CT spectral parameters and serum tumour markers to differentiate histological types of cancer histology.

AIM: To evaluate lung cancer histology type using computed tomography (CT) spectral quantitative parameters combining with serum tumour markers.

MATERIALS AND METHODS: Patients with suspicious lung cancer underwent CT spectral imaging and serum tumour markers. CT spectral quantitative parameters including attenuation value, the slope of spectral curve (λ), iodine concentration, water concentration, and effective atomic number (Zeff) were acquired. Serum levels of tumour markers including carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), squamous cell carcinoma antigen (SCC-Ag), and cytokeratin fragment CYFRA21-1 were also obtained. All the values were compared among different histological types of lung cancer. The diagnostic efficiencies of serum tumour markers, CT spectral parameters, and a combination of them were computed by statistical analysis.

RESULTS: CEA and NSE levels were higher in adenocarcinoma and neuroendocrine tumour, respectively, while SCC-Ag and CYFRA21-1 levels were higher in squamous cell cancer. There was no significant difference in attenuation among the groups (p>0.05), whereas λ in the arterial phase, and Zeff and IC in both the arterial and venous phases were significantly different among groups (p<0.05). According to the area under the receiver operating characteristic (ROC) curve (AUC) and Youden's index, the diagnostic efficiency of serum tumour markers were higher than that of CT spectral parameters. Moreover, AUCs of combined serum and CT indicators were larger than that of combined serum markers and combined CT spectral parameters between squamous cell cancer and adenocarcinoma as well as between squamous cell cancer and neuroendocrine tumour.

CONCLUSION: CT spectral quantitative parameters and serum tumour markers are valuable in evaluating histological types of lung cancer. In combination they can significantly improve diagnostic efficiency.