Abstract 1631: Bioluminescence imaging enhances analysis of drug response in a patient-derived xenograft model of pediatric acute lymphoblastic leukemia

Abstract

Abstract Despite the success of current therapy, 15-20% of pediatric acute lymphoblastic leukemia (ALL) patients fail treatment. Since the number of new agents for testing outweighs the availability of patients for clinical trials, robust preclinical models are required to prioritize effective drugs. Patient-derived xenografts (PDXs) of ALL in immune-deficient mice are a clinically relevant testing model as they recapitulate original disease characteristics and allow monitoring of engraftment and drug response by flow cytometric enumeration of the% human vs mouse CD45+ cells in the mouse peripheral blood (PB). However, factors such as leukemic harbor and relapse sites are overlooked, and this model may be biased towards drugs that clear circulating lymphocytes. This study aimed to overcome these shortfalls by introducing a bioluminescence imaging (BLI) capability into a pediatric ALL PDX to enhance monitoring of leukemia burden. Firefly luciferase was incorporated via lentiviral transduction, and BLI and PB measures used to track engraftment and response to an induction-type drug regimen; vincristine (0.15 mg/kg weekly x 4), dexamethasone (5 mg/kg daily x 5 × 4) and L-asparaginase (1,000 KU/kg daily x 5 × 4) (VXL). Additional groups received 0.5, 0.25 and 0.125x this dose for 2 weeks. Engraftment by BLI and PB correlated strongly (R2 = 0.8, p = <0.0001) yet BLI detected engraftment 6 weeks before PB. VXL response was scored using PB measures and stringent response criteria used by the Pediatric Preclinical Testing Program, modeled after the clinical setting. The 0.125, 0.25, 0.5 and 1x groups scored Progressive Disease 2, Stable Disease, Complete Response (CR) and Maintained CR, respectively. BLI was able to quantify regression with greater range and sensitivity than PB, highlighting a large difference in regression between the 1x and 0.5x groups. A median 3.2-fold decrease in whole-body bioluminescence (photons/sec/steradian/cm2) was seen in the 0.5x group at treatment end relative to Day 0, compared to a >2,000-fold decrease in the 1x group. Autopsy of mice at Day 28 revealed a median of 0.8% blasts in the bone marrow (BM) in the 1x group compared to 35% in the 0.5x group. Since the clinical definition of a CR is <5% blasts in the BM at Day 28, the 0.5x group did not achieve a CR by this classification, despite scoring a CR by PB measurements. BLI supported this finding, revealing >2000-fold decrease in BM bioluminescence in the 1x group relative to Day 0, and a 1.3-fold increase in the 0.5x group. These results show that BLI enhances the sensitivity of drug efficacy testing in this PDX model, and allows quantification of efficacy in discreet leukemic harbor sites such as the BM. This work suggests BLI can increase the stringency of preclinical drug testing to accurately prioritize agents for clinical trials. Citation Format: Luke Jones, Jennifer Richmond, Kathryn Evans, Richard B. Lock. Bioluminescence imaging enhances analysis of drug response in a patient-derived xenograft model of pediatric acute lymphoblastic leukemia. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1631. doi:10.1158/1538-7445.AM2015-1631