Is there a prognostic difference among patients with T3N0-3M0 non-small-cell lung cancer with different T3 descriptors?

• Published in: European journal of cardio-thoracic surgery 2023-03, Vol.63 (3)
• Main Authors: Cai, Jing-Sheng, Yang, Fan, Wang, Xun
• Format: Article
• Language: English
• Subjects: Carcinoma, Non-Small-Cell Lung - surgery; Humans; Lung Neoplasms - surgery; Neoplasm Staging; Prognosis; Proportional Hazards Models
• ISSN: 1873-734X; 1873-734X
• DOI: 10.1093/ejcts/ezac558

Our goal was to evaluate the survival disparities among the patients with T3N0-3M0 non-small-cell lung cancer with different T3 descriptors and to further externally validate the current T3 category. Overall survival and cancer-specific survival were compared using the Kaplan-Meier method with a log-rank test. A univariable and a multivariable Cox regression model were performed to determine the prognostic factors. A series of 28,519 eligible cases was included. There were 17,971 cases with tumours that were larger than 5 cm but equal to or less than 7 cm (T-size group); 3,028 cases with tumours with chest wall/pericardium/phrenic nerve invasion (T-invasion group); 4,600 cases with tumours with additional tumour nodules in the same lobe (T-add group); and 2,900 cases with tumours that had at least 2 T3 descriptors (T-multiple group). The survival data indicated that patients in the T-add group had the best survival rates compared with other patients both in the entire cohort and in the subgroup analyses. Multivariable Cox models indicated that T3 descriptor was an important prognostic factor. Of the patients with different T3 descriptors, patients in the T-add group had the best prognosis, followed by patients in the T-size and in the T-invasion groups, and patients in the T-multiple group had the worst prognosis both in the pathological and clinical tumor-node-metastasis (TNM) stage cohorts. Patients with T3N0-3M0 non-small-cell lung cancer with different T descriptors had inconsistent survival rates. T-add yielded the best survivals, followed by T-size and T-invasion, and T-multiple was associated with the worst survivals. Our results were exploratory in nature and need to be further validated.