Abstract
Background/Aim: The purpose of this study was to elucidate the status and clinical significance of skip lymph node metastasis (LNM) in colorectal cancer. Patients and Methods: This is a retrospective cohort study from a high-volume cancer center in Japan. A total of 1,060 patients with primary stage III colorectal cancer (CRC) underwent radical resection were included in the study and divided into skip LNM (LNM) (skip+) and non-skip LNM (skip−) groups. Propensity score matching was used to balance potential confounding factors. The cancer-specific survival (CSS) and relapse-free survival rate (RFS) between the two groups before and after matching were evaluated. Results: Before matching, patients in the skip+ group had a greater number of LNM (pN2: 48.3% vs. 31.1%; p<0.001), more pathological stage IIIC (43.2% vs. 17.6%; p<0.001), and a higher rate of recurrence (38.1% vs. 25.8%; p=0.0045). Consequently, the CSS (p=0.034) and RFS (p=0.005) were worse compared to the skip− group. However, after matching, there were no significant differences in prognosis for CSS (p=0.529) and RFS (p=0.515). In multivariate analysis, skip+ was not an independent risk factor for prognosis. However, the skip+ LNM group was more likely to have local lymph node recurrence. Conclusion: Skip+ LNM was more likely to happen in later stage CRCs, leading to significantly poorer survival outcomes. However, when all background factors were matched, the prognosis of the skip+ group was not worse. Surgeons need to be more alert to the possibility of local recurrence in patients with skip LNM.
The lymphatic system is one of the main routes of tumor spread, and lymph node metastasis (LNM) is an important prognostic factor of colorectal cancer (CRC) (1). Theoretically, cancer cells are progressive, and LNM should appear first in the area closest to the tumor. However, sometimes LNM can be detected far away from the primary tumor while not being found near the tumor, which is called skip LNM. This unique pattern of LNM has been reported in various cancers such as gastric, breast, and lung malignancies (2). Skip metastasis in CRC has also been reported in some studies (3, 4), but data on the prognosis and mechanism of this phenomenon are inconsistent due to the limited number of patients in each study (5).
The purpose of this study was to elucidate the differences in clinical pathology and prognosis between skip LNM and non-skip LNM CRC groups using a propensity score matched analysis.
Patients and Methods
From April 2007 to December 2020, a total of 1,487 patients with primary stage III CRC underwent radical resection; of these, 427 patients with multiple advanced cancers, familial polyposis coli, non-radical resection, and non-D3 radical resection were excluded. Finally, a total of 1,060 patients with CRC were divided into skip LNM (skip+) and non-skip LNM (skip−) groups (Figure 1). The Ethics Committee of the Saitama Medical University International Medical Center approved the study, and all study participants provided their informed consent.
Schematic of patient allocation/study design. CRCs: Colorectal cancers; Skip−: no skip lymph node metastasis; Skip+: skip lymph node metastasis.
The treatment of specimens. This study was based on data collected prospectively in a high-volume cancer center in Japan according to the Japanese Society for Cancer of the Colon and Rectum (JSCCR). Patients with pathological stage III colon cancer who underwent D3 lymphadenectomy were identified. The freshly resected CRC specimens were divided into apical lymph nodes (LNs), intermediate LNs, and pericolic LNs (Figure 2) according to the Japanese General Rules for Clinical and Pathological Studies on Cancer of the Colon, Rectum and Anus (1). Clinical information was used to determine the stage of the tumor according to the seventh edition of the Union for International Cancer Control (UICC) Tumors, Nodes, Metastasis (TNM) classification as described in the American Joint Committee on Cancer Staging Manual. Two Japanese pathologists reviewed all colorectal cancer specimens.
Lymph node (LN) classification according to the Japanese General Rules.
Definition of skip LNM. In colorectal cancer, LNM generally occurs along the main vessels supplying blood to the tumor. Skip node metastases occur when lymph nodes closer to the tumor are negative, while distant nodes are positive. For example, the occurrence of negative pericolic LNM, but positive distal intermediate or apical lymph nodes, was defined as skip LNM (skip+). The development of positive pericolic lymph nodes and negative intermediate lymph nodes, but positive apical lymph nodes, was also defined as skip+. In rectal cancer below the peritoneal reflection, positive lateral nodes that were along the internal, external, and common iliac arteries were also defined as skip+ (Table I).
Definition of skip lymph node (LN) metastasis.
Propensity score matching was used to balance the base variables for comparative analysis using polynomial logistic regression models (Table II). Age, sex, tumor location, surgical method, neoadjuvant chemotherapy, amount of operative blood loss, depth of invasion, preoperative carcinoembryonic antigen (CEA) levels and harvested lymph nodes, LNM, lymphatic invasion, perineural infiltration, vascular invasion, and pathological staging were included. After matching, logistic regression analysis was performed to determine differences in the cancer specific survival rate (CSS) and relapse free survival rate (RFS) between skip− and skip+ LNM.
Clinical characteristics and surgical results before and after propensity score matching.
All statistical analyses were performed on Mac computers using SPSS version 22 software (IBM Corp, Tokyo, Japan). The chi-square test and Fisher exact test were used to test for differences in categorical variables. The Kaplan-Meier method was used to estimate the CSS and RFS. A value of p<0.05 indicated statistical significance.
Results
There were 1,060 patients with stage III colorectal cancer in this study. Prognostic survival analysis was performed for the farthest LNM group. The 5-year CSS in the pericolic, intermediate, and apical LNM groups were 91.5%, 85.5%, and 76.6%. Similarly, the 5-year RFS of the three groups were 77.6%, 64.2%, and 53.2%, respectively, with statistically significant differences (p=0.000) (Figure 3A and B).
Tendency in LNM and survival outcomes. (A) Cancer-specific survival rate (CSS), (B) Relapse-free survival rate (RFS).
A total of 1,060 patients were then divided into two groups. There were 118 patients (11.13%) in the skip+ group and 942 patients (88.87%) in the skip− group. In the skip− group, 776 patients (73.2%) had only pericolic LNM, 34 patients (12.64%) had metastasis of intermediate and pericolic LNM, and 32 patients (3.02%) had metastasis of pericolic LNM, intermediate LNs, and apical LNM. Among the 118 patients in the skip+ group, 16 patients (1.51%) had no intermediate LNM but had pericolic and apical LNM, 39 patients (3.68%) had just intermediate LNM, five patients (0.47%) had intermediate and apical LNM, and 14 patients (1.3%) had apical LNM. Lateral LNM occurred in 44 patients (4.1%) (Table I).
The clinical and pathological data of the 942 skip− patients and 118 skip+ patients were analyzed retrospectively. Univariate analysis between the skip− and skip+ groups were conducted. The skip+ group had more neoadjuvant therapy (8.5% vs. 1.8%; p<0.001) and a higher intraoperative bleeding amount of more than 100 ml (29.7% vs. 13.2%; p<0.001) (Table II). In postoperative pathological results, the proximal resection margin in the Skip+ group was slightly longer (13.6±5.76 vs. 12.4±4.8 cm; p=0.037), there were more harvested lymph nodes (29.7±13.6 vs. 27.0±11.7; p=0.045) as well as lymph node metastases pN2 (48.3% vs. 31.1%; p<0.001) and a higher percentage of p Stage IIIC (43.2% vs. 17.6%; p<0.001) (Table III). Consequently, the CSS (p=0.034) and RFS (p=0.005) were worse in the skip− group (Figure 4A and B).
Pathological characteristics before and after propensity score matching.
Survival outcomes for patients with skip− and skip+ before propensity score matching. (A) Cancer-specific survival rate (CSS), (B) Relapse-free survival rate (RFS).
To accurately describe the influence of each parameter on the prognosis of patients, Cox regression analysis was performed before matching. Tumor pathologic classification [odds ratio (OR)=2.8; 95% confidence interval (CI)=1.5-4.9; p=0.00065]; Stage IIIIC (OR=4.1; 95%CI=1.3-13; p=0.018); CEA level (OR=1.8; 95%CI=1.2-2.6; p=0.0064); age (OR=1; 95%CI=1-1; p=0.017), body mass index (OR=0.94; 95%CI=0.89-1; p=0.038) were independent risk factors for CSS (Table IV). Meanwhile, we found that the CEA level (OR=1.5; 95%CI=1.1-1.9; p=0.003); stage IIIB (OR=2.6; 95%CI=1.4-4.8; p=0.002), stage IIIC (OR=7; 95%CI=3.5-14; p=0.000), and venous invasion (OR=1.4; 95%CI=1-2; p=0.048) were independent risk factors for RFS. Regarding CSS or RFS, whether LNM was skip− vs. skip+ was not found to be an independent prognostic factor (Table IV).
Multivariate logistic regression analysis for cancer-specific survival and relapse-free survival before propensity score matching.
PSM with a ratio of 1:1 was carried out. Univariate comparison was performed between 113 patients in the skip− and skip+ groups. There were no significant differences in the clinicopathological characteristics between the two groups (Table II and Table III). Surprisingly, there were also no significant differences in CSS (74.3% vs. 83.2%; p=0.529) and RFS (65.5% vs. 62.8%; p=0.515) after matching (Figure 5A and B).
Survival outcomes for patients with skip− and skip+ after propensity score matching. (A) Cancer specific survival rate (CSS), (B) Relapse-free survival rate (RFS).
After multivariate analysis of the prognostic factors in matched patients, tumor location (colon vs. rectum) was an independent prognostic factor for CSS (OR=0.26; 95%CI= 0.12-0.6; p=0.0015). Stage IIIC (OR=4.5; 95%CI=1.3-16; p=0.021), gross type, 0-1 vs. 2-5 (OR=0.27; 95%CI=0.09-0.79; p=0.018), preoperative CEA levels (OR=1.8; 95%CI=1-3.1; p=0.036) and tumor location (colon vs. rectum; OR=0.49, 95%CI= 0.28-0.87; p=0.014) were independent prognostic factors for RFS. In terms of CSS and RFS, whether LNM was skip− vs. skip+ was not noted to be an independent risk factor (Table V).
Multivariate logistic regression analysis for cancer-specific survival and relapse-free survival after propensity score matching.
Finally, we studied the distribution of recurrence between the two groups. Before matching, the total recurrence rate of the skip+ group was significantly higher (38.1% vs. 25.8%, p=0.0045), which was mainly in lung metastasis (15.3% vs. 9.3%, p=0.0435) and local LNM (14.4% vs. 4.4%, p<0.001). After matching, there was no significant difference in the total recurrence rate between the two groups [37.2% vs. 38.9%, not significant (N.S.)]. However, the skip+ group still showed a higher occurrence of local LNM (15.0% vs. 3.5%, p=0.003) (Table VI).
Tumor recurrence status.
Discussion
This study found that the farther away the LNM was from the tumor, the worse the CSS and RFS. LNM is a very important pathological parameter to evaluate the prognosis of patients with colorectal cancer. The TNM classification is further subclassified according to the number of LNM (6). However, the traditional Japanese colorectal cancer protocol tends to use the location of LNM as a very important prognostic factor for colorectal cancer (7, 8). If there is apical LNM, no matter the total number of LNM, it is directly defined as N3.
Without any matching or adjustment, the results of this study indicate that more lymph nodes were harvested in the skip+ group and the LNM was later in staging, which may explain the poor prognosis of patients in the skip+ group. These findings are contrary to previous reports (5). The two studies occurred at different times, and the method of lymph node classification was also different; these may be some of the main reasons for the conflicting results.
In contrast to colon cancer, rectal cancer also has a unique metastasis route called lateral LNM. In Japan, the lateral lymph node is regarded as the regional lymph node of rectal cancer, and radical surgical resection is preferred if the preoperative computed tomography (CT) scan is suspicious for LNM in this area. However, in Europe and the United States, lateral LNM is regarded as distant metastasis, and radiotherapy and chemotherapy may be preferred (6, 9). Lateral lymph node is not within the mesentery of the rectum, thus we consider lateral lymph node metastases as skip lymph node metastasis. Clearly, rectal cancer appears to be more prone to skip LNM than colon cancer.
Another interesting finding was that there was no difference in prognosis between the two groups after all patient backgrounds were matched, and skip metastasis was not an independent risk factor for prognosis in multivariate analyses. The Japanese D3 excision approach emphasizes excision of lymph nodes at the root of the artery and mesentery within 10 cm along the longitudinal length of the excised colon. Therefore, with or without skip LNM, D3 excision can result in good outcomes (10, 11). Despite the recent tendency to limit sentinel lymph node biopsy in breast cancer (12) in order to promote rapid recovery, there are very few sentinel lymph nodes that could be used for radical resection of colorectal cancer (13, 14). This may impair the resection quality of CRC, and radical surgery for colorectal cancer should continue to be standard practice.
Some authors have hypothesized that skip LNM may occur because of more advanced disease that leads to lymphatic obstruction, which allows lymphatic drainage to bypass the obstructed lymph node (4, 15). However, this study showed that, of the 61 patients with LNM in T1 colorectal cancer, only 6 had skip metastases (9.8%), while the skip metastasis rates among the T2 (8.1%), T3 (11.4%), and T4 (12.6%) groups did not increase according to the depth of cancer invasion. Therefore, it is difficult to determine whether the cancer cells bypass the sentinel lymph node and skip to metastasis.
We also found that the local recurrence rate was significantly higher in the skip+ group and involved mainly regional lymph node recurrence. Therefore, whether there is a more developed LNM network in skip LNM and whether surgeons should consider expanding radical mesenteric resection when performing surgery needs to be further discussed in the future.
What is the pattern of LNM in colorectal cancer? In the Halsted model, cancer cell spread follows a process of gradual migration: that is, first to pericolic lymph nodes, then to intermediate and apical lymph nodes or distant organs. According to this hypothesis, total mesorectal excision (TME) or complete mesocolic excision (CME) for colorectal cancer can achieve complete cure (16). In contrast, the Fisher model suggests that spread of infiltrating lymph nodes occurs randomly, with distant organ metastases occurring regardless of the location of the infiltrating lymph nodes. Our findings suggest that skip LNM in colorectal cancer is more consistent with the Fisher model.
As we know, the lengths of the intestinal segment and area of mesenteric excision during D3 radical resection of colorectal cancer in Japan were less than those under the guidance of TME and CME recommendations (17). Therefore, it remains to be verified whether CME techniques, which have a wider range than D3 radical resection, have a better prognosis than D3 radical resection for preoperative diagnosis of skip+ LNM. For patients with colorectal cancer with skip LNM, surgeons need to pay attention to local LNM after D3 lymphadenectomy, and CME plus D3 radical lymphadenectomy may be the best surgical choice.
As with most retrospective studies, selection bias may be a limitation of our study. Although it included 13 years of follow-up data from a large cancer center in Japan, more evidence from large-scale multicenter clinical trials is required. Second, although we recommended postoperative adjuvant chemotherapy according to the guidelines, the degree of implementation and drug effect of postoperative individualized chemotherapy will also affect the prognosis of patients.
Conclusion
Skip+ LNM was more likely to occur in later stage CRCs, leading to significantly poorer survival outcomes. However, when all background factors were matched, the prognosis of the skip+ group was not worse than that of the skip− group. Surgeons need to be more alert to the possibility of local recurrence in patients with skip LNM.
Footnotes
Authors’ Contributions
Bolun Song and Liming Wang drafted the manuscript and provided original pictures; Yinggang Chen and Yasumitsu Hirano reviewed the manuscript. All Authors reviewed the manuscript.
Conflicts of Interest
The Authors have no competing interests to declare with respect to this study.
Funding
This study was supported by the Sanming Project of Medicine in Shenzhen (NO. SZSM201911012).
- Received May 31, 2023.
- Revision received June 28, 2023.
- Accepted June 29, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.











