23EX直线加速器 7年故障统计分析经验

目的 研究故障统计分析在23EX加速器日常维护中的应用。方法 收集2008-2015年间23EX加速器的故障数据。分析23EX加速器故障率与时间关系,利用帕累托图统计分析影响23EX加速器稳定性的主要子系统和主要零部件。结果 23EX加速器运行 7年共发生故障318次,更换零部件358个。加速器投入使用后1.5年内故障率逐渐增加,最高时达38次/半年,之后有所下降并保持相对稳定,1年中第3季度是加速器故障率最高时间段。影响23EX加速器稳定性的主要子系统是MLC和机械系统,分别占总故障数66.4%和11.9%;影响23EX加速器稳定性的主要零部件是MLC叶片螺母和叶片马达,分别占总更换零部件的38.8%和28.5%。结论 加速器故障统计分析有助于维护人员发现故障的分布规律,找出影响加速器稳定性的主要因素,为制定日常维护计划、改善维护策略提供依据。     

不同CT扫描条件模拟定位对放射治疗计划的影响

目的:利用不均匀等效模体,探讨不同CT扫描条件对CT值及照射跳数(MU)的影响.方法:采用大小2种CT孔径(80和70 cm)、2种模体几何摆放顺序及2种扫描电压(120和140 KV),测量不同组合,比较各自的CT值,建立相应的CT-电子密度(ED)转换曲线,选取盆腔、胸部、头颈部各10例患者的CT图像模拟适形计划(CRT)和调强计划(IMRT),分析照射MU数值的偏差.结果:对于小孔径CT,无论扫描电压、模体几何位置如何变化,其MU数值相差均≤0.1％；大孔径CT扫描电压改变对MU值无影响,模体几何摆放顺序的改变有影响,但＜0.3％；CRT计划和IMRT计划各自偏差值相同.结论:对于精确放疗计划系统,CT扫描条件改变和测量模体位置改变均可引起照射MU数值误差.若使用大孔径CT进行模拟定位,需根据其特性建立合理的电子密度曲线,并应用在计划制定中.     

基于R语言BP神经网络瓦里安NovalisTx直线加速器MLC系统故障预测模型研究

目的 构建并研究瓦里安NovalisTx直线加速器MLC系统故障预测BP神经网络模型。方法 取加速器临床使用18个月MLC系统故障统计数据为研究对象,以加速器使用总时间、月治疗患者数量、日均开机工作时间、RapidArc计划数量及加速器保养后时间间隔为输入故障因素,以故障频次预测为输出结果,采用R语言AMORE包构建MLC系统故障预测BP神经网络模型并对其进行仿真验证。结果 模型采用3层网络实现输入输出转换,其输入层5个节点、隐层13个节点、输出层1个节点;输入层至隐层、隐层至输出层分别选用tansig、purelin传递函数;模型设定最大训练学习次数150次,实际使用111次,设定误差3%,实际误差2.7%,表明其收敛较好。该模型对18个月临床故障数据仿真验证结果表明预测数据与实际数据较为接近。结论 基于R语言BP神经网络故障预测模型实现了MLC系统故障因素与故障频次间映射关系描述,可为设备故障规律了解和备件库存管理提供参考。     

肺低剂量区体积预测急性放射性肺炎价值探讨

目的 观察胸部肿瘤三维适形放疗患者放射性肺炎发生情况,分析其与各临床、剂量学因素关系,探讨低剂量区体积对放射性肺炎的预测价值.方法 2005-2008年本科收治的中晚期非小细胞肺癌(NSCLC)及食管癌患者共161例接受了三维适形放疗,其中局部晚期NSCLC患者53例,处方剂量60 Gy分30～34次,均行长春瑞滨+顺铂同期化疗;食管癌患者108例,处方剂量58～70 Gy分29～35次,单纯放疗46例,余62例接受亚叶酸钙+氟尿嘧啶+顺铂同期化疗.对急性放射性肺炎进行Spearman等级相关分析、Logistic因素分析及受试者工作特征(ROC)曲线分析.结果 随访率100%.全组急性放射性肺炎总发生率为57.8%(93例),其中NSCLC组为94%(50例,4、5级各1例),食管癌组为39.8%(43例,无≥4级病例).等级相关分析结果显示患者性别(r=0.19,P=0.016)、大体肿瘤体积(r=0.52,P=0.000))、平均肺剂量(r=0.33,P=0.000)、肺正常组织并发症概率(r=0.30,P=0.000)、接受5、10、15、20、25、30 Gy照射的肺体积百分比(肺V5～V30,r=0.21～0.29,P=0.000～0.027)均与放射性肺炎发生相关.Logistic因素分析结果显示肺V5(X2=7.07,P=0.008)、大体肿瘤体积(X2=10.21,P=0.001)是预测≥2级放射性肺炎最有价值指标.ROC曲线分析结果显示曲线下面积为0.684,P=0.000;曲线界值为V5=55%.肺V5≥55%组与<55%组≥2级放射性肺炎发生率分别为43%(36/84)和18%(14/77).结论 平均肺剂量、正常组织并发症概率、V5～V30可较好预测放射性肺炎的发生,其中V5可能是最有价值的预测性指标.当V5>55%时≥2级的急性放射性肺炎的发生率可能会明显增加,制定治疗计划时除平均肺剂量、V20、V30外,还应将低剂量区体积限制在适当范围内.     

P-POSSUM评分在预测食管癌术后死亡风险中的临床研究

目的:研究评估改良的POSSUM(Portsmouth-POSSUM,P-POSSUM)在食管癌患者术后死亡风险中预测能力.方法:对2012年1月到2012年12月四川省肿瘤医院556例接受食管癌手术的患者进行回顾性分析,收集资料进行P-POSSUM评分,预测术后死亡发生率,与实际值进行比较,并用实际例数/预测例数(O/E值)及ROC曲线验证该评分系统的预测能力.结果:P-POSSUM评分预测的死亡发生率为5.22％(29例),与实际的死亡发生率3.42％(19例)比较,差异无统计学意义(x2=2.177,P=0.140),O/E为0.66,ROC曲线下面积(AUC)为0.713,P=0.002.结论:P-POSSUM评分在食管癌术后死亡风险方面具有较好的预测能力,对临床诊治具有一定指导意义.     

原发性肝癌术后发生静脉血栓栓塞症的危险因素及Caprini风险预测模型的应用

目的:研究原发性肝癌术后发生静脉血栓栓塞症（venous thromboembolism,VTE）的危险因素,并验证和改良Caprini模型对肝癌术后患者VTE发生的预测能力。方法:对我院收治的452例肝癌患者进行回顾性分析,根据术后1月内是否发生VTE而分为VTE组和非VTE组。采用多因素Logistic回归以用来筛选VTE发生的独立危险因素。采用受试者操作特征曲线（receiver operating characteristic,ROC）和曲线下面积（area under ROC curve,AUC）来描述和比较传统Caprini模型和改良Caprini模型对VTE发生预测的准确性。结果:共有41例原发性肝癌患者术后出现VTE,整体发生率为9.07%。单因素分析示BMI、糖尿病患病率、门静脉癌栓发生率、手术时间、二次手术率、以及Caprini评分可能与VTE的发生有关。多因素分析示BMI（OR=1.14,P=0.01）、手术时间（OR=10.91,P=0.001）、有门静脉癌栓（OR=4.98,P=0.001）、二次手术（OR=7.85,P=0.01）和Caprini评分（OR=2.63,P=0.001）是VTE发生的独立危险因素。改良后的Caprini模型和一般Caprini模型在预测VTE时的AUC分别为0.912和0.811;当取最大约登指数时,二者敏感度分别为85.37%和63.41%,特异度分别为85.64%和87.59%。结论:BMI、手术时间、门静脉癌栓、二次手术是原发性肝癌患者术后VTE发生的独立影响因素,联合上述四种指标可以显著提高Caprini模型对VTE的预测能力。     

剂量调整的BEACOP方案治疗淋巴瘤不良反应分析

目的 研究剂量调整的BEACOP(DA-BEACOP)方案用于预后不良淋巴瘤患者的不良反应.方法 对2002年8月至2012年2月间接受1次以上DA-BEACOP方案化疗的89例预后不良淋巴瘤患者发生的治疗相关性不良反应进行回顾性分析,方案组成:CHOP方案联合依托泊苷和(或)博来霉素,有或无利妥昔单抗.结果 74例(83.1％)患者发生3～4级化疗不良反应,主要为骨髓抑制(73例,82.0％),未发生治疗相关死亡,女性患者3～4级化疗不良反应的发生率比男性高,分别为94.2％(33/35)、75.9％(41/54)(P=0.024).曾接受3次及以上化疗者较接受2次及以下化疗者易出现3～4级骨髓抑制,发生率分别为95.9％(47/49)、75.0％(30/40)(P=0.004),曾接受放疗的患者较未接受放疗者3～4级骨髓抑制风险高,发生率分别为100.0％(20/20)、78.2 ％(54/69)(P=0.022).结论 DA-BEACOP方案在淋巴瘤患者治疗中的不良反应具有可控性.     

大孔径CT模拟定位机扫描条件对CT值影响分析

目的:分析大孔径CT模拟定位机扫描条件对CT值的影响,为模拟定位工作中不同扫描条件的选择提供参考。方法:采用飞利浦大孔径CT模拟定位机扫描CIRS-062M电子密度模体并建立CT值-相对电子密度曲线。分析管电流、管电压、层厚、层间距、分辨率、准直器、旋转时间、重建模型、过滤器、几何位置多种扫描条件对CT值的影响。结果:管电压变化对CT值影响最为显著,90 kV、140 kV管电压CT值平均偏差66.1 HU、-24.0 HU,差异主要集中在高密度区域。过滤器YA、YB对CT值影响明显,CT值平均偏差-40.7 HU、-39.9 HU。不同扫描位置对CT值有一定影响。结论:管电压、部分过滤器、扫描位置对CT值的影响不可忽略,临床工作中需规范上述参数的使用,以免造成剂量计算偏差。     

计算机断层扫描图像放射组学分析鉴别诊断磨玻璃密度肺腺癌的浸润性

[目的]通过放射组学分析方法研究鉴别在计算机断层扫描(computed tomography,CT)图像中表现为亚实性磨玻璃结节(ground-glass nodules,GGNs),是属于侵袭性肺腺癌(invasive pulmonary adenocarcinomas,IPAs)或非IPA,并结合传统CT图像定性特征与其他临床特征制定诊断IPA诺模图模型。[方法]回顾性收集2015年2月至2019年4月在新乡医学院第一附属医院进行手术确诊的88例患者,共计100个亚实性结节(56个IPA和44个非IPA)。选取增强CT动脉期图像进行3D结节感兴趣区的分割并计算定量放射组学特征。使用逻辑回归分析将一组常规临床风险因素和放射医生视觉评估的定性CT成像特征与放射学特征进行比较。建立3种诊断模型,即使用临床风险因素和CT定性特征的基础模型,使用包含具有统计学意义的放射组学特征模型,以及结合所有重要特征的诺模图模型,并根据接受者操作特性曲线(receiver operating characteristic curve,ROC)对三种模型的诊断效能分别进行比较。[结果]除了3个视觉评估的CT定性成像特征外,还发现从数百个放射学特征中选择的另外三个定量特征(P<0.05)与诊断IPA显著性相关。ROC曲线下面积(area under the curve,AUC)的显示采用诊断诺模图模型区分IPA与非IPA的性能最佳(AUC=0.903),均高于基础模型(AUC=0.853,P=0.0009)或放射组学模型(AUC=0.769,P<0.0001)。决策曲线分析也表明在临床诊断中使用此诺模图模型的潜在益处。[结论]除临床评估的CT图像定性特征外,定量放射学特征为鉴别IPA和非IPA提供了有效帮助,基于以上两类重要特征的诊断列线图模型在临床上可用于术前决策。     

Ⅲ期非小细胞肺癌同步放化疗所致急性食管损伤研究

目的 观察三维适形放疗(3DCRT)联合长春瑞滨+顺铂(NP)同步化疗Ⅲ期非小细胞肺癌(NSCLC)的急性放射性食管炎发生情况,并对相关因素进行分析,以求得到合理的预测性指标.方法 37例Ⅲ期NSCLC患者接受3DCRT及NP方案同步化疗,放疗处方剂量60 Cy,1.8～2.0Cy/次,5次/周,共30～34次,于放疗开始的第1、5周给予同步化疗,观察急性放射性食管炎发生情况并进行相关因素及受试者工作特征(ROC)曲线分析.结果 全组食管炎总发生率为92%(34例),其中0级3例,1级11例,2级9例,3级14例,无4级发生.单因素分析显示食管接受的平均剂量、LETT_(40)、LETT_(45)、LETT_(50)、LETT_(55)、LETT_(60)、V_(40)、V_(45)、V_(50)、V_(55)、V_(60)均与放射性食管炎具有较好相关性,能预测急性放射性食管炎发生;多因素分析结果提示食管V_55是预测放射性食管炎的最有价值的指标.ROC曲线分析结果示曲线下面积为0.906(P=0.000),曲线界值V_(55)=30%;食管V_(55)>30%组与≤30%组≥2级放射性食管炎发生率分别为100%(15/15)和36%(8/22).结论 对局部晚期NSCLC进行3DCRT结合同步化疗可出现较严重放射性食管炎,食管接受的平均剂量、LETT_(40)、LETT_(45)、LETT_(50)、LETT_(55)、LETT_(60)、V_(40)、V_(45)、V_(50)、V_(55)、V_(60)可较好预测放射性食管炎的发生,V_(55)可能是最有价值的预测性指标,当V_55>30%时2、3级急性放射性食管炎的发生率可能会明显增加.     

The six-year operation faults statistics analysis and prediction of Philips Brilliance big bore CT

Objective To analyze the 6-year operation faults of PHILIPS Brilliance big bore CT, identify the common problems, make corresponding maintenance plans, reduce the incidence of failures, and carry out simulation prediction of the occurrence rate of failures in the next few years. Methods The failure data of Brilliance big bore CT from June 2012 to June 2018 were collected, and the curve estimation function in SPASS 19.0 software and the pareto diagram were used to analyze the relationship between the number of failures, time and failure types, and the prediction was made. Results A total of 28 faults occurred during the 6-year opeation of Brilliance big bore CT. During the first half year, five times of faults occurred with the highest fault rate and then tended to stabilize. The linear function model was obtained using the curve estimation:y=-0.033x+2.099(y for the number of fault, unit for times, x for the unit of time for half a year), the model of R²=0.003. In the next three years, approximately twice faults occurred within half year. The pareto chart showed that 16 faults occurred during data collection, including 3 faults in the treatment bed and 3 faults in the power supply system, respectively. The accumulative ratio of the above three faults was 71.4%, which were the main fault sources. Conclusion The fault statistical analysis of Brilliance big bore CT is helpful for department maintenance personnel to better understand CT, develop effective maintenance programs, reduce the occurrence of faults, and predict the incidence of faults in the future.     

Carcinoma of the nasopharynx: survival and patterns of recurrence

We have analyzed survival, disease-free (NED) survival, and patterns of recurrence for 76 patients with carcinoma of the nasopharynx treated from June 1964 through July 1979. Patients with Stages I through IV, who were treated for cure, comprise the study group. Megavoltage teletherapy treatment was delivered to the primary bilateral neck and supraclavicular areas. Dose to the nasopharynx ranged from less than 55 Gy to 70 Gy with several patients receiving higher doses. The observed 5 year NED survival rates were: Stage I, 75%; Stage II, 20%; Stage III, 42%; and Stage IV, 31%. During a 5-year minimum follow-up for all patients, 46 or 60.5% failed. Ninety-five percent of these failures occurred within 3 years; only two patients relapsed after more than 5 years.     

Curative surgery for adenocarcinoma of the pancreas/ampulla of Vater: the role of adjuvant pre or postoperative radiation therapy

From 1972 to 1981, 7 patients received adjuvant external beam radiation therapy before (5 patients) or after (2 patients) a curative Whipple operation for adenocarcinoma of the pancreas (5 patients) or ampulla of Vater (2 patients). Via supervoltage, 4000-4520 rad/20-25 fractions/4-5 1/2 weeks were delivered to the tumor bed and draining lymphatics. Two patients (of 4 at risk) are 5-year survivors; 3 are alive-without-disease at 1-2 years follow-up. Distant metastases have developed to date in 3 patients. A marginal miss occurred at the edge of radiation portal in 1 patient, and no true in-field failures have occurred. Complications occurred in 2 patients. Because of the high incidence of local-regional failures after curative surgery alone for adenocarcinoma of the pancreas or ampulla of Vater, the above results suggest that adjuvant radiation therapy should be considered as a modality to potentially improve treatment results.     

瓦里安CLINAC-Ⅸ型直线加速器联锁UDRS原理及不常见故障处理

UDRS联锁是瓦里安高能直线加速器的常见剂量联锁,由于系统的复杂性,该故障的出现几乎一直困扰很多工程师,需要找出解决的方法。本文详细分析了瓦里安CLinac-Ⅸ型直线加速器电路构成和UDRS联锁发生的原理,总结出UDRS联锁故障的检修方法。通过对少见故障处理方法的分享,为解决类似机型出现UDRS故障时提供方法参考。     

Outcome of childhood acute promyelocytic leukemia with all-trans-retinoic acid and chemotherapy.

PURPOSE: To determine the results of treatment combining all-trans-retinoic acid (ATRA) and chemotherapy (CT) in childhood acute promyelocytic leukemia (APL). PATIENTS AND METHODS: Children (< 18 years) with newly diagnosed APL were included in the APL93 trial, treated by ATRA followed or combined with daunorubicin-cytarabine, and then randomly assigned between no maintenance, intermittent ATRA, continuous CT, or both. RESULTS: Of the 576 patients included in APL93 trial, 31 (5%) were children, including 22 girls (71%) and nine boys (29%). Thirty of the children (97%) obtained complete remission (CR). ATRA syndrome occurred in four children (13%), who all achieved CR, and headaches occurred in 12 children (39%), with signs of pseudotumor cerebri in five children (16%). Seven patients (23%) relapsed. None of the eight patients who received both ATRA and CT for maintenance relapsed. All relapsing patients achieved a second CR. Twenty-two patients remained in first CR after 43+ to 96+ months, six remained in second CR after 17+ to 66+ months, and three patients had died. The 5-year event-free survival (EFS), relapse, and overall survival rates were 71%, 27%, and 90%, respectively. No difference between adults and children included in the APL93 trial was seen for CR rate, 5-year relapse rate, EFS, and overall survival, but significantly better survival was seen in children after adjustment on WBC counts (P =.02) and incidence of microgranular M3 variant (P =.04). CONCLUSION: ATRA combined with CT for induction and also probably for maintenance provides as favorable results in children with APL as in adults and currently constitutes the reference first-line treatment in both age groups.     

Long-term survivors in metastatic non-small-cell lung cancer: an Eastern Cooperative Oncology Group Study

Between December 1979 and June 1983 the Eastern Cooperative Oncology Group (ECOG) treated 893 good-performance status patients with metastatic non-small-cell lung cancer (NSCLC) on one of seven phase III combination chemotherapies. The overall median survival was 23.5 weeks with no significant differences between treatments. One hundred sixty-eight patients (19%) survived greater than 1 year and 36 (4%) for greater than 2 years. The etoposide-platinum combination had the highest proportion of 1-year survivors (25%). Mitomycin-vinblastine-platinum (MVP), which had demonstrated the highest response rate, had significantly fewer 1-year survivors (12%) than any other regimen (P = .003). Analysis of pretreatment characteristics that distinguished patients who survived greater than 1 year from those who did not demonstrated that an initial performance status of 0, no bone metastases, female sex, no subcutaneous metastases, non-large-cell histology, less than 5% prior weight loss, no symptoms of shoulder or arm pain, and no liver metastases were predictors of longer survival. Of particular interest was the finding that response duration was significantly longer (P = .002) for those patients who experienced a longer time to best response. In addition, patients who survived greater than 1 year experienced greater degrees of nonlethal toxicity, in particular, gastrointestinal and hematologic, than patients who did not survive 1 year, (P = .006). A detailed chart review of 32 2-year survivors and 32 matched controls demonstrated that maintenance or improvement of performance status and maintenance of serum albumin levels at 3 months from the initiation of treatment were both important predictors of longer survival.     

Optimized outcome prediction in breast cancer by combining the 70-gene signature with clinical risk prediction algorithms

Clinical guidelines for breast cancer treatment differ in their selection of patients at a high risk of recurrence who are eligible to receive adjuvant systemic treatment (AST). The 70-gene signature is a molecular tool to better guide AST decisions. The aim of this study was to evaluate whether adding the 70-gene signature to clinical risk prediction algorithms can optimize outcome prediction and consequently treatment decisions in early stage, node-negative breast cancer patients. A 70-gene signature was available for 427 patients participating in the RASTER study (cT1-3N0M0). Median follow-up was 61.6 months. Based on 5-year distant-recurrence free interval (DRFI) probabilities survival areas under the curve (AUC) were calculated and compared for risk estimations based on the six clinical risk prediction algorithms: Adjuvant! Online (AOL), Nottingham Prognostic Index (NPI), St. Gallen (2003), the Dutch National guidelines (CBO 2004 and NABON 2012), and PREDICT plus. Also, survival AUC were calculated after adding the 70-gene signature to these clinical risk estimations. Systemically untreated patients with a high clinical risk estimation but a low risk 70-gene signature had an excellent 5-year DRFI varying between 97.1 and 100 %, depending on the clinical risk prediction algorithms used in the comparison. The best risk estimation was obtained in this cohort by adding the 70-gene signature to CBO 2012 (AUC: 0.644) and PREDICT (AUC: 0.662). Clinical risk estimations by all clinical algorithms improved by adding the 70-gene signature. Patients with a low risk 70-gene signature have an excellent survival, independent of their clinical risk estimation. Adding the 70-gene signature to clinical risk prediction algorithms improves risk estimations and therefore might improve the identification of early stage node-negative breast cancer patients for whom AST has limited value. In this cohort, the PREDICT plus tool in combination with the 70-gene signature provided the best risk prediction.     

放疗在早期鼻型NK/T细胞淋巴瘤综合治疗中作用及预后分析

目的 探讨放疗在ⅠE～ⅡE期结外鼻型NK/T细胞淋巴瘤综合治疗中的作用及其预后因素.方法 回顾分析1990-2006年收治的177例患者,其中单纯化疗37例(中位4周期),化疗(中位3周期)+放疗(中位剂量52 Gy)128例,单纯放疗(中位剂量58 Gy)6例,放疗(中位剂量54 Gy)+化疗(中位5周期)6例.结果 首程化疗后有效(CR+PR)率为65.3%,接受放疗后的为92.8%(x2=28.63,P<0.01).接受放疗的局部控制率(80.9%)优于单纯化疗者(50.0%;x2=14.39,P<0.01);5年总生存率分别为53.4%和18.3%(x2=23.38,P<0.01),无进展生存率分别为45.0%和10.9%(x2=23.46,P<0.01).首程化疗后有效与无效(SD+PD)者接受放疗的局部控制率、5年总生存率均明显优于单纯化疗者[83.5%与76.2%优于50.0%(x2=14.13,P<0.01;x2=5.78,P<0.01)、56.2%与48.6%优于18.3%(x2=28.87,P<0.05;x2=4.80,P<0.05)].结论 放疗比化疗能显著提高早期结外鼻型NK/T细胞淋巴瘤的有效率、局部控制率和生存率,对化疗后局部肿瘤无效者也有显著疗效.根治性放疗应成为早期鼻型NK/T细胞淋巴瘤的首选治疗手段.     

Prognostic Nomogram for Overall Survival in Early Stage Extranodal Natural Killer/T Cell Lymphoma Treated With High-Dose Radiotherapy

BackgroundThe aim of this study was to develop a prognostic nomogram for early stage extranodal natural killer/T-cell lymphoma, nasal type (ENKL) treated with high-dose radiotherapy (RT).Patients and MethodsA total of 81 patients at 2 cancer centers with stage I to IIE ENKL who received chemotherapy (CT) and high-dose RT were retrospectively analyzed. The development of the nomogram was on the basis of the Cox proportional hazards model. We implemented the concordance index (C-index) and performed a calibration curve to determine its predictive and discriminatory capacity and compared our nomogram with the International Prognostic Index (IPI) and the Korean Prognostic Index (KPI).ResultsThe nomogram included 4 important variables and used a multivariate analysis: lactate dehydrogenase, primary tumor invasion, tumor response, and CT regimen. The 5-year OS rate and progression-free survival were 64.7% and 57.5%, respectively for the entire group. The C-index of the nomogram for overall survival (OS) prediction was 0.87, and it was superior to the predictive power of the IPI and KPI. The calibration curve showed that the nomogram accurately predicted the 5-year OS.ConclusionThe proposed nomogram could provide an individualized risk estimate of the OS for early stage ENKL treated with CT and high-dose RT.