Abstract B047: Pancreatic cancer model for drug development and personalized treatment: Progress of the Tumor Model and Drug Development Core (TMDDC) of CaRE2

Abstract

Challenges with studies of patient-derived xenografts (PDX) to screen best agents for treatment include the prolonged growth time and the long follow-up for evaluation of tumor response. A liquid biopsy assay that might within days evaluate tumor response could revolutionize selection of cancer treatment regimens. Personalized therapy would overcome disparities, genetic and ethnic, that impact patients. We have developed a bead-based branched-DNA assay with the needed accuracy and precision to detect and quantify cfDNA. The assay is available as a Clinical Laboratory Improvement Amendments (CLIA)-ready kit to measure both tumor and host cell kill. In this study, we measured host and tumor circulating cfDNA in mice bearing human pancreatic cancer cell xenografts, showing that cfDNA increases shortly after radiation (IR) or gemcitabine treatment. The assay successfully compared gemcitabine to an acetyl gemcitabine developed in the Agyare Lab. NSG mice were subcutaneously implanted with either PANC1 or BxPc3 human pancreatic cancer cells. When tumor volume reached 1000 mm3, mice received either 12 Gy total body radiation (TBI), or Gemcitabine (GEM) or acetyl-GEM at 150 mg/kg BW. Blood samples were collected prior to treatment and early time points post-treatment. Plasma murine (host) and human (tumor) cfDNA concentrations were based on the measurement of the B1 (murine) or Alu (human) retrotransposon sequences in the cfDNA using the QuantiDNA RadTox cfDNA Test Kits (Diacarta, Inc., CA, USA). Both host (B1) and tumor (Alu) cfDNA increased after treatment with radiation or GEM/acetyl-GEM. A greater increase in B1 after IR indicates more host cell damage, compared to GEM in both cell lines. Perhaps unexpectedly, there was enhanced tumor response to a second dose of GEM when given 4 days after the first dose as demonstrated by the greater Alu cfDNA concentrations. The acetyl-GEM derivative had a similar tumor response but lower toxicity than GEM suggesting an improved therapeutic gain. QuantiDNA tests for host and tumor cfDNA provided a quantitative method to evaluate tumor response in a murine xenograft model. Testing for B1 allowed us to monitor murine host toxicity. Testing for Alu allowed the discovery of a two-hit mechanism for GEM response. Both IR and GEM, common treatments for pancreatic cancer, resulted in increased circulating human Alu cfDNA within hours of treatment. Therefore, this test shows promise as a model to identify treatment regimens with greater therapeutic gain. Paul Okunieff, Steven G. Swarts, Lori Rice, Hernan Flores-Rozas, Steven B. Zhang, Zhenhuan Zhang, Christine Jarosik, Edward Agyare. Pancreatic cancer model for drug development and personalized treatment: Progress of the Tumor Model and Drug Development Core (TMDDC) of CaRE2 [abstract]. In: Proceedings of the 19th AACR Conference on the Science of Cancer Health Disparities; 2026 Oct 18-21; Atlanta, GA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2026;35(10 Suppl):Abstract nr B047.