Plasma PD-L1 as a biomarker in the clinical management of glioblastoma multiforme-a retrospective cohort study

Background and objectives: Glioblastoma multiforme (GBM) is the most aggressive, malignant, and therapy-resistant tumor of the brain. Blockade therapy targeting the programmed cell death protein 1 (PD-1)/programmed death ligand (PD-L1) axis is currently under investigation for the clinical management of the GBM. This study has quantified the plasma levels of PD-L1 as a biomarker for the clinical management of GBM.

Methods: A cohort (n = 128) of Pakistani adult glioblastoma patients together with age- and sex-matched healthy controls was used for quantification of pre-surgery levels of plasma PD-L1. PD-L1 protein and mRNA were measured by PD-L1 platinum enzyme-linked immunosorbent assay and quantitative real-time PCR, respectively. Receiver operating characteristic (ROC) curve analysis was used to compute area under the curve (AUC) for specificity and sensitivity analyses. The Kaplan-Meier survival analysis was employed to compute overall survival.

Results: PD-L1 protein and mRNA were significantly higher in GBM compared to the healthy controls (p < 0.0001). Mean PD-L1 concentration for the GBM was found to be 48.98 ± 2.290 pg/ml compared to 27.63 ± 1.281 pg/ml for controls. Gene expression analysis showed statistically significant upregulation (p < 0.0001) of PD-L1 in blood of GBM compared to healthy controls. Plasma PD-L1 showed an AUC of 0.840 (p < 0.0001; 95% CI = 0.7716 to 0.9090) where a cutoff value higher than 46 pg/ml demonstrated 100% specificity and 57.81% sensitivity. Higher pre-surgery levels of PD-L1 were found to be associated with overall poor survival [p < 0.0001; HR (log-rank) = 0.08; 95% CI = 0.04 to 0.15]. Age, gender, and ethnic background were not found to be associated with plasma PD-L1 levels.

Conclusion: The study concludes that blood-based measurements of PD-L1 in GBM can be a promising prognostic marker and therapeutic target besides a rapid and relatively non-invasive screening tool for routine clinical management. Future work extending the analysis to larger cohorts through multi-center collaborations involving pre-treatment and post-treatment groups is required to fully explore the usefulness of circulating PD-L1 for effective clinical applications.