesis [97, 121–123]. In this respect, it is suggested that a large hyperplastic polyp located in the proximal colon should be completely removed, and colon should be examined thoroughly for any potential synchronous advanced neoplasia, and such patients require follow-up [97, 118].

Unlike left-sided “hyperplastic” polyps, SSA/Ps are often located on the right side of the colon (71–75 % at the proximal splenic flexure), and larger than hyperplastic polyps, and nonpedunculated [91, 110, 124, 125]. Also, these right-sided polyps were reported to occur more often in smokers and with a family history of CRC compared to those with left colon localization [125]. Endoscopically, these lesions often show a regular surface covered with yellowish mucus [97, 126]. Some polyps resemble the surrounding intact mucosa, and show regularly arranged circular dots that can be easily distinguished from adenomas by high-resolution chromoendoscopy; however, they look like MVHP [88, 126]. As mentioned before, molecularly SSA/P, CpG island methylation (CIMP) MSI, and MSS are considered as precursors of sporadic carcinomas [90, 101, 110]. In SSA/Ps, the main characteristics of BRAF mutation include loss of MLH1 gene expression due to promotor hypermethylation and nuclear translocation of β catenin [110]. It shows a high rate of DNA methylation (CIMP) (75 %) and BRAF mutation (80 %), but a low rate of KRAS mutation (6 %) [91].

Morphologically, these polyps show elongated crypts, epithelial serration, and crescendo type of dilatation, which is seen more clearly in the base of the crypt [110]. The base of the crypt shows an irregular inverted T-shape or an L-shape form, reduced amount of lamina propria between crypts, and an inverted growth pattern beneath the epithelial surface [87, 110]. Crypts may rarely show herniation through muscularis mucosa leading to “pseudoinvasion” or “inverted growth pattern.” Although its pathogenesis remains unknown, these features may also be present in HPs, but they are more common in SSA/Ps [98]. Due to goblet cell or gastric foveolar cell differentiation at the base of the crypt, proliferative zone is asymmetrical, and often not located at the base of crypt [111]. Presence of two or three consecutive crypts with these features can be interpreted as SSA/P [110]. Immunohistochemically SSA/Ps are characterized by CK20-positivity in epithelial surface and irregular and scattered distribution of Ki67-positive cells at the base of the crypt, while basal part of some crypts may have CK-positivity instead of Ki67-positivity [106]. Furthermore, there may be increased expression of gastric-type mucins, including MUC5AC and MUC6 [88, 110, 114]. SSPs may show nuclear atypia in many stages and in every part, but especially in the upper one-third of the crypt [88]. There is little or no stratification at the one end of the spectrum, low rate of mitosis, and marked superficial maturation [87]. SSA/P may show focal cytologic dysplasia. As with conventional adenoma, low or high degree dysplasia or progression to carcinoma may be observed [97]. Furthermore, cuboidal cells with open vesicular-type chromatin and marked nucleus and increased cytoplasmic eosinophilia at the other end of the spectrum may be seen but is rare (Fig. 11.6). These polyps are more aggressive than ordinary adenomas [97]. These findings suggest a diagnosis of TSA, but, there are no long filiform or villous projections contrarily [96, 110]. SSA/P and TSA are usually easy to distinguish, but they can be described as “serrated polyp, unclassified” when differentiation is not possible [88].

Unfortunately, the natural history and malignancy potential of SSA/Ps still remain unclear, and studies on these subjects are ongoing [87, 90, 127, 128] Right-sided polyps greater than 1 cm were reported to have a high risk of becoming cancerous [123]. Another study reported that the risk for development of cancer in right-sided polyps was 20–30 % [129]. While there are some studies suggesting that the risk for progression to carcinoma was mostly related to the size of the lesion, some studies have indicated that dysplasia, loss of DNA repair, and recurrent MSI even in small SSA/Ps might result in a faster progression to neoplasm [87, 127, 128]. It is reported that it takes about 15 years for SSA/Ps to become cancerous [128].

All serrated lesions between the sigmoid and proximal colon and those measuring greater than 0.5 cm with rectosigmoid location should be completely removed during colonoscopy [98]. The unremoved polyps should be reevaluated at an interval of 2–6 months, and completely removed upon examination of the clinician and the pathologist [130].

The term “hyperplastic polyposis” was first coined in 1980 by Williams et al. [137]. Serrated polyposis was first described in 1996 by Jeevaratnam et al. [138] as familial serrated polyposis syndrome. Although it is commonly recognized as hyperplastic polyposis, morphology of the polyps is often in the form of SSA/P [139]. However, since hyperplastic polyps can be found in combination with TSA, sessile serrated polyps with dysplasia and traditional adenomas, the term “serrated polyposis” is now widely preferred [88, 140]. SPS is commonly diagnosed at a mean age of 52–56 years [99, 140–142]. Although some publications report that it is more common in females, others indicate that it occurs most commonly in males [140, 142]. So there is no gender superiority [99]. It is often asymptomatic; however, complaints of bleeding can be seen in large polyps [97]. Endoscopically, the polyps are often smaller than 0.5 cm, sessile, and rarely large and pedunculated [140]. Small polyps are frequently scattered throughout the colon, while the large polyps are located in proximal colon [88]. There are two clinical types: Type 1 includes numerous large SSA/Ps, dysplastic sessile serrated polyps, HPs and even traditional adenomas, with significantly increased risk of cancer [101]. Type 2 includes numerous (usually smaller than 0.5 cm) HPs (with microvesicular and/or goblet cell), with slightly increased risk of cancer [88].

Genetically, its etiology and molecular features remain unknown. Although there are indications that those with many polyps in younger age have familial history, and it is a genetic disease [143], most of the patients are sporadic [98]. Some studies indicated that serrated neoplasia pathway and BRAF mutation may be effective in development of carcinoma in patients with SPS [99, 120, 142, 144]. Since MUTYH-associated polyposis shows multiple SSA/P similar to SPS, it is proposed that MUTYH gene might have also been involved in development of SPS [140, 145]. It is suggested that Type 1 may be associated with BRAF mutation, and Type 2 with KRAS mutation [88].

According to the recent WHO criteria, any of the following criteria should be present for a diagnosis of SPS: (1) at least five serrated polyps located in proximal of sigmoid colon, two of them must be larger than 10 mm; (2) any number of serrated polyps located in proximal of sigmoid colon in person with first-degree relatives with diagnosed SPS; or (3) more than 20 serrated polyps distributed throughout the colon [88].

A recent study compared immunohistochemical staining patterns of p53 among polyps of SPS and SSA/P, HP, and ordinary adenomas, and found that polyps were stained for SSA/P [139]. Polyps rarely show MVHP morphology, and are greater than normal in size. Cytologically, carcinoma may develop more rapidly in dysplastic SSA/P [88, 97, 101].

Although the risk for CRC is certainly unknown in SPS, many studies indicates the increased risk [98, 99, 140, 142, 143]. A retrospective study indicated that the risk for CRC was increased by 7 % in patients on surveillance [142]. Other reports have indicated that the incidence of CRC is increased during diagnosis in patients with SPS, and CRC may develop in 25–70 % of patients during surveillance [99, 140, 142, 143]. The CRC in these patients are located predominantly in the proximal colon [140, 143].

Treatment for SPS is variable, and not standardized [142]. Early and frequent endoscopic follow-up and surgical colectomy are among the treatment options. Both CRC and SPS risks are increased in patients with polyp who have a first-degree relative with SPS, and therefore they should undergo colonoscopy every 5 year or more frequently [98, 99, 142]. If colonoscopic surveillance is not appropriate, then right hemicolectomy or total colectomy is recommended [98]. Colectomy is recommended for multiple polyps, particularly in the proximal colon [98]. In conclusion, the treatment should be individualized based on the age of the patient, comorbidity, number and size of polyps, type of polyp, and presence of dysplasia and carcinoma [88]. The risk for adenocarcinoma is remarkably increased in untreated patients with SPS [97, 101].

Although adenomas are almost always asymptomatic, overt or occult rectal hemorrhage may occur in many patients [146, 147]. The incidence of adenoma often increases with age. It is most common in the fifth decade of life [146, 148, 149]. The risk for having at least one adenoma after the age of 50 years was reported to be around 30–50 % [147]. While some studies indicate that it is more common in males [150, 151], other studies do not support it, and report that it is more common in females [152, 153]. High levels of Vitamin D and calcium were reported to reduce the risk for colorectal adenoma, particularly those in the proximal colon [154, 155].

Molecularly, adenomatous polyposis coli (APC) gene involved in development of adenomas is a tumor-suppressor gene located on the chromosome 5q21-22 [156, 157]. Numerous pathogenic mutations (somatic and genetic) in the pathway including β-catenin and APC protein complex were reported to be involved [158]. β-catenin is involved in cell adhesion binding via E cadherin, playing a pivotal role in the Wnt/Wg signal pathway [158]. In normal cells, the level of cytoplasmic β catenin is downregulated by APC-mediated degradation [159]. If the APC protein is not functional, β-catenin level rises in the cytoplasm, enters the nucleus, enabling transcription of many oncogenes including c-myc through Wnt/Wg signal pathway [160]. Thus, inactivation of normally functioning APC/β-catenin complex results in epithelial proliferation in dysplastic epithelium from the base of the crypts, and development of adenoma [156]. Additional mutations and loss of heterozygosity inactivate the tumor suppressor gene p53 and SMAD2 and SMAD4 on chromosome 18q, leading to carcinoma [101, 161]. As a result, large TA and TVA are originated from TA and they proceed to CRC with CIMP negative and KRAS mutation, respectively [101] (Fig. 11.9).

The adenomatous polyps of the colon are categorized as TA, TVA, and villous adenoma (VA) based on their architectural nature. Basically, they are classified as pedunculated or sessile/flat (nonpedunculated), and pedunculated polyps are more common [162]. However, some authors reported that most of the polyps were endoscopically sessile (87 %) [153, 163]. Most of the polyps have TA morphology, followed by TVA, and VA, respectively [153, 163, 164]. Some studies reported that TVA was more common [151]. Polyps are often located in the left colon [151, 163], mostly measuring 0.5–1 cm in diameter [151, 153] (Fig. 11.10).

Adenomas are morphologically defined as dysplastic clonal proliferation of colonic epithelium. Being microscopically categorized as tubular, tubulovillous, or villous, adenomas are differentiated based on a reasonable rule; villous lesions contain at least 75 % villi, while tubular lesions have less than 25 % [165, 166] (Fig. 11.11). In some reports, the rates have been given as 80 % and 20 %, respectively [156, 167]. Tubulovillous lesions contain between 25 and 75 % villi [165, 166] (Fig. 11.12). The rate of villous differentiation increases with the increasing size of adenoma. Villous lesions represent an important parameter in diagnosis of advanced adenomas [89, 130, 164, 168–171].

Crypt apoptosis, neutrophilic infiltrate of the lamina propria and crypts (focal active cyrptitis) are common in conventional adenomas [153]. Neutrophilic infiltrate of the lamina propria can be seen without focal active cryptitis [153]. This finding is evident particularly in high-grade adenoma and mainly in invasive areas [153]. Some adenomas, particularly pedunculated adenomas, may show dilated and ruptured crypts with mucin extravasation into the lamina propria. This feature is often associated with misplacement of the epithelium into the submucosa (will be discussed further) [89, 172, 173].

Dysplasia was previously classified in three groups as mild, moderate, or high [167]. At present, a dual classification is used as low- (mild or moderate) and high-grade (severe, including in-situ carcinoma) intraepithelial dysplasia [165, 174]. The advantages of this classification are as follows: (1) it reduces interobserver variability in interpretation of dysplasia by pathologists; (2) it provides clinical pathological compliance for surveillance and treatment options; and (3) it prevents misinterpretation of the term “in-situ carcinoma” as indicative of a malignant behavior, and unnecessary resections of the colon.

Microscopic definition of adenoma includes at least a low-grade of dysplasia. Some data report that high-grade dysplasia is more common, and they mainly have TVA and VA morphology [151]. Although it was reported that high-grade dysplasia and carcinoma were more common in flat adenomas, with a higher malignant degeneration, it is still controversial [165, 175, 176]. The size of the polyp is not associated with dysplasia; sometimes even small adenomas may show high-grade dysplasia [168].

Low-grade dysplasia is defined as the nuclei that are pseudostratified or partially stratified and reach lower half of the cell cytoplasm in architecturally noncomplex crypts and epithelium [156]. Mitotic activity can be seen, but atypical mitosis, significant loss of polarity and pleomorphism are minimal. Crypts are arranged in parallel to the surface. High-grade dysplasia is defined as marked pseudostratification or stratification of the neoplastic nuclei, which extend into the luminal aspect of the cytoplasm, show significant pleomorphism, increased mitotic activity, significant increase in nuclear to cytoplasmic ratio, and loss of polarity [167] (Fig. 11.13). As architectural changes, back-to-back arrangement of glands can be seen [167]. As it progress to neoplasia, glands lose their previous structure, becoming more irregular and complex. In addition, neoplastic nuclei become paler and nucleolus gets prominent.

Advanced adenoma refers to adenomas that are larger than 1 cm with a villous structure and high-grade dysplasia [89, 130, 164, 168–171]. Advanced adenomas have a high risk of CRC, and it is emphasized that these patients require close clinical surveillance [89, 130, 169–171].

Single-cell infiltration, small gland proliferation, or significant enlargement of back-to-back glands with cribriform architecture, and their invasion into lamina propria or muscularis mucosa accompanied with desmoplasia is defined as intramucosal adenocarcinoma [165] (Figs. 11.14 and 11.15). For the differentiation of intramucosal adenocarcinoma from high-grade dysplasia, intramucosal carcinomas show significant invasion into lamina propria and it is accompanied by desmoplasia [165]. Invasive carcinoma occurs with penetration beyond muscularis mucosa and invasion to the submucosal peduncle [165] (Fig. 11.16). Several studies have shown that high-grade dysplasia, even adenomas including intramucosal adenocarcinoma, has no metastatic potential [162]. Thus, if these lesions can be removed with intact margins, polypectomy alone is an adequate treatment [162].

Half or two-third of adenomatous polyps, which are clinically the most common and important polyps, have a high risk for CRC [169]. The rate of recurrence was reported to be about 15 % in adenomas [153]. Patient’s age, histologic type, size and number of the polyps are important parameters for progression [150, 169]. When compared to individuals with one polyp, there is a twofold increased risk in those with three– to four polyps, and a fourfold increased risk in those with five or more polyps for carcinoma development [150]. As previously reported, advanced adenomas represented a big risk for development of invasive CRC [89, 130, 169–171]. Prognosis of polyps depending on their diameter remains controversial. While some argue that the diameter has no impact on local recurrence and lymphoid node metastasis in patients under surveillance following the treatment [162, 177], there are some other studies that have suggested that the risk for malignancy increases with increasing size of the polyp [178]. Another study that compared polyps smaller than 1 cm in size with diminutive (<0.6 cm) polyps reported that polyps smaller than 1 cm were more neoplastic than diminutive polyps, and the risk for carcinoma was increased in the patients older than 60 years, left colon polyps, and in males [179]. Although most of the adenomas are located distal to the splenic flexure, advanced neoplasia or adenoma has been shown in proximal to the splenic flexure in these patients, and examination of the entire colon is needed since it may show an advanced neoplasia without presence of any distal polyp [147, 163, 180, 181]. Thus, colonoscopy is an optimal screening method over all other modalities such as anoscopy, sigmoidoscopy, occult blood testing in stool, and barium enema [147]. Screening programs with colonoscopy to prevent colorectal adenocarcinoma are based on distribution of the adenomas in the colon, age of the patient, and family history [98, 130]. According to current consensus, colonoscopy screening starts at age 50 or 10 years earlier than the age of a first-degree relative who developed cancer [148, 174]. Based on the number of adenomas, on type and diameter of polyps during the first colonoscopy, and on family history, it is grouped as low-, moderate-, and high-risk adenoma, and it is repeated every 1–10 year depending on the risk of the adenoma removed during colonoscopy [130, 149, 168, 182]. Surveillance chart recommended by several studies is summarized in Table 11.2 [130, 147, 168, 182].