Abstract 2224: A panel of autoantibodies against multiple tumor-associated antigens in the early immunodiagnosis of lung cancer

Abstract

Abstract Lung cancer is the most common cancers worldwide and with very low 5-year survival rate due to lack of effective early diagnostic methods. In our previous study, nine autoantibodies against tumor-associated antigens (TAAs) were identified to be potential biomarkers in detection of lung cancer and this study aimed to find the optimal panel of TAAs to detect lung cancer in the early stage. Methods A total of 1,130 serum samples were included in this study, including patients with lung cancer, normal control and benign lung diseases patients. Autoantibodies against PSIP1,TOP2A,ACTR3,RPS6KA5,HMGB3,MMP12,GREM1,ZWINT and NUSAP1 were detected by enzyme-linked immunosorbent assay. Logistic regression models were generated from the training set and further validated in another independent set. We also evaluated the ability of the model to detecting benign lung diseases and early-stage lung cancer. Results Two greater models were selected. One model including 5 autoantibodies panel (GREM1,HMGB3,ZWINT,TOP2A and PSIP1) and another model including 3 autoantibodies panel (GREM1,HMGB3 and PSIP1). Finally, the panel of 3 autoantibodies showed high diagnostic accuracy with areas under the curve (AUC) of 0.711(95 % CI 0.674-0.746) in training cohort and 0.891(95% CI 0.845-0.927) in validation cohort, respectively. The AUC of the model of 3 autoantibodies panel was 0.948(95% CI 0.912-0.972) in discriminate lung cancer and benign pulmonary disease. This model could discriminate early-stage lung cancer patients from normal controls, with AUC of 0.686(95 % CI 0.633-0.735) in training cohort and AUC of 0.911(95 % CI 0.849-0.954) in validation cohort, and the overall AUC for early-stage lung cancer was 0.779(95 % CI 0.738-0.816) when the two cohorts were combined. Conclusions In summary, it would be more cost-effective and more efficient to use the model with 3 autoantibodies panel which has a high diagnostic performance for lung cancer detection, especially for early-stage lung cancer. And the model may apply a new help for the discrimination for benign pulmonary disease and lung cancer. Note: This abstract was not presented at the meeting. Citation Format: Di Jiang, Tingting Wang, Lu Pei, Peng Wang, Hua Ye, Chunhua Song, Kaijuan Wang, Jianying Zhang, Liping Dai. A panel of autoantibodies against multiple tumor-associated antigens in the early immunodiagnosis of lung cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 2224.