INTRODUCTION: To assess the cost-effectiveness of alectinib versus crizotinib as first-line treatments for advanced anaplastic lymphoma kinase (ALK)-positive non-small cell lung cancer (NSCLC) patients from the perspective of China's healthcare system. METHODS: A Markov model was developed to assess the clinical outcomes and costs of alectinib and crizotinib, which included five health states: progression-free (PF) without central nervous system (CNS) progression, PF with CNS progression, post-progression (PP) without CNS progression, PP with CNS progression, and death. Clinical data for transition probabilities were obtained from the ALEX trial at the updated data cutoff. Healthcare resource utilization and costs were derived from clinical expert opinions and published literature. One-way sensitivity analysis and probabilistic sensitivity analysis were conducted to assess the uncertainty of the results. Scenario analyses were conducted including using clinical data from the ALESIA trial in Asian patients, using utilities from the ALEX trial, and choosing different parametric survival models. RESULTS: In base case analysis, alectinib yielded an additional 1.04 quality-adjusted life years (QALYs) with incremental costs of $54,827, resulting in an incremental cost-effectiveness ratio (ICER) of $52,869/QALY. In scenario analysis, the ICER was $56,787/QALY using clinical data from the ALESIA trial. In probabilistic sensitivity analysis, the probabilities of alectinib being cost-effective were 0.4% and 43.7% when the willingness-to-pay (WTP) thresholds were $28,109/QALY and $50,000/QALY, respectively. CONCLUSION: Alectinib could prolong the mean time of PF and delay the time to CNS progression. However, because of its high drug cost, alectinib was unlikely to be cost-effective for untreated ALK-positive NSCLC patients in China.