Clinical Study Of Magnetic Resonance Diffusion-Weighted Imaging Assessing And Predicting The Outcome Of Chemotherapy On Lung Cancer Patients

Objective:To explore the value of diffusion-weighted imaging (DWI) as an early response indicator and predictor in lung cancer patients treated with chemotherapy.Materials and methods:33patients pathologically diagnosed as lung cancer were included. MRI imaging was performed before,3to4weeks and3months after treatment in all patients. Conventional MRI sequence and DWI (b=0,800s/mm2) were performed at the first two times. At the last time, conventional MRI sequence only. The measurable lesions were classified as responding and non-responding according to RECIST1.1Version.26cases in responding group,9cases in non-responding group. The difference of mean ADC value and the long diameter were compared before and after chemotherapy with Paired-Sample T test; the correlations between the percentage change of mean ADCs and the percentage change of diameters were evaluated with Pearson correlation coefficient, as well as between pre-chemotherapy mean ADCs and the percentage change of diameters. The difference of the long diameter prior to chemotherapy, mean ADCs prior to chemotherapy, the percentage of the diameters change and the percentage of mean ADCs change at the time of the first two were compared between response and non-response group respectively with Independent Sample T test; and ROC curve was used to analyze the outcome of chemotherapy on lung cancer patients in differentiation response from non-response. Then select the optimal value and confirm the diagnostic threshold. We also classified most of the lesions as squamous carcinoma group and adenocarcinoma group according to pathological type. The differences of mean ADCs prior to chemotherapy and the percentage of the mean ADCs changes at the time of the first two were compared between squamous carcinoma group and adenocarcinoma group with Independent Sample T test respectly. Results:1. There totally were35lesions. After3to4weeks of treatment, early statistically significant increases of mean ADCs were observed. While the diameters of lesion were of no statistical changes (P>0.05).2. There was negative correlation between percentage change of ADCs and percentage change of diameter (either after3to4weeks or3months of treatment)[(r=-0.340, P=0.046),(r=-0.408, P=0.015)]. But there was no significant correlation between pretreatment ADCs and the changes of diameters percentage changes of diameter (neither after3to4weeks nor3months of treatment).3. There were26lesions in response group and9in non-response group. There was no statistically difference of the diameters prior to chemotherapy, ADCs prior to chemotherapy and the percentage of the diameters change of response lesions between response and non-response group. The percentage of the ADCs change of response lesions was significantly higher than that of non-response ones (P=0.021). The best increase percentage of ADCs at3to4weeks after therapy as to differentiate between response and non-response group was9.2%, ROC analysis shows the area under curve (AUC) were0.774, sensitivity and specificity were65.4%and88.9%respectively.4. There were15lesions in squamous carcinoma group and13in adenocarcinoma group. There was statistically difference of ADCs pior to chemotherapy between squamous carcinoma group and adenocarcinoma group. There was no statistically difference of the percentage of the ADCs change at the time of the first two were compared between the two group.Conclusion:The mean ADCs can be more sensitive to monitor the early changes of lesions, which were undergoing chemotherapy. Measurement of tumor ADCs can be used as indicators for assessing the efficacy of chemotherapy. The percentage of the ADCs changes of tumor had ability to predict response of chemotherapy. For squamous carcinomas and adenocarcinomas of different pathology type, the differences of pre-chemotherapy ADCs between them were obvious, but similar of response to chemotherapy.