WJCR World Journal of Cancer Research 2164-9049 Scientific Research Publishing 10.12677/WJCR.2024.141005 WJCR-79488 WJCR20240100000_79479594.pdf 医药卫生 基于深度学习的影像组学在预测结直肠癌基因状态中的应用进展 Application Advances in Deep Learning-Based Imaging Radiomics for Predicting Colorectal Cancer Gene Status 鹏飞 2 1 晓明 2 1 青海大学附属医院胃肠外科,青海 西宁 null 09 01 2024 14 01 27 34 © Copyright 2014 by authors and Scientific Research Publishing Inc. 2014 This work is licensed under the Creative Commons Attribution International License (CC BY). http://creativecommons.org/licenses/by/4.0/

结直肠癌(CRC)在中国的发病率和死亡率持续上升,多数病人在确诊时已属于中晚期。个性化的治疗策略和预后分析在很大程度上依赖于对CRC患者遗传特征的了解。近年来,影像组学已经成为一种有价值的工具,它通过图形分析和特征提取在肿瘤基因表达和影像表型之间建立了联系。随着机器深度学习的介入,这种非侵入性的技术可在术前预测肿瘤相关的基因型。目前,越来越多的研究致力于研究图像特征和CRC基因型之间的关系,从而为CRC基因型的诊断和预测提供更高的准确性。本综述旨在总结基于CRC的影像技术在预测结直肠癌遗传状态方面的临床应用、进展和目前的局限性。最终目的是加强临床医师对CRC成像技术的理解,提高CRC的诊断、预后和治疗方面的潜力。 The incidence and mortality of colorectal cancer (CRC) continue to rise in China, and most patients are in the middle to advanced stages at the time of diagnosis. Personalized treatment strategies and prognostic analyses rely heavily on the knowledge of genetic characteristics of CRC patients. In re-cent years, imaging genomics has emerged as a valuable tool that establishes a link between tumor gene expression and imaging phenotype through graphical analysis and feature extraction. With the intervention of deep machine learning, this non-invasive technique can predict tumor-associated genotypes preoperatively. Currently, more and more studies are devoted to investigating the relationship between image features and CRC genotypes, thus providing higher accuracy in diagnosis and prediction of CRC genotypes. The aim of this review is to summarize the clinical applications, advances, and current limitations of CRC-based imaging techniques in predicting the genetic status of colorectal cancer. The ultimate goal is to enhance clinicians’ understanding of CRC imaging techniques and to improve the potential of CRC in terms of diagnosis, prognosis, and treatment.

结直肠癌,影像组学,深度学习,基因状态,微卫星不稳定性, Colorectal Cancer Imaging Radiomics Deep Learning Gene Status Microsatellite Instability
摘要

结直肠癌(CRC)在中国的发病率和死亡率持续上升,多数病人在确诊时已属于中晚期。个性化的治疗策略和预后分析在很大程度上依赖于对CRC患者遗传特征的了解。近年来,影像组学已经成为一种有价值的工具,它通过图形分析和特征提取在肿瘤基因表达和影像表型之间建立了联系。随着机器深度学习的介入,这种非侵入性的技术可在术前预测肿瘤相关的基因型。目前,越来越多的研究致力于研究图像特征和CRC基因型之间的关系,从而为CRC基因型的诊断和预测提供更高的准确性。本综述旨在总结基于CRC的影像技术在预测结直肠癌遗传状态方面的临床应用、进展和目前的局限性。最终目的是加强临床医师对CRC成像技术的理解,提高CRC的诊断、预后和治疗方面的潜力。

关键词

结直肠癌,影像组学,深度学习,基因状态,微卫星不稳定性

Application Advances in Deep Learning-Based Imaging Radiomics for Predicting Colorectal Cancer Gene Status <sup> </sup>

Pengfei Li, Xiaoming Ma*

Department of Gastrointestinal Surgery, Affiliated Hospital of Qinghai University, Xining Qinghai

Received: Dec. 11th, 2023; accepted: Jan. 11th, 2024; published: Jan. 19th, 2024

ABSTRACT

The incidence and mortality of colorectal cancer (CRC) continue to rise in China, and most patients are in the middle to advanced stages at the time of diagnosis. Personalized treatment strategies and prognostic analyses rely heavily on the knowledge of genetic characteristics of CRC patients. In recent years, imaging genomics has emerged as a valuable tool that establishes a link between tumor gene expression and imaging phenotype through graphical analysis and feature extraction. With the intervention of deep machine learning, this non-invasive technique can predict tumor-associated genotypes preoperatively. Currently, more and more studies are devoted to investigating the relationship between image features and CRC genotypes, thus providing higher accuracy in diagnosis and prediction of CRC genotypes. The aim of this review is to summarize the clinical applications, advances, and current limitations of CRC-based imaging techniques in predicting the genetic status of colorectal cancer. The ultimate goal is to enhance clinicians’ understanding of CRC imaging techniques and to improve the potential of CRC in terms of diagnosis, prognosis, and treatment.

Keywords:Colorectal Cancer, Imaging Radiomics, Deep Learning, Gene Status, Microsatellite Instability

Copyright © 2024 by author(s) and beplay安卓登录

This work is licensed under the Creative Commons Attribution International License (CC BY 4.0).

http://creativecommons.org/licenses/by/4.0/

1. 引言

目前,中国结直肠癌(Colorectal Cancer, CRC)的发病率及死亡率仍处上升趋势,根据2020年中国癌症统计报告显示:在所有恶性肿瘤中,我国结直肠癌发病率、死亡率在全部恶性肿瘤中分别位居第2及第5位,且多数病人在确诊时已属于中晚期 [ 1 ] 。新辅助放化疗(Neoadjuvant Chemoradiotherapy, nCRT)联合全直肠系膜切除术(Total Mesorectal Excision, TME)是局部进展期直肠癌的标准治疗方法 [ 2 ] [ 3 ] 。然而,现有研究表明,结直肠癌患者对该治疗的敏感性呈现显著的个体差异 [ 4 ] [ 5 ] 。作为一种替代效果有限的常规治疗方案,靶向治疗在结直肠癌(CRC)患者中的应用要求对基因状态进行评估。传统的活检组织标本基因检测被视为金标准,然而,其存在高成本、脱氧核糖核酸(DNA)可用性差和周转时间长等问题。此外,检出的基因信息无法全面代表肿瘤的时空异质性 [ 6 ] 。影像组学作为一门新兴的跨学科研究领域,它结合了先进的计算机视觉技术和深度学习算法,以非侵入性的方式探索结直肠癌等恶性肿瘤的基因型与影像表型之间的相关性。该领域通过高效的图像特征自动提取方法,以及对这些特征进行深入学习和分析的计算模型,致力于解码肿瘤的遗传背景。此方法不仅为肿瘤的生物标志物发现提供了新途径,也为个性化治疗方案的制定和精准医疗的实施提供了有力支持,展现了其在医学影像分析和临床应用中的巨大潜力。

2. 影像组学

影像组学通过图像采集,应用大量的自动化数据特征化算法,将感兴趣区域(ROI)的影像数据转化为具有高分辨率、可挖掘的特征空间数据,最终转化为定量数据,用于描述ROI的特征 [ 7 ] 。近年来在肿瘤良恶性鉴别、术前分期、疗效评价等方面已显示出预测生存、肿瘤反应、副作用、病毒状态和基因组、蛋白质组等信息的潜力 [ 8 ] [ 9 ] 。

基于深度学习的影像组学

深度学习作为机器学习的一个子集,在包括影像组学在内的各个领域都显示出了显著的成效 [ 10 ] 。通过构建层叠的人工神经网络架构,自主地从数据中学习并萃取深层次特征,摒弃了对人工特征的依赖。该方法论采用逐层加工的策略,对输入数据进行连续的变换与高阶抽象,进而实现对数据复杂性的有效编码与解析 [ 11 ] 。在影像组学的背景下,深度学习算法被应用于分析影像数据,以提取高信息量的特征。这些算法包括卷积神经网络(Convolutional Neural Networks, CNN)、递归神经网络(Recurrent Neural Networks, RNN)和混合模型等。通过学习和提取影像数据中与遗传状态相关的特征,并结合基因组、蛋白质组或转录组数据,这些算法能够准确预测患者的遗传状态。这些模型的发展对于个体化治疗决策和病人分层提供了重要的临床意义。

预测结直肠癌(CRC)的基因分型在治疗决策和病人分层方面扮演重要的临床角色。大量研究显示,CRC的遗传变异与药物敏感性、治疗反应和预后密切相关。通过准确确定患者的基因分型,医生能够更精确地定制个体化的治疗方案,从而提高治疗效果并降低不必要的治疗。此外,基因分型还可用于对患者进行分层,以实现精准医学的目标,更好地满足患者的个体化需求。因此,基因分型在CRC的临床应用中具有显著的临床意义。

3. 影像组学预测CRC基因型 3.1. 预测患者KRAS基因状态

Kirsten rat sarcoma viral oncogene (KRAS)作为癌症中最著名的致癌基因之一,在癌症中具有最高的突变率,与一系列高致死率的癌症密切相关。KRAS位于12号染色体上,是一个原癌基因,编码KRAS蛋白。KRAS蛋白作为表皮生长因子受体(epidermal growth factor receptor, EGFR)功能信号的下游分子,与结直肠癌(CRC)的发生、增殖、迁移、扩散和血管生成密切相关 [ 12 ] [ 13 ] 。携带该突变的患者对于EGFR抑制剂呈现不敏感的特性。基因表达状态在很大程度上决定了对EGFR抗体治疗的敏感性,而KRAS在其中扮演着重要的角色,并且是一个关键的预后生物标志物。准确地识别KRAS突变状态可以为精确治疗的决策提供支持 [ 14 ] [ 15 ] 。

然而,传统的基因分型预测方法存在一些局限性。例如,分子分析通常需要对肿瘤样本进行基因测序,这涉及昂贵的实验费用和复杂的操作流程 [ 16 ] 。此外,目前的金标准方法,如组织活检,虽然能提供可靠的基因信息,但其侵入性和有限的可行性限制了其在临床实践中的广泛应用 [ 17 ] 。单一组织样本的基因分型可能无法完全代表整个肿瘤的遗传异质性。然而,基于深度学习的影像组学利用计算机视觉和深度学习算法,通过自动化和高通量的影像特征提取,为预测结直肠癌(CRC)的基因分型提供了一种新的非侵入性方法。深度学习模型能够从大规模影像数据中学习和提取与基因分型相关的特征,这种方法能够综合考虑肿瘤的时间和空间异质性,从而提供更全面的基因分型预测。

传统影像学技术在评估CRC中的KRAS突变时,主要通过视觉特征进行评估,如病变大小、形态、血管性、代谢特征和淋巴结转移等 [ 18 ] 。然而,不同研究对于CRC基因突变与传统成像特征之间的相关性存在差异,可能是由于主观因素、图像质量和扫描参数的影响所致 [ 19 ] [ 20 ] [ 21 ] [ 22 ] [ 23 ] 。同时,正电子发射断层扫描研究结果也存在争议,关于18F-FDG摄取与KRAS基因状态的关联性以及PET/CT定量指标的准确性在临床中的应用仍需进一步研究验证 [ 24 ] [ 25 ] [ 26 ] [ 27 ] 。

通过深度学习技术的应用,Song [ 28 ] 等学者基于36名结直肠癌(CRC)患者的T2磁共振成像数据,构建了一种名为多任务双流注意网络(MTDSAN)的深度学习模型。该模型在识别KRAS基因突变状态方面表现出较高的平均准确率(89.95 ± 1.23%)、平均敏感性(89.29 ± 1.79%)、平均特异度(90.53 ± 2.45%)以及平均曲线下面积(AUC, 95.73 ± 0.52%)。此外,研究团队还对病灶进行了分割,也取得了较高的准确率。Taguchi [ 29 ] 等研究人员则通过收集40例已验证的结直肠癌(CRC)患者数据,结合KRAS突变检测、增强CT和18F-FDG PET图像,探究了CT纹理参数和18F-FDG PET图像上的最大标准摄取值(SUVmax)对KRAS突变状态的预测性能。研究结果显示,基于CT纹理参数的机器学习模型的曲线下面积(AUC)为0.82,在预测KRAS突变状态方面优于18F-FDG PET图像上的SUVmax方法。Wu [ 30 ] 等学者则通过对279例结直肠癌(CRC)患者的门静脉期CT图像进行了分析,比较了传统放射组学特征和深度学习特征在预测方面的能力。研究结果表明,将传统放射组学特征与深度学习特征相结合可以显著提高预测性能。融合了传统放射组学特征和深度学习的放射组学特征的c指数及验证队列鉴别能力分别为0.815 (95% CI: 0.766~0.868)、0.832 (95% CI: 0.762~0.905),明显高于传统放射组学特征的c指数及验证队列鉴别能力分别为0.719 (95% CI: 0.658~0.776)、0.720 (95% CI: 0.625~0.813)。深度学习和机器学习技术在预测结直肠癌KRAS突变状态方面表现出潜力,并有可能超越传统影像学检查方法的能力。然而,仍需要进一步的研究来验证这些模型的有效性和临床适用性。

3.2. 预测患者MSI状态

微卫星(Microsatellites, MSI)是指细胞脱氧核糖核酸(DNA)中的短核苷酸重复序列。微卫星不稳定性(MSI)是指DNA序列中简单重复序列的碱基长度和(或)重复次数的增加或减少,导致遗传不稳定性的现象 [ 31 ] 。这种不稳定性是由于脱氧核糖核酸错配修复基因(如MLH1、MSH2、MSH6和PMS2等基因)的失活突变引起的,它是结直肠癌(CRC)发生和发展的重要致癌途径之一 [ 32 ] 。MSI不仅对患者的预后产生影响,还对选择靶向药物具有指导作用。特别是在晚期CRC患者中,免疫检查点抑制剂治疗在MSI阳性的情况下具有显著疗效。因此,对MSI状态进行准确评估对于指导精准治疗至关重要 [ 33 ] [ 34 ] [ 35 ] 。然而,对微卫星不稳定性(MSI)状态的评估主要依赖于结肠镜活检或手术切除标本进行免疫组化或聚合酶链反应。尽管术后获得的错配修复蛋白表达水平信息对预处理计划的影响有限,但这仍然是目前主要的评估方法 [ 36 ] 。由于活检样本的获取受到限制,可能无法充分反映肿瘤内的异质性,因此可能导致误判。此外,使用活检和手术进行MSI状态评估存在一定的假阴性结果的风险。考虑到这些方法是有创性的过程,患者在接受这些过程时面临手术相关并发症的风险。此外,这些方法并不适用于反复监测MSI状态的变化 [ 37 ] 。因此,对于MSI状态的评估仍需要进一步的研究以提高准确性和非侵入性。

Kim [ 38 ] 等利用术前18F氟脱氧葡萄糖(FDG)正电子发射断层扫描(PET)/计算机断层扫描(CT)的放射组学结合机器学习方法,对233例术前接受FDG PET/CT检查的结直肠癌(CRC)患者进行了研究。研究团队建立了基于PET的放射组学特征用于预测患者的微卫星不稳定性(MSI)状态。放射组学特征的预测能力通过在测试集中计算受试者工作特征曲线下面积(AUROC)来评估。研究还将结合机器学习的放射组学特征与常规PET参数的预测性能进行了比较。结果显示,基于这两种放射学特征构建的MSI状态的AUROC值相似(分别为0.815和0.867;p = 0.490)。然而,随后进行的Logistic回归分析表明,结合机器学习的放射组学特征是独立预测训练集MSI状态的因子。在AUROC评估方面,结合机器学习的放射组学特征优于代谢肿瘤体积(0.867 vs 0.794, p = 0.015)。这些结果表明,结合机器学习的放射组学特征在预测CRC患者的MSI状态方面具有潜力,并且可能优于传统的代谢参数评估方法。Chen [ 39 ] 等人进行了一项多中心研究涉及837名接受增强CT扫描的结直肠癌(CRC)术前患者。该研究的目的是开发与微卫星不稳定性(MSI)相关的放射组学特征。研究团队采用遗传增强的人工神经网络模型,通过遗传算法选择10个特征并构建了一个特征标签。在内部和外部验证队列中,该特征标签表现出良好的性能,其曲线下面积(AUC)值分别为0.788和0.775。相比之下,该特征标签的性能与联合模型(AUC为0.777和0.767)相当,优于基于年龄和肿瘤位置的临床模型(AUC为0.768和0.623)。生存分析显示,该特征标签可以根据预后对II期CRC患者进行分层(HR: 0.402, p = 0.029)。这些发现表明,使用基因增强的人工神经网络模型构建的放射组学特征在预测CRC患者的MSI状态和预后方面具有潜力,其表现优于传统的临床模型。Fan [ 40 ] 这项回顾性研究包括119名病理证实为II期结直肠癌(CRC)的中国患者,收集微卫星不稳定性(MSI)状态,以及术前对比增强CT图像。使用矩阵实验室(matrix laboratory, MATLAB)从每个整个原发肿瘤的分割体积的门静脉相CT图像中提取放射学特征,并使用最小绝对收缩率和选择运算符逻辑回归模型生成放射学特征。使用放射学特征、临床因素和综合模型来评估MSI状态的预测效能。使用接收器操作特征曲线下的面积、准确性、敏感性和特异性来评估预测性能。与单独使用任何一个特征相比,结合了临床因素和放射学特征的联合模型取得了最佳的整体表现,其曲线下面积、敏感性和特异性分别为0.752、0.663和0.841。明显优于单独使用临床因素的性能0.598、0.371、0.825及单独使用放射学特征的性能则分别为0.688、0.517和0.858。

基于上述文献研究,目前已初步尝试利用影像组学评估结直肠癌(CRC)患者的微卫星不稳定性(MSI)状态。通过对传统图像进行定量分析,影像组学能够充分挖掘图像中的特征信息,从而提高无创评估CRC患者MSI状态的准确性,以判断患者的预后。这种方法在临床上显示出成为广泛应用的新兴影像技术的潜力。

3.3. 预测患者其他基因状态

在结直肠肝转移(CRLM)的个体化治疗决策中,化疗反应的预测至关重要。为了预测CRLM患者对化疗的反应,Shi [ 41 ] 等人进行了一项多中心研究,纳入了159名肝转移性结直肠癌(CRLM)患者,收集了肺部和腹部对比增强CT (CECT)扫描数据。从每个患者的CT扫描的门静脉期(PVP)图像中提取了放射学特征。研究团队采用了七种机器学习算法,分别建立了基于语义学、放射组学及其组合的三种评分。他们采用人工神经网络(ANN)方法,用于预测RAS和BRAF基因的语义和放射学特征。基于放射组学、语义特征和综合评价,构建了三个评分。综合评分成功区分了野生型和突变型患者,在主要队列及验证队列中的AUC分别为0.95和0.79。这项研究表明,将放射组学与语义特征结合起来,对CRLM的RAS (KRAS、NRAS)和BRAF基因突变状态进行改进的非侵入性评估是具有优势的。Yang [ 42 ] 等通过收集了治疗前接受了KRAS/NRAS/BRAF基因突变检测和对比度增强的CT扫描的患者资料。从整个原发肿瘤的门静脉期CT图像中提取了放射学特征。使用单变量分析评估基因突变与临床特征、肿瘤分期和组织学分化之间的关系。采用RELIEFF和支持向量机(SVM)方法选择关键特征来构建放射组学特征。研究结果显示,放射学特征与KRAS/NRAS/BRAF基因突变之间存在明显的关联(p < 0.001)。通过曲线下面积、敏感性和特异性进行衡量,在主要队列中,KRAS/NRAS/BRAF基因突变的预测性能分别为0.869、0.757和0.833,优于验证队列中的0.829、0.686和0.857。这项研究为结直肠癌(CRC)患者提供了一种经过定量、高效和非侵入性测试的创新方法,以优化治疗决策支持。研究结果证实了通过应用影像组学技术获取CRC相关表型的可行性。

4. 总结与展望

深度学习及影像组学在临床应用中尚属较新领域,技术发展迅速,学者采用的方法差异较大,包括CT、MRI和PET等不同模态的图像均显示出了效能和优势。然而,目前的研究多为回顾性和单中心研究,可能存在样本选择偏倚,不能真实反映临床病例的分布情况,影响预测模型的精度。同时,不同中心具有不同的机器参数、扫描设置和诊断规则,单中心研究限制了预测模型的普适性。为确保影像组学特征在不同人群和临床环境中的稳定性和可靠性,未来的研究需要进行大样本量、多中心和标准化的深入研究,细化不同肿瘤类型及亚型的分析。将更多的影像组学研究与肿瘤基因突变信息、临床预后信息相结合,有助于提高影像组学在临床中的应用率,发挥其在结直肠癌患者的诊疗方面的特有优势。

文章引用

李鹏飞,马晓明. 基于深度学习的影像组学在预测结直肠癌基因状态中的应用进展Application Advances in Deep Learning-Based Imaging Radiomics for Predicting Colorectal Cancer Gene Status[J]. 世界肿瘤研究, 2024, 14(01): 27-34. https://doi.org/10.12677/WJCR.2024.141005

参考文献 References Cao, W., Chen, H.D., Yu, Y.W., et al. (2021) Changing Profiles of Cancer Burden Worldwide and in China: A Second-ary Analysis of the Global Cancer Statistics 2020. Chinese Medical Journal (England), 134, 783-791.
https://doi.org/10.1097/CM9.0000000000001474
Valentini, V., Glimelius, B., Haustermans, K., et al. (2014) EURECCA Consensus Conference Highlights about Rectal Cancer Clinical Management: The Radiation Oncologist’s Expert Review. Radiotherapy and Oncology, 110, 195-198.
https://doi.org/10.1016/j.radonc.2013.10.024
Fujita, S., Mizusawa, J., Kanemitsu, Y., et al. (2017) Mesorectal Excision with or without Lateral Lymph Node Dissection for Clinical Stage II/III Lower Rectal Cancer (JCOG0212): A Multicenter, Randomized Controlled, Noninferiority Trial. Annals of Surgery, 266, 201-207.
https://doi.org/10.1097/SLA.0000000000002212
Biller, L.H. and Schrag, D. (2021) Diagnosis and Treatment of Metastatic Colorectal Cancer: A Review. JAMA, 325, 669-685.
https://doi.org/10.1001/jama.2021.0106
Zhou, J., Ji, Q. and Li, Q. (2021) Resistance to Anti-EGFR Therapies in Metastatic Colorectal Cancer: Underlying Mechanisms and Reversal Strategies. Journal of Experimental & Clinical Cancer Research, 40, Article No. 328.
https://doi.org/10.1186/s13046-021-02130-2
Herreros-Villanueva, M., Chen, C.C., Yuan, S.S., et al. (2014) KRAS Mutations: Analytical Considerations. Clinica Chimica Acta, 431, 211-220.
https://doi.org/10.1016/j.cca.2014.01.049
Lambin, P., Rios-Velazquez, E., Leijenaar, R., et al. (2012) Radi-omics: Extracting More Information from Medical Images Using Advanced Feature Analysis. European Journal of Cancer, 48, 441-446.
https://doi.org/10.1016/j.ejca.2011.11.036
Engin, G., Sharifov, R., Gural, Z., et al. (2012) Can Diffu-sion-Weighted MRI Determine Complete Responders after Neoadjuvant Chemoradiation for Locally Advanced Rectal Cancer? Diagnostic and Interventional Radiology, 18, 574-581.
https://doi.org/10.4261/1305-3825.DIR.5755-12.1
Aerts, H.J., Grossmann, P., Tan, Y., et al. (2016) Defining a Radiomic Response Phenotype: A Pilot Study Using Targeted Therapy in NSCLC. Scientific Reports, 6, Article No. 33860.
https://doi.org/10.1038/srep33860
Litjens, G., Kooi, T., Bejnordi, B.E., et al. (2017) A Survey on Deep Learning in Medical Image Analysis. Medical Image Analysis, 42, 60-88.
https://doi.org/10.1016/j.media.2017.07.005
Zhang, Q., Yang, L.T., Chen, Z., et al. (2018) A Survey on Deep Learning for Big Data. Information Fusion, 42, 146-157.
https://doi.org/10.1016/j.inffus.2017.10.006
Roth, A.D., Tejpar, S., Delorenzi, M., et al. (2010) Prognostic Role of KRAS and BRAF in Stage II and III Resected Colon Cancer: Results of the Translational Study on the PETACC-3, EORTC 40993, SAKK 60-00 Trial. Journal of Clinical Oncology, 28, 466-474.
https://doi.org/10.1200/JCO.2009.23.3452
Dienstmann, R., Connor, K., Byrne, A.T., et al. (2020) Precision Therapy in RAS Mutant Colorectal Cancer. Gastroenterology, 158, 806-811.
https://doi.org/10.1053/j.gastro.2019.12.051
Imamura, Y., Morikawa, T., Liao, X., et al. (2012) Specific Mu-tations in KRAS Codons 12 and 13, and Patient Prognosis in 1075 BRAF Wild-Type Colorectal Cancers. Clinical Can-cer Research, 18, 4753-4763.
https://doi.org/10.1158/1078-0432.CCR-11-3210
Jones, R.P., Sutton, P.A., Evans, J.P., et al. (2017) Specific Mutations in KRAS Codon 12 Are Associated with Worse Overall Survival in Patients with Advanced and Recurrent Colorectal Cancer. British Journal of Cancer, 116, 923-929.
https://doi.org/10.1038/bjc.2017.37
Di Fiore, F., Charbonnier, F., Lefebure, B., et al. (2008) Clinical Interest of KRAS Mutation Detection in Blood for Anti-EGFR Therapies in Metastatic Colorectal Cancer. British Journal of Cancer, 99, 551-552.
https://doi.org/10.1038/sj.bjc.6604451
Marisa, L., De Reynies, A., Duval, A., et al. (2013) Gene Expression Classification of Colon Cancer into Molecular Subtypes: Characterization, Validation, and Prognostic Value. PLOS Med-icine, 10, e1001453.
https://doi.org/10.1371/journal.pmed.1001453
Pershad, Y., Govindan, S., Hara, A.K., et al. (2017) Using Naive Bayesian Analysis to Determine Imaging Characteristics of KRAS Mutations in Metastatic Colon Cancer. Diag-nostics (Basel), 7, Article No. 50.
https://doi.org/10.3390/diagnostics7030050
Xu, Y., Xu, Q., Sun, H., et al. (2018) Could IVIM and ADC Help in Predicting the KRAS Status in Patients with Rectal Cancer? European Radiology, 28, 3059-3065.
https://doi.org/10.1007/s00330-018-5329-y
Jo, S.J. and Kim, S.H. (2019) Association between Oncogenic RAS Mutation and Radiologic-Pathologic Findings in Patients with Primary Rectal Cancer. Quantitative Imaging in Medicine and Surgery, 9, 238-246.
https://doi.org/10.21037/qims.2018.12.10
Gultekin, M.A., Turk, H.M., Besiroglu, M., et al. (2020) Relation-ship between KRAS Mutation and Diffusion Weighted Imaging in Colorectal Liver Metastases; Preliminary Study. Eu-ropean Journal of Radiology, 125, Article No. 108895.
https://doi.org/10.1016/j.ejrad.2020.108895
Song, C., Shen, B., Dong, Z., et al. (2020) Diameter of Superior Rectal Vein-CT Predictor of KRAS Mutation in Rectal Carcinoma. Cancer Management and Research, 12, 10919-10928.
https://doi.org/10.2147/CMAR.S270727
Promsorn, J., Chadbunchachai, P., Somsap, K., et al. (2021) Imaging Features Associated with Survival Outcomes among Colorectal Cancer Patients with and without KRAS Mutation. Egyptian Journal of Radiology and Nuclear Medicine, 52, Article No. 15.
https://doi.org/10.1186/s43055-020-00393-x
Lv, Y., Wang, X., Liang, L., et al. (2019) SUVmax and Metabolic Tumor Volume: Surrogate Image Biomarkers of KRAS Mutation Status in Colorectal Cancer. OncoTargets and Therapy, 12, 2115-2121.
https://doi.org/10.2147/OTT.S196725
Arslan, E., Aksoy, T., Gursu, R.U., et al. (2020) The Prognostic Value of (18)F-FDG PET/CT and KRAS Mutation in Colorectal Cancers. Molecular Imaging and Radionuclide Therapy, 29, 17-24.
https://doi.org/10.4274/mirt.galenos.2019.33866
He, P., Zou, Y., Qiu, J., et al. (2021) Pretreatment (18)F-FDG PET/CT Imaging Predicts the KRAS/NRAS/BRAF Gene Mutational Status in Colorectal Cancer. Journal of Oncology, 2021, Article ID: 6687291.
https://doi.org/10.1155/2021/6687291
Liu, X., Wang, S.C., Ni, M., et al. (2022) Correlation between (18)F-FDG PET/CT Intra-Tumor Metabolic Heterogeneity Parameters and KRAS Mutation in Colorectal Cancer. Ab-dominal Radiology (NY), 47, 1255-1264.
https://doi.org/10.1007/s00261-022-03432-5
Song, K., Zhao, Z., Ma, Y., et al. (2022) A Multitask Du-al-Stream Attention Network for the Identification of KRAS Mutation in Colorectal Cancer. Medical Physics, 49, 254-270.
https://doi.org/10.1002/mp.15361
Taguchi, N., Oda, S., Yokota, Y., et al. (2019) CT Texture Analy-sis for the Prediction of KRAS Mutation Status in Colorectal Cancer via a Machine Learning Approach. European Jour-nal of Radiology, 118, 38-43.
https://doi.org/10.1016/j.ejrad.2019.06.028
Wu, X., Li, Y., Chen, X., et al. (2020) Deep Learning Features Improve the Performance of a Radiomics Signature for Predicting KRAS Status in Patients with Colorectal Cancer. Aca-demic Radiology, 27, e254-e262.
https://doi.org/10.1016/j.acra.2019.12.007
Jiricny, J. (2006) The Multifaceted Mismatch-Repair System. Na-ture Reviews Molecular Cell Biology, 7, 335-346.
https://doi.org/10.1038/nrm1907
Bao, X., Zhang, H., Wu, W., et al. (2020) Analysis of the Molecular Nature Associated with Microsatellite Status in Colon Cancer Identifies Clinical Implications for Immunotherapy. The Journal for ImmunoTherapy of Cancer, 8, e001437.
https://doi.org/10.1136/jitc-2020-001437
De’Angelis, G.L., Bottarelli, L., Azzoni, C., et al. (2018) Microsatel-lite Instability in Colorectal Cancer. Acta Biomedica, 89, 97-101. Mei, W.J., Mi, M., Qian, J., et al. (2022) Clinico-pathological Characteristics of High Microsatellite Instability/Mismatch Repair-Deficient Colorectal Cancer: A Narrative Review. Frontiers in Immunology, 13, Article ID: 1019582.
https://doi.org/10.3389/fimmu.2022.1019582
Cohen, R., Buhard, O., Cervera, P., et al. (2017) Clinical and Molecular Characterisation of Hereditary and Sporadic Metastatic Colorectal Cancers Harbouring Microsatellite Instabil-ity/DNA Mismatch Repair Deficiency. European Journal of Cancer, 86, 266-274.
https://doi.org/10.1016/j.ejca.2017.09.022
Luchini, C., Bibeau, F., Ligtenberg, M.J.L., et al. (2019) ESMO Recommendations on Microsatellite Instability Testing for Immunotherapy in Cancer, and Its Relationship with PD-1/PD-L1 Expression and Tumour Mutational Burden: A Systematic Review-Based Approach. Annals of Oncology, 30, 1232-1243.
https://doi.org/10.1093/annonc/mdz116
Bhargava, R. and Madabhushi, A. (2016) Emerging Themes in Image Informatics and Molecular Analysis for Digital Pathology. Annual Review of Biomedical Engineering, 18, 387-412.
https://doi.org/10.1146/annurev-bioeng-112415-114722
Kim, S., Lee, J.H., Park, E.J., et al. (2023) Prediction of Microsatellite Instability in Colorectal Cancer Using a Machine Learning Model Based on PET/CT Radiomics. Yonsei Medical Journal, 64, 320-326.
https://doi.org/10.3349/ymj.2022.0548
Chen, X., He, L., Li, Q., et al. (2023) Non-Invasive Prediction of Mi-crosatellite Instability in Colorectal Cancer by a Genetic Algorithm-Enhanced Artificial Neural Network-Based CT Ra-diomics Signature. European Radiology, 33, 11-22.
https://doi.org/10.1007/s00330-022-08954-6
Fan, S., Li, X., Cui, X., et al. (2019) Computed Tomogra-phy-Based Radiomic Features Could Potentially Predict Microsatellite Instability Status in Stage II Colorectal Cancer: A Preliminary Study. Academic Radiology, 26, 1633-1640.
https://doi.org/10.1016/j.acra.2019.02.009
Shi, R., Chen, W., Yang, B., et al. (2020) Prediction of KRAS, NRAS and BRAF Status in Colorectal Cancer Patients with Liver Metastasis Using a Deep Artificial Neural Network Based on Radiomics and Semantic Features. American Journal of Cancer Research, 10, 4513-4526. Yang, L., Dong, D., Fang, M., et al. (2018) Can CT-Based Radiomics Signature Predict KRAS/NRAS/BRAF Mutations in Colo-rectal Cancer? European Radiology, 28, 2058-2067.
https://doi.org/10.1007/s00330-017-5146-8
Baidu
map