Background Immune checkpoint inhibitors have improved clinical responses and overall survival for patients with non-small cell lung cancer (NSCLC). However, the response is not equal and known NSCLC biomarkers are not sufficient in predicting therapy outcome. Deep proteomic analysis of NSCLC patient‘s plasma treated with anti-PD-1-blockade using a state-of-the-art data independent acquisition mass spectrometry (DIA-MS) is a powerful and unbiased way of identifying protein signatures associated with disease stage or response to treatment. However, to unravel these associations large-scale omics data should be analyzed with respect to available clinical information. To achieve this goal, we have used an approach previously applied by Uhlen et al., 2017¹ for transcriptomic datasets. In this approach survival data is used to set the most optimal thresholds for candidate biomarkers.

Methods 125 plasma samples were analyzed by capillary flow liquid chromatography coupled to DIA-MS. Data were extracted with latest Spectronaut^TM and proteins were quantified. Each recorded protein intensity was used as a threshold for two groups of samples for which Kaplan-Meier estimates were generated using ‘survival’² package in R. Benjamini-Hochberg correction was applied and p-values with corresponding intensity cut-offs were extracted to generate panels of potential biomarkers.

Results 125 plasma samples (in total 75 baseline and 50 after 8-weeks treatment) from advanced NSCLC patients treated with an anti-PD-1 inhibitor following at least 1 prior line of treatment were analyzed. 727 unique proteins were quantified across all samples. Data analysis was performed separately for each line of treatment and treatment status resulting in more than 100’000 p-values. For each group, panels of proteins with best performance in separating progression free survivals were defined at FDR of 0.10, giving 64 unique proteins which were mapped to acute phase response, platelet degranulation and complement activation. Several of these proteins were listed in the Early Detection Research Network database of the National Cancer Institute, and one of them – LYPD3, was a potential therapeutic target in a preclinical study for NSCL treatment.³ Selected proteins were then used to cluster patients into cohorts that showed association with the response to therapy.

Conclusions Deep proteomic profiling of plasma samples using DIA-MS in conjunction with clinical outcome enables a holistic and stringent analysis of potential circulating biomarkers. Such analysis generates functional insights into the plasma proteome that enable deeper understanding and comprehensive integration of clinical data with proteomics markers at different disease stages and treatment phases.

References

1. Uhlen M, Zhang C, Lee S, Sjöstedt E, Fagerberg L, Bidkhori G, Benfeitas R, Arif M, Liu Z, Edfors F, Sanli K, von Feilitzen K, Oksvold P, Lundberg E, Hober S, Nilsson P, Mattsson J, Schwenk J.

2. Therneau TM, Grambsch PM. Modeling Survival Data: Extending the Cox Model. Springer. 2000, New York, ISBN 0-387-98784-3.

3. Willuda J, Linden L, Lerchen H, Kopitz C, Stelte-Ludwig B, Pena C, Lange C, Golfier S, Kneip C, Carrigan P E, Mclean K, Schuhmacher J, von Ahsen O, Müller J, Dittmer F, Beier R, El Sheikh S, Tebbe J, Leder G, Apeler H, Jautelat R, Ziegelbauer K, Kreft B, Preclinical Antitumor Efficacy of BAY 1129980-a Novel Auristatin-Based Anti-C4.4A (LYPD3) Antibody-Drug Conjugate for the Treatment of Non-Small Cell Lung Cancer. Mol Caner Ther 2017;16(5):893–904