Background: ALK is a prognostic and predictive tumor marker in non-small cell lung carcinoma (NSCLC), and is more often found in lung adenocarcinomas. Methods: The clinical and pathological data of 87 patients confirmed to have NSCLC by pathology or cytology were selected from April 2014 to January 2017 at the Tumor Hospital of Hebei Province. Results: Of the 87 ALK-positive-patients, 47 patients were treated with oral administration of crizotinib. The objective response rate (ORR) was 61.7%, the disease control rate (DCR) was 93.6%, and the mPFS was 19 months. In an analysis of the number of metastatic sites, the patients who had more than three metastatic sites, the ORR, DCR, and mPFS were 63.9%, 94.5%, and 19 months, compared with the 45.5%, 91%, and 11 months in the patients with less sites (P = 0.040). For patients of 60 years or older, ORR and DCR were 40% and 100%, the other group was 71.9% and 90.6%, respectively(P = 0.036). The timing of treatment was analyzed. At the first application of crizotinib, ORR and DCR were 78.2% and 100% corresponding 45.8% and 87.5% at the second and final application of crizotinib group (P = 0.022). Baseline brain metastases were present in 25.5% (12/47) of patients in this study. 9 of the patients who developed disease progression during crizotinib treatment had new brain metastases or increased preexisting cranial foci. Most of them took the treatment strategy of continuing crizotinib or replacing the second/third generation ALK-TKI treatment combined with local radiotherapy for brain metastases. Conclusions: The efficacy of crizotinib in patients with advanced NSCLC is related to the number of metastatic organs, age and timing of treatment. The use of crizotinib is prone to intracranial progression, and progression of simple brain metastases is not an indication that crizotinib is discontinued. Patients will continue to benefit from combination of local radiotherapy.