Abstract 4087: Efficacy of three checkpoint Inhibitors, five chemotherapeutic agents, and the experimental drug thiarabine against MC38 colon cancer expressing human PD-L1 in transgenic C57BL/6 mice expressing human PD-1 and PD-L1 checkpoint genes

Abstract

Abstract Human checkpoint genes PD-1 and PD-L1 are the target of extensive research for treating human cancers. Checkpoint inhibitor therapy shows great promise in the clinic. Combining checkpoint inhibitors with standard chemotherapeutic agents is an important for these inhibitors. Animal models that use human clinical antibodies and chemotherapeutic agents will be beneficial for evaluating new therapies. A novel transgenic mouse model is described that expresses human PD-1 and PD-L1 in place of the murine genes. Additionally, a murine MC38 colon tumor cell line was modified to express human PD-L1. We initially evaluated early treatment efficacy of human check point antibodies nivolumab, pembrolizumab, and atezolizumab in the transgenic animal model with modified MC38 colon cancer cells. Five hundred thousand transgenic MC38 cells were implanted in the C57BL/6 PD-1/PD-L1 knock in mice and allowed to grow for three days. Tumor-implanted mice were treated with nivolumab and pembrolizumab at 100 µg per animal and atezolizumab at 1 mg per animal on Days 3, 7, 10, and 14. Treatment with nivolumab and pembrolizumab caused tumor regression by Day 17. Growth inhibition was 84%, 58%, and 94% on Day 17 for nivolumab, atezolizumab, and pembrolizumab, respectively, compared to the control animals. There was no significant body weight loss and no signs of toxicity in any of the treated animals. To extend the utility of the model we implanted mice and allowed the tumors to grow to 63 to 126 mg before treatment initiation (Day 10). Similar results were observed in established tumors as for early treatment with the PD-1 inhibitors nivolumab and pembrolizuman. In contrast there was little activity from the PD-L1 inhibitor atezolizumab in established tumors. To prepare for combination studies, we also tested cisplatin, irinotecan, gemcitabine, decitabine, rucaparib, and thiarabine near the MTD for each agent with various treatment schedules against the established tumors. The doses for cisplatin, irinotecan and decitabine resulted in excess animal deaths and require reduced doses for combination therapies. Gemcitabine, decitabine, and thiarabine (in house development drug) were too active with complete regressions observed; the doses for combination treatment will be reduced. Due to the effectiveness of the PD-1 checkpoint inhibitors the number of doses will be reduced from four to two treatments in combination studies; no change for atezolizumab. We have shown that PD-1 inhibitors are effective in treating genetically modified MC38 colon tumors in transgenic mice in early treatment and in established tumors while the PD-L1 inhibitor was only effective in early treatment. We also have identified doses and schedules of five chemotherapeutic agents to use in future combination studies in this model. Citation Format: Murray A. Stackhouse, Ted Green, Jay Liu, Charlotte Hammond, LaJuana Durbin, Ana Chen, Jerry Zhou, Mike Koratich. Efficacy of three checkpoint Inhibitors, five chemotherapeutic agents, and the experimental drug thiarabine against MC38 colon cancer expressing human PD-L1 in transgenic C57BL/6 mice expressing human PD-1 and PD-L1 checkpoint genes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4087.