The Role Of The Number Of Involved Structures In Thymic Epithelial Tumors: Results From The ESTS Database.

Purpose: The role of the number of involved structures in thymic epithelial tumors(TET) is under investigation for inclusion in staging systems. Aim of this study is to analyse the prognostic role of the number of involved structures in patients included in the ESTS thymic database underwent surgical resection.

Methods: Clinical and pathological data of patients from ESTS thymic database underwent surgery for TET from 1/2000 to 7/2019 were reviewed and analysed. Among the records of 2506 patients operated on, data of patients with a pathological diagnosis of TET with other structures infiltration were extracted. Patients with incomplete data regarding tumour and surgical characteristics, thymomectomy and incomplete follow-up data were excluded. Pathological reports were adapted at the 8th TNM staging system for TET and Masaoka-Koga staging systems. Number of involved structures(NIS) included different infiltration layers counted in total (i.e., in case of lung infiltration, the previous mediastinal pleura invasion was also counted, resulting in two involved structures). Patients clinical data, tumour characteristics and number of involved structures were collected and correlated to cancer specific survival (CSS) using Kaplan Meier curves. The log-rank test was used to assess differences between subgroups. Multivariable model was built using logistic regression analysis.

Results: The final analysis was conducted on 303 patients. Clinical and pathological characteristics are reported in table 1. Histology resulted thymoma in 216(71.3%) and NET/thymic carcinoma(TC) in 87(28.7%)patients.
The pleura and the pericardium resulted the most frequent infiltrated structures, respectively in 198(65.3%) and 185(61.1%) cases, while lung, great vessels and phrenic nerve resulted involved in 96(31.7%),74(24.4%) and 31(10.2%) patients.
A single involved structure was present in 120(39.6%) patients, multiple in 183(60.4%)(range 2-7).
Median follow-up resulted 57.9 months(1-449), with 90 patients died, but only 46 related to TET recurrence appearance.
Five and 10years (5Y-10Y)CSS resulted 82% and 89%. At univariable analysis, neoadjuvant therapy, Masaoka stage 3, metastases absence, Myasthenia Gravis absence, pTNM I-II-III (versus IV), complete resection, thymoma histology (vs NET/Carcinoma) and NIS≤2 resulted favourable prognostic factors(table1).
Patients with >2 NIS presented a significant worse CSS compared to patients with ≤2 NIS: 5Y-10Y CSS 89.5% and 83.5% vs 93.2% and 91.2%(p=0.04)(figure1).
Moreover, patients with pericardium+other structures involvement presented a significant worse CSS compared to patients with pericardium alone: 5Y-10Y CSS 87.8% and 84.9% vs 96.4% and 92.9%(p=0.02).
Multivariable analysis confirmed as negative independent prognostic factors for CSS incomplete resection (HR:2.543,95%CI1.010-6.407,p=0.048) and NIS>2 (HR:1.395,95%CI1.021-1.905,p=0.036), while TC histology raises the statistical significance (HR:3.22,95%CI 0.985-11.200,p=0.053).

Conclusion: Our study showed that incomplete resection and a number of involved structures > 2 are negative independent prognostic factors in patients with infiltrative TET, suggesting that the number of involved structures may be taken in account for prognosis stratification and should be considered for further staging systems.

Identify the source of the funding for this research project: none