Neoadjuvan Kemoradyoterapi ile tedavi edilen rektum kanseri hastalarında lenf nodu tutulumunu predikte eden faktörler

Abstract

Purpose/Objective(s): Following neoadjuvant chemoradiotherapy (NART), some rectal cancer patients may have pathologic lymph node involvement (ypN+) despite complete response in the primary tumor, and should not be candidates for non-operative management. The aim of this study was to determine risk factors for ypN+ in rectal cancer patients treated with NART. Materials/Methods: Data of rectal cancer patients treated with NART followed by surgery were evaluated. Patients with synchronous colon cancer, radiotherapy to surgery interval longer than 13 weeks, and those without lymph node (LN) dissection were excluded from the analysis. 92 patients were included. Median age was 59 (30-83). Female/male ratio was 35/57. Tumor location was proximal, mid- and distal rectum in 3, 45 and 44 patients, respectively. Tumor histology was adenocarci- noma in 79 patients, signet ring cell in 6 and mucinous carcinoma in 7. Clinical T stage was T2, T3 and T4 in 3, 81 and 8 patients, respectively. 90 patients had clinical LNs larger than 4 mm (10-34 mm: n=46). Median tumor and LNs SUVmax were 15 (3.9-46) and 2 (0.5-17.6), respectively. Median tumor GTV was 73 (28.9-370) ml. Median tumor length and thickness were 6 (3-14) cm and 16.5 (7-30) mm, respectively. All patients received 56 Gy to the primary tumor and 50.4 Gy to the lymph node regions, simultaneously in 28 fractions, and concurrent capecitabine. 27 patients had ypN+. Clinical and radiological features at initial presentation were analyzed to predict ypN+ using Chi-square or Mann-Whitney U tests. Logistic regression analysis was performed to determine the factors affecting ypN+. ROC analysis was performed to determine the cutoff values. Results: Age (p=0.027), tumor SUVmax (p=0.001), largest clinical LN size (p=0.007) and number of LNs ≥10 mm in diameter (p=0.033) were signifi- cantly different between ypN0 and ypN+ patients in univariate analysis (Table). Only tumor SUVmax (HR (%95 CI): 0.828 (0.731-0.937), p=0.003) and largest clinical LN size (HR (%95 CI): 1.338 (1.023-1.750), p=0.033) were significant in multivariate analysis. Area under the ROC analysis curve (AUC) to determine ypN+ was 68.8% for LN size and 79.3% for tumor SUVmax. Cutoff values of LN size and tumor SUVmax for ypN+ were 10.5 mm (p=0.005) and 12.57 (p<0.001), respectively. Although ypN+ rates were higher in patients with signet ring cell (83%) and mucinous (43%) his- tology compared to adeno carcinoma (24%), it was not statistically significant. Conclusion: Rectal cancer patients with lower tumor SUVmax and larger clinical lymph nodes at initial presentation are more likely to have patho- logic lymph node involvement following neoadjuvant chemoradiotherapy.