RESUMO
OBJECTIVES: This study aimed to explore the prognostic disparity among T4N0-2M0 non-small-cell lung cancer (NSCLC) patients with different T4 descriptors. METHODS: T3-4N0-2M0 NSCLC patients were included. Patients were assigned to 7 subgroups: T3, T4 tumours with size larger than 70 mm (T4-size), T4 tumours with aorta/vena cava/heart invasion (T4-blood vessels), T4 tumours with vertebra invasion (T4-vertebra), T4 tumours with carina/trachea invasion (T4-carina/trachea), T4 tumours with additional tumour nodules in different lobes of ipsilateral lung (T4-add) and T4 tumours had at least 2 T4 descriptors (T4-multiple). Univariable and multivariable Cox analyses were used to explore the effect of T4 category on overall survival. Kaplan-Meier method with log-rank test was used to compare survival differences among subgroups. Propensity score matching was used to minimize the bias caused by imbalanced covariates between groups. RESULTS: A total of 41 303 eligible T3-4N0-2M0 NSCLC cases were included (17 057 T3 cases and 24 246 T4 cases). There were 10 682 cases, 573 cases, 557 cases, 64 cases, 2888 cases and 9482 cases in the T4-size, T4-blood vessels, T4-vertebra, T4-carina/trachea, T4-add and T4-multiple subgroups, respectively. Multivariable Cox analyses revealed that T4-add patients had the best prognosis in the entire cohort and in several subgroups. In the matched cohort of T4-add and T4-size and T4-add and T3, the survival of T4-add patients was superior to that of T4-size patients (P < 0.001) but was comparable to that of T3 patients (P = 0.115). CONCLUSIONS: Among NSCLC patients with different T4 descriptors, T4-add patients had the best prognosis. T4-add patients and T3 patients had similar survivals. Herein, we proposed that T4-add patients should be downstaged from T4 to T3 category. Our results served as a novel supplement to the proposals for the T category revision.