Diagnostic value of dual-layer spectral detector CT in differentiating lung adenocarcinoma from squamous cell carcinoma

Ronghua Mu; Zhuoni Meng; Zixuan Guo; Xiaoyan Qin; Guangyi Huang; Xuri Yang; Hui Jin; Peng Yang; Meimei Deng; Xiaodi Zhang; Xiqi Zhu

doi:10.3389/fonc.2022.868216

Diagnostic value of dual-layer spectral detector CT in differentiating lung adenocarcinoma from squamous cell carcinoma

Front Oncol. 2022 Dec 2:12:868216. doi: 10.3389/fonc.2022.868216. eCollection 2022.

Authors

Ronghua Mu¹, Zhuoni Meng¹, Zixuan Guo¹, Xiaoyan Qin², Guangyi Huang², Xuri Yang², Hui Jin², Peng Yang², Meimei Deng², Xiaodi Zhang³, Xiqi Zhu²

Affiliations

¹ Department of Radiology, Graduate School of Guilin Medical University, Guilin, China.
² Department of Radiology, Nanxishan Hospital of Guangxi Zhuang Autonomous Region, Guilin, China.
³ Philips (China) Investment Co., Ltd., Chengdu Branch, Chengdu, China.

Abstract

Background and objective: The pathological type of non-small cell lung cancer is considered to be an important factor affecting the treatment and prognosis. The purpose of this study was to investigate the diagnostic value of spectral parameters of dual-layer spectral detector computed tomography (DLCT) in determining efficacy to distinguish adenocarcinoma (AC) and squamous cell carcinoma (SC), and their combined diagnostic efficacy was also analyzed.

Methods: This is a single-center prospective study, and we collected 70 patients with lung SC and 127 patients with lung AC confirmed by histopathological examination. Morphological parameters, plain scan CT value, biphasic enhanced CT value, and spectral parameters were calculated. The diagnostic efficiency of morphological parameters, spectral parameters, and spectral parameters combined with morphological parameters was obtained by statistical analysis.

Results: In univariate analysis, seven morphological CT features differed significantly between SC and AC: tumor location (distribution), lobulation, spicule, air bronchogram, vacuole sign, lung atelectasis and/or obstructive pneumonia, and vascular involvement (all p < 0.05). In the arterial phase and the venous phase, the spectral parameters of AC were higher than those of SC (AP-Zeff: 8.07 ± 0.23 vs. 7.85 ± 0.16; AP-ID: 1.41 ± 0.47 vs. 0.94 ± 0.28; AP-NID: 0.13 ± 0.04 vs. 0.09 ± 0.03; AP-λ: 3.42 ± 1.10 vs. 2.33 ± 0.96; VP-Zeff: 8.26 ± 0.23 vs. 7.96 ± 0.16; VP-ID: 1.18 ± 0.51 vs. 1.16 ± 0.30; VP-NID: 0.39 ± 0.13 vs. 0.29 ± 0.08; VP-λ: 4.42 ± 1.28 vs. 2.85 ± 0.72; p < 0.001). When conducting multivariate analysis combining CT features and DLCT parameters with the best diagnostic efficacy, the independent predictors of AC were distribution on peripheral (OR, 4.370; 95% CI, 1.485-12.859; p = 0.007), presence of air bronchogram (OR, 5.339; 95% CI, 1.729-16.484; p = 0.004), and presence of vacuole sign ( OR, 7.330; 95% CI, 1.030-52.184; p = 0.047). Receiver operating characteristic curves of the SC and AC showed that VP-λ had the best diagnostic performance, with an area under the curve (AUC) of 0.864 and sensitivity and specificity rates of 85.8% and 74.3%, respectively; the AUC was increased to 0.946 when morphological parameters were combined, and sensitivity and specificity rates were 89.8% and 87.1%, respectively.

Conclusion: The quantitative parameters of the DLCT spectrum are of great value in the diagnosis of SC and AC, and the combination of morphological parameters and spectral parameters is helpful to distinguish SC from AC.

Keywords: X-ray computed tomography; dual layer detector; energy spectrum; lung cancer; pathological classification.