胃肠胰神经内分泌肿瘤CT影像组学研究进展
Advances in CT Radiomics Research on Gastroenteropancreatic Neuroendocrine Neoplasms
摘要: 胃肠胰神经内分泌肿瘤(gastroenteropancreatic neuroendocrine neoplasms, GEP-NENs)是一类具有高度多样性和复杂性的罕见肿瘤,其发病率逐年增加。CT检查是诊断、评估、随访GEP-NENs的重要影像学方法,而影像组学作为一种新兴的无创分析方法,可从诊断图像中提取定量且具有可重复性的特征参数,已被广泛运用于各类肿瘤的诊断、疗效评估、预后预测等多个方面。本文就CT影像组学在GEP-NENs鉴别诊断、病理分级预测、生物学行为预测、疗效评估及预后预测的研究进展进行综述,以期为后续研究方向提供思路。
Abstract: Gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs) represent a rare tumor category characterized by high diversity and complexity, with an escalating annual incidence. Computed tomography (CT) serves as a crucial imaging modality for diagnosing, evaluating, and monitoring GEP-NENs. As an emerging non-invasive analytical approach, radiomics enables the extraction of quantitative and reproducible feature parameters from diagnostic images. It has been widely applied in various aspects of tumor diagnosis, treatment efficacy assessment, and prognosis prediction. This review summarizes the research progress of CT radiomics in the differential diagnosis, pathological grading prediction, biological behavior prediction, treatment efficacy assessment, and prognosis prediction of GEP-NENs, aiming to provide insights for future research directions.
文章引用:王欣, 汤丽平. 胃肠胰神经内分泌肿瘤CT影像组学研究进展[J]. 临床医学进展, 2026, 16(3): 2100-2108. https://doi.org/10.12677/acm.2026.1631001

1. 引言

神经内分泌肿瘤(Neuroendocrine neoplasms, NENs)是一类具有高度异质性的罕见肿瘤,起源于神经内分泌细胞和肽能神经元[1] [2]。NENs可以发生于人体的各个部位,约3/2的病例发生在胃肠胰腺系统中,故胃肠胰神经内分泌肿瘤(gastroenteropancreatic neuroendocrine neoplasms, GEP-NENs)是NENs最主要的亚型[3],其中50%肠道,30%胰腺[4] [5]。近几十年来,由于病理诊断技术的进步和影像学检查的普遍应用,全球GEP-NENs的发病率呈现上升趋势[3] [6]。美国NENs的发病率从1973年的1.09/10万到2012年的6.98/10万升高了6.4倍[5] [7],而在我国台湾地区,NENs的发病率从1996年的0.244/10万增加到2015年的3.162/10万[8]

计算机断层扫描(computed tomography, CT)具有扫描速度快、空间分辨率好、使用成本低等优点,在GEP-NENs的临床诊断、病理分级、病情评估及随访监测中应用广泛[9]。但其在评估肿瘤异质性、预测生物学行为等方面仍存在一定的局限性[10]。影像组学作为一种先进的分析技术,包括图像采集和分割、特征提取和同质化处理、特征选择、分析和建立预测模型这几个步骤,旨在从诊断图像中提取定量、可重复性好的特征信息,包括人眼难以识别或量化的复杂部分[11] [12]。基于CT的影像组学不仅能非侵入性地量化肿瘤及其微环境的异质性,还有与基因组学等分子特征相关联,从而在肿瘤无创分型、预后预测及疗效评估等多个方面中展现出独特的优势[13]。本文就CT影像组学在GEP-NENs鉴别诊断、病理分级、生物学行为预测、疗效评估及预后预测的研究进展进行综述,以期为后续研究方向提供思路。

2. 鉴别诊断

关于GEP-NENs与其他肿瘤的鉴别诊断,现有研究主要集中于胰腺神经内分泌瘤(pancreatic neuroendocrine tumors, PNETs)与胰腺导管腺癌(pancreatic ductal adenocarcinomas, PDACs)之间[14]。胰腺癌作为消化系统最常见且最具有侵袭性的恶性肿瘤之一,最常见的类型是PDACs,其次即为PNETs,两者在治疗方案选择、疾病进展及预后等方面存在显著差异[15] [16]。然而,PDACs和PNETs在常规CT影像学上表现相似,这使得仅通过放射科医师的视觉评估易出现误诊[17]。He等[18]通过ITK-SNAP软件2D逐层手动勾画肿瘤感兴趣区域(regions of interest, ROI)并使用LASSO回归筛选了7个影像特征,构建了区分不典型非功能性胰腺神经内分泌瘤(nonfunctional pancreatic neuroendocrine tumors, NF-PNETs)与PDACs的临床影像模型、影像组学模型及两者结合的综合模型,发现综合模型在独立验证队列中效能最佳(AUC = 0.884),显著优于单纯临床模型,且通过列线图实现了可视化预测。Zhang等[19]则进一步提升使用LIFEx影像组学专用软件3D手动绘制ROI,融合五种特征选择算法与九种分类算法构建了45种诊断模型,实现了从单算法到多算法组合的进步。其中基于梯度提升决策树(Gradient Boosting Decision Tree, GBDT)筛选特征并结合随机森林(Random Forest, RF)进行分类的模型表现最佳,训练集与验证集的AUC分别达到0.971和0.930。现有研究共同表明,使用CT影像组学方法能显著提升PNETs与PDACs的鉴别效能,为临床精准诊疗提供了可靠的依据[20]

除此之外,有研究者运用CT影像组学来鉴别胃神经内分泌癌(gastric neuroendocrine carcinomas, g-NECs)与胃腺癌(gastric adenocarcinomas, GAC)。Wang等[21]的研究表明,基于静脉期CT影像提取的GLCM能量、熵等纹理特征结合肿瘤边界清晰度与淋巴结转移状态所构建的影像组学列线图模型,在区分g-NECs与GAC方面表现出较高诊断效能,训练集AUC达0.821。He等[22]在Wang等研究者的基础上进一步深化,创新性地将胃混合性神经–非神经内分泌癌(gastric mixed adenocarcinoma-neuroendocrine carcinoma, g-MANEC)纳入鉴别体系,并构建了融合传统CT特征(肿瘤坏死和淋巴结转移)与影像组学评分的综合模型,该模型通过一致性检验、LASSO回归等一系列更严格的特征筛选流程,在区分g-NEC与GAC方面取得更高AUC (训练集为0.887,验证集为0.852),并首次实现g-MANEC与GAC的亚组鉴别,其在训练集和验证集中的AUC分别为0.823、0.762。而国内研究者基于CT影像组学进行直肠神经内分泌瘤(rectal neuroendocrine tumors, RNETs)和直肠腺癌(rectal adenocarcinoma, READ)的鉴别,于直肠增强CT平扫及静脉期图像中手动勾画ROI,通过应用主成分分析方法(principal components analysis, PCA)和方差分析方法(analysis of variance, ANOVA)筛选特征,构建逻辑回归(logistic regression, LR)、RF及决策树(decision tree, DT)三种分类器模型,结果显示基于静脉期图像的LR模型显示出最佳的分类效能,AUC为0.89,敏感性0.76,特异性0.75 [23]。以上研究为上述临床治疗与预后迥异的恶性肿瘤提供了稳定、无创的术前鉴别工具,进一步拓展了影像组学在神经内分泌肿瘤精准诊断中的应用场景。

3. 病理分级

传统CT依赖肿瘤大小、强化方式、有无坏死及侵犯范围等形态学特征评估分级,但其主观性强且难以捕捉肿瘤微观异质性,导致对GEP-NENs的病理诊断效能有限。而CT影像组学通过量化肿瘤纹理、密度分布、边缘形态等深层次特征,可精准反映肿瘤细胞增殖活性与分化程度的差异,为术前无创分级提供了新途径。基于组织分化程度和细胞增殖活性,2022年第五版《WHO内分泌与神经内分泌肿瘤分类》将GEP-NENs分为胃肠胰神经内分泌瘤(gastroenteropancreatic neuroendocrine tumors, GEP-NETs)、胃肠胰神经内分泌癌(gastroenteropancreatic neuroendocrine carcinoma, GEP-NECs)和胃肠胰混合性神经内分泌–非神经内分泌肿瘤(gastroenteropancreatic mixed neuroendocrine/non-neuroendocrine neoplasms, GEP-MiNENs) [24]。其中,根据Ki-67增殖指数和核分裂象,GEP-NETs可进一步分为G1,低级别[Ki-67增殖指数 ≤ 2%和(或)核分裂象 < 2/10高倍视野];G2,中级别[Ki-67指数3%~20%和(或)核分裂象2-20/10高倍视野];G3,高级别[Ki-67增殖指数 > 20%和(或)核分裂象 > 20/10高倍视野] [25]。G3级NETs和NEC因在诊断时多已出现转移,被欧洲神经内分泌肿瘤协会(European Neuroendocrine Tumor Society, ENETS)指南分类为高级别NENs,相较G1~G2级NENs的5年生存率更低,需要更加积极的治疗方式[26]。由此,准确的病理分级诊断在GEP-NENs患者的个体化治疗和预后评估中起着十分重要的作用。

大多数GEP-NENs影像组学的研究侧重于PNENs病理分级预测。多项研究已经验证了CT影像组学在PNENs病理分级诊断中的有效性,并逐步形成了从单一模态或期相分析向多模态、多期相、多参数融合建模发展的明确趋势。

根据模型预测目标这一核心临床问题,现有研究可归纳为三大方向:第一类,旨在鉴别高级别(G2~G3级)与低级别(G1级)肿瘤,这是最常见的应用场景。Ricci等[27]基于增强CT三期相图像半自动手动分割并自动化提取三维纹理特征,选择LASSO回归筛选特征构建模型,专门用于预测“高级别(非G1级)”PNENs。Liu等[28]进一步通过融合CT动脉期与MRI T2加权图像的影像组学特征,使用Wilcoxon秩和检验和最小冗余最大相关性(Max-Relevance and Min-Redundancy, mRMR)两步骤筛选3D特征,应用线性判别(linear discriminant analysis, LAD)分类器构建预测模型,成功区分G1与G2~3级NF-PNETs。Wang等[29]结合平扫CT影像组学特征与T分期等临床指标,Ye等[30]提取并由mRMR算法筛选动脉期及静脉期双期图像特征,通过对比择优选用RF算法开发影像组学机器学习模型,Javed等[31]同样使用增强CT双期图像和RF算法,融合包含10个特征的影像组学评分与肿瘤大小,均构建了针对鉴别G1级与G2~G3级PNETS的高效预测模型,其验证集AUC分别为0.875、0.779及0.80,显示出此类模型稳健的鉴别能力。第二类,专注于G1与G2级的精细鉴别。Bian等[32]仅基于门静脉期CT图像LASSO回归筛选特征构建的模型,在区分G1与G2级(总体AUC = 0.86)及针对 ≤ 2 cm的小PNETs (AUC = 0.81)中均表现出色。Zhao等[33]则通过mRMR筛选融合平扫期与静脉期的多类纹理特征,构建的支持向量机(Support Vector Machine, SVM)模型实现了对G1与G2的高精度鉴别,其验证集AUC达0.876,敏感性90.9%,特异性88.9%。第三类,探索更复杂的分级策略或广义分组。Pulvirenti等[34]采用创新的两阶段SVM模型,先区分G1~G2与G3级,再在G1~G2组内区分G1与G2级,模拟了病理分级的多层次决策过程。Chiti等[35]的研究则遵循ENETS的预后分类思路,将G3级与NEC统一归类为“高级别”,与“低级别(G1~G2级)”进行区分,其动脉期模型展现了最优预测性能。上述关于PNENs病理分级的研究,在模型构建方法上,研究普遍体现出多特征融合提升性能的一致性认识。然而,在影像期相与模态的选择上仍存在策略差异,这些差异表明,针对不同的分级目标,最优的影像采集方案可能不同,目前尚未形成共识。同时,算法选择方面同样呈现出多样性,虽然多种机器学习方法均被证实有效,但最优算法尚未统一。尽管各研究在模型构建方法、特征选择策略及所用影像期相上存在差异,但其构建的模型普遍显示出较高的预测性能。

相较于PNENs,胃及肠道NENs的CT影像组学分级研究相对较少,但已有初步探索显示其应用潜力。王睿及其团队[36]针对81例胃神经内分泌肿瘤(gastric neuroendocrine neoplasms, g-NENs)患者进行研究,提取并筛选动脉期及静脉期影像组学特征,发现采用XGboost算法建立联合模型对g-NENs G1~G2级和G3级具有良好的鉴别能力。国外有研究者[37]回顾性分析了61例直肠神经内分泌肿瘤(rectal neuroendocrine neoplasms, R-NENs)患者的增强CT图像,提取动脉期及静脉期图像纹理参数,研究结果发现多个动脉期及静脉期直方图参数(均值、中位数、第10、25、75和90百分位数)在G1与G2/G3/NEC间差异有统计学意义。

4. 生物学行为预测

GEP-NENs在生物学行为上呈现出高度异质性[38],治疗方案的制定需要综合考虑肿瘤生物学特征(如淋巴结转移、远处转移)。近年来,CT影像组学因其可提取潜在的病理生理信息,逐渐成为肿瘤生物学行为分析的有力工具[39]。Gu等[40]多中心收集了320名NF-PNETs患者术前增强CT图像并提取特征,发现结合CT影像组学与深度学习特征构建的多中心预测模型可高效预测NF-PNETs淋巴结转移风险(内部验证AUC达0.88,外部验证AUC达0.91),并能有效指导不同肿瘤大小(≤2 cm与>2 cm)患者的治疗决策。类似地,Ahmed等人[41]整合了增强CT影像组学特征与肿瘤大小、病理分级,基于RF算法构建模型,对G1~G2级NF-PNETs患者淋巴结转移的预测在验证集上AUC为0.80,并可准确识别出85%肿瘤直径 < 2 cm患者的淋巴结转移。此外,也有研究基于PNETs患者动脉期增强CT图像建立模型,可稳健预测假包膜状态(训练集AUC 0.75,验证集AUC 0.74) [42]。其他学者还使用CT影像组学特征构建模型预测PNENs患者微血管浸润、术后肝转移风险等关键生物学行为[43] [44]

值得注意的是,鲜少有胰腺之外的GEP-NENs生物学行为的研究。Blazevic等人[45]的工作填补了部分空白,通过提取病灶周围肠系膜的影像组学特征,构建了预测小肠神经内分泌肿瘤(small intestinal neuroendocrine tumors, SI-NETs)转移性肠系膜肿块所致肠道并发症的模型,AUC达到0.81,显示出该技术在非胰腺GEP-NENs中同样具有应用前景。

5. 疗效评估

在GEP-NENs患者的治疗中,早期评估治疗反应有助于治疗方案的选择,对于临床实践具有重要意义。传统评价方法难以全面反映肿瘤内部异质性与治疗应答的早期生物学改变,而CT影像组学通过高通量提取并分析图像特征,为无创性预测治疗反应提供了新思路[11]

5.1. 生长抑素类似物治疗

生长抑素类似物(somatostatin analogues, SSAs)已被国内外指南和共识推荐为生长缓慢、Ki67指数 ≤ 10%的SSTR阳性的晚期GEP-NETs的一线治疗,可以抑制50%~80%患者的肿瘤生长[46]-[48]。然而,部分高分化GEP-NETs患者对SSAs治疗反应不佳,治疗后病情出现进展,故早期识别此类患者至关重要。Polici等[49]分析55例接受SSAs治疗的GEP-NETs肝转移患者的基线肝脏CT影像组学特征,开发了结合Ki-67指数与影像组学的联合预测模型,结果显示联合模型AUC为0.814,表明利用基于CT的影像组学特征可早期识别出SSAs治疗无效或进展高风险的晚期GEP-NETs患者。

5.2. 靶向治疗

在靶向治疗的新时代,依维莫司作为一种口服的哺乳动物雷帕霉素靶蛋白(mammalian target of rapamycin, mTOR)的选择性抑制剂,已被用于治疗晚期NETs [50],有研究者[51]回顾性分析了25例转移性NETs患者(胰腺15例,回肠9例,肺1例)经依维莫司治疗前的肝脏增强CT图像,采用全肝脏3D手动分割并提取影像组学特征,构建模型进行内部与合成外部验证,结果显示筛选出的10个放射组学特征能够有效区分依维莫司治疗的应答者与非应答者,其中动脉期GLCM_Correlation和GLCM_Imc1表现最佳(内部验证AUC 0.86~0.84,外部验证AUC 0.84~0.90),并成功基于GLSZM_GrayLevelVariance和GLSZM_ZonePercentage构建可独立于临床参数预测患者死亡风险的影像组学模型。

舒尼替尼是一种抗血管生成的多靶点酪氨酸激酶抑制剂(tyrosine kinase inhibitors, TKIs),已被临床指南推荐用于治疗进展期G1~G2 PNETs [48],但临床诊疗中缺乏有效的疗效预测方法。由此,Chen等[52]针对38例PNETs患者建立了基于治疗前CT图像的影像组学模型,用以预测舒尼替尼治疗后的疗效和无进展生存期(progression-free survival, PFS),结果发现影像组学模型预测肿瘤缩小的效能在训练集和验证集中均表现良好(AUC分别为0.915和0.770)。进一步生存分析提示,影像组学特征与PFS相关(P = 0.020)。上述研究证实,影像组学在靶向治疗疗效预测方面具有较高准确性和临床应用潜力。

5.3. 放射性核素治疗

肽受体介导的放射性核素治疗(peptide receptor radionuclide therapy, PRRT)是针对NETs的一种全身治疗方法,利用NETs细胞表面生长抑素受体(somatostatin receptor, SSTR)的过度表达,将放射性核素靶向结合到肿瘤细胞产生杀伤作用[53] [54]。现有临床研究中,对比高剂量奥曲肽,177Lu-DOTATATE被证实可显著延长晚期转移性GEP-NETs患者的PFS和总生存期(overall survival, OS) [55]。Behmanesh等[56]研究整合治疗前CT影像组学特征与临床生物标志物,构建了能够高准确率预测NETs患者对177Lu-DOTATATE治疗反应的模型,其中基于XGBoost特征选择方法构建的RF模型的准确率达89%。上述研究结果提示影像组学特征可以应用于PRRT治疗反应的预测,实现个体化治疗。

6. 预后预测

GEP-NENs的预后取决于病灶大小、肿瘤病理分级、功能状态、局部浸润、远处转移等多种临床和病理因素[57],依赖于WHO分级和AJCC TNM分期,而CT影像组学为GEP-NENs的预后预测提供了突破传统方法的新工具。

当前研究一致表明,从术前CT图像中提取的影像组学特征是预测GEP-NENs术后无复发生存期(recurrence-free survival, RFS)与OS的独立风险因素。Homps等[58]提取37例PNETs患者影像组学特征,筛选并构建影像组学指数,结果发现该指数 > 0.4的患者的中位RFS显著缩短(36个月VS 84个月)。进一步地,Heo等[59]研究者扩大数据量并加入外部验证,开发并验证了整合年龄、AJCC分期与动脉期CT影像组学特征的联合模型,其预测RFS和OS的一致性指数分别为0.734、0.781,较临床模型明显升高。与之类似,Yang等人[60]单中心回顾性纳入182例g-NENs患者,3D手动全层分割增强CT图像ROI并通过LASSO-Cox回归分析筛选特征,研究结果显示动静脉期影像组学特征形成的联合模型在预测OS方面表现最佳,可将g-NENs 患者分层为高风险组和低风险组。进一步地,Yang等研究者在另一篇文章中使用双中心数据开发深度学习影像组学模型,进一步证明了CT影像组学用于g-NENs的预后预测中的效能[61]。另一些研究者[62]将纳入研究的NENs范围进一步扩大至胃肠胰腺,将临床数据与多期相影像组学特征整合构建的联合模型同样有助于预测预后,其效能(训练集AUC达0.885)显著优于任何单一模型。

7. 小结

综上所述,CT影像组学作为一种新兴的定量影像分析方法,在GEP-NENs的诊断与全程管理中展现出重要的应用潜力。当前研究已初步证实,其在肿瘤鉴别诊断、病理分级评估、生物学行为预测、治疗反应监测及预后判断等方面具有一定的辅助价值,为GEP-NENs个体化诊疗提供了新的无创工具。然而,该领域仍面临诸多挑战,多数研究为单中心、回顾性研究,且主要聚焦于胰腺、病理诊断方面,存在样本量有限、特征提取与建模方法不一、缺乏外部验证等局限性,未来研究应致力于推动多中心、前瞻性队列的建立,加强影像组学特征与其他生物学信息的整合,并结合深度学习等先进算法提升模型的效能和泛化能力。此外,标准化影像采集流程、特征提取及分析过程也是实现影像组学向临床实践转化的重要前提。

CT影像组学有望成为GEP-NENs精准影像评估体系的有力补充,其进一步发展需依靠多学科的深度协作,以最终实现对其诊断、治疗及预后预测等全流程优化,推动GEP-NENs临床管理向更精准、动态的方向迈进。

NOTES

*通讯作者。

参考文献

[1] Zhang, X., Fan, Y., Jing, R., Getu, M.A., Chen, W., Zhang, W., et al. (2024) Gastroenteropancreatic Neuroendocrine Neoplasms: Current Development, Challenges, and Clinical Perspectives. Military Medical Research, 11, Article No. 35. [Google Scholar] [CrossRef] [PubMed]
[2] Xu, Z., Wang, L., Dai, S., Chen, M., Li, F., Sun, J., et al. (2021) Epidemiologic Trends of and Factors Associated with Overall Survival for Patients with Gastroenteropancreatic Neuroendocrine Tumors in the United States. JAMA Network Open, 4, e2124750. [Google Scholar] [CrossRef] [PubMed]
[3] Shi, M., Fan, Z., Xu, J., Yang, J., Li, Y., Gao, C., et al. (2021) Gastroenteropancreatic Neuroendocrine Neoplasms G3: Novel Insights and Unmet Needs. Biochimica et Biophysica Acta (BBA)—Reviews on Cancer, 1876, Article ID: 188637. [Google Scholar] [CrossRef] [PubMed]
[4] Fernandez, C.J., Agarwal, M., Pottakkat, B., Haroon, N.N., George, A.S. and Pappachan, J.M. (2021) Gastroenteropancreatic Neuroendocrine Neoplasms: A Clinical Snapshot. World Journal of Gastrointestinal Surgery, 13, 231-255. [Google Scholar] [CrossRef] [PubMed]
[5] Dasari, A., Shen, C., Halperin, D., Zhao, B., Zhou, S., Xu, Y., et al. (2017) Trends in the Incidence, Prevalence, and Survival Outcomes in Patients with Neuroendocrine Tumors in the United States. JAMA Oncology, 3, Article No. 1335. [Google Scholar] [CrossRef] [PubMed]
[6] Das, S. and Dasari, A. (2021) Epidemiology, Incidence, and Prevalence of Neuroendocrine Neoplasms: Are There Global Differences? Current Oncology Reports, 23, Article No. 43. [Google Scholar] [CrossRef] [PubMed]
[7] Zheng, R., Zhao, H., An, L., Zhang, S., Chen, R., Wang, S., et al. (2023) Incidence and Survival of Neuroendocrine Neoplasms in China with Comparison to the United States. Chinese Medical Journal, 136, 1216-1224. [Google Scholar] [CrossRef] [PubMed]
[8] Chang, J.S., Chen, L., Shan, Y., Chu, P., Tsai, C. and Tsai, H. (2021) An Updated Analysis of the Epidemiologic Trends of Neuroendocrine Tumors in Taiwan. Scientific Reports, 11, Article No. 7881. [Google Scholar] [CrossRef] [PubMed]
[9] Tobias, J. and Keutgen, X.M. (2024) Diagnostics and Imaging for Pancreatic Neuroendocrine Tumors. Surgical Clinics of North America, 104, 883-890. [Google Scholar] [CrossRef] [PubMed]
[10] 牟玮, 田捷. PET/CT、SPECT/CT影像组学: 沟通宏观影像和微观分子的桥梁[J]. 中华核医学与分子影像杂志, 2024, 44(2): 65-67.
[11] Mayerhoefer, M.E., Materka, A., Langs, G., Häggström, I., Szczypiński, P., Gibbs, P., et al. (2020) Introduction to Radiomics. Journal of Nuclear Medicine, 61, 488-495. [Google Scholar] [CrossRef] [PubMed]
[12] De Muzio, F., Pellegrino, F., Fusco, R., Tafuto, S., Scaglione, M., Ottaiano, A., et al. (2023) Prognostic Assessment of Gastropancreatic Neuroendocrine Neoplasm: Prospects and Limits of Radiomics. Diagnostics, 13, Article No. 2877. [Google Scholar] [CrossRef] [PubMed]
[13] Chen, M., Copley, S.J., Viola, P., Lu, H. and Aboagye, E.O. (2023) Radiomics and Artificial Intelligence for Precision Medicine in Lung Cancer Treatment. Seminars in Cancer Biology, 93, 97-113. [Google Scholar] [CrossRef] [PubMed]
[14] Staal, F.C.R., Aalbersberg, E.A., van der Velden, D., Wilthagen, E.A., Tesselaar, M.E.T., Beets-Tan, R.G.H., et al. (2022) GEP-NET Radiomics: A Systematic Review and Radiomics Quality Score Assessment. European Radiology, 32, 7278-7294. [Google Scholar] [CrossRef] [PubMed]
[15] Berbís, M.Á., Godino, F.P., Rodríguez-Comas, J., Nava, E., García-Figueiras, R., Baleato-González, S., et al. (2023) Radiomics in CT and MR Imaging of the Liver and Pancreas: Tools with Potential for Clinical Application. Abdominal Radiology, 49, 322-340. [Google Scholar] [CrossRef] [PubMed]
[16] Becker, A.E. (2014) Pancreatic Ductal Adenocarcinoma: Risk Factors, Screening, and Early Detection. World Journal of Gastroenterology, 20, Article No. 11182. [Google Scholar] [CrossRef] [PubMed]
[17] Kimura, T., Miyamoto, H., Fukuya, A., Kitamura, S., Okamoto, K., Kimura, M., et al. (2016) Neuroendocrine Carcinoma of the Pancreas with Similar Genetic Alterations to Invasive Ductal Adenocarcinoma. Clinical Journal of Gastroenterology, 9, 261-265. [Google Scholar] [CrossRef] [PubMed]
[18] He, M., Liu, Z., Lin, Y., Wan, J., Li, J., Xu, K., et al. (2019) Differentiation of Atypical Non-Functional Pancreatic Neuroendocrine Tumor and Pancreatic Ductal Adenocarcinoma Using CT Based Radiomics. European Journal of Radiology, 117, 102-111. [Google Scholar] [CrossRef] [PubMed]
[19] Zhang, T., Xiang, Y., Wang, H., Yun, H., Liu, Y., Wang, X., et al. (2022) Radiomics Combined with Multiple Machine Learning Algorithms in Differentiating Pancreatic Ductal Adenocarcinoma from Pancreatic Neuroendocrine Tumor: More Hands Produce a Stronger Flame. Journal of Clinical Medicine, 11, Article No. 6789. [Google Scholar] [CrossRef] [PubMed]
[20] De Robertis, R., Mascarin, B., Bardhi, E., Spoto, F., Cardobi, N. and D’Onofrio, M. (2025) Radiomics in Differential Diagnosis of Pancreatic Tumors. European Journal of Radiology Open, 14, Article ID: 100651. [Google Scholar] [CrossRef] [PubMed]
[21] Wang, R., Liu, H., Liang, P., Zhao, H., Li, L. and Gao, J. (2021) Radiomics Analysis of CT Imaging for Differentiating Gastric Neuroendocrine Carcinomas from Gastric Adenocarcinomas. European Journal of Radiology, 138, Article ID: 109662. [Google Scholar] [CrossRef] [PubMed]
[22] He, X., Yang, S., Ren, J., Wang, N., Li, M., You, Y., et al. (2024) Synergizing Traditional CT Imaging with Radiomics: A Novel Model for Preoperative Diagnosis of Gastric Neuroendocrine and Mixed Adenoneuroendocrine Carcinoma. Frontiers in Oncology, 14, Article ID: 1480466. [Google Scholar] [CrossRef] [PubMed]
[23] 赵金莉. 基于CT图像的放射组学对直肠神经内分泌肿瘤和直肠腺癌的鉴别诊断价值[D]: [硕士学位论文]. 沈阳: 中国医科大学, 2021.
[24] Rindi, G., Mete, O., Uccella, S., Basturk, O., La Rosa, S., Brosens, L.A.A., et al. (2022) Overview of the 2022 WHO Classification of Neuroendocrine Neoplasms. Endocrine Pathology, 33, 115-154. [Google Scholar] [CrossRef] [PubMed]
[25] 中国临床肿瘤学会指南工作委员会. 中国临床肿瘤学会(CSCO)神经内分泌肿瘤诊疗指南2024 [M]. 北京: 人民卫生出版社, 2024.
[26] Sorbye, H., Grande, E., Pavel, M., Tesselaar, M., Fazio, N., Reed, N.S., et al. (2023) European Neuroendocrine Tumor Society (ENETS) 2023 Guidance Paper for Digestive Neuroendocrine Carcinoma. Journal of Neuroendocrinology, 35, e13249. [Google Scholar] [CrossRef] [PubMed]
[27] Ricci, C., Mosconi, C., Ingaldi, C., Vara, G., Verna, M., Pettinari, I., et al. (2021) The 3-Dimensional-Computed Tomography Texture Is Useful to Predict Pancreatic Neuroendocrine Tumor Grading. Pancreas, 50, 1392-1399. [Google Scholar] [CrossRef] [PubMed]
[28] Liu, C., Bian, Y., Meng, Y., Liu, F., Cao, K., Zhang, H., et al. (2022) Preoperative Prediction of G1 and G2/3 Grades in Patients with Nonfunctional Pancreatic Neuroendocrine Tumors Using Multimodality Imaging. Academic Radiology, 29, e49-e60. [Google Scholar] [CrossRef] [PubMed]
[29] Wang, X., Qiu, J., Tan, C., Chen, Y., Tan, Q., Ren, S., et al. (2022) Development and Validation of a Novel Radiomics-Based Nomogram with Machine Learning to Preoperatively Predict Histologic Grade in Pancreatic Neuroendocrine Tumors. Frontiers in Oncology, 12, Article ID: 843376. [Google Scholar] [CrossRef] [PubMed]
[30] Ye, J., Fang, P., Peng, Z., Huang, X., Xie, J. and Yin, X. (2023) A Radiomics-Based Interpretable Model to Predict the Pathological Grade of Pancreatic Neuroendocrine Tumors. European Radiology, 34, 1994-2005. [Google Scholar] [CrossRef] [PubMed]
[31] Javed, A.A., Zhu, Z., Kinny-Köster, B., Habib, J.R., Kawamoto, S., Hruban, R.H., et al. (2024) Accurate Non-Invasive Grading of Nonfunctional Pancreatic Neuroendocrine Tumors with a CT Derived Radiomics Signature. Diagnostic and Interventional Imaging, 105, 33-39. [Google Scholar] [CrossRef] [PubMed]
[32] Bian, Y., Jiang, H., Ma, C., Wang, L., Zheng, J., Jin, G., et al. (2020) Ct-based Radiomics Score for Distinguishing between Grade 1 and Grade 2 Nonfunctioning Pancreatic Neuroendocrine Tumors. American Journal of Roentgenology, 215, 852-863. [Google Scholar] [CrossRef] [PubMed]
[33] Zhao, Z., Bian, Y., Jiang, H., Fang, X., Li, J., Cao, K., et al. (2020) CT-Radiomic Approach to Predict G1/2 Nonfunctional Pancreatic Neuroendocrine Tumor. Academic Radiology, 27, e272-e281. [Google Scholar] [CrossRef] [PubMed]
[34] Pulvirenti, A., Yamashita, R., Chakraborty, J., Horvat, N., Seier, K., McIntyre, C.A., et al. (2021) Quantitative Computed Tomography Image Analysis to Predict Pancreatic Neuroendocrine Tumor Grade. JCO Clinical Cancer Informatics, 5, 679-694. [Google Scholar] [CrossRef] [PubMed]
[35] Chiti, G., Grazzini, G., Flammia, F., Matteuzzi, B., Tortoli, P., Bettarini, S., et al. (2022) Gastroenteropancreatic Neuroendocrine Neoplasms (GEP-NENs): A Radiomic Model to Predict Tumor Grade. La Radiologia Medica, 127, 928-938. [Google Scholar] [CrossRef] [PubMed]
[36] 王睿, 梁盼, 余娟, 等. 基于极端梯度上升算法的联合诊断模型对胃神经内分泌肿瘤病理分级的诊断效能[J]. 中华医学杂志, 2021, 101(34): 2717-2722.
[37] Liang, P., Xu, C., Tan, F., Li, S., Chen, M., Hu, D., et al. (2020) Prediction of the World Health Organization Grade of Rectal Neuroendocrine Tumors Based on CT Histogram Analysis. Cancer Medicine, 10, 595-604. [Google Scholar] [CrossRef] [PubMed]
[38] 陈洛海, 梁赟, 陈洁. 2023年度神经内分泌肿瘤治疗研究进展[J]. 肿瘤综合治疗电子杂志, 2024, 10(2): 14-19.
[39] Gillies, R.J., Kinahan, P.E. and Hricak, H. (2016) Radiomics: Images Are More than Pictures, They Are Data. Radiology, 278, 563-577. [Google Scholar] [CrossRef] [PubMed]
[40] Gu, W., Chen, Y., Zhu, H., Chen, H., Yang, Z., Mo, S., et al. (2023) Development and Validation of CT-Based Radiomics Deep Learning Signatures to Predict Lymph Node Metastasis in Non-Functional Pancreatic Neuroendocrine Tumors: A Multicohort Study. eClinicalMedicine, 65, Article ID: 102269. [Google Scholar] [CrossRef] [PubMed]
[41] Ahmed, T.M., Zhu, Z., Yasrab, M., Blanco, A., Kawamoto, S., He, J., et al. (2024) Preoperative Prediction of Lymph Node Metastases in Nonfunctional Pancreatic Neuroendocrine Tumors Using a Combined CT Radiomics-Clinical Model. Annals of Surgical Oncology, 31, 8136-8145. [Google Scholar] [CrossRef] [PubMed]
[42] Wang, Y., Gu, W., Huang, D., Zhang, W., Chen, Y., Xu, J., et al. (2025) Evaluation of Tumor Pseudocapsule Using Computed Tomography-Based Radiomics in Pancreatic Neuroendocrine Tumors to Predict Prognosis and Guide Surgical Strategy: A Cohort Study. International Journal of Surgery, 111, 4454-4463. [Google Scholar] [CrossRef] [PubMed]
[43] Ma, M., Gu, W., Liang, Y., Han, X., Zhang, M., Xu, M., et al. (2024) A Novel Model for Predicting Postoperative Liver Metastasis in R0 Resected Pancreatic Neuroendocrine Tumors: Integrating Computational Pathology and Deep Learning-radiomics. Journal of Translational Medicine, 22, Article No. 768. [Google Scholar] [CrossRef] [PubMed]
[44] Mori, M., Palumbo, D., Muffatti, F., Partelli, S., Mushtaq, J., Andreasi, V., et al. (2022) Prediction of the Characteristics of Aggressiveness of Pancreatic Neuroendocrine Neoplasms (PanNENs) Based on CT Radiomic Features. European Radiology, 33, 4412-4421. [Google Scholar] [CrossRef] [PubMed]
[45] Blazevic, A., Starmans, M.P.A., Brabander, T., Dwarkasing, R.S., van Gils, R.A.H., Hofland, J., et al. (2021) Predicting Symptomatic Mesenteric Mass in Small Intestinal Neuroendocrine Tumors Using Radiomics. Endocrine-Related Cancer, 28, 529-539. [Google Scholar] [CrossRef] [PubMed]
[46] Rinke, A., Müller, H., Schade-Brittinger, C., Klose, K., Barth, P., Wied, M., et al. (2009) Placebo-Controlled, Double-Blind, Prospective, Randomized Study on the Effect of Octreotide LAR in the Control of Tumor Growth in Patients with Metastatic Neuroendocrine Midgut Tumors: A Report from the PROMID Study Group. Journal of Clinical Oncology, 27, 4656-4663. [Google Scholar] [CrossRef] [PubMed]
[47] Caplin, M.E., Pavel, M., Ćwikła, J.B., Phan, A.T., Raderer, M., Sedláčková, E., et al. (2014) Lanreotide in Metastatic Enteropancreatic Neuroendocrine Tumors. New England Journal of Medicine, 371, 224-233. [Google Scholar] [CrossRef] [PubMed]
[48] 中国抗癌协会神经内分泌肿瘤整合诊治指南(精简版) [J].中国肿瘤临床, 2023, 50(8): 385-397.
[49] Polici, M., Caruso, D., Masci, B., Marasco, M., Valanzuolo, D., Dell’Unto, E., et al. (2024) Radiomics in Advanced Gastroenteropancreatic Neuroendocrine Neoplasms: Identifying Responders to Somatostatin Analogs. Journal of Neuroendocrinology, 37, e13472. [Google Scholar] [CrossRef] [PubMed]
[50] Lee, L., Ito, T. and Jensen, R.T. (2018) Everolimus in the Treatment of Neuroendocrine Tumors: Efficacy, Side-Effects, Resistance, and Factors Affecting Its Place in the Treatment Sequence. Expert Opinion on Pharmacotherapy, 19, 909-928. [Google Scholar] [CrossRef] [PubMed]
[51] Caruso, D., Polici, M., Rinzivillo, M., Zerunian, M., Nacci, I., Marasco, M., et al. (2022) CT-Based Radiomics for Prediction of Therapeutic Response to Everolimus in Metastatic Neuroendocrine Tumors. La Radiologia Medica, 127, 691-701. [Google Scholar] [CrossRef] [PubMed]
[52] Chen, L., Wang, W., Jin, K., Yuan, B., Tan, H., Sun, J., et al. (2022) Special Issue “the Advance of Solid Tumor Research in China”: Prediction of Sunitinib Efficacy Using Computed Tomography in Patients with Pancreatic Neuroendocrine Tumors. International Journal of Cancer, 152, 90-99. [Google Scholar] [CrossRef] [PubMed]
[53] Alevroudis, E., Spei, M., Chatziioannou, S.N., Tsoli, M., Wallin, G., Kaltsas, G., et al. (2021) Clinical Utility of 18F-FDG PET in Neuroendocrine Tumors Prior to Peptide Receptor Radionuclide Therapy: A Systematic Review and Meta-Analysis. Cancers, 13, Article No. 1813. [Google Scholar] [CrossRef] [PubMed]
[54] Shaheen, S., Moradi, F., Gamino, G. and Kunz, P.L. (2020) Patient Selection and Toxicities of PRRT for Metastatic Neuroendocrine Tumors and Research Opportunities. Current Treatment Options in Oncology, 21, Article No. 25. [Google Scholar] [CrossRef] [PubMed]
[55] Strosberg, J., El-Haddad, G., Wolin, E., Hendifar, A., Yao, J., Chasen, B., et al. (2017) Phase 3 Trial of 177Lu-Dotatate for Midgut Neuroendocrine Tumors. New England Journal of Medicine, 376, 125-135. [Google Scholar] [CrossRef] [PubMed]
[56] Behmanesh, B., Saray, A.A., Deevband, R.M., et al. (2024) Radiomics Analysis for Clinical Decision Support in 177Lu-Dotatate Therapy of Metastatic Neuroendocrine Tumors Using CT Images. Journal of Biomedical Physics & Engineering, 14, 423-434.
[57] Manfredi, S., Walter, T., Baudin, E., Coriat, R., Ruszniewski, P., Lecomte, T., et al. (2017) Management of Gastric Neuro-Endocrine Tumours in a Large French National Cohort (GTE). Endocrine, 57, 504-511. [Google Scholar] [CrossRef] [PubMed]
[58] Homps, M., Soyer, P., Coriat, R., Dermine, S., Pellat, A., Fuks, D., et al. (2023) A Preoperative Computed Tomography Radiomics Model to Predict Disease-Free Survival in Patients with Pancreatic Neuroendocrine Tumors. European Journal of Endocrinology, 189, 476-484. [Google Scholar] [CrossRef] [PubMed]
[59] Heo, S., Park, H.J., Kim, H.J., Kim, J.H., Park, S.Y., Kim, K.W., et al. (2024) Prognostic Value of CT-Based Radiomics in Grade 1-2 Pancreatic Neuroendocrine Tumors. Cancer Imaging, 24, Article No. 28. [Google Scholar] [CrossRef] [PubMed]
[60] Yang, Z., Han, Y., Cheng, M., Wang, R., Li, J., Zhao, H., et al. (2023) Prognostic Value of Computed Tomography Radiomics Features in Patients with Gastric Neuroendocrine Neoplasm. Frontiers in Oncology, 13, Article ID: 1143291. [Google Scholar] [CrossRef] [PubMed]
[61] Yang, Z., Han, Y., Li, F., Zhang, A., Cheng, M. and Gao, J. (2023) Deep Learning Radiomics Analysis Based on Computed Tomography for Survival Prediction in Gastric Neuroendocrine Neoplasm: A Multicenter Study. Quantitative Imaging in Medicine and Surgery, 13, 8190-8203. [Google Scholar] [CrossRef] [PubMed]
[62] An, P., Zhang, J., Li, M., Duan, P., He, Z., Wang, Z., et al. (2022) Clinical Data-CT Radiomics-Based Model for Predicting Prognosis of Patients with Gastrointestinal Pancreatic Neuroendocrine Neoplasms (GP-NENs). Computational and Mathematical Methods in Medicine, 2022, Article ID: 4186305. [Google Scholar] [CrossRef] [PubMed]