Fig 2: Western blotting and IF staining confirmed the immunoreactivity of LUAD plasma to PRTN3. (A)Western blotting of anti-PRTN3 autoantibody in human plasma. The cropping strips of lanes 1-9 representing LUAD plasma showed strong reactivity with PRTN3 recombinant protein. Lanes 10-12 (BPN) and lanes 13-15 (NC), the cropping strips of 6 random human plasma with negative reactivity to PRTN3 recombinant protein. Lane16, anti-PRTN3 antibody used as positive control. (B)Immunofluorescence staining of PRTN3 in LUAD cells. Phosphate-buffered saline (PBS) was used as negative control; monoclonal anti-PRTN3 autoantibody was used as positive control; a representative anti-PRTN3 autoantibody positive LUAD plasma was used as LUAD; the same LUAD plasma used in LUAD pre-absorbed was pre-absorbed with recombinant PRTN3 protein; a representative anti-PRTN3 autoantibody negative NC plasma was used as NC; the same normal control plasma used in NC pre-absorbed was pre-absorbed with recombinant PRTN3 protein, and subsequently utilized for immunofluorescence staining assay. Scale bars, 10μm. Obtained at 40× by microscope. LUAD, lung adenocarcinoma; BPN, benign pulmonary nodule; NC, normal control; PC, positive control.

Fig 3: PRTN3 protein was highly expressed in LUAD tissues. (A) Representative IHC staining images of PRTN3 protein in NC tissue, para-cancerous tissue and LUAD tissue from tissue array (obtained at 20× by microscope). (B) Statistical analyses of the IHC scores of PRTN3 expression in NC tissues, para-cancerous tissues and LUAD tissues. (C) The expression profiles of PRTN3 in NC tissues and different pathological grades of LUAD tissues. NC, normal control; Para, para-carcinoma. ****P < 0.0001. Scale bars, 50μm. Lines represented quartile and median.

Fig 4: The scatter plots of the SBI and diagnostic performance of anti-PRTN3 autoantibodies in LUAD. (A–D) Distribution of anti-PRTN3 IgG and anti-PRTN3 IgM in the training set (A, B) and validation set (C–F) The ROC analysis of anti-PRTN3 IgG (E) and IgM (F) in differentiating LUAD from NC in the training set. (G, H) Diagnostic performances of anti-PRTN3 IgG (G) and IgM (H) in the validation set. (I, J) The positive frequency of anti-PRTN3 IgG (I) and IgM (J) in the training set. (K, L) The positive frequency of anti-PRTN3 IgG (K) and IgM (L) in the validation set. LUAD, lung adenocarcinoma; BPN, benign pulmonary nodule; NC, normal control; ****P < 0.0001; ***P < 0.001. Lines represented median.

Fig 5: The value of combined anti-PRTN3 autoantibodies and CEA in the diagnosis of LUAD and early LUAD. (A-D) AUC of anti-PRTN3 IgG autoantibody, anti-PRTN3 IgM autoantibody and combination of anti-PRTN3 IgG and IgM autoantibody for the diagnosis of NC and BPN from LUAD (A, B) and early LUAD (C, D) in the validation set. (E-H) Diagnostic performances of CEA alone, combination of anti-PRTN3 IgG and IgM autoantibody, and the combination of anti-PRTN3 autoantibodies and CEA for the diagnosis of NC and BPN from LUAD (E, F) and early LUAD (G, H) in the validation set. LUAD, lung adenocarcinoma; BPN, benign pulmonary nodule; NC, normal control.