Earlytreatment Evaluation Of Lymphoma Response To Chemotherapy By Perfusion Computed Tomography

Objective:The aim of this study was to search for chemotherapy induced perfusion changes of lymphoma at earlytreatment ersus baseline perfusion computed tomography(PCT). Material and Methods:Collected from April 2015 to February 2016,a total of 50 consecutive patients who received PCT examinations before treatment(baseline) in Shanxi Tumor Hospital, using the Pefution GSI software(GE Healthcare),41 patients also received PCT examinations during treatment(earlytreatment: after 2 cycles of chemotherapy) and enhancement CT during treatment(midtreatment: after 4 cycles of chemotherapy). Blood flow(BF), Blood Volume(BV), Mean Transit Time(MTT)and Permeability Surface(PS) at baseline and earlytreatment were determined respectively. Treatment response was categorized according to the size of the lesion baseline and midtreatment into Complete Response( CR), Partial Response(PR) was defined as Responders(26);Stable Disease(SD)and Progressive Disease(PD) was defined as Nonresponders(14); Death as Nonresponders(1) reference the standard of RECIST. Statistical analysis(software: SPSS17.0) was performed, normality testing at first, followed by using the paired t test, P<0.05 is believed as having statistically significant. Receiver operating characteristic(ROC) curve analyses were performed to determine the cut-off values,sensitivity and specificity. Results:Baseline values of BF, BV, and PS in lesions as compared with the muscle were statistically higher. SIZE,BF,BV and PS in baseline and earlytreatment were statistically significant, the latter is significantly reduced than the former.The paired t test showed that SIZE, BF,BV and PS significantly decreased in earlytreatment when compared with the baseline about the responders(26/41), Whether it is HL or NHL. In nonresponders(15/41), only SIZE significantly decreased in earlytreatment when compared with the baseline. There were significant differences in BF before and after treatment in HL(3/15). Using the independent-Sample t test, SIZE, BF, BV and PS decline rate higer in responders than in the nonresponders, the difference is statistically significant. By ROC curve research that for the identification of nonresponse were reached at less than 49.7% reduction in PS(AUC, 0.905; sensitivity, 84.6%; specificity, 86.7%), less than 26.4% reduction in BF(AUC, 0.885; sensitivity, 88.5%; specificity, 86.7%), less than 19% reduction in BV(AUC, 0.785; sensitivity, 92.3%; specificity, 53.3%),less than 77.1% reduction in SIZE(AUC, 0.733; sensitivity, 69.2%; specificity, 80%). Baseline values of PS in responders as compared with the nonresponders were statistically significant, By ROC curve research that the cut-off value of 31,76 ml/100g/min in PS(AUC, 0.708; sensitivity, 69.2%; specificity, 66.7%).Earlytreatment values of BV and PS in responders as compared with the nonresponders were statistically significant, AUC<0.5, no predictive value. SIZE、BF、MTT difference was not statistically significant(P>0.05). Conclusion:BF, BV and PS in baseline can be used to differentiate lymphoma and normal muscle. The rate of decline in the early perfusion parameters of lymphoma chemotherapy can be used as a quantitative indicator for evaluating the early effect of chemotherapy. The decline rate of PS and BF was more sensitive and specific than that of SIZE in the evaluation of lymphoma. The PS in baseline for lymphoma has a certain value in forecast of the Curative Effect. So far, it has not been confirmed that the perfusion value and SIZE of earlytreatment can predict the effect of chemotherapy.