OBJECTIVES: Targeted therapy with ALK inhibitor crizotinib offers significant improvement in clinical outcome for treatment of EML4–ALK fusion positive non-small cell lung cancer (NSCLC) patients. We estimated the cost-effectiveness of companion EML4-ALK genetic testing in combination with crizotinib treatment in the second-line setting for advanced NSCLC in Ontario. METHODS: We performed a cost-effectiveness analysis using a Markov model from a Ministry of Health perspective and a lifetime horizon. Transition probabilities and mortality rates were calculated based on the data of a recent second-line randomized trial of crizotinib versus chemotherapy (Shaw et al. New Engl J Med 2013). Costs were obtained from OCCI database, public labs and Princess Margaret Hospital. All parameters were varied separately in one-way and selected two-way sensitivity analyses. Various scenarios to assess the impact of model assumptions about testing and treatment were conducted. RESULTS: The use of pemetrexed and docetaxel in ALK-rearranged NSCLC, based on our preliminary model, could yield as much as 0.539 QALY and 0.429 QALY respectively, assuming no crossover from chemotherapy to crizotinib. Average costs per patient based on the preliminary model are estimated at CAD $19,388 for pemetrexed and $$33,226for docetaxel, with incremental cost-effectiveness ratios of $333,595/QALY and $125,812/QALY gained respectively. The results of the one-way sensitivity analysis indicated that the primary drivers of the ICER were the utilities and cost of crizotinib treatment. The model was least sensitive to IHC and FISH genetic test costs, re-biopsy cost, probability of progression while on pemetrexed treatment and probability of re-biopsy. CONCLUSIONS: EML4–ALK genetic testing in combination with crizotinib treatment for all NSCLC patients eligible for chemotherapy is not economically attractive in the current setting. Lower drug costs would be required to make this strategy economically feasible.