[Early diagnosis of subtype in early clinical stage lung adenocarcinoma by using an autoantibody panel and computed tomography].

To explore the value of CT and autoantibody panel to diagnosis the subtype in early clinical stage lung cancer,especially lepidic predominant adenocarcinoma (LPA),and to provide the correct information for the clinical and the prognosis evaluation. A retrospective study was conducted of 60 patients (total 63 subsolid nodules,which included 39 PSN and 24 pGGN) who underwent surgical resection or needle biopsy for stage Ⅰa or Ⅰb lung adenocarcinoma at Affiliated Tumor Hospital of Zhengzhou University between June 2017 and April 2018,age from 28 to 82 years old,and the average age was (52±7) years. All patients underwent a pretreatment chest LDCT and the peripheral blood of all patients were used to detect the tumor related antibody (including p53, GAGE7, PGP9.5, CAGE, MAGEA1, SOX2, GBU4-5) through enzyme linked immunosorbent assay. All the patients were divided into LPA group (43 subsolid nodules, which included AIS 10 subsolid nodules, MIA 11 subsolid nodules, LPA 22 subsolid nodules) and invasive predominant adenocarcinoma (IPA) group (20 subsolid nodules). The information of CT scanning was measured and assessed in lung and mediastinal windows with double blind method. The mean computed tomography (m-CT) value and the solid component/tumor ratio in three-dimensional (3D) and two-dimensional (2D) planes were measured and analyzed using computer-aided diagnosis (CAD) system. There were 20 partial solid nodules in IPA group, 19 partial solid nodules in LPA group and 24 ground-glass nodules (χ(2)=19.278,0.000). There were 4 circular nodules, 16 irregular nodules in the IPA group, 21 circular nodules, 5 oval nodules and 7 irregular nodules in the LPA group χ(2)=8.587,0.003). The incidence of burr, vascular aggregation and bronchial truncation in IPA group was higher than that in LPA group (40.0% vs 16.3%, 70.0% vs 18.6%, 30.0% vs 2.3%, χ(2)=4.234,15.860,10.580, 0.040,0.000, 0.001). The incidence of clear tumor lung interface in patients in LPA group was significantly higher than that in patients in IPA group (97.7% vs 65.0%, χ(2)=13.146,0.00). Of all the quantitative analysis of nodules,the m-CT value, the solid component/tumor ratios in three-dimensional (3D) and two-dimensional (2D) planes in IPA group were higher than those of LPA group ((-180±156) vs (-410±213) HU, 0.44±0.32 vs 0.14±0.26, 0.54±0.26 vs 0.18±0.26, 4.208, -3.951、-5.166, 0.000, 0.000, 0.000). Among the 60 patients with lung cancer, there were 33 cases with positive antibody in peripheral blood, with a positive rate of 55.0%. The positive rate of 7-AABs was 70.0% in IPA group and 44.2% in LPA group, which had no statistical difference (χ(2)=3.647, 0.056), the positive expression of tumor-associated antibodies was independent of the patient's age, CT value and nodular solid components and lung nodular volume ratio and area ratio, 0.05, only in correlation with pleural traction (χ(2)=3.866, 0.049). Compared with IPA, the imaging features of LDCT about the mGGN and PGGN appearance, clear tumor-lung interface, low m-CT and the solid component/tumor ratio in two-dimensional or three-dimensional (3D) planes are benefit for the diagnosis of the LPA; the expression of tumor-associated antibody group is independent of the age of the patient and the number of nodular solid components, and is only related to pleural depression, which is not conducive to the identification of LPA and IPA.