Five tumor-associated autoantibodies expression levels in serum predict lung cancer and associate with poor outcome

Machine Learning for Early Discrimination Between Lung Cancer and Benign Nodules Using Routine Clinical and Laboratory Data

BACKGROUND

Lung cancer poses a global health threat necessitating early detection and precise staging for improved patient outcomes. This study focuses on developing and validating a machine learning-based risk model for early lung cancer screening and staging, using routine clinical data.

METHODS

Two medical center, observational, retrospective studies were conducted, involving 2312 lung cancer patients and 653 patients with benign nodules. Machine learning techniques, including differential analysis and feature selection, were employed to identify key factors for modeling. The study focused on variables such as nodule density, carcinoembryonic antigen (CEA), age, and lifestyle habits. The Logistic Regression model was utilized for early diagnoses, and the XGBoost model was utilized for staging based on selected features.

RESULTS

For early diagnoses, the Logistic Regression model achieved an area under the curve (AUC) of 0.716 (95% confidence interval [CI] 0.607-0.826), with 0.703 sensitivity and 0.654 specificity. The XGBoost model excelled in distinguishing late-stage from early-stage lung cancer, exhibiting an AUC of 0.913 (95% CI 0.862-0.963), with 0.909 sensitivity and 0.814 specificity. These findings highlight the model's potential for enhancing diagnostic accuracy and staging in lung cancer.

CONCLUSION

This study introduces a novel machine learning-based risk model for early lung cancer screening and staging, leveraging routine clinical information and laboratory data. The model shows promise in enhancing accuracy, mitigating overdiagnosis, and improving patient outcomes.

An analysis of the influencing factors of false negative autoantibodies in patients with non-small cell lung cancer

Objectives

To analyze the clinical significance of seven autoantibodies (P53, PGP9.5, SOX2, GAGE7, GBU4-5, MAGE, and CAGE) in patients with non-small cell lung cancer (NSCLC) and the factors that influence false-negative results.

Methods

Seven autoantibodies were measured in the serum of 502 patients with non-small cell lung cancer (NSCLC) using ELISA, and their correlations with age, sex, smoking history, pathological type, clinical stage, and PD-L1 gene expression were analyzed. The clinicopathological data of the false-negative and positive groups for the seven autoantibodies were compared to determine the influencing factors.

Results

P53 antibody expression level was correlated with lobulation sign, PGP9.5 antibody expression level with sex and vascular convergence; SOX2 antibody expression level with pathological type, clinical stage, and enlarged lymph nodes; and MAGE antibody expression level with the pathological type (P<0.05). False-negative autoantibodies are prone to occur in lung cancer patients with ground-glass nodules, no enlarged lymph nodes, no vascular convergence, and PD-L1 gene expression <1% (P <0.05).

Conclusion

Detection of seven autoantibodies was clinically significant in patients with NSCLC. However, poor sensitivity should be considered in clinical diagnoses to prevent missed diagnoses.

A protein microarray-based serum proteomic investigation reveals distinct autoantibody signature in colorectal cancer

PURPOSE

Colorectal cancer (CRC) has been reported as the second leading cause of cancer death worldwide. The 5-year annual survival is around 50%, mainly due to late diagnosis, striking necessity for early detection. This study aims to identify autoantibody in patients' sera for early screening of cancer.

EXPERIMENTAL DESIGN

The study used a high-density human proteome array with approximately 17,000 recombinant proteins. Screening of sera from healthy individuals, CRC from Indian origin, and CRC from middle-east Asia origin were performed. Bio-statistical analysis was performed to identify significant autoantibodies altered. Pathway analysis was performed to explore the underlying mechanism of the disease.

RESULTS

The comprehensive proteomic analysis revealed dysregulation of 15 panels of proteins including CORO7, KCNAB1, WRAP53, NDUFS6, KRT30, and COLGALT2. Further biological pathway analysis for the top dysregulated autoantigenic proteins revealed perturbation in important biological pathways such as ECM degradation and cytoskeletal remodeling etc. CONCLUSIONS AND CLINICAL RELEVANCE: The generation of an autoimmune response against cancer-linked pathways could be linked to the screening of the disease. The process of immune surveillance can be detected at an early stage of cancer. Moreover, AAbs can be easily extracted from blood serum through the least invasive test for disease screening.

Improvement in the performance of an autoantibody panel in combination with heat shock protein 90a for the detection of early‑stage lung cancer

The early diagnosis of lung cancer is closely associated with the decline of mortality. A panel consisting of seven lung cancer-related autoantibodies (7-AABs) has been shown to be a reliable and specific indicator for the early detection of lung cancer, with a specificity of ~90% and a positive predictive value of ~85%. However, its low sensitivity and negative predictive value limit its wide application. To improve its diagnostic value, the diagnostic efficiencies of 7-AABs in combination with non-specific tumor markers were retrospectively investigated for the detection of early-stage lung cancer. A total of 217 patients with small lung nodules who presented with ground-glass opacity or solid nodules as well as 30 healthy controls were studied. The concentrations of 7-AABs and heat shock protein 90a (HSP90a) were assessed using ELISA. Automated flow fluorescence immune analysis was used for the assessment of CEA, CYFRA21-1, CA199 and CA125 levels. The results showed that 7-AABs + HSP90a possessed a remarkably improved diagnostic efficiency for patients with small pulmonary nodules or for patients with lung nodules of different types, which suggested that 7-AABs in combination with HSP90a could have a high clinical value for the improvement of the diagnostic efficiency of early-stage lung cancer.

The combination of a seven-autoantibody panel with computed tomography scanning can enhance the diagnostic efficiency of non-small cell lung cancer

Background

Non-small cell lung cancer (NSCLC) is still of concern in differentiating it from benign disease. This study aims to validate the diagnostic efficacy of a novel seven-autoantibody (7-AAB) panel for the diagnosis of NSCLC.

Methods

We retrospectively enrolled 2650 patients who underwent both the 7-AAB panel test and CT scanning. We compared the sensitivity, specificity, and PPV of 7-AAB, CT, and PET-CT in the diagnosis of NSCLC in different subgroups. Then, we established a nomogram based on CT image features and the 7-AAB panel to further improve diagnostic efficiency. Moreover, we compared the pathological and molecular results of NSCLC patients in the 7-AABs positive group and the negative group to verify the prognostic value of the 7-AAB panel.

Results

The strategy of a “both-positive rule” combination of 7-AABs and CT had a specificity of 95.4% and a positive predictive value (PPV) of 95.8%, significantly higher than those of CT or PET-CT used alone (P&lt;0.05). The nomogram we established has passed the calibration test (P=0.987&gt;0.05) with an AUC of 0.791. Interestingly, it was found that the 7-AABs positive group was associated with higher proportion of EGFR mutations (P&lt;0.001), lower pathological differentiation degrees (P=0.018), more advanced pathological stages (P=0.040) and higher Ki-67 indexes (P=0.011) in patients with adenocarcinoma.

Conclusion

This study shows that combination of a 7-AAB panel with CT has can significantly enhance the diagnostic efficiency of lung cancer. Moreover, the 7-AAB panel also has potential prognostic value and has reference significance for the formulation of the treatment plan.

Autoantibodies - enemies, and/or potential allies?

Autoantibodies are well known as potentially highly harmful antibodies which attack the host via binding to self-antigens, thus causing severe associated diseases and symptoms (e.g. autoimmune diseases). However, detection of autoantibodies to a range of disease-associated antigens has enabled their successful usage as important tools in disease diagnosis, prognosis and treatment. There are several advantages of using such autoantibodies. These include the capacity to measure their presence very early in disease development, their stability, which is often much better than their related antigen, and the capacity to use an array of such autoantibodies for enhanced diagnostics and to better predict prognosis. They may also possess capacity for utilization in therapy, in vivo. In this review both the positive and negative aspects of autoantibodies are critically assessed, including their role in autoimmune diseases, cancers and the global pandemic caused by COVID-19. Important issues related to their detection are also highlighted.

The diagnostic value of a seven-autoantibody panel and a nomogram with a scoring table for predicting the risk of non-small-cell lung cancer

The early detection of non-small-cell lung cancer (NSCLC) remains a common concern. The aim of our study was to validate the diagnostic value of a seven-autoantibody (7-AAB) panel compared with radiological diagnosis for NSCLC. We constructed a nomogram and a scoring table based on the 7-AAB panel's result to predict the risk of NSCLC. We prospectively enrolled 268 patients who presented with radiological lesions and underwent both the 7-AAB panel test and pathological diagnosis by surgical resection. A comparison between the 7-AAB panel and radiological diagnosis was performed. A nomogram and a scoring table based on the 7-AAB panel's result to predict the risk of NSCLC were constructed and internally validated. The 7-AAB panel test had a specificity of 90.2% and a positive predictive value (PPV) of 92.7%, which were significantly higher than those of the radiological diagnosis. The 7-AAB panel also showed a preferable sensitivity in patients with early-stage disease. Seven factors, including the 7-AAB panel results, were integrated into the nomogram. For more convenient application, we formulated a scoring table based on the nomogram. The area under the receiver operating characteristic curve was 0.840 and 0.860 in the training group and validation group, respectively, which was higher than that using the 7-AAB panel or radiological diagnosis alone. This study reveals that our 7-AAB panel has clinical value in the diagnosis of NSCLC. The utility of our nomogram and the scoring table indicated that they have the potential to assist clinicians in avoiding unnecessary treatment or needless follow-up.