PO.CL01.08 · 临床研究
通过整合基于 CRISPR 的细胞外囊泡 RNA 和蛋白质分析推进液体活检
Advancing liquid biopsies via integrated CRISPR-based extracellular vesicle RNA and protein analyses
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
背景：我们开发了 REPAIR，一个基于尿液的整合平台，将细胞外囊泡（EV）mRNA 和蛋白质谱分析与 CRISPR 相结合，以实现非侵入性、多参数的疾病监测。膀胱癌作为我们的临床应用案例，鉴于其晚期时的高死亡率，仍然是监测和治疗成本最高的癌症。目前对局限性疾病的监测依赖于侵入性膀胱镜检查，缺乏用于风险分层和频繁治疗监测的分子特异性。
方法：我们大幅推进了近期以 mRNA 为重点的 CRISPR-Cas13 检测（该检测在血浆中实现了亚阿托摩尔级灵敏度和单核苷酸分辨率；Nature Biotech，2025 [封面文章]；PMID: 39375445），具体措施包括：（1）整合 EV mRNA 加蛋白质的双重检测；（2）使用我们开发的创新性基于圆盘的微流控技术实现样本处理自动化；以及（3）转化为以尿液作为临床基质。作为概念验证，我们设计了针对 EV 蛋白质（CD63、EGFR、EpCAM、PD-L1）和 mRNA（GAPDH、LASS2、GALNT1）的 REPAIR 探针，这些标志物也与膀胱癌相关。对于蛋白质检测，使用适配体作为亲和配体。这些适配体最初与互补 RNA 序列（起始 RNA）杂交。当适配体与靶点结合时，起始 RNA 被释放并通过 REPAIR 检测。对于 mRNA 检测则采用了非基于适配体的策略。使用统一的 REPAIR 方案分析了来自 85 例接受治疗的膀胱癌患者（多为非肌层浸润性）和 98 例健康对照的尿液样本。
结果：多参数谱分析显示，与对照尿液相比，癌症患者的标志物表达显著升高（所有标志物 p<0.05）。整合蛋白质和 mRNA 标志物的 Logistic 回归实现了 98% 的检测准确率（AUC=0.95；95% CI：0.92-0.98）。单标志物分析显示性能较差，证明了基于 CRISPR 的多参数 EV 方法的关键价值。整合读数在 60 分钟内即可获得。
结论：REPAIR 极大地促进了尿液中 EV 蛋白质-RNA 的同步谱分析——这是一种用于膀胱癌监测和治疗监测的非侵入性、可扩展的液体活检。这些发现有望为一种经济负担巨大且监测极具挑战的疾病建立一个基于纳米技术的同次就诊精准监测平台。目前正在其他以尿液作为近端体液的恶性肿瘤中开展实验。
查看英文原文 English abstract
BACKGROUND : We developed REPAIR, an integrated urine-based platform combining extracellular vesicle (EV) mRNA and protein profiling with CRISPR to enable non-invasive, multiparametric disease monitoring. Bladder cancer, our clinical use case, remains the costliest cancer to surveil and treat given the high mortality when advanced. Current surveillance of localized disease relies on invasive cystoscopy, lacking the molecular specificity for risk stratification and frequent treatment monitoring.
METHODS : We significantly advanced our recent mRNA-focused CRISPR-Cas13 assay achieving sub-attomolar sensitivity and single-nucleotide resolution in plasma ( Nature Biotech , 2025 [cover article]; PMID: 39375445), by: (1) integrating dual EV mRNA plus protein detection; (2) automating sample processing with innovative disc-based microfluidics we developed; and (3) translating to urine as the clinical matrix. As proof-of-concept, we designed REPAIR probes targeting EV proteins (CD63, EGFR, EpCAM, PD-L1) and mRNAs (GAPDH, LASS2, GALNT1) also relevant to bladder cancer. For protein detection, aptamers were used as affinity ligands. These aptamers were initially hybridized with complementary RNA sequences (initiator RNAs). When the aptamers bound to targets, the initiator RNAs were released and detected through REPAIR. A non-aptamer based strategy was employed for mRNA detection. Urine samples from 85 bladder cancer patients (mostly non-muscle invasive) undergoing treatment and 98 healthy controls were analyzed using the unified REPAIR protocol.
RESULTS : Multiparametric profiling revealed significantly elevated marker expression in cancer versus control urine (p<0.05 for all markers). Integrated logistic regression combining protein and mRNA markers achieved 98% detection accuracy (AUC=0.95; 95% CI: 0.92-0.98). Single-marker analysis showed inferior performance, demonstrating the critical value of a multiparametric CRISPR-based EV approach. Integrated readouts were achieved within 60 minutes.
CONCLUSION : REPAIR greatly facilitates simultaneous EV protein-RNA profiling in urine-a non-invasive, scalable liquid biopsy for bladder cancer surveillance and treatment monitoring. These findings potentially establish a nanotechnology-based platform for same-visit precision monitoring in a disease with an extraordinary economic burden and surveillance challenges. Current experiments in other malignancies where urine serves as a proximal fluid are ongoing.
利益披露 Disclosure
C. M. Castro,
Qiagen, Inc. Independent Contractor.
Canon Medical Independent Contractor.
Alight, Inc Independent Contractor.
Teladoc Independent Contractor.
J. Song, None.