Epithelial Neoplasms of the Colorectum

Gross Findings

Adenomas smaller than 1 cm in diameter tend to be evenly distributed throughout the colorectum. In contrast, adenomas 1 cm or larger in diameter are more commonly found in the sigmoid colon and rectum. In general, the polyp size is correlated with the histologic type—tubular, tubulovillous, or villous. Among adenomas smaller than 1 cm in diameter, more than 90% are tubular adenomas, 7% are tubulovillous, and 2% are villous. In contrast, only 50% of adenomas larger than 2 cm are tubular, whereas 38% are tubulovillous and 12% villous. Grossly, colorectal adenomas may appear pedunculated, sessile, or flat. Tubular adenomas range in size from individual crypt adenomas to large exophytic lesions several centimeters in diameter. Larger tubular adenomas may become polypoid with a lobulated surface and a stalk at the base of the lesion. In addition to the surface lobulations, areas of hemorrhage may be present resulting from erosion of the surface epithelium. In contrast, villous adenomas are commonly sessile lesions with a broad base and grossly evident papillary architecture.

Flat adenomas are a morphologic type of adenoma that appear as flat or slightly raised plaques on the mucosal surface. By definition, they usually show a concave surface that does not exceed twice the thickness of the adjacent non-neoplastic mucosa.

Microscopic Findings

Similar to the rest of the GI tract, adenomas are graded using the two-tiered system—low-grade and high-grade dysplasia. By definition, adenomas harbor at least low-grade dysplasia. Low-grade dysplasia is characterized by atypical glands lined by cells with cigar-shaped or penicillate, pseudostratified, and hyperchromatic nuclei ( Fig. 12.1 ). These cytologic changes involve both the crypts and the surface epithelium. The individual cells have increased nuclear-to-cytoplasmic ratio and frequent mitotic activity. High-grade dysplasia encompasses additional architectural and cytologic changes ( Fig. 12.2 ). Architecturally, the crypts show irregular branching, gland crowding, and cribriform architecture. Cytologically, the nuclei show increased atypia with enlarged, hyperchromatic, vesicular nuclei; prominent nucleoli; and loss of nuclear polarity (basement membrane of the cell is no longer perpendicular to the long axis of the nucleus). Mitoses, including atypical mitotic figures, are prominent.

Flat adenomas are in essence tubular adenomas that lack a polypoid growth or an exophytic growth pattern. Rather than the complex branching of adenomatous glands, villous adenomas show tall, fingerlike projections of adenomatous epithelium with a central fibrovascular core that extends from the polyp base to the surface.

A variety of cell types may be encountered in adenomas, including Paneth cells, goblet cells, and endocrine cells. The finding of these cell types in an adenomatous polyp reflect the ability of the neoplastic cells to partially differentiate into a variety of cell types. Goblet cells within adenomatous polyps are often dystrophic, seen as cells or goblet cells with atypical nuclei oriented away from the basement membrane of the gland. Endocrine cells are present in at least half of adenomas and can best be discerned with the use of special stains. Paneth cells are present in 10% of adenomas, recognized for their characteristic eosinophilic granules. Excessive Paneth cells may be seen in patients with polyposis syndrome, especially FAP. Some adenomas may contain foci of squamous morular metaplasia ( Fig. 12.3 ), osseous metaplasia, melanocytes, or even gastric type epithelium.

Gross Evaluation

Gross description of a polyp should include size, presence or absence of a stalk, and intactness of specimen. Pathologists should avoid using the terms pedunculated, sessile , and semi-pedunculated because these terms refer to in vivo endoscopic assessment of these lesions. It is a common practice to ink the polyp base margin which is cauterized during polyp removal. The benefit of this process is that it allows for easy identification of margin on microscopic examination. However, improper identification of the polyp base and inaccurate inking can contribute to inaccuracy in assessment of the margin status. Therefore, cauterized tissue visualized under the microscope is the best guide for assessing the distance of invasive adenocarcinoma from the margin.

Histologic Reporting Parameters

Risk assessment of malignant polyp (low or high risk of adverse outcome) is performed using histologic parameters that include margin status, lymphovascular invasion, and tumor grade. Unfavorable histologic features that predict recurrent cancer or lymph node metastasis include tumor at or near the margin (≤1 mm), grade 3 (poorly differentiated adenocarcinoma), or lymphovascular invasion. In theory, the presence of any of these features is an indication for surgery. In the absence of these features, the polypectomy procedure is considered to be completely curative.

• Tumor Grade

  Similar to colorectal resection specimens, the grade of invasive adenocarcinoma is assessed using the American Joint Committee on Cancer (AJCC) eighth edition classification system. It is based on the degree of glandular differentiation:

  Grade 1: well-differentiated adenocarcinoma, greater than 95% glandular differentiation or with well-formed glands with open lumina

  Grade 2: moderately differentiated adenocarcinoma; 50% to 95% glandular differentiation

  Grade 3: poorly differentiated adenocarcinoma, signet ring cell or mucinous adenocarcinoma, less than 50% glandular differentiation or solid sheets of neoplastic glands

  Grade 4: undifferentiated; no gland formation or mucin; no squamous or neuroendocrine differentiation

  Grade 1 and 2 are considered low-grade tumors.

• Lymphovascular Invasion

  Lymphatic space invasion requires the presence of tumor cells within endothelial-lined channels without significant red blood cells ( Fig. 12.6A ). Venous invasion is defined by the presence of tumor emboli within endothelial-lined channels surrounded by smooth muscle. There is a significant interobserver variability in assessing lymphovascular invasion. Lymphatic invasion should be distinguished from retraction artifact and micropapillary differentiation. Additional serial levels or ancillary stains such as D2-40 ( Fig. 12.6B ) and Movat stain may be used to confirm lymphovascular space invasion. Extramural venous invasion (tumor emboli within veins beyond the muscle layer), characterized by the “orphan artery sign” and “protruding tongue sign,” represent a poor prognostic factor.

  

• Margin Status

  Cancer at or near the margin of resection is associated with an adverse outcome. The status of margin can only be assessed in intact polypectomy specimens. The resection margin is the actual free edge of submucosal connective tissue showing thermal artifact (see Fig. 12.5B , black arrow ). There is no consensus as to the definition of what constitutes a positive margin of resection. The National Comprehensive Cancer Network 2018 Colon Cancer Guidelines define positive margin as follows: (1) tumor less than 1 mm from the transected margin, (2) tumor less than 2 mm from the transected margin, and (3) tumor cells present within the diathermy of the transected margin. It is therefore recommended that when the tumor is very close but not at the transected margin, the exact distance of the tumor from the cauterized margin be included in the report.

  When invasive adenocarcinoma is present at or near the margin of resection (≤1 mm), the risk of relapse is between 21% and 33%. However, studies have shown that even in cases with “doubtfully complete” resections, the clinical outcome is favorable. This is because of the assumption that diathermy coagulation destroys an additional 3 to 5 mm of submucosal tissue beyond what is visualized by microscopy. Occasionally, the specimen orientation and processing make it difficult, if not impossible, to accurately assess the resection margin. In these cases, the report should clearly state the limitations of histologic assessment and encourage the clinician to assess completeness of excision by correlating with in vivo endoscopic findings of resection.

• Tumor Budding

  Tumor budding is considered as an independent adverse prognostic factor in CRC. It is associated with a higher TNM (tumor, node, metastasis) stage, high tumor grade, and presence of lymphovascular invasion and therefore, ultimately, with lymph node and distant metastasis. In malignant polyps, the presence of high tumor budding is associated with an increased risk of lymph node metastasis. Therefore, it has been suggested that the finding of high tumor budding may be an indication for surgical resection. A recent International Tumor Budding Consensus Conference (ITBCC) defined tumor budding as a single tumor cell or a cluster of up to four tumor cells (see section on tumor budding later). This finding is variably reported by pathologists.

• Level or Depth of Invasion of Adenocarcinoma in a Polyp

It is not a routine practice to report depth of invasion in malignant polypectomy specimens. The Haggitt system is advocated for assessing depth of invasion in pedunculated malignant polyps. Per the Haggitt system for pedunculated malignant polyps, level 1 is defined as invasion into the polyp head, level 2 is invasion into the polyp neck, level 3 is carcinoma invading any part of polyp stalk, and level 4 is invasion beyond the stalk but above the muscularis propria. Specimen orientation, tissue shrinkage after fixation, interobserver disagreement between level 1 and level 2 invasion, and lack of muscularis propria in these specimens are some of the inherent issues that make it difficult to apply this system in routine practice. Therefore, many practices do not report Haggitt levels on pedunculated polyps.

The Kudo or Kikuchi method is used for assessing depth of invasion in sessile malignant colorectal polyps. Histologic assessment is performed by creating an imaginary division of the submucosal region into thirds (sm1, sm2, and sm3) and then determining the depth of invasion. However, accurate division of the submucosal space requires the presence of muscularis propria, and therefore this classification is most commonly employed only in transanal endoscopy microsurgery resections, which usually sample superficial layer of muscularis propria. Studies have shown that the incidence of lymph node metastasis increases from 1% to 3% in sm1 cancers to 8% and 23% in sm2 and sm3 cancers, respectively. Newer studies suggest that submucosal invasion greater than 1 mm represents an adverse prognostic factor and needs to be reported.

Gross Findings

Grossly, HPs are dome-shaped nodules that are usually smaller than 5 mm in diameter. They are often multiple and are most frequently found in the sigmoid colon and rectum. HPs are most commonly found on the crest of mucosal folds and are similar in color to the surrounding normal mucosa.

Microscopic Findings

Hyperplastic polyps contain serrated colonic crypts lined by goblet and absorptive cells. The crypt serrations are usually restricted to the upper half or two-thirds of the crypt epithelium, resulting in the characteristic “saw-toothed” or serrated appearance ( Fig. 12.10 ). The basal aspect of the crypts is normal. The number of absorptive cells lining the crypts is greater than the number of goblet cells. Mitoses are common but are confined to the bottom half of crypts (proliferative zone of the crypt). The basement membrane is often thickened, and this finding is best appreciated at the polyp surface ( Fig. 12.11 ). The epithelial cells typically show basally located small and hyperchromatic nuclei with only slight enlargement. Scattered Paneth or endocrine cells may also be seen within the crypt bases. Larger HPs often show changes of mucosal prolapse. This is characterized by splaying of the muscularis mucosae fibers at the crypt bases, with extension of the smooth muscle fibers into the lamina propria and around individual crypts ( Fig. 12.12 ). This often results in dilation of the crypt bases and can cause diagnostic confusion with SSP/A.

Hyperplastic polyps are composed of three histologic subtypes: microvesicular type, goblet cell–rich type, and mucin-depleted type. Microvesicular type HPs are the most common subtype and are characterized by microvesicular mucin with few goblet cells, including some dystrophic goblet cells ( Fig. 12.13 ). Goblet cell–rich HPs show prominent goblet cells and subtle serrations of the crypts ( Fig. 12.14 ). Mucin-depleted HPs are the least common subtype of HP and show reactive epithelial mucin loss with minimal cytoplasm ( Fig. 12.15 ). Currently, there is no evidence to support subtyping HPs for prognostic purposes.