Developing and Validating an Integrated Gross Tumor Volume (GTV)-TNM Stratification System for Supplementing Unresectable Locally Advanced Non-small Cell Lung Cancer Treated With Concurrent Chemoradiotherapy

Abstract PurposeThe gross tumor volume (GTV) could be an independent prognostic factor for unresectable locally advanced non-small cell lung cancer (LANSCLC). We aimed to develop and validate a novel integrated GTV-TNM stratification system to supplement LANSCLC sub-staging in patients treated with concurrent chemoradiotherapy (CCRT).MethodsWe performed a retrospective review of 340 patients with unresectable LANSCLC receiving definitive CCRT. All included patients were divided into two randomized cohorts. Then the Kaplan-Meier method and Cox regression were calculated to access the prognostic value of the integrated GTV-TNM stratification system, which was further validated by the area under the receiver operating characteristic curve (AUC) score and F1-score.ResultsThe optimal outcome-based GTV cut-off values (70 and 180 mL) of the modeling cohort were used to determine each patient’s integrated GTV-TNM stratum in the whole cohort. Our results indicated that a lower integrated GTV-TNM stratum could had better overall survival and progression-free survival (all P<0.001), which was recognized as an independent prognostic factor. Also, its prognostic value was robust in both the modeling and validation cohorts. Furthermore, the prognostic validity of the integrated GTV-TNM stratification system was validated by significantly improved AUC score (0.636 vs. 0.570, P=0.027) and F1-score (0.655 vs. 0.615, P<0.001), compared with TNM stage. The discordance rates between the GTV-TNM stratum and TNM stage were 49.3% and 49.6% in the modeling and validation cohort, respectively.ConclusionsWe proposed a novel integrated GTV-TNM stratification system to supplement unresectable LANSCLC sub-staging due to its prognostic value independent of TNM stage and other clinical characteristics, suggesting that it could be considered in pretreatment assessment and individual treatment decision-making process.