多参数MRI与微超声对临床显著前列腺癌诊断准确性的比较:前瞻性研究双变量荟萃分析
该系统综述与荟萃分析仅纳入前瞻性研究,比较微超声与多参数MRI检测临床显著前列腺癌的准确性。共纳入8项前瞻性研究(n=2626),微超声汇总敏感度0.87、特异度0.25,mpMRI敏感度0.88、特异度0.30;敏感度无显著差异(P=0.72),特异度mpMRI更优(P=0.003)。
要点
- 微超声检测csPCa的敏感度与mpMRI相当(0.87 vs 0.88,P=0.72)。
- mpMRI特异度显著优于微超声(0.30 vs 0.25,P=0.003)。
- 微超声或可作为可及性替代或补充模态,但需更多前瞻性研究验证。
核心结论
- 在确诊活检前筛查临床显著前列腺癌时,微超声敏感度与多参数MRI相当,但多参数MRI特异度更优。
- 微超声可能作为可及性替代或补充模态,但需进一步高质量前瞻性研究强化比较证据。
研究设计与样本
- 系统综述与荟萃分析,方案注册号INPLASY202540027,遵循PRISMA和PICOTT框架。
- 纳入2012年至2025年3月发表的前瞻性队列研究和随机对照试验,共8项研究(n=2626例患者),含1项随机对照试验和7项前瞻性队列。
- 以活检或前列腺切除标本为参考标准,采用双变量随机效应模型估计汇总敏感度、特异度和sROC曲线。
主要发现
- 微超声汇总敏感度0.87(95%CI: 0.80-0.92),特异度0.25(95%CI: 0.17-0.36)。
- 多参数MRI汇总敏感度0.88(95%CI: 0.81-0.93),特异度0.30(95%CI: 0.18-0.46)。
- 两种模态的sROC置信区域重叠。
- 荟萃回归显示敏感度无显著差异(P=0.72),特异度差异显著 favoring mpMRI(P=0.003)。
- 不同患病率场景下PPV为0.41-0.46,NPV为0.72-0.80。
临床怎么用
- 微超声可作为mpMRI可及性受限时的替代或补充筛查工具,用于确诊活检前csPCa检测。
- 因微超声特异度较低,需结合临床情境解读阳性结果,避免过度活检。
局限性
- 仅纳入前瞻性研究,纳入研究数量有限(8项)。
- 部分汇总特异度置信区间较宽,精确性受限。
- 需更多高质量前瞻性研究强化比较证据。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Prostate cancer (PCa) remains a leading cause of cancer-related mortality in men. While multiparametric MRI (mpMRI) is an established tool for detecting clinically significant PCa (csPCa), it is limited by cost, access, and acquisition time. Micro-ultrasound (Micro-US) offers real-time imaging with potential advantages in accessibility and integration into routine care. This systematic review and meta-analysis (SR/MA) aimed to compare the diagnostic accuracy of Micro-US versus mpMRI in detecting csPCa, based exclusively on prospective evidence. METHODS: A protocol-registered SR/MA (INPLASY202540027) was conducted following PRISMA and PICOTT frameworks. Prospective cohort studies and randomized controlled trials published between 2012 and March 2025 comparing micro-US and mpMRI for csPCa detection, using biopsy or prostatectomy specimens as reference standards, were included. Bivariate random-effects models were used to estimate pooled sensitivity, specificity, and summary ROC curves. Positive/negative predictive values (PPV/NPV) were calculated using pooled prevalence and literature-based prevalence values. Meta-regression assessed modality differences and potential effect modifiers. RESULTS: Eight prospective studies (n = 2626 patients) met the inclusion criteria, 1 randomized controlled trial and 7 prospective cohorts. Micro-US demonstrated a pooled sensitivity of 0.87 (95%CI: 0.80-0.92) and specificity of 0.25 (95% CI: 0.17-0.36), while mpMRI showed a sensitivity of 0.88 (95% CI: 0.81-0.93) and specificity of 0.30 (95% CI: 0.18-0.46). sROC confidence regions overlapped for both modalities. Meta-regression detected no significant difference in sensitivity (P = 0.72) but a significant difference in specificity favoring mpMRI (P = 0.003). PPVs were modest (0.41-0.46), and NPVs were high (0.72-0.80) across prevalence scenarios. CONCLUSION: Micro-US demonstrates sensitivity comparable to mpMRI for csPCa screening before confirmatory biopsy, although mpMRI retains superior specificity. Micro-US may serve as an accessible alternative or complementary modality, but further high-quality prospective studies are needed to strengthen comparative evidence.
引用
- Garcia-Becerra CA, Arias-Gallardo MI, Juarez-Garcia JE, et al. Comparative diagnostic accuracy of multiparametric-MRI and Micro-ultrasound for clinically significant prostate cancer-a bivariate meta-analysis of prospective studies. Prostate Cancer Prostatic Dis. 2026. PMID: 41634333. doi:10.1038/s41391-026-01079-7
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。