Expression of programmed death-1 (PD-1) and programmed death ligand-1 (PD-L1) is associated with poor glioblastoma outcomes.
Expression of programmed death-1 (PD-1) and programmed death ligand-1 (PD-L1) is associated with poor glioblastoma outcomes, according to results presented at the 20th Annual Scientific Meeting of the Society for Neuro-Oncology, held November 19-22 in San Antonio, Texas.
“PD-L1 expression in [glioblastoma] is frequent but is largely confined to a minority subpopulation-similar to other malignancies that have been profiled for PD-L1 expression,” reported lead study author Edjah K. Nduom, MD, of the MD Anderson Cancer Center in Houston, and coauthors, in a poster presentation. “Higher expression of PD-L1 is correlated with worse outcome, and proper stratification of PD-L1–positive and negative patients may become an important criterion for high-quality immunotherapeutic trials in [glioblastoma]. Expression of PD-L1 on infiltrating lymphocytes also suggests that a unique immunosuppressive pathway may operate in [glioblastoma].”
Immune checkpoint-inhibitor therapies that target immune response–inhibiting CTLA-4 and PD-1/PD-L1 can facilitate glioblastoma tumor regression and phase II clinical trials are ongoing, Dr. Nduom’s team noted. But whether or not these tumor-regression responses are more likely in patients whose tumors express PD-L1 remains controversial.
“At issue is the validation of PD-L1 biomarker assays and prognostic impact,” Dr. Nduom and colleagues reported.
The team measured PD-L1 expression in 94 patients’ glioblastoma tumors, using immunohistochemical assays.
“We determined the total number of PD-L1-expressing cells within the glioblastomas and validated this finding using ex vivo glioblastoma flow cytometry, with further analysis of T-cell populations,” they reported. They then compared these data with patient survival times, cytotoxicity markers, mRNA, and protein expression data sets from The Cancer Genome Atlas (TCGA).
“The majority of glioblastoma patients (61%) had tumors with at least 1% PD-L1–positive cells,” the coauthors noted in an accompanying abstract.
Median survival time among patients in the PD-L1 high-expressing group was 11.4 months compared with 14.9 months for the low-expressing group (P = .023), median survival time for patients in the PD-1 high-expressing group was 10.6 months compared with 14.7 months for the low-expressing group (P = .028), the team reported. When both PD-L1 and PD-1 expression levels were analyzed together, based on TCGA datasets for PD-1 and PD-L1 mRNA expression levels, the median survival time for patients with high-expressing tumors was 9.7 months vs 15.9 months for the low-expressing group (P = .003).
“Using mRNA expression from the TCGA database, we determined that there was a significant association between PD-1 mRNA expression and mRNA expression of [T cell markers] CD3D, CD3E, CD3G, CD4, CD8A, and CD8B,” the researchers reported. “Significant correlations were found between PD-L1 expression and CD3G and CD4.”
Cytotoxicity marker expression levels in glioblastoma also predicted survival times. High expression of granzyme A was associated with a median 10.5-month survival time vs 14 months in the low-expressing group, the team reported. High expression of pore-forming protein (perforin-1) was associated with a median survival time of 10.6 months vs 15.1 months for the low-expression group. Expression of perforin-1 correlated positively with PD-L1 and granzyme A expression.
“Expression of PD-L1 is prognostic by multivariate analysis for a negative glioblastoma outcome by both immunochemistry and mRNA expression analysis of TCGA data,” the team concluded.