OBJECTIVES: To speculate the cumulated ICER of EGFR-TKIs treatments in ten years for advanced NSCLC patients with different EGFR gene mutation status and assist the government to make decision on drug reimbursement in order to achieve the optimal value of the limited healthcare resource. METHODS: TreeAge Pro 2011 was used and one month was set as a stage. A total of 120 stages were circulated to evaluate the 10-year cost and effectiveness. RESULTS: According to roll back analysis, the total cost for chemotherapy was $28273 with the prolongation of 15.814 month in cumulated 10 years, and $37405 with the prolongation of 17.102 months for EGFR-TKIs. The ICER was $1304.72/LGM (life-gained month) through Markov cohort analysis and $1297.38/LGM through Monte Carlo analysis. Neither of the ICER values exceeded the predefined WTP (three times of GDP per capita per month). According to the probability sensitivity analysis, there was a probability of 87.1% for EGFR-TKIs being more cost-effective than chemotherapy. For cities or provinces with a GDP per capita per month more than $378.75, there was a probability of more than 50% for EFR-TKIs being more cost-effective. Chengdu, the city from which our clinical data came, gained GDP per capita per month in 2014 of $942.83, Therefore, its WTP would raise to $2862.5 (three times of $942.83) which largely exceeds the ICER either speculated by Markov cohort ($1304.72/LGM) or by Monte Carlo analysis ($1297.38/LGM). CONCLUSIONS: As WTP was predefined as three times of GDP per capita per month,EGFR-TKIs showed a probability of 87.1% being more cost-effective than chemotherapy for advanced NSCLC patients with mutated EGFR gene under the continuing fund support from China Charity Federation. Taking the example of Chengdu, EGFR-TKIs was 99% probably superior to chemotherapy in cost-effectiveness for NSCLC patients with mutated EGFR gene.