Background: Non-small cell lung cancer (NSCLC) is a deadly human malignancy, and previous studies support the contribution of microRNAs (miRNAs) to cancer assessment. It has been reported that miR-1231 can be used as a biomarker to assess prognosis in different cancers. However, the prognostic value of miR-1231 in NSCLC patients with comorbid diabetes mellitus (DM) remains unclear. The present study evaluated the risk factors for NSCLC with DM and developed a predictive model for it.
Methods: A real-world study was conducted, including data from 108 patients with NSCLC combined with DM from April 1, 2010, to June 1, 2015. MiR-1231 was recorded during hospital admission. Cox-proportional hazards model was applied for survival analysis of risk factors for cancer-related mortality and to create nomograms for prediction. The accuracy of the model was evaluated by C-index and calibration curves.
Results: The mortality rate in the high miR-1231 level (≥ 1.775) group was 57.4%. On the basis of univariate analysis, we put factors (P < 0.05) into multivariate regression models, and high miR-1231 levels (P < 0.001, HR = 0.57), surgery (P < 0.001, HR = 0.37) and KPS score > 80 (P = 0.01, HR = 0.47) had a better prognosis and were considered as independent protective factors. These independently relevant factors were used to create nomograms to predict long-term patient survival. Nomogram showed good accuracy in risk estimation with a guide-corrected C-index of 0.691.
Conclusion: MiR-1231 reduced the risk of cancer-related death in patients with combined NSCLC and DM. Nomogram based on multivariate analysis showed good accuracy in estimating the overall risk of death.
Keywords: Diabetes mellitus; Non-small cell lung cancer; Overall survival; The prediction model; miR-1231.
© The Author(s) 2020.