scCNVseq Datasets Supporting Tumor Intrinsic Mechanisms of Antigen Escape to Anti-BCMA and Anti-GPRC5D Targeted Immunotherapies in Multiple Myeloma

Multiple myeloma (MM) immune escape resulting from B cell maturation antigen (BCMA) target loss is considered to be a rare event that mediates MM resistance to anti-BCMA chimeric antigen receptor T cell (CAR T) or bispecific T cell engager (TCE) therapies. Emerging clinical data also suggests that downregulation of G protein coupled receptor family C group 5 member D (GPRC5D), another promising target antigen expressed on MM cells, is observed in patients at relapse post anti-GPRC5D CAR T, but the genomic mechanisms that underlie GPRC5D loss has not been described. In order to examine the tumor intrinsic factors that promote MM antigen escape, we performed combined bulk whole genome sequencing and single cell copy number variation analysis on CD138+ cells from bone marrow aspirates of patients before and after relapse from anti-BCMA or -GPRC5D CAR T/ TCE. We describe five cases with distinct biallelic events on TNFRSF17 at MM relapse after CAR T/ TCE. In addition to focal biallelic deletions at the TNFRSF17 locus acquired at relapse, BCMA negative clones can emerge from the selective expansion of subclones with homozygous TNFRSF17 loss that exist prior to any anti-BCMA therapy exposure. Furthermore, we corroborate with functional data to demonstrate that three different non-truncating mutations in the extracellular domain of BCMA negates the efficacies of BCMA directed TCEs and mediate disease relapse in patients. With respect to GPRC5D, we report four cases of MM relapse with biallelic loss of GPRC5D following anti-GPRC5DxCD3e. Our data support that immunoselection of BCMA negative or mutant clones post anti-BCMA therapies may be more frequent than currently accepted in the field, and that an all or none screening approach for BCMA expression is inadequate to detect pertinent mutations that affect patient response to targeted therapies. We also highlight the importance of developing immunotherapies targeting novel and non-redundant epitopes or antigens in MM Overall design: Experiments were conducted on primary multiple myeloma patients bone marrow samples provided after informed consent, in accordance with the Declaration of Helsinki and following the approval by the Medical Center Institutional Review Board. Unbiased mRNA and DNA profiling were conducted by using scRNA-seq and sc-CNV-seq from the GemCode system (10x Genomics) accordingly to the manufacturer's protocols. For single-cell DNA library generation for CNV single-cell suspensions of primary MM cells were processed according to 10X Genomics Reagent Kits User Guide (CG000153). Single cells were partitioned in a hydrogel matrix by combining with a CB polymer to form Cell beads (CBs) using a microfluidic chip. Post a first encapsulation, CBs were treated to lyse the encapsulated cells and denature the genomic DNA (gDNA). The denatured gDNAs in the CB were then accessible to amplification and barcoding. A second microfluidic encapsulation step was then required to partition the CB with 10x barcode Gel Beads (GBs) to generate GEMs. Immediately after barcoding and amplification, 10x barcoded fragments were pooled and attached to standard Illumina adaptors. For all the single cell methods, quality control and quantification was performed using a Kapa Library Quantification qPCR kit (Kapa Biosystems) on a BioRad qPCR instrument prior to preparing a single pool containing equal amounts of each library. This pool was then subjected to on-board cluster formation and sequencing on an Illumina NextSeq 500 sequencer with a high-output v2.5 150 sequencing kit for RNA-seq and 300 cycle sequencing kit for CNV-seq as per the standard Illumina protocols. After sequencing, the bcl data was converted to fastq data files using the Illumina BCL2FASTQ utility. Samples were processed with CellRanger suite v3.0 and downstream analyses were realized with Seurat, Loupe and custom R packages.

多发性骨髓瘤（Multiple myeloma, MM）因B细胞成熟抗原（B cell maturation antigen, BCMA）靶标丢失引发的免疫逃逸，既往被认为是介导MM对抗BCMA嵌合抗原受体T细胞（chimeric antigen receptor T cell, CAR T）或双特异性T细胞衔接蛋白（bispecific T cell engager, TCE）治疗产生耐药的罕见事件。此外，新兴临床数据显示，在接受抗GPRC5D CAR T治疗后复发的患者体内，可观察到MM细胞上另一极具潜力的靶标抗原——G蛋白偶联受体家族C组5成员D（G protein coupled receptor family C group 5 member D, GPRC5D）的表达下调，但GPRC5D丢失背后的基因组机制至今尚未被阐明。

为探究促进MM抗原逃逸的肿瘤内在因素，我们对接受抗BCMA或抗GPRC5D CAR T/TCE治疗后复发患者的骨髓抽吸物中的CD138阳性细胞，分别在复发前后进行了批量全基因组测序与单细胞拷贝数变异分析。我们报道了5例在接受CAR T/TCE治疗后复发的MM患者中，TNFRSF17基因出现不同双等位基因变异事件的病例。除了复发时获得的TNFRSF17位点局灶性双等位基因缺失外，BCMA阴性克隆可由预先存在纯合TNFRSF17缺失的肿瘤亚克隆选择性扩增产生，而这类亚克隆在接受任何抗BCMA治疗前就已存在。此外，我们通过功能实验验证证实：BCMA胞外域的3种不同非截短突变，会抵消抗BCMA TCE的治疗效果，并介导患者疾病复发。

针对GPRC5D，我们报道了4例接受抗GPRC5D×CD3ε治疗后复发的MM患者，其体内出现GPRC5D双等位基因丢失。我们的研究数据表明，抗BCMA治疗后免疫筛选出的BCMA阴性或突变克隆的发生率，可能比当前肿瘤学界的认知更高；且仅通过“全有或全无”的BCMA表达筛查模式，不足以检测出影响患者靶向治疗应答的相关突变。我们同时强调，开发针对MM新型非冗余表位或抗原的免疫治疗手段具有重要临床意义。

### 实验整体设计
本研究所有操作均遵循《赫尔辛基宣言》，且经医学中心伦理审查委员会批准，所有原发性多发性骨髓瘤患者骨髓样本均在获取知情同意后提供。我们按照制造商提供的实验方案，使用GemCode系统（10x Genomics）的单细胞RNA测序（single-cell RNA sequencing, scRNA-seq）与单细胞拷贝数变异测序（single-cell copy number variation sequencing, sc-CNV-seq）技术，对样本进行无偏倚的mRNA与DNA组学分析。

针对用于拷贝数变异分析的单细胞DNA文库构建：原代MM细胞的单细胞悬液按照10x Genomics试剂试剂盒用户指南（CG000153）进行处理。通过微流控芯片将单细胞与CB聚合物结合，使其包裹于水凝胶基质中，形成细胞微珠（Cell beads, CBs）。首次包裹完成后，对细胞微珠进行处理以裂解包裹的细胞并变性基因组DNA（genomic DNA, gDNA）。细胞微珠内变性的基因组DNA随后可进行扩增与条形码标记。随后需进行第二次微流控包裹步骤，将细胞微珠与10x条形码凝胶微珠（Gel Beads, GBs）结合以生成GEMs。

条形码标记与扩增完成后，立即将带有10x条形码的DNA片段混合，并连接至标准Illumina测序接头。针对所有单细胞测序流程，在混合等量各文库以构建统一测序文库前，我们使用Kapa Library Quantification qPCR试剂盒（Kapa Biosystems）在BioRad qPCR仪器上完成质量控制与定量检测。随后按照Illumina标准流程，将该混合文库在Illumina NextSeq 500测序仪上进行上机簇生成与测序：RNA-seq采用高产出v2.5 150循环测序试剂盒，CNV-seq采用300循环测序试剂盒。

测序完成后，使用Illumina BCL2FASTQ工具将bcl格式的测序数据转换为fastq格式文件。我们使用CellRanger套件v3.0对样本进行预处理，并通过Seurat、Loupe与自定义R包完成后续生物信息学分析。