Value of CT quantitative parameters in preoperative malignancy assessment and prognostic prediction for lung cancer patients
CHEN Wenbao
DING Xin
XIA Banghong
HU Shaohui
Abstract：Objective To evaluate the clinical value of preoperative CT imaging-derived quantitative parameters in assessing pathological malignancy and predicting postoperative survival outcomes in lung cancer patients.Methods A total of 215 patients who underwent surgical treatment for lung cancer at our hospital between April 2020 and April 2024 were enrolled.All cases were confirmed as lung cancer via postoperative pathology.Preoperative high-resolution CT scans were performed,and tumor-related quantitative parameters were analyzed,including sphericity,eccentricity,surface-to-volume ratio,solid component ratio,mean CT value,standard deviation(SD),skewness,kurtosis,energy,and entropy.Patients were divided into low-grade and high-grade malignancy groups based on pathological results.Differences in CT parameters between groups were compared.Multivariate Logistic regression identified independent risk factors for high-grade malignancy,while receiver operating characteristic(ROC)curves evaluated diagnostic efficacy for preoperative malignancy assessment.Follow up until May 2025,based on the follow-up results,patients were categorized into survival group of 146 cases and mortality group of 63 cases.Clinical data differences were analyzed,and multivariate Logistic regression identified independent risk factors for poor postoperative outcomes.ROC curves further assessed prognostic prediction accuracy.Results Based on pathological findings,87 patients were classified as low-grade malignancy and 128 as high-grade malignancy.Significant differences were observed in sphericity,eccentricity,surface-to-volume ratio,solid component ratio,mean CT value,SD,energy,and entropy(all P<0.05).Multivariate analysis identified surface-to-volume ratio,solid component ratio,and entropy as independent risk factors for high-grade malignancy,while sphericity and energy acted as protective factors(all P<0.001).ROC analysis demonstrated areas under the curve(AUCs)of 0.747(entropy),0.788(SD),0.794(solid component ratio),0.768(kurtosis),and 0.774(surface-to-volume ratio).The combined model achieved an AUC of 0.891(95%CI:0.850-0.932),significantly outperforming individual metrics(DeLong test:Z=2.162,P=0.008).During follow-up,6 patients were lost to follow-up.The mortality group showed higher proportions of poorly/undifferentiated tumors and significant differences in surface-to-volume ratio,solid component ratio,SD,energy,and entropy(all P<0.05).Multivariate analysis confirmed solid component ratio,and entropy as independent predictors of poor prognosis(all P<0.001).ROC analysis yielded AUCs of 0.824(solid component ratio),0.794(SD),and 0.852(entropy).The combined model achieved an AUC of 0.896(95%CI:0.850-0.942),demonstrating superior performance over single indicators(DeLong test:Z=3.525,P=0.001).Conclusions CT quantitative parameters exhibit robust efficacy in preoperative malignancy assessment and prognostic prediction for lung cancer patients.Surface-to-volume ratio,solid component ratio,and entropy are independent risk factors for high-grade malignancy,while sphericity and energy serve as protective factors.The combined diagnostic performance outperforms individual parameters.Solid component ratio,SD,and entropy also independently predict poor postoperative outcomes,with combined analysis offering enhanced predictive accuracy.
Keywords：CT quantitative parametersLung cancerPreoperative assessmentPrognosis prediction
Publication Date：2025-08-20
Online Publishing Date：2025-09-15(First online date of this platform, not the publication date of the document)
Pages：6( 390-395 )