The Prognostic Role of the Number of Involved Structures in Thymic Epithelial Tumors: Results from the ESTS Database

BackgroundThe role of the number of involved structures (NIS) in thymic epithelial tumors (TETs) has been investigated for inclusion in future staging systems, but large cohort results still are missing. This study aimed to analyze the prognostic role of NIS for patients included in the European Society of Thoracic Surgeons (ESTS) thymic database who underwent surgical resection.MethodsClinical and pathologic data of patients from the ESTS thymic database who underwent surgery for TET from January 2000 to July 2019 with infiltration of surrounding structures were reviewed and analyzed. Patients' clinical data, tumor characteristics, and NIS were collected and correlated with CSS using Kaplan-Meier curves. The log-rank test was used to assess differences between subgroups. A multivariable model was built using logistic regression analysis.ResultsThe final analysis was performed on 303 patients. Histology showed thymoma for 216 patients (71.3%) and NET/thymic carcinoma [TC]) for 87 patients (28.7%). The most frequently infiltrated structures were the pleura (198 cases, 65.3%) and the pericardium in (185 cases, 61.1%), whereas lung was involved in 96 cases (31.7%), great vessels in 74 cases (24.4%), and the phrenic nerve in 31 cases (10.2%). Multiple structures (range, 2-7) were involved in 183 cases (60.4%). Recurrence resulted in the death of 46 patients. The CSS mortality rate was 89% at 5 years and 82% at 10 years. In the univariable analysis, the favorable prognostic factors were neoadjuvant therapy, Masaoka stage 3, absence of metastases, absence of myasthenia gravis, complete resection, thymoma histology, and no more than two NIS. Patients with more than two NIS presented with a significantly worse CSS than patients with no more than two NIS (CSS 5- and 10-year rates: 9.5% and 83.5% vs 93.2% and 91.2%, respectively; p = 0.04). The negative independent prognostic factors confirmed by the multivariable analysis were incomplete resection (hazard ratio [HR] 2.543; 9