| Objective:The purpose of this study was to explore the distribution character of local regional failure in esophageal carcinoma after intensity modulation radiotherapy by image fusion technique, and investigate the susceptive factors of local regional failure. In order to provide reference for improving the local control and survival of esophageal carcinoma after definitive radiation by researching the relationship between local regional failure and dose-volume parameters of target volume.Methods: 103 patients with esophageal carcinoma treated with definitive radiation were reviewed from January 2009 through January 2013 at the third radiotherapy department of The Fourth Hospital of Hebei Medical University, their clinical and physics data were completed. There were 65 males and 38 females; 100 squamous cell carcinoma patients and 3 other pathology category patients; Sixty five patients received involved-field irradiation(IFI) and 38 patients received elective nodal prophylactic irradiation(ENI). All patients were irradiated with a prescription dose 54-66 Gy, median 60 Gy. According to the result of follow-up, CT image of patient who diagnosed local regional failure was transmitted to radiation treatment planning system and their original CT-based radiation treatment plans were reviewed. The position relations of the local failure position with the original treatment plans’ GTV, CTV, PTV, CTV1, PTV1 were observed and compared according to the anatomic structure. Univariate and multivariate statistical methods were used to analyze the clinical factors of local failure.The dose and volume parameters of original treatment plan were compared between local failure patients and non-local failure patients. The dose parameters including D98, D95, D50, D2 of GTV, CTV, PTV, the additional dose parameters: D98, D95, D50, D2 of CTV1, PTV1 for ENI. The volume parameters:V60, V55, V50 of GTV, CTV, PTV and additional volume parameter:V50, V45 of CTV1, PTV1 for ENI. The data was analyzed by SPSS13.0.Continuous and discontinuous data were analyzed by t test and chi-square test.Survival analysis was performed by Kaplan-Meier method, and logistic regression was used for multivariate analysis, P<0.05 was defined to be statistically significant.Results:1 All the patients were followed-up to December 31 th 2014, the follow-up period was 6.3-71 months, median follow-up was 32 months. For all patients, the 1-,3-,5-year overall survival rates and median survival time were 87%, 43%, 20% and 32 months, respectively. Fifty-one patients had local regional failure, 24 patients had distant metastasis and 40 patients had no evidence of disease at last follow-up. 1-,3-,5-year overall survival rates and median survival time of 51 local failure patients were 78%, 33%,11% and 18 months,respectively, they were significantly lower than 95%, 76%, 54% and 43 months with 52 non-local failure patients.(P=0.00).1-,3-,5-year overall survival rates and median survival time of 65 receiving IFI patients were 87%, 37%, 17% and 28 months,respectively, and 89%, 54%, 27% and 37 months of 38 receiving ENI patients.(P=0.09). 1-,3-,5-year overall survival rates and median survival time of 24 distant metastasis patients were 68%, 25%, 4% and 13 months,respectively.1-,3-,5-year overall survival rates and median survival time of 40 no evidence of any failure patients were 97%, 78%, 60% and 48 months,respectively.Of the 51 patients who had local regional failure, 46 patients had failure in GTV, 1 patient had failure in CTV, 2 patients had failure in PTV and 2 patients had failure out of PTV.2 Univariate analysis showed T stage, N stage, TNM stage were the influence factors of local region failure. Multivariate analysis showed patient who had late T stage(P=0.01), elder age(P=0.011), cervical-or upper-thoracic primary lesion(P=0.027)were easier to experience local regional failure and all these factors were the independent factors of local region failure.3 The dose and volume parameters of target volume were compared between 51 local region failure patients and 52 non-local failure patients, it was found that dose parameters GTV D98-D2, CTV D98-D2, PTV D50-D2 and volume parameters GTV V60, CTV V60 of non-local failure group were higher than those of local failure group, the difference was statistically significant. Volume parameters(GTV, CTV, PTV) V60 of local failure group were lower than 95%. Dose and volume parameters GTV D2, CTV D2, GTV V60, CTV V60 of 46 patients who had local failure in GTV were obviously higher than same stage of 46 patients in non-local failure group, with a statistically significant difference. And target volume parameter(GTV, CTV, PTV) V60 of local failure group were lower than 95%.Further analyzed in different stratification, with T1+2 stage, dose parameters CTV D98-D2, PTV D95-D2 and volume parameters CTV V60, PTV V60 of 24 non-local failure patients were higher than those of 13 local failure patients, the difference was statistically significant, and target volume parameter(GTV, CTV, PTV) V60 of local failure group were lower than 95%. When the primary tumor location was at lower-thoracic, dose parameters GTV D50-D2 and volume parameters CTV V60, PTV V60 of 15 non-local failure patients were higher than 8 local failure patients, with a statistically significant difference,target volume parameter V60 of local failure group were lower than 95%. In the group under 65 years old, dose parameters GTV D98-D2, CTV D98-D2, PTV D50-D2 and volume parameters GTV V60, CTV V60, PTV V60 of 27 non-local failure patients were higher than 18 local failure patients, difference was statistically significant, target volume parameter(GTV, CTV, PTV)V60 of local failure group were lower than 95%.4 Thirty-eight patients received ENI were compared with 65 patients received IFI in the entire cohort, dose parameters GTV D98-D2, CTV D98-D2, PTV D98-D2 and volume parameters GTV V60, CTVV60-55, PTV V60-55 of ENI were obviously higher than those of IFI, with statistically significant difference,the target volume parameter(GTV, CTV, PTV) V60 of IFI group were lower than 95%.In the local failure group, dose parameters PTV D98-D95 and volume parameters CTV V60, PTV V60-50 of 16 ENI patients were higher than 35 IFI patients, with a statistically significant difference, and the target volume parameter(GTV, CTV, PTV) V60 of IFI group were lower than 95%. In the non-local failure group, 22 ENI patients was compared with 30 IFI patients, dose parameters GTV D98-D2, CTV D98-D2, PTV D98-D2 and volume parameter PTV V60 of ENI were higher than IFI, with a statistically significant difference, and the PTV V60 of IFI was lower than 95%. In the ENI group, dose parameters GTV D98-D2, CTV D50-D2, PTV D50-D2 of 22 non-local failure patients were higher than 16 local failure patients, with a statistically significant difference. In the IFI group, volume parameters GTV V60, CTV V60, PTV V60 of 30 non-local failure patients were higher than 35 local failure patients, with a statistically significant difference, and the target volume parameter(GTV, CTV, PTV) V60 of local failure group were lower than 95%.Conclusions:1 Local regional failure was still the major failure model for esophageal carcinoma treated by definitive radiation with IMRT. Most of local failure occured within GTV.2 It was easier to experience the local regional failure for patient who had late T stage, elder age, cervical or upper thoracic tumor location.3 For reducing the local failure rates, the target volume contained prescription dose, especially PTV can not be lower than 95% prescription dose, that was consistent with the require of ICRU. |