1
|
Liu M, Zhou Z, Liu F, Wang M, Wang Y, Gao M, Sun H, Zhang X, Yang T, Ji L, Li J, Si Q, Dai L, Ouyang S. CT and CEA-based machine learning model for predicting malignant pulmonary nodules. Cancer Sci 2022; 113:4363-4373. [PMID: 36056603 DOI: 10.1111/cas.15561] [Citation(s) in RCA: 6] [Impact Index Per Article: 3.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 03/04/2022] [Revised: 08/22/2022] [Accepted: 08/25/2022] [Indexed: 12/15/2022] Open
Abstract
Computed tomography (CT), an efficient radiological technology, is used to detect lung cancer in the clinic. Carcinoembryonic antigen (CEA), a common tumor biomarker, is applied in the detection of various tumors. To highlight the advantages of two-dimensional techniques and assist clinicians in optimizing lung cancer diagnostic schemes, we established a favorable model combining CT and CEA. In the study, univariate analysis was performed to screen independent predictors in a training cohort of 271 patients with malignant pulmonary nodules (MPNs) and 92 with benign pulmonary nodules (BPNs). Six machine learning-based models involving five CT predictors (mediastinal lymph node enlargement, lobulation, vascular notch sign, spiculation, and nodule number) and lnCEA were constructed and validated in an independent cohort of 129 participants (92 MPNs and 37 BPNs) by SPSS Modeler. A nomogram and the Delong test were generated by R software. Finally, the model established by logistic regression had highest diagnostic efficiency (area under the curve [AUC] = 0.912). Moreover, the diagnostic ability of the logistic model in the validation cohort (AUC = 0.882, 80.4% sensitivity, 75.7% specificity) was higher than that of the Peking University model (AUC = 0.712, 68.5% sensitivity, 70.3% specificity) and the Mayo model (AUC = 0.745, 62.0% sensitivity, 75.7% specificity). Interestingly, for the participants with intermediate (10-30 mm) and CEA-negative nodule, the model reached an AUC of 0.835 (72.3% sensitivity, 83.3% specificity). The AUC for the early lung cancer was as high as 0.822 with 67.3% sensitivity and 78.9% specificity. As a conclusion, this promising model presents a new diagnostic strategy for the clinic to distinguish MPNs from BPNs.
Collapse
Affiliation(s)
- Man Liu
- Department of Respiratory and Sleep Medicine, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China.,Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China
| | - Zhigang Zhou
- Department of Radiology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| | - Fenghui Liu
- Department of Respiratory and Sleep Medicine, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| | - Meng Wang
- Department of Radiology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| | - Yulin Wang
- Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China
| | - Mengyu Gao
- Department of Radiology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| | - Huifang Sun
- Department of Radiology, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| | - Xue Zhang
- Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China
| | - Ting Yang
- Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China.,BGI College, Zhengzhou University, Zhengzhou, China
| | - Longtao Ji
- Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China.,BGI College, Zhengzhou University, Zhengzhou, China
| | - Jiaqi Li
- Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China
| | - Qiufang Si
- Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China.,BGI College, Zhengzhou University, Zhengzhou, China
| | - Liping Dai
- Henan Institute of Medical and Pharmaceutical Sciences & Henan Key Medical Laboratory of Tumor Molecular Biomarkers, Zhengzhou University, Zhengzhou, China.,BGI College, Zhengzhou University, Zhengzhou, China
| | - Songyun Ouyang
- Department of Respiratory and Sleep Medicine, The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
| |
Collapse
|
2
|
Li L, Guo C, Wan JL, Fan QS, Xu XL, Fu YF. The use of carcinoembryonic antigen levels to predict lung nodule malignancy: a meta-analysis. Acta Clin Belg 2022; 77:227-232. [PMID: 32703103 DOI: 10.1080/17843286.2020.1797330] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 10/23/2022]
Abstract
OBJECTIVES To assess the diagnostic value of serum carcinoembryonic antigen (CEA) as a diagnostic biomarker that can be used to differentiate between benign and malignant lung nodules (LNs). METHODS PubMed, Cochrane Library, and Embase were reviewed from January 2000 to April 2020 for eligible studies. Stata v12.0 was used to conduct this meta-analysis. RESULTS Our initial literature search identified 511 potentially relevant studies, of which 11 were ultimately included in the present meta-analysis. Ten studies were retrospective and only 1 study was prospective. Overall these studies incorporated 2760 patients and 2760 total LNs (1733 malignant, 1027 benign). Pooled sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), and diagnostic odds ratio (DOR) values for these studies were 0.33 (95% CI: 0.20-0.49), 0.92 (95% CI: 0.85-0.96), 3.96 (95% CI: 2.84-5.54), 0.73 (95% CI: 0.62-0.87), and 5.42 (95% CI: 3.77-7.78), respectively. The area under curve (AUC) value was 0.77, consistent with moderate diagnostic accuracy. We detected significant heterogeneity when calculating pooled sensitivity (I2 = 95.9%, P = 0.00), specificity (I2 = 92.0%, P = 0.00), PLR (I2 = 61.7%, P = 0.00), NLR (I2 = 92.8%, P = 0.00), and DOR (I2 = 93.8%, P = 0.00). No significant evidence of publication bias was detected via Deeks' funnel plot asymmetry test (P = 0.371). Meta-regression analysis revealed different reference standards to be closely associated with both sensitivity and specificity. CONCLUSIONS Serum CEA can achieve moderate diagnostic performance as a means of differentiating between malignant and benign LNs.
Collapse
Affiliation(s)
- Lei Li
- Department of Oncology, Binzhou Medical University Hospital, Binzhou, China
| | - Chen Guo
- Department of Oncology, Binzhou Medical University Hospital, Binzhou, China
| | - Jin-Liang Wan
- Department of Oncology, Binzhou Medical University Hospital, Binzhou, China
| | - Qing-Shuai Fan
- Department of Oncology, Binzhou Medical University Hospital, Binzhou, China
| | - Xiao-Liang Xu
- Department of Pediatric Surgery, Binzhou Medical University Hospital, Binzhou, China
| | - Yu-Fei Fu
- Department of Radiology, Xuzhou Central Hospital, Xuzhou, China
| |
Collapse
|