Hepatic metastases from colorectal carcinoma frequently recur after resection and hepatic micrometastases most likely are important in the development of such recurrences. The objectives of the current study were to assess the feasibility of the immunohistochemical detection of hepatic micrometastases from colorectal carcinoma and to determine their clinical significance.
Fifty-three patients underwent curative hepatic resection for colorectal carcinoma metastases. Multiple tissue sections were cut from the advancing margin of the largest hepatic metastasis in each patient and were stained with an antibody against cytokeratin-20 to detect hepatic micrometastases, which were defined as discrete microscopic cancerous lesions surrounding the dominant metastasis.
Normal hepatocytes and intrahepatic bile duct epithelia stained negative for cytokeratin-20 in all patients, whereas the largest hepatic tumors stained positive in 46 patients (86.8%). Among the 46 patients with hepatic tumors that were positive for cytokeratin-20, hepatic micrometastases were found immunohistochemically in 32 patients (69.6%). The presence of hepatic micrometastases was associated with a larger number of macroscopic hepatic metastases (P = 0.047) and patients with hepatic micrometastases were found to demonstrate a higher probability of intrahepatic recurrence (P = 0.003) compared with those patients without hepatic micrometastases. In addition, patients with hepatic micrometastases demonstrated a worse survival (10-year survival rate of 21.9%) compared with those patients without hepatic micrometastases (10-year survival rate of 64.3%) (P = 0.017).
Hepatectomy is the treatment of choice for patients with resectable colorectal carcinoma metastases.1–4 However, > 50% of patients eventually develop intrahepatic recurrences, even after curative hepatectomy.4–6 Hepatic micrometastases missed during resection may play an important role in the development of such recurrences. To our knowledge, hepatic micrometastases in colorectal carcinoma patients have been documented in only a few studies in which the authors used conventional hematoxylin and eosin staining alone for detection and did not reveal their clinical implications.7, 8 The efficacy of immunohistochemistry for the detection of micrometastases in lymph nodes or bone marrow has been shown in various malignancies9–17 and we hypothesize that immunohistochemical staining using an anticytokeratin antibody will increase the detection rate of hepatic micrometastases from colorectal carcinoma above that of traditional hematoxylin and eosin staining. To our knowledge there have been no reports published to date regarding the immunohistochemical detection of hepatic micrometastases.
The objectives of the current study were to assess the feasibility of the immunohistochemical detection of hepatic micrometastases from colorectal carcinoma using an antibody specific for cytokeratin-20 (CK-20)18 and to determine the clinical significance of such micrometastases.
MATERIALS AND METHODS
Between August 1989 and August 1996, a total of 66 patients with hepatic metastases from primary colorectal carcinoma were admitted for surgical intervention to the Division of Digestive and General Surgery in Niigata University Medical Hospital. A study period was selected to evaluate the impact of hepatic micrometastasis on long-term survival. Of these 66 patients, 53 who underwent a curative partial hepatic resection were included in the current retrospective study. Curative resection was defined as the removal of all macroscopic hepatic tumors without regard to resection margin status, which was judged to be histologically positive when tumor tissue was exposed to the cut surface of the liver (n = 7) and was judged to be negative when the tumor tissue was concealed by the liver parenchyma (n = 46). There were 22 women and 31 men, with a median age of 59 years (range, 32–78 years) at the time of hepatic resection. The primary tumor location was the colon in 33 patients and the rectum in 20 patients. Using Dukes staging criteria,19 2 patients were determined to have Stage A tumors, 18 were determined to have Stage B tumors, and 33 were determined to have Stage C tumors at the time of the initial colorectal resection. The primary tumor was adenocarcinoma in 52 patients and mucinous carcinoma in 1 patient; the histopathologic grade was determined to be well differentiated in 30 patients, moderately differentiated in 22 patients, and poorly differentiated in 1 patient. All patients had undergone curative resection of the primary colorectal tumor.
Twenty-one patients had synchronous liver metastases at the time of the initial colorectal resection, whereas 32 patients were reported to have metachronous tumors. Hepatectomy procedures included anatomic resection with or without nonanatomic resection (n = 40) and nonanatomic resection alone (n = 13). Adjuvant chemotherapy after hepatic resection depended on the surgeon's preference; 33 patients received oral 5-fluorouracil or 1 of its derivatives within 1 year of surgery. No patients received adjuvant radiotherapy. Clinical records and survival data were obtained for all patients. The follow-up period ranged from 6–148 months (median, 44 months) after hepatectomy.
Definition of Hepatic Micrometastases
Hepatic micrometastases were defined as discrete microscopic cancerous lesions ranging from a single cell to clusters of cells within the hepatic parenchyma or portal tracts surrounding the dominant macroscopic hepatic tumor (Fig. 1). Any macroscopic hepatic lesions detected on preoperative imaging, during surgery, or by inspection of resected specimens were excluded.
Immunohistochemical Detection of Hepatic Micrometastases
Multiple tissue blocks were cut from the advancing margin of the largest hepatic metastasis in each patient; the number of blocks ranged from two to seven blocks (median, four blocks) for each patient. A 3 μm-thick tissue section was cut from each block and was treated with 0.1% trypsin (Sigma Chemical Company, St. Louis, MO) in 0.1% calcium chloride (pH 7.8) at 37 °C for 20 minutes prior to immunostaining. After the blocking of endogenous peroxidase, the sections were incubated overnight at 4 °C with a 1:50 dilution of mouse monoclonal antibody against for CK-20, Ks 20.8 (Dakopatts, Glostrup, Denmark). Sections then were incubated at room temperature for 30 minutes with goat antimouse immunoglobulins conjugated to a peroxidase-labeled amino acid polymer, Simple Stain Max PO (M) (Nichirei, Tokyo, Japan). Diaminobenzidine was used as the chromogen, and the sections were counterstained with hematoxylin. All sections were reviewed independently by two experienced histopathologists who were blinded to the clinical details.
Factors Influencing Hepatic Micrometastasis
To elucidate factors influencing the presence of hepatic micrometastasis, 11 clinicopathologic variables were examined: patient gender, patient age at hepatectomy, location of primary tumor, Dukes stage, histopathologic grade of the primary tumor, adjuvant chemotherapy after primary colorectal surgery, timing of detection of hepatic metastases, size of the largest hepatic metastasis, number of macroscopic hepatic metastases, distribution of macroscopic hepatic metastases, and hepatectomy margin status.
For deceased patients, the cause of death was determined based on medical records. Cumulative probabilities of survival and of intrahepatic or extrahepatic recurrence were estimated using the Kaplan–Meier method. A two-tailed log-rank test was used to evaluate differences in cumulative probabilities. The Fisher exact test was used to test associations between hepatic micrometastases (or CK-20–positivity) and 11 clinicopathologic variables. All statistical evaluations were performed using the SPSS 9.0J (SPSS Japan Inc., Tokyo, Japan) software package. Values of P < 0.05 were considered statistically significant.
Posthepatectomy follow-up revealed intrahepatic recurrences in 26 of 53 patients (49.1%) and extrahepatic recurrences in 29 patients (54.7%). At the time of last follow-up, 35 patients had died. The median length of overall survival was 32 months.
Immunohistochemical Detection of Hepatic Micrometastases
Although normal hepatocytes and intrahepatic bile duct epithelia in resected specimens were found to stain negatively for CK-20 in all patients, the largest hepatic tumors stained positively in 46 patients (86.8%) and negative in 7 patients (13.2%). CK-20 positivity of the largest hepatic tumors demonstrated no significant correlations with any of the 11 clinicopathologic variables (data not shown).
Hepatic micrometastases were detected immunohistochemically in 32 of the 46 patients (69.6%) with CK-20–positive hepatic tumors; these were located within hepatic sinusoids in 12 patients or within portal tracts in 12 patients, or within both hepatic sinusoids and portal tracts in 8 patients (Fig. 1). No hepatic micrometastases were detected immunohistochemically in any of the seven patients with hepatic tumors that were negative for CK-20.
Association between Hepatic Micrometastasis and Other Factors
The association between the presence of hepatic micrometastases and the 11 clinicopathologic variables for the 46 patients with CK-20–positive hepatic tumors is summarized in Table 1. Only an increased number of macroscopic hepatic metastases demonstrated a significant correlation; hepatic micrometastases were observed more frequently in patients with multiple liver tumors (P = 0.047).
Table 1. Association between Hepatic Micrometastasis and 11 Clinicopathologic Variables for 46 Patients with CK-20–Positive Hepatic Tumors
Adjuvant chemotherapy after primary colorectal surgery
Timing of detection of macroscopic hepatic metastases
Size of the largest hepatic metastasis (cm)
Number of macroscopic hepatic metastases
Distribution of macroscopic hepatic metastases
Hepatectomy margin status
Impact of Hepatic Micrometastasis on Recurrence and Long-Term Survival
Of the 46 patients with CK-20–positive hepatic tumors, the cumulative probability of intrahepatic recurrence after hepatic resection was 69.4% at 8 years in patients with hepatic micrometastases and only16.4% at 10 years in those patients without hepatic micrometastases. Overall, patients with hepatic micrometastases were found to have a significantly higher probability of intrahepatic recurrence compared with patients without hepatic micrometastases (P = 0.003) (Fig. 2). Similarly, patients with hepatic micrometastases also demonstrated a higher probability of extrahepatic recurrence, although the difference was not found to be statistically significant (P = 0.070) (Fig. 3). The cumulative probability of extrahepatic recurrence after hepatic resection was 70.6% at 10 years in patients with hepatic micrometastases versus 37.7% at 10 years in patients without hepatic micrometastases. Finally, those patients with hepatic micrometastases appeared to have a poorer survival rate than those without hepatic micrometastases (P = 0.017) (Fig. 4). The median length of patient survival after hepatic resection was 18 months with a 10-year survival rate of 21.9% in patients with hepatic micrometastases, compared with 73 months with a 10-year survival rate of 64.3% in patients without hepatic micrometastases.
Although we aggressively resect liver metastases occurring in patients with colorectal carcinoma, intrahepatic recurrences frequently are observed.20–23 Hepatic micrometastases most likely are important in the development of such recurrences. To our knowledge, the current study is the first to demonstrate that immunohistochemically detected hepatic micrometastases predict an increased risk of intrahepatic recurrence after hepatic resection and are poor prognostic indicators of survival in patients with colorectal carcinoma.
CK-20, a member of the cytokeratin family, is expressed predominantly in gastric foveolar and intestinal epithelia, urothelium, and Merkel cells.18 In 1992, Moll et al.18 reported that CK-20–positivity was observed in the vast majority of colonic adenocarcinomas (89 of 93 cases). Although CK-20 expression frequently is observed in colorectal carcinoma tissue, it is absent in normal hepatic parenchyma and bile duct epithelia.18, 24, 25 The results of the current study confirmed this observation; although the majority of colorectal carcinoma hepatic metastases were positive for CK-20, hepatocytes and bile duct epithelia were found to be negative. The high specificity of this method enabled us to discern hepatic micrometastases from the background liver tissue easily. Although some hepatic micrometastases can be detected by the use of conventional hematoxylin and eosin staining,7, 8 the detection of a single cancerous cell or small clusters of tumor cells and their differentiation from normal bile duct epithelia often is impossible (unpublished data). Thus, immunohistochemical staining appears to be a much more practicable method.
For patients with colorectal carcinoma hepatic metastases that stained positive for CK-20, the probability of intrahepatic recurrence was significantly different between those patients with hepatic micrometastases and those patients without hepatic micrometastases, as was overall survival. Not surprisingly, the presence of hepatic micrometastases was associated with a larger number of hepatic metastases and one can assume that micrometastasis is an indicator of widespread hepatic involvement that heralds an increased risk of intrahepatic recurrence and a poorer patient prognosis. Although the results of the current study are limited by its retrospective nature and the small sample size involved, to our knowledge the current report is the largest of its kind to investigate the clinical significance of hepatic micrometastases in patients with primary colorectal carcinoma. Unfortunately these results are only applicable to those patients with hepatic tumors that are positive for CK-20.
The immunohistochemical detection of hepatic micrometastases is feasible in patients with colorectal carcinoma liver metastases. Hepatic micrometastasis most likely indicates widespread hepatic involvement and thus is predictive of an increased risk of intrahepatic recurrence after hepatic resection as well as a poorer patient prognosis.
The authors thank to Mr. Takashi Hatano and Ms Ayako Sato for their technical assistance.