Abstract
Abstract
Background: Outgrowth of new blood vessels (neovascularization) allows tumors to supply themselves with oxygen and nutrients, and to rapidly metastasize throughout the body. Triple negative breast cancer (TNBC) is particularly susceptible to neovascularization. However, success with anti-angiogenics is highly variable and often patient-specific. This is particularly true as anti-angiogenics are being combined with immunotherapies. Thus, there is a huge unmet need for clinicians to test and predict clinical efficacy of anti-angiogenics at the individual patient level, prior to treatment.
Methods: Here, we characterize a patient-autologous, ex-vivo tumor model, termed CANscript, as a platform to study the intratumor microvascular density (iMVD) of breast cancer samples (N=15). To profile iMVD we used immunohistochemical (IHC) analysis of CD34, an early biomarker of neovascularization. We then introduced anticancer and anti-angiogenic agents (e.g. Avastin) for 72 hours, and subsequently quantified phenotypic response to drugs by testing viability, cell death, proliferation and morphology. These quantitative data were then fed into a machine learning algorithm that provides a clinical response prediction (M-Score).
Results: We determined that ex-vivo culture reliably retains baseline heterogeneity of iMVD based on expression of CD34+ nodes per visual field by IHC. Furthermore, we show that anticancer and anti-angiogenic agents will dynamically alter iMVD, ex-vivo, in a patient-specific manner. Finally, we show that prediction of clinical response using the 'M-Score' algorithm associates with diminished expression of CD34 per visual field of IHC after drug pressure.
Summary: Neovascularization and iMVD are features of aggressive cancers, such as TNBC. CANscript provides a rapid assessment of clinical response to anticancer drugs, many of which induce their antitumor effect by targeting the tumor vasculature. We show that pharmacodynamics of antiangiogenics can be captured during acute ex-vivo culture under drug pressure, which associate to clinical response prediction. Therefore, we highlight the ability of CANscript as a platform to predict clinical response to anti-angiogenic drugs, and may therefore be a logical 'testing ground' to predict clinical efficacy of antiangiogenic drugs combined with immunotherapies.
Citation Format: Smalley M, Alam N, Murmu N, Somashekhar S, Ulaganathan B, Thayakumar A, Maciejko L, Ganesh J, Lawson M, Gertje H, Shanthappa BU, Goldman A. A live tissue platform allows dynamic measurement of neovascularization and prediction of clinical response in human breast cancer samples, ex vivo [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P6-07-03.