分享:
分享到微信朋友圈
X
临床研究
磁共振结合临床指标对优化O-RADS MRI 4分肿块风险分层的价值
李彩红 李易 刘柳 周印 杨雅莹 毛芸

Cite this article as: LI C H, LI Y, LIU L, et al. Value of MRI combining with clinical indicators in optimizing the risk stratification of O-RADS MRI Score 4[J]. Chin J Magn Reson Imaging, 2024, 15(11): 103-109.本文引用格式:李彩红, 李易, 刘柳, 等. 磁共振结合临床指标对优化O-RADS MRI 4分肿块风险分层的价值[J]. 磁共振成像, 2024, 15(11): 103-109. DOI:10.12015/issn.1674-8034.2024.11.016.


[摘要] 目的 探讨MRI特征联合临床指标[糖类抗原125(carbohydrate antigen 125, CA125)、绝经状态、年龄]对优化卵巢-附件影像报告和数据系统(Ovarian-Adnexal Reporting and Data System, O-RADS)MRI评分4分肿块风险分层的价值,并评估能否提高O-RADS MRI评分系统诊断性能。材料与方法 回顾性纳入57例进行术前盆腔MRI增强检查并经组织病理学证实的卵巢-附件肿块患者的影像及临床资料,所有肿块的O-RADS MRI评分均为4分,由两名经验丰富的放射科医师评估得出,结果不一致时协商决定。以病理结果为金标准,对O-RADS MRI 4分肿块良、恶性组的MRI和临床指标进行差异性分析,将有差异统计学意义的指标利用分类与回归决策树(classification and regression tree, CART)构建模型,用于进一步细分4分肿块。绘制受试者工作特征(receiver operating characteristic, ROC)曲线评价决策树模型的预测准确性。分别评估优化4分肿块前后O-RADS MRI评分系统的诊断性能,并比较两者的曲线下面积(area under the curve, AUC)。计算不同经验水平阅片者之间优化后预测结果的一致性。结果 (1)57例O-RADS MRI 4分肿块中,良性22例,恶性35例。实性组织T2WI低信号常见于良性肿块(P<0.001)。乳头状突起和不规则增厚的囊壁/分隔常见于恶性肿块(P<0.001,P=0.008)。CA125>35 U/mL多见于恶性肿块(P<0.05)。决策树模型预测4分肿块良恶性的AUC为0.984(95% CI:0.908~1.000),敏感度97.1%、特异度90.9%、准确度94.7%。(2)应用决策树模型对4分肿块优化后,在整个人群中,O-RADS MRI评分系统的AUC从0.838提升至0.945(P<0.001);在绝经前妇女中,AUC从0.818提升至0.934(P<0.001);在绝经后妇女中,AUC从0.871提升至0.962(P=0.008)。不同经验水平医师之间的优化后预测结果一致性极好(Kappa值分别为0.887、0.869)。结论 在本研究中,基于实性组织MRI特征联合临床指标CA125开发的预测模型有助于优化O-RADS MRI 4分肿块风险分层,并显著提高了O-RADS MRI评分系统的诊断效能,尤其在绝经前女性中。
[Abstract] Objective To investigate the value of MRI characteristics combined with clinical indicators [carbohydrate antigen 125 (CA125), menopausal status, age] in optimizing the Ovarian-Adnexal Reporting and Data System (O-RADS) MRI score 4 mass risk stratification and whether it can improve the diagnostic performance of the O-RADS MRI scoring system.Materials and Methods Totally 57 ovarian adnexal masses scored 4 according to O-RADS MRI were retrospectively analyzed. All masses underwent preoperative pelvic MRI enhancement imaging and were confirmed by histopathology. They were evaluated by two experienced radiologists and determined through consultation when the results were inconsistent. The pathological results were used as the gold standard to analyze the differences of MRI and clinical indicators in the O-RADS MRI score 4 group of benign and malignant masses. The classification and regression tree (CART) was employed to construct a model for statistically significant indicators for the further subdivision of the O-RADS MRI 4 mass. Receiver operating characteristic (ROC) analysis was used to evaluate the prediction accuracy of the decision tree model. To evaluate the diagnostic effect of O-RADS MRI scoring system before and after O-RADS MRI score 4 mass optimization, and compare the difference of area under the curve (AUC). The consistency of the optimized prediction results among different viewers was calculated.Results (1) Among 57 O-RADS MRI score 4 masses, 22 masses were benign, and 35 masses were malignant. Solid tissue showed hypointense on T2WI was more common in benign mass well (P<0.001). Papillary projections and irregularly thickened cyst wall or septations were more frequent in malignant mass (P<0.001, P=0.008). The CA125 level in malignant mass was often greater than 35 U/mL (P<0.05). The AUC of the decision tree model for predicting benign and malignant tumors was 0.984 (95% CI: 0.908-1.000), with a sensitivity of 97.1%, specificity of 90.9% and accuracy of 94.7%. (2) The AUC of the O-RADS MRI scoring system increased from 0.838 to 0.945 (P<0.001) in the whole population after optimizing the O-RADS MRI 4 mass with the decision tree model; In premenopausal women, the AUC increased from 0.818 to 0.934 (P<0.001). In postmenopausal women, the AUC increased from 0.871 to 0.962 (P=0.008). There was excellent agreement between the optimized predictions among physicians with different experience levels (Kappa=0.887, 0.869).Conclusion In this study, a predictive model developed based on solid tissue MRI features combined with CA125 levels helped optimize the risk stratification of O-RADS MRI score 4 mass and significantly improved the diagnostic performance of the O-RADS MRI scoring system, especially in premenopausal women.
[关键词] 卵巢-附件肿块;卵巢-附件报告和数据系统;良恶性病变;鉴别诊断;磁共振成像
[Keywords] ovarian-adnexal mass;Ovarian-Adnexal Reporting and Data System;benign and malignant lesions;differential diagnosis;magnetic resonance imaging

李彩红 1   李易 1   刘柳 1   周印 1   杨雅莹 2, 3, 4, 5   毛芸 1*  

1 重庆医科大学附属第一医院放射科,重庆 400016

2 重庆医科大学基础医学院病理学教研室,重庆 400016

3 重庆医科大学病理诊断中心,重庆 400016

4 重庆医科大学临床病理研究室,重庆 400016

5 重庆医科大学附属第一医院病理科,重庆 400016

通信作者:毛芸,E-mail: maoyun1979@163.com

作者贡献声明:毛芸负责设计本研究的方案,并对稿件重要内容进行了修改;李彩红起草并撰写稿件,分析和解释本研究的数据;李易、刘柳、杨雅莹、周印参与获取本研究的数据,对稿件重要内容进行了修改;全体作者一致同意发表最终的修改稿,同意对本研究的所有方面负责,并确保本研究的准确性和诚信。


收稿日期:2024-08-02
接受日期:2024-11-10
中图分类号:R445.2  R737.31 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2024.11.016
本文引用格式:李彩红, 李易, 刘柳, 等. 磁共振结合临床指标对优化O-RADS MRI 4分肿块风险分层的价值[J]. 磁共振成像, 2024, 15(11): 103-109. DOI:10.12015/issn.1674-8034.2024.11.016.

0 引言

       卵巢癌是妇科恶性肿瘤中死亡率仅次于宫颈癌的第二大肿瘤,晚期患者五年生存率低于50%,而及时接受规范化治疗的早期患者5年生存率可超过90%[1, 2, 3]。卵巢-附件肿块术前良恶性鉴别对实现精准个性化治疗及预后管理至关重要。对恶性肿瘤通常采取全面手术切除并进行分期,必要时需辅以放疗或化疗等[4]。良性肿块患者则应避免接受不必要或不适当的广泛手术治疗,尤其是绝经前女性应优先考虑保留生育能力的治疗方式,以降低对年轻女性生育能力的影响和医疗资源的浪费[5, 6, 7]

       影像学检查是术前评估卵巢-附件肿块良恶性的关键手段[8, 9],超声作为首选方法,2020年美国放射学会(American College of Radiology, ACR)发布了基于超声特征的卵巢-附件影像报告和数据系统(Ovarian-Adnexal Reporting and Data System, O-RADS)超声风险分层和管理共识指南并已得到广泛认可[10]。但仍至少有18%~31%的肿块超声无法确定[11]。MRI在识别肿块来源和成分方面展现出显著优势,能够为超声不确定肿块提供更准确的诊断[12, 13, 14]。THOMASSIN-NAGGARA等[15]基于卵巢-附件肿块的MRI形态、信号及功能学特征开发了O-RADS MRI风险分层评分,采用5分制来构建超声不确定肿块的恶性肿瘤风险分层,评分≥4分定义为恶性肿块。然而,O-RADS MRI 4分肿块仍存在较高不确定性,恶性肿瘤阳性预测值约50%[11, 15, 16]。同时,O-RADS MRI 4分在低至高风险人群(3~5分)中占比约25.76%~26.07%[11, 16]。我们推测可能源于以下原因:首先,O-RADS MRI 4分仅根据“实性组织强化程度”单一特征来定义,但实际情况中,良、恶性肿块的强化程度存在重叠;其次,O-RADS MRI评分系统尚缺乏临床指标,导致评估不全面。因此,增加额外的指标对O-RADS MRI 4分进一步细分具有重要临床意义。

       目前,针对O-RADS MRI 4分评分的不足,国内外尚缺乏相关研究。因此,本研究的目标是针对O-RADS MRI 4分肿块开发更精准的诊断模型。我们将纳入额外的MRI特征,并与临床指标[糖类抗原125(carbohydrate antigen 125, CA125)、绝经状态和年龄]相结合,以探索这种综合评估方法是否有助于优化O-RADS MRI 4分肿块的风险分层。从而为临床医生制订精准个性化治疗方案提供依据,以提高生存率和患者获益。

1 材料与方法

1.1 研究对象

       本回顾性临床研究遵守《赫尔辛基宣言》,经重庆医科大学附属第一医院伦理委员会批准,免除受试者知情同意,批准文号:2024-105-01。收集2014年至2023年于重庆医科大学附属第一医院行盆腔MRI增强检查并经手术病理证实的卵巢-附件肿块患者的影像及临床资料,并记录患者年龄、绝经状态(绝经定义:末次月经后1年或达绝经年龄)、血清CA125水平。纳入标准:(1)经手术切除且病理证实的卵巢-附件肿块;(2)术前未行放射治疗或化学治疗;(3)术前一周进行MRI增强检查和血清CA125检查。排除标准:(1)病理诊断不明确;(2)怀孕或伴急性症状;(3)MRI图像缺失的,以及无法进行O-RADS评分的;(4)O-RADS MRI 1~2分。具体纳入排除数据流程如图1所示。

图1  研究对象纳入排除流程图。CA125:糖类抗原125;O-RADS:卵巢-附件影像报告和数据系统。
Fig. 1  Inclusion and exclusion flowchart of study population. CA125: carbohydrate antigen 125; O-RADS: Ovarian-Adnexal Reporting and Data System.

1.2 研究方法

1.2.1 MRI扫描方案

       采用GE 3.0 T磁共振扫描仪(Singa HD Excite, GE Healthcare, USA),8通道腹部相控阵线圈,仰卧位固定。扫描序列及主要参数:(1)轴位T1WI,SE序列,TR 560~760 ms,TE 10~14 ms,视野(FOV)200 mm×240 mm,矩阵384×180,层厚6 mm。(2)轴位、冠状位及矢状位T2WI,FRFSE序列,TR 2500~3500 ms,TE 100~105 ms,FOV 220 mm×240 mm,矩阵320×224,层厚4 mm。(3)轴位扩散加权成像(diffusion weighted imaging, DWI),SSEPI序列,TR 3200~4100 ms,TE 83~88 ms,矩阵320×256,层厚5 mm,b值取0和1000 s/mm2。(4)增强扫描,LAVA序列,TR 4.0 ms,TE 1.9 ms,矩阵320×320,FOV 220 mm×240 mm,层厚3 mm,扫描5期轴位图像,扫描时间17 s/期,延迟期扫描矢状位及冠状位,静脉注射对比剂采用马根维显(Ga-DTPA, Bayer Pharma AG, Germany),注射剂量为0.2 mL/kg,流速2~3 mL/s。

1.2.2 CA125检测方法

       采集患者空腹静脉血3~5 mL,使用罗氏Cobas e602全自动电化学发光免疫分析仪,检测血清CA125水平,CA125试剂盒购自罗氏诊断产品(苏州)有限公司,CA125正常参考区间0~35 U/mL。

1.2.3 MRI图像分析

       两名从事腹部MRI诊断且经验丰富的放射科医师A(19年经验)和B(10年经验)在不知道病理及临床表现的情况下,依据ACR发布的O-RADS MRI评分指南[17]对每个卵巢-附件肿块MRI图像进行独立盲法评估,筛出O-RADS MRI 3~5分肿块。结果不一致时,回顾MRI图像协商得出最终评分。根据2020年ACR O-RADS MRI委员会发布的白皮书中的术语及其定义对卵巢-附件肿块进行描述[18],重点记录O-RADS MRI 4分肿块的以下特征:(1)实性组织类型乳头状突起(定义为囊壁或分隔上强化的实性组织,呈分枝状结构外观)、壁结节(定义为囊壁或分隔上>3 mm的强化实性组织,呈结节状外观)、不规则增厚囊壁/分隔、较大部分实性组织(定义为在形态上不呈现为“乳头状突起、壁结节、不规则增厚囊壁/分隔”的强化实性组织)、实性组织≥80%;(2)实性组织T2WI及DWI信号;(3)液体类型 单纯(与脑脊液/尿液信号一致)、非单纯。然后,由医师A、B和一名初级医师C(3年经验)分别在O-RADS MRI评分的基础上,针对O-RADS MRI 4分肿块使用决策树模型再次评估。本研究MRI扫描未采集灌注加权图像,无法绘制时间-强度曲线,故按照指南要求,对强化的实现组织按照注射对比剂后30~40 s的强化程度评为O-RADS MRI 4分和5分。O-RADS MRI评分≥4分为恶性。

1.2.4 参考标准

       以术后病理诊断结果作为临床诊断的金标准。将患者分为良性组和恶性组,术后病理证实为低度恶性及交界性时,归为恶性肿瘤范畴。

1.3 统计分析

       采用SPSS 27.0和MedCalc 22.018软件进行统计学分析。符合正态分布的计量资料以均数±标准差表示,计数资料以例(%)表示。以病理结果为金标准,采用卡方检验或Fisher精确检验比较O-RADS MRI 4分肿块良、恶性组间MRI及临床指标的差异性。差异有统计学意义的指标作为自变量,4分肿块良恶性分类为因变量,采用分类与回归决策树(classification and regression tree, CART)算法构建决策树模型;应用基尼系数最小化方法,评估自变量对模型预测能力的贡献,通过正态化重要性来量化各变量对模型预测能力的贡献度,以此确定变量优先级,确保重要特征优先分枝;最大深度设定为5,父节点最小个案数为10,子节点最小个案数为5。绘制受试者工作特征(receiver operating characteristic, ROC)曲线并计算曲线下面积(area under the curve, AUC)评估决策树模型预测准确性。分别评估优化4分肿块前后O-RADS MRI评分系统的诊断性能,采用DeLong检验比较两者AUC。使用Kappa检验,计算3名不同经验水平医生之间优化后预测结果的一致性,当Kappa≤0.2为一致性差,0.21<Kappa≤0.4为一致性一般,0.4<Kappa≤0.6为一致性中等,0.6<Kappa≤0.8为一致性较好,0.8<Kappa≤1.0为一致性极好[19]P<0.05表示差异有统计学意义。

2 结果

2.1 一般资料及病理

       本研究共纳入57例O-RADS MRI 4分肿块,其中良性组22例(38.6%),恶性组35例(61.4%)。实性组织T2WI信号(P<0.001)、乳头状突起(P<0.001)、不规则增厚囊壁/分隔(P=0.008)、较大部分实性组织(P=0.004)、实性组织≥80%(P<0.001)、CA125水平(P=0.004)在良、恶性组间差异有统计学意义(表1)。

       病理结果分布如下:良性22例,包括浆液性囊腺纤维瘤2例、浆液性腺纤维瘤1例、子宫内膜异位囊肿1例、卵泡膜细胞瘤2例、纤维-卵泡膜细胞瘤4例、阔韧带平滑肌瘤12例;恶性35例,包括交界性浆液性囊腺瘤4例、交界性黏液性囊腺瘤4例、交界性Brenner瘤1例、浆液性囊腺癌7例、黏液性囊腺癌1例、子宫内膜样癌1例、输卵管浆液性腺癌1例、透明细胞癌4例、颗粒细胞瘤5例、莱迪细胞瘤1例、卵黄囊瘤2例、无性细胞瘤3例、阔韧带纤维肉瘤1例。

表1  MRI特征及临床指标在O-RADS MRI 4分肿块良恶性组间差异
Tab. 1  Comparison of MRI features and clinical indicators between benign and malignant groups of O-RADS MRI 4 masses

2.2 决策树模型建立

       决策树的根节点为57例O-RADS MRI 4分肿块,决策树深度为4,叶节点个数为5;根据变量对模型预测能力的贡献度确定优先级,最后进入CART的变量及其正态化重要性分别为:实性组织T2WI信号(100.0%)、乳头状突起(43.0%)、不规则增厚囊壁/分隔(36.6%)、CA125水平(36.5%)。CART决策树中,左支表示良性肿块,其余表示恶性肿块。

       决策树模型(图2)表明,若实性组织T2WI呈低信号,可预测为良性。若实性组织T2WI呈非低信号,则需要进一步观察是否存在乳头状突起或不规则增厚囊壁/分隔。若存在这些特征,可预测为恶性。若既不存在乳头状突起或不规则增厚囊壁/分隔,并且T2WI上呈非低信号,则需查看CA125水平。若≤35 U/mL预测为良性,否则即为恶性。代表性MRI图像如图3, 4, 5所示。

       level: 12.9 U/mL, postoperative pathology: broad ligament leiomyoma of left ovary with cystic change. 3A: Axial T2-weighted imaging shows low signal solid tissue; 3B: Solid tissue is hyperintense on DWI; 3C: On post-contrast MRI, solid tissue enhances but less than the outer myometrium. Fig. 4 A 23-year-old premenopausal female with O-RADS MRI 4 adnexal mass demonstrating papillary projection confirmed as malignant, CA125 level: 959 U/mL, postoperative pathology: borderline serous cystadenoma of right ovary. 4A: Axial T2-weighted imaging shows T2 isointense papillary projection; 4B: Papillary projection is hyperintense on diffusion-weighted imaging; 4C: On post-contrast MRI, papillary projection enhances avidly but slightly less than the outer myometrium. Fig. 5 A 53-year-old postmenopausal female with O-RADS MRI 4 adnexal mass demonstrating irregular thickening of the sac wall and septum confirmed as malignant, CA125 level: 16.5 U/mL, postoperative pathology: adult granulosa cell tumor of the right ovary. 5A: Sagittal T2-weighted imaging shows a multilocular cystic mass with irregularly thickened cyst wall and septa, mixed intracapsular signals with liquid-liquid plane; 5B: solid tissue is hyperintense on diffusion-weighted imaging; 5C: On post-contrast MRI, irregular thickening of the sac wall and septum enhances avidly but slightly less than the outer myometrium. O-RADS: Ovarian-Adnexal Reporting and Data System; CA125: carbohydrate antigen 125; DWI: diffusion weighted imaging.

图2  O-RADS MRI 4分肿块良恶性预测CART算法树形图。O-RADS:卵巢-附件影像报告和数据系统;CA125:糖类抗原125;CART:分类与回归决策树。
Fig. 2  O-RADS MRI 4 masses benign and malignant prediction CART algorithm tree diagram. O-RADS: Ovarian-Adnexal Reporting and Data System; CA125: carbohydrate antigen 125; CART: classification and regression tree.
图3  女,42岁,绝经前,O-RADS MRI 4分,显示为实性组织T2WI低信号,CA125水平:12.9 U/mL,术后病理:左侧阔韧带平滑肌瘤伴囊性变。3A:轴位T2WI示病灶实性组织呈低信号;3B:DWI示实性组织弥散受限;3C:MRI增强示实性组织强化低于子宫外肌层。
图4  女,23岁,绝经前,O-RADS MRI 4分,显示为乳头状突起,CA125水平:959 U/mL,术后病理:右卵巢交界性浆液性囊腺瘤。4A:轴位T2WI示等信号的乳头状突起;4B:DWI示乳头状突起弥散受限;4C:MRI增强示乳头状突起明显强化,但略低于子宫外肌层。
图5  女,53岁,绝经后,O-RADS MRI 4分,显示为不规则增厚囊壁/分隔,CA125水平:16.5 U/mL,术后病理:右侧卵巢成年型颗粒细胞瘤。5A:矢状位T2WI示多房囊性肿块伴不规则增厚的囊壁及分隔,囊腔内信号混杂伴液-液平面;5B:DWI示实性组织弥散受限;5C:MRI增强示不规则增厚的囊壁及分隔明显强化,强化略低于子宫外肌层。O-RADS:卵巢-附件影像报告和数据系统;CA125:糖类抗原125;DWI:扩散加权成像。
Fig. 3  A 42-year-old premenopausal female with O-RADS MRI 4 adnexal mass demonstrating low signal solid tissue on T2WI confirmed as benign, CA125 level: 12.9 U/mL, postoperative pathology: broad ligament leiomyoma of left ovary with cystic change. 3A: Axial T2-weighted imaging shows low signal solid tissue; 3B: Solid tissue is hyperintense on DWI; 3C: On post-contrast MRI, solid tissue enhances but less than the outer myometrium.
Fig. 4  A 23-year-old premenopausal female with O-RADS MRI 4 adnexal mass demonstrating papillary projection confirmed as malignant, CA125 level: 959 U/mL, postoperative pathology: borderline serous cystadenoma of right ovary. 4A: Axial T2-weighted imaging shows T2 isointense papillary projection; 4B: Papillary projection is hyperintense on diffusion-weighted imaging; 4C: On post-contrast MRI, papillary projection enhances avidly but slightly less than the outer myometrium.
Fig. 5  A 53-year-old postmenopausal female with O-RADS MRI 4 adnexal mass demonstrating irregular thickening of the sac wall and septum confirmed as malignant, CA125 level: 16.5 U/mL, postoperative pathology: adult granulosa cell tumor of the right ovary. 5A: Sagittal T2-weighted imaging shows a multilocular cystic mass with irregularly thickened cyst wall and septa, mixed intracapsular signals with liquid-liquid plane; 5B: solid tissue is hyperintense on diffusion-weighted imaging; 5C: On post-contrast MRI, irregular thickening of the sac wall and septum enhances avidly but slightly less than the outer myometrium. O-RADS: Ovarian-Adnexal Reporting and Data System; CA125: carbohydrate antigen 125; DWI: diffusion weighted imaging.

2.3 诊断性能比较

       决策树模型预测O-RADS MRI 4分肿块恶性风险的ROC曲线(图6)显示,AUC为0.984(95% CI:0.908~1.000),敏感度为97.1%,特异度为90.9%,准确度为94.7%。应用决策树模型对4分肿块优化后,在整个人群中,O-RADS MRI评分系统的AUC从0.838提升至0.945,差异具有统计学意义(P<0.001);优化前后O-RADS MRI评分的敏感度、特异度、阳性预测值、阴性预测值、准确度分别为:97.9% vs. 97.2%、69.8% vs. 91.9%、84.4% vs. 95.2%、95.2% vs. 95.2%、87.4% vs. 95.2%。在不同绝经状态人群中,优化后的O-RADS MRI评分系统的AUC均高于优化前,且差异均有统计学意义(表2)。

图6  CART模型ROC曲线。CART:分类与回归决策树;ROC:受试者工作特征;AUC:曲线下面积。
Fig. 6  ROC curve of CART model. CART: classification and regression tree; ROC: receiver operating characteristic; AUC: area under the curve.
表2  优化前后的O-RADS MRI评分在不同人群中的诊断性能
Tab. 2  Diagnostic performance of O-RADS MRI score before and after optimization in different populations

2.4 不同阅片者之间优化后预测结果的一致性评估

       三位医生之间优化后预测结果的一致性评估结果显示,有经验的医师A、B与初级医师C之间一致性均为极好,Kappa值分别为0.887、0.869。

3 讨论

       本研究采用MRI特征联合临床指标对4分肿块进行评估,发现基于实性组织T2WI信号、乳头状突起、不规则增厚囊壁/分隔、CA125水平构建的决策树模型可以很好地预测4分肿块恶性风险。在应用决策树模型对4分肿块优化后,O-RADS MRI评分系统的AUC从0.838提升至0.945,恶性肿块阳性预测值从84.4%提升至95.2%,同时阴性预测值稳定为95.2%。

3.1 MRI特征在鉴别O-RADS MRI 4分肿块良恶性的价值

       MRI形态及信号特征为更好地鉴别卵巢-附件良恶性肿块提供了关键线索,在O-RADS MRI 4分肿块中也是如此。我们的决策模型中纳入的MRI特征在以前也都被用作卵巢-附件肿块良恶性鉴别的重要指标[15, 17, 20]。首先,本研究发现,实性组织呈T2WI低信号是预测良性肿块的重要指标。正如既往研究所示,实性组织呈T2WI低信号常见于纤维瘤、纤维-卵泡膜细胞瘤、(囊)腺纤维瘤、阔韧带平滑肌瘤、良性Brenner瘤[20, 21, 22]。在ACR发布的O-RADS MRI评分指南中,使用T2WI低信号和DWI低信号(b=1000 s/mm2)的实性组织来定义由纤维组织类病变,O-RADS MRI评分为2[17]。虽然,实性组织“双低信号”特征在预测良性肿块有很高的特异性,但值得注意的是,表现为“双低信号”的典型纤维组织病变仅占少数。我们通过“实性组织T2WI低信号”对4分肿块中非典型表现的纤维组织类病变和阔韧带平滑肌瘤等进行定义,最终成功识别了50%(11/22)的良性肿块。同时,本研究还发现,在实性组织T2WI非低信号肿块中,“乳头状突起”或“不规则增厚的囊壁/分隔”的存在,可以很好地预测恶性肿瘤。乳头状突起常见于恶性上皮性肿瘤,尤其在交界性肿瘤中[23, 24, 25]。THOMASSIN-NAGGARA等[15]和BERNARDIN等[26]的研究均表明不规则增厚囊壁/分隔常提示恶性肿瘤(恶性阳性似然比为9.83)。虽然部分囊腺纤维瘤也可表现为不规则增厚囊壁/分隔,但其T2WI常呈低信号[22, 27]。本研究中,存在“乳头状突起”或“不规则增厚囊壁/分隔”且实性组织呈T2WI非信号的27个肿块均为恶性肿瘤。因此,本研究采用了一种分步骤的评估策略,首先对O-RADS MRI 4分肿块的实性组织T2WI信号进行评估,随后进一步评估是否存在“乳头状突起”或“不规则的囊壁/分隔”,将显著提升诊断的准确性。

       此外,我们注意到O-RADS MRI 4分肿块中,具有“较大部分实性组织”或“实性组织≥80%”的肿块虽然更常见于良性病变,但恶性肿瘤中也有相当比例(分别为45.5%和32.0%)表现出这些特征,因此这些指标在区分4分肿块的良恶性上并不十分可靠。另外,我们还观察到DWI信号强度对于鉴别O-RADS MRI 4分肿块良恶性没有帮助。尽管既往有研究表明DWI有助于区分良性与恶性肿块,且在O-RADS MRI 4分肿块中设定表观扩散系数(apparent diffusion coefficient, ADC)阈值可能有助于区分良恶性[28, 29, 30]。但KIM等[31]的一项荟萃分析显示,由于卵巢肿瘤的多样性和复杂性,很多良性肿瘤也可表现为扩散受限,且不同研究中DWI缺乏标准化,难以确定ADC值临界值,所以这在实践中并不具可行性。

3.2 临床指标在鉴别O-RADS MRI 4分肿块良恶性的价值

       与文献结果一致,我们发现CA125水平升高常提示恶性肿瘤。CA125是临床广泛认可的卵巢癌血清肿瘤标志物,但CA125单独用于鉴别卵巢-附件肿块良恶性的效果并不理想[32]。在卵巢的某些良性病变如子宫内膜异位症和良性病变急性事件以及生理条件下也出现阳性表达[33, 34]。因此,本研究综合运用MRI特征和CA125水平对O-RADS MRI 4分的卵巢-附件肿块进行良恶性预测,先利用MRI特征进行初步评估,对MRI难以区分的肿块再结合CA125水平进行判断。研究结果显示,该方法对4分肿块良恶性鉴别具有高敏感度(97.1%)和特异度(90.9%),AUC达到0.984。

       此外,年龄是卵巢癌重要的独立危险因素,特别是在绝经后,患病率显著上升[35]。然而,在本研究中,我们观察到4分肿块良恶性组均以绝经前占比较多,在绝经前妇女中优化后的O-RADS MRI评分系统AUC提升幅度也显著高于绝经后妇女。这是因为,我们的研究结果来自特定的O-RADS MRI 4分肿块人群。在O-RADS MRI 4分中,良性肿瘤以阔韧带平滑肌瘤/纤维类肿瘤为主,恶性肿瘤以交界性肿瘤为主,这些肿瘤通常好发于绝经前女性[36, 37, 38]

3.3 本研究的局限性

       我们的研究存在一定局限性:第一,单中心回顾性研究,存在一定选择偏差;第二,本研究中4分肿块样本量相对较小(n=57),未能设置验证组,未来需要行前瞻性大样本的外部验证;第三,本研究的资料均为非动态对比增强(dynamic contrast-enhanced, DCE)MRI增强扫描图像,无法与DCE MRI方案的诊断效能行比较分析,在未来进行具体的比较研究将是很有意义的。

4 结论

       综上所述,基于实性组织MRI特征和CA125水平构建的决策树模型在优化O-RADS MRI 4分肿块风险分层方面具有重要价值,可能有助于优化患者的临床决策。

[1]
SIEGEL R, GIAQUINTO A N, JEMAL A. Cancer statistics, 2024[J]. CA A Cancer J Clin, 2024, 74: 12-49. DOI: 10.3322/caac.21820.
[2]
WEBB P M, JORDAN S J. Global epidemiology of epithelial ovarian cancer[J]. Nat Rev Clin Oncol, 2024, 21(5): 389-400. DOI: 10.1038/s41571-024-00881-3.
[3]
BRAY F, LAVERSANNE M, SUNG H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2024, 74(3): 229-263. DOI: 10.3322/caac.21834.
[4]
GONZÁLEZ-MARTÍN A, HARTER P, LEARY A, et al. Newly diagnosed and relapsed epithelial ovarian cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up[J]. Ann Oncol, 2023, 34(10): 833-848. DOI: 10.1016/j.annonc.2023.07.011.
[5]
MAY T, OZA A. Conservative management of adnexal masses[J]. Lancet Oncol, 2019, 20(3): 326-327. DOI: 10.1016/S1470-2045(18)30939-2.
[6]
WOLFMAN W, THURSTON J, YEUNG G, et al. Guideline No. 404: initial investigation and management of benign ovarian masses[J/OL]. J Obstet Gynaecol Can, 2020, 42(8): 1040-1050.e1 [2024-06-05]. https://pubmed.ncbi.nlm.nih.gov/32736855/. DOI: 10.1016/j.jogc.2020.01.014.
[7]
POPLAWSKI R, MA K. Benign ovarian cysts in premenopausal women[J]. Obstet Gynaecol Reprod Med, 2022, 32(10): 234-239. DOI: 10.1016/j.ogrm.2022.08.003.
[8]
TIMMERMAN D, PLANCHAMP F, BOURNE T, et al. ESGO/ISUOG/IOTA/ESGE Consensus Statement on pre-operative diagnosis of ovarian tumors[J]. Int J Gynecol Cancer, 2021, 31(7): 961-982. DOI: 10.1136/ijgc-2021-002565.
[9]
ROSELAND M E, MATUREN K E, SHAMPAIN K L, et al. Adnexal mass imaging: contemporary guidelines for clinical practice[J]. Radiol Clin North Am, 2023, 61(4): 671-685. DOI: 10.1016/j.rcl.2023.02.002.
[10]
ANDREOTTI R F, TIMMERMAN D, STRACHOWSKI L M, et al. O-RADS US risk stratification and management system: a consensus guideline from the ACR ovarian-adnexal reporting and data system committee[J]. Radiology, 2020, 294(1): 168-185. DOI: 10.1148/radiol.2019191150.
[11]
THOMASSIN-NAGGARA I, PONCELET E, JALAGUIER-COUDRAY A, et al. Ovarian-adnexal reporting data system magnetic resonance imaging (O-RADS MRI) score for risk stratification of sonographically indeterminate adnexal masses[J/OL]. JAMA Netw Open, 2020, 3(1): e1919896 [2024-06-05]. https://pubmed.ncbi.nlm.nih.gov/31977064/. DOI: 10.1001/jamanetworkopen.2019.19896.
[12]
SADOWSKI E A, STEIN E B, THOMASSIN-NAGGARA I, et al. O-RADS MRI after initial ultrasound for adnexal lesions: AJR expert panel narrative review[J]. AJR Am J Roentgenol, 2023, 220(1): 6-15. DOI: 10.2214/AJR.22.28084.
[13]
LEE S I, KANG S K. MRI improves the characterization of incidental adnexal masses detected at sonography[J/OL]. Radiology, 2023, 307(1): e222866 [2024-04-27]. https://pubmed.ncbi.nlm.nih.gov/36413134/. DOI: 10.1148/radiol.222866.
[14]
STRACHOWSKI L M, JHA P, PHILLIPS C H, et al. O-RADS US v2022: an update from the American college of radiology's ovarian-adnexal reporting and data system US committee[J/OL]. Radiology, 2023, 308(3): e230685 [2024-06-10]. https://pubmed.ncbi.nlm.nih.gov/37698472/. DOI: 10.1148/radiol.230685.
[15]
THOMASSIN-NAGGARA I, AUBERT E, ROCKALL A, et al. Adnexal masses: development and preliminary validation of an MR imaging scoring system[J]. Radiology, 2013, 267(2): 432-443. DOI: 10.1148/radiol.13121161.
[16]
RIZZO S, COZZI A, DOLCIAMI M, et al. O-RADS MRI: a systematic review and meta-analysis of diagnostic performance and category-wise malignancy rates[J/OL]. Radiology, 2023, 307(1): e220795 [2024-06-10]. https://pubmed.ncbi.nlm.nih.gov/36413127/. DOI: 10.1148/radiol.220795.
[17]
SADOWSKI E A, THOMASSIN-NAGGARA I, ROCKALL A, et al. O-RADS MRI risk stratification system: guide for assessing adnexal lesions from the ACR O-RADS committee[J]. Radiology, 2022, 303(1): 35-47. DOI: 10.1148/radiol.204371.
[18]
REINHOLD C, ROCKALL A, SADOWSKI E A, et al. Ovarian-adnexal reporting lexicon for MRI: a white paper of the ACR ovarian-adnexal reporting and data systems MRI committee[J]. J Am Coll Radiol, 2021, 18(5): 713-729. DOI: 10.1016/j.jacr.2020.12.022.
[19]
夏邦世, 吴金华. Kappa一致性检验在检验医学研究中的应用[J]. 中华检验医学杂志, 2006, 29(1): 83-84. DOI: 10.3760/j:issn:1009-9158.2006.01.030.
XIA B S, WU J H. Application of Kappa consistency test in laboratory medicine research[J]. Chin J Lab Med, 2006, 29(1): 83-84. DOI: 10.3760/j:issn:1009-9158.2006.01.030.
[20]
WILSON M P, KATLARIWALA P, LOW G. Solid hypoechoic adnexal lesions with acoustic shadowing warrant an MRI recommendation in the O-RADS risk stratification and management system[J/OL]. Radiology, 2020, 296(1): E11-E13 [2024-06-10]. https://pubmed.ncbi.nlm.nih.gov/32315270/. DOI: 10.1148/radiol.2020200437.
[21]
PARK S BIN. Features of the hypointense solid lesions in the female pelvis on T2-weighted MRI[J]. J Magn Reson Imaging, 2014, 39(3): 493-503. DOI: 10.1002/jmri.24512.
[22]
AVESANI G, ELIA L, ANGHELONE A G, et al. Features of cystadenofibroma on magnetic resonance imaging: an update using the O-RADS lexicon and considering diffusion-weighted and perfusion imaging[J/OL]. Eur J Radiol, 2022, 154: 110429 [2024-06-10]. https://pubmed.ncbi.nlm.nih.gov/35797789/. DOI: 10.1016/j.ejrad.2022.110429.
[23]
FLICEK K T, VANBUREN W, DUDIAK K, et al. Borderline epithelial ovarian tumors: what the radiologist should know[J]. Abdom Radiol, 2021, 46(6): 2350-2366. DOI: 10.1007/s00261-020-02688-z.
[24]
TAYLOR E C, IRSHAID L, MATHUR M. Multimodality imaging approach to ovarian neoplasms with pathologic correlation[J]. Radiographics, 2021, 41(1): 289-315. DOI: 10.1148/rg.2021200086.
[25]
VALENTIN L, AMEYE L, SAVELLI L, et al. Unilocular adnexal cysts with papillary projections but no other solid components: is there a diagnostic method that can classify them reliably as benign or malignant before surgery?[J]. Ultrasound Obstet Gynecol, 2013, 41(5): 570-581. DOI: 10.1002/uog.12294.
[26]
BERNARDIN L, DILKS P, LIYANAGE S, et al. Effectiveness of semi-quantitative multiphase dynamic contrast-enhanced MRI as a predictor of malignancy in complex adnexal masses: radiological and pathological correlation[J]. Eur Radiol, 2012, 22(4): 880-890. DOI: 10.1007/s00330-011-2331-z.
[27]
WONG B Z Y, CAUSA ANDRIEU P I, SONODA Y, et al. Improving risk stratification of indeterminate adnexal masses on MRI: what imaging features help predict malignancy in O-RADS MRI 4 lesions?[J/OL]. Eur J Radiol, 2023, 168: 111122 [2024-06-10]. https://pubmed.ncbi.nlm.nih.gov/37806193/. DOI: 10.1016/j.ejrad.2023.111122.
[28]
HOTTAT N A, VAN PACHTERBEKE C, VANDEN HOUTE K, et al. Magnetic resonance scoring system for assessment of adnexal masses: added value of diffusion-weighted imaging including apparent diffusion coefficient map[J]. Ultrasound Obstet Gynecol, 2021, 57(3): 478-487. DOI: 10.1002/uog.22090.
[29]
雷岩, 宋彬, 刘恋恋, 等. 磁共振多参数定量分析在区分卵巢-附件O-RADS MRI4分病变良恶性的价值[J]. 中国医学计算机成像杂志, 2023, 29(1): 50-57. DOI: 10.3969/j.issn.1006-5741.2023.01.012.
LEI Y, SONG B, LIU L L, et al. Value of quantitative analysis of multiparameter MRI in differentiation of benign and malignant ovarian lesions with O-RADS MRI score 4[J]. Chin Comput Med Imag, 2023, 29(1): 50-57. DOI: 10.3969/j.issn.1006-5741.2023.01.012.
[30]
HOTTAT N A, BADR D A, VAN PACHTERBEKE C, et al. Added value of quantitative analysis of diffusion-weighted imaging in ovarian-adnexal reporting and data system magnetic resonance imaging[J]. J Magn Reson Imaging, 2022, 56(1): 158-170. DOI: 10.1002/jmri.28003.
[31]
KIM H J, LEE S Y, SHIN Y R, et al. The value of diffusion-weighted imaging in the differential diagnosis of ovarian lesions: a meta-analysis[J/OL]. PLoS One, 2016, 11(2): e0149465 [2024-06-12]. https://pubmed.ncbi.nlm.nih.gov/26907919/. DOI: 10.1371/journal.pone.0149465.
[32]
ZHANG M H, CHENG S S, JIN Y, et al. Roles of CA125 in diagnosis, prediction, and oncogenesis of ovarian cancer[J/OL]. Biochim Biophys Acta Rev Cancer, 2021, 1875(2): 188503 [2024-06-12]. https://pubmed.ncbi.nlm.nih.gov/33421585/. DOI: 10.1016/j.bbcan.2021.188503.
[33]
BUAMAH P. Benign conditions associated with raised serum CA-125 concentration[J]. J Surg Oncol, 2000, 75(4): 264-265. DOI: 10.1002/1096-9098(200012)75:4<264:aid-jso7>3.0.co;2-q.
[34]
SÖLÉTORMOS G, DUFFY M J, HASSAN S O ABU, et al. Clinical use of cancer biomarkers in epithelial ovarian cancer: updated guidelines from the European Group on tumor markers[J]. Int J Gynecol Cancer, 2016, 26(1): 43-51. DOI: 10.1097/IGC.0000000000000586.
[35]
ALI A T, AL-ANI O, AL-ANI F. Epidemiology and risk factors for ovarian cancer[J]. Prz Menopauzalny, 2023, 22(2): 93-104. DOI: 10.5114/pm.2023.128661.
[36]
THOMASSIN-NAGGARA I, BELGHITTI M, MILON A, et al. O-RADS MRI score: analysis of misclassified cases in a prospective multicentric European cohort[J]. Eur Radiol, 2021, 31(12): 9588-9599. DOI: 10.1007/s00330-021-08054-x.
[37]
HUCHON C, BOURDEL N, ABDEL WAHAB C, et al. Borderline ovarian tumors: French guidelines from the CNGOF. Part 1. Epidemiology, biopathology, imaging and biomarkers[J/OL]. J Gynecol Obstet Hum Reprod, 2021, 50(1): 101965 [2024-06-12]. https://pubmed.ncbi.nlm.nih.gov/33160106/. DOI: 10.1016/j.jogoh.2020.101965.
[38]
DEVINS K M, YOUNG R H, WATKINS J C. Sclerosing stromal tumour: a clinicopathological study of 100 cases of a distinctive benign ovarian stromal tumour typically occurring in the young[J]. Histopathology, 2022, 80(2): 360-368. DOI: 10.1111/his.14554.

上一篇 初探T1 mapping联合DWI识别早期肾间质纤维化的价值
下一篇 mp-MRI影像组学术前预测子宫内膜癌微卫星不稳定性的应用研究
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2