A 25% rate of recurrence after performing complete resection in node-negative colon cancer patients suggests that their nodal staging is frequently suboptimal. Moreover, the value of occult cancer cells in tumor-free lymph nodes still remains uncertain. The authors evaluated the prognostic significance of the pathologic parameters, including the lymph node occult disease (micrometastases) detected by immunohistochemistry, in patients with node-negative colon cancer.
The study included 160 patients with curatively resected stage I or II colon cancer and they were without rectal cancer. 2852 lymph nodes were re-examined by re-do hematoxylin and eosin (H-E) staining and immunohistochemical staining. The detection rates were compared with the clinicopathologic characteristics and with the cancer-specific survival.
Occult metastases were detected in 8 patients (5%). However, no clinicopathologic parameter was found to be correlated with the presence of micrometastasis. Twenty patients developed recurrence at a median follow-up of 45.7 months: 14 died of colon cancer and 9 died from noncancer-related causes. Univariate analysis showed that lymphatic invasion and the number of retrieved lymph nodes significantly influenced survival, and multivariate analysis revealed that the stage, the number of retrieved lymph nodes and lymphatic invasion were independently related to the prognosis.
Inadequate lymph node retrieval and lymphatic invasion were found to be associated with a poorer outcome for node-negative colon cancer patients. The presence of immunostained tumors cells in pN0 lymph nodes was found to have no significant effect on survival, but these tumor were identified by re-do H-E staining. Maximal attention should be paid to the total number of lymph nodes that are retrieved during surgery for colon cancer patients.
Colorectal carcinoma (CRC) is the second most common gastrointestinal malignancy in Korea, and the incidence of CRC continues to steadily increase (
The aim of this study was to examine a wide range of clinicopathologic variables in a retrospective series of patients with curatively resected stage I or II colon cancer, and we also wanted to determine whether these findings were associated with tumor recurrence.
Between March 1996 and June 2005, 160 patients who met the following inclusion criteria underwent resection for their colon cancer: (1) colon carcinoma was proven by biopsy (rectal cancer and rectosigmoid colon cancer were excluded) and the treatment was curative colon resection, (2) R0 resection was done (on the pathologic examinations), (3) stage I or II disease (any T stage with no lymph node or distant metastasis); (4) no other synchronous or metachronous malignant tumor, (5) the carcinoma did not arise from familial adenomatous polyposis, hereditary nonpolyposis colorectal cancer, ulcerative colitis or Crohn's disease. One hundred three (64.4%) men and 57 (35.6%) women were included in this study. The median age of the patients was 62 years (range: 24~86 years)(
The mean follow-up period was 45.7 months (range: 1~137 months). All the patients were assessed for local control and distant metastases by performing clinical examination, chest radiography, abdominal ultrasonography and abdominal computed tomography scanning and measuring the carcinoembryonic antigen (CEA) serum level. Follow-up examinations were performed every 3 months during the first two years, biannually for the next three years and then annually. The survival times were calculated from the date of surgery to the date of death or the last follow-up, and these dates were confirmed using the data registered at the National Cancer Center.
With regard to the cancer's location, two anatomic sites were considered, i.e., the proximal colon (from the cecum to the transverse colon), and the distal colon (from the splenic flexure to the sigmoid colon). The tumor's gross shape was classified as exophytic or nonexophytic. The tumor's maximal diameter was categorized as ≤ or >5 cm. The tumors were classified as mucinous or nonmucinous, according to the amount of the mucinous component (≥ and <50% of the tumor volume, respectively). Well and moderately differentiated tumors were categorized as low-grade tumors, whereas poorly differentiated, undifferentiated and mucinous tumors were categorized as high-grade.
The original histologic slides of the tumor tissues were reviewed by an experienced pathologist. 2852 LNs from 160 patients were examined (mean number of LNs: 17.8 LNs/patient). For each LN, two new slices of 5 µm thickness were obtained from the original paraffin blocks. The sections were deparaffinized in xylene and then they were rehydrated. One section was stained with H-E and the other was subjected to immunohistochemistry (IHC). The IHC staining procedures were conducted using an automated immunostainer (Vision Biosystems, VIC, Australia). The sections were incubated at a dilution of 1 : 400 with a monoclonal antibody for anti-cytokeratin 20 (CK20; clone K 20.8, Dako, Carpentaria, CA). The immunostaining was developed using 3,3'-diaminobenzidine as a chromogen. Appropriate positive controls were added to each automated IHC run to confirm the sensitivity and specificity of the antibody (sections of the CK20-positive CRC tissue served as the positive controls). The immunostained slides were evaluated by a pathologist who had no knowledge of the pathologic data or the patients' outcomes.
The primary endpoints were tumor recurrence and the overall survival. Statistical correlations between micrometastasis and the pathological variables were assessed using the chi-square test and Fisher's exact test, and the relationships between the pathologic variables and survival were estimated using the Kaplan-Meier method. The survival differences between the groups were analyzed using the log-rank test. The Cox proportional hazards regression model was used to assess the clinicopathologic factors that could independently influence survival. P values <0.05 were considered to be statistically significant.
2852 LNs from 160 patients were examined after staining them with anti-CK20 monoclonal antibody. Cytokeratin 20-positive neoplastic cells were found in 12 of the 2852 LNs (0.4%) from 8 of the 160 patients (5.0%). In addition, all the corresponding H-E re-stained sections confirmed the presence of these neoplastic cells. Micrometastases (≥0.2 mm but <2.0 mm) were detected in 7 of the LNs, whereas isolated tumor cells (<0.2 mm) were not found. Macrometastases (>2 mm) were found in 5 of the LNs. A retrospective review of the original histologic slides revealed that neoplastic cells were present in 2 of the 12 CK20-positive LNs at the original evaluations (
The clinicopathologic characteristics of the patients with or without CK20-positive LNs are listed in
During the observation period, 20 patients developed tumor recurrence and 23 patients died after surgery. Nine of these 23 patients succumbed to diseases other than colon cancer; therefore, the survival analysis was performed on the remaining 14 cases. These 14 patients more often had hematogenous metastasis (n=14) than local recurrence (n=6). The sites of recurrence were as follows: six hepatic metastases, four pulmonary metastases, four peritoneal metastases and six local recurrences (anastomotic or local lymphatic).
Univariate analysis was utilized to evaluate the impacts of CK20-positive staining in the LNs and the various clinicopathologic parameters on the prognosis of patients with nodenegative colon cancer (
The terms "micrometastasis" and "macrometastasis" were originally defined to be metastatic deposits of breast carcinoma cells that measured less than or greater than 2 mm, respectively (
Special techniques such as IHC or reverse transcriptase polymerase chain reaction (RT-PCR) can be used to identify the micrometastases within LNs that are not detectable by conventional H-E staining. Several studies have employed IHC with using monoclonal antibodies directed against carcinoembryonic antigen (CEA), epithelial membrane antigen (EMA) or cytokeratin (CK). O'Brien et al. (
In the present study, occult cancer cells in the previously considered negative LNs were detected by H-E staining in 5.0% of the patients, and these patients were consequently restaged as pN1. However, the disease-free and overall survival rates of the restaged pN1 patients on univariate analysis were not significantly different from those of the pN0 patients. Significant correlations between the presence of occult tumor cells and survival have been reported by some studies (
However, the prognostic significance of the LN micrometastases that are detected by IHC in colorectal cancer patients is controversial. In contrast, based on the findings obtained with using molecular genetic techniques, some investigators have suggested there is a positive correlation between the presence of micrometastases and a poor prognosis in patients with nodenegative colorectal cancer (
In the present study, all the occult tumor cells detected on IHC were confirmed by performing examinations of the corresponding H-E re-stained sections. As many as 70% of the tumor-positive LNs have been reported to contain metastases of <0.5 cm in diameter (
However, the microscopic examinations involving only one histologic section of a LN are clearly inadequate for the detection of micrometastases and also possibly for the detection of macrometastases. In the present study, macrometastases were detected in 5 of the LNs with occult tumor cells.
It is also important to recognize that patients may be at a high risk of recurrence regardless of micrometastatic involvement. A poorly differentiated tumor, lymphatic and venous invasion, extension to adjacent organs and inadequate regional LN retrieval all have adverse effects, and those patients with such features should be recommended for adjuvant treatment (
In the present study, we examined the tumor-negative LNs in patients with stage I and II colon cancer. IHC staining was performed, and the predictive values of the clinicopathologic parameters were investigated. CK20-positive neoplastic cells were found in 5.0% of the patients, but these were not found to have a significant effect on the disease-free survival or overall survival. However, all the occult tumor cells detected by IHC were confirmed in the corresponding H-E re-stained sections. We estimate that including one more section on conventional H-E staining would upstage pN0 disease to pN1 disease in 5% of the node-negative colon cancer patients. Furthermore, LN retrieval and lymphatic invasion were found to have adverse effects on survival. We recommend that surgeons and pathologists seriously consider examining an adequate number of LNs and a sufficient number of cut sections in patients suffering with stage 1 and 2 colon cancer.
This work was supported by the second phase of the
Disease-free survival according to (A) the stage (stage I vs. stage II) and (B) the number of retrieved LNs (<12 vs. ≥12). Both parameters were found to be significantly associated with disease-free survival.
Overall survival according to (A) the number of retrieved LNs (<12 vs. ≥12) and (B) lymphatic invasion (absent vs. present). Both parameters were found to be significantly associated with disease-free survival.
Clinicopathologic characteristics of the patients with stage I and II colon cancer
The immunohistochemical staining and the re-do H-E staining findings
*immunohistochemistry, †occult tumor cells, ‡lymph nodes.
Clinicopathologic characteristics of the patients and the presence of CK20-positive tumor cells in their LNs
CK20: monoclonal, anti-human cytokeratin 20.
Univariate analysis of survival for the stage I and II colon cancer patients
*lymph node.
Multivariate Cox proportional hazards analysis of the survival factors for stage I and II colon cancer
*hazard ratio, †confidence interval.