MA05.02 Log Odds of Positive Lymph Nodes Predicts Overall Survival and the Benefit of Postoperative Radiotherapy in Malignant Pleural Mesothelioma

Abstract

Nodal categories of malignant pleural mesothelioma (MPM) are mostly adopted from lung cancer staging criteria and the N descriptors in the eighth edition of TNM staging system have not been fully verified. We aimed to evaluate the effectiveness of the current N descriptors and a novel prognosticator—the log odds of positive lymph nodes (LODDS)—in predicting overall survival (OS) and postoperative radiotherapy (PORT) benefit in MPM. Patients in the Surveillance, Epidemiology, and End Results (SEER) database with MPM undergoing surgery and lymph nodes examination were extracted and restaged according to the 8th edition TNM staging system. LODDS was calculated as loge[(positive nodes count+0.5)/(negative nodes count+0.5)]. X-tile software determined the optimal cut-point for LODDS. Log-rank tests along with Cox regression analyses were adopted for survival analyses. Harrell's C-index statistic measured discriminatory ability and prognostic performance. A total of 534 patients were enrolled in this study. N descriptors were unevenly distributed. Most cases were staged as N0 (51.9%) and N1 (47.0%), with only 1.1% staged as N2. The eighth edition N descriptors failed to clarify the survival difference between adjacent categories and were incapable of predicting PORT benefit. The cut-points for LODDS were classified as follows: LODDS1 (≤-2.61), LODDS2 (-2.56≤LODDS≤0.62), and LODDS3 (≥0.87). The median survival of LODDS1 was 23.1 months compared with 17.9 months (HR=1.397, P=0.005) and 13.0 months (HR=2.317, P<0.001) for LODDS2 and LODDS3, respectively. The survival curves stratified by LODDS separated nicely without overlapping and the benefit of PORT was limited to cases with LODDS3 (≥0.87). LODDS also provided better C-index than the conventional N descriptors. LODDS performs better than N descriptors for predicting survival and benefits of PORT in resected MPM, and it could be considered as a potential parameter to compensate for defects in the 8th AJCC TNM staging for MPM.