Comparison of the T descriptor of the 7th and 8th editions of the TNM staging system for non-small cell lung cancer after radiotherapy: An analysis from the TCP modelling perspective

Abstract

The tumour, node, metastasis (TNM) staging system employs a configuration that depicts the anatomic extent of cancer, thereby enhancing the exchange of information between oncologists. This study aims to incorporate the T descriptor of the TNM classification system into a tumour control probability (TCP) model. A mechanistic TCP model based on the linear-quadratic concept combined with the Poisson statistic was developed to assess variations in local tumour control according to the 7th and 8th editions of the TNM classification system. The simulation involved a single cohort of 354 patients with non-small cell lung cancer (NSCLC). The estimated mean gross tumour volumes (GTV) were 2.3–500 cm3 and 0.4–350 cm3 in accordance with the recommendations of the 7th and 8th editions, respectively. The predicted TCP for the T1a vs. T1b group after applying 2 Gy in 30 fractions was 48% ± 1.09% vs. 43% ± 1.05% according to the 7th edition and 51.5% ± 0.80% vs. 47.8% ± 0.85% according to the 8th edition. The differences in the predicted TCP between the adjacent T category groups in the 8th edition were all significant. Similar findings were observed in the 7th edition, except for the T3 vs. T4 stage. The TCP model outcomes based on the T descriptors of the 8th TNM revealed an enhanced model fit, a higher R2 and a lower RMSE compared with those of the 7th TNM, implying the superior discriminatory ability of the later edition.