6572 Background: Broad genomic testing is necessary for treatment of patients with stage IV non-small cell lung cancer (NSCLC). This quality improvement initiative aims to implement a precision medicine service for NSCLC patients at a hybrid academic-community practice oncology network. Methods: Following IRB approval, network tumor registries were queried for retrospective newly diagnosed stage IV NSCLC from 01/2016 through 12/2018. PREDICT was implemented in 08/2021. It consists of: 1) system-wide reflex testing of newly diagnosed stage IV NSCLC patients by an in-house solid tumor focused assay (hybrid DNA/RNA next-generation sequencing (NGS) panel previously reported) and PD-L1 testing, 2) PREDICT navigator, 3) molecular tumor board (MTB), 4) integrated information portal for real-time updates on samples processing, results and treatment recommendations by the MTB. We compared prospective data (08/2021 through 01/2022) after PREDICT with retrospective practice assessment. Comparisons between the groups were conducted using independent samples t-test / Wilcoxon rank sum test for continuous variables and Chi-square test/Fisher’s exact test for categorical variables. Results: Of 861 retrospective patients identified in the compiled registry of stage IV NSCLC, 626 were eligible. Since PREDICT launch in 08/2021, 103 eligible prospective patients have been identified. Prospective patients are slightly older (mean age 70.9 vs 68 years old; p = 0.013), with no other significant demographic or clinical differences identified. Rates of NGS testing obtained within 90 days of biopsy date (BxD) increased significantly (94.1% vs 60.8%; p < 0.0001) after PREDICT. Turnaround times (TAT) from BxD to test results were significantly shorter for both NGS (12 vs 18 days; p < 0.0001) and PD-L1 (7 vs 10 days; p = 0.007) after PREDICT. A trend towards higher rates of actionable alterations (EGFR, ALK, ROS1, RET, BRAF, MET14 skipping, NTRK1/2/3) was noted: 19.6% vs 13% (p = 0.071). Targeted therapy use increased from 6.8% to 15.6% (p = 0.002) in the overall cohort. No differences in time to treatment initiation (TTI) after PREDICT have been identified to this point, with a median of 34 and 35 days for the prospective and retrospective groups, respectively. Conclusions: Implementing a precision medicine service for thoracic oncology patients has led to significantly higher rates of NGS testing for patients with stage IV NSCLC in a large hybrid academic-community practice network. Launching of this initiative resulted in significantly shorter TAT for both NGS and PD-L1 test results. A trend towards higher rates of actionable alterations has been identified. Targeted therapy use has increased significantly overtime, potentially due to higher availability of precision medicine drugs in the current era.