Ultra-high frequency ultrasound imaging and quantification of microvascular flow in xenograft renal cell carcinoma in an avian chorioallantoic membrane model

TL;DR

This study develops and validates an ultra-high frequency ultrasound imaging pipeline with motion compensation and clutter filtering to quantify microvascular flow in CAM tumor models, enabling rapid assessment of treatment response.

Contribution

The paper introduces a novel, less computationally intensive UHFUS imaging pipeline combining motion compensation and interframe subtraction for CAM tumor microvascular flow detection.

Findings

MC+IS effectively detects flow changes post-treatment.

UHFUS methods show significant flow decrease in treated tumors.

Pipeline enables rapid, high-throughput assessment of therapy response.

Abstract

Patient derived xenograft (PDX) tumor models initiated in avian chorioallantoic membranes (CAM) are under investigation to evaluate the effectiveness of therapeutic options with the objective of personalizing treatments. CAM PDXs paired with ultra-high frequency ultrasound (UHFUS) imaging could potentially constitute prospective high throughput assays that can rapidly assess tumor volume and vascular response to therapy. To date, little work has been conducted to adapt and validate UHFUS flow imaging methods to CAM tumor models. Here we report the development and evaluation of an imaging pipeline for UHFUS detection of microvascular flow in a CAM tumor model using interframe subtraction (IS) to suppress tissue clutter. The IS pipeline included a tissue motion compensation (MC) stage prior to clutter filtering and was compared to a singular value decomposition (SVD) clutter filter. The performance was evaluated using UHFUS data acquired in phantom and in vivo Sunitinib-treated renal cell carcinoma. MC substantially reduced tissue motion effects. MC+IS was comparable to MC+SVD filtering at detecting flow within tumors. The results for both IS and SVD filters were dependent on the details of implementation. The UHFUS imaging methods detected a significant decrease in blood flow metrics in treated versus control tumors. An effective imaging pipeline was developed for the assessment of the treatment response of CAM PDX models in a clinically relevant timeframe. The MC+IS approach implemented on B-scan image derived data is less computationally intensive and can be used with widely available UHFUS systems.