Maximilian Brunner, Susanne Merkel, Klaus Weber, Axel Denz, Georg F Weber, Robert Grützmann, Christian Krautz
Preoperative predictors of regional lymph node metastasis in right-sided colon carcinoma, namely cN status and preoperative CEA, have limited sensitivity and specificity and are therefore insufficient for reliable prediction of node-positive disease. Given the limited reliability of preoperative nodal staging, selective restriction of complete mesocolic excision based on presumed node-positive disease cannot currently be recommended. Preoperative lymph node understaging is associated with worse overall and disease-free survival, although the underlying mechanisms remain unclear.
BACKGROUND: Complete mesocolic excision (CME) is widely adopted in specialized colorectal centers for right-sided colon carcinoma, with increasing evidence supporting its oncological relevance. However, as some studies suggest that particularly patients with stage III disease benefit from CME, the question arises whether CME could be considered selectively for stage III patients. This would require reliable preoperative detection of locoregional nodal metastases. This study evaluated the performance of preoperative nodal staging, identified predictors of pathological regional lymph node involvement and assessed the prognostic impact of staging discrepancies.
METHODS: We retrospectively analyzed 400 patients undergoing right hemicolectomy with CME at our institution between 2008 and 2024. Uni- and multivariate logistic regression analyses were performed to identify independent predictors of lymph node metastases, which were then combined into a prediction modal. Risk factors for preoperative staging discrepancies were evaluated and the impact of understaging and overstaging on overall survival (OS) and disease-free survival (DFS) was assessed using multivariable Cox regression.
RESULTS: The overall rate of regional lymph node metastasis in the cohort was 30%. Among 334 patients with available imaging-based nodal staging, preoperative cN + status was significantly associated with regional lymph node metastases (p < 0.001), but showed limited diagnostic performance (sensitivity 59.4%, specificity 65.7%, AUC 0.625). In multivariate analysis, elevated preoperative CEA (OR 2.7, 95% CI 1.1-6.8, p = 0.034) was identified as an independent predictor of lymph node metastases, in addition to cN + -status (OR 3.0, 95% CI 1.3-6.8, p = 0.010). A combined prediction model (cN + CEA) did not substantially enhance overall discriminative power (sensitivity 68.0%, specificity 57.2%, AUC 0.636). Preoperative imaging understaging occurred in 12% of patients, whereas overstaging occurred in 24%. Lower cT category and a higher rate of normal CA19-9 were significantly associated with understaging, while higher cT category was associated with overstaging. Understaged pN + patients had worse 5-year OS (47.6% vs. 79.7%, p < 0.001) and DFS (42.4% vs. 75.8%, p < 0.001) compared with correctly staged pN + patients. Multivariable Cox regression identified understaging as independent predictors of OS (HR 5.0 [95% CI 2.2 - 11.4], p < 0.001) and DFS (HR 3.7 [95% CI 1.8 - 7.8], p < 0.001), whereas overstaging was not associated with reduced survival (p = 0.339 respectively p = 0.581).
CONCLUSION: Preoperative predictors of regional lymph node metastasis in right-sided colon carcinoma, namely cN status and preoperative CEA, have limited sensitivity and specificity and are therefore insufficient for reliable prediction of node-positive disease. Given the limited reliability of preoperative nodal staging, selective restriction of complete mesocolic excision based on presumed node-positive disease cannot currently be recommended. Preoperative lymph node understaging is associated with worse overall and disease-free survival, although the underlying mechanisms remain unclear.