Early stage colon cancer

Abstract

Evidence has now accumulated that colonoscopy and removal of polyps, especially during screening and surveillance programs, is effective in overall risk reduction for colon cancer. After resection of malignant pedunculated colon polyps or early stage colon cancers, long-term repeated surveillance programs can also lead to detection and removal of asymptomatic high risk advanced adenomas and new early stage metachronous cancers. Early stage colon cancer can be defined as disease that appears to have been completely resected with no subsequent evidence of involvement of adjacent organs, lymph nodes or distant sites. This differs from the clinical setting of an apparent “curative” resection later pathologically upstaged following detection of malignant cells extending into adjacent organs, peritoneum, lymph nodes or other distant sites, including liver. This highly selected early stage colon cancer group remains at high risk for subsequent colon polyps and metachronous colon cancer. Precise staging is important, not only for assessing the need for adjuvant chemotherapy, but also for patient selection for continued surveillance. With advanced stages of colon cancer and a more guarded outlook, repeated surveillance should be limited. In future, novel imaging technologies (e.g., confocal endomicroscopy), coupled with increased pathological recognition of high risk markers for lymph node involvement (e.g., “tumor budding”) should lead to improved staging and clinical care.

Core tip: Evidence has now accumulated that colonoscopy and removal of polyps, especially during screening and surveillance programs, is effective in overall risk reduction for colon cancer. After resection of malignant pedunculated colon polyps or early stage colon cancers, long-term repeated surveillance programs can also lead to detection and removal of asymptomatic high risk advanced adenomas and new early stage metachronous cancers. In future, novel imaging technologies (e.g., confocal endomicroscopy), coupled with increased pathological recognition of high risk markers for lymph node involvement (e.g., “tumor budding”) should lead to improved staging and clinical care.

INTRODUCTION

Adenocarcinoma of the colon, including rectum, is a major cause of morbidity and mortality among all internal malignant diseases in men and women. When the disease is at an advanced stage with documented metastatic involvement of lymph nodes or other organs, the prognosis is especially dismal. A number of different staging criteria have been used to estimate the depth of cancer penetration in the colon as well as the extent of extra-colonic disease involvement. Currently, a commonly used staging method for colon cancer is based on the TNM (tumor/node/metastases) system as delineated by the American Joint Committee on Cancer (AJCC), now with a staging manual and atlas in its 7^th edition[1]. These different AJCC stages are summarized in Table 1.

Table 1.

Colon cancer staging

AJCC stage	TNM stage	TNM criteria
Stage 0	Tis N0 M0	Tumor confined to mucosa
Stage I	T1 N0 M0	Tumor invades submucosa
Stage I	T2 N0 M0	Tumor invades muscularis propria
Stage IIA	T3 N0 M0	Tumor invades subserosa
Stage IIB	T4 N0 M0	Tumor invades adjacent organs
Stage IIIA	T1-2 N1 M0	Tumor metastases to 1-3 nodes
Stage IIIB	T3-4 N1 M0	Tumor metastases to 1-3 nodes
Stage IIIC	Any T, N2, M0	Tumor metastases to 4 or more nodes
Stage IV	Any T or N, M1	Metastases to distant sites

AJCC: American Joint Committee on Cancer; TNM: Tumor/Nodes/Metastases. Other classification methods include: Dukes System: A, tumor confined to intestinal wall; B, tumor invading through the intestinal wall; C, tumor with lymph node involvement; D, tumor with distant metastases; and Astler-Coller System: A, Tumor limited to mucosa; B1, Tumor through muscularis mucosa but not muscularis propria; B2, Tumor beyond muscularis propria; C1, B1 with lymph node metastases; C2, B2 with lymph node metastases; D, Distant metastases. Other criteria include: venous and lymphatic invasion and differentiation.

EARLY STAGE COLON CANCER

Early stage colon cancer can be defined as disease that appears to have been completely resected with no subsequent evidence of involvement of adjacent organs, lymph nodes or distant sites. This definition differs from the clinical setting of an apparent “curative” resection later pathologically upstaged following detection of malignant cells extending into adjacent organs, peritoneum, lymph nodes, or other distant sites, including the liver.

This highly-selected group with disease localized in the colon still remains at especially high risk for subsequent development of colon polyps and metachronous colon cancer. Conceptually, this definition of early stage disease reflects increasing use of colonoscopic surveillance as an important tool in an emerging management approach. Precise staging, however, is critical, not only in assessing the need for adjuvant chemotherapy, but also for the selection of patients for continued surveillance. In patients with advanced stages of colon cancer and a more guarded outlook, repeated surveillance should be limited.

IMAGING METHODS

Although imaging methods are important in defining suspected areas of involvement, complete staging currently requires pathological assessment of resected tissue, particularly to define early stage disease. Usually staging has been estimated after surgical removal of the colon cancer, however, experience has shown that complete staging is also possible after endoscopic resection of a malignant pedunculated polyp that has minimal invasion. For these malignant polyps, however, deep histopathological assessment is not possible and lymph nodes are not removed. Further upstaging of colon cancer may result from employment of ultrasound, computed tomography (CT), magnetic resonance imaging or position emission tomography with pathological confirmation. In contrast, studies have already confirmed that methods such as fecal immunochemical testing (FIT) or CT have limited value in the detection of early stage colon cancer. For example, a high rate of false-negative results with FIT for early stage cancers was recently recorded[2] and CT was shown to have a low sensitivity for diagnosis of early T1 or T2 cancers[3].

Studies to explore staging using evolving endoscopic methods have also appeared. For example, a recent report[4] compared new techniques for assessment of the actual depth of colon cancer invasion. Magnification chromoendoscopy and endoscopic ultrasound were found to have similar accuracy in estimating the depth of invasion, but neither procedure was believed to currently have sufficient diagnostic accuracy for use as a reliable or recommended standard[4]. Further investigative efforts are needed to explore novel and emerging imaging developments, particularly endoscope-based or probe-based confocal endomicroscopic methods. These offer the possibility for more rapid (and possibly for economical) differentiation of neoplastic from non-neoplastic colonic disease, earlier diagnosis of colorectal cancer, further evaluation of degree of differentiation and estimation of invasion depth for early colorectal cancer[5-8].

OUTCOME OF STAGING

Evidence has accumulated to show that a more advanced cancer stage is correlated with a worse clinical outcome. In patients with localized and limited disease confined to the submucosa or muscularis propria, the overall 5 year survival is about 70%. With more advanced disease extending beyond the subserosa into adjacent structures, peritoneum, lymph nodes or distant sites, the overall 5 year survival is about 30%. Even in early stage colorectal cancer, bowel perforation from the tumor itself or anastomotic leakage following surgery is associated with increased recurrence rates and an impaired disease-free survival[9].

Early detection of colon cancer has been an important goal for physicians evaluating patients at increased risk for colon cancer. Colonoscopic regimens of surveillance have emerged based on good evidence that morbidity and mortality can be improved[10,11]. A number of guidelines have been developed for endoscopic surveillance of high risk groups to detect colon cancer. Some high risk categories have included a documented personal and family history of colon adenomas and colon cancer as well as inflammatory bowel disease. Among these high risk groups, a prior history of a completely resected colon cancer is a special group that should be considered for regular surveillance, particularly for those with early stage disease[12]. Most important, recent publications have provided good evidence that colonoscopy is associated with reduced colorectal cancer mortality[13,14]. In addition, persistent and sustained reduction in colorectal cancer mortality has been attributed, in large part, to the effect of polypectomy[14]. For malignant colorectal polyps with localized submucosal invasion, similar long-term results have been recorded, although a risk for new colon polyps, including advanced adenomas, and metachronous colon cancer persists[15].

SURVEILLANCE AFTER COLON CANCER RESECTION

Earlier randomized clinical trials compared intense with less intense surveillance after a “curative” resection[16-20]. Unfortunately, a number of methodological flaws in these studies were noted[21], particularly the inclusion of both early- (i.e., node-negative) and late- (i.e., node-positive) stage disease together in the comparison groups, regardless of the intensity of later surveillance. Perhaps, in these earlier studies, evaluation of more homogeneous populations, particularly with early-stage colon cancer, would have shown a positive effect of surveillance because prognosis for patients with nodal involvement, invasion of other structures and distant metastases would be expected to be much more limited[21]. Moreover, a more recent Cochrane evaluation has suggested a survival benefit for selected patients with more intense follow-up[22]. Finally, long-term studies of symptomatic early stage colon cancer patients followed over more than 10 years[23] demonstrated no locally recurrent disease. However, in the same study[23], there was still an ongoing risk for new and asymptomatic neoplasms, including advanced adenomas and early-stage metachronous colon cancers.

RISK OF LYMPH NODE METASTASES

A number of factors critical to accurate clinical and pathological staging have been explored in recent years, especially definition of high risk factors for lymph node involvement, if only early stage colon cancer with submucosal invasion (or T1) disease appears to be present. These factors include lymphatic invasion, venous invasion, tumor budding, poor tumor differentiation, extent (especially width) of submucosal invasion, complete disruption of muscularis mucosa. Indeed, some studies have suggested that up to 16% with localized submucosal invasive disease may already have lymph node metastases[24-30].

For malignant pedunculated colon polyps, Haggitt et al[24] initially proposed a 4-level classification defined by increasing depths of cancer invasion into the submucosa, particularly if deeper than the polyp stalk. Level 4 invasion into the submucosa was thought to represent the highest risk for lymph node metastases. Some have used alternative measures of depth of invasion to ensure complete electrocautery removal of malignant pedunculated polyps. For example, a distance from the leading invasive margin of the cancer to the cautery line of more than 2 mm has been empirically used as a guideline of an adequate resection of a pedunculated lesion with a stalk. If the cautery line is involved with malignant cells after removal of a malignant polyp, colectomy should be done.

For non-polypoid malignant lesions with submucosal invasion, assessment is more difficult. In these, level 4 invasion was traditionally defined[24]. Others have suggested a different classification schema, especially for surgically-resected specimens, defined by submucosal depth of invasion (i.e., specifically, sm1, sm2, sm3) with greatest depth of invasion having greatest risk for lymph node involvement[27,31]. For endoscopic resection, complete removal of the submucosa may be more difficult pathologically to define, although a retrospective evaluation of colorectal cancer initially treated with endoscopic resection suggested that a positive vertical (rather than lateral) resection margin and inadequate lifting sign were positively correlated with risk of residual tumor and lymph node metastases[32]. Other pathological risk factors for node metastases have also been emphasized include venous or lymphatic invasion, moderately or poorly differentiated tumor grade, tumor “budding” at the submucosal invasive front of the cancer, or a completely cancer-disrupted muscularis mucosa[33]. A high CEA value may also be predictive of metastatic disease[34,35]. Because of this increased risk for node involvement after endoscopic resection with these high risk factors, colectomy may be recommended to ensure complete cancer removal and permit more detailed node sampling for metastatic disease.

TUMOR BUDDING AND OTHER RISK FACTORS

“Tumor budding” is an independent prognostic indicator of risk for lymph node involvement, especially in early TNM stage colorectal cancer, as recently emphasized by expert pathologists[36]. This description of “tumor budding” was attributed to Imai who first postulated that this particular pathological feature of an invasive colon cancer represented a sudden or rapid growth of the leading or invasive edge of a carcinoma, in part, related to an interaction between epithelial and mesenchymal elements at the tumor margin[36]. Evidence has accumulated that tumor budding as well as high tumor grade or lymphovascular invasion are independent risk factors for lymph node metastases in patients with submucosally invasive colon cancer[37,38]. Patients with none of these high risk pathological features had only rare lymph node metastases (less than 1%) whereas the risk increased substantially with one (i.e., about 20%) or multiple (i.e., almost 40%) risk factors. In addition, this study showed that absence of extensive, particularly lateral, submucosal invasion (specifically, < 4 mm in width and < 2 mm in depth), had no apparent risk of metastases to lymph nodes (using anti-cytokeratin immunohistochemical staining method for detection of lymph node micrometastases) if other high risk markers were absent. Similar observations have been independently reported[39-42], including a recent evaluation following endoscopic removal of submucosal invasive T1 colorectal cancers[43].

In future, the clinical relevance of other clinical and pathological methods of evaluation for staging, including stage II colon cancer, will need additional evaluation. These include number of lymph nodes surgically harvested[44-47], techniques used for lymph node evaluation (including detection of micrometastases with novel immunohistochemical stains and polymerase chain reaction methods)[48-51] as well as definition of the precise role of sentinel node mapping for node sampling[52-54] and final staging.

CONCLUSION

Colonoscopy screening and surveillance have a documented benefit in reducing the risk of colon cancer. As a result, more early stage colon cancers will be detected in surveillance programs and treated with endoscopic methods. Emerging imaging technologies, such as confocal endomicroscopic methods, may lead to further refinements in definition of patients with early stage disease as well their management. Pathological staging to define early stage disease also continues to evolve, particularly with the increased recognition of risk factors for lymph node disease in early stage colon cancers and immunohistochemical methods for lymph node evaluation, especially detection of lymph node micrometastases.