Correlating the effects of VeriStrat result, anti-PD1 therapy, and neutrophil-to-lymphocyte ratio (NLR) on progression-free survival (PFS) in patients with stage IV non-small cell lung cancer (NSCLC).

Abstract

e20660 Background: Biodesix VeriStrat (VS) is a mass spectrometry assay found to be prognostic in previously treated and front-line NSCLC patients. A label assigns a “good” or “poor” result and is believed to in part represent host inflammatory reaction to tumor. This study looks to correlate VS good (VS-G) or poor (VS-P) label with the neutrophil to lymphocyte ratio (NLR) and outcomes in front-line NSCLC patients who were screened for actionable mutations as part of a companion diagnostic. Methods: Retrospective chart review was conducted of 76 patients with advanced NSCLC who had pretreatment VS testing. Included patients had at least 3-month follow up. Chi squared analysis was used to correlate VS label with patient characteristics and Cox proportional hazards was used to correlate PFS with VS label, NLR and other baseline characteristics in patients receiving anti-PD1 targeted therapy as part of their first treatment. Results: Median PFS was 6.31 months (follow up range 3-17 months). Of 76 tested patients, 23 never received treatment and 47 received anti-PD1 therapy. 51 had VS-G and 23 had VS-P (2 indeterminate). Using two standard NLR cutoffs of 3.5 and 5, 29 and 47, and 42 and 34 had values above and below the cutoff, respectively. VS-G vs VS-P correlated with performance status (PS) as well as NLR at both cut points and on a continuous scale. (P = 0.031, 0.043, 0.045, and 0.056 respectively, VS-G correlating with good PS and low NLR). There was no correlation with PDL1 or body mass index. On multivariate analysis VS label and NLR were also associated with PFS and OS (P = < 0.001 and P = 0.046). Among a subset of 47 patients treated with ICIs, VS-G and NLR < 5 were associated with longer PFS (P = 0.010 and 0.010) though PDL1 result was not. Conclusions: This is the first description of the performance of the VS mass spectrometry assay in front-line patients whose regimen included an anti-PD1 agent. The VS label was able to predict for PFS in this patient population despite relatively short follow up. VS-G label also correlated with low NLR by multiple methods analyzed. Further evaluation of the impact of cytotoxic chemotherapy to anti-PD1 is warranted.