2026 年第 10 期 第 21 卷

超声造影参数联合临床指标预测去势抵抗性前列腺癌的价值研究

Study on the value of contrast-enhanced ultrasound parameters combined with clinical indicators in predicting castration-resistant prostate cancer

作者：孟佳琪1红华2

英文作者：Meng Jiaqi1 Hong Hua2

单位：1内蒙古医科大学，呼和浩特010110；2内蒙古自治区人民医院超声医学科，呼和浩特010010

英文单位：1Inner Mongolia Medical University Hohhot 010110 China; 2Department of Ultrasound Medicine Inner Mongolia People′s Hospital Hohhot 010010 China

关键词：前列腺癌；Ki-67；超声造影；病理分级

英文关键词：Prostatecancer;Ki-67;Contrast-enhancedultrasound;Pathologicalgrading

• 摘要：

• 目的  探讨超声造影参数联合临床指标预测去势抵抗性前列腺癌（CRPC）的价值。方法 收集2018年7月至2024年7月在内蒙古自治区人民医院经病理证实为前列腺癌且行内分泌治疗1～2年的148例患者的临床资料行回顾性分析。以随访18个月为节点，根据是否进展为CRPC分为未进展组（100例）和进展组（48例）。筛选CRPC进展的相关因素并建立预测模型，比较不同模型的效能。结果 多因素分析显示，最低前列腺特异性抗原（PSA）≥1 μg/L、雄激素剥夺治疗后PSA降至最低值所需时间（TTN）<6个月、Ki-67高表达、Gleason评分高危、骨转移以及峰值强度是CRPC进展的独立危险因素（均P<0.05）。基于上述指标构建3种预测模型：临床模型（最低PSA、TTN、Ki-67、Gleason评分、骨转移）、超声造影参数模型（峰值强度）及联合模型（临床指标+峰值强度）。受试者工作特征曲线分析显示，联合模型的曲线下面积（AUC）为0.822（95%置信区间：0.755～0.890），敏感度为91.7%，特异度为60.0%；临床模型AUC为0.812（95%置信区间：0.738～0.890），特异度为76.0%；超声造影参数模型AUC为0.640（95%置信区间：0.545～0.735），特异度为66.0%，联合模型预测效能最优。结论 利用超声造影参数联合临床指标建立的预测模型可以早期准确预测CRPC的发生。

• Objective To explore the value of contrast-enhanced ultrasound parameters combined with clinical indicators in predicting castration-resistant prostate cancer (CRPC). Methods Clinical data of 148 patients pathologically confirmed as prostate cancer and receiving 1-2 years of endocrine therapy at Inner Mongolia People′s Hospital from July 2018 to July 2024 were collected for retrospective analysis. With 18 months of follow-up as the endpoint, patients were divided into non-progression group (100 cases) and progression group (48 cases) according to whether CRPC progression occurred. Related factors for CRPC progression were screened and prediction models were established, and the performance of different models was compared. Results Multivariate analysis showed that nadir prostate-specific antigen (PSA) ≥1 μg/L, time to PSA nadir (TTN) <6 months after androgen deprivation therapy, high Ki-67 expression, high-risk Gleason score, bone metastasis and peak intensity were independent risk factors for CRPC progression (all P<0.05). Three prediction models were constructed based on the above indicators: clinical model (nadir PSA, TTN, Ki-67, Gleason score, bone metastasis), contrast-enhanced ultrasound parameter model (peak intensity) and combined model (clinical indicators + peak intensity). Receiver operating characteristic curve analysis showed that the combined model had an area under the curve (AUC) of 0.822 (95% confidence interval: 0.755-0.890), a sensitivity of 91.7% and a specificity of 60.0%; the clinical model had an AUC of 0.812 (95% confidence interval: 0.738-0.890) and a specificity of 76.0%; the contrast-enhanced ultrasound parameter model had an AUC of 0.640 (95% confidence interval: 0.545-0.735) and a specificity of 66.0%. The combined model exhibited the optimal predictive performance. Conclusion The prediction model established by contrast-enhanced ultrasound parameters combined with clinical indicators can early and accurately predict the onset of CRPC.