| Background:Nasopharyngeal carcinoma (NPC) is one of the common malignant tumors in Southeast Asia, especially in Guangdong, Guangxi, Hainan, Jiangxi, Fujian and Hunan of China. Radiation therapy is recognized as the main treatment for NPC. Because of many organs at risk (OARs) adjacent to the nasopharynx, conventional or three-dimensional conformal radiotherapy technique can not avoid high dose of irradiation to OARs and high-grade radiation-induced toxicities. With the advances of radiotherapy technique, intensity-modulated radiotherapy (IMRT) was developed in the end of the20th century. With IMRT technique, the output intensity of each segment within beams can be modulated to make the treated dose volume to conform the tumor target volume in three-dimensions and reduce the radiation dose to OARs, thus improve the local-regional control probability and lower the radiation-induced toxicities. In NPC patient, the primary disease locates in the central of head with many surrounding OARs, in addition to the irregular shape of the target volume. IMRT was estabilished as standard radiotherapy technique for NPC soon. Many studies showed that the tumor local control rate was significantly increased and the radiation-induced toxicities reduced, with3-year local control rate up to90%and significantly lowered radiation-induced toxicities for patients with NPC treated using IMRT.Radiation-induced toxicities might seriously affect the quality of life in nasopharyngeal carcinoma patients after radiotherapy. Almost100%of the patients received conventional radiotherapy suffered severe xerostomia, over60%of patients would experience otitis media and hearing loss, and more than50%of patients would develop dysphagia and dysphonia5years after radiation therapy. Radiation-induced encephalopathy, dental caries and osteonecrosis of the jaws are also common radiation sequela. IMRT can deliver sufficient radiation dose conforming to the irregular target volume of nasopharyngeal carcinoma and reduced dose to the OARs around the target, thus provide technical support for improving the tumor local control and reducing the radiation-induced toxicities. However, any new technique needs the accumulation of users’knowledge and experience.The application of IMRT not only call higher request on the knowledge of sectional anatomy/imaging and radiation biology, but also pose new challenge to understanding of the radiotolerance of the OARs. Although many clinical results showed that IMRT could significantly improve the local-regional control and reduce the radiation toxicities in comparison to conventional radiotherapy, we still have a long way to go to take the most advantages of this new technique. To provide clinical dosimetry database for further reduction of radiation-induced toxicities in NPC patients treated with IMRT, we collected the data of radiation-induced toxicities, clinical characteristics and radiation dosimetry of nasopharyngeal carcinoma patients receiving IMRT from March2010to September2011, and analyzed the association between the acute radiation-induced toxicities (before radiotherapy,3months and6months after radiotherapy; including radiation-induced ototoxicity, salivary gland toxicity, oral toxicity and pharyngeal-laryngeal toxicity) and clinical and dosimetric factors.Objective:IMRT is a newly estabilished radiation technique for NPC. The purpose of this study is to analyse the relativity between radiation-induced toxicities and clinical dosimetric factors, and provide clinical dosimetric evidence for further reduction of radiation-induced toxicities in NPC patients treated with IMRT.Meterials and Methods:A total of80NPC patients treated with IMRT by our treatment panel from March2010to September2011were included in this study. The histologic diagnoses of all cases in this group were confirmed by pathologist of Nanfang hospital.75cases were diagnosed as non-keratinizing undifferentiated carcinoma (93.75%) and5cases non-keratinizing differentiated carcinoma (6.25%). The clinical stage distribution was as follows (according to AJCC7th staging criteria,2010):Stage Ⅰ7cases (8.8%), stage Ⅱ14cases(17.5%), stage Ⅲ34cases (42.5%) and stage Ⅳ25cases (31.3%). All underwent MRI scan head and neck, PET scan, chest X-ray and abdominal B ultrasound or whole body PET/CT scan. Impedance audiometry and pure tone audiometry were performed pre-therapy as a baseline. The mean values of bone and air conduction for pure tone audiometry at500,1000,2000,4000Hz were used as endpoints. The hearing threshold was evaluated according to the A-B gap change. It was considered clinically significant if the pre-and post-radiotherapy A-B gap change was equal to or greater than10dB. The results of tympanograms for Acoustic impedance were divided into A, B or C type:type A indicates normal, type B indicates dropsy of the middle ear cavity, and type C indicates eustachian tube disfunction. No change of acoustic impedance post-radiotherapy in comparison to pre-radiotherapy was recorded as A-A, B-B or C-C; changes of acoustic impedance form type A to C, A to B or C to B were recorded as A-C, A-B or C-B, which indicates deterioration; and B-C, C-A or B-A indicates improvement. According to ICRU50and62Reports, target volumes and OARs were delineated with MRI as routine reference imaging for head and upper neck (to hyoid bone level) for IMRT. Inferior neck (below the hyoid bone level) and supraclavicular region were treated with conventional split AP half-beam. The median follow-up time of this series was12months (6-18months). According to RTOG criteria, we graded xerostomia, pharyngalgia and hearing changes at the end of radiotherapy,3months and6months after radiotherapy. The dosimetric parameters of OARs (the average dose, V30, V40and V50of parotid; the average dose, V30, V50, D5and D10of oral cavity; the average dose, V40, V45, V50, D5, and D10of oropharynx; and the point dose at the center of tympanic cavity, internal acoustic pore, cochlear, vestibular, eustachian tube isthmus and the average dose of these five points) were measured and recorded from the dose statistics tables and dose-volume histogram on the treatment planning system.Statistic analysis:SPSS13.0version software was used for statistical analyses. The comparison among three groups was performed with one-way ANOVA.Using Fisher ANOVA when the dependent variable was of homogeneity of variance, and using Correction Welch when the dependent variable was of inhomogeneity of variance. The comparison between two groups was performed with LSD method when the dependent variable was of homogeneity of variance, and using Dunnett’s T3method when the dependent variable is of inhomogeneity of variance. The univariate and multivariate Binary Logistic regression were used to analyze the correlation of patient’s hearing loss within six months post radiotherapy with age, gender, tumor T stage, chemotherapy, the point dose at the center of tympanic cavity and eustachian tube isthmus.Variables with P<0.10were included in the multivariate analysis, and P<0.05was considered statistically significant. K Related Sample nonparametric test was used to compare the xerostomia at the end of radiotherapy,3months and6months post-radiotherapy. Univariate and multivariate Ordinal Logistic regression were used to analyze the correlation between salivary glands toxicity(xerostomia), oropharynx (mucositis, pharyngalgia), oral toxicity (oral mucositis) and clinical factors such as patient’s age, gender, clinical stage, T stage, chemotherapy and radiation dosimetry factors. Variables with P<0.10in univariate analysis were included in multivariate analysis, and P<0.05was considered statistically significant. ROC curve was used to obtain the best diagnosis critical point of the factors related to xerostomia at the end of radiotherapy, acute oral mucositis. oropharyngeal mucositis and pharyngalgia.Results:All80patients complied with the required follow-ups. In the80patients with160ears,14ears (8.75%) ameliorated,102ears had no changes (63.75%) and44ears aggravated (27.5%) in electro-audiometer post-radiotherapy. The difference of the point dose at the center of tympanic cavity, internal acoustic pore, cochlear, vestibular, eustachian tube isthmus and the average dose of these five point among the three groups(no change’s group, aggravated’s group and ameliorated group) have statistical significance(F values were7.713,8.786,9.661,8.627,4.990and10.810, respectively; P values were0.002,0.001,0.001,0.001,0.008and <0.001, respectively). Pairwise comparison results showed that the point dose at the center of tympanic cavity, internal acoustic pore, cochlear, vestibular, eustachian tube isthmus and the average dose of these five point between the no change group and the aggravated group have statistical significance (P values were0.001,0.001,<0.001,<0.001,0.002and<0.001, respectively). But there was no statistical significance between the no change’s group and the ameliorated group, as well as the ameliorated group and aggravated group (P values are greater than0.05). The acoustic impedance of18ears (11.25%) ameliorated,95ears had no changes (59.38%) and47ears aggravated (29.38%) post-radiotherapy. The difference of the point dose at the center of tympanic cavity, internal acoustic pore, cochlear, vestibular, eustachian tube isthmus and the average dose of these five point among the three groups (no change’s group, aggravated’s group and ameliorated group) have statistical significance (F values were8.063,9.038,9.062,8.001,4.688and10.679, respectively; P values were0.001,0.001,0.001,0.001,0.011and0.001, respectively). Pairwise comparison results showed that the differences of the point dose at the center of tympanic cavity, internal acoustic pore, cochlear, vestibular, eustachian tube isthmus and the average dose of these five point had statistical significance (P values were0.001,0.001,0.001,0.001,0.003and0.001, respectively) between the no change’s group and the aggravated’s group, but there was no statistical significance between the no change’s group and the ameliorated group, as well as the ameliorated group and aggravated group (P values were greater than0.05). The average dose to the center of tympanic cavity, internal acoustic pore, cochlear, vestibular and eustachian tube isthmus was34Gy,39Gy,39Gy,31Gy and58Gy, respectively, in patients with no change in the electro-audiometer and acoustic impedance after radiotherapy; and was39Gy,45Gy,46Gy,37Gy and62Gy, respectively, inpatients with worsen results in electro-audiometer and acoustic impedance. There were65ears (40.63%) had hearing loss and95ears (59.37%) have no significant change within6months after radiotherapy. In univariate binary classification Logistic regression analysis, age, T stage, the point dose at the center of tympanic cavity and eustachian tube isthmus were related to the hearing loss6months after radiotherapy (P values were less than0.10). The multivariate binary classification Logistic regression analysis showed only the age and the point dose at the center of tympanic cavity were related to the the hearing loss6months after radiotherapy (P values were O.001and0.026; OR values were1.056and1.075)The xerostomia incidence rate of Grade0～Ⅲ were0(0%),11(13.8%),54(67.5%) and15(18.8%) at the end of radiotherapy,2(2.5%),41(51.3%),37(46.3%) and0(0%)3months after radiotherapy and17(21.3%),56(70%),7(8.8%) and0(0%)6months after radiotherapy, respectively. K Related Sample Nonparametric Test confirmed that with the time prolonged after radiotherapy, xerostomia gradually improved (Mean ranks were2.74,1.99and1.26, respectively; P<0.001).Univariate ordinal Logistic regression analysis showed that age, weight loss, the change of the distance between the lower edges of bilateral earlobe after radiotherapy, the average dose of parotid gland and the V30, V40and V50of parotid were related to the xerostomia at the end of radiotherapy (P values were less than0.10). Age, T stage, weight loss after radiotherapy, average dose of parotid gland, the parotid V40and V50were related to the xerostomia3months after radiotherapy (P values were less than0.10). Age, weight loss and the change of the distance between the lower edges of bilateral earlobe after radiotherapy, average dose of parotid, the parotid V30,V40and V50are related to the xerostomia6months after radiotherapy(P values were less than0.10). Multivariate ordinal Logistic regression analysis showed that the age and the average dose of parotid gland were related to the xerostomia at the end of radiotherapy (P values were0.016and0.008; OR values were1.052and1.568) and3months after radiotherapy (P values were0.003and0.005; OR values were1.078and1.784). The age, weight loss after radiotherapy and the average dose of parotid were related to the xerostomia6months after radiotherapy (P values were0.001, 0.046and <0.001, respectively; OR values were1.124,1.288and3.438, respectively). The best diagnosis critical point of the average dose to parotid obtained was27.2Gy for xerostomia at the end of radiotherapy (obtained from ROC curve, the area under the curve was0.852(P<0.001).The acute radiation-induced oral cavity mucositis was as follows:6cases (7.5%),33cases (41.3%),31cases (38.8%),10cases (12.5%) and0cases (0%) with Grade0-IV mucositis, respectively. Univariate ordinal Logistic regression analysis revealed that the clinical stages, average dose to oral cavity, V30、V50、D5and D10were related to the acute oral cavity mucositis occurred during the radiotherapy (P values were less than0.10).Multivariate ordinal Logistic regression analysis showed that only the average dose to oral cavity was related to the acute oral cavity mucositis (P=0.025, OR=1.490). ROC curve demonstrated that32.19Gy was the best diagnosis critical point of the average dose to oral cavity for acute oral cavity mucositis (the area under the curve was0.776(P<0.001)The acute oropharynx mucositis was as follows:Grade0-IV in0case (0%),9cases (11.3%),36cases (45%),35cases (43.8%) and0case (0%), respectively. Univariate ordinal Logistic regression analysis revealed that clinical stage, induction chemotherapy, concurrent chemotherapy, the average dose to oropharynx, oropharyngeal V40, V45, V50, D5and D10were related to the acute radiation-induced oropharynx mucositis (P values were less than0.10).Multivariate ordinal Logistic regression analysis showed that only concurrent chemotherapy and the oropharyngeal were in connection with the acute oropharynx mucositis (P values were0.032and0.043; OR values were0.242and1.151).The best diagnosis critical point of the oropharynx V50was11.03%for acute oropharyngeal mucositis (obtained form ROC curve, and the area under the curve was0.826(P=0.001)Pharyngalgia during radiotherapy course was as follows:Grade0-3in0case (0%),26cases (32.5%),37cases (46.3%) and17cases (21.3%), respectively. Univariate ordinal Logistic regression analysis showed that the clinical stage, induction chemotherapy, the average dose to oropharynx, V40, V45, V50, D5and D10were related to the grade of pharyngalgia during the radiotherapy (P values were less than0.10). Multivariate ordinal Logistic regression analysis showed that induction chemotherapy, the average dose to oropharynx and D5were related to the grade of pharyngalgia. ROC curve showed the best diagnosis critical point of the average dose and D5to oropharynx for grade2-3pharyngalgia were38.1Gy and54.385Gy, respectively. Area under the curve were0.911and0.938, respectively (P<0.001)Conclusion:1. After IMRT, radiation-induced otitis media with effusion in NPC patients was related to the dose to middle and inner ear structures. We suggest that the dose to the center of tympanic cavity, internal acoustic pore, cochlear, vestibular and eustachian tube isthmus are limited to34Gy,39Gy,39Gy,31Gy and58Gy, respectively.2. Xerostomia after IMRT was related to the patient age, average dose to parotids, and body weight loss during radiotherapy. Parotid gland’s function could gradually recover with the time prolonged after radiotherapy.Most of the patients were only0-I grade xerostomia at6months after radiotherapy.The grade of xerostomia post-radiotherapy was significantly greater in patients with the average dose to parotids more than27.2Gy. We suggest that the average dose to parotids should be limited to27Gy and the nutrition support during radiotherapy need to be enhanced.3. Acute radiation-induced oral cavity mucositis is related to the average dose to oral cavity. The grade of acute oral cavity mucositis was significantly higher when the average dose to oral cavity was greater than32.19Gy and we suggest that32Gy might be set as limit average-dose to oral cavity. 4. Acute radiation-induced oropharyngeal mucositis was related to concurrent chemotherapy and V50of pharynx and larynx. The grade of acute radiation-induced oropharyngeal mucositis was significantly greater in patients with V50>11%and concurrent chemotherapy.5. The degree of pharyngalgia was related to induction chemotherapy, the average dose and D5to oro-hypopharyx and larynx. The grade of pharyngalgia was significantly higher when the average dose to oro-hypopharyx and larynx was more than38Gy or the D5was greater than54Gy or induction chemotherapy. |
PDF Full Text Request