The impacts of primary tumor location on risk factors for recurrence in patients with advanced stage II colon cancer undergoing curative resection


 Purpose

It has been reported that colon tumors show different characteristics according to the location of the primary tumor (left vs. right side). The aim of this study was to identify and compare clinicopathological risk factors for colon cancer recurrence based on primary tumor location, specifically, the right versus left colon.
Methods

A total of 235 patients with stage II colon cancer (including rectosigmoid cancer) who underwent curative resection with D2 or above lymph node dissection in multiple acute hospitals were analyzed retrospectively to compare clinicopathological risk factors and prognosis for recurrence of tumors originating in the right or left colon. Multivariate analysis of overall survival (OS) and recurrence-free survival (RFS) were modeled using the Cox-hazard model.
Results

For right-sided tumors, multivariate analysis revealed that no independent risk factors were associated with recurrence. In contrast, the risk factors associated with recurrence in left-sided tumors were perineural invasion and venous invasion. The 5-year overall survival rates and recurrence-free survival rates for patients with right and left side-originating tumors were comparable (OS: 93.0% and 93.3%, RFS: 93.0% and 88.3%, respectively).
Conclusion

Different clinicopathological risk factors for recurrence were identified in patients with right- vs. left-sided stage II colon cancer.


Introduction
It has been reported that despite the right and left sides of the colon constituting the same organ, they have distinct characteristics. [1] These distinctions, thought to be associated with genetic and molecular differences, have certain implications for postoperative follow-up and adjuvant chemotherapy planning.
Retrospective subgroup analysis of large clinical trials investigating RAS-positive stage IV unresectable colon cancer demonstrated that the patients in the right-sided group showed worse prognosis than those in the left-sided group. [2][3][4] Conversely, no consensus has yet been reached regarding differences in prognosis between the right-and left-side of the colon in patients with resectable colon cancer. [5][6][7] For patients who have resected stage II colon cancer, adjuvant chemotherapy is not considered as standard therapy. However, according to major groups such as the American Society of Clinical Oncology (ASCO), the National Comprehensive Cancer Network (NCCN), and the European Society for Medical Oncology (ESMO), adjuvant chemotherapy could be considered for patients with "high risk" clinicopathological features, after discussing the risks and bene ts.
[8] If right-and left-sided colon cancer have different clinicopathological risk factors for recurrence, appropriate adjuvant chemotherapy and follow-up plans should be considered separately for each group. Therefore, the aim of this study was to compare postoperative prognosis and independent clinicopathological risk factors for recurrence between right-and left-sided stage II colon cancer.

Patients
The ethics review boards of Hirosaki University and the qualifying medical facilities approved the study protocol. This study adhered to the principles of the Declaration of Helsinki. All patients provided informed consent to have their chart data used for research purposes. Group performance status score of 2 or higher, positive cytology or metastases to other organs, or who received neoadjuvant chemotherapy or radiotherapy were excluded from this study. Patients were classi ed based on the primary tumor location. Tumors between the cecum and splenic exure were considered right-sided in origin, while tumors between the descending colon and rectosigmoid colon were considered left-sided in origin. Staging and exact primary tumor locations were de ned according to the American Joint Committee on Cancer tumor-node-metastasis (TNM) criteria 8th edition. [9] Of the 240 patients with stage II colon cancer (including rectosigmoid) who underwent open radical curative surgical resection with D2 and D3 lymph node dissection, we excluded 5 cases due to inadequate pathological description. Therefore, a total of 235 patients were enrolled in this study. According to the Japanese Classi cation of Colorectal Carcinoma, regional lymph nodes were categorized into three groups based on location. [10] D1 included pericolic lymph nodes close to the bowel wall, D3 included the main lymph nodes at the ori ce of the feeding artery, and D2 included intermediate nodes along the feeding arteries between the D1 and D3 nodes. Based on the preference of the physicians, 190 of the 235 patients received oral 5-uorouracil (5-FU) at 200 or 300 mg per day for 1 year as adjuvant chemotherapy.
Patient information was extracted from the electronic medical chart system in each hospital. M.K. extracted the patients' data and A.O. double-checked it to prevent misclassi cation. The pathological slides of the enrolled patients were obtained, and a retrospective histopathological reevaluation was performed by two board-certi ed Japanese pathologists according to the criteria of the TMN 8th edition.

Surveillance
In accordance with the Japanese Society for Cancer of the Colon and Rectum (JSCCR) guidelines, [11] patients were followed up until 5-years after surgery, recurrence, or death, whichever came rst. The patients were evaluated with routine physical examinations, measurement of serum tumor markers such as carcinoembryonic antigen and carbohydrate 19-9 (every 3 months during the rst 3 years after the surgery and once every 6 months from the fourth year onward), thoracoabdominal computed tomography scans (every 6 months), and colonoscopy (1 and 3 years after the surgery).

Designs and outcomes
The following clinicopathological parameters were analyzed to determine if they represented independent risk factors for recurrence: tumor depth, histology, number of dissected LNs, lymphatic invasion, venous invasion, budding, [12] perineural invasion (PNI) [13], Extramural cancer deposit [14], chemothrapy. Overall survival (OS) and recurrence-free survival (RFS) were also analyzed. Patient survival was con rmed in outpatient clinic visits or by telephone call.

Statistical Analyses
The chi-squared test was used for categorical univariate analysis. Survival analyses were performed using the Kaplan-Meier method, and between-group differences were assessed with the log-rank test.
Hazard Ratios were estimated using Cox proportional hazards model. The backward method was used to select retained factors (p<0.1) during multivariate analysis. All statistical tests were two-tailed, and pvalues <0.05 were considered statistically signi cant. Statistical analyses were performed using SPSS software version 22 (IBM Corp., Armonk, NY, USA).     Table 3 shows the risk factors for recurrence in right-sided stage II colon cancer patients. Multivariate analysis identi ed no independent recurrence risk factor on the right side.

Discussion And Conclusions
In this study, multivariate analysis identi ed different independent risk factors for tumor recurrence depending on the location of the original tumor. We also found that there was no signi cant difference in 5-year OS or RFS between patients with right-vs. left-sided stage II colon cancer after curative resection.
Different independent risk factors for recurrence were identi ed by multivariate analysis between rightand left-sided stage II colon cancer patients. When the tumor is located on the right side of the colon, no independent risk factor was selected, whereas venous invasion and PNI were selected on the left which suggest the difference in prognosis factors between the right and left colon. Although both the right-and left-side of the colon constitute the same organ, their origins are entirely different. The right-side of the colon arises from the midgut and is vascularized by the superior mesenteric artery (SMA). In contrast, the left-sided colon is derived from the hindgut and is vascularized by the inferior mesenteric artery (IMA). From this embryological perspective, it is reasonable to expect that not only the expression of molecular subtypes but also independent risk factors for recurrence are different between right-and left-sided colon cancer patients.
Our study also showed that there was no signi cant difference in 5-year RFS or OS rates between patients with right-vs. left-sided stage II colon cancer. The in uence of the primary tumor location was found to be consistent with previously reported studies of metastatic unresectable colon cancer. [2][3][4] However, the impact of the primary tumor location on the prognosis of resectable colon cancer is still unclear, and a consensus has not yet been reached. [5][6][7] We propose that these discrepancies arise from the association between tumor location and expression of biological molecular subtypes. Supporting this, it has been reported that BRAF and RAS mutations are more common in patients with right-sided colon cancer compared to those with left-sided colon cancer. [15] One randomized clinical trial in Europe demonstrated that among patients with stage III colon cancer with a RAS or BRAF mutant phenotype, disease-free survival (DFS) was better in the right-vs. the left-sided group. In contrast, the European study also showed that in those who were wild-type for RAS and BRAF, DFS was better in the left-vs. the rightsided group.
[16] Hence, molecular subtypes seem to have a major impact on prognosis and are in uenced by the location of the tumor. In addition, clinicopathological parameters such as budding [12], EX [14], venous invasion, and perineural invasion [13] are also known to have an impact on prognosis. We hypothesize that these factors also have some connection with both molecular subtype and tumor location. To achieve a consensus on the impact of tumor location on the prognosis of resectable colon cancer, we should consider not only the location of the primary tumor, but also the molecular subtype and clinicopathological parameters.
According to the guidelines of major groups such as ASCO, This study has some limitations. First, the patients in this study were treated in the 1990s and 2000s. Thus, patients received oral 5-FU as adjuvant chemotherapy according to the assigned physicians' preference, despite oxaliplatin-based regimens currently being considered the standard regimen. Second, this study had a retrospective observational design, which may have resulted in misclassi cation of data or selection bias.
Different independent risk factors for recurrence were identi ed in right-and left-sided stage II colon cancer, although there was no signi cant difference in the 5-year postoperative prognosis between patients with right-and left-sided stage II colon cancer after curative resection. The right -and left-sided colon should be regarded as different organs, and high-risk clinicopathological features of stage II colon cancer should be assessed separately for each side. These ndings could have an impact on indications for adjuvant chemotherapy and therapeutic regimens for patients with stage II colon cancer. Further prospective studies that assess risk factors for recurrence in the right-and left-sided groups, including evaluation of patients receiving modern adjuvant chemotherapy, are warranted.
Declarations Figure 1 Five-year RFS and OS comparing right-and left-sided group.