Abstract
OBJECTIVE: To investigate whether lateral lymph node (LLN) dissection (LLND) can be exempted in patients with good tumor regression grading (TRG) after neoadjuvant chemoradiotherapy (nCRT)? METHODS: A retrospective study was conducted on consecutive patients with advanced rectal cancer who underwent nCRT and total mesorectal resection plus selective LLND at our institution. The primary outcomes are the relationship between LLN metastasis (LLNM) and magnetic resonance imaging TRG (mrTRG) and the relationship between LLNM and pathological TRG (pTRG). RESULTS: A total of 91 patients were included, of which 24 patients (26.4%) had LLNM, 67 patients (73.6%) had no LLNM. There were significant differences of the maximum short-axis of LLN before and after nCRT, short-axis reduction rate of the LLN with maximum short-axis, length diameter reduction rate of primary tumor, mrTRG, and pTRG between the two groups. Multivariate logistic regression showed that mrTRG (P = 0.026) and pTRG (P = 0.013) were independent predictors for LLNM. The combination used by mrTRG and the maximum short-axis of LLNs ≥ 8 mm before nCRT and the maximum short-axis of LLN ≥ 5 mm after nCRT achieved specificity of 0.970, positive predictive value (PPV) of 0.867, and negative predictive value (NPV) of 0.855. The same combination used by pTRG achieved the specificity of 0.970, PPV of 0.857 and NPV of 0.844. CONCLUSION: The suspected positive LLNs tend to be sterilized by nCRT in patients who have a very good response to nCRT. It is rational to avoid LLND in patients whose primary tumor and LLNs both show good response to nCRT.