PSA灰区临床局限性前列腺癌包膜外侵犯的MRI影像组学-血清miR-181c-5p诊断模型的建立与泛化能力验证-东南大学学报医学版

PSA灰区临床局限性前列腺癌包膜外侵犯的MRI影像组学-血清miR-181c-5p诊断模型的建立与泛化能力验证

单位：1. 邯郸市第一医院 CT. MRI室, 河北邯郸 056000;
2. 邯郸市第一医院检验科, 河北邯郸 056000;
3. 石河子大学第一附属医院医学影像中心, 新疆石河子 832000

关键词：前列腺特异抗原灰区临床局限性前列腺癌包膜外侵犯 MRI影像组学 miR-181c-5p 诊断模型

分类号：R737.25

出版年·卷·期（页码）：2026·45·第四期（564-575）

摘要：

目的:采用MRI影像组学-血清微小RNA(miR)-181c-5p构建前列腺特异抗原(PSA)灰区临床局限性前列腺癌包膜外侵犯(EPE)的诊断模型,并对其泛化能力进行验证。方法:前瞻性选取2021年2月至2023年12月邯郸市第一医院收治的284例PSA灰区临床局限性前列腺癌患者作为建模集,根据手术病理结果是否存在EPE分为EPE组(85例)和无EPE组(199例)。比较两组基线资料和血清miR-181c-5p水平,筛选MRI影像组学特征,LASSO-多因素Logistic回归分析EPE的影响因素,构建PSA灰区临床局限性前列腺癌EPE的术前诊断模型。另选石河子大学第一附属医院2024年1月至2025年10月的121例该病患者作为外部验证集(EPE 38例,无EPE 83例)。采用受试者工作特征(ROC)曲线、校准曲线、决策曲线评估模型的泛化能力。结果:EPE组前列腺特异性抗原密度(PSAD)、穿刺Gleason评分、T2期占比、外周带肿瘤占比、尖部受累占比、影像组学评分(Rad-score)均高于无EPE组,前列腺体积、miR-181c-5p低于无EPE组(P＜0.05)。LASSO-多因素Logistic回归分析显示,穿刺Gleason评分、临床T分期、尖部受累、miR-181c-5p、Rad-score是EPE的相关影响因素(P＜0.05);基于穿刺Gleason评分、临床T分期、尖部受累、miR-181c-5p、Rad-score分别构建EPE的术前诊断模型,即临床基础模型、临床+miR模型、临床+MRI模型、联合模型。ROC曲线显示,建模集和外部验证集中联合模型预测的曲线下面积分别为0.866(95%CI:0.816~0.916)、0.840(95%CI:0.761~0.918);校准曲线显示,建模集和外部验证集中联合模型评估EPE发生率与实际发生率基本一致;决策曲线显示,建模集和外部验证集中联合模型在5%~99%、4%~99%的阈值概率范围内,其评估EPE具有较好的临床净获益。结论:基于穿刺Gleason评分、临床T分期、尖部受累、miR-181c-5p、Rad-score构建的PSA灰区临床局限性前列腺癌EPE的术前诊断模型,具有良好的预测效能和泛化能力,可为评估患者术前EPE提供参考。

Objective: To build a combined MRI radiomics and serum miR-181c-5p model to predict extracapsular extension(ECE) in clinically localized prostate cancer patients with prostate-specific antigen(PSA) in the gray zone, and to verify the model's generalizability. Methods: A total of 284 patients with clinically localized prostate cancer and PSA levels within the gray zone who were admitted to Handan First Hospital between February 2021 and December 2023 were prospectively enrolled as the training cohort. Patients were stratified into an ECE group(n=85) and a non-ECE group(n=199) based on the presence or absence of extracapsular extension(ECE) on postoperative pathological specimens. Baseline clinical data and serum miR-181c-5p levels were compared between the two groups, and MRI radiomic features were screened. Least absolute shrinkage and selection operator(LASSO) regression followed by multivariate Logistic regression was performed to identify independent risk factors for ECE, and a preoperative diagnostic model for ECE was established for patients with PSA gray zone clinically localized prostate cancer. An additional 121 patients treated at The First Affiliated Hospital of Shihezi University from January 2024 to October 2025 were recruited as the external validation cohort, including 38 patients with ECE and 83 without ECE. Receiver operating characteristic(ROC) curves, calibration curves, and decision curve analysis(DCA) were applied to assess the model's generalizability. Results: Patients in the ECE group exhibited significantly higher prostate-specific antigen density(PSAD), biopsy Gleason score, percentage of T2-stage disease, percentage of peripheral zone lesions, percentage of apical involvement, and radiomic score(Rad-score), alongside smaller prostate volume and lower serum miR-181c-5p levels relative to the non-ECE group(all P<0.05). Two-step LASSO and multivariate Logistic regression analysis identified biopsy Gleason score, clinical T stage, apical involvement, miR-181c-5p, and Rad-score as independent risk factors for ECE(all P<0.05). Four preoperative predictive models for ECE were constructed using the five predictors above: a clinical-only model, a clinical-miR combined model, a clinical-MRI radiomics combined model, and an integrated multi-index model. ROC curve analysis revealed that the combined model yielded AUC values of 0.866(95%CI: 0.816-0.916) in the training cohort and 0.840(95%CI: 0.761-0.918) in the external validation cohort. Calibration curves demonstrated strong consistency between model-predicted ECE probabilities and actual observed ECE rates across both cohorts. Decision curve analysis confirmed that the combined model provided favorable clinical net benefit for ECE prediction over wide threshold probability ranges of 5%-99%(training cohort) and 4%-99%(external validation cohort). Conclusion: Built on biopsy Gleason score, clinical T stage, apical involvement, serum miR-181c-5p and Rad-score, this preoperative ECE prediction model for clinically localized prostate cancer patients with PSA in the gray zone shows excellent predictive performance and generalizability for preoperative ECE assessment.

参考文献：

[1] 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 A Cancer J Clin,2024,74(3):229-263.
[2] GUO E,XU L,ZHANG D,et al.Diagnostic performance of MRI in detecting prostate cancer in patients with prostate-specific antigen levels of 4-10 ng/mL:a systematic review and meta-analysis[J].Insights Imag,2024,15(1):147.
[3] KOBAYASHI M,MATSUOKA Y,UEHARA S,et al.Utility of positive core number on MRI-ultrasound fusion targeted biopsy in combination with PI-RADS scores for predicting unexpected extracapsular extension of clinically localized prostate cancer[J].Int J Urol,2024,31(7):739-746.
[4] OLIVEIRA T,AMARAL FERREIRA L,MARTO C M,et al.The role of multiparametric MRI in the local staging of prostate cancer[J].Front Biosci(Elite Ed),2023,15(3):21.
[5] XIAO V G,KRESNANTO J,MOSES D A,et al.Quantitative MRI in the local staging of prostate cancer:a systematic review and meta-analysis[J].J Magn Reson Imag,2024,59(1):255-296.
[6] WU S Y,WANG Y,FAN P,et al.Bi-parametric MRI-based quantification radiomics model for the noninvasive prediction of histopathology and biochemical recurrence after prostate cancer surgery:a multicenter study[J].Abdom Radiol,2025,50(9):4320-4330.
[7] WU G,SUN Y,XIANG Z,et al.Preclinical study using circular RNA 17 and micro RNA 181c-5p to suppress the enzalutamide-resistant prostate cancer progression[J].Cell Death Dis,2019,10(2):37.
[8] WU J,YU X,ZUO Z,et al.Predictive value of magnetic resonance imaging radiomics combined with serum PSA for the extracapsular extension of prostate cancer prediction[J].Arch Españoles De Urol,2024,77(2):217.
[9] 中国抗癌协会泌尿男生殖系统肿瘤专业委员会前列腺癌学组.前列腺癌筛查中国专家共识(2021年版)[J].中国癌症杂志,2021,31(5):435-440.
[10] EPSTEIN J I,EGEVAD L,AMIN M B,et al.The 2014 international society of urological pathology(ISUP) consensus conference on gleason grading of prostatic carcinoma:definition of grading patterns and proposal for a new grading system[J].Am J Surg Pathol,2016,40(2):244-252.
[11] PRIESTER A,MOTA S M,GRUNDEN K P,et al.Extracapsular extension risk assessment using an artificial intelligence prostate cancer mapping algorithm[J].BJUI Compass,2024,5(10):1100-1111.
[12] 周金才,居峰,倪颖,等.前列腺健康指数在不同tPSA区间段诊断前列腺癌的临床价值研究[J].现代医学,2023,51(7):918-922.
[13] 摆俊博,陶宁,邱瑞莹,等.去势抵抗性前列腺癌的危险因素及神经侵犯对疾病进展的评估价值[J].现代泌尿外科杂志,2022,27(3):210-217.
[14] SEKITO S,ONISHI T,OKAMOTO T,et al.Predictive factors for extracapsular extension of prostate cancer to select the candidates for nerve-sparing radical prostatectomy[J].Indian J Surg Oncol,2024,15(2):213-217.
[15] DING Y,MO C,DING Q,et al.Prediction of T staging in PI-RADS 4-5 prostate cancer by combination of multiparametric MRI and 68Ga-PSMA-11 PET/CT[J].BMC Urol,2023,23(1):206.
[16] MA M W,WANG K,GAO X S,et al.Integration of multiparameter MRI into conventional pretreatment risk factors to predict positive surgical margins after radical prostatectomy[J].Clin Genitourin Cancer,2024,22(2):281-290.e1.
[17] YANG L,JIN P,QIAN J,et al.Value of a combined magnetic resonance imaging-based radiomics-clinical model for predicting extracapsular extension in prostate cancer:a preliminary study[J].Transl Cancer Res,2023,12(7):1787-1801.
[18] GUERRA A,WANG H,ORTON M R,et al.Prediction of extracapsular extension of prostate cancer by MRI radiomic signature:a systematic review[J].Insights Imag,2024,15(1):217.
[19] HUANG G,ALBERS P,MOOKERJI N,et al.Three-dimensional spatial localization and volume estimation of prostate tumors using(18)F-PSMA-1007 PET/CT versus multiparametric MRI[J].Eur J Nucl Med Mol Imaging,2025,52(5):1642-1648.
[20] PEDRAZA A M,FERGUSON E L,RAMOS-CARPINTEYRO R,et al.Single-port robotic transvesical partial prostatectomy:a novel approach for focal treatment in prostate cancer[J].J Endourol,2025,39(3):261-270.
[21] 徐鸿昊,杜其聪,马元昊,等.双参数MRI瘤周影像组学对前列腺癌包膜外侵犯的诊断价值[J].中华放射学杂志,2025,59(9):1055-1062.
[22] 乔晓梦,胡尘翰,杨丽勤,等.基于双参数MRI影像组学模型在PSA灰区csPCa的应用研究[J].实用放射学杂志,2023,39(9):1462-1466.
[23] SYED R U,ALSHAMMARI M D,BANU H,et al.Targeting the autophagy-miRNA axis in prostate cancer:toward novel diagnostic and therapeutic strategies[J].Naunyn Schmiedeberg's Arch Pharmacol,2024,397(10):7421-7437.
[24] KANG S,OU C,YAN A,et al.Long noncoding RNA SNHG5 induces the NF-κB pathway by regulating miR-181c-5p/CBX4 axis to promote the progression of non-small cell lung cancer[J].Arch De Bronconeumología,2023,59(1):10-18.
[25] BAI Y,ZHANG Z,BI J,et al.miR-181c-5p/DERL1 pathway controls breast cancer progression mediated by TRAF6-linked K63 ubiquitination of AKT[J].Cancer Cell Int,2024,24(1):204.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】

提示：您还未登录，请登录！点此登录