Abstract 6653: Analysis of selective response of anti-PD-L1 treatment in huPBMC-NCG mice reconstituted with different donors

Abstract

Abstract Despite the success of programmed death ligand 1 (PD-L1) and programmed death 1 (PD-1) blockage as cancer therapies, the overall clinical response rates are still unsatisfying, with only a 25% response rate in most cancers. The underlying mechanisms of non-responsiveness towards anti-PD-L1 treatment have been extensively studied and associated with many factors, including PD1/PD-L1 expression level, tumor mutation burden, IFN signaling, loss of MHC-I, tumor microenvironment (immune infiltration, suppressive immune system), and so on. Immunodeficient mice such as the NSG and NCG are commonly used to study the efficacy of cancer immunotherapies. These mice, reconstituted with either human PBMC or HSC, act as an avatar to provide a window of the potential agent efficacy at the preclinical level. However, due to the challenge of obtaining an HSC/PBMC donor, and often cost and time can also be a limiting factor, many studies only use 2-5 human donors. Here we present data demonstrating the significant variation observed between donor-to-donor, highlighting the importance of testing multiple donors. We reconstituted immune system and tumor tissue humanized mouse models using peripheral blood mononuclear cells (PBMCs), and cell line-derived xenografts (CDX) in severe immunodeficiency NCG mice. Anti-tumor efficacy of anti-PD-L1 (Tecentriq) in huPBMC-NCG mice engrafted with several tumor cells including MDA-MB-231, HCC827, NUGC4, and some other CDX cell lines was tested in our system. In the cases of MDA-MB-231, HCC827, and NUGC4, different donors were enrolled in the same study, and tumor growth inhibition measured by tumor volume (TGITV) of 3% to 52.40% (MDA-MB-231), 5% to 33% (HCC827) and 4.75% to 20.74% (NUGC4) were observed. Especially, In the case of MDA-MB-231, 6 donors were enrolled in this study at the same time. About 40% of immune infiltration was first observed in tumors. High expression of PD1 on immune cells and PDL1 on tumor cells were also detected from different donors simultaneously. Although the similar immune infiltration and PD1/PDL1 expression, the TGITV of 6 donors varied between 3% and 52.40%, suggesting the donor-dependent response to anti-PD-L1 treatment, which is similar to the clinical trials. Our data showed diverse in vivo responses when multiple human PBMC donors were used for preclinical agent evaluation. While unsurprising, it showed the critical need to include various human PBMC donors when an agent is in preclinical status because the donors' immune profile may dictate the study's outcome. This can be divided into two parts, on the one hand, donor screening for favorable efficacy response donors in advance is an alternative approach to ensure the positive efficacy of anti-PDL1 blockade relative therapy. On the other hand, the low-efficacy response donors also can be candidates for anti-PD1/PDL1 combination therapy. Citation Format: Tingting Gu, Weiwei Yu, Hongyan Sun, Fang Zhu, Huiyi Wang, Mengting Wang, Shuai Li, Jianming Xu, Santi Chen, Zhiying Li, Mark Wade Moore, Cunxiang Ju, Hongyu Wang, Jing Zhao, Xiang Gao. Analysis of selective response of anti-PD-L1 treatment in huPBMC-NCG mice reconstituted with different donors [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6653.