| Objective:Precision radiotherapy is an important local treatment for those patients with colorectal liver metastases(CRLM).The aim of this study is to evaluate the efficacy and safety of stereotactic body radiation therapy(SBRT)and intensity modulated radiation therapy(IMRT)for CRLM patients.Methods and Materials:This study retrospectively analyzed 120CRLM patients who were either unsuitable for or unwilling to receive operation for liver metastases between July 2014 and July 2021 at the Third Xiangya Hospital of Central South University,including 35 patients who locally received SBRT and 28 patients who locally received IMRT and 57 patients who locally received no treatment.Besides,37 pairs of CRLM patients with local SBRT or IMRT and without local treatment were matched by propensity score matched(PSM)and analyzed.Then,their clinical data,imaging data and survival data were monitored and collected.Kaplan-Meier method was used to calculate the overall survival(OS),local control(LC)and progress-free survival(PFS)of CRLM patients who received SBRT or IMRT.OS condition was compared between“precision radiotherapy group”and“no local treatment group”,the survival curve was plotted,the COX regression model was analyzed by univariate and multivariate,the efficacy and survival of the subgroups of the indicators with statistical differences in multivariate analysis were compared by Log-Rank method and the toxic effects of patients during radiotherapy were recorded.Besides,the survival prognosis models of clinical risk score(CRS)and tumor burden score(TBS)were also compared for the accuracy of the OS in the CRLM population who received precision radiation therapy in this study.Results:For 35 CRLM patients undergoing SBRT,the median OS was 39 months,and the OS rates of 1-,3-and 4-years were 93.2%,54.7%,and 23.5%,respectively.The univariate analysis showed that the volume of GTV(p=0.010)was related to OS,and the multivariate analysis revealed that GTV volume(p=0.016)was an independent risk factor for their OS.The OS of CRLM patients was significantly shortened for those whose Volume of GTV exceeded 30.0cm~3(3-year OS rate:3.8%vs 57.6%).For 28 patients undergoing IMRT,the median OS was 23.3months,and the OS rates of 1-,3-and 4-years were 78.4%,49.8%,and19.9%,respectively.There was no statistical difference found in univariate analysis.Multivariate analysis showed that CEA level(p=0.039)was an independent risk factor for their OS.The OS of CRLM patients was shortened for those with elevated CEA levels(3-year OS rate:35.1%vs 70.4%).The median OS in“precision radiotherapy group”after PSM was 38.0 months for 37 pairs of CRLM patients,and the OS rate in the 1-,3-,and 4-year was 86.8%,52.0%,and 17.3%,respectively.The median OS in“no local treatment group”after PSM was 24.0 months for37 pairs of CRLM patients,and the OS rates in the 1-,3-,and 4-year were91.3%,2.8%,and 0%,respectively.There was statistical difference observed in OS between them(p=0.003).The objective response rate(ORR)of the wild lesions in 35 CRLM patients with SBRT was 94.2%,the median local control time was not reached,and the 1-and 2-year LC rates were 96.0%and 81.9%,respectively.Univariate analysis showed that the maximum diameter of the liver metastases(p=0.025)was related to the LC.Multivariate analysis suggested that the maximum diameter of the liver metastases(p=0.047)was an independent risk factor for their LC.The LC of CRLM patients was significantly reduced for those whose largest diameter exceeded 3.0 centimeter(2-years LC rate:23.4%vs88.9%).The ORR of 27 IMRT patients was 93.6%,the median local control time was 27.8 months,and the LC rates of 1-and 2-years were93.8%and 64.6%,respectively.There was no statistical difference found in univariate and multivariate analyses.The median PFS of 35 patients with SBRT was 10.2 months,and the PFS rates of 1-and 2-years were32.5%and 9.6%,respectively.Univariate analysis showed that the combination of systemic therapy(p=0.019)and the primary lesion site(p=0.005)were related to PFS.Multivariate analysis showed that the primary lesion site(p=0.029)was an independent risk factor for their PFS.The PFS of CRLM patients originating from right half colon was shorter than left half colon,and the PFS was the longest for rectum cancer(1-year PFS rate:0 vs 24.8%vs 53.3%).The median PFS of 27 patients with IMRT was 13.0 months,and the PFS of 1-and 2-years was 52.8%and8.3%,respectively.Univariate analysis showed that lymph node metastasis(p=0.008)and extrahepatic metastasis(p=0.043)related to PFS.Multivariate analysis showed that lymph node metastasis(p=0.005)and extrahepatic metastasis(p=0.024)were the independent risk factors for their PFS.Those without lymph node metastases were significantly reduced in their PFS(half-year PFS rate:25.0%vs 74.9%),and those with extrahepatic metastases were significantly reduced as well.(1-year PFS rate:30.8%vs 64.9%).There were controllable toxic effects in those patients undergoing radiotherapy,and one patient got gastric horn ulcer,which improved after treatments.There were no toxic side reactions of level 3 or above.The C-index of the TBS scoring model was 0.649,the AIC was 189.1,and the C-index of the CRS scoring model was 0.526 and the AIC was 191.3,which indicates that the accuracy of OS for TBS scoring model was higher than the CRS scoring model for those CRLMs who received precision radiotherapy in our study.Conclusions:1.SBRT and IMRT are the safe and effective local treatment modality for unoperative CRLM patients,and their survival time is significantly longer than those who had not been treated locally.2.For those CRLM patients with SBRT,GTV volume is an independent risk factor for their OS,the maximum diameter is an independent risk factor for their LC,and the primary lesion site is an independent risk factor for their PFS.3.For those CRLM patients with IMRT,CEA level is an independent risk factor for their OS,while lymph node metastasis and extrahepatic metastasis are the independent risk factors for their PFS.4.The TBS scoring model had a better prediction performance than the CRS scoring model for OS in our study,which can provide a reference for the risk stratification of pre-radiotherapy risk in CRLM patients. |
PDF Full Text Request