Abstract Background: Immune-checkpoint blockades provide durable responses and improved long-term survival in a subset of advanced non-small-cell lung cancer (NSCLC) patients. However, predictive markers of response to immune-checkpoint blockades are a significant unmet clinical need. The objective of this study was to identify clinical predictors of disease progression and treatment response among NSCLC patients treated with immune-checkpoint blockades. Methods: Pre-treatment/baseline predictors included patient demographics, clinical data, driver mutations, blood chemistry, and hematology data. Using stepwise backward elimination and Classification and Regression Tree (CART) analyses, parsimonious models were developed to identify the most predictor factors associated with progression-free survival (PFS) and hyperprogressive disease (HPD) defined as patients that exhibited a greater than two-fold increase in tumor growth rate in less than 2 months. Results: This analysis included 228 NSCLC patients treated with single agent or double agent immunotherapies. Univariable analyses identified 15 covariates significantly associated with PFS which was reduced to 3 covariates by backward elimination. Among these 3 covariates, CART analyses revealed three patient subgroups based on those who had prior systemic treatment (PST) and baseline absolute neutrophil count (ANC). Patients without PST and low ANC had significantly (log-rank p-value < 0.001) improved 36-month PFS (22.8%; Hazard Ratio [HR] = 1.0) compared to patients without PST and high ANC (13.6% 36-month PFS; HR = 2.37) or patients with PST irrespective of ANC (11.3% 36-month PFS; HR = 2.18). For the HPD analyses, HPD patients had significantly worse survival compared to patients who had PD without HPD (median survival = 3.2 months vs. 8.4 and 0% 36-month PFS vs. 25.3%, respectively; p < 0.001). In multivariable analysis with HPD versus non-HPD as the dependent variable, the full model with two clinical covariates yielded an Area under Receiver Operating Characteristic (AUROC) of 0.783 whereas the parsimonious model included only Royal Marsden Hospital (RMH) prognostic score with an AUROC of 0.712. Conclusion: Because of the complexity in objective immunotherapy response and acquired resistance, there is a pressing challenge to identify which patients are least likely to respond. The models identified in this study have potential important translational implications to identify highly vulnerable NSCLC patients that experience poor outcomes and hyperprogressive disease in the immunotherapy setting. Note: This abstract was not presented at the meeting. Citation Format: Matthew B. Schabath, Ilke Tunali, Jhanelle E. Gray, Robert J. Gillies. Predictors of disease progression and treatment response among lung cancer patients treated with immunotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 3301.