血清肿瘤标志物及肺癌相关自身抗体联合检测在肺结节良恶性鉴别中的应用价值

The Value of Combined Serum Tumor Markers and Cancer-Associated Autoantibodies for Differentiating Benign and Malignant Pulmonary Nodules

摘要: 目的：探讨血清肿瘤标志物及肺癌相关自身抗体检测在肺结节良恶性鉴别中的临床价值。方法：回顾性纳入2023年2月至2025年10月因肺结节接受手术治疗并获得明确病理诊断的患者200例，其中肺癌组160例、肺部良性疾病组40例。比较两组9种血清肿瘤标志物及7种肺癌相关自身抗体检测结果，分析单项及联合检测的诊断效能，并探讨其与肺癌患者临床病理特征的关系。采用ROC曲线评价单项指标和联合模型，DeLong检验比较AUC，决策曲线分析(DCA)评价临床净获益。结果：肺癌组与肺部良性疾病组9种肿瘤标志物单项阳性率差异均无统计学意义(均P > 0.05)。肺癌自身抗体中仅GAGE7阳性率高于良性组(P = 0.016)。7种自身抗体联合检测的灵敏度、特异度和诊断符合率分别为32.50%、95.00%和45.00%；肿瘤标志物联合自身抗体检测的灵敏度和诊断符合率升至81.88%和71.00%，但特异度仅为27.50%。7种自身抗体联合检测阳性率与结节直径、淋巴结转移及TNM分期差异有统计学意义(均P < 0.05)。9种肿瘤标志物模型、7种自身抗体模型和16项指标联合模型的交叉验证AUC分别为0.497、0.657和0.660；后两种模型的AUC均高于9种肿瘤标志物模型(P = 0.044、0.002)，但两者间差异无统计学意义(P = 0.934)。DCA未显示各模型在较宽阈值范围内具有稳定的临床净获益。结论：7种肺癌相关自身抗体联合检测在肺结节良恶性鉴别中具有较高特异度，但灵敏度及总体鉴别效能有限，其在肺结节良恶性鉴别中的辅助价值仍需进一步验证。肿瘤标志物联合自身抗体检测可提高诊断灵敏度，但尚未显示稳定的临床净获益。

Abstract: Objective: To investigate the clinical value of combined detection of serum tumor markers and lung cancer-associated autoantibodies in differentiating benign and malignant pulmonary nodules. Methods: A total of 200 patients who underwent surgery for pulmonary nodules and received a definitive pathological diagnosis between February 2023 and October 2025 were retrospectively included, comprising 160 patients with lung cancer and 40 with benign pulmonary diseases. The results of nine serum tumor markers and seven lung cancer-associated autoantibodies were compared between the two groups. The diagnostic performance of individual indicators and combined models was evaluated, and their associations with clinicopathological characteristics in patients with lung cancer were explored. Receiver operating characteristic curves were used to assess individual indicators and combined models, DeLong’s test was applied to compare AUCs, and decision curve analysis was performed to evaluate clinical net benefit. Results: There were no significant differences in the individual positivity rates of the nine tumor markers between the lung cancer and benign pulmonary disease groups (all P > 0.05). Only the positivity rate of GAGE7 differed significantly between the two groups (P = 0.016). The seven-autoantibody panel showed a sensitivity of 32.50%, specificity of 95.00%, and diagnostic accuracy of 45.00%. Combining tumor markers with autoantibodies increased the sensitivity and diagnostic accuracy to 81.88% and 71.00%, respectively, but reduced the specificity to 27.50%. The positivity rate of the seven-autoantibody panel differed significantly according to tumor diameter, lymph node metastasis status, and TNM stage (all P < 0.05). The cross-validated AUCs of the nine-tumor-marker model, seven-autoantibody model, and 16-indicator model were 0.497, 0.657, and 0.660, respectively. The latter two models had higher AUCs than the nine-tumor-marker model (P = 0.044 and 0.002, respectively), whereas no significant difference was observed between them (P = 0.934). DCA did not demonstrate a stable clinical net benefit for any model across a broad range of threshold probabilities. Conclusion: Combined detection of seven lung cancer-related autoantibodies has high specificity for differentiating benign from malignant pulmonary nodules; however, its sensitivity and overall discriminatory performance are limited. Its auxiliary value in the differential diagnosis of pulmonary nodules requires further validation. Combining tumor markers with autoantibodies may improve diagnostic sensitivity, but has not yet demonstrated a stable clinical net benefit.