An online clinical model and risk stratification system to predict progression-free survival for advanced non-small-cell lung cancer patients treated with PD-(L)1 inhibitor

• Published in: Heliyon Vol. 9; no. 10; p. e20465
• Main Authors: Tu, Zegui, Yu, Yang, Tian, Tian, Li, Caili, Luo, Jieyan
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.10.2023; Elsevier
• Subjects: Immunotherapy; Non-small-cell lung cancer; PD-(L)1 inhibitor; Progression-free survival; Risk assessment; Non-small-cell lung cancer; Progression-free survival; Risk assessment; PD-(L)1 inhibitor; Immunotherapy
• ISSN: 2405-8440; 2405-8440
• DOI: 10.1016/j.heliyon.2023.e20465

Our study aimed to build a risk stratification system predicting the progression-free survival (PFS) to classify patients into diverse prognostic subgroups for advanced non-small-cell lung cancer patients treated with PD-(L)1 inhibitor. 404 patients from our center were enrolled in this study and 70% patients (n = 282) were randomly assigned into the training cohort and other 30% patients (n = 122) into the validation cohort. A testing cohort contained 81 patients from other centers were used to assess the generalizability of model. Cox regression analyses were used to identify the most significant clinical parameters. The model's performance was assessed by using concordance index (C-index), calibration curves, Decision Curve Analyses (DCAs), net reclassification improvement (NRI), integrated discrimination improvement (IDI) analyses, and survival curve. Five clinical parameters were identified as the most significant predictors by using cox regression. We then integrated them into a Nomogram to Evaluate the relative PFS of ICIs Treatment (NEPIT). The C-index of NEPIT in the training cohort, the validation cohort and testing cohort was 0.789 (95%CI: 0.750–0.828), 0.745 (95%CI: 0.706–0.784), and 0.766 (95%CI: 0.744–0.788), respectively. The calibration curves presented a good congruence between the predictions and actual observations. The Decision Curve Analyses (DCAs) reflected positive net benefits can be obtained for NEPIT. The results from NRI and IDI analyses showed that the NEPIT could improve predictive power of TPS. In addition, the further constructed risk stratification system could effectively categorize patients into different risk subgroups. The tools developed in this study would have value in guiding the optimal patient selection for precision care.