Clinical factors associated with high PD-L1 expression in patients with early-stage non-small cell lung cancer.

Abstract

Background: Superior outcomes have been obtained for neoadjuvant treatment with immune checkpoint inhibitors (ICI) plus chemotherapy over neoadjuvant chemotherapy alone, especially in patients with high programmed cell death ligand 1 (PD-L1) expression. However, it is not always possible to obtain sufficient tumor specimens for biomarker testing before surgery. In this study, we explored clinical factors that can predict high PD-L1 expression.

Methods: We retrospectively enrolled 340 lung cancer patients who received pulmonary resection between 2014 and 2023 and who had PD-L1 expression data. Chi-squared tests and logistic regression analyses were used to identify clinical factors associated with high PD-L1 status.

Results: Univariable and multivariable analyses revealed that smoking, high maximum standardized uptake value (SUVmax) of 18F-fluorodeoxyglucose positron emission computed tomography (18F-FDG PET/CT), and high plasma fibrinogen are independent predictors of high PD-L1 expression. A predictive score for high PD-L1 expression (ranging from 0 to 3) was developed based on these parameters. Notably, only 5% of patients with a score of 0 exhibited high PD-L1 expression, whereas this proportion increased to 53% for patients with a score of 3.

Conclusion: These results showed that plasma fibrinogen, smoking history, and SUVmax are predictors of high PD-L1 expression, providing a basis for identifying patients expected to benefit from neoadjuvant ICI treatment.