Abstract POSTER-TECH-1124: Modeling antibody penetration into metastatic ovarian tumors after intravenous or intraperitoneal delivery

Abstract

Abstract Although targeted therapy with monoclonal antibodies has been successful against certain types of cancer, these treatments have had little success in metastatic ovarian cancer. Limited response may in part be due to suboptimal delivery of antibodies to tumors within the peritoneal cavity, as well as heterogeneous expression of surface receptors. Recent studies have shown that intraperitonal (i.p.) delivery of cisplatin has a survival advantage over intravenous (i.v.) delivery, yet few studies have examined how route of delivery influences the efficacy of antibody treatment. We have combined experimental and mathematical models to explore whether between-patient variation in tumor vascular density can predict which route of delivery is optimal for targeted treatment of metastatic ovarian tumors. The vascular density of paired bowel and omental metastatic ovarian tumors from ten patients was calculated and found to be uniform for tumors from the same patient, but vascular density among patients ranged from 1% to 10% indicating that vascular density may influence a patient’s response to treatment. Antibody penetration was measured in 3-D spheroid cultures and in vivo in an orthotopic mouse model of ovarian cancer metastasis. For our spheroid model, SKOV3.ip-GFP human ovarian cancer cells that overexpress erbB2 were allowed to form spheroids for 48 hours and then incubated with a fluorescently tagged anti-erbB2 therapeutic antibody, pertuzumab. Pertuzumab bound to the surface of the spheroids within one hour of treatment, but penetration into the center of the spheroid was slow, only reaching the center of the spheroid by 24 hours. In contrast, the chemotherapeutic agent doxorubicin completely penetrated the spheroids within 90 minutes. In our mouse model, SKOV3.ip cells were injected directly into the peritoneal cavity. Fluorescently-tagged pertuzumab was delivered via intravenous or intraperitoneal injection two weeks post-injection by which time tumors were seeded throughout the peritoneal cavity. Excised tumors were imaged on a two-photon microscope to observe antibody penetration. By 24 hours post-injection, antibody had fully penetrated into peritoneal tumors regardless of delivery route, suggesting that any advantage for i.p. delivery will likely occur at early time points. Using parameters derived from our experimental data, we have built a 3-D mathematical model, the OvTM (Ovarian Tumor Model), which simulates cell-cell interactions within an ovarian tumor. The simulated tumor can be avascular to simulate penetration of drugs into spheroids within the ascites fluid, or it can be populated with vessels that are matched to the vascular density of each patient’s tumors. We can use this individualized model to investigate the effect of different modes of drug delivery on drug distribution. Ultimately, we hope that our model will provide improved methods for stratifying patients by treatment route in order to optimize antibody penetration into metastatic ovarian tumors. Citation Format: Mara P. Steinkamp, Kimberly Kanigel Winner, Melanie Moses, Yi Jiang, Bridget S. Wilso. Modeling antibody penetration into metastatic ovarian tumors after intravenous or intraperitoneal delivery [abstract]. In: Proceedings of the 10th Biennial Ovarian Cancer Research Symposium; Sep 8-9, 2014; Seattle, WA. Philadelphia (PA): AACR; Clin Cancer Res 2015;21(16 Suppl):Abstract nr POSTER-TECH-1124.