The Value Of CT Image Texture Analysis In Early Assessment Of Chemotherapeutic Effect In Patients With Advanced Lung Adenocarcinoma:A Preliminary Study

Objective: To analyse the change trend of texture features and the possible causes by observing the gray level co-occurrence matrix(GLCM)parameters of computed tomography-based texture analysis(CTTA)before and after 2 cycles of chemotherapy in patients with advanced lung adenocarcinoma.To investigate the value of GLCM parameters of CTTA to predict and evaluate the early efficacy of chemotherapy in patients with advanced lung adenocarcinoma.Materials and methods: Patients with advanced lung adenocarcinoma were collected retrospectively from January 2013 to October 2018.All patients underwent chest CT examination before and after 2 cycles of chemotherapy.A total of 6 GLCM parameters(joint average,joint entropy,joint energy,contrast,inverse differences moment(IDM)and correlation)were extracted.Patients were classified into response group(PR and CR)and non-response group(SD and PD)according to RECIST 1.1 standard.The paired t-test or Wilcoxon rank sum test was used to compare the texture parameters between pre-chemotherapy and post-chemotherapy.The two independent samples t-test or Mann-Whitney U test was used to compare the texture parameters,?parameters and %parameters between the response group and the non-response group.The diagnostic efficacy was evaluated by using the receiver operating characteristic curve.The two-side significant test level was set up as ?=0.05.Results: A total of 59 patients were enrolled,including 20 patients in response group and 39 patients in non-response group.There were 4 parameters(joint average,contrast,IDM and correlation)were significantly different between pre-chemotherapy and post-chemotherapy in the response group(all P<0.05).There were 4 parameters(joint entropy,joint energy,contrast and IDM)were significantly different between pre-chemotherapy and post-chemotherapy in the non-response group(all P<0.05).There was no significant difference in the parameter between the response group and the non-response group before chemotherapy(all P>0.05).The joint average of non-response group was higher than that of response group after 2 cycles of chemotherapy,and the difference was statistically significant(P=0.012).AUC=0.690 for joint average after 2 cycles of chemotherapy,with an optimal sensitivity and specificity of 95.0% and 41.0%,respectively,for a cut-off value of 17.647.?contrast,?IDM,%contrast and %IDM were significantly different between the response group and the non-response group(all P<0.05).AUC=0.668 for ?contrast,with an optimal sensitivity and specificity of 70.0% and 69.2%,respectively,for a cut-off value of ?0.759.AUC=0.701 for ?IDM,with an optimal sensitivity and specificity of 60.0% and 82.1%,respectively,for a cut-off value of 0.063.AUC=0.660 for %contrast,with an optimal sensitivity and specificity of 70.0% and 64.1%,respectively,for a cut-off value of 0.322.AUC=0.703 for %IDM,with an optimal sensitivity and specificity of 65.0% and 74.4%,respectively,for a cut-off value of 0.088.Conclusion: The GLCM parameters of lung adenocarcinoma were significantly changed before and after chemotherapy in both response and non-response groups.The joint average of non-response group was higher than that of response group after 2 cycles of chemotherapy.And ?IDM and %IDM revealed high diagnostic efficacy.Our results suggest that parameters of GLCM,?parameters and %parameters after 2 cycles of chemotherapy could help to evaluate the early efficacy of chemotherapy of lung adenocarcinoma.As a non-invasive quantification technique,CTTA is expected to provide more valuable information for evaluating the chemotherapy efficacy of lung adenocarcinoma,and to help adjust the treatment plan in time and improve the prognosis.