Prognostic and predictive role of soluble programmed death ligand-1 in head and neck cancer

Abstract

Objectives: The aim of the study was to investigate clinical significance of soluble PD-L1 (sPD-L1) serum level in head and neck cancer and to evaluate its role as a possible prognostic and predictive biomarker. Methods: A prospective analysis of sPD-L1 levels in 60 patients diagnosed and treated due to malignant and non-malignant lesions in the region of head and neck was performed in peripheral blood by an ELISA test. Results: The range of sPD-L1 in the study group was 0.16–1.63 ng/mL, mean 0.64 ± 0.32. There were no differences in the mean sPD-L1 regarding patients’ age, sex, and the localization of the lesion. Statistically significant difference was revealed in the average sPD-L1 level (p = 0.006) depending on the histopathological advancement of the lesions, 0.704 ± 0.349 and 0.512 ± 0.177 respectively in the malignant and benign group. The separate analysis of laryngeal lesions confirmed statistical difference in sPD-L1 (p = 0.002) for the malignant lesions (0.741 ± 0.353) compared with the benign (0.489 ± 0.175). The sPD-L1 level of 0.765 ng/mL or higher, revealed 35% sensitivity and 95.5% specificity for the diagnosis of head and neck malignant lesions (AUC = 0.664, 95% CI 0.529‒0.8, p-value = 0.039). The 1-year DFS was 83.3% in the group of patients with low sPD-L1 levels (< 0.765 ng/mL) and 53.8% in patients with high sPD-L1 (≥0.765 ng/mL). The 2-year OS were 68% and 69.2% respectively in both groups. The log-rank test confirmed statistically significant prognostic value of sPD-L1 level for 1-year DFS (p-value = 0.035). Conclusions: sPD-L1 is a promising prognostic and early recurrence predictive biomarker for head and neck cancers, most significantly for laryngeal lesions. Level of evidence: 3.

Keywords

• Soluble biomarker
• Programmed death ligand 1
• Head and neck cancer
• Laryngeal cancer
• sPD-L1