The Experimental Study For MSCT Perfusion Imaging In Assessment Anatomical Target Of Body Tumour

bjective:In precision radiotherapy, delineation of target volume is one of the most important and basic steps , it is also the controversial focus in published studies. At present, the existing imaging technology was not clearly show infiltration and scope of infiltration, So, the determination of CTV is one of the most difficult challenges for radiotherapists.Generally,the radiotherapists delineate targets depending evaluation of risk and scope of infiltration,clinical experience,norm of ward. It is possible that the difference exists between conformal radiotherapy and real target.CT perfusion imaging can indirectly mirror tumor vascularity,reconstruct the pcolor perfusion maps reflecting tumor perfusion,show tumor outline more precisely depenging quantifying assessment tumor blood flow,blood volume,peak enhanced and so on. CT perfusion imaging can detect tumour whose morpha doesn't change,but only hemodynamics changes. Consequently, it can supply scientifc basis for the defining of CTV; However, many uncertainties persist. for example:which does the pcolor perfusion map show tomour edge closest to pathologic edge? whether does the microscopic extension exist beyond tomour edge which is mirrored by perfusion maps? How much is the margin of the perfusion maps to CTV? The purpose of our study were to discuss the application of MSCT perfusion imaging in target delineation of Intensity Modulated Radiation Therapy for body tumour, and lay the foudation for the individualization and objectivity in radiotherapy target delineation, contrasting tomour size between imaging-based and pathology–based through perfusion imaging for the soft tissue tumour of rabbits in the right proximal thigh . Methods : 1 34 New Zealand white rabbits were etablished the soft tissue tumour models in the right proximal thighs from June 2008 to December 2008. 2 About 14 days later, CT plain scan and perfusion scan were performed on these rabbits repectively, then the images were transmitted to Phillips Brilliance 190P workstation. the functional maps, reflecting tumor perfusion, were calculated and showed including BV maps,BF maps,PE maps,TTP maps. 3 Under default window setting , radiotherapists manually measured maximum values of long axis and short axis of the tumour maximum layer in contranst enhanced CT maps,BV maps,BF maps,PE maps , respectively. 4 The macroscopic specimens were also measured maximum values of long and short axis in the tumour maximum layer. Specimen conventional fixing, production, and HE staining, respectively. Infiltration were measured from the edge of the tumor with the light microscope eyepiece, and each one of the largest value were records. 5 Statistical analysis used SPSS 13.0 for Windows statistical software, the results were with mean±standard deviation (x_±s) . The two group datas were compared with paired t-test. The analysis of linear correlation was performed between GTV病理,CTV and GTV影像. p<0.05 is a significant difference. Results: 1 The comparison between GTVCECT and GTV 病理: long axis is 23.4±2.9 mm,23.3±2.5 mm,short axis is 16.1±2.8mm,15.6±2.2mm respectively,The mean values of long and short axis of GTV病理are larger than GTVCECT, however, the difference are not statistically significant(p>0.05). The result for the analysis of linear correlation between GTVCECTand GTV病理was: r长=0.890, p<0.01;r短=0.900, p<0.01. 2 The comparison between GTVBV,GTVBF,GTVPE and GTV病理: the long and short axis of GTVBV is 29.1±2.7 mm,20.1±2.8 mm respectively,the long and short axis of GTVBF is 28.8±2.4 mm,19.7±2.7mm respectively,the long and short axis of GTVPE is 29.1±2.6 mm,19.6±2.8 mm respectively. The mean values of long and short axis of GTVBV,GTVBF,GTVPE are larger than GTV病理, the difference are statistically significant(p<0.01). 3 The comparison between GTVBV,GTVBF,GTVPE and CTV : the long and short axis of CTV is 30.2±3.3 mm,20.8±3.2 mm respectively. The result for the analysis of linear correlation between GTVCECT and CTV was: r长=0.466, p>0.05; r短=0.776, p<0.01. the long and short axis of GTVCECT are shorter than CTV, the difference are statistically significant(p<0.01). The result for the analysis of linear correlation between GTVBV and CTV was: r长=0.884, p<0.01; r短=0.938, p<0.01. The result for the analysis of linear correlation between GTVBF and CTV was: r长=0.747, p<0.01; r短=0.871, p<0.01. The result for the analysis of linear correlation between GTVPEand CTV was: r长=0.715, p<0.01; r短=0.816, p<0.01. In 18 cases ,the long axises of 5 cases for GTVBV are larger than CTV, 13 cases are shorter than CTV; the short axises of 4 cases for GTVBV are larger than CTV, 14 cases are shorter than CTV. In 18 cases ,the long axises of 3 cases for GTVBF are larger than CTV, 15 cases are shorter than CTV; the short axises of 4 cases for GTVBF are larger than CTV, 14 cases are shorter than CTV. In 18 cases ,the long axises of 5 cases for GTVPE are larger than CTV, 13 cases are shorter than CTV; the short axises of 3 cases for GTVPE are larger than CTV, 15 cases are shorter than CTV. 4 The proportion of the margin of GTVCECT,GTVBV,GTVBF,GTVPE to CTV: the long and short axis of GTVCECT all need to add the margin to CTV in 15 cases, the proportion is 34±13%, 35±13% respectively. In 18 cases , the long axis of GTVBV of 13 cases need to add the margin to CTV, the proportion is 6±3%; the short axis of GTVBV of 14 cases need to add the margin to CTV, the proportion is 6±4%. In 18 cases , the long axis of GTVBF of 15 cases need to add the margin to CTV,the proportion is 9±2%; the short axis of GTVBF of 14 cases need to add the margin to CTV, the proportion is 9±5%. In 18 cases , the long axis of GTVPE of 13 cases need to add the margin to CTV, the proportion is 8±4%; the short axis of GTVPE of 15 cases need to add the margin to CTV, the proportion is 11±6%.Conclusion: 1 GTVCECT is in close agreement with GTV病理, In target delineation for 3DCRT and IMRT, we could use the GTVCECT instead of GTV病理. 2 BV maps,BF maps,PE maps can show tumor outline, are larger and clearer than contranst enhanced CT maps.TTP maps cann't show tumor outline. The long and short axis of GTVBV,GTVBF,GTVPE are larger than GTV病理(p<0.01) . 3 BV maps are closely cosisitent with CTV.The proportion of the margin of GTVBV to CTV is smallest.