Objective:To explore the feasibility of autosegmentation and reconstruction of interstitial needles in intracavitary / interstitial brachytherapy (IC-ISBT) for cervical cancer based on deep learning.Methods:The data of 180 treatment plans from 98 patients who received IC-ISBT were retrospectively collected and divided into the training, validation, and testing sets in a 16:1:1 ratio. Masks of needles were created using the dwell positions of radiation sources, and a 3D U-Net model was trained. The performance of the model was evaluated using the Dice similarity coefficient (DSC). Absolute and relative accuracy rates were used to assess the results of this method, and the position bias was used to evaluate the precision of predictions in the transversal plan of CT scans. Wilcoxon rank-sum test was performed to evaluate the reconstruction efficiency by comparing the time required for automated versus manual reconstruction.Results:DSC of the model was 0.93 ± 0.02. The absolute and relative accuracy rates were 0.44 ± 0.09 and 0.95 ± 0.03, respectively. The distance deviation on the CT horizontal plane was (0.58 ± 0.54) mm. The average time of autosegmentation and reconstruction was (6.2 ± 0.4) s, leading to a significant reduction in time consumption compared with manual construction ( P<0.001). Conclusions:Based on deep learning, using the dwell positions of radiation sources for data annotation, combined with post-processing algorithms, accurate automated segmentation and digital reconstruction of needles in IC-ISBT three-dimensional CT images can be achieved, significantly improving reconstruction efficiency.
Objective:To analyze the treatment efficacy, safety and dose parameters of optimized hippocampus-avoidance prophylactic cranial irradiation (HA-PCI) in limited-stage small cell lung cancer (LS-SCLC) and explore the corresponding dosimetric parameters under the condition of narrowing the hippocampus avoidance region as hippocampus region plus 2 mm in three dimensions.Methods:Clinical data of patients with LS-SCLC receiving HA-PCI (hippocampus avoidance region defined as hippocampus region plus 2 mm in three dimensions) in Cancer Hospital Chinese Academy of Medical Sciences from August 2014 to June 2020 were retrospectively analyzed. Dose parameters of HA-PCI and adverse events were analyzed using descriptive statistics analysis. Changes of neurocognitive function, such as mini-mental state examination (MMSE) and Hopkins verbal learning test-revised (HVLT-R) scores, were evaluated by analysis of variance and Kruskal-Wallis H test. Overall survival (OS), progression-free survival (PFS) and intracranial PFS (iPFS) were calculated using Kaplan-Meier method. The cumulative incidence of local-regional recurrence (LRR), extracranial distant metastases (EDM), and locoregional recurrence (LR) were investigated under competing risk analysis. Results:A total of 112 patients were included, the median follow-up time was 50 months (95% CI: 45.61-54.38). The median volume of hippocampus was 4.85 ml (range: 2.65-8.34 ml), with the average dose ≤9 Gy in 106 patients (94.6%), ≤8 Gy in 92 patients (82.1%). The median volume of hippocampus avoidance area was 15.00 ml (range: 8.61-28.06 ml), with the average dose ≤12 Gy in 109 patients (97.3%), ≤10 Gy in 101 patients (90.2%). The 2-year cumulative LRR, EDM, LR rates were 16.9%, 23.2% and 28.5%, respectively. The 5-year cumulative LRR, EDM, LR rates were 23.2%, 26.9% and 33.3%, respectively. The 2-year iPFS, PFS and OS rates were 66.1% (95% CI: 57.9%-75.4%), 53.6% (95% CI: 45.1%-63.7%) and 80.4% (95% CI: 73.3%-88.1%), respectively. The most common grade I-Ⅱ adverse events were nausea (33.9%) and dizziness (31.3%), and only 1 patient developed grade Ⅲ nausea and dizziness. MMSE ( n=57) and HVLT-R tests ( n=56) showed no significant decline. Conclusions:Optimized HA-PCI can achieve similar dose limitation with favorable efficacy and light toxicity. No significant decline is observed in short-term neurocognitive function in evaluable patients.
Abstract Purpose We aimed to investigate the efficacy and side effects of concurrent chemoradiotherapy, with or without nimotuzumab, for the treatment of locally advanced nasopharyngeal carcinoma after neoadjuvant chemotherapy. Methods This study retrospectively enrolled 109 patients with NPC from our hospital from July 2019 to May 2021.All patients were treated with docetaxel, cisplatin, and fluorouracil(TPF) neoadjuvant chemotherapy for 2 cycles, and concurrent chemoradiotherapy was performed 2 weeks after chemotherapy. According to whether nimotuzumab was added in concurrent chemoradiotherapy, they were divided into the nimotuzumab group and the control group, with 52 cases in the nimotuzumab group and 57 cases in the control group.The efficacy and adverse reactions of the two groups were retrospectively analyzed. Results The objective remission and complete remission rates in the nimotuzumab and control groups were 100% vs 98.2% (p = 1.000), and 92.3% vs 78.9% (p = 0.049), respectively. The 3-year distant metastasis-free survival of the nimotuzumab and control groups was 91.6% and 77.3% (p = 0.047), respectively.The 3-year progression-free survival, locoregional relapse-free survival, and overall survival of the nimotuzumab and control groups were 87.6% vs 75.5% (p = 0.110), 90.5% vs 86.9% (p = 0.566), and 94.5% vs 87.1% (p = 0.295), respectively. In the nimotuzumab group, subgroup analysis showed that patients aged < 60 years (hazard ratio [HR] = 0.350, 95% confidence interval [CI]: 0.131–0.934, p = 0.036) and those with a neutrophil-to-lymphocyte ratio (neutrophil/lymphocyte ratio) ≤ 4 (HR = 0.365, 95% CI: 0.144–0.923, p = 0.033) achieved a better result. Additionally, multivariate analysis demonstrated that neutrophil/lymphocyte ratio was an independent risk factor for disease progression (HR = 7.485, p = 0.012) and distant metastasis (HR = 17.540, p = 0.009).No grade 4 adverse reactions were observed in either group. Grade 3 oral mucosal reactions, as well as pharyngeal and esophageal reactions were slightly higher in the nimotuzumab group than in the control group, but the difference was not statistically significant. No significant differences were observed in the incidence of adverse reactions such as leukopenia, HB reduction, thrombocytopenia between the two groups (P > 0.05). Conclusion The concurrent chemoradiotherapy plus nimotuzumab after neoadjuvant chemotherapy for locally advanced nasopharyngeal carcinoma achieved a higher complete remission rate and significantly improved distant metastasis-free survival compared with concurrent chemoradiotherapy alone. Additionally, an increasing trend was observed in progression-free survival, and the incidence of side effects was similar in both groups.
近年来,互联网的技术迅速发展,宽带普及度逐渐提高,互联网在医疗领域的应用也在不断开展.为推进分级诊疗制度建设和医疗联合体(以下简称医联体)建设,构建优质高效的医疗卫生服务体系,逐步实现城市医联体网格化布局管理,国家发布了相关指导意见和工作方案,因此,在区域内医联体中开展放疗工作是趋势.为使区域内医联体中的医疗机构能够达到相适应的放疗水平与诊疗质量,开展远程质量保证是构建优质放疗医联体的重要组成部分.因此,国家癌症中心/国家肿瘤质控中心组织专家制定了本指南.指南涉及远程放射治疗平台架构、平台性能、人员和管理制度等内容.
目的 探讨术后单纯放疗和术后放化疗治疗软组织肉瘤(STS)的临床结局和不良反应方面的差异,以及影响STS患者预后的因素.方法 回顾性分析浙江省肿瘤医院2012年5月至2019年5月首诊确诊为原发性STS的患者,术后接受辅助放疗,伴或不伴术后化疗.共入组100例患者,将其分为术后单纯放疗组(52例)与术后放化疗组(48例),中位随访时间为65个月(24～124个月).统计两组患者的无局部复发生存(LRFS)期、无远处转移生存(DMFS)期、总生存(OS)期和治疗相关不良反应.采用Kaplan-Meier法计算生存率,log-rank检验进行单因素分析,Cox模型行多因素分析.结果 多因素分析显示,肿瘤最长径是肿瘤局部复发的独立预测因素(HR=4.80,95％CI=1.16～19.85,P=0.031),同时也是远处转移(HR=4.67,95％CI 为 1.53～14.26,P=0.007)和患者 OS 期(HR=4.10,95％CI为1.35～12.48,P=0.013)的独立预测因素.另外,接受放化疗患者的骨髓抑制程度显著高于单纯放疗患者(P＜0.001).结论 在患者样本量有限的情况下,与单纯放疗相比,放化疗在远处转移或生存率方面没有改善,不良反应增加,但总体耐受性尚好.有必要进行大规模人群的前瞻性随机研究,并对组织学亚型进行亚组分析,以确保获得更有参考价值的结果.
目的 探究一种以剂量体积直方图参数为优化变量,使用多目标优化算法的剂量优化配准算法,以提高锥形线束CT(CBCT)图像验证匹配的准确性.方法 回顾性分析哈尔滨医科大学附属肿瘤医院2022年收治的6例肺癌和5例宫颈癌患者的28套CBCT图像.以骨性配准的结果为剂量配准算法的初始点,在其周围的三维空间内计算各位移点的通量加权灰度均方差,并使用无监督k均值聚类方法筛选候选位移点.使用有限尺寸笔形束算法计算各候选位移点的三维剂量分布,并提取直方图剂量指标作为多目标优化算法的优化变量.使用多目标遗传算法求解Pareto最优解集后,根据预设的目标权重方案求解最优唯一解.结果 采用剂量优化配准后,肺癌病例计划靶区(PTV)的D90％、D95％、D98％、Dmean、适形度(CI)指标,较常规配准策略分别平均提高0.23 Gy、0.49 Gy、1.05 Gy、0.15 Gy、0.03,危及器官的剂量差异无统计学意义.宫颈癌病例PTV的D90、D95％、D98％、Dmean、CI指标,较常规配准策略分别平均提高0.72 Gy、1.15 Gy、2.53 Gy、0.24 Gy、0.05,部分危及器官评估指标降低1.06～1.81 Gy.结论 剂量优化配准方法可有效提高靶区覆盖度并降低危及器官剂量,减少刚性配准算法的残余误差,可作为自适应放疗技术的流程之一.
目的 体外构建鼻咽癌放射抵抗细胞株,为进一步研究鼻咽癌放射抵抗的分子机制提供实验基础.方法 采用剂量梯度法照射诱导建立放射抵抗细胞模型5-8F-IR.光学显微镜观察细胞形态变化;克隆形成实验检测细胞克隆形成能力;CCK-8实验检测细胞活力;5-溴-2-脱氧尿嘧啶(EdU)实验检测细胞增殖能力;彗星实验及免疫荧光实验检测细胞DNA损伤修复能力;流式细胞术检测细胞凋亡水平及周期分布;蛋白质印迹法检测DNA损伤相关蛋白γH2AX及凋亡相关蛋白胱天蛋白酶3(Caspase-3)的蛋白表达水平.结果 剂量梯度照射后5-8F-IR细胞形态较亲本5-8F细胞形态变长,5-8F-IR细胞的克隆形成能力(P＜0.01)、细胞活力(P＜0.001)、细胞增殖能力(P＜0.05)、DNA损伤修复能力(P＜0.05)均明显优于亲本细胞5-8F,照射后5-8F-IR细胞的凋亡率明显低于5-8F细胞(P＜0.01),未接受照射时5-8F-IR细胞处于S期的百分比较5-8F细胞升高,接受照射后5-8F-IR细胞出现明显的G2/M期阻滞(P＜0.01),同时照射后5-8F-IR细胞的DNA损伤相关蛋白γH2AX(P＜0.001)及凋亡相关蛋白Caspase-3(P＜0.05)蛋白表达水平明显低于5-8F细胞.结论 人鼻咽癌细胞株5-8F经剂量梯度法照射诱导建立的5-8F-IR细胞具有放射抗拒性并显示出与亲代5-8F细胞不同的生物学特性,为探索鼻咽癌放射抵抗机制提供了研究工具.
手术联合术后立体定向放射外科(SRS)治疗提高了可切除脑转移瘤患者的总生存,成为可切除脑转移瘤重要的治疗模式之一.与之相比,术前SRS具有减少术中有活性的肿瘤细胞播散到治疗腔、降低照射剂量、减少对周围正常脑组织的照射、保证了围术期治疗的顺利进行等理论优势,而且被近年的临床研究证实其可降低局部复发率、放射性脑坏死和软脑膜转移的发生率.另外,其适应证、剂量分割及联合药物治疗等实施细节有待更多前瞻性研究证实.本文对该治疗策略的研究进展进行综述.
目的 探索脱氧胞苷激酶(dCK)在放射诱导三阴性乳腺癌(TNBC)铁死亡中的调控作用.方法 用人乳腺癌细胞株MDA-MB-231构建dCK基因沉默和不同磷酸化表型的细胞模型,并给予铁死亡诱导剂Erastin和/或铁死亡抑制剂Fer-1联合或不联合X线放射处理.通过MTT法检测细胞活性,活性氧荧光探针(DCFH-DA)检测活性氧水平,通过蛋白质印迹法(Western blot)检测dCK、转铁蛋白、转铁蛋白受体(TfR1)、铁转运蛋白(FPN)、铁蛋白重链1(FTH1)的蛋白表达量.采用SPSS 17.0和Origin 2021软件对数据进行分析.计量资料符合正态分布,以(-x)±s表示.两组之间的比较使用Student t检验进行,而三组及以上的比较使用单向方差分析.结果 在MDA-MB-231细胞中,放射诱导细胞死亡,铁死亡诱导剂Erastin显著促进放射诱导的细胞死亡,铁死亡抑制剂Fer-1能够逆转放射诱导的细胞死亡.与对照细胞相比,dCK基因沉默细胞的放射诱导的细胞死亡增加,活性氧水平降低,Erastin联合放射诱导的细胞死亡减少,活性氧水平减弱,Fer-1可使放射诱导的细胞死亡程度降低,并且Fer-1无法抑制放射对活性氧的诱导作用.与对照细胞相比,在野生型dCK(dCK-WT)或dCK过磷酸化(dCK-S74E)的dCK基因沉默细胞中,放射诱导细胞死亡减少,活性氧水平降低,FTH1表达量降低,加入放射进一步降低FTH1的表达水平.此外,在这些细胞中,Erastin促进放射诱导的细胞死亡和活性氧水平增加,Fer-1对放射诱导活性氧和细胞死亡的逆转程度明显增强.结论 dCK磷酸化促进了放射诱导TNBC细胞的铁死亡,靶向dCK可能是一种克服TNBC治疗中辐射抗性的新治疗方式.
目的 分析初诊非小细胞肺癌(NSCLC)脑转移(BM)患者一线免疫检查点抑制剂联合放疗的疗效.方法 回顾性分析河南省肿瘤医院2017年12月至2021年6月首诊确诊为NSCLC BM的患者66例的资料.患者一线应用程序性死亡受体1(PD-1)/程序性死亡受体配体1(PD-L1)抑制剂联合或不联合脑部放疗.采用Kaplan-Meier法计算生存率,log-rank检验进行单因素分析,Cox模型行多因素分析.结果 全组患者中位随访13(范围4～49)个月,中位生存期24个月,1、2、3年总生存(OS)率分别是75.8％、44.7％、33.6％.1年脑内无进展生存(iPFS)分别是68.8％.放疗免疫组和免疫组脑内客观应答率icORR分别是71.4％、52.6％,脑内疾病控制率(icDCR)分别是96.4％、89.4％(P=0.122、0.291).单因素分析显示:预后评估评分(GPA)高的患者1、2、3年OS率分别是100％、85.7％、35.7％,GPA低的患者则分别是68.3％、33.6％、33.6％(P=0.115);放疗免疫组患者1、2、3年OS率分别是 88.7％、58.7％、44.0％,免疫组患者则分别是 63.6％、20.9％、20.9％(P=0.022);BM 有症状患者 1、2、3年OS率分别是60.9％、33.8％、33.8％,无症状患者则分别是84.6％、53.1％、39.8％(P=0.021).多因素分析显示,无BM症状、脑部放疗是OS率的独立影响因素(P=0.038、0.037).结论 一线免疫检查点抑制剂联合脑部放疗治疗NSCLC BM可以提高患者的生存率.建议BM患者尽早开始放疗,尤其是有脑部症状的患者.
目的 探讨联合自动针道优化及3D打印技术研制腔道适形插植导板的方法,评价该技术对剂量及疗效的提升与改进作用.方法 回顾性分析广西医科大学附属肿瘤医院2020年9月-2022年9月收治的晚期宫颈癌患者15例.患者存在阴道残端复发、肿瘤偏心性生长、子宫锥切术后等特征,CT扫描获取三维影像,完成针道优化后,通过放射治疗3D打印系统软件(PMT3D)对预计划数据进行解析,抓取驻留点的坐标而建立插植通道,并基于患者生理结构特征生成腔道适形插植导板的尺寸及形状,3D打印后实施插植放疗.评估适形指数等剂量学参数,对治疗前、后的肿瘤体积变化进行配对t检验.结果 15例患者采用腔道适形插植导板进行治疗,其治疗计划适形指数CI为0.74± 0.26,放射治疗总剂量高危临床靶区(HR-CTV)D90％(EQD2,α/p=10)为(85.5±6.8)Gy,膀胱、直肠、小肠、结肠的 D2 cm3(EQD2,α/β=3)分别为(72.2±4.2)、(65.8±6.1)、(65.2±4.4)、(69.8±3.7)Gy,符合临床要求.治疗后,肿瘤体积缩小明显.适形插植导板与阴道腔体贴合度较好,少体积的空气间隙并未影响剂量.非平行入针可以提高腔体空间及插植针的利用率.结论 基于自动针道优化及3D打印技术的适形腔道插植导板研制方法兼顾效率和效果,符合现代精准放疗的要求,具有临床实用价值.
中低位局部晚期(Ⅱ-Ⅲ期)直肠癌侧方淋巴结转移(LLN)率高、预后差.目前,关于LLN治疗方案东西方争议较大:以日本为代表的东方学者认为LLN是区域性疾病,当肿瘤位于腹膜返折下且侵及肌层时无论有无LLN,均推荐预防性盆腔侧方淋巴结清扫(LPLND);而欧美学者认为LLN是全身性疾病,推荐新辅助放化疗(nCRT)联合全直肠系膜切除(TME).然而,近年来有研究发现,nCRT或LPLND均不能显著降低LLN患者的局部复发率,而nCRT联合LPLND治疗的预后更好,也有研究发现增加转移淋巴结放疗剂量可以提高局部控制率.本文综述了该类人群的治疗现状,以期为临床治疗提供依据.
乳腺癌术后放疗提高了患者的总体生存率,但心脏因其解剖位置在放疗期间不可避免地受到照射,存在放射性心脏损伤的风险.心脏被视为辐射敏感的"高危器官",其受照剂量应尽可能低.既往研究主要评估心脏整体受照剂量对远期心血管事件的影响,而最新研究发现,平均心脏剂量无法准确反映心脏受照程度.近年来放疗后心脏亚临床损伤日益受到重视,旨在早期识别隐匿性心脏损伤.此外,特定心脏亚结构剂量与心律失常的关系尚不清楚.本文重点阐述了平均心脏剂量预测放射性心脏损伤的局限性、早期识别心脏损伤的指标及乳腺癌放射性心脏损伤的影响因素,并关注不同心脏亚结构放射损伤与心律失常的关系,以期实现乳腺癌患者心脏风险精细化管理,降低乳腺癌患者的非癌死亡率.
近距离治疗(BT)技术是使放射源进入或贴近治疗部位,利用近源处剂量极高随着距离的增加剂量快速跌落,实现肿瘤局部治疗的一种技术.近距离治疗是宫颈癌放射治疗的重要组成部分.近距离治疗过程复杂且需要一系列步骤,可能导致患者在施源器置入状态下等待时间过长而产生施源器相对于肿瘤的位置以及器官充盈状态的改变.近些年,人工智能(AI)在医学领域有了极大的发展,机器学习以及神经网络等模型广泛应用于近距离治疗的各个环节,如施源器置入、图像采集、靶区和危及器官勾画、施源器重建、剂量优化和治疗实施等步骤.在缩短流程总体耗时、提高操作同质性及提高治疗精度等方面作用显著.本文总结了近年来AI在宫颈癌近距离治疗中的应用、发展前景以及面临的挑战,对AI在临床近距离治疗中的应用进行展望并提供新的思路.
2023年11月22日,中国科学院公布2023年增选当选院士名单,中山大学肿瘤防治中心常务副主任、副院长,放射肿瘤学/鼻咽癌首席专家马骏教授增选为中国科学院生命科学和医学学部院士. 马骏院士是我国具有重要国际影响力的鼻咽癌诊治专家和研究型医生,38年来他聚焦高发于我国的鼻咽癌(占全球病例47％),发现了"先GP全身化疗,后局部放射治疗"的"增效"理论并制定了新策略,大幅提高了晚期鼻咽癌患者的生存率;提出"低危患者,可降低治疗强度"的"减毒"理论,创立了"豁免化疗及缩小放疗范围"的治疗新策略,改善了患者生活质量.
目的 探讨局部中晚期鼻咽癌诱导化疗后放疗开始时间与预后之间的关系及其最佳开始时间.方法 回顾性分析2017-2020年中山大学附属第五医院收治的局部中晚期鼻咽癌患者101例,所有患者接受了至少2个疗程的诱导化疗后行根治性放疗.放疗等待时间定义为诱导化疗结束至第1次放疗开始所等待的时间.通过Cox回归分析法明确放疗等待时间以及其他因素(年龄、性别、分期等)与预后指标包括无进展生存(PFS)、无局部复发生存(LRFS)、无远处转移生存(DMFS)之间的关系.以中位放疗等待时间3周作为界限,所有患者被分成≤3周和＞3周两组,通过Kaplan-Meier生存分析法比较两组的PFS、LRFS、DMFS.P＜0.05为差异有统计学意义.结果 Cox回归分析发现,放疗等待时间与PFS、LRFS、DMFS相关(P＜0.05).生存分析发现,放疗等待时间≤3周组的PFS、LRFS、DMFS明显优于＞3周组(P＜0.05).同样在T3期、N2期和治疗前EB病毒DNA复制水平升高的情况下,放疗等待时间≤周组的PFS、LRFS、DMFS均优于＞3周组(P＜0.05).结论 放疗等待时间是影响局部中晚期鼻咽部患者预后的因素之一.诱导化疗后放疗开始时间越早预后越好,最好控制在3周以内.
目的 观察食管鳞癌根治术后纵隔淋巴结复发患者采用同步加量调强放疗(SIB-IMRT)的长期疗效和晚期不良反应.方法 前瞻性研究2019年6月至2021年12月期间就诊于上海瑞金医院放疗科的食管鳞癌术后纵隔淋巴结复发(≤5枚)的患者共20例,其中Ⅰ、Ⅱ期研究各纳入10例患者,4、3、13例患者分别接受58.8 Gy分28次、64.4Gy分28次、70Gy分28次3个剂量等级的SIB-IMRT.使用Kaplan-Meier法计算入组患者的总生存(OS)率、局部控制率(LCR)及无进展生存(PFS)率,并对不良反应进行分析.结果 纵隔淋巴结最常见的复发部位为2R和4L,分别占35％和25％.中位随访32个月,复发后接受挽救性SIB-IMRT者的1、2、3年OS率分别为100％、88％、78％,1、2、3年PFS率分别为85％、78％、78％.最常见的血液学不良反应是白细胞减少和贫血.最常见的非血液学不良反应是食管炎,但未发现3级及以上食管炎、肺炎和心脏不良反应.随访中3例接受58.8 Gy分28次剂量的患者在治疗后2年出现肿瘤远处转移死亡,另外1例接受70 Gy分28次剂量的患者在治疗后16个月出现肿瘤远处转移死亡.结论 对于食管鳞癌根治术后纵隔淋巴结复发患者,采用SIB-IMRT挽救性同步放化疗方案具有较高的有效性和安全性.
近期中华医学会杂志社学术期刊出版平台在后台监测到部分用户使用虚假的手机号和Email地址注册账号,这些账号的投稿IP地址与作者所在单位所属行政区域严重偏离,涉嫌第三方机构代写代投.此类行为属于严重的学术不端,我们已将排查到的稿件信息通报各编辑部,杂志社新媒体部也将对此类账号做封禁处理,相关稿件一律做退稿处理.
近期中华医学会杂志社学术期刊出版平台在后台监测到部分用户使用虚假的手机号和Email地址注册账号,这些账号的投稿IP地址与作者所在单位所属行政区域严重偏离,涉嫌第三方机构代写代投.此类行为属于严重的学术不端,我们已将排查到的稿件信息通报各编辑部,杂志社新媒体部也将对此类账号做封禁处理,相关稿件一律做退稿处理.
Objective:To evaluate the value of enhanced CT radiomics feature model for predicting 5-year overall survival (OS) of esophageal squamous cell carcinoma patients after radiotherapy.Methods:Clinical data of 218 patients with esophageal squamous cell carcinoma treated with radical chemoradiotherapy in the Fourth Hospital of Hebei Medical University from July 2016 to December 2017 were retrospectively analyzed. Patients were randomly divided into the training group ( n=153) or a validation group ( n=65) at a 7 vs. 3 ratio. Enhanced CT radiomics features were extracted. The data in the training group was used to construct the prediction model, and the data in the validation group were utilized to validate the efficiency of this model for predicting the 5-year OS of patients. The predictive performance of this model was assessed by the receiver operating characteristic (ROC) curve, consistency index (C-index), and decision curve analysis (DCA). Results:The 1-, 3-, 5-year OS rates were 67.0%, 33.4%, 24.9%. Five radiomic features were selected from extracted features in the training group to construct the radiomic signature (RS) for predicting 5-year OS. The area under the ROC curve (AUC) was 0.760 in the training group and 0.707 in the validation group, and the C-index was 0.680 and 0.684, respectively. The radiomics nomogram, which incorporated the RS with clinical risk factors, were established to predict the 5-year OS of esophageal squamous cell carcinoma patients after radiotherapy. The AUC was 0.782 in the training group and 0.751 in the validation group, and the C-index was 0.708 and 0.688, respectively. According to the optimal cutoff of the model, all patients were divided into the high risk and low risk groups. The 1-, 3-, 5-year OS rates were 86.5%, 65.4%, 28.9% in the low risk group, and 58.4%, 17.8%, 5.9% in the high risk group, and the differences were statistically significant (all P<0.001). Similar conclusions were obtained in the validation group (all P<0.001). Conclusion:Enhanced CT radiomics features can be utilized to construct the prediction model for 5-year OS of esophageal squamous cell carcinoma patients after radiotherapy, which can be applied in clinical practice.
