Document resource
Background Immune checkpoint inhibitor (ICI) therapy has revolutionized the treatment of lung cancer, though only a subset of patients experience long-term response. High PDL-1 expression is utilized to triage treatment, but this alone does not determine efficacy. To this end, further molecular characterization of responders and non-responders is necessary. Blood-based proteomic and glycoproteomic biomarkers have shown promise in many cancers, though direct clinical applications remain limited.Methods The SomaScan and GlycoVision platforms were employed to assess n=10753 proteomic and n=607 glycopeptide biomarkers in advanced lung cancer patients (n=33 non-small cell lung cancer and n=2 small cell) treated with Pembrolizumab, Nivolumab or Durvalumab, either alone or in conjunction with chemotherapy (n=6 ICI and n=29 ICI+chemo). Pre-treatment samples were provided by the Translational Immunology Laboratory at the University of Sydney. The cohort consisted of 21 males and 14 females, with ages ranging from 56 to 83. The majority of patients were assessed as PDL-1 low (n=24 of 30 patients reported). Median progression-free survival (PFS) was 11.3 months (95% CI: 6.8-NR). Cox proportional hazards models and the Kolmogorov-Smirnov test at 6 and 12 months were utilized to assess biomarker significance individually, and the likelihood ratio test (LRT) was used to assess additivity between platforms.Results The SomaScan assay identified n=174 proteins individually associated with response, and GlycoVision identified n=20 glycoprotein biomarkers, including five glycopeptides from alpha-2-macroglobulin (A2MG). All combinations of these two sets of individual markers were assessed for additivity, and in total n=37 combinations reached interaction p-value < 0.05 and LRT p-value < 0.01. The Kaplan-Meier curves for the top six combinations are plotted for the subset of patients with PDL-1 low staining ( figure 1).Conclusions Blood-based proteomic and glycoproteomic biomarkers show promising associations with response in ICI-treated patients with advanced lung cancer. Furthermore, combinations of markers from the two platforms can be utilized in conjunction to segment the patient population into those expected to either progress quickly or have durable response. Ultimately, such biomarker combinations could be utilized to extend ICI treatment to currently unapproved populations, such as patients not expressing PDL-1.Ethics Approval The study was approved by The University of Sydney Ethics Board, approval number X18-0525 & 2019/ETH08954.Abstract 40 Figure 1Kaplan-Meier curves for the top six combinations of SomaScan and GlycoVision biomarkers