Assessing Therapeutic Nanoparticle Accumulation in Tumors Using Nanobubble-Based Contrast-Enhanced Ultrasound Imaging

Abstract

This study explores the challenges associated with nanoparticle-based drug delivery to the tumor parenchyma, focusing on the widely utilized enhanced permeability and retention effect (EPR). While EPR has been a key strategy, its inconsistent clinical success lacks clear mechanistic understanding and is hindered by limited tools for studying relevant phenomena. This work introduces an approach that employs multiparametric dynamic contrast-enhanced ultrasound (CEUS) with a nanoscale contrast agent for noninvasive, real-time examination of tumor microenvironment characteristics. We demonstrate that CEUS imaging can: (1) evaluate tumor microenvironment features, (2) be used to help predict the distribution of doxorubicin-loaded liposomes in the tumor parenchyma, and (3) be used to predict nanotherapeutic efficacy. CEUS using nanobubbles (NBs) was carried out in two tumor types of high (LS174T) and low (U87) vascular permeability. LS174T tumors consistently showed significantly different time intensity curve (TIC) parameters, including area under the rising curve (AUC_R, 2.7×) and time to peak intensity (TTP, 1.9×) compared to U87 tumors. Crucially, a recently developed decorrelation time (DT) parameter specific to NB CEUS dynamics successfully predicted the distribution of doxorubicin-loaded liposomes within the tumor parenchyma (r = 0.86 ± 0.13). AUC_R, TTP, and DT were used to correlate imaging findings to nanotherapeutic response with 100% accuracy in SKOV-3 tumors. These findings suggest that NB-CEUS parameters can effectively discern tumor vascular permeability, serving as a biomarker for identifying tumor characteristics and predicting the responsiveness to nanoparticle-based therapies. The observed differences between LS174T and U87 tumors and the accurate prediction of nanotherapeutic efficacy in SKOV-3 tumors indicate the potential utility of this method in predicting treatment efficacy and evaluating EPR in diseases characterized by pathologically permeable vasculature. Ultimately, this research contributes valuable insights into refining drug delivery strategies and assessing the broader applicability of EPR-based approaches.

Keywords: EPR; companion nanoparticles; nanobubbles; oncology; tumors; ultrasound; ultrasound contrast agents.

Figure 1

Experimental setup and parameters. (A) Mice were inoculated with tumor cells, and the tumors were permitted to grow until they reached at least approximately 250 mm³. Mice were then imaged with NB-contrast-enhanced ultrasound (CEUS) and B-mode imaging. One day post CEUS imaging, mice were injected with doxorubicin-loaded liposomes, which circulated for 24 h before sacrifice via exsanguination. (B) Depiction of vascular differences between U87 and LS174T tumor types. (C) Graphical depiction of TIC parameters and their meaning based on Gu et al. DT analysis is based on work by Cooley et al. and Wegierak et al.⁻

Figure 2

Representative TICs and ultrasound images of LS174T and U87 tumors. (A) Representative B-mode and NLC mode images (acquired simultaneously) of an LS174T tumor at baseline (t = 0 s), peak intensity, and 50% of peak intensity in the wash-out. (B) Corresponding LS174T TIC and TIC-derived parameters, normalized by subtracting baseline intensity. (C) Representative B-mode and NLC mode images of a U87 tumor at baseline (t = 0 s), peak intensity, and 50% of peak intensity in the wash-out. (D) Corresponding U87 TIC and TIC-derived parameters, normalized to baseline intensity. a.u. represents arbitrary units, produced by Vevo LAB proprietary software.

Figure 3

TIC parameter analysis between LS174T and U87 tumors. (A) Peak intensity (PI). (B) Time to peak intensity (TTP). (C) Mean transit time (MTT). (D) Area under the curve (AUC). (E) Area under the rising curve (AUC_R). (F) Area under the falling curve (AUC_F). (G) P-values for all analyzed parameters. * indicates p < 0.05, and error bars represent standard deviation. For (A–F), each symbol represents an individual tumor.

Figure 4

Decorrelation time (DT) analysis. (A) DT for each tumor, found by averaging the DT map pixels within the tumor ROI. (B) Standard deviation of DT between all intratumoral pixels. (C) Representative DT map split into quadrants. (D) Histological doxorubicin-generated fluorescence intensity map from the same tumor and approximately the same imaging plane as (C). The tumor is divided into corresponding quadrants. (E) Representation of the linear regression of the average DT vs average fluorescence intensity, corresponding to the tumor in (C, D). (F) Correlation coefficient (r) of all tumors with an average DT > 1 s (A) that had corresponding histological data available. Corresponding quadrants of DT and fluorescence intensity, as in (C, D), were compared to generate the correlation coefficients. U87 and LS174T tumors are labeled; * indicates p < 0.05 and error bars represent standard deviation.

Figure 5

Representation of intratumoral heterogeneity. (A) NLC of an LS174T tumor at baseline (t = 0 s), peak intensity, and 50% of the peak intensity in the wash-out. The ROIs of the tumor have been split into the entire tumor (white dashed lines), Region 1 (pink dashed lines), and Region 2 (blue dashed lines). (B) Fluorescence intensity originating from doxorubicin-loaded liposomes of the corresponding tumor, yellow ROI is magnified in each subsequent image. In the most magnified, rightmost image, the black regions are predominantly blood vessels, surrounded by doxorubicin-generated signal. (C) H&E histological image of the corresponding tumor with the same magnification and ROI as (B). The large, elongated white spaces predominantly correspond to blood vessels. (D) TIC of the ROI of the whole tumor (white dashed lines). (E) TICs when the tumor is divided into Regions 1 and 2. (F) Normalized TIC parameters for each ROI.

Figure 6

TICs and DT maps of LS174T and U87 tumors over 2 weeks of growth. (A–C) TICs and DT maps of three LS174T tumors between Week 1 and Week 2 of growth postinoculation. (D–F) TICs and DT maps of three U87 tumors between Week 2 and Week 3 of growth postinoculation. (G) Average DT of each tumor between the first and second week of imaging. (H) Slope of average DT between the first and second week of imaging.

Figure 7

Evaluating therapeutic efficacy with DT, AUC_R, and TTP. Comparison between responder and nonresponder mice for (A) DT, (B) AUC_R, and (C) TTP. (A–C) The red dotted line represents the threshold for which a mouse is predicted to respond based on LS174T and U87 data. Each symbol represents an individual tumor, and standard deviation is represented with error bars. (D) Visual representation of DT mapping in responder and nonresponder mice.