Soluble PD-L1 shows no association to relapse and overall survival in early stage non-small cell lung cancer (NSCLC)

• Published in: Lung cancer (Amsterdam, Netherlands) Netherlands), 2024-10, Vol.196, p.107955, Article 107955
• Main Authors: Mildner, F.O., Sykora, M.M., Hackl, H., Amann, A., Zelger, B., Sprung, S., Buch, M.L., Nocera, F., Moser, P., Maier, H., Augustin, F., Manzl, C., Kocher, F., Pircher, A., Lindenmann, J., Mykoliuk, I, Raftopoulou, S., Kargl, J., Wolf, D., Sopper, S., Gamerith, G.
• Format: Article
• Language: English
• Subjects: ADAM10 Protein - genetics; ADAM10 Protein - metabolism; Adult; Aged; Aged, 80 and over; Amyloid Precursor Protein Secretases - metabolism; B7-H1 Antigen - genetics; B7-H1 Antigen - metabolism; Biomarker; Biomarkers, Tumor; Carcinoma, Non-Small-Cell Lung - genetics; Carcinoma, Non-Small-Cell Lung - metabolism; Carcinoma, Non-Small-Cell Lung - mortality; Carcinoma, Non-Small-Cell Lung - pathology; Female; Humans; Lung Neoplasms - genetics; Lung Neoplasms - mortality; Lung Neoplasms - pathology; Male; Membrane Proteins - genetics; Membrane Proteins - metabolism; Middle Aged; Neoplasm Recurrence, Local; Neoplasm Staging; NSCLC; Prognosis; Relapse prediction; sPD-L1; Tumor immune microenvironment
• ISSN: 0169-5002; 1872-8332; 1872-8332
• DOI: 10.1016/j.lungcan.2024.107955

•sPD-L1 shows in vitro immunosuppressive capacity by reducing IFN-γ production in T cells.•ADAM10, a metalloproteinase associated with shedding of PD-L1, expression in the TCGA shows a correlation to OS in early stage NSCLC.•A high sPD-L1 before surgery showed a predictive potential in our first cohort but could not be validated in two independent control cohorts.•The high variance in measured levels of sPD-L1 with different ELISA assays displays the need for a standardization of sPD-L1 measurement.•Alternative sources, except the tumor, of sPD-L1. Cancer immune evasion is critical in non-small cell lung cancer (NSCLC) and has been targeted by immunotherapy. High soluble (s)PD-L1 is associated with reduced survival and treatment failure in advanced stages. Here we evaluated the effects of sPD-L1 on T cells, relapse free survival, and overall survival in early stage NSCLC. In vitro T cell stimulation was performed in the presence of sPD-L1 to evaluate its immunomodulatory activity. Data from The Cancer Genome Atlas (TCGA) were investigated for PD-L1 splice variants and enzymes involved in proteolytic cleavage (i.e. ADAM10). Plasma from 74 NSCLC (stage IA-IIIB), as well as an additional 73 (control cohort) patients was collected prior to curative surgery. Thereafter sPD-L1 levels from an immunosorbent assay were correlated with patient outcome. In vitro sPD-L1 inhibited IFN-γ production and proliferation of T cells and induced a terminal effector CD4 T cell subtype expressing CD27. Data from the TCGA demonstrated that elevated mRNA levels of ADAM10 is a negative predictor of outcome in NSCLC patients. To investigate the clinical relevance of these in vitro and TCGA findings, we quantified sPD-L1 in the plasma of early-stage NSCLC patients. In the first cohort we found significantly higher sPD-L1 levels in relapsing NSCLC patients, with a multivariate analysis revealing high sPD-L1 (>1000 pg/mL) as an independent predictor of survival. However, these findings could not be validated in two independent control cohorts. Although in vitro and TCGA data support the suppressive effect of sPD-L1 we were unable to translate this in our clinical setting. These results may be due to the small patient number and their heterogeneity as well as the lack of a standardized sPD-L1 ELISA. Our inconclusive results regarding the value of sPD-L1 in early stage NSCLC warrant assay validation and further investigation in larger (neo-)adjuvant trials.