Consolidation immunotherapy of immune checkpoint inhibitor (ICI) following chemoradiotherapy (CRT) is the current standard of care for patients with unresectable locally advanced non-small cell lung cancer (NSCLC) as it improves both progression-free survival and overall survival. However, a substantial proportion of patients still experience disease recurrence despite consolidation ICI. It is important for personalized treatment to predict the efficacy of consolidation ICI. PD-L1 expression is used as a predictive biomarker for ICI response and efficacy in advanced NSCLC, but its role in patients with stage III disease is unclear. One important reason is PD-L1 testing performed on pre-CRT tissue may not reflect changes in PD-L1 expression after CRT. CT-based radiomics approaches have been successfully applied to generate imaging biomarkers as decision support tools for clinical practice. The hypothesis of this study is that CT radiomics model can assess PD-L1 status after CRT and predict the efficacy of CRT combined with ICI in unresectable locally advanced NSCLC.

This is an observational longitudinal prospective study.

CT scan is performed before radiotherapy, during radiotherapy and at the end of radiotherapy in patients with unresectable locally advanced NSCLC who undergo CRT. Radiomic features were extracted from CT images and baseline PD-L1 expression is assessed. CT-based radiomics models is developed to assess PD-L1 expression and predict the efficacy of ICI.

Inclusion Criteria:

• Histological or cytological proven non-small cell lung cancer
• Unresectable stage III according to American Joint Committee of Cancer stage (the eighth edition)
• 18 years or older

Exclusion Criteria:

• Previous thoracic radiotherapy
• Palliative treatment

Patients diagnosed with stage III NSCLC undergo radical intervention.