Prognostic and predictive significance of soluble programmed death ligand 1 in bronchoalveolar lavage fluid in stage IV non-small cell lung cancer

Abstract

Background: Patients with non-small cell lung cancer (NSCLC) have been shown to exhibit elevated levels of soluble programmed death-ligand 1 (sPD-L1) in the blood, associated with poor survival in NSCLC. The bronchoalveolar lavage fluid (BALF) composition reflects the tumor microenvironment of lung cancer. In this study, we investigated sPD-L1 levels in BALF and its role as a prognostic and predictive marker in patients with stage IV NSCLC.

Methods: We prospectively obtained BALF from lung cancer patients who underwent bronchoscopy between January 2020 and September 2022 at Chungnam National University Hospital (CNUH). Finally, 94 NSCLC stage IV patients were included in this study. Soluble PD-L1 levels in BALF were measured using a human PD-L1 Quantikine ELISA kit.

Results: The correlation between PD-L1 expression in tumor cells and sPD-L1 in BALF was weakly positive (rho =0.314, P=0.002). The median overall survival (OS) of the low sPD-L1 in BALF group was 16.47 months [95% confidence interval (CI): 11.15-21.79 months], which is significantly longer than 8.87 months (95% CI: 0.0-19.88 months, P=0.001) in the high sPD-L1 in BALF group. In 64 patients treated with or without immune checkpoint inhibitors (ICIs), sPD-L1 in BALF was significantly associated with progression-free survival (PFS) and OS. In the subgroup analysis of 31 patients treated with ICI, the objective response rate (ORR) in the low sPD-L1 BALF group was significantly higher than in high sPD-L1 in BALF group (ORR: 60.9% vs. 12.5%, P=0.02).

Conclusions: Soluble PD-L1 in BALF is a potential prognostic indicator for patients with stage IV NSCLC and a predictive marker for ICI treatment response.

Keywords: Non-small cell lung cancer (NSCLC); biomarker; bronchoalveolar lavage fluid (BALF); soluble programmed death-ligand 1 (sPD-L1).

Figure 1

Flow diagram of the patients included in the study. *, we analyzed the overall survival of these patients (N=94); ^†, subgroup analysis was performed for patients with or without ICI treatment (N=64); ^‡, subgroup analysis was performed for patients who received ICI treatment (N=31). BALF, bronchoalveolar lavage fluid; CNUH, Chungnam National University Hospital; SCLC, small cell lung cancer; NSCLC, non-small cell lung cancer; ICI, immune checkpoint inhibitor.

Figure 2

Distribution and correlation analysis of PD-L1 and sPD-L1 expression in BALF. (A) Scatterplot showing sPD-L1 levels in BALF in all patients (N=94). (B) Graph showing weakly positive correlation between PD-L1 expression in tumor cells and sPD-L1 levels in BALF (P=0.002, Spearman’s rho value =0.314). (C) Mean sPD-L1 in BALF based on tumor PD-L1 expression. (D) Mean tumor PD-L1 expression based on sPD-L1 in BALF. BALF, bronchoalveolar lavage fluid; PD-L1, programmed death-ligand 1; sPD-L1, soluble programmed death-ligand 1.

Figure 3

OS in all patients based on soluble PD-L1 levels in BALF (N=94). (A) ROC curve of sPD-L1 levels for survival rate in total patients (N=94) showing an AUC value of 0.626 (P=0.045) based on the 7.35 pg/mL cutoff. (B) Graph showing significantly longer median OS of the low sPD-L1 in BALF group (16.47 months; 95% CI: 11.15–21.79) than in the high sPD-L1 in BALF group (8.87 months; 95% CI: 0.0–19.88). AUC, area under the curve; OS, overall survival; sPD-L1, soluble programmed death-ligand 1; PD-L1, programmed death-ligand 1; BALF, bronchoalveolar lavage fluid; ROC, receiver operating characteristic; CI, confidence interval.

Figure 4

PFS and OS in patients with or without ICI treatment (N=64). (A) The median PFS of the low sPD-L1 in BALF group was 10.20 months (95% CI: 6.45–13.95 months), which was significantly longer than 3.90 months (95% CI: 1.38–6.43 months) in the high sPD-L1 in BALF group (P=0.001). (B) The median OS of the low sPD-L1 in BALF group was 20.57 months (95% CI: 13.86–27.27 months), which was significantly longer than 10.60 months (95% CI: 5.87–15.33 months) in the high sPD-L1 in BALF group (P=0.003). PFS, progression-free survival; sPD-L1, soluble programmed death-ligand 1; BALF, bronchoalveolar lavage fluid; OS, overall survival; ICI, immune checkpoint inhibitor; CI, confidence interval.

Figure 5

ORR, PFS, and OS in patients who received immunotherapy (N=31). (A) Bar graph showing significantly higher ORR in patients with low sPD-L1 in BALF (60.9%) than in those with high sPD-L1 in BALF (12.5%) (P=0.02). (B) Graph showing significantly longer median PFS in the low sPD-L1 in BALF (7.57 months; 95% CI: 3.65–11.48) than in the high sPD-L1 in BALF group (2.0 months; 95% CI: 0.0–5.74). (C) Graph showing significantly longer median OS in the low sPD-L1 in BALF (not reached) than in the high sPD-L1 in BALF group (4.17 months; 95% CI: 0.0–14.38). ORR, objective response rate; sPD-L1, soluble programmed death-ligand 1; BALF, bronchoalveolar lavage fluid; PFS, progression-free survival; OS, overall survival; CI, confidence interval.

Figure 6

Graphical summary of the study. NSCLC, non-small cell lung cancer; BAL, bronchoalveolar lavage; PD-L1, programmed death-ligand 1; sPD-L1, soluble programmed death-ligand 1.