Poster T1 - Neoadjuvant Vascular Normalization Augments Focused Ultrasound-Mediated Model Drug Delivery to Solid Tumors

Introduction: Poor delivery and penetration of intravenously administered drugs into solid tumors is often rate-limiting to effective therapeutic outcomes. Focused ultrasound (FUS) and microbubbles may be employed to augment drug delivery to solid tumors; however, heterogeneous perfusion and high interstitial pressure limit the potential efficacy of this approach. Excessive pro-angiogenic signaling, which leads to aberrant tumor vessel function, is the primary culprit behind these limitations. Low-dose anti-angiogenic drugs are capable of reversing vessel leakiness, tumor hypoxia, high interstitial pressure, and poor pericyte coverage of vessels, in a phenomenon called vascular normalization (VN). Here, we tested whether neoadjuvant normalization of tumor vasculature could augment subsequent FUS-mediated drug delivery and penetration of two model drugs.

Materials and

Methods: DC101, a VEGFR2 blocking antibody capable of eliciting vascular normalization, was intraperitoneally administered to 4T1 tumor-bearing BALB/c mice. VN was confirmed by CD31 immunofluorescence and contrast-enhanced ultrasound. T1-mapping MRI was used to spatiotemporally quantify the influence of vascular normalization on the delivery of two model drugs, both with and without FUS. We employed Gadovist as a model small molecule drug (600 Da) analogous to a chemotherapeutic or small molecule inhibitor and Gadospin D as a model biologic drug (17 kDa upon administration, with an ~5 nm diameter). We quantified and analyzed these two molecules’ delivery as the integration of their concentration over time as a metric of cumulative exposure. Additionally, we evaluated the subsequent distribution of the model drugs throughout the tumor microenvironment.

Results, Conclusions, and Discussions:

Results: DC101 normalized tumor vasculature, eliciting a reduced vessel diameter and density (Fig. 1A-D) and increased perfused tumor fraction (Fig 1E,F), without affecting bulk tumor growth (Fig. 1G). Both DC101 and FUS individually augmented total model small molecule drug (Gadovist) exposure by ~30% (Fig. 1H). Gadovist delivery with FUS was marginally improved by vascular normalization ~25% (Fig. 1H). Additionally, this combination improved the distribution of drug throughout the tumor, evidenced by trending reductions of the T1 MRI mapping distribution’s skewness (Fig. 1I), coefficient of variation (COV) (Fig. 1J), and kurtosis (Fig. 1K). For the model biologic drug (Gadospin-D), neither FUS nor vascular normalization alone enhanced delivery (Fig. 1L); however, when combined with vascular normalization, total FUS-mediated Gadospin-D exposure increased ~50% (Fig. 1L). Visualizations depicting the distribution of Gadospin-D throughout tumors are provided in Fig. 1M. Despite the ability of DC101 to potentiate total FUS-mediated Gadospin-D delivery, the uniformity of model drug distribution throughout the tumor was not changed (Fig. 1N-P).

Discussion: By leveraging quantitative T1 MRI in a novel way, we determined that the ability for VN and FUS to increase intratumoral model drug concentration and improve its distribution throughout the tumor is dependent on model drug size. The VN+FUS combination treatment yields a modest increase in small model drug delivery, yet markedly improves the uniformity of the drug’s distribution throughout the tumor. In contrast, we observe a much starker increase in total delivery of the larger model drug; however, this occurred with little improvement in the uniformity of the model drug distribution. These delivery differences are likely rooted in the differently sized model drugs’ reliance on diffusion versus convection for transport.

Conclusions: Neoadjuvant vascular normalization can markedly augment the therapeutic efficacy of FUS-mediated drug delivery to solid tumors; however, the relative ability of the VN + FUS combination to augment total drug delivery vs. the spatial uniformity of delivery throughout the tumor microenvironment depends strongly on the size of the model drug being delivered.

Acknowledgements (Optional): MRS was supported by an NSF GRFP. Also supported by NIH R01EB030409 and R01EB030007to RJP.