Early cancer detection aims to find tumors before they progress to an incurable stage. Currently, circulating tumor DNA (ctDNA) is a biomarker used to gather information about previously diagnosed tumors to monitor treatment efficacy and recurrence. To determine the potential health benefits of ctDNA detection for lung cancer screening in an asymptomatic population, we developed a new natural history model of lung cancer evolution and ctDNA shedding that accounts for the heterogeneity in growth, shedding, and mutation rates. We performed in-silico clinical trials for annual and semi‑annual ctDNA-based screening protocols and compared the results in terms of diagnosis size, detection stage and five years survival. Our analytic framework allows us to estimate how ctDNA-based screening programs could improve lung cancer‑specific survival, and can be extended to other cancers.