Cell-based models of 'Cytokine Release Syndrome' endorse CD40L and GM-CSF knockout in CAR T cells as mitigation strategy

Chimeric antigen receptor (CAR) T cells are engineered lymphocytes able to recognize and eradicate cancer cells. While CAR T cell therapy has shown promising outcomes in patients with hematologic malignancies, it can cause undesirable side-effects such as cytokine release syndrome (CRS). CRS is triggered by CAR T cell-based activation of monocytes, which are stimulated via the CD40L–CD40R axis or via uptake of GM-CSF to secrete pro-inflammatory cytokines. Mouse models have been used to model CRS but they are labor intensive and not amenable to screening approaches. To overcome this challenge, we established two cell-based CRS in vitro models that entail the co-culturing of leukemic B cells with CD19-targeting CAR T cells and primary monocytes from the same donor. Upon antigen encounter, CAR T cells upregulated CD40L and GM-CSF which stimulated the monocytes to release IL-6. To endorse these models, we demonstrated that neutralizing antibodies or genetic disruption of the CD40L and/or CSF2 loci in CAR T cells using CRISPR-Cas technology significantly reduced IL-6 secretion by bystander monocytes without affecting the cytolytic activity of the engineered lymphocytes. Overall, our cell-based models were able to recapitulate CRS in vitro, allowing us to validate mitigation strategies based on antibodies or genome editing. Overall design: CAR T cells, produced via lentiviral transduction were treated with two RNP complexes (Cas9) targeting close sites within CSF2 and/or CD40L. CAST-Seq was performed on cells treated with either one or both nuclease pairs looking at the targeted loci in each sample to assess the formation of stable translocations between the on-target site potential off-target sites.

嵌合抗原受体（Chimeric Antigen Receptor, CAR）T细胞是一类可识别并清除癌细胞的工程化淋巴细胞。CAR-T细胞疗法在血液系统恶性肿瘤患者中已展现出颇具前景的治疗效果，但也可能引发细胞因子释放综合征（Cytokine Release Syndrome, CRS）等不良副作用。CRS的触发机制为：基于CAR-T细胞的单核细胞活化，单核细胞可通过CD40L-CD40R轴或摄取粒细胞-巨噬细胞集落刺激因子（Granulocyte-Macrophage Colony-Stimulating Factor, GM-CSF）被激活，进而分泌促炎细胞因子。既往研究虽已利用小鼠模型构建CRS体外模型，但此类模型操作繁琐且无法适配筛选类实验方案。为解决这一难题，本研究建立了两种基于细胞的CRS体外模型：将白血病B细胞与靶向CD19的CAR-T细胞、来自同一供体的原代单核细胞进行共培养。当遭遇抗原时，CAR-T细胞会上调CD40L与GM-CSF的表达，进而刺激单核细胞释放白细胞介素6（Interleukin-6, IL-6）。为验证该模型的有效性，本研究证实，利用CRISPR-Cas技术对CAR-T细胞中的CD40L和/或CSF2基因座进行中和抗体处理或基因敲除，可显著降低旁观者单核细胞的IL-6分泌水平，且不会影响工程化淋巴细胞的细胞溶解活性。总体而言，本研究所建立的细胞模型可在体外重现CRS病理过程，从而能够验证基于抗体或基因组编辑的缓解策略。整体实验设计：通过慢病毒转导制备的CAR-T细胞，用两种靶向CSF2和/或CD40L基因座内邻近位点的核糖核蛋白（Ribonucleoprotein Complex, RNP）复合物（Cas9）进行处理。对仅接受单一核酸酶复合物或同时接受两种复合物处理的细胞进行CAST-Seq检测，分析各样本中的靶向基因座，以评估靶向位点与潜在脱靶位点之间稳定易位的形成情况。