PD-L1 expression is often used to determine the eligibility of patients for treatment with anti-PD-1 or anti-PD-L1 antibodies. Nonetheless, many trials reveal responses to these agents in patients with PD-L1-negative disease. Now, data from a retrospective study demonstrate that N-linked glycosylation of PD-L1, which interferes with antibody-based detection methods, might explain this inconsistency.
Lead author Mien-Chie Hung explains “According to previous studies, PD-L1 is a relatively small glycoprotein with heavy N-glycosylation on its extracellular domain”. Therefore, “antibodies generated to recognize pure recombinant PD-L1 peptide or protein antigens could have a differential binding affinity to the PD-L1 glycoprotein in human tissue owing to N-glycosylation”.
To investigate this effect, researchers analysed PD-L1 expression before and after deglycosylation with peptide-N-glycosidase F in pathological specimens from 200 patients with solid tumours, including breast, lung, colon, prostate and pancreatic cancers. A >2-fold increase in H-score (reflecting PD-L1-positivity) was observed in 47.5% of samples following deglycosylation. Subsequent analysis of non-small-cell lung cancer specimens obtained from three independent cohorts with known PD-L1 tumour proportion scores (TPS) confirmed the clinical relevance of PD-L1 glycosylation: 22.5% and 16% of patients had increases in TPS from <1% to ≥5% and ≥49%, respectively, following deglycosylation — scores that would have made these patients eligible for treatment with nivolumab or pembrolizumab.
An analysis of samples from 95 patients who received anti-PD-1 or anti-PD-L1 antibodies revealed statistically significant positive correlations of both PD-L1 H score and PD-L1 TPS with overall survival after (P = 0.033 and P = 0.005, respectively) but not before (P = 0.80 and P = 0.29) deglycosylation.
Hung summarizes “our method identified a population (~16%) of anti-PD-L1 responders whose PD-L1 status changed from negative to positive”, adding that “deglycosylation might be an effective method of improving the predictive power of PD-L1 as a biomarker”.
References
Original article
Lee, H. H. et al. Removal of N-linked glycosylation enhances PD-L1 detection and predicts anti-PD-1/PD-L1 therapeutic efficacy. Cancer Cell https://doi.org/10.1016/j.ccell.2019.06.008 (2019)
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Sidaway, P. Deglycosylated PD-L1 might be a better biomarker. Nat Rev Clin Oncol 16, 592 (2019). https://doi.org/10.1038/s41571-019-0261-x
Published:
Issue Date:
DOI: https://doi.org/10.1038/s41571-019-0261-x
This article is cited by
-
A comparability study of natural and deglycosylated PD-L1 levels in lung cancer: evidence from immunohistochemical analysis
Molecular Cancer (2021)
-
Proximity ligation assay mediated rolling circle amplification strategy for in situ amplified imaging of glycosylated PD-L1
Analytical and Bioanalytical Chemistry (2021)
-
Establishing peripheral PD-L1 as a prognostic marker in hepatocellular carcinoma patients: how long will it come true?
Clinical and Translational Oncology (2021)
-
The impact of PD-L1 N-linked glycosylation on cancer therapy and clinical diagnosis
Journal of Biomedical Science (2020)