RESUMEN
Objective:To propose an automatic planning approach for Eclipse15.6 planning system based on Eclipse scripting application programming interface (ESAPI) and evaluate its clinical application.Methods:20 patients with nasopharyngeal carcinoma and 20 cases of rectal cancer were selected in the clinical planning. The developed automatic planning script SmartPlan and RapidPlan were used for automatic planning and dosimetric parameters were compared with manual planning. The differences were compared between two groups by using Wilcoxon signed rank test. Results:The dosimetric results of automatic and manual plans could meet clinical requirements. There was no significant difference in target coverage in nasopharyngeal carcinoma planning between two groups ( P>0.05), and automatic plans were superior to manual plans in organs at risk sparing ( P<0.05). Except for the homogeneity index of PTV and the maximum dose of bowel in rectal cancer plans, the other dosimetric parameters of the automatic plans were better than those of the manual plans (all P<0.05). Conclusions:Compared with the manual plans, the automatic plans have the same or similar target coverage, similar or better protection of organs at risk, and more convenient implementation. The developed SmartPlan based on ESAPI has clinical feasibility and effectiveness.
RESUMEN
Objective:To help clinicians simplify the post-processing operations of structures by developing rapid processing software for target area and organs at risk structures based on ESAPI.Methods:SmartStructure script software was developed based on ESAPI, verified and evaluated in clinical work. 10 cases of rectal cancer receiving neoadjuvant radiotherapy, 10 breast cancer treated with postoperative radiotherapy, 10 cervical cancer receiving postoperative radiotherapy, 10 nasopharyngeal carcinoma receiving radical radiotherapy and 10 lung stereotactic body radiotherapy (SBRT) were selected, and different types of tumors had different post-processing operations of structures. In each case, three methods were used for post-processing of structures. In the control group (manual group), normal manual processing was employed. In the experimental group 1(SmaStru-N group), scripts without templates were utilized. In the experimental group 2(SmaStru-P group). scripts combined with templates were adopted. The processing time of the three methods was compared. Clinicians scored the scripting software from multiple aspects and compared the feeling scores of scripting software and manual operation.Results:All three methods can be normally applied in clinical settings. The error rate in the manual group was 7.0%, 3.0% in the SmaStru-N group 0% in the SmaStru-P group, respectively. Compared with the manual method, SmaStru-N shortened the processing time of target area and organs at risk by 60.9% and 93.3% for SmaStru-P. In addition, SmartStructure was superior to manual method in terms of using feeling scores. Clinicians gave lower score for the" applicability" and" simplicity" , and higher score on the" accuracy" and" efficiency" .Conclusions:Compared with conventional manual structure processing method, SmartStructure software can rapidly and accurately process all structures of the target area and organs at risk, and its advantages become more obvious with the increasing number of structures that need to be processed. SmartStructure software can meet clinical requirements, reduce the error rate, elevate processing speed, improve the working efficiency of clinicians, providing basis for the development of adaptive radiotherapy.
RESUMEN
Objective:To compare the dosimetric parameters and plan complexity between newly-delicated HyperArc (HA) and conventional volumetric-modulated arc therapy (VMAT) in the treatment of brain metastases.Methods:For 26 patients with brain metastases, HA, conventional coplanar (Cop) and non-coplanar (Non-cop) VMAT plans with a prescription dose of 9 Gy 3fx or 6 Gy 5fx were generated. The dosimetric parameters for planning target volume (PTV), RTOG conformity index (RTOG CI), Paddick CI, homogeneity index (HI), gradient index (GI), maximum dose (D max) of brainstem and dose-volume parameters of brain-PTV(V 2Gy-V 26Gy) were statisticaly compared among these three approaches. In addition, the monitor unit (MU) and the plan complexity parameters (including MCSv and AlPO) were statistically compared. Results:To prevent missed targets during treatment, all plans were established with RTOG CI of greater than 1.1. For Paddick CI, HA provided significantly higher conformity (0.89±0.019) than Non-cop (0.87±0.036, P=0.001) and Cop (0.88±0.017, P=0.003) VMAT. For GI, the fastest dose fall-off was noted in HA (3.35±0.64), followed by conventional Non-cop VMAT of (3.70±0.80), and conventional Cop VMAT of (4.90±1.85)(all P<0.05). For the brainstem sparing, HA plan performed better than Non-cop plan[(604.14±531.61) cGy vs.(682.75±558.22) cGy, P<0.05)]. For normal brain tissue sparing, HA approach showed significant reduction than conventional Cop and Non-cop VMAT (both P<0.05). For MU, HA approach (2 872.60 ± 566.93) was significantly lower than those of Non-cop VMAT (3 771.28 ± 1 022.38, P<0.05) and Cop VMAT (4 494.08 ± 1 323.09, P<0.05). In terms of plan complexity, the MCSv of Cop plan was the lowest, indicating that the complexity was the highest ( P<0.05). The AlPO of HA was significantly higher than that of Non-cop VMAT ( P<0.05), suggesting that the complexity of HA plan was lower ( P<0.05). Conclusion:For the treatment of brain metastases, HA provides better conformity, more rapid dose fall-off, better sparing of brainstem and normal brain tissues and less plan complexity compared with conventional VMAT.
RESUMEN
The segmentation of organs at risk is an important part of radiotherapy. The current method of manual segmentation depends on the knowledge and experience of physicians, which is very time-consuming and difficult to ensure the accuracy, consistency and repeatability. Therefore, a deep convolutional neural network (DCNN) is proposed for the automatic and accurate segmentation of head and neck organs at risk. The data of 496 patients with nasopharyngeal carcinoma were reviewed. Among them, 376 cases were randomly selected for training set, 60 cases for validation set and 60 cases for test set. Using the three-dimensional (3D) U-NET DCNN, combined with two loss functions of Dice Loss and Generalized Dice Loss, the automatic segmentation neural network model for the head and neck organs at risk was trained. The evaluation parameters are Dice similarity coefficient and Jaccard distance. The average Dice Similarity coefficient of the 19 organs at risk was 0.91, and the Jaccard distance was 0.15. The results demonstrate that 3D U-NET DCNN combined with Dice Loss function can be better applied to automatic segmentation of head and neck organs at risk.
RESUMEN
Radiation-induced dermatitis (RD) is one of the most common complications induced by radiotherapy in cancer patients.The incidence and severity of RD are largely associated with the risk factors of patients.RD can severely affect the quality of life,even discontinue the treatment and reduce the tumor control rate in severe patients.Currently,randomized control trials evaluating the preventive measures for acute RD have been rarely conducted.The clinical interventions for acute RD significantly vary.In this article,the pathogenesis,risk factors,clinical manifestations,evaluation,prevention and treatment of RD were investigated.
RESUMEN
Objective To explore the feasibility of the volume modulated arc therapy ( VMAT) auto-planning based on template library ( TL). Methods VMAT plans of 68 patients diagnosed with postoperative rectal cancer in Eclipse system were retrospectively selected. The prescription dose was 50 Gy/25F. In 19 patients, the feature values of target and organs at risk were extracted as the vectors. The final optimized restricted conditions were saved as the TL. Then, the plans of 15 rectal cancer patients (10 cases from TL and 5 outside TL) were automatically optimized. According to the multi-dimensional vector similarity principle, the similarity parameter αwas defined. The designed program automatically selected the optimal-object template in an in-house software developed with Matlab. The dosimetric parameters of the auto-optimized plans with the optimal-object template (ATP) and the clinical plan (CP) were compared by the paired t-test. The changes in the dosimetric parameters and similarity parameter α were statistically compared by Pearson′s correlation analysis. The linear fitting of the dosimetric parameters with α was used by least squares method to explore the tendency of the ATP dosimetric parameters relative to CP. Results The dosimetric parameters of ATP for 13 cases were slightly worse than those of CP. The conformal index (P=0. 004), heterogeneous index (P= 0. 015),V40(P= 0. 003) and mean dose (P= 0. 022) of the intestine significantly differed. The α values of these 13 cases were 2. 67,2. 60,2. 60,2. 49,2. 67,2. 74,2. 72,2. 48, 2. 53,2. 86,2. 68,2. 56 and 2. 63. The α value was significantly correlated with the V40 or mean dose of the bladder (r= 0. 649 and 0. 603,P= 0. 016 and 0. 029).Along with the increase of α value ,V40 and mean dose of the intestine for ATP were gradually deteriorated than those of CP . The remaining dosimetric para-meters of ATP were close to or superior to those of CP. Conclusions The results of ATP are slightly worse than those of CP, whereas can satisfy the clinical requirements. The TL, the quality of TL and ATP remain to be further optimized.
RESUMEN
Objective To study the difference of the constructed doses between electronic portal imaging device (EPID) and dynalogs files of linac for in vivo phantom dosimetry.Methods Twelve pelvic patients treated with volumetric modulated arc therapy (VMAT) plans were selected and the information of each plan was copied to theCheese phantom to recalculate the doses before delivered on Varian RapidArc Linac.TheCheese phantom was placed on the isocenter and the electronic portal image (EPI) formed by the EPID was sent to EPIgray software to reconstruct the actual delivered doses.Meanwhile,dynalogs files were respectively imported to the Mobius software to reconstruct the actual delivered doses too.The point dose in the center of each VMAT plan (the center of the effective sensitive volume of ionization chamber) was measured by the A1SL ionization chamber.At the same time,the dose of sensitive volume of ionization chamber from treatment planning systcm (TPS) was recorded.Results The relative deviation between the dose from TPS and the measurement results by the ionization chamber was 1.13％.The difference between the reconstructed doses of EPID-based or the dynalogs file-based with the measurement results by the ionization chamber was not statistically significant (P ＞ 0.05).Conclusions The two methods of dose reconstruction can provide reference for in vivo dosimetry of VMAT.
RESUMEN
Cerebral radiation necrosis (CRN) is a serious complication of radiotherapy for intracranial tumors and skull base tumors.Since there lacked effective therapeutic methods in the past,CRN was once considered to be progressive and irreversible.With the development of histopathology and neuroimaging,the development and progression of CRN is gradually clarified,and new therapeutic methods have been developed.In recent years,the scholars at home and abroad have tried to use bevacizumab (a humanized monoclonal antibody targeting vascular endothelial growth factor),nerve growth factor,and ganglioside in the treatment of CRN and have achieved definite therapeutic effects.In some patients,cerebral necrosis was even repaired and reversed.This article reviews the incidence,pathophysiology,treatment,and prognosis of CRN.
RESUMEN
Objective To compare the dosimetric differences between helical tomotherapy (HT) and volume-modulated arc therapy (VMAT) in the treatment of upper thoracic esophageal carcinoma (UTEC).Methods A total of 10 patients with UTEC were randomly selected.HT plan and double-arc VMAT plan were designed and optimized for each patient.The prescription dose was 50 Gy/30 fractions for gross target volume (GTV), 66 Gy/30 fractions for planned target volume (PTV).The dose distribution and conformal index (CI), homogeneity index (HI) of target volume, the D1％, D5％, Dg5％, D99％, and dose of organ at risk (OAR) were analyzed by using the dose volume histogram (DVH).The monitor units and delivery time were also evaluated.Results For GTV and PTV, the D99％ of HT plans were slightly higher than those of VMAT plans (t =4.476, 3.756, P ＜ 0.05) , but no significant differences in D1％ , D5％ , D95％ , HI and CI (P ＞ 0.05) were found.The V10, V15, V20 and mean lung dose (MLD) to the total-lung of HT plans were all significantly lower than those of VMAT plans (t =-3.369,-4.824, -4.869,-3.657, P ＜ 0.05).There were no significant differences for V5, V30 and Dmax of cord (P ＞ 0.05).The monitor units and delivery time of VMAT plans were significantly lower than those of HT plans (t =13.970, 7.982, P ＜ 0.05).Conclusions Both HT and VMAT are appropriate for esophageal cancer radiotherapy.HT significantly reduces the radiation dose of the total-lung, while VMAT has obvious advantages in efficiency.
RESUMEN
Background and purpose:Neoadjuvant chemotherapy with docetaxel was gaining more attention in the treatment for locoregionally advanced nasopharyngeal carcinoma (NPC). The prediction of the sensitivity to neoadjuvant chemotherapy with docetaxel could enable clinicians to individualize treatment protocols for NPC. Recently we found that 99mTc-MIBI imaging in NPC could predict tumor response to chemotherapy with cisplatin plus 5-FU. However, there was no study to support similar findings in NPC patients receiving chemotherapy containing docetaxel. This study was to evaluate the value of double-phase 99mTc-MIBI SPECT/CT in predicting response to neoadjuvant chemotherapy with docetaxel-based regimen for nasopharyngeal carcinoma. Methods:Thirty-one nasopharyngeal carcinoma patients participated in this prospective study. Before treatment, early and delayed single-photon emission computed tomography/compute tomography (SPECT/CT) images were obtained instantly and 2 hours after an intravenous injection of 25-30 mCi 99mTc-MIBI. All patients received neoadjuvant chemotherapy consisting of docetaxel, cisplatin plus 5-FU for two cycles. The relationships between efficacy of neoadjuvant chemotherapy and the early uptake ratio, late uptake ratio and washout rate of 99mTc-MIBI were evaluated.Results:According to the MRI, the early uptake of 99mTc-MIBI (2.67±0.83) in the lesions which were sensitive to chemotherapy was significantly higher compared with that (1.69±0.46) in the insensitive lesions(P=0.003). The difference of the late uptake between the sensitive(1.46±0.39) and the insensitive (1.06±0.62) was also statistically significant (P=0.026). However, the washout rate of 99mTc-MIBI was not significantly different between the two groups (P=0.23). Through ROC curve analysis, the AUC for early uptake of 99mTc-MIBI and late uptake were 0.84. The sensitivity, specificity, positive predictive value and negative predictive value were 74.2%, 87.5%, 95.8%and 46.7%for early uptake when the cut off value of 1.97 was used. Conclusion:The uptake of 99mTc-MIBI in both early phase and late phase could predict the efficacy of neoadjuvant chemotherapy with docetaxel-based regimen.
RESUMEN
Objective Nasopharyngeal carcinoma patients with stage N3 disease are prone to develop distant metastasis even treated with standard concurrent chemoradiotherapy(CRT).The aim of this study is to compare the ettlcacy of difierent chemotherapy sequences in these patients.Methotis All patients with histologically proven,carcinoma of the nasopharynx treated between July 1999 and November 2003 were restaged according to the AJCC 2002 stage classification system.A total of 114 patients had AJCC N3 diseases were analyzed retrospectively.Patients were treated by conventional RT technique using 6 MV photons or 60 Coγ-ray with 1.8-2.0 Gy per fraction,5 fractions a week,to a planned dose of 70 Gy.The prophylactic irradiation dose of the neck wss 54-60 Gy.Any positive lymph node was boosted to a total dose of 60-68 Gy.All patients received cisplatin-based chemotherapy of difierent sequences but 9 patients RT alone.CRT regimen was delivered in 37 patients,neoadjuvant chemotherapy(NACT)+CRT regimen in 53 patients and CRT+adjuvant chemotherapy(AC)regimen in 15 patients.Results The prophylactic irradiation dose of the neck wss 54-60 Gy.Any positive lymph node was boosted to a total dose of 60-68 Gy.All patients received cisplatin-based chemotherapy of difierent sequences but 9 patients received RT alone.CRT regimen was delivered in 37 patients,neoadjuvant chemotherapy(NACT)+CRT regimen in 53 patients and CRT+adjuvant chemotherapy(AC)regimen in 15 patients.Results The median follow up time was 54 months(3-117months).The 5-year overall survival rate was 59.1%in whole groups,and with 17%,51%,68%and 71%in RT,CRT,NACT+CRT and CRT+AC group,respectively(X2=15.44,P=0.001).The 5-year relapse-free survival rates were 83%,77%,88%and 93%in RT,CRT,NACT+CRT and CRT+AC group,respectively(X2=2.34,P=0.505).The 5-year metastasis-free survival rates were 17%,54%,72%and 80%in RT,CRT,NACT+CRT and CRT+AC group,respectively(X2=19.28,P=0.000).Conclusions The NACT+CRT and CRT+AC regimens were more effective than CRT alone for N3 disease in the current study.Large prospective,randomized clinieal studies are warranted.
RESUMEN
Objective To evaluate set-up errors by megavoltage cone-beam CT in head and neck cancers treated with precision radiotherapy. Methods From April 2007 to March 2008, 22 patients with nasopharyngeal carcinoma (15 patients) ,parotid carcinoma (4 patients) and brain glioma (3 patients) were enrolled, among whom 7 patients underwent three-dimensional conformal radiotherapy (3DCRT) and 15 received intensity modulated radiotherapy (IMRT). The radiation dose was 56. 0 -70.4 Gy in 28 -32 fractions within 6 -7 weeks. Megavoltage cone-beam scan was performed weekly before treatment. The isocenter displacement was calculated by comparing megavoltage cone-beam CT and planning CT in left-right (LR), cranio-caudal (CC), and anterior-posterior (AP) directions. Results Totally 129 sets of megavoltage cone-beam CT images were obtained for 22 patients. The frequency of isocenter shift more than 0. 3 cm,0. 4 cm and 0. 5 cm were 28,15 and 9, respectively. The maximum mean set-up error was found in CC, which was about 0. 1 cm more than that in LR and AP. The frequency of isocenter shift direction was almost identical in LR and CC, which was more frequent (about 75%) in the posterior direction. Conclusions During the course of radiation of brain tumor and head and neck cancer, the enlarging tendency of set-up errors has been found in all the three directions. The isocenter shift in AP was more frequent to the posterior direction.
RESUMEN
Objective To summarize our experience and treatment results of nasopharyngeal carcinoma treated in a single institution. Methods From Jan. 2000 to Dec.2003,1837 patients with histologically proven nasopharyngeal carcinoma(NPC) were retrospectively analyzed. The disease was staged according to the Fuzhou stage classification. 885 patients received cisplatin (DDP) based chemotherapy. All patients received radiotherapy to the nasopharynx and neck. The dose was 30.6-74.0 Gy, 1.8-2.0 Gy per fraction over 3.5-8.0 weeks to the primary site with 60Co γ rays or 6 MV X-rays. The dose to lymph nodes was 60-68 Gy. The residual disease was boosted by 192Ir afterloading brachytherapy,small external beam fields, conformal radiotherapy,or X-knife. Results The median follow-up time was 54(3-90) months. The 5-year overall survival(OS), disease-free survival (DFS), relapse-free survival (RFS) and distant metastasis free survival(DMSF) rates were 67.42% ,63.25% ,86.47% and 80.31% ,respectively. Clinical stage was the most significant prognostic factor,and OS was 88% ,74.8% ,65.9% ,52.4% and 20% for stage Ⅰ ,stage Ⅱ,stage Ⅲ,stage ⅣA and stage ⅣB,respectively. Gender,T,N and TNM stage were the significant prognostic factors of OS in multivariate analysis. Conclusions For NPC patients,the 5-year OS of 67.4% is achieved by conventional radiotherapy technique in our institution. Both univariate and multivariate analysis shows that gender and clinical stage are the significant prognostic factors of OS.
RESUMEN
Elective neck irradiation is usually used as the modality of choice for the treatment of phase N 0 head and neck cancers. The increasing use of 3 dimensional conformal treatment planning has created an urgent need for new guidelines for the classification and delineation of the neck node areas. Surgical literature has provided us with valuable information in this field. The development of imaging technology has also offered us more detailed information. Several surgical and imaging methods have been proposed in this article, with the hope to give reference to most radiation oncologists. [
RESUMEN
Objective To evaluate the limitation of image-based classification and delineation of cervical nodes proposed in 1999. Methods 259 consecutive nasopharyngeal carcinoma patients received contrast-enhanced transverse CT scan from July to November in 2003, the scanning range extended from the skull base to the inferior border of the clavicle, with thickness set at 5?mm per slice from the skull base to the oropharynx, and 1?cm per slice from the oropharynx to the clavicle. Interpretation of the images and assessment of the involved nodes distribution were performed by both radiation oncologists together with diagnostic radiologists according to 1999 image-based classification and delineation of the neck. Then we evaluated whether this classification could fully cover all the lymphatic drainage areas of the neck. Results 218 cases had nodal involvements. Nodal distributions were 0 in levelⅠA, 6 in levelⅠB, 136 in level ⅡA, 171 in level ⅡB, 78 in level Ⅲ, 20 in level Ⅳ, 33 in level VA, 27 in level VB,0 in level Ⅵ and Ⅶ, 5 in supraclavicle and 102 in retropharyngeal space, respectively. Among another 57 cases who belonged to none of the above levels, 2 cases had lesions in the preauricular area, the rest 55 medial to the border of internal carotid artery, 2?cm from inferior to the skull base to the hyoid bone. Conclusions The 1999 image-based classification of the neck nodes,being essentially rational, did not fully cover the retropharyngeal space. It would be better to shift the inferior boundary of the retropharyngeal space to the level of the hyoid bone.