Aditi Gupta, Abstract Co-Author: Nothing to Disclose

Kelly Dulin, Abstract Co-Author: Nothing to Disclose

Samantha Jaffe, Abstract Co-Author: Nothing to Disclose

Mark Forsberg, Abstract Co-Author: Nothing to Disclose

Jaydev Kardam Dave PhD, MS, Abstract Co-Author: Nothing to Disclose

Flemming Forsberg PhD, Presenter: Equipment support, Toshiba Corporation Research collaboration, General Electric Company Research collaboration, Analogic Corporation Research collaboration, Toshiba Corporation Speakers Bureau, Toshiba Corporation

Manasi Dahibawkar BSc, Abstract Co-Author: Nothing to Disclose

Valgerdur Halldorsdottir MSc, Abstract Co-Author: Nothing to Disclose

Anya Isabelle Forsberg, Abstract Co-Author: Nothing to Disclose

Andrew Marshall, Abstract Co-Author: Nothing to Disclose

Priscilla Machado MD, Abstract Co-Author: Grant, Toshiba Corporation Equipment support, Toshiba Corporation

Traci B. Fox MS, RT, Abstract Co-Author: Nothing to Disclose

Ji-Bin Liu MD, Abstract Co-Author: Research Grant, GluMetrics, Inc

To compare different methods for quantifying tumor neovascularity based on immunohistochemical markers of angiogenesis to contrast-enhanced subharmonic ultrasound imaging (SHI).

Twenty-eight (28) athymic, nude, female rats were implanted with 5 x 106 breast cancer cells (MDA-MB-231) in the mammary fat pad. The ultrasound contrast agent Definity (Lantheus Medical Imaging, N Billerica, MA) was injected in a tail vein (dose: 36 μl) and pulse-inversion SHI was performed in triplicate using a modified Sonix RP scanner (Analogic Ultrasound, Richmond, BC, Canada) with a L9-4 linear array (transmitting and receiving frequencies of 8 and 4 MHz, respectively). Specimens were extracted and sliced corresponding to the imaging planes and stained for endothelial cells (CD31), vascular endothelial growth factor (VEGF), and cyclooxygenase-2 (COX-2). Tumor neovascularity was quantified in 3 different ways 1) over the entire tumor 2) in small sub-regions of interest (ROIs) and 3) in the tumor periphery (within 2 mm of the margin) and centrally. Results from specimens and SHI were compared using a linear regression analysis.

Of the 28 rats implanted 23 (82 %) exhibited tumor growth and were successfully studied. SHI depicted the tortuous morphology of tumor neovessels and delineated small areas of necrosis. The angiogenic expression in the tumor periphery was consistently higher than in the center for all 3 immunohistochemical markers (p < 0.014). SHI measures of tumor vascularity did not correlate with the immunohistochemical markers when assessed over the entire tumor area or over sub-ROIs (p > 0.3). However, when the specimens were divided into a central and a peripheral region VEGF was found to correlate with SHI in both areas (r = 0.45 and r = 0.56; p < 0.04). The strongest correlation in this model was between SHI and COX-2 in the periphery of the tumors (r = -0.61; p = 0.004).

When comparing quantitative measures of tumor neovascularity derived from immunohistochemical markers to SHI from xenograft models, sub-ROIs corresponding to the biologically active region (i.e., the tumor periphery) appear to account for tumor heterogeneity.

In the future SHI may be used to monitor response for patients treated with anti-COX-2 therapies.

Gupta, A, Dulin, K, Jaffe, S, Forsberg, M, Dave, J, Forsberg, F, Dahibawkar, M, Halldorsdottir, V, Forsberg, A, Marshall, A, Machado, P, Fox, T, Liu, J, Quantifying Tumor Neovascularity with Immunohistochemical Markers Compared to Subharmonic US Imaging.  Radiological Society of North America 2014 Scientific Assembly and Annual Meeting, - ,Chicago IL. http://archive.rsna.org/2014/14045783.html