Clinical application of high-resolution CT combined with Ki-67 expression in predicting poorly differentiated invasive non-mucinous lung adenocarcinoma
ZHOU Yi1, ZHANG Bei2, HUANG KeFeng1, YANG Bo3
1. Department of Radiology, the 991th Hospital of PLA Joint Logistics Support Force, Xiangyang 441001, China; 2. Department of Radiology, Shaanxi Provincial Tumor Hospital, Xi’an 710061, China; 3. Department of Radiology, General Hospital of PLA Central Theater Command, Wuhan 430012, China
Abstract：Objective To explore the predictive value of high-resolution CT (HRCT) image features combined with Ki-67 expression level in poorly differentiated invasive non-mucinous adenocarcinoma. Methods A total of 197 patients with invasive non-mucinous lung adenocarcinoma were selected from the General Hospital of PLA Central Theater Command and the 991th Hospital of PLA Joint Logistics Support Force from July 2015 to June 2023. According to the pathological results, they were divided into two groups, with 150 cases in highly and moderately differentiated group and 47 cases in poorly differentiated group. Ki-67 expression level and HRCT characteristic indicators of cancerous nodule lesions were analyzed and compared between the two groups, such as nodule type, size, location, burr, lobed, margin, cavity, relationship with blood vessels, air bronchial sign, false cavity sign. Results There were significant differences in gender, nodule type, nodule size, pleural indentation sign positive, pseudo cavity sign, bronchial inflation sign and Ki-67 expression level between the two groups (P<0.05).Ki-67 expression level (OR=6.057, P＜0.05), nodule nature (OR=6.523, P＜0.05) and nodule size (OR=3.119, P＜0.05)were screened out as independent risk factors for poorly differentiated invasive non-mucinous lung adenocarcinoma(P<0.05). The best cut-off value of Ki-67 expression level and nodule size were 14.5% and 21.3 mm, respectively. The diagnostic efficacy, sensitivity and specificity of Ki-67 combined with Ki-67 expression level and nodule size were 0.888, 76.6% and 89.3%, which were superior to Ki-67 expression level or nodule size alone. Conclusions Ki-67 expression level, nodule nature, nodule size are independent risk factors in predicting poorly differentiated invasive non-mucinous lung adenocarcinoma, and nodule size combined with the expression level of Ki-67 can improve the diagnostic accuracy of poorly differentiated cancer cells.
Hansell D M,Bankier A A,MacMahon H, et al. Fleischner society: glossary of terms for thoracic imaging[J]. Radiology,2008,246(3): 697-722.
[5]
Weisbrod G L, Chamberlain D, Herman S J. Cystic change (pseudocavitation) associated with bronchioloalveolar carcinoma: a report of four patients[J].J Thorac Imag,1995, 10(2): 106-111.
Snoeckx A, Reyntiens P, Desbuquoit D, et al. Evaluation of the solitary pulmonary nodule: size matters, but donot ignore the power of morphology[J]. Insights Imaging, 2018, 9(1): 73-86.
Wu Y J,Liu Y C,Liao C Y,et al.A comparative study to evaluate CT-based semantic and radiomic features in preoperative diagnosis of in-vasive pulmonary adenocarcinomas manifesting as subsolid nodules[J].Sci Rep,2021,11: 66.
[12]
Aherne E A,Plodkowski A J,Montecalvo J,et al.What CT character-istics of lepidic predominant pattern lung adenocarcinomas correlate with invasiveness on pathology?[J].Lung Cancer,2018,118: 83- 89.
[13]
Li M, Wang Y, Chen Y, et al. Identification of preoperative prediction factors of tumor subtypes for patients with solitary ground-glass opacity pulmonary nodules[J]. J Cardiothorac Surg, 2018, 13(1):1-6.
Niazi M K, Tavolara T E, Arole V, et al. Identifying tumor in pancreatic neuroendocrine neoplasms from Ki67 images using transfer learning[J]. PlOS one,2018,13(4):e0195621.
2024, Vol. 35 
