BACKGROUND/AIMS: Recently, many studies have demonstrated that various tumor-associated autoantibodies have been detected in early stages of lung cancer. Therefore, we conducted a meta-analysis to comprehensively evaluate available evidence on the diagnostic value of autoantibodies against tumor-associated antigens in lung cancer. METHODS: We systematically searched PubMed, Scopus, Web of Science and other databases through 23 March 2018. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2. We used the bivariate mixed-effect models to calculate pooled values of sensitivity, specificity, positive likelihood ratios, negative likelihood ratios, diagnostic odds ratios and associated 95% confidence intervals. Summary receiver operating characteristic (SROC) curves were used to summarize overall test performance. Deek's funnel plot was used to detect publication bias. RESULTS: Review of 468 candidate articles identified fifty-three articles with a total of 11,515 patients for qualitative review and meta-analysis. Pooled sensitivity, specificity and area under the SROC curve were as follows for tumor-associated autoantibodies against the following proteins: p53, 0.19, 0.98, 0.82; NY-ESO-1, 0.17, 0.98, 0.90; Survivin, 0.19, 0.99, 0.96; c-myc, 0.14, 0.98, 0.45; Cyclin B1, 0.18, 0.98, 0.91; GBU4-5, 0.07, 0.98, 0.91; CAGE, 0.14, 0.98, 0.90; p16, 0.08, 0.97, 0.91; SOX2, 0.14, 0.99, 0.93; and HuD, 0.17, 0.99, 0.82. CONCLUSION: Each tumor-associated autoantibody on its own showed excellent diagnostic specificity for lung cancer but inadequate sensitivity. Our results suggest that combinations or panels of tumor-associated autoantibodies may provide better sensitivity for diagnosing lung cancer, and the diagnostic accuracy of tumor-associated autoantibodies should be validated in more studies.
BACKGROUND/AIMS: Recently, many studies have demonstrated that various tumor-associated autoantibodies have been detected in early stages of lung cancer. Therefore, we conducted a meta-analysis to comprehensively evaluate available evidence on the diagnostic value of autoantibodies against tumor-associated antigens in lung cancer. METHODS: We systematically searched PubMed, Scopus, Web of Science and other databases through 23 March 2018. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies-2. We used the bivariate mixed-effect models to calculate pooled values of sensitivity, specificity, positive likelihood ratios, negative likelihood ratios, diagnostic odds ratios and associated 95% confidence intervals. Summary receiver operating characteristic (SROC) curves were used to summarize overall test performance. Deek's funnel plot was used to detect publication bias. RESULTS: Review of 468 candidate articles identified fifty-three articles with a total of 11,515 patients for qualitative review and meta-analysis. Pooled sensitivity, specificity and area under the SROC curve were as follows for tumor-associated autoantibodies against the following proteins: p53, 0.19, 0.98, 0.82; NY-ESO-1, 0.17, 0.98, 0.90; Survivin, 0.19, 0.99, 0.96; c-myc, 0.14, 0.98, 0.45; Cyclin B1, 0.18, 0.98, 0.91; GBU4-5, 0.07, 0.98, 0.91; CAGE, 0.14, 0.98, 0.90; p16, 0.08, 0.97, 0.91; SOX2, 0.14, 0.99, 0.93; and HuD, 0.17, 0.99, 0.82. CONCLUSION: Each tumor-associated autoantibody on its own showed excellent diagnostic specificity for lung cancer but inadequate sensitivity. Our results suggest that combinations or panels of tumor-associated autoantibodies may provide better sensitivity for diagnosing lung cancer, and the diagnostic accuracy of tumor-associated autoantibodies should be validated in more studies.