Advertisement

P07.07 Computed Tomography of Lymph Nodes to Predict Occult pN2 Disease in Non-Small-Cell Lung Cancer

      Introduction

      Occult pN2 disease is the presence of non-small-cell lung cancer (NSCLC) metastases in mediastinal lymph nodes (MLNs) staged cN0 or cN1. The tumor diameter and consolidation/tumor ratio (CTR) increase the risk of occult pN2 disease. However, only few studies have evaluated lymph nodes with preoperative computed tomography (CT).

      Methods

      From 2010 to 2019, 2,581 patients with NSCLC underwent pulmonary lobectomy and MLN dissection at the Kanagawa Cancer Center. Of these, we retrospectively reviewed 272 cases of NSCLC. Inclusion criteria were as follows: invasive size<5 cm (cT0/1); cN0/1; and CTR>50%. Among the examined pN2 cases, we reviewed preoperative CT imaging factors of the MLNs that presented with the maximum sum of long and short diameters. CT factors pertaining to the long and short diameters of MLN and the CT value (maximum/minimum) of MLNs were collected. Clinical factors predicting occult pN2 disease were also evaluated. Multivariate logistic regression analyses were performed to examine the factors affecting occult pN2 disease. The receiver operating characteristic (ROC) curve analysis was performed to determine the optimal cut-off values of factors that significantly affected occult pN2 in multivariate analyses.

      Results

      Eighty of the 272 cases were diagnosed as pN2 (occult pN2), while the others were diagnosed as pN0/1. The study population included 138 men and 134 women (median age: 68 years). The smoking history was positive in 150 cases. Histologically, there were 250 cases of adenocarcinoma, 11 cases of squamous cell carcinoma, and 11 other cases. Clinically, 230 cases were cT1, and 42 cases were cT2. The average CTR was 77% (range: 50%–100%). The median long and short diameters of the MLNs were 10 (range: 3–32) and 6 (range: 2–22) mm, respectively, while the median maximum and minimum CT values were 120 (range: −46–233) and −73 (range: −406–101) HU, respectively. On multivariate logistic regression analysis, the long (p=0.037) and short (p=0.018) diameters significantly affected occult pN2, but were adjusted by CTR. The optimal cut-off obtained from the ROC curve analysis for the lymph node size was 12 mm in the long diameter and 7 mm in the short diameter, with areas under the curve of 0.761 and 0.777, respectively. The sensitivity and specificity of occult N2 were 48.8% and 90.6%, respectively, when both factors were positive.

      Conclusion

      The long and short diameters of MLNs on preoperative CT significantly predicted occult N2 disease.

      Keywords

      Occult N2 disease, diameter of MLN, preoperative computed tomography