Patients and Clinical Information
This retrospective study was approved by the Institutional Ethics Review Committee of the First Affiliated Hospital of Zhengzhou University. Patients with TETs who underwent preoperative contrast-enhanced CT from October 2013 to October 2020 at the First Affiliated Hospital of Zhengzhou University were recruited. Patients were identified by querying the electronic medical record database at the Department of Thoracic Surgery and Pathology. Of the 269 patients, 41 did not receive preoperative contrast-enhanced CT scan and 19 were excluded due to preoperative neoadjuvant chemotherapy, radiotherapy or other treatments. A total of 209 patients with TET were included in the final analysis (Figure 1). Clinical information, including gender, age, smoking history, drinking history, date of surgery, presence of MG before surgery, the content of tumor markers in peripheral blood (carcinoembryonic antigen (CEA), normal value <5 μg/L; carbohydrate antigen 724 (CA724), normal value <8.20 U/ml; carbohydrate antigen 199 (CA199), normal value <37.00 U/ml; carbohydrate antigen 125 (CA125), normal value <35.00 U/ml; cytokeratin-19 (CK19) fragment, normal value <3.30 ng/ml; neuron-specific enolase, normal value <17.00 ng/ml; and NSCLC antigen 21-1, 0.1 < normal value < 3.3 ng/ml), postoperative recurrence, postoperative complications, patient follow-up dates and outcomes, Masaoka-Koga stage and WHO classification of thymoma and TC and TNM stage of TETs, were retrospectively obtained from the hospital record database. All patients underwent thymic tumor resection, and postoperative pathological examination results were available. Pleural and pulmonary nodules were diagnosed using a combined identification strategy based on pathological findings or follow-up examination (11, 12). Postoperative pathological analysis was performed according to standard procedures (13, 14). Macroscopic and microscopic pathological examination revealed pleura and pulmonary nodules around the tumor. Most of the distant nodules were high-density hard nodules. Some cases were confirmed by needle biopsy and the rest were followed up for at least one year. Patients were followed up in the outpatient department. In the first year, the first outpatient follow-up was 1 month after surgery, the second was 4 months and the third was 10 months postoperatively, every 1 year for the second year and yearly thereafter if the patient recovered well. Outpatient follow-up examination items included routine 64-slice spiral CT scan and tumor markers such as CEA, CA724, etc. A total of 164 patients were followed up from the date of surgery until 2022. Patients were divided into two subgroups according to average age: ≤50 and >50 years.
CT Imaging
All the patients underwent preoperative cross-section spiral CT examinations to evaluate suspected mediastinal tumors. Chest CT examinations were performed using a 64-row detector CT scanner (Toshiba Aquilion 64, Otawara-shi, Japan) or a 16-row detector CT scanner (Toshiba Aquilion 16, Shipu, Tokyo, Japan). All examinations were performed after administration intravenous contrast medium. All images were taken with patients in the supine position and breathing quietly. Acquisition parameters were: 0.6-1.25 mm detector collimation. Images were viewed at window settings optimized for assessment of mediastinum (window width, 300-400 HU; window level, 30-50 HU) and lungs (window width, 1,000-1,500 HU; window level, -600 to -700 HU). The area from the thoracic inlet to caudally include the kidney was scanned. All images were reviewed by two radiologists with more than 10 years of experience in CT imaging who were experts in oncologic and thoracic imaging. The radiologists were blinded to clinical and histopathological data of patients to avoid bias. Image interpretation criteria used standard reporting terms defined by ITMIG for anterior mediastinal masses suspected to be TETs (15). CT imaging features were evaluated, including lesion location in the mediastinum (thoracic inlet: anterior superior, anterior inferior), size (mm), tumor shape (focal mass, polygonal mass), calcification (yes; no), tumor internal density (homogeneous, heterogeneous), tumor contour (round, lobulated, irregular), with or without fat plane separating tumor from vessels (yes or no), infiltration of surrounding fat (yes or no), adjacent lung changes (yes or no), degree of abutment of vessel circumference (<50%, ≥50%), presence of effusion (yes or no), lymphadenopathy (yes or no), pleural nodularity (yes or no) and lung nodule (yes or no).
Masaoka-Koga and TNM Staging Systems and WHO Histological Classification
In 2017, ITMIG and IASLC established the 8th edition of TNM staging (6-8), which applies to thymic epithelial neoplasms, TC, thymic neuroendocrine neoplasms, and compound thymic carcinoma. TNM stage Ⅲ and Ⅳ lesions were considered more aggressive and had a poorer prognosis than stage I and II lesions. There were 172 cases of thymomas, 25 of TCs and 12 of NETTs. The Masaoka-Koga staging system is suitable for thymoma and TC. The final pathological stage (16-18) was determined according to the modified Masaoka-Koga staging system, combined with surgical records and pathological reports. Masaoka-Koga stage III and IV lesions were considered to be more aggressive than stage I and II lesions. Thymomas and TCs were classified as type A, AB, B1, B2, B3 and C according to 2015 revised WHO histological classification (19, 20). WHO types B2, B3 and C were considered more malignant than types A, AB and B1. WHO classification type B2, B3 and C and A, AB and B1 were classified as high-risk and low-risk groups, respectively. When a tumor exhibited multiple histological features, it was classified according to major components (21).
Statistical Analysis
Independent two-sample t-test was used to compare continuous variables and the χ2 test or Fisher’s exact test was used to compare categorical variables. Age was assessed as a categorical variable: ≤50 and >50 years. P<0.05 was considered statistically significant. Binary logistic regression analysis was used to evaluate CT imaging features, clinical characteristics and MG as risk factors for TNM, Masaoka-Koga staging system and WHO histological classification. To determine which factors include in the final model, forward stepping selection was adopted and the significance level was set as P = 0.05. The odds ratio (OR) and 95% confidence interval (CI) were used to measure the strength of the association of risk factors. All data were analyzed using the SPSS Software package (Version 24.0, SPSS, Inc., Chicago, IL, USA) and GraphPad Prism (Version 8.0.2).