Objective:To explore the correlation between tumor markers and prognosis of patients with idiopathic inflammatory myopathy (IIM) associated interstitial lung disease (ILD).Methods:A total of 149 patients who were no less than 18 years old and diagnosed with IIM-ILD from July 2017 to September 2019 in the First Affiliated Hospital of Zhengzhou University were consecutively enrolled in the study. Ten patients were lost to follow-up. The remaining 139 cases were regarded as research objects. Patients were divided into survival group or death group according to their one-year survival status. Then their baseline characteristics were compared. Univariate Cox regression analyses of age, gender, cancer, inflammatory indexes, muscle zymogram, tumor markers, ferritin, melanoma differentiation-associated gene 5 (MDA5) antibody and treatment regimens were conducted to identify prognostic risk factors of one-year mortality. Corrected multivariable cox regression was applied to screen the independent risk factors associated with one-year mortality of IIM-ILD. According to the cut-off value of carcinoembryonic antigen (CEA) and neuron specific enolase (NSE) (6 μg/L and 28 μg/L, respectively), patients were divided into high-level groups and low-level groups. Kaplan Meier survival curve were generated to compare one-year survival rate of high-level groups and low-level groups. On the basis of qualitative results of MDA5 antibody, patients were split into two groups with positive MDA5 antibody or negative MDA5 antibody. The differences of CEA, NSE levels between the two groups and the correlation between CEA, NSE levels and ferritin were analyzed.Results:Age, lactate dehydrogenase (LDH), CEA, carbohydrate antigen (CA) 199, NSE and ferritin in the death group were higher than those in the survival group, while the rate of immunosuppressant administration was lower than that in survival group ( P<0.05). Univariate regression analyses showed that CEA, cytokeratin 19 fragment (CYFRA211) and NSE were risk factors for one-year mortality of IIM-ILD. Adjusted by age, treatment regimens and tumor, multivariate regression analysis showed that CEA [ HR=1.112, 95% CI (1.017-1.214), P=0.019] and NSE [ HR=1.033, 95% CI (1.002-1.064), P=0.034] were independent risk factors for one-year mortality. One-year survival rate of the group with CEA≥6 μg/L was lower than that in the group with CEA<6 μg/L (Logrank test, P<0.001). Similarly, one-year survival rate of the group with NSE≥28 μg/L was lower than that in the group with NSE<28 μg/L (Logrank test, P<0.001). In addition, the CEA level in patients with positive MDA5 antibody was higher than that in patients with negative MDA5 antibody ( P<0.001). However, there was no correlation between NSE and MDA5 antibody. Moreover, serum levels of CEA ( r=0.299, P=0.002) and NSE ( r=0.349, P<0.001) were positively correlated with ferritin. Conclusions:Tumor markers have predictive value for the prognosis of IIM-ILD. Higher CEA and NSE are independent risk factors for poor prognosis in patients with IIM-ILD.