Molecularly targeted photoacoustic endoscopy with fiber-scanning side-view probe for in vivo staging of early mucosal tumors

分子靶向光声内窥镜结合光纤扫描侧视探头用于早期黏膜肿瘤的体内分期

阅读:1

Abstract

Accurate in vivo staging of early gastrointestinal (GI) cancers is critical for selecting between local and systemic therapy. We present a molecularly targeted photoacoustic endoscopy (PAE) system that combines a compact, fiber-scanning side-view probe with a HER2-targeted near-infrared (NIR) contrast agent (KSP∗-IRDye800) to assess mucosal and submucosal tumor invasion in vivo. The 4.2 mm diameter probe uses a piezoelectric (PZT) bender to steer a laser beam laterally (±16°) and achieve high-resolution imaging. The system provides 363 μm lateral and 119 μm axial resolution at a depth of 3.1 mm and supports 3D volumetric image acquisition via rotational scanning and linear pullback. In vivo imaging was performed in CPC;Apc mice that spontaneously develop colonic adenomas. The targeted contrast agent demonstrated a significantly higher peak target-to-background (T/B) ratio (3.0 ± 0.3, RSD = 10 %) than indocyanine green (ICG, 1.37 ± 0.1), with peak uptake at 1.5 h post-injection. Adenoma dimensions measured by PAE correlated strongly with histology (ρ = 0.97 for width, ρ = 0.90 for depth), and 3D reconstructions accurately delineated tumor margins. Ex vivo validation confirmed imaging performance and molecular specificity. This work demonstrates the feasibility of targeted PAE for high-resolution, minimally invasive staging of early GI tumors. The system's resolution and depth performance are sufficient to distinguish between T1(a) and T1(b) lesions. Integration of molecular contrast with miniaturized photoacoustic imaging enables real-time assessment of tumor invasion depth and has potential to improve diagnostic accuracy and therapeutic decision-making during endoscopy.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。