To study the clinical features of patients with lung metastasis from colorectal cancer and investigate risk factors on prognosis.

The SEER*Stat software was used to identify 4 202 colorectal cancer patients with organs metastasis from 2010 to 2013. The differences in clinicopathological features between the lung metastasis group (894 cases) and the non-lung metastasis group (3 308 cases) were compared by chi-square test and binary logistic regression. The Kaplan-Meier method was used to univariate analysis for lung metastasis patients. The deference of accumulative survival rate was evaluated by the log-rank test. The significant factors in univariate analysis were included in the multivariate cox regression model to get the independent risk factors.

Following factors had statistically significant differences between the two groups in chi-square test: race, sex, the primary site, tumor stage and node stage. Binary logistic regression results showed that race (OR: 1.211, 95%CI: 1.040~1.411, P=0.014), sex (OR: 1.209, 95%CI: 1.042~1.403, P=0.013), the primary site (OR: 1.147, 95%CI: 1.067~1.234, P<0.001) and Node stage (OR: 0.907, 95%CI: 0.826~0.995, P=0.039) were independent impact factors affecting lung metastasis. The median survival time of 894 patients was 11 months. The overall 1-year survival was 65.2% and 3-year survival was 24.4%. The factors of years of diagnosis, the primary site, histological grade, T stages, N stages, radiotherapy, CEA, the number of positive lymph nodes and other organs metastasis significantly influenced the survival rate by univariate analysis. The factors of years of diagnosis (HR: 1.631, 95%CI: 1.344~1.980, P<0.001), histological grade (HR: 1.405, 95%CI: 1.194~1.652, P<0.001), Node stages (HR: 1.253, 95%CI: 1.111~1.414, P<0.001), CEA (HR: 1.744, 95%CI: 1.306~2.330, P<0.001), the number of positive lymph nodes (HR: 1.764, 95%CI: 1.408~2.208, P<0.001) and other organs metastasis (HR: 1.980, 95%CI: 1.578~2.483, P<0.001) were available independent prognostic factors through multivariate analysis.

The factors of race, sex, the primary site and Node stage are independent impact factors affecting lung metastasis. The factors of years of diagnosis, histological grade, Node stages, CEA, the number of positive lymph nodes and other organs metastasis can be used to predict the prognosis of patients with lung metastasis from colorectal cancer independently.