e19364 Background: Checkpoint Inhibitors (CPI) has revolutionized cancer treatment but financial toxicity can be incurred due to high cost. We performed a population-based analysis examining the cost of CPI vs. conventional chemotherapy (CT) in patients with NSCLC to capture the real-world value of CPI on an individual and community level. Methods: Population based claims database of privately insured individuals, MarketScan was interrogated. Pts in years 2015 and 2016 with ICD 9 or 10 diagnosis of lung cancer who were continuously enrolled in insurance plan for at least 12 months were included. CPT codes for CPIs and CT were used. Outcome Measure was Health Service Use ( surrogate for treatment related toxicity) such as Inpatient Admissions and Emergency Department Use. Cost Categories (US Dollars) noted were Total Costs,Total Out-of-Pocket Costs,Total Outpatient Infusion Costs, Outpatient Infusion Out-of-Pocket Costs, Inpatient Admissions Costs and Emergency Department Use Costs.Statistical tests “paired t tests” and “unpaired t tests” were used to compare the above variables using SAS software. Results: In 2015 of 22,193 with lung cancer 554 received CPI vs CT in 21,639. In 2016 of 22,810 pts 2478 received CPI vs CT in 20,332. Data for CPI vs. CT in years 2015 (a) and 2016 ( b) respectively. p value for each pair was <0.0001 unless as noted below 1. Mean Total Cost: a) 228504 vs. 140,970 b) 202202 vs. 147801 2. Mean Total Out of Pocket Cost: a) 3381 vs. 3308 p=0.4829 b) 3500 vs. 3343 p=0.0021 3. Mean Total Outpatient Infusion Cost: a) 51158 vs. 18634 b) 60087 vs. 18999 4. Mean Total Out of Pocket Outpatient Infusion Cost: a)159 vs. 282 b) 359 vs. 263 5. Inpatient Admission rate: a) 33.75% vs. 59.40% b) 43% vs. 58% 6. Mean Inpatient Admission Cost: a) 53045 vs 69810 p =0.0025 b) 54124 vs 73903 7. ED Visit rate: a) 38.2% vs. 61.4% b) 50% vs. 61% 8. Mean ED Cost: a) 4117 vs. 5265 p=0.0379 b) 4947 vs. 6025 p=0.0001. Conclusions: Despite the higher individual total cost in pts receiving CPI vs CT; the cost of CPI could potentially be offset at healthcare level by lesser expenditure through lower ED visits and inpatient hospitalization. Though the out of pocket costs/patient was statistically different, the mean numerical difference was small between CPI vs CT. Future prospective trials of CPI should also incorporate costs to individual and healthcare systems in addition to evaluating the impact of CPI on Quality of life vs CT to understand the real “value” of CPI. This novel study can serve as a benchmark for future population based studies.