9056 Background: A pro-inflammatory diathesis as measured by the neutrophil-to-lymphocyte ratio (NLR) heralds an adverse disease course in non-small cell lung cancer (NSCLC). Whether the NLR identifies patients who derive a differential degree of benefit from immunotherapy versus chemotherapy is not known. Methods: This post hoc analysis used data from the phase III OAK trial, which randomized previously treated patients with NSCLC to receive atezolizumab or docetaxel. The main objective was to assess the differential impact of pre-treatment NLR on overall survival (OS) depending on the treatment modality. In addition, we assessed patients genomic characteristics according to inflammatory status using circulating free (cf)DNA NGS analysis. Results: A total of 600 and 575 patients with an available NLR were included in the atezolizumab and docetaxel cohort, with a median NLR of 4 (IQR: 2.6-6.7) for the pooled population. NLR ≥ 4 was associated with positive smoking status (88.6% vs 78.1%, p < 0.01), male sex (66.4% vs 57.6%, p = 0.01), worse performance status (71.3% vs 55.2%, p < 0.01), higher number of metastatic sites (63.2% vs 51.6%, p = 0.01), squamous histology (32.1% vs 21.4%, p < 0.01), and tissue KRAS mutation (30% vs 18.7%, p = 0.02), but not with PD-L1 expression, nor with tissue EGFR/ALK status. Pre-treatment NLR of ≥ 4 was more strongly associated with mortality following atezolizumab with an adjusted hazard ratio (HR) of 1.64 (95%CI:1.35-2.01) compared to docetaxel (HR 1.32, 95%CI: 1.08-1.60, multivariable (MVA) interaction p = 0.08. Exclusion of EGFR/ALK positive patients further increased the prognostic ability of baseline NLR in favor of atezolizumab (HR 1.67, 95%CI: 1.35-2.06), as compared with the docetaxel arm (HR 1.24, 95%CI: 1.02-1.52, MVA interaction p = 0.02). The HR for the risk of death for patients with NLR≥ 4/PD-L1 negative tumours (compared to NLR < 4/PD-L1 positive) was significantly higher in the atezolizumab cohort (HR 2.28, 95%CI: 1.72-3.03) than in the docetaxel cohort (HR 1.42, 95%CI: 1.08-1.86, MVA interaction p = 0.01). NGS pretreatment cfDNA data showed that patients with a high blood tumor mutational burden (cut-off 16 Mut/Mb) had a higher median NLR (4.6 vs 3.7, p = 0.01). After adjusting for multiple comparisons, none among the selected variants of interest (EGFR, KRAS, TP53, KEAP1, STK11, SMARCA4, ARID1A and targeted DDR genes), were significantly associated with the NLR. Conclusions: In this post-hoc analysis, a baseline low NLR identifies patients with NSCLC who derive a greater survival benefit from atezolizumab as compared to those identified in the docetaxel cohort, irrespective of genomic features. Patients with a low NLR and PD-L1 positive tumors derive the greatest benefit with immunotherapy and the NLR could complement PD-L1 expression in tailoring treatment in this setting. Clinical trial information: NCT02008227.