Objectives: The aim of this study was to investigate magnetic resonance imaging (MRI) with αvß3-integrin-targeted ultrasmall superparamagnetic iron oxide nanoparticles (RGD-USPIO) for the in vivo monitoring of early antiangiogenic therapy effects in experimental breast cancer.
Materials and methods: Orthotopic human breast cancer (MDA-MB-231) xenograft-bearing severe combined immunodeficiency mice were imaged before and after a 1-week therapy with the vascular endothelial growth factor receptor-antibody bevacizumab or placebo (n = 10 per group, daily intraperitoneal injections of bevacizumab or a volume-equivalent placebo solution, respectively) on a clinical 3 T scanner (Magnetom Skyra; Siemens Healthcare, Erlangen, Germany) before and 60 minutes after the intravenous injection of RGD-USPIO (P04000; Guerbet, Villepinte, France). R2 relaxometry employing a T2-weighted spin-echo sequence with 4 echo times (echo time, 20/40/60/80 milliseconds; repetition time, 3800 milliseconds; matrix, 128 × 128; field of view, 50 × 50; slice thickness, 1.2 mm; time to acquisition, 25 minutes) was used as semiquantitative measure to determine RGD-USPIO endothelial binding. In addition, the T2-weighted images were used to perform volumetric tumor response assessments. Imaging results were validated by ex vivo multiparametric immunohistochemistry with regard to αvß3-integrin expression, microvascular density (CD31), proliferation (Ki-67), and apoptosis (TUNEL).
Results: RGD-USPIO endothelial binding was significantly reduced after vascular endothelial growth factor inhibition, compared with the control group in which an increased endothelial binding was detected ([INCREMENT]R2Therapy = -0.80 ± 0.78 s; [INCREMENT]R2Control = +0.27 ± 0.59 s; P = 0.002). Correspondingly, immunohistochemistry revealed a significantly lower αvß3-integrin expression (91 ± 30 vs 357 ± 72; P < 0.001), microvascular density (CD31, 109 ± 46 vs 440 ± 208; P < 0.001), tumor cell proliferation (Ki-67, 4040 ± 1373 vs 6530 ± 1217; P < 0.001), as well as significantly higher apoptosis (TUNEL, 11186 ± 4387 vs 4017 ± 1191; P = 0.004) in the therapy compared with the control group. Contrary to the changes in αvß3-integrin expression detected by RGD-USPIO MRI, morphology-based tumor response assessments did not show a significant intergroup difference in tumor volume development over the course of the experiment (ΔVolTherapy +71 ± 40 μL vs ΔVolControl +125 ± 81 μL; P > 0.05).
Conclusions: RGD-USPIO MRI allows for the noninvasive assessment of αvß3-integrin expression in the investigated breast cancer model. RGD-USPIO MRI may be applicable for the in vivo monitoring of early antiangiogenic therapy effects in experimental breast cancer, generating possible complementary molecular imaging biomarkers to morphology-based tumor response assessments.