Single-Cell Analysis Reveals Ide-cel and Cilta-cel Characteristics That Influence Efficacy

Single-Cell Analysis Reveals Ide-cel and Cilta-cel Characteristics That Influence Efficacy | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Article Single-Cell Analysis Reveals Ide-cel and Cilta-cel Characteristics That Influence Efficacy Ciara Freeman, Jerald Noble, Xiaofei Song, Meghan Menges, Julieta Abraham-Miranda, and 13 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-4994668/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Chimeric antigen receptor T-cells targeting BCMA have revolutionized the treatment of relapsed/refractory multiple myeloma (RRMM) with two approved products, idecabtagene vicleucel (ide-cel) and ciltacabtagene autoleucel (cilta-cel). To explore biological differences, we analyzed pre-infusion products (IP) and CAR-enriched peripheral blood mononuclear cells (PBMCs) at expansion using single-cell RNA sequencing (scRNAseq) from 52 samples. Post-quality control 247,500 cells (117,530 CD4, 80,939 CD8) were analyzed. We found that ide-cel IPs from durable responders (DR) had higher construct expression, enhanced NFKB signaling, and anti-apoptotic signatures, correlating with improved progression free survival. CAR + ide-cel PBMCs in DRs showed upregulated ribosomal genes and higher CD27, KLF2, TCF7 expression. Relative to ide-cel, cilta-cel CAR + cells showed higher expression of CD27, GZMK, TCF7, and a 4-fold increase in CAR expression. In addition, the TCR repertoire was less clonal and more diverse. This study elucidates the distinct characteristics of ide-cel and cilta-cel, offering insights into their differing clinical efficacy. Health sciences/Medical research/Translational research Biological sciences/Cancer/Haematological cancer/Myeloma Biological sciences/Cell biology/Cell signalling Single-cell RNA sequencing Ide-cel Abecma Cilta-cel Carvykti CAR T-cell B-Cell Maturation Antigen Promoter CAR density clonal diversity Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Introduction Multiple myeloma remains incurable, however, significant advances over the past two decades have led to major improvements in the depth and durability of responses achieved with modern treatment combinations[ 1 – 3 ]. For patients who have progressed after exposure to the major classes of agents, the introduction of two chimeric antigen receptor T-cell (CAR-T) products which target B-cell maturation antigen (BCMA) has radically improved outcomes and led to approvals for those with relapsed and refractory multiple myeloma (RRMM)[ 4 – 6 ]. More recently, both products demonstrated a significant benefit in randomized trials, outperforming standard-of-care regimens, leading to expanded indications and altering the therapeutic paradigm for RRMM [ 7 – 9 ]. While these products have not been evaluated head-to-head, indirect comparisons have been performed which suggested there may be improved depth and durability of responses with ciltacabtagene autoleucel (cilta-cel) compared to idecabtagene vicleucel (ide-cel)[ 10 ]. Clinical trial outcomes also suggest differential efficacy, although this could be influenced by differences in the enrolled patient populations and trial design. These products employ antigen (BCMA) binding domains that differ in both modality and target binding valency, and in aspects of the manufacturing process. Understanding which attributes are beneficial for patients could potentiate enhanced treatment selection[ 11 ]. In addition, neither agent has proven curative, thus understanding the mechanisms underpinning therapeutic success and failure are paramount. We sought to evaluate the characteristics of the products using single-cell RNA sequencing (scRNAseq) analysis in a large, real-world cohort of patients with RRMM. Results Patient and product characteristics We identified 46 RRMM patients treated per standard of care, with BCMA targeting CAR-T cell products (Extended Data Table 1). Only ide-cel treated patients, N = 40, had pre-infusion products (IP) available for sequencing. We identified 6 matched peripheral blood post-infusion samples obtained during early expansion-phase (days 7–14 post-infusion) from which CAR-positive cells were extracted and encapsulated (Fig. 1 A/Supplementary Fig. 1) for scRNAseq. We identified 6 additional patients treated with cilta-cel with post-infusion samples during the same period, and similarly extracted CAR-positive cells. Median time of collection for ide-cel treated patients was 9 days (range 7–14) and for cilta-cel treated patients 13 days (range 11–14) aligned with rising lymphocyte count measurements in peripheral blood. All patients in the cohort were triple-class exposed, and had received a minimum of four prior lines of therapy (median 6, range 4–13). In total, 28% had evidence of high tumor burden (≥ 50% involvement in restaging bone marrow biopsy prior to lymphodepletion [LD]) and 17% had evidence of extramedullary disease on pre-treatment imaging. The median duration of follow up for all living patients was 19.87 months (range 7.1-35.87) and median PFS for the whole cohort was reached at 15.27 months (95% CI: 8.83–22.47). The median OS for this cohort was not reached, but 18-months estimated at 77% (95% CI 61–87) (Supplementary Figs. 2A/B). Cytokine release syndrome (CRS) of any grade occurred in 38 (83%) of the cohort, all grade 1 (N = 26) or grade 2 (N = 12). Similarly, immune effector cell-associated neurotoxicity syndrome (ICANS) occurred in 6 patients (13%), grade 2 (N = 3), grade 3 (N = 2) and 1 grade 4 event (in a patient with pre-treatment evidence of CNS involvement). Of the patients treated with cilta-cel, one patient developed delayed neurotoxicity with evidence of movement and neurocognitive treatment-emergent adverse events that had not resolved at the time of death. At last assessment prior to death, this patient remained in complete response and had not yet reached the landmark timepoint. Single cell RNAseq libraries were generated from 52 samples in total; composed of 40 unique patient ide-cel infusion products (IP), and CAR-enriched peripheral blood mononuclear cells (PBMCs) from 6 ide-cel and 6 cilta-cel treated patients (Fig. 1 A). Patients were stratified as having durable response (DR) if they were still alive, without receipt of additional myeloma therapy, and without evidence of progression at 9 months following treatment. This cutoff was selected on the basis of clinical trial and real-world data suggesting that the median progression free survival post ide-cel infusion is approximately 8.8 months[ 4 , 6 , 12 ]. Patients were considered to have non-durable response (NDR) if they died from myeloma or had evidence of disease progression prior to this 9 month cutoff, apart from the patient that developed delayed neurotoxicity as outlined (Fig. 1 B). All patients had sufficient follow up to either progress (and thus be labelled NDR) or reach the 9-month landmark cutoff at the time of analysis, and of those who received ide-cel, 26 (65%) of patients had DR and 14 (35%) NDR (Fig. 1 B). Following quality control procedures and removal of batch effects (Online Methods) 247,500 cells were analyzed. Individual cells were classified as CD4 or CD8 using an in-silico gating method from Li et al[ 13 ] (Supplementary Fig. 3A) resulting in 117,530 CD4 cells and 80,939 CD8 cells (Supplementary Fig. 3B). A population of monocytes was identified; however, the majority (2536/3137) were isolated from one patient PBMC sample and were not included in subsequent analyses. The CD4 and CD8 cells were then separately integrated, clustered, and annotated based on known markers (Supplementary Figs. 4/5). This process identified 11 CD4 subtypes and 12 CD8 subtypes (Figs. 1 C-F). Pseudotime trajectory analysis supported annotations of cell subtypes and showed a clear differentiation trajectory in CD8 cells but not CD4 cells (Supplementary Fig. 6) Ide-cel product has a high CD4/CD8 ratio and response is associated with more CD4 cells that have higher CAR expression and activation associated gene expression profiles The majority of cells from ide-cel infusion product (IP) were identified as CD4, a feature which changed after infusion as the product expanded in peripheral blood (Fig. 2 A/B). CAR-positive (CAR+) cells extracted from PBMC of both ide-cel and cilta-cel patients during early expansion phase proximal to infusion were predominately CD8 (Fig. 2 A/B). We confirmed these findings by immunophenotyping both the IP and PBMC samples used to generate the scRNAseq libraries (Fig. 2 B, Supplementary Fig. 7) and found the CD4:CD8 ratios to be highly correlated between scRNAseq data and immunophenotyping (Fig. 2 C). To detect CAR + cells in single-cell RNA-seq data, a reference transcriptome was created by incorporating ide-cel and cilta-cel construct sequences from respective patents into the GRCh38 human transcriptome, with samples aligned to this reference and cells deemed CAR positive if they had at least one read aligning to the respective construct (Online Methods).) We also confirmed the frequency of CAR + cells using this approach (Supplementary Fig. 9A/B) and by immunophenotyping (Supplementary Figs. 8/9). The proportion of CAR + CD4 cells in the ide-cel IP, and the expression of the CAR construct in CAR + CD4 cells, was higher in patients who had DR compared with NDR (Fig. 2 D-E). We next sought to identify potential determinants of response in the infusion product based on cell subtype composition and compared the proportion of cells that were in each of the identified CD4 or CD8 clusters, restricted to CAR expressing cells, between DR and NDR. The composition of infusion product cell subtypes was similar between responders and non-responders (Supplementary Figs. 10/11), with products from non-responding (NDR) patients having higher proportions of proliferating CD4 cells (p = 0.063) and a cluster of CD4 cells characterized by high expression of glycolysis genes [CD4gly] (p = 0.0047). Given the paucity of composition differences in IP cell subtypes between DR and NDR patients, we sought to identify how they might vary transcriptionally and identified differentially expressed genes (DEGs) and biological pathway differences between responders in CD4 and CD8 cells and their respective subtypes. We identified upregulation of several gene sets associated with key functions in CAR + CD4 cells in patients with DR reflecting their ability to generate a robust and coordinated immune response. Key pathways included those involved in signaling and cytokine production (IL-6/JAK/STAT3, IL-2/STAT5, cytokine-cytokine receptor interaction, inflammatory response), immune cell activation (TGF-β signaling, KRAS signaling, and IFN-γ/IFN-α pathways, T-cell and Toll-like receptor signaling) and anti-apoptosis pathways TNF-α signaling via NFKB and noncanonical NFκB signaling, promoting survival and persistence (Fig. 2 F).The CAR + CD8 cells from DR patients had upregulation of interferon pathways and genes involved in MHC class II and ribosomal pathways. In addition, we noted upregulation of TCF7 regulon in CAR + CD4 and CAR + CD8 cells of DR patients which has been associated with a more favorable naïve T-cell state in CD19-directed CAR-T [ 14 ] (Fig. 2 F). In contrast, CAR + CD4 cells from NDRs demonstrated upregulation of cell proliferation genes, mTORC1 signaling, oxidative phosphorylation, glycolysis, and the TCA cycle (Fig. 2 F) suggestive of excessive proliferation while CD8 effector cells from NDR had activated hypoxia, p53 and TGFβ pathways, also suggestive of pre-existing dysfunction. Key upregulated genes in CAR + CD4 IP cells from those with DR included IFN-γ, MAL, CD2, CD69, and GZMA (Fig. 2 G, Extended Data Fig. 1 A), key for T cell signaling, activation, proliferation, and cytotoxicity suggesting a CD4 compartment primed for the generation of a more effective and comprehensive immune response post-infusion[ 15 , 16 ]. Additionally NEAT1, a long noncoding RNA whose suppression results in impaired CD4 cell differentiation via the STAT3 axis [ 17 ], and the pro-survival gene BIRC3 were upregulated in CAR + CD4 IP cells of DRs (Extended Data Fig. 1 A). Genes associated with non-durable responses in CAR + CD4 IP cells include MCM5, MCM7, and CDT1, which are involved in DNA replication and cell cycle progression, indicating potential dysregulation in cell proliferation in NDR (Extended Data Fig. 1 A). Using pseudobulk aggregation of read counts from CAR + CD4 cells of each patient, we validated the statistical significance of 8/10 of these genes (Extended Data Fig. 1 B). Differentially expressed genes between responders in specific CD4 IP subtypes have been reported in markdown format and hosted on github (Online reports). In CAR + CD8 IP cells, fewer genes were differentially expressed between responders (Fig. 2 H). Genes upregulated in CAR + CD8 DR IP included HLA class II gene HLA-DQA2, GZMK, and BIRC3, associated with enhanced antigen presentation, cytotoxic functions, and cell survival, respectively (Fig. 2 H, Extended Data Fig. 2 A). Conversely, CAR + CD8 IP cells from NDR patients had upregulation of LGALS3, and LIME1 (Extended Data Fig. 2 A), indicating potential impaired effector function via LAG3 [ 18 ] and premature activation of T-cell migration [ 19 ], respectively. GZMK and BIRC3 maintained their significance between responders when comparing expression at the pseudobulk level (Extended Data Fig. 2 B). Differentially expressed genes between responders in specific CD8 IP subtypes have been reported in markdown format and hosted on github (Online reports). Upregulation of pro-survival pathways in the ide-cel infusion product is correlated with higher density of CAR expression and improved survival outcomes To provide a sample-level signature for biological pathways of interest, we performed pseudobulk aggregation of single-cell read counts by sample, and by cell subtype within each sample, followed by single-sample gene set enrichment analysis (ssgsea). We found that products from patients with durable responses had transcriptional profiles associated with memory T cells (FOXO1 regulon) with downregulation of glycolysis in CAR + CD4 and CD8 cells (Extended Data Fig. 3 ). Additionally, cytokine/cytokine receptor pathway and cytotoxicity genes were increased in both CAR + CD4 and CD8 DR IP (Extended Data Fig. 3 ). We noted that NFKB signaling, tonic signaling, and anti-apoptosis gene signatures were upregulated in CAR + CD4 and CD8 from DR IP cells and in particular cell subtypes (Fig. 3 A, Extended Data Fig. 3 ). As there is a known promotion of CAR-T survival via 4-1BB signaling[ 20 ], we investigated the relationship between expression of the CAR and NFKB signaling and the anti-apoptosis gene signature, finding a strong correlation between CAR expression and these signatures in CAR + CD4 cells (Fig. 3 B/C) but not in CAR + CD8 cells (Supplementary Fig. 12). Tonic signaling, based on expression of a specific gene signature linked to tonic signaling [ 21 ] was also correlated with CAR expression and anti-apoptosis (Fig. 3 D/E, Extended Data Fig. 4 ). Given the association observed with these gene signatures and durable responses, we investigated their association with survival outcomes. We stratified patients as expressing these signatures above (high) or below (low) the median for all IP samples and identified that NFKB signaling and anti-apoptosis signatures in CAR + CD4 and CD8 cells were both significantly associated with progression-free survival (PFS) (Figs. 3 F-I) and the anti-apoptosis signature significantly discriminated between groups for overall survival (OS) (Extended Data Fig. 5 ). Additionally, the tonic signaling signature in CAR + CD4 and CD8 cells was associated with OS but not PFS (Extended Data Fig. 6 ). Ide-cel exhibits marked changes in cellular composition, transcriptional profiles, and clonotype dynamics following infusion Following infusion, ide-cel CAR-T demonstrate a substantial shift in CD4:CD8 ratios as previously outlined (Fig. 2 A/B, Supplementary Fig. 7), and shifts within subtypes of both CAR + CD4 and CD8 cell compartments (Supplementary Figs. 13/14). Proportionally, there was an increase in CD4em in PBMCs relative to IP (p = 0.001) and relative decreases in Th2, a population of cells defined by high expression of histone genes (CD4hist), and Tregs (Supplementary Fig. 13). Within the CD8 compartment, there were increases in effector memory (CD8em) and terminal effector memory (CD8tem) cells and decreases in stem central memory (CD8scm), CD8tc2 [ 22 , 23 ], and proliferating cells (Supplementary Fig. 14). To elucidate determinants of response following infusion, we assessed the transcriptional differences in CAR + ide-cel PBMCs between DR and NDR patients. Ribosomal genes were upregulated in the CAR + CD4 and CD8 PBMCs of DR patients and the FOXO1 regulon was upregulated in CAR + CD4 PBMCs of DR patients (Fig. 4 A). Glycolysis, fatty acid metabolism, and cell proliferation pathways were upregulated in CAR + CD4 and CD8 PBMCs of NDR patients (Fig. 4 A). CAR + CD8scm PBMCs in NDR patients were also enriched for an exhaustion signature and upregulation of interferon signaling (Fig. 4 A). Although CD8scm typically do not express exhaustion markers, the fact that these genes are differentially upregulated in NDR suggest that they are more prone to the development of exhaustion, which may contribute to their attenuated therapeutic effect. Interrogating individual genes, in patients with DR we identified upregulation of genes associated with the maintenance of a more naïve phenotype, including KLF2, CD27, TCF7, and DUSP2 in CAR + CD4 and CD8 cells (Figs. 4 B-D), and multiple ribosomal genes in the CAR + CD8 cells of DRs. In contrast, ENO1, CD38, and multiple metallothionein genes were upregulated in CD8 cells of NDRs (Figures C-D). Additionally, expression of the ide-cel construct was again upregulated in CD4 and CD8 cells of DRs (Figs. 4 B-D). Following infusion into patients, CAR + cells exhibited increases in clonality (Fig. 5 A). We analyzed transcriptional differences between clones composing a small percentage of the total clonal population in a sample (0–0.1%) versus expanding clones (> 0.1%) in all CAR + IP and PBMC ide-cel samples (Fig. 5 B, Supplementary Fig. 15). As expected, CAR + CD4 IP cells that expanded in vivo exhibited an upregulation of genes associated with protein translation and cell proliferation. Non-expanding CAR + CD4 IP cells had upregulation of hypoxic pathways and interferon signaling (Fig. 5 C). No non-expanding clones were detected in CAR + CD4 PBMC cells. Conversely, genes associated with immune response, NFKB signaling, and the TCF regulon were upregulated in the IP of expanding CD8 cells, whereas ribosomal genes and genes associated with proliferation were upregulated in non-expanding CD8 cells (Fig. 5 C). We next assessed the transcriptional differences between CAR + clones that were present in both IP and PBMC (paired) or only present at either time point (unpaired) in the IP and PBMC samples of 6 ide-cel patients (Fig. 5 D, Supplementary Fig. 16). Pathway enrichment analysis revealed unpaired clones were enriched in hypoxia, apoptosis, interferon response, and p53 pathways in CAR + CD4 and CD8 IP and PBMC cells. Paired clones were enriched in MHC class II genes in CAR + CD4 and CD8 IP and PBMC cells and ribosomal genes in CAR + CD8 IP and PBMC cells (Fig. 5 E). Additionally, paired clones were more likely to expand in CD4 IP, CD8 IP, and CD8 PBMC cells (chi square p-value < 0.0001, Fig. 5 F). Ide-cel features associated with efficacy are accentuated in cilta-cel; including higher CAR expression, enhanced NFKB and ribosomal signaling, and greater clonal diversity Given the improved depth and durability of responses reported in clinical trials enrolling RRMM patients treated with cilta-cel when compared to ide-cel [ 4 , 5 , 7 , 8 , 10 , 12 ], we investigated cell type composition, gene expression, and biological pathway differences in CAR-T cells enriched from PBMCs collected during expansion-phase between the two products. Compared to ide-cel, the expanding CAR + CD4 compartment of cilta-cel demonstrated a higher proportion of CD4em cells (p = 0.0087) while ide-cel was proportionally higher in CD4scm (p = 0.043) (Fig. 6 A, Supplementary Fig. 17). In contrast, CD8scm were higher proportionally in cilta-cel (p = 0.0087) and CD8tem trended towards an increase in ide-cel, although not reaching significance (p = 0.13, Fig. 6 B, Supplementary Fig. 18). We identified gene sets previously associated with key functions upregulated in cilta-cel when compared to ide-cel, notably in NKFB signaling and ribosomal pathways (Extended Data Fig. 7A). Compared to CAR + CD4 cells of ide-cel, CAR + CD4 cells of cilta-cel cells had higher expression of CD27, GZMK, TCF7 (Fig. 6 C, Extended Data Fig. 7B-C). CD27 and GZMK were also significantly upregulated in CAR + CD8 cells of cilta-cel (Fig. 6 C). The expression of the CAR construct was significantly higher in cilta-cel across both CD4 and CD8 subpopulations (Fig. 6 D) and we confirmed this difference in expression via flow cytometry (Fig. 6 E, Supplementary Fig. 19). Mean fluorescent intensity (MFI) was approximately four-fold higher, which suggests more CAR per cell even accounting for the additional binding domains of the cilta-cel construct. Pseudobulk aggregation of single-cell read counts by sample followed by ssgsea confirmed NFKB signaling and ribosomal pathways to be significantly higher in cilta-cel when compared with ide-cel (Fig. 6 F). We also noted that the TCR repertoire of cilta-cel was far less clonal than ide-cel. The ide-cel TCR repertoires were comprised of medium and larger expanding clones (Fig. 5 G) while the TCR repertoire was significantly more diverse in CAR + CD8 cells of cilta-cel (Fig. 5 H). We also found the diversity of the TCR repertoire to be higher in CAR + cells than in the CAR- cells in all PBMC and IP cells (Supplementary Fig. 20). Discussion This single cell analysis provides key insights into the product characteristics that are important for durable responses, shedding light on the different outcomes that have been reported in clinical trials to date. Our findings from the infusion product suggest that cells highly expressing NFKB signaling signatures and the downstream, pro-survival target genes of this pathway (e.g. BIRC3)[ 24 ], in addition to gene signatures associated with immune response and tonic signaling, prime CAR-T for both activation and survival following infusion. Additionally, we show that upregulation of cell proliferation pathways in CAR + CD4 IP cells is associated with poor outcome following treatment. In CAR-T with CD28 co-stimulatory domains, tonic signaling can lead to exhaustion during ex vivo expansion[ 25 ]. However, its effect in CAR-T utilizing 4-1BB co-stimulatory domains remains less well defined. Rodriguez-Marquez et al. demonstrated that increased density of CAR (CAR high T-cells) produced by higher numbers of viral integrations could trigger tonic signaling[ 26 ]. These CAR high T-cells also produced more cytokines and demonstrated an increase in cytotoxicity in vitro . Our data also showed increased CAR-expression in the pre-infusion products of patients that had durable responses. Although there exists concern that tonic signaling has potentially negative consequences, others have challenged this dogma[ 21 , 27 ]. In this context, tonic signaling may be desirable in particular for constructs with a 4-1BB co-stimulatory domain[ 21 , 25 , 28 , 29 ]. Singh et al. (2021) found that tonic 4-1BB signaling can be protective against dysfunction and actually enhance CAR T cell function, which would support our findings[ 28 ]. The upregulation of ribosomal pathways in the expanding products of durable responders also highlights the importance of sustained protein synthesis for durable anti-tumor activity. This hypothesis is supported by gene expression, functional correlation, and proteomic studies, which demonstrate that activated T cells allocate substantial bioenergetic resources to ribosome biogenesis, increasing ribosomal output more than 13-fold following activation[ 30 – 35 ]. Taken together, these data suggest that a strong signal generated through the CAR combined with increased ribosomal capacity can lead to improved function. In addition, we confirmed the positive influence of higher levels of CD27 expression on both CAR + CD4 and CD8 PBMC cells, shown to promote T cell expansion not by affecting cell cycle activity, but by stimulating the survival of activated T cells[ 36 ]. Harnessing the potential of ectopic CD27 expression is already under evaluation with CD27-Armored BCMA-CAR T showing promising results[ 37 ]. A striking finding in our data was that these mechanisms were upregulated in CAR + PBMCs obtained from patients treated with cilta-cel when compared to ide-cel CAR + cells. We observed more CD27 and GZMK expression in both CAR + CD4 and CD8 cells, and increased ribosomal and NKFB pathway expression in cilta-cel. In conjunction with this, we also observed dramatic differences in the expression of the CAR construct. The detailed specifics of manufacturing of both products are proprietary, however, it is known that they differ in their promoter utilized in their vector [ 38 – 40 ]. Cilta-cel utilizes an elongation factor-1α (EF-1α) promoter whereas ide-cel uses an MND promoter (myeloproliferative sarcoma virus MPSV enhancer, negative control region NCR deletion, d1587rev primer binding site replacement). Ho et al. (2021) performed a series of experiments comparing these two promoters’ effect on CAR density and cell functionality[ 41 ]. The EF-1α resulted in significantly higher density of CAR expression, more cytokine secretion, and equivalent cytotoxicity. Authors concluded that MND could generate a safer product, which clinically has been shown in real-world experience with reduced high grade toxicity and non-relapse mortality observed in patients treated with ide-cel[ 6 , 12 ]. However, this is at the expense of depth and durability of response which, in the case of cilta-cel, may be driven by higher levels of CAR expression and the concomitant downstream effects. Other relevant differences include the greater proportion of small clones and greater clonal diversity seen in the expanding cilta-cel compared with ide-cel products. This could confer some additional advantages to the former, as central and effector memory CAR-T in patients with long-term persistence remain highly polyclonal, whereas patients with limited CAR-T persistence were found to have less diversity[ 42 ]. We appreciate there are some aspects that remain unexamined in this work. We did not evaluate peripheral blood samples taken across several different timepoints and selected samples were estimated at peak expansion based on increasing lymphocyte counts. We were unable to examine the pre-infusion cilta-cel product or long term persistence, and this is certainly an area for future research. This is, to the best of our knowledge, the largest single cell analysis of real-world anti-BCMA CAR-T therapy that has been performed to date and provides key insights into the differences that underpin the clinical results that are being reported as the trial and real-world data matures. We identify features associated with patient outcomes in the starting product and that could be optimized for future iterations of therapy. These data will serve as a resource for further academic investigations. Increased transparency in providing detailed information post-CAR-T approval, including critical aspects of manufacturing or other information about the product generated for patients, could optimize research efforts, improving patient outcomes. Online Methods Patient sample collection All patients had a diagnosis of RRMM and consented to prospective sample collection and research database protocols which were Institutional Review Board (IRB) approved by the University of South Florida (USF). All patient samples were collected at Moffitt Cancer Center under approved Institutional Review Board (IRB) protocols. Excess infusion product was collected via elution of CAR-T infusion product bags following patient treatment. Patient PBMC of post-infusion ide-cel and cilta-cel patients was collected at 7-14 days post-infusion and 11-14 days post infusion for ide-cel and cilta-cel patients, respectively. Flow cytometry Cryopreserved cells from ide-cel infusion product were removed from liquid nitrogen and rapidly thawed in a 37˚C water bath prior to being transferred to 10ml of pre-warmed complete media to remove excess of DMSO. Cells were then centrifugated 5 min at 1500 rpm, the cell pellet was washed twice with PBS, and resuspended in 100 µl of a solution containing 1X Live/Dead Fixable green cell stain (Invitrogen, ThermoFisher Scientific) and 1 µL of human Fc block (BD) and incubated for 30 min at room temperature. Surface staining was performed for 30 min at 4°C with antibody mix in MACS buffer with 0.5% BSA (Miltenyi Biotec). Cells were then fixed using IC Fixation Buffer (eBioscience) for 30 min at RT, washed 1X Permeabilization Buffer (eBioscience), and intracellular staining was performed for 30 min at 4°C with antibody mix in 1X Permeabilization Buffer (eBioscience). The following monoclonal antibodies against human antigens were obtained from BD Biosciences: anti-CD3 (SK7), anti-CD8 (SK1), and anti-CD4 (L200). Cells were then assessed for the fraction of CD3 + CD4 + and CD3 + CD8 + cells. Frozen PBMC samples from 12 myeloma patients (6 Abecma treated and 6 Carykti treated) collected at estimated time of peak CAR T cell expansion were thawed and a Pan T cell selection was performed per manufacturer’s protocol (Miltenyi). The negative fraction representing enriched T cells was surface stained with fluorochome conjugated antibodies against CD3, CD4, CD8 (BD) and FITC-conjugated soluble BCMA (AcroBio) for 30 minutes in the dark at 4C. Cells were washed twice in MACS buffer and resuspended in 90uL MACS buffer + 10uL anti-FITC magnetic microbeads per sample and incubated at 4C for 15 minutes (Miltenyi). Labeled cells were run through MACS MS columns per manufacturer’s protocol and the positive fraction representing enriched CAR T cells was collected. Cells were incubated with Live/Dead Near IR fixable cell stain (ThermoFisher) for 30 min at room temperature. Cells were washed and incubated in 100uL/tube BD Cytofix buffer for 30 minutes at 4C, washed, resuspended in FACS buffer and stored at 4C in the dark until acquisition. Samples were acquired on a BD FACS Symphony flow cytometer and analyzed using FlowJo software. Single-cell RNA-seq and V(D)J library preparation and sequencing Single-cell RNA-sequencing was performed using the 10X Genomics Chromium System (10X Genomics, Pleasanton, CA) by the Molecular Genomics Core at the Moffitt Cancer Center and Research Institute. Cryopreserved infusion product cells were rapidly thawed in a 37C water bath, washed twice by centrifugation, and resuspended in MACS buffer (Miltenyi). Cryopreserved PBMC from peak expansion time points were thawed and washed twice in MACS buffer. A Pan T cell isolation followed by a CAR T cell enrichment by FITC-conjugated soluble BCMA staining and subsequent anti-FITC microbead magnetic column separation was performed as described in the flow cytometry methods above. Following cell thawing with or without enrichment as indicated, cell suspensions were washed twice with 1X PBS (calcium and magnesium-free) containing 0.04% weight/volume BSA. The cells were then resuspended in the same buffer following the 10X Genomics cell preparation guide. The cell viability and counts were obtained by AO/PI dual fluorescent staining and visualization on the Nexcelom Cellometer K2 (Nexcelom Bioscience LLC, Lawrence, MA). Cells were then loaded onto the 10X Genomics Chromium Single Cell Controller at a concentration of 1,000 cells/µl to encapsulate 5,000 cells per sample. Single cells, reagents, and 10X Genomics gel beads were encapsulated into individual nanoliter-sized Gelbeads in Emulsion (GEMs), and reverse transcription of poly-adenylated mRNA was performed inside each droplet at 53°C. The cDNA libraries were then completed in a single bulk reaction by following the 10X Genomics Chromium NextGEM Single Cell 5’ Reagent Kit v2 user guide. The TCR V(D)J enrichments from cDNA were performed using the 10x Genomics Chromium single cell human TCR amplification kit. 50,000 sequencing reads per cell for gene expression and 5,000 sequencing reads per cell for V(D)J were generated on the Illumina NovaSeq6000 instrument. The demultiplexing and barcode processing were performed using CellRanger (v7.1.0, 10x Genomics). Generation of a reference transcriptome containing ide-cel and cilta-cel sequences To detect CAR+ cells in single cell RNA-seq data, we generated a reference transcriptome containing sequences from ide-cel and cilta-cel constructs. The nucleotide sequence of the ide-cel vector was acquired from its respective patent [43]. Nucleotide sequences of the cilta-cel vector containing the CD8A signal peptide, VHH1 CAR BCMA binding domain, G4S linker, V1HI2 CAR BCMA binding domain, CD8A hinge domain, CD8A transmembrane domain, CD137 cytoplasmic domain, and CD3ζ cytoplasmic domain were acquired its respective patent [44] and joined together to serve as a cilta-cel reference sequence. The sequences for each CAR product were then added to the GRCh38 human transcriptome. Two single cell RNA-seq samples from ide-cel infusion product and cilta-cel PBMCs were then aligned to this reference. Correction for incorrect SNPs in the sequences used for each product were done using a pipeline similar to Haradhvala et al. [45]. The code for this pipeline and the sequences used for ide-cel and cilta-cel are available at https://github.com/jeraldnoble . Single cell RNA-seq preprocessing Raw single-cell RNA-seq reads were aligned to the GRCh38 human transcriptome containing ide-cel and cilta-cel CAR sequences using Cell Ranger (v7.1.0, 10x Genomics). Cells were deemed CAR positive they originated from an ide-cel sample and had at least 1 read aligning to the ide-cel construct, or if they originated from a cilta-cel sample had at least 1 read aligning to the cilta-cel construct, and were considered CAR negative otherwise. Genes detected in less than 10 cells were removed from analysis. Low quality cells were filtered from the data set via requiring: at least 1000 UMIs per cell, at least 500 genes detected per cell, log10 genes per UMI of at least 80%, and less than 15% of the mitochondrial genes per cell Doublets were predicted for each sample using Scrublet [46] and DoubletFinder [47] . Cells that were predicted as being doublets by either program were removed from the data set. Following these quality control measures, all subsequent operations were implemented using functions within Seurat V5 [48]. The top 3000 highly variable genes (HVGs) were identified using the “vst” method within the FindVariableFeatures function. T-cell receptor genes, immunoglobulin genes, and the ide-cel and cilta-cell constructs were removed from the HVGs to prevent clustering based on monoclonal T-cells and the expression of the CAR constructs. Cells were scored for S and G2/M cell cycle phases using the CellCycleScoring function and a list of cell cycle genes (Supplementary File 1). Next, we regressed out the effects of the percentage of mitochondrial genes, UMIs per cell, S cell cycle phase scores, and G2/M cell cycle phase scores using the ScaleData function. To remove batch effects and inter-sample heterogeneity, integration was performed using Harmony integration within the IntegrateLayers function using the top 50 principal components. Following integration, a shared nearest neighbor graph was constructed based on the integrated data using 50 principal components. Then, clusters were generated using the Louvian algorithm within the FindClusters function with a resolution of 0.75. Identification of CD4 and CD8 cell populations To classify CD4 and CD8 cells, we leveraged a K-nearest neighbors graph with K=100 using the top 20 batch effect corrected principal components as in Li et al [13] to generate the smoothed expression of CD3, CD4, and CD8. Genes used to calculate the smoothed expression for CD3 were CD3D, CD3E, CD3G, and CD247, for CD8 we used CD8A and CD8B, and used CD4 for CD4. We then generated kernel density estimate plots of the smoothed expression values to define thresholds to classify CD4, CD8, double negative (CD4-CD8-), and double positive (CD4+CD8+) cells. Because populations of CD4 and CD8 cells did not cluster separately (Supplementary Figure S2A), we isolated CD4 and CD8 cells into separate data sets and repeated the above integration and clustering procedures. Cell type annotation Following integration and clustering of defined CD4 and CD8 populations, signatures for T-cell types were calculated using the AddModuleScore function with a marker gene list for various T-cell subtypes and states and additional marker gene list from Chu et al. (2023) [49] and Anderson et al. (2023) [50] (Supplementary File 2). Additionally, marker genes for each cluster were identified using the FindAllMarkers function requiring a log fold change threshold of 0.25 and only reporting genes with a positive fold change in each cluster. Using the marker gene lists and the marker genes generated by FindAllMarkers ,11 cell subtypes were identified for CD4 cells and 12 cell subtypes were identified CD8 cells (Supplementary Figures 4-5). Differential gene expression analysis All single-cell differential gene expression analyses were conducted in CAR+ cells using the FindMarkers function. The log fold change parameter was set to 0 when calling this function to enable downstream gene set enrichment analysis and genes were expressed in at least 10% of either of the populations being compared. Genes were considered differentially expressed if they exhibited a log2 fold change > 0.5 and had a Bonferroni adjusted p-value < 0.01. Gene set enrichment analysis Log2 fold change calculations for each gene were generated by FindMarkers as stated above and used for gene set enrichment analysis (GSEA). As inputs we utilized the Hallmark gene sets and KEGG reference gene sets available via msigdbr (igordot.github.io/msigdbr/). T-cell phenotype gene sets from Chu et al. (2023) [49], the FOXO1 regulon from Doan el al. (2024) [51], the TCF7 regulon from Chen et al. (2021) [52], the tonic signaling signature from Boroughs et al. (2020) [21], and GOBP canonical and non-canonical NFKB signaling gene sets from gsea-msigdb.org were additionally used for enrichment analysis (Supplementary File 3). Single sample gene set enrichment analysis Read counts for each sample, and each cell subtype within each sample, were aggregated using a pseudobulk approach. Read counts were then normalized using variance stabilizing transformation within DESeq2 [53]. Gene signatures were then calculated with GSVA [54] using the genes sets in Supplementary File 3. Single cell V(D)J analyses Single cell VDJ sequencing libraries from each sample were processed using CellRanger (v7.1.0, 10x Genomics) with the GRCh38 vdj reference (https://cf.10xgenomics.com/supp/cell-vdj/refdata-cellranger-vdj-GRCh38-alts-ensembl-7.1.0.tar.gz). Filtered contig files were combined and processed using scRepertoire version 1.11.0 [55] . Samples were combined via the combineTCR function with parameters “removeNA = TRUE” and “filterMulti = TRUE.” Declarations Acknowledgments This work was supported in part by the American Cancer Society, internal funding, the National Cancer Institute (P30CA076292, PI Cleveland), and generous donations from the Hyer family and the Thiel family. F.L.L. is supported in part by the Leukemia and Lymphoma Society as a Clinical Scholar. The authors thank the Flow Cytometry Core, Molecular Genomics Core and Tissue Core at Moffitt Cancer Center. Author Contributions JN: data analysis, curation writing the first and final manuscript drafts, XS: biostatistical input and review, MM: immunophenotyping and data generation, CLF: experimental design, oversight, data analysis, curation and writing of all manuscript drafts. All authors reviewed data and contributed to the manuscript review and editing and approved the submitted version. Declaration of interests JN: none XS: none MM: none JAM: none SC: none GdA: none OACP: reports Honoraria/consulting Legend Biotech USA Inc, BMS HL: reports a role with the speakers’ bureau for Sanofi MA: reports consulting- Janssen, BMS, Sanofi and research support - BMS TN: reports clinical trial support (to the institution) by Novartis, clinical trial support (drug only supply to the institution) by Karyopharm, Consultancy from ImmunoGen, advisory board: Medexus KHS: reports consultancy, advisor, and/or speaker roles with Adaptive Biotech, Janssen, BMS, Takeda, Sanofi, and Glaxo Smith Kline; research funding with Karyopharm and Abbvie, and funds from BMS, Amgen, and Janssen-funded clinical trials RB: reports Honoraria/consulting BMS Janssen Pfizer and research funding: BMS, Janssen, Abbvie, Regeneron, Karyopharm BB: speakers’ bureau for Sanofi pharmaceuticals AGC: serves on the advisory boards for Janssen and Sanofi; and reports a role with the speakers’ bureau for Sanofi JK: none RA: none DKH: reports research funding from Bristol-Myers Squibb, Karyopharm, and Adaptive Biotech; Consulting or advisory role for Bristol-Myers Squibb, Janssen, Pfizer, and Karyopharm. D.K.H is also supported by the Pentecost Family Myeloma Research Center. YB: none FLL: reports Scientific Advisory Role/Consulting Fees: A2, Allogene, Amgen, Bluebird Bio, BMS, Calibr, Caribou, Cowen, EcoR1, Gerson Lehrman Group (GLG), Iovance, Kite Pharma, Janssen, Legend Biotech, Novartis, Sana, Umoja, Pfizer. Data Safety Monitoring Board: Data and Safety Monitoring Board for the NCI Safety Oversight CAR T-cell Therapies Committee. Research Contracts/Grants: Kite Pharma (Institutional), Allogene (Institutional), CERo Therapeutics (Institutional), Novartis (Institutional), BlueBird Bio (Institutional), 2SeventyBio (Institutional), BMS (Institutional), National Cancer Institute (R01CA244328 MPI: Locke; P30CA076292 PI: Cleveland), Leukemia and Lymphoma Society Scholar in Clinical Research (PI: Locke) Patents, Royalties, Other Intellectual Property: Several patents held by the institution in my name (unlicensed) in the field of cellular immunotherapy. Education or Editorial Activity: Aptitude Health, ASH, BioPharma Communications CARE Education, Clinical Care Options Oncology, Imedex, Society for Immunotherapy of Cancer CLF: reports Honoraria/consulting BMS/Celgene, ONK therapeutics & Janssen; research funding from BMS and Janssen. References Braunlin, M., et al., Trends in the multiple myeloma treatment landscape and survival: a U.S. analysis using 2011–2019 oncology clinic electronic health record data . Leukemia & Lymphoma, 2020: p. 1–10. Nidhi Mankan, A.K.N., Improvements in myeloma specific survival over the last two decades. , in ASCO . 2022, JCO: Chicago. Maples, K.T., S.A. Scott, and S. Lonial, Novel approaches to the treatment of multiple myeloma . Clin Adv Hematol Oncol, 2023. 21(5): p. 247–256. Munshi, N.C., et al., Idecabtagene Vicleucel in Relapsed and Refractory Multiple Myeloma . New England Journal of Medicine, 2021. 384(8): p. 705–716. Berdeja, J.G., et al., Ciltacabtagene autoleucel, a B-cell maturation antigen-directed chimeric antigen receptor T-cell therapy in patients with relapsed or refractory multiple myeloma (CARTITUDE-1): a phase 1b/2 open-label study . The Lancet, 2021. 398(10297): p. 314–324. Hansen, D.K., et al., Idecabtagene Vicleucel for Relapsed/Refractory Multiple Myeloma: Real-World Experience From the Myeloma CAR T Consortium . J Clin Oncol, 2023. 41(11): p. 2087–2097. San-Miguel, J., et al., Cilta-cel or Standard Care in Lenalidomide-Refractory Multiple Myeloma . New England Journal of Medicine, 2023. 389(4): p. 335–347. Rodriguez-Otero, P., et al., Ide-cel or Standard Regimens in Relapsed and Refractory Multiple Myeloma . New England Journal of Medicine, 2023. Costa, L.J., et al., How I treat triple-class refractory multiple myeloma . Br J Haematol, 2022. 198(2): p. 244–256. Martin, T., et al., Matching-adjusted indirect comparison of efficacy outcomes for ciltacabtagene autoleucel in CARTITUDE-1 versus idecabtagene vicleucel in KarMMa for the treatment of patients with relapsed or refractory multiple myeloma . Current Medical Research and Opinion, 2021. 37(10): p. 1779–1788. Bruno, B., et al., European Myeloma Network perspective on CAR T-Cell therapies for multiple myeloma . Haematologica, 2021. 106(8): p. 2054–2065. Sidana, S., et al., Real World Outcomes with Idecabtagene Vicleucel (Ide-Cel) CAR-T Cell Therapy for Relapsed/Refractory Multiple Myeloma . Blood, 2023. 142: p. 1027. Li, X., et al., A single-cell atlas of CD19 chimeric antigen receptor T cells . Cancer Cell, 2023. 41(11): p. 1835–1837. Chen, G.M., et al., Integrative Bulk and Single-Cell Profiling of Premanufacture T-cell Populations Reveals Factors Mediating Long-Term Persistence of CAR T-cell Therapy . Cancer Discovery, 2021. 11(9): p. 2186–2199. Antón, O., et al., An essential role for the MAL protein in targeting Lck to the plasma membrane of human T lymphocytes . Journal of Experimental Medicine, 2008. 205(13): p. 3201–3213. Rossi, J., et al., Preinfusion polyfunctional anti-CD19 chimeric antigen receptor T cells are associated with clinical outcomes in NHL . Blood, 2018. 132(8): p. 804–814. Shui, X., et al., Knockdown of lncRNA NEAT1 inhibits Th17/CD4(+) T cell differentiation through reducing the STAT3 protein level . J Cell Physiol, 2019. 234(12): p. 22477–22484. Kouo, T., et al., Galectin-3 Shapes Antitumor Immune Responses by Suppressing CD8 + T Cells via LAG-3 and Inhibiting Expansion of Plasmacytoid Dendritic Cells . Cancer Immunol Res, 2015. 3(4): p. 412–23. Park, I., et al., The Transmembrane Adaptor Protein LIME Is Essential for Chemokine-Mediated Migration of Effector T Cells to Inflammatiory Sites . Mol Cells, 2020. 43(11): p. 921–934. Philipson, B.I., et al., 4-1BB costimulation promotes CAR T cell survival through noncanonical NF-κB signaling. Sci Signal, 2020. 13(625). Boroughs, A.C., et al., A Distinct Transcriptional Program in Human CAR T Cells Bearing the 4-1BB Signaling Domain Revealed by scRNA-Seq. Molecular Therapy, 2020. 28(12): p. 2577–2592. Dobrzanski, M.J., et al., Tc1 and Tc2 effector cell therapy elicit long-term tumor immunity by contrasting mechanisms that result in complementary endogenous type 1 antitumor responses . J Immunol, 2004. 172(3): p. 1380–90. Mittrücker, H.-W., A. Visekruna, and M. Huber, Heterogeneity in the Differentiation and Function of CD8 + T Cells . Archivum Immunologiae et Therapiae Experimentalis, 2014. 62(6): p. 449–458. Verzella, D., et al., Life, death, and autophagy in cancer: NF-kappaB turns up everywhere . Cell Death Dis, 2020. 11(3): p. 210. Long, A.H., et al., 4-1BB costimulation ameliorates T cell exhaustion induced by tonic signaling of chimeric antigen receptors . Nat Med, 2015. 21(6): p. 581–90. Rodriguez-Marquez, P., et al., CAR density influences antitumoral efficacy of BCMA CAR T cells and correlates with clinical outcome . Sci Adv, 2022. 8(39): p. eabo0514. Hege, K., Context matters in CAR T cell tonic signaling . Nature Medicine, 2021. 27(5): p. 763–764. Singh, N., et al., Antigen-independent activation enhances the efficacy of 4-1BB-costimulated CD22 CAR T cells . Nature Medicine, 2021. 27(5): p. 842–850. Frigault, M.J., et al., Identification of chimeric antigen receptors that mediate constitutive or inducible proliferation of T cells . Cancer Immunol Res, 2015. 3(4): p. 356–67. Wolf, T., et al., Dynamics in protein translation sustaining T cell preparedness . Nat Immunol, 2020. 21(8): p. 927–937. Araki, K., et al., Translation is actively regulated during the differentiation of CD8(+) effector T cells . Nat Immunol, 2017. 18(9): p. 1046–1057. Claiborne, M.D., et al., Persistent CAD activity in memory CD8(+) T cells supports rRNA synthesis and ribosomal biogenesis required at rechallenge . Sci Immunol, 2022. 7(71): p. eabh4271. Zhou, L., et al., T cell proliferation requires ribosomal maturation in nucleolar condensates dependent on DCAF13 . Journal of Cell Biology, 2023. 222(10). Hukelmann, J.L., et al., The cytotoxic T cell proteome and its shaping by the kinase mTOR . Nat Immunol, 2016. 17(1): p. 104–12. Tan, T.C.J., et al., Suboptimal T-cell receptor signaling compromises protein translation, ribosome biogenesis, and proliferation of mouse CD8 T cells. Proceedings of the National Academy of Sciences, 2017. 114(30): p. E6117-E6126. Hendriks, J., Y. Xiao, and J. Borst, CD27 promotes survival of activated T cells and complements CD28 in generation and establishment of the effector T cell pool . J Exp Med, 2003. 198(9): p. 1369–80. Xu, Y., et al., CD27-Armored BCMA-CAR T Cell (CBG-002) Therapy for Relapsed and Refractory Multiple Myeloma: A Phase I Clinical Trial . Blood, 2023. 142(Supplement 1): p. 3468–3468. Mitra, A., et al., From bench to bedside: the history and progress of CAR T cell therapy . Frontiers in Immunology, 2023. 14. Ciltacabtagene autoleucel: regenerative medical product submitted for marketing approval conducted by the Pharmaceuticals and Medical Devices Agency (PMDA). . Committee on Regenerative Medicine Products and Biotechnology 7/6/2024]; Available from: extension:// efaidnbmnnnibpcajpcglclefindmkaj/https://www.pmda.go.jp/files/000252610.pdf . DELL'ARINGA, T.C.R.H.E.H.Y.J.S.K.E.T.J.F.N.M.R.L.J., Uses of anti-bcma chimeric antigen receptors , in https://patents.google.com/patent/WO2021091978A1/en?oq=WO+2021%2f091978+A1 . 2020. Ho, J.Y., et al., Promoter usage regulating the surface density of CAR molecules may modulate the kinetics of CAR-T cells in vivo . Mol Ther Methods Clin Dev, 2021. 21: p. 237–246. Biasco, L., et al., Clonal expansion of T memory stem cells determines early anti-leukemic responses and long-term CAR T cell persistence in patients . Nat Cancer, 2021. 2(6): p. 629–642. Thompson, E.G. and N.T. Martin, Uses of chimeric antigen receptor (car) t-cell therapies in combination with inhibitors of inflammation-related soluble factors . 2024, Google Patents. Schecter, J.M. and X. Fan, BCMA-targeted CAR-T cell therapy for multiple myeloma . 2022, Google Patents. Haradhvala, N.J., et al., Distinct cellular dynamics associated with response to CAR-T therapy for refractory B cell lymphoma . Nat Med, 2022. 28(9): p. 1848–1859. Wolock, S.L., R. Lopez, and A.M. Klein, Scrublet: Computational Identification of Cell Doublets in Single-Cell Transcriptomic Data . Cell Syst, 2019. 8(4): p. 281–291 e9. McGinnis, C.S., L.M. Murrow, and Z.J. Gartner, DoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors . Cell Syst, 2019. 8(4): p. 329–337 e4. Hao, Y., et al., Dictionary learning for integrative, multimodal and scalable single-cell analysis . Nat Biotechnol, 2023. Chu, Y., et al., Pan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance . Nat Med, 2023. 29(6): p. 1550–1562. Anderson, N.D., et al., Transcriptional signatures associated with persisting CD19 CAR-T cells in children with leukemia . Nat Med, 2023. 29(7): p. 1700–1709. Doan, A.E., et al., FOXO1 is a master regulator of memory programming in CAR T cells . Nature, 2024. Chen, G.M., et al., Integrative Bulk and Single-Cell Profiling of Premanufacture T-cell Populations Reveals Factors Mediating Long-Term Persistence of CAR T-cell Therapy . Cancer Discov, 2021. 11(9): p. 2186–2199. Love, M.I., W. Huber, and S. Anders, Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 . Genome Biol, 2014. 15(12): p. 550. Hanzelmann, S., R. Castelo, and J. Guinney, GSVA: gene set variation analysis for microarray and RNA-seq data . BMC Bioinformatics, 2013. 14: p. 7. Borcherding, N., N.L. Bormann, and G. Kraus, scRepertoire: An R-based toolkit for single-cell immune receptor analysis . F1000Res, 2020. 9: p. 47. Additional Declarations Yes there is potential Competing Interest. OACP: reports Honoraria/consulting Legend Biotech USA Inc, BMS HL: reports a role with the speakers’ bureau for Sanofi MA: reports consulting- Janssen, BMS, Sanofi and research support - BMS TN: reports clinical trial support (to the institution) by Novartis, clinical trial support (drug only supply to the institution) by Karyopharm, Consultancy from ImmunoGen, advisory board: Medexus KHS: reports consultancy, advisor, and/or speaker roles with Adaptive Biotech, Janssen, BMS, Takeda, Sanofi, and Glaxo Smith Kline; research funding with Karyopharm and Abbvie, and funds from BMS, Amgen, and Janssen-funded clinical trials RB: reports Honoraria/consulting BMS Janssen Pfizer and research funding: BMS, Janssen, Abbvie, Regeneron, Karyopharm BB: speakers’ bureau for Sanofi pharmaceuticals AGC: serves on the advisory boards for Janssen and Sanofi; and reports a role with the speakers’ bureau for Sanofi DKH: reports research funding from Bristol-Myers Squibb, Karyopharm, and Adaptive Biotech; Consulting or advisory role for Bristol-Myers Squibb, Janssen, Pfizer, and Karyopharm. D.K.H is also supported by the Pentecost Family Myeloma Research Center. FLL: reports Scientific Advisory Role/Consulting Fees: A2, Allogene, Amgen, Bluebird Bio, BMS, Calibr, Caribou, Cowen, EcoR1, Gerson Lehrman Group (GLG), Iovance, Kite Pharma, Janssen, Legend Biotech, Novartis, Sana, Umoja, Pfizer. Data Safety Monitoring Board: Data and Safety Monitoring Board for the NCI Safety Oversight CAR T-cell Therapies Committee. Research Contracts/Grants: Kite Pharma (Institutional), Allogene (Institutional), CERo Therapeutics (Institutional), Novartis (Institutional), BlueBird Bio (Institutional), 2SeventyBio (Institutional), BMS (Institutional), National Cancer Institute (R01CA244328 MPI: Locke; P30CA076292 PI: Cleveland), Leukemia and Lymphoma Society Scholar in Clinical Research (PI: Locke) Patents, Royalties, Other Intellectual Property: Several patents held by the institution in my name (unlicensed) in the field of cellular immunotherapy. Education or Editorial Activity: Aptitude Health, ASH, BioPharma Communications CARE Education, Clinical Care Options Oncology, Imedex, Society for Immunotherapy of Cancer CLF: reports Honoraria/consulting BMS/Celgene, ONK therapeutics & Janssen; research funding from BMS and Janssen. Supplementary Files SupplementaryFigures.pdf Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-4994668","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Article","associatedPublications":[],"authors":[{"id":351570733,"identity":"86212373-29bb-45a6-828a-a4e4fc8b85e5","order_by":0,"name":"Ciara Freeman","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAABEUlEQVRIiWNgGAWjYBAC/mYw9R/KrWBgYG8+2IBXi8RhMHUAyj3DwMBzLLGx4QAu9UBg4ICshbENpCWBEb8WduZnH34wHMjnZ+9O/Fw4b5scDxtz++MPDDb58g44tDCzGc/sYfhjObPn7GbpmdtuG/OwMYIclma5EYdVBswMxgw8DAcMDG7kbpDm3XY7cb98I0jLYQPDBlxa2D8z/gFqsb//dvNv3jm3E3sgtuDR4sxjzAy2RYJ3mzRvA5IWeRzelzjMU8wsY3DAQOJM7jZrnmMQv8w4Y5BmYIBDC3//8c2MbyoOGPC3n918m6fmNjDE2B98qKiwMZDH4TCo87CJGBzApwUrwG/LKBgFo2AUjCAAADiiXB8mWEGYAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0002-0662-6480","institution":"H. Lee Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":true,"prefix":"","firstName":"Ciara","middleName":"","lastName":"Freeman","suffix":""},{"id":351570734,"identity":"cace3d02-3b5c-4d37-918c-5a1e96c8d3d7","order_by":1,"name":"Jerald Noble","email":"","orcid":"","institution":"H. Lee Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Jerald","middleName":"","lastName":"Noble","suffix":""},{"id":351570735,"identity":"ba8287a8-1cb3-427b-92cd-f62b3f8c26b2","order_by":2,"name":"Xiaofei Song","email":"","orcid":"","institution":"H. Lee Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Xiaofei","middleName":"","lastName":"Song","suffix":""},{"id":351570736,"identity":"6d555eb6-72e1-4e24-950e-e9cab187bd5b","order_by":3,"name":"Meghan Menges","email":"","orcid":"https://orcid.org/0009-0007-7446-9492","institution":"MOFFITT","correspondingAuthor":false,"prefix":"","firstName":"Meghan","middleName":"","lastName":"Menges","suffix":""},{"id":351570737,"identity":"d507df72-6cdb-43aa-8fc2-51ec2eee40a0","order_by":4,"name":"Julieta Abraham-Miranda","email":"","orcid":"","institution":"H. Lee Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Julieta","middleName":"","lastName":"Abraham-Miranda","suffix":""},{"id":351570738,"identity":"b61d112e-7b1d-4356-90ea-ec03b91d773a","order_by":5,"name":"Salvatore Corallo","email":"","orcid":"https://orcid.org/0009-0001-2450-4022","institution":"H. Lee Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Salvatore","middleName":"","lastName":"Corallo","suffix":""},{"id":351570739,"identity":"bfd2e09b-666d-49e1-964e-d7409fd24e09","order_by":6,"name":"Gabriel De Avila","email":"","orcid":"","institution":"H Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Gabriel","middleName":"","lastName":"De Avila","suffix":""},{"id":351570740,"identity":"367b742b-2cc0-4274-8993-f9bb7a5ea126","order_by":7,"name":"Omar Castaneda","email":"","orcid":"https://orcid.org/0000-0003-0423-5870","institution":"H. Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Omar","middleName":"","lastName":"Castaneda","suffix":""},{"id":351570741,"identity":"1c7ff0f0-36dd-4ad4-8122-21f0ef7da100","order_by":8,"name":"Hien Liu","email":"","orcid":"","institution":"H. Lee Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Hien","middleName":"","lastName":"Liu","suffix":""},{"id":351570742,"identity":"bfe14c10-6bd8-4446-964f-a3b0af900e10","order_by":9,"name":"Melissa Alsina","email":"","orcid":"","institution":"H. Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Melissa","middleName":"","lastName":"Alsina","suffix":""},{"id":351570743,"identity":"7a04f975-2a62-456b-a05b-b185db22fa99","order_by":10,"name":"Taiga Nishihori","email":"","orcid":"","institution":"H Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Taiga","middleName":"","lastName":"Nishihori","suffix":""},{"id":351570744,"identity":"3c0bf701-70a8-4ff5-b9d4-20bd7ebc62fb","order_by":11,"name":"Kenneth Shain","email":"","orcid":"","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Kenneth","middleName":"","lastName":"Shain","suffix":""},{"id":351570745,"identity":"ec892a4e-6bb6-4ffa-bbe2-d96f4470c10c","order_by":12,"name":"Rachid Baz","email":"","orcid":"","institution":"H Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Rachid","middleName":"","lastName":"Baz","suffix":""},{"id":351570746,"identity":"3daa562f-0fa8-46e6-83bf-3f638aa75b49","order_by":13,"name":"Brandon Blue","email":"","orcid":"https://orcid.org/0000-0003-0639-3768","institution":"H Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Brandon","middleName":"","lastName":"Blue","suffix":""},{"id":351570747,"identity":"1bd54723-7877-464a-bbb5-0bfabed7ca93","order_by":14,"name":"Ariel Grajales Cruz","email":"","orcid":"","institution":"H Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Ariel","middleName":"Grajales","lastName":"Cruz","suffix":""},{"id":351570748,"identity":"74a5eddd-3af6-4b80-a354-f57809ee3834","order_by":15,"name":"Doris Hansen","email":"","orcid":"https://orcid.org/0000-0002-9813-3578","institution":"H. Lee Moffitt Cancer Center and Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Doris","middleName":"","lastName":"Hansen","suffix":""},{"id":351570749,"identity":"a8114e57-0db7-4451-9d64-03700bce3e9e","order_by":16,"name":"Reginald Atkins","email":"","orcid":"","institution":"Moffitt Cancer Center \u0026 Research Institute","correspondingAuthor":false,"prefix":"","firstName":"Reginald","middleName":"","lastName":"Atkins","suffix":""},{"id":351570750,"identity":"8cede067-9851-4688-9956-c454b45d0fbc","order_by":17,"name":"Frederick Locke","email":"","orcid":"https://orcid.org/0000-0001-9063-6691","institution":"Moffitt Cancer Center","correspondingAuthor":false,"prefix":"","firstName":"Frederick","middleName":"","lastName":"Locke","suffix":""}],"badges":[],"createdAt":"2024-08-29 04:55:17","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-4994668/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-4994668/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":66392536,"identity":"c303afe6-9c5a-4663-9557-d93a67147c1b","added_by":"auto","created_at":"2024-10-11 09:13:00","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":1715302,"visible":true,"origin":"","legend":"\u003cp\u003eStudy design, cohort summary, and cohort cell subtype summary.\u003c/p\u003e\n\u003cp\u003eA Experimental design summary. B Swimmer plot depicting progression free survival (PFS) and post-relapse survival of the 46 patients in the study. C-D CD4 and CD8 cell subtypes identified via scRNAseq, the number of total cells of each subtype, and the number of CAR+ cells of each subtype. E-F CAR+ CD4 and CD8 cell subtype compositions of each sample in the study. Stacked bars depict fractions of each cell subtype and tiles depict additional information for each sample.\u003c/p\u003e","description":"","filename":"fig1.png","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/e1d85db4a2f6dd64c3ce71ee.png"},{"id":66392724,"identity":"b3d78588-5e7e-4c5f-be5d-831151e09b37","added_by":"auto","created_at":"2024-10-11 09:21:00","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":1484913,"visible":true,"origin":"","legend":"\u003cp\u003eCD4:CD8 ratios of samples in the study, IP CAR+ cells, IP ide-cel expression, and differential gene expression results between responders in CAR+ ide-cel IP.\u003c/p\u003e\n\u003cp\u003eA-B CD4:CD8 ratios in ide-cel IP and PBMCs and cilta-cel PBMCs via scRNAseq and flow cytometry, respectively. C Spearman correlation between scRNAseq and flow cytometry of CD4:CD8 ratios in samples from A \u0026amp; B. D Percentage of CAR+ cells in ide-cel IP between responders in CD4 and CD8 cells. E Expression of the ide-cel construct in CAR+ ide-cel IP cells between responders in CD4 and CD8 cells. F GSEA results based on fold change differences of genes between responders in CAR+ CD4 and CD8 IP cells and their respective cell subtypes. G-H Volcano plots depicting results of differential gene expression analysis between responders in CAR+ CD4 and CD8 IP cells, respectively. Genes with log2 fold change \u0026gt; (+/-)0.5 and Bonferroni-corrected p-value \u0026lt; 0.01 are colored based on the direction they are upregulated. Colored genes are labeled where space permits.\u003c/p\u003e","description":"","filename":"fig2.png","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/1f265a92a80196c846a99d4a.png"},{"id":66392537,"identity":"f49920df-dc6f-49a5-b284-fedbb99c4686","added_by":"auto","created_at":"2024-10-11 09:13:00","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":889115,"visible":true,"origin":"","legend":"\u003cp\u003ePathway expression differences between responders in CAR+ CD4 and CD8 IP cells and PFS between patients with high and low NFKB signaling and anti-apoptosis signatures in CAR+ CD4 IP cells.\u003c/p\u003e\n\u003cp\u003eA Expression of NFKB signaling, anti-apoptosis, and tonic signaling signatures via pseudobulk aggregation of gene expression followed by ssgsea in CAR+ CD4 and CD8 IP cells between responders. Wilcoxon rank sum test p-values reported B-D Spearman correlation between NFKB signaling (B), anti-apoptosis (C), and tonic signaling signatures (D) and ide-cel CAR construct expression in all CAR+ CD4 cells with points labeled by 9 month response. E Heatmap depicting the expression of NFKB signaling, anti-apoptosis, tonic signaling, and ide-cel CAR construct expression in all CAR+ CD4 cells and their association with 9 month response. F-I PFS differences between patients with high and low NFKB signaling (F-G) and anti-apoptosis (H-I) signatures in CAR+ CD4 IP cells.\u003c/p\u003e","description":"","filename":"fig3.png","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/cccfb2bcfe5899bc902f68c2.png"},{"id":66392539,"identity":"c7878ef7-0ee4-4248-abda-cefec14c64e4","added_by":"auto","created_at":"2024-10-11 09:13:00","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":1257711,"visible":true,"origin":"","legend":"\u003cp\u003eDifferential gene expression results between responders in CAR+ CD4 and CD8 ide-cel PBMC cells.\u003c/p\u003e\n\u003cp\u003eA GSEA results based on fold change differences of genes between responders in CAR+ CD4 and CD8 PBMC cells and their respective cell subtypes. B-C Volcano plots depicting results of differential gene expression analysis between responders in CAR+ CD4 and CD8 PBMC cells, respectively. Genes with log2 fold change \u0026gt; (+/-)0.5 and Bonferroni-corrected Wilcoxon p-value \u0026lt; 0.01 are colored based on the direction they are upregulated. Colored genes are labeled where space permits. D Violin plots depicting expression differences between responders in genes of interest between responders in CAR+ CD4 and CD8 PBMC cells. All Bonferroni-corrected Wilcoxon p-values are \u0026lt; 0.0001.\u003c/p\u003e","description":"","filename":"fig4.png","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/f52e865939edfd2fbdc5ca40.png"},{"id":66392542,"identity":"37b2b2ab-2f74-4284-a291-ed3c63f4cd11","added_by":"auto","created_at":"2024-10-11 09:13:00","extension":"png","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":774419,"visible":true,"origin":"","legend":"\u003cp\u003eTCR clonotype dynamics in CAR+ CD4 and CD8 IP and PBMC cells.\u003c/p\u003e\n\u003cp\u003eA TCR clonal population summary of all samples in the study. B Fractions of expanded and non-expanded clones in CAR+ CD4 and CD8 IP and PBMC cells of ide-cel. C GSEA results based on fold change differences of genes between expanded and non-expanded in CAR+ CD4 and CD8 PBMC cells in ide-cel. D Percentage of CAR+ clones that were detected in both IP and PBMC samples in 6 ide-cel patients with matched samples. E GSEA results based on fold change differences of genes between CAR+ CD4 and CD8 clones that were (Paired), or were not (Unpaired), detected in both IP and PBMC samples in 6 ide-cel patients with matched samples. Results are reported separately for CAR+ CD4 and CD8 IP clones that were detected in PBMC and CAR+ CD4 and CD8 PBMC clones were detected in IP.\u003c/p\u003e","description":"","filename":"fig5.png","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/7328a2287f73b3e3841f62fd.png"},{"id":66392723,"identity":"0f9af704-4a3f-4f05-8356-f130b1cfbc84","added_by":"auto","created_at":"2024-10-11 09:21:00","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":620309,"visible":true,"origin":"","legend":"\u003cp\u003eProduct comparison between CAR+ ide-cel and cilta-cel PBMCs.\u003c/p\u003e\n\u003cp\u003eA-B Cell subtype composition differences between ide-cel and cilta-cel CAR+ CD4 (A) and CD8 (B) PBMCs. C Expression differences in key genes between ide-cel and cilta-cel in CAR+ CD4 and CD8 PBMCs. All Bonferroni-corrected Wilcoxon p-values \u0026lt; 0.0001. D CAR construct expression differences between ide-cel and cilta-cel via pseudobulk aggregation of gene expression in CAR+ CD4 and CD8 PBMCs. E CAR construct expression differences between ide-cel and cilta-cel via flow cytometry. F Ssgsea pathway expression differences in NKFB signaling and ribosomal gene expression between ide-cel and cilta-cel in CAR+ CD4 and CD8 PBMCs. G TCR clonal population differences between ide-cel and cilta-cel in CAR+ PBMCs. H TCR diversity measured via inverse Simpson’s index between ide-cel and cilta-cel in CAR+ CD4 and CD8 PBMCs. All p-values reported are via Wilcoxon rank sum test unless stated otherwise.\u003c/p\u003e","description":"","filename":"fig6.png","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/334b07f1265c7d57421280f7.png"},{"id":68215144,"identity":"108e135c-b6e7-4ac8-a896-334d1294296f","added_by":"auto","created_at":"2024-11-04 19:23:44","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":7296013,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/74b21634-dd6d-4936-98f8-bb51743c5b46.pdf"},{"id":66392543,"identity":"e011bcf8-e163-4031-9f71-54df1bce2ddd","added_by":"auto","created_at":"2024-10-11 09:13:00","extension":"pdf","order_by":8,"title":"","display":"","copyAsset":false,"role":"supplement","size":9141105,"visible":true,"origin":"","legend":"","description":"","filename":"SupplementaryFigures.pdf","url":"https://assets-eu.researchsquare.com/files/rs-4994668/v1/fe49ea430224af5ba9db0c65.pdf"}],"financialInterests":"\u003cb\u003eYes\u003c/b\u003e there is potential Competing Interest.\nOACP: reports Honoraria/consulting Legend Biotech USA Inc, BMS \r\nHL: reports a role with the speakers’ bureau for Sanofi\r\nMA: reports consulting- Janssen, BMS, Sanofi and research support - BMS\r\nTN: reports clinical trial support (to the institution) by Novartis, clinical trial support (drug only supply to the institution) by Karyopharm, Consultancy from ImmunoGen, advisory board: Medexus\r\nKHS: reports consultancy, advisor, and/or speaker roles with Adaptive Biotech, Janssen, BMS, Takeda, Sanofi, and Glaxo Smith Kline; research funding with Karyopharm and Abbvie, and funds from BMS, Amgen, and Janssen-funded clinical trials\r\nRB: reports Honoraria/consulting BMS Janssen Pfizer and research funding: BMS, Janssen, Abbvie, Regeneron, Karyopharm\r\nBB: speakers’ bureau for Sanofi pharmaceuticals\r\nAGC: serves on the advisory boards for Janssen and Sanofi; and reports a role with the speakers’ bureau for Sanofi\r\nDKH: reports research funding from Bristol-Myers Squibb, Karyopharm, and Adaptive Biotech; Consulting or advisory role for Bristol-Myers Squibb, Janssen, Pfizer, and Karyopharm. D.K.H is also supported by the Pentecost Family Myeloma Research Center.\r\nFLL: reports Scientific Advisory Role/Consulting Fees: A2, Allogene, Amgen, Bluebird Bio, BMS, Calibr, Caribou, Cowen, EcoR1, Gerson Lehrman Group (GLG), Iovance, Kite Pharma, Janssen, Legend Biotech, Novartis, Sana, Umoja, Pfizer. Data Safety Monitoring Board: Data and Safety Monitoring Board for the NCI Safety Oversight CAR T-cell Therapies Committee. Research Contracts/Grants: Kite Pharma (Institutional), Allogene (Institutional), CERo Therapeutics (Institutional), Novartis (Institutional), BlueBird Bio (Institutional), 2SeventyBio (Institutional), BMS (Institutional), National Cancer Institute (R01CA244328 MPI: Locke; P30CA076292 PI: Cleveland), Leukemia and Lymphoma Society Scholar in Clinical Research (PI: Locke) Patents, Royalties, Other Intellectual Property: Several patents held by the institution in my name (unlicensed) in the field of cellular immunotherapy. Education or Editorial Activity: Aptitude Health, ASH, BioPharma Communications CARE Education, Clinical Care Options Oncology, Imedex, Society for Immunotherapy of Cancer\r\nCLF: reports Honoraria/consulting BMS/Celgene, ONK therapeutics \u0026 Janssen; research funding from BMS and Janssen.","formattedTitle":"Single-Cell Analysis Reveals Ide-cel and Cilta-cel Characteristics That Influence Efficacy","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMultiple myeloma remains incurable, however, significant advances over the past two decades have led to major improvements in the depth and durability of responses achieved with modern treatment combinations[\u003cspan additionalcitationids=\"CR2\" citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e]. For patients who have progressed after exposure to the major classes of agents, the introduction of two chimeric antigen receptor T-cell (CAR-T) products which target B-cell maturation antigen (BCMA) has radically improved outcomes and led to approvals for those with relapsed and refractory multiple myeloma (RRMM)[\u003cspan additionalcitationids=\"CR5\" citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e]. More recently, both products demonstrated a significant benefit in randomized trials, outperforming standard-of-care regimens, leading to expanded indications and altering the therapeutic paradigm for RRMM [\u003cspan additionalcitationids=\"CR8\" citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e9\u003c/span\u003e]. While these products have not been evaluated head-to-head, indirect comparisons have been performed which suggested there may be improved depth and durability of responses with ciltacabtagene autoleucel (cilta-cel) compared to idecabtagene vicleucel (ide-cel)[\u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e]. Clinical trial outcomes also suggest differential efficacy, although this could be influenced by differences in the enrolled patient populations and trial design. These products employ antigen (BCMA) binding domains that differ in both modality and target binding valency, and in aspects of the manufacturing process. Understanding which attributes are beneficial for patients could potentiate enhanced treatment selection[\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e11\u003c/span\u003e]. In addition, neither agent has proven curative, thus understanding the mechanisms underpinning therapeutic success and failure are paramount. We sought to evaluate the characteristics of the products using single-cell RNA sequencing (scRNAseq) analysis in a large, real-world cohort of patients with RRMM.\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003ePatient and product characteristics\u003c/p\u003e \u003cp\u003eWe identified 46 RRMM patients treated per standard of care, with BCMA targeting CAR-T cell products (Extended Data Table\u0026nbsp;1). Only ide-cel treated patients, N\u0026thinsp;=\u0026thinsp;40, had pre-infusion products (IP) available for sequencing. We identified 6 matched peripheral blood post-infusion samples obtained during early expansion-phase (days 7\u0026ndash;14 post-infusion) from which CAR-positive cells were extracted and encapsulated (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA/Supplementary Fig.\u0026nbsp;1) for scRNAseq.\u0026nbsp;We identified 6 additional patients treated with cilta-cel with post-infusion samples during the same period, and similarly extracted CAR-positive cells. Median time of collection for ide-cel treated patients was 9 days (range 7\u0026ndash;14) and for cilta-cel treated patients 13 days (range 11\u0026ndash;14) aligned with rising lymphocyte count measurements in peripheral blood. All patients in the cohort were triple-class exposed, and had received a minimum of four prior lines of therapy (median 6, range 4\u0026ndash;13). In total, 28% had evidence of high tumor burden (\u0026ge;\u0026thinsp;50% involvement in restaging bone marrow biopsy prior to lymphodepletion [LD]) and 17% had evidence of extramedullary disease on pre-treatment imaging. The median duration of follow up for all living patients was 19.87 months (range 7.1-35.87) and median PFS for the whole cohort was reached at 15.27 months (95% CI: 8.83\u0026ndash;22.47). The median OS for this cohort was not reached, but 18-months estimated at 77% (95% CI 61\u0026ndash;87) (Supplementary Figs.\u0026nbsp;2A/B). Cytokine release syndrome (CRS) of any grade occurred in 38 (83%) of the cohort, all grade 1 (N\u0026thinsp;=\u0026thinsp;26) or grade 2 (N\u0026thinsp;=\u0026thinsp;12). Similarly, immune effector cell-associated neurotoxicity syndrome (ICANS) occurred in 6 patients (13%), grade 2 (N\u0026thinsp;=\u0026thinsp;3), grade 3 (N\u0026thinsp;=\u0026thinsp;2) and 1 grade 4 event (in a patient with pre-treatment evidence of CNS involvement). Of the patients treated with cilta-cel, one patient developed delayed neurotoxicity with evidence of movement and neurocognitive treatment-emergent adverse events that had not resolved at the time of death. At last assessment prior to death, this patient remained in complete response and had not yet reached the landmark timepoint.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eSingle cell RNAseq libraries were generated from 52 samples in total; composed of 40 unique patient ide-cel infusion products (IP), and CAR-enriched peripheral blood mononuclear cells (PBMCs) from 6 ide-cel and 6 cilta-cel treated patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Patients were stratified as having durable response (DR) if they were still alive, without receipt of additional myeloma therapy, and without evidence of progression at 9 months following treatment. This cutoff was selected on the basis of clinical trial and real-world data suggesting that the median progression free survival post ide-cel infusion is approximately 8.8 months[\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Patients were considered to have non-durable response (NDR) if they died from myeloma or had evidence of disease progression prior to this 9 month cutoff, apart from the patient that developed delayed neurotoxicity as outlined (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). All patients had sufficient follow up to either progress (and thus be labelled NDR) or reach the 9-month landmark cutoff at the time of analysis, and of those who received ide-cel, 26 (65%) of patients had DR and 14 (35%) NDR (Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003eFollowing quality control procedures and removal of batch effects (Online Methods) 247,500 cells were analyzed. Individual cells were classified as CD4 or CD8 using an in-silico gating method from Li et al[\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e] (Supplementary Fig.\u0026nbsp;3A) resulting in 117,530 CD4 cells and 80,939 CD8 cells (Supplementary Fig.\u0026nbsp;3B). A population of monocytes was identified; however, the majority (2536/3137) were isolated from one patient PBMC sample and were not included in subsequent analyses. The CD4 and CD8 cells were then separately integrated, clustered, and annotated based on known markers (Supplementary Figs.\u0026nbsp;4/5). This process identified 11 CD4 subtypes and 12 CD8 subtypes (Figs.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eC-F). Pseudotime trajectory analysis supported annotations of cell subtypes and showed a clear differentiation trajectory in CD8 cells but not CD4 cells (Supplementary Fig.\u0026nbsp;6)\u003c/p\u003e \u003cp\u003eIde-cel product has a high CD4/CD8 ratio and response is associated with more CD4 cells that have higher CAR expression and activation associated gene expression profiles\u003c/p\u003e \u003cp\u003eThe majority of cells from ide-cel infusion product (IP) were identified as CD4, a feature which changed after infusion as the product expanded in peripheral blood (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA/B). CAR-positive (CAR+) cells extracted from PBMC of both ide-cel and cilta-cel patients during early expansion phase proximal to infusion were predominately CD8 (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA/B). We confirmed these findings by immunophenotyping both the IP and PBMC samples used to generate the scRNAseq libraries (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB, Supplementary Fig.\u0026nbsp;7) and found the CD4:CD8 ratios to be highly correlated between scRNAseq data and immunophenotyping (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eC). To detect CAR\u0026thinsp;+\u0026thinsp;cells in single-cell RNA-seq data, a reference transcriptome was created by incorporating ide-cel and cilta-cel construct sequences from respective patents into the GRCh38 human transcriptome, with samples aligned to this reference and cells deemed CAR positive if they had at least one read aligning to the respective construct (Online Methods).) We also confirmed the frequency of CAR\u0026thinsp;+\u0026thinsp;cells using this approach (Supplementary Fig.\u0026nbsp;9A/B) and by immunophenotyping (Supplementary Figs.\u0026nbsp;8/9). The proportion of CAR\u0026thinsp;+\u0026thinsp;CD4 cells in the ide-cel IP, and the expression of the CAR construct in CAR\u0026thinsp;+\u0026thinsp;CD4 cells, was higher in patients who had DR compared with NDR (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eD-E).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eWe next sought to identify potential determinants of response in the infusion product based on cell subtype composition and compared the proportion of cells that were in each of the identified CD4 or CD8 clusters, restricted to CAR expressing cells, between DR and NDR. The composition of infusion product cell subtypes was similar between responders and non-responders (Supplementary Figs.\u0026nbsp;10/11), with products from non-responding (NDR) patients having higher proportions of proliferating CD4 cells (p\u0026thinsp;=\u0026thinsp;0.063) and a cluster of CD4 cells characterized by high expression of glycolysis genes [CD4gly] (p\u0026thinsp;=\u0026thinsp;0.0047). Given the paucity of composition differences in IP cell subtypes between DR and NDR patients, we sought to identify how they might vary transcriptionally and identified differentially expressed genes (DEGs) and biological pathway differences between responders in CD4 and CD8 cells and their respective subtypes.\u003c/p\u003e \u003cp\u003eWe identified upregulation of several gene sets associated with key functions in CAR\u0026thinsp;+\u0026thinsp;CD4 cells in patients with DR reflecting their ability to generate a robust and coordinated immune response. Key pathways included those involved in signaling and cytokine production (IL-6/JAK/STAT3, IL-2/STAT5, cytokine-cytokine receptor interaction, inflammatory response), immune cell activation (TGF-β signaling, KRAS signaling, and IFN-γ/IFN-α pathways, T-cell and Toll-like receptor signaling) and anti-apoptosis pathways TNF-α signaling via NFKB and noncanonical NFκB signaling, promoting survival and persistence (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF).The CAR\u0026thinsp;+\u0026thinsp;CD8 cells from DR patients had upregulation of interferon pathways and genes involved in MHC class II and ribosomal pathways. In addition, we noted upregulation of TCF7 regulon in CAR\u0026thinsp;+\u0026thinsp;CD4 and CAR\u0026thinsp;+\u0026thinsp;CD8 cells of DR patients which has been associated with a more favorable na\u0026iuml;ve T-cell state in CD19-directed CAR-T [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e] (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF). In contrast, CAR\u0026thinsp;+\u0026thinsp;CD4 cells from NDRs demonstrated upregulation of cell proliferation genes, mTORC1 signaling, oxidative phosphorylation, glycolysis, and the TCA cycle (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eF) suggestive of excessive proliferation while CD8 effector cells from NDR had activated hypoxia, p53 and TGFβ pathways, also suggestive of pre-existing dysfunction.\u003c/p\u003e \u003cp\u003eKey upregulated genes in CAR\u0026thinsp;+\u0026thinsp;CD4 IP cells from those with DR included IFN-γ, MAL, CD2, CD69, and GZMA (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eG, Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA), key for T cell signaling, activation, proliferation, and cytotoxicity suggesting a CD4 compartment primed for the generation of a more effective and comprehensive immune response post-infusion[\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e, \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e]. Additionally NEAT1, a long noncoding RNA whose suppression results in impaired CD4 cell differentiation via the STAT3 axis [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e], and the pro-survival gene BIRC3 were upregulated in CAR\u0026thinsp;+\u0026thinsp;CD4 IP cells of DRs (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Genes associated with non-durable responses in CAR\u0026thinsp;+\u0026thinsp;CD4 IP cells include MCM5, MCM7, and CDT1, which are involved in DNA replication and cell cycle progression, indicating potential dysregulation in cell proliferation in NDR (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eA). Using pseudobulk aggregation of read counts from CAR\u0026thinsp;+\u0026thinsp;CD4 cells of each patient, we validated the statistical significance of 8/10 of these genes (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003eB). Differentially expressed genes between responders in specific CD4 IP subtypes have been reported in markdown format and hosted on github (Online reports).\u003c/p\u003e \u003cp\u003eIn CAR\u0026thinsp;+\u0026thinsp;CD8 IP cells, fewer genes were differentially expressed between responders (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH). Genes upregulated in CAR\u0026thinsp;+\u0026thinsp;CD8 DR IP included HLA class II gene HLA-DQA2, GZMK, and BIRC3, associated with enhanced antigen presentation, cytotoxic functions, and cell survival, respectively (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eH, Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA). Conversely, CAR\u0026thinsp;+\u0026thinsp;CD8 IP cells from NDR patients had upregulation of LGALS3, and LIME1 (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA), indicating potential impaired effector function via LAG3 [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] and premature activation of T-cell migration [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e], respectively. GZMK and BIRC3 maintained their significance between responders when comparing expression at the pseudobulk level (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eB).\u003c/p\u003e \u003cp\u003eDifferentially expressed genes between responders in specific CD8 IP subtypes have been reported in markdown format and hosted on github (Online reports).\u003c/p\u003e \u003cp\u003eUpregulation of pro-survival pathways in the ide-cel infusion product is correlated with higher density of CAR expression and improved survival outcomes\u003c/p\u003e \u003cp\u003eTo provide a sample-level signature for biological pathways of interest, we performed pseudobulk aggregation of single-cell read counts by sample, and by cell subtype within each sample, followed by single-sample gene set enrichment analysis (ssgsea). We found that products from patients with durable responses had transcriptional profiles associated with memory T cells (FOXO1 regulon) with downregulation of glycolysis in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 cells (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). Additionally, cytokine/cytokine receptor pathway and cytotoxicity genes were increased in both CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 DR IP (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). We noted that NFKB signaling, tonic signaling, and anti-apoptosis gene signatures were upregulated in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 from DR IP cells and in particular cell subtypes (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eA, Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e). As there is a known promotion of CAR-T survival via 4-1BB signaling[\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e], we investigated the relationship between expression of the CAR and NFKB signaling and the anti-apoptosis gene signature, finding a strong correlation between CAR expression and these signatures in CAR\u0026thinsp;+\u0026thinsp;CD4 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eB/C) but not in CAR\u0026thinsp;+\u0026thinsp;CD8 cells (Supplementary Fig.\u0026nbsp;12). Tonic signaling, based on expression of a specific gene signature linked to tonic signaling [\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e] was also correlated with CAR expression and anti-apoptosis (Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eD/E, Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eGiven the association observed with these gene signatures and durable responses, we investigated their association with survival outcomes. We stratified patients as expressing these signatures above (high) or below (low) the median for all IP samples and identified that NFKB signaling and anti-apoptosis signatures in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 cells were both significantly associated with progression-free survival (PFS) (Figs.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003eF-I) and the anti-apoptosis signature significantly discriminated between groups for overall survival (OS) (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003e). Additionally, the tonic signaling signature in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 cells was associated with OS but not PFS (Extended Data Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003e).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eIde-cel exhibits marked changes in cellular composition, transcriptional profiles, and clonotype dynamics following infusion\u003c/p\u003e \u003cp\u003eFollowing infusion, ide-cel CAR-T demonstrate a substantial shift in CD4:CD8 ratios as previously outlined (Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003eA/B, Supplementary Fig.\u0026nbsp;7), and shifts within subtypes of both CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 cell compartments (Supplementary Figs.\u0026nbsp;13/14). Proportionally, there was an increase in CD4em in PBMCs relative to IP (p\u0026thinsp;=\u0026thinsp;0.001) and relative decreases in Th2, a population of cells defined by high expression of histone genes (CD4hist), and Tregs (Supplementary Fig.\u0026nbsp;13). Within the CD8 compartment, there were increases in effector memory (CD8em) and terminal effector memory (CD8tem) cells and decreases in stem central memory (CD8scm), CD8tc2 [\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e, \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e], and proliferating cells (Supplementary Fig.\u0026nbsp;14).\u003c/p\u003e \u003cp\u003eTo elucidate determinants of response following infusion, we assessed the transcriptional differences in CAR\u0026thinsp;+\u0026thinsp;ide-cel PBMCs between DR and NDR patients. Ribosomal genes were upregulated in the CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 PBMCs of DR patients and the FOXO1 regulon was upregulated in CAR\u0026thinsp;+\u0026thinsp;CD4 PBMCs of DR patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Glycolysis, fatty acid metabolism, and cell proliferation pathways were upregulated in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 PBMCs of NDR patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). CAR\u0026thinsp;+\u0026thinsp;CD8scm PBMCs in NDR patients were also enriched for an exhaustion signature and upregulation of interferon signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eA). Although CD8scm typically do not express exhaustion markers, the fact that these genes are differentially upregulated in NDR suggest that they are more prone to the development of exhaustion, which may contribute to their attenuated therapeutic effect.\u003c/p\u003e \u003cp\u003eInterrogating individual genes, in patients with DR we identified upregulation of genes associated with the maintenance of a more na\u0026iuml;ve phenotype, including KLF2, CD27, TCF7, and DUSP2 in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 cells (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB-D), and multiple ribosomal genes in the CAR\u0026thinsp;+\u0026thinsp;CD8 cells of DRs. In contrast, ENO1, CD38, and multiple metallothionein genes were upregulated in CD8 cells of NDRs (Figures C-D). Additionally, expression of the ide-cel construct was again upregulated in CD4 and CD8 cells of DRs (Figs.\u0026nbsp;\u003cspan refid=\"Fig4\" class=\"InternalRef\"\u003e4\u003c/span\u003eB-D).\u003c/p\u003e \u003cp\u003eFollowing infusion into patients, CAR\u0026thinsp;+\u0026thinsp;cells exhibited increases in clonality (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eA). We analyzed transcriptional differences between clones composing a small percentage of the total clonal population in a sample (0\u0026ndash;0.1%) versus expanding clones (\u0026gt;\u0026thinsp;0.1%) in all CAR\u0026thinsp;+\u0026thinsp;IP and PBMC ide-cel samples (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eB, Supplementary Fig.\u0026nbsp;15). As expected, CAR\u0026thinsp;+\u0026thinsp;CD4 IP cells that expanded in vivo exhibited an upregulation of genes associated with protein translation and cell proliferation. Non-expanding CAR\u0026thinsp;+\u0026thinsp;CD4 IP cells had upregulation of hypoxic pathways and interferon signaling (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC). No non-expanding clones were detected in CAR\u0026thinsp;+\u0026thinsp;CD4 PBMC cells. Conversely, genes associated with immune response, NFKB signaling, and the TCF regulon were upregulated in the IP of expanding CD8 cells, whereas ribosomal genes and genes associated with proliferation were upregulated in non-expanding CD8 cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eC).\u003c/p\u003e \u003cp\u003eWe next assessed the transcriptional differences between CAR\u0026thinsp;+\u0026thinsp;clones that were present in both IP and PBMC (paired) or only present at either time point (unpaired) in the IP and PBMC samples of 6 ide-cel patients (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eD, Supplementary Fig.\u0026nbsp;16). Pathway enrichment analysis revealed unpaired clones were enriched in hypoxia, apoptosis, interferon response, and p53 pathways in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 IP and PBMC cells. Paired clones were enriched in MHC class II genes in CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 IP and PBMC cells and ribosomal genes in CAR\u0026thinsp;+\u0026thinsp;CD8 IP and PBMC cells (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eE). Additionally, paired clones were more likely to expand in CD4 IP, CD8 IP, and CD8 PBMC cells (chi square p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.0001, Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003eIde-cel features associated with efficacy are accentuated in cilta-cel; including higher CAR expression, enhanced NFKB and ribosomal signaling, and greater clonal diversity\u003c/p\u003e \u003cp\u003eGiven the improved depth and durability of responses reported in clinical trials enrolling RRMM patients treated with cilta-cel when compared to ide-cel [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e, \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e, \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e7\u003c/span\u003e, \u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e, \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e10\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e], we investigated cell type composition, gene expression, and biological pathway differences in CAR-T cells enriched from PBMCs collected during expansion-phase between the two products.\u003c/p\u003e \u003cp\u003eCompared to ide-cel, the expanding CAR\u0026thinsp;+\u0026thinsp;CD4 compartment of cilta-cel demonstrated a higher proportion of CD4em cells (p\u0026thinsp;=\u0026thinsp;0.0087) while ide-cel was proportionally higher in CD4scm (p\u0026thinsp;=\u0026thinsp;0.043) (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eA, Supplementary Fig.\u0026nbsp;17). In contrast, CD8scm were higher proportionally in cilta-cel (p\u0026thinsp;=\u0026thinsp;0.0087) and CD8tem trended towards an increase in ide-cel, although not reaching significance (p\u0026thinsp;=\u0026thinsp;0.13, Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eB, Supplementary Fig.\u0026nbsp;18).\u003c/p\u003e \u003cp\u003eWe identified gene sets previously associated with key functions upregulated in cilta-cel when compared to ide-cel, notably in NKFB signaling and ribosomal pathways (Extended Data Fig.\u0026nbsp;7A). Compared to CAR\u0026thinsp;+\u0026thinsp;CD4 cells of ide-cel, CAR\u0026thinsp;+\u0026thinsp;CD4 cells of cilta-cel cells had higher expression of CD27, GZMK, TCF7 (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC, Extended Data Fig.\u0026nbsp;7B-C). CD27 and GZMK were also significantly upregulated in CAR\u0026thinsp;+\u0026thinsp;CD8 cells of cilta-cel (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eC). The expression of the CAR construct was significantly higher in cilta-cel across both CD4 and CD8 subpopulations (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eD) and we confirmed this difference in expression via flow cytometry (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eE, Supplementary Fig.\u0026nbsp;19). Mean fluorescent intensity (MFI) was approximately four-fold higher, which suggests more CAR per cell even accounting for the additional binding domains of the cilta-cel construct. Pseudobulk aggregation of single-cell read counts by sample followed by ssgsea confirmed NFKB signaling and ribosomal pathways to be significantly higher in cilta-cel when compared with ide-cel (Fig.\u0026nbsp;\u003cspan refid=\"Fig6\" class=\"InternalRef\"\u003e6\u003c/span\u003eF).\u003c/p\u003e \u003cp\u003eWe also noted that the TCR repertoire of cilta-cel was far less clonal than ide-cel. The ide-cel TCR repertoires were comprised of medium and larger expanding clones (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eG) while the TCR repertoire was significantly more diverse in CAR\u0026thinsp;+\u0026thinsp;CD8 cells of cilta-cel (Fig.\u0026nbsp;\u003cspan refid=\"Fig5\" class=\"InternalRef\"\u003e5\u003c/span\u003eH). We also found the diversity of the TCR repertoire to be higher in CAR\u0026thinsp;+\u0026thinsp;cells than in the CAR- cells in all PBMC and IP cells (Supplementary Fig.\u0026nbsp;20).\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThis single cell analysis provides key insights into the product characteristics that are important for durable responses, shedding light on the different outcomes that have been reported in clinical trials to date. Our findings from the infusion product suggest that cells highly expressing NFKB signaling signatures and the downstream, pro-survival target genes of this pathway (e.g. BIRC3)[\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e], in addition to gene signatures associated with immune response and tonic signaling, prime CAR-T for both activation and survival following infusion. Additionally, we show that upregulation of cell proliferation pathways in CAR\u0026thinsp;+\u0026thinsp;CD4 IP cells is associated with poor outcome following treatment. In CAR-T with CD28 co-stimulatory domains, tonic signaling can lead to exhaustion during \u003cem\u003eex vivo\u003c/em\u003e expansion[\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e]. However, its effect in CAR-T utilizing 4-1BB co-stimulatory domains remains less well defined. Rodriguez-Marquez et al. demonstrated that increased density of CAR (CAR\u003csup\u003ehigh\u003c/sup\u003e T-cells) produced by higher numbers of viral integrations could trigger tonic signaling[\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e]. These CAR\u003csup\u003ehigh\u003c/sup\u003e T-cells also produced more cytokines and demonstrated an increase in cytotoxicity \u003cem\u003ein vitro\u003c/em\u003e. Our data also showed increased CAR-expression in the pre-infusion products of patients that had durable responses. Although there exists concern that tonic signaling has potentially negative consequences, others have challenged this dogma[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e]. In this context, tonic signaling may be desirable in particular for constructs with a 4-1BB co-stimulatory domain[\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e, \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e, \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e, \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e]. Singh et al. (2021) found that tonic 4-1BB signaling can be protective against dysfunction and actually enhance CAR T cell function, which would support our findings[\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eThe upregulation of ribosomal pathways in the expanding products of durable responders also highlights the importance of sustained protein synthesis for durable anti-tumor activity. This hypothesis is supported by gene expression, functional correlation, and proteomic studies, which demonstrate that activated T cells allocate substantial bioenergetic resources to ribosome biogenesis, increasing ribosomal output more than 13-fold following activation[\u003cspan additionalcitationids=\"CR31 CR32 CR33 CR34\" citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e]. Taken together, these data suggest that a strong signal generated through the CAR combined with increased ribosomal capacity can lead to improved function.\u003c/p\u003e \u003cp\u003eIn addition, we confirmed the positive influence of higher levels of CD27 expression on both CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 PBMC cells, shown to promote T cell expansion not by affecting cell cycle activity, but by stimulating the survival of activated T cells[\u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e36\u003c/span\u003e]. Harnessing the potential of ectopic CD27 expression is already under evaluation with CD27-Armored BCMA-CAR T showing promising results[\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eA striking finding in our data was that these mechanisms were upregulated in CAR\u0026thinsp;+\u0026thinsp;PBMCs obtained from patients treated with cilta-cel when compared to ide-cel CAR\u0026thinsp;+\u0026thinsp;cells. We observed more CD27 and GZMK expression in both CAR\u0026thinsp;+\u0026thinsp;CD4 and CD8 cells, and increased ribosomal and NKFB pathway expression in cilta-cel. In conjunction with this, we also observed dramatic differences in the expression of the CAR construct.\u003c/p\u003e \u003cp\u003eThe detailed specifics of manufacturing of both products are proprietary, however, it is known that they differ in their promoter utilized in their vector [\u003cspan additionalcitationids=\"CR39\" citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e]. Cilta-cel utilizes an elongation factor-1α (EF-1α) promoter whereas ide-cel uses an MND promoter (myeloproliferative sarcoma virus MPSV enhancer, negative control region NCR deletion, d1587rev primer binding site replacement). Ho et al. (2021) performed a series of experiments comparing these two promoters\u0026rsquo; effect on CAR density and cell functionality[\u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e41\u003c/span\u003e]. The EF-1α resulted in significantly higher density of CAR expression, more cytokine secretion, and equivalent cytotoxicity. Authors concluded that MND could generate a safer product, which clinically has been shown in real-world experience with reduced high grade toxicity and non-relapse mortality observed in patients treated with ide-cel[\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e, \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. However, this is at the expense of depth and durability of response which, in the case of cilta-cel, may be driven by higher levels of CAR expression and the concomitant downstream effects.\u003c/p\u003e \u003cp\u003eOther relevant differences include the greater proportion of small clones and greater clonal diversity seen in the expanding cilta-cel compared with ide-cel products. This could confer some additional advantages to the former, as central and effector memory CAR-T in patients with long-term persistence remain highly polyclonal, whereas patients with limited CAR-T persistence were found to have less diversity[\u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e42\u003c/span\u003e].\u003c/p\u003e \u003cp\u003eWe appreciate there are some aspects that remain unexamined in this work. We did not evaluate peripheral blood samples taken across several different timepoints and selected samples were estimated at peak expansion based on increasing lymphocyte counts. We were unable to examine the pre-infusion cilta-cel product or long term persistence, and this is certainly an area for future research.\u003c/p\u003e \u003cp\u003eThis is, to the best of our knowledge, the largest single cell analysis of real-world anti-BCMA CAR-T therapy that has been performed to date and provides key insights into the differences that underpin the clinical results that are being reported as the trial and real-world data matures. We identify features associated with patient outcomes in the starting product and that could be optimized for future iterations of therapy. These data will serve as a resource for further academic investigations. Increased transparency in providing detailed information post-CAR-T approval, including critical aspects of manufacturing or other information about the product generated for patients, could optimize research efforts, improving patient outcomes.\u003c/p\u003e"},{"header":"Online Methods","content":"\u003cp\u003e\u003cstrong\u003ePatient sample collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll patients had a diagnosis of RRMM and consented to prospective sample collection and research database protocols which were Institutional Review Board (IRB) approved by the University of South Florida (USF). All patient samples were collected at Moffitt Cancer Center under approved Institutional Review Board (IRB) protocols. Excess infusion product was collected via elution of CAR-T infusion product bags following patient treatment. Patient PBMC of post-infusion ide-cel and cilta-cel patients was collected at 7-14 days post-infusion and 11-14 days post infusion for ide-cel and cilta-cel patients, respectively.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFlow cytometry\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCryopreserved cells from ide-cel infusion product were removed from liquid nitrogen and rapidly thawed in a 37˚C water bath prior to being transferred to 10ml of pre-warmed complete media to remove excess of DMSO. Cells were then centrifugated 5 min at 1500 rpm, the cell pellet was washed twice with PBS, and resuspended in 100 µl of a solution containing 1X Live/Dead Fixable green cell stain (Invitrogen, ThermoFisher Scientific) and 1 µL of human Fc block (BD) and incubated for 30 min at room temperature. Surface staining was performed for 30 min at 4°C with antibody mix in MACS buffer with 0.5% BSA (Miltenyi Biotec). Cells were then fixed using IC Fixation Buffer (eBioscience) for 30 min at RT, washed 1X Permeabilization Buffer (eBioscience), and intracellular staining was performed for 30 min at 4°C with antibody mix in 1X Permeabilization Buffer (eBioscience). The following monoclonal antibodies against human antigens were obtained from BD Biosciences: anti-CD3 (SK7), anti-CD8 (SK1), and anti-CD4 (L200). Cells were then assessed for the fraction of CD3\u003csup\u003e+\u003c/sup\u003eCD4\u003csup\u003e+\u0026nbsp;\u003c/sup\u003eand CD3\u003csup\u003e+\u003c/sup\u003eCD8\u003csup\u003e+\u0026nbsp;\u003c/sup\u003ecells.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFrozen PBMC samples from 12 myeloma patients (6 Abecma treated and 6 Carykti treated) collected at estimated time of peak CAR T cell expansion were thawed and a Pan T cell selection was performed per manufacturer’s protocol (Miltenyi). The negative fraction representing enriched T cells was surface stained with fluorochome conjugated antibodies against CD3, CD4, CD8 (BD) and FITC-conjugated soluble BCMA (AcroBio) for 30 minutes in the dark at 4C. Cells were washed twice in MACS buffer and resuspended in 90uL MACS buffer + 10uL anti-FITC magnetic microbeads per sample and incubated at 4C for 15 minutes (Miltenyi). Labeled cells were run through MACS MS columns per manufacturer’s protocol and the positive fraction representing enriched CAR T cells was collected. Cells were incubated with Live/Dead Near IR fixable cell stain (ThermoFisher) for 30 min at room temperature. Cells were washed and incubated in 100uL/tube BD Cytofix buffer for 30 minutes at 4C, washed, resuspended in FACS buffer and stored at 4C in the dark until acquisition. Samples were acquired on a BD FACS Symphony flow cytometer and analyzed using FlowJo software.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle-cell RNA-seq and V(D)J library preparation and sequencing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSingle-cell RNA-sequencing was performed using the 10X Genomics Chromium System (10X Genomics, Pleasanton, CA) by the Molecular Genomics Core at the Moffitt Cancer Center and Research Institute. \u0026nbsp;Cryopreserved infusion product cells were rapidly thawed in a 37C water bath, washed twice by centrifugation, and resuspended in MACS buffer (Miltenyi). Cryopreserved PBMC from peak expansion time points were thawed and washed twice in MACS buffer. A Pan T cell isolation followed by a CAR T cell enrichment by FITC-conjugated soluble BCMA staining and subsequent anti-FITC microbead magnetic column separation was performed as described in the flow cytometry methods above. Following cell thawing with or without enrichment as indicated, cell suspensions were washed twice with 1X PBS (calcium and magnesium-free) containing 0.04% weight/volume BSA. The cells were then resuspended in the same buffer following the 10X Genomics cell preparation guide. The cell viability and counts were obtained by AO/PI dual fluorescent staining and visualization on the Nexcelom Cellometer K2 (Nexcelom Bioscience LLC, Lawrence, MA). Cells were then loaded onto the 10X Genomics Chromium Single Cell Controller at a concentration of 1,000 cells/µl to encapsulate 5,000 cells per sample. \u0026nbsp;Single cells, reagents, and 10X Genomics gel beads were encapsulated into individual nanoliter-sized Gelbeads in Emulsion (GEMs), and reverse transcription of poly-adenylated mRNA was performed inside each droplet at 53°C. The cDNA libraries were then completed in a single bulk reaction by following the 10X Genomics Chromium NextGEM Single Cell 5’ Reagent Kit v2 user guide. The TCR V(D)J enrichments from cDNA were performed using the 10x Genomics Chromium single cell human TCR amplification kit. 50,000 sequencing reads per cell for gene expression and 5,000 sequencing reads per cell for V(D)J were generated on the Illumina NovaSeq6000 instrument. The demultiplexing and barcode processing were performed using CellRanger (v7.1.0, 10x Genomics).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGeneration of a reference transcriptome containing ide-cel and cilta-cel sequences\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo detect CAR+ cells in single cell RNA-seq data, we generated a reference transcriptome containing sequences from ide-cel and cilta-cel constructs. The nucleotide sequence of the ide-cel vector was acquired from its respective patent [43]. Nucleotide sequences of the cilta-cel vector containing the CD8A signal peptide, VHH1 CAR BCMA binding domain, G4S linker, V1HI2 CAR BCMA binding domain, CD8A hinge domain, CD8A transmembrane domain, CD137 cytoplasmic domain, and CD3ζ cytoplasmic domain were acquired its respective patent [44] and joined together to serve as a cilta-cel reference sequence. The sequences for each CAR product were then added to the GRCh38 human transcriptome. Two single cell RNA-seq samples from ide-cel infusion product and cilta-cel PBMCs were then aligned to this reference. Correction for incorrect SNPs in the sequences used for each product were done using a pipeline similar to Haradhvala et al. [45]. The code for this pipeline and the sequences used for ide-cel and cilta-cel are available at https://github.com/jeraldnoble .\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle cell RNA-seq preprocessing\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRaw single-cell RNA-seq reads were aligned to the GRCh38 human transcriptome containing ide-cel and cilta-cel CAR sequences using Cell Ranger (v7.1.0, 10x Genomics). Cells were deemed CAR positive they originated from an ide-cel sample and had at least 1 read aligning to the ide-cel construct, or if they originated from a cilta-cel sample had at least 1 read aligning to the cilta-cel construct, and were considered CAR negative otherwise. Genes detected in less than 10 cells were removed from analysis. Low quality cells were filtered from the data set via requiring: at least 1000 UMIs per cell, at least 500 genes detected per cell, log10 genes per UMI of at least 80%, and less than 15% of the mitochondrial genes per cell Doublets were predicted for each sample using Scrublet [46] and DoubletFinder [47] . Cells that were predicted as being doublets by either program were removed from the data set.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eFollowing these quality control measures, all subsequent operations were implemented using functions within Seurat V5 [48]. The top 3000 highly variable genes (HVGs) were identified using the “vst” method within the \u003cem\u003eFindVariableFeatures\u0026nbsp;\u003c/em\u003efunction. T-cell receptor genes, immunoglobulin genes, and the ide-cel and cilta-cell constructs were removed from the HVGs to prevent clustering based on monoclonal T-cells and the expression of the CAR constructs. Cells were scored for S and G2/M cell cycle phases using the \u003cem\u003eCellCycleScoring\u003c/em\u003e function and a list of cell cycle genes (Supplementary File 1). Next, we regressed out the effects of the percentage of mitochondrial genes, UMIs per cell, S cell cycle phase scores, and G2/M cell cycle phase scores using the \u003cem\u003eScaleData\u0026nbsp;\u003c/em\u003efunction. To remove batch effects and inter-sample heterogeneity, integration was performed using Harmony integration within the \u003cem\u003eIntegrateLayers\u0026nbsp;\u003c/em\u003efunction using the top 50 principal components. Following integration, a shared nearest neighbor graph was constructed based on the integrated data using 50 principal components. Then, clusters were generated using the Louvian algorithm within the \u003cem\u003eFindClusters\u003c/em\u003e function with a resolution of 0.75.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIdentification of CD4 and CD8 cell populations\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo classify CD4 and CD8 cells, we leveraged a K-nearest neighbors graph with K=100 using the top 20 batch effect corrected principal components as in Li et al [13] to generate the smoothed expression of CD3, CD4, and CD8. Genes used to calculate the smoothed expression for CD3 were CD3D, CD3E, CD3G, and CD247, for CD8 we used CD8A and CD8B, and used CD4 for CD4. We then generated kernel density estimate plots of the smoothed expression values to define thresholds to classify CD4, CD8, double negative (CD4-CD8-), and double positive (CD4+CD8+) cells. Because populations of CD4 and CD8 cells did not cluster separately (Supplementary Figure S2A), we isolated CD4 and CD8 cells into separate data sets and repeated the above integration and clustering procedures.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCell type annotation\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFollowing integration and clustering of defined CD4 and CD8 populations, signatures for T-cell types were calculated using the \u003cem\u003eAddModuleScore\u003c/em\u003e function with a marker gene list for various T-cell subtypes and states and additional marker gene list from Chu et al. (2023) [49] and Anderson et al. (2023) [50] (Supplementary File 2). Additionally, marker genes for each cluster were identified using the \u003cem\u003eFindAllMarkers\u0026nbsp;\u003c/em\u003efunction requiring a log fold change threshold of 0.25 and only reporting genes with a positive fold change in each cluster. Using the marker gene lists and the marker genes generated by \u003cem\u003eFindAllMarkers\u003c/em\u003e,11 cell subtypes were identified for CD4 cells and 12 cell subtypes were identified CD8 cells (Supplementary Figures 4-5).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDifferential gene expression analysis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eAll single-cell differential gene expression analyses were conducted in CAR+ cells using the \u003cem\u003eFindMarkers\u0026nbsp;\u003c/em\u003efunction. The log fold change parameter was set to 0 when calling this function to enable downstream gene set enrichment analysis and genes were expressed in at least 10% of either of the populations being compared. Genes were considered differentially expressed if they exhibited a log2 fold change \u0026gt; 0.5 and had a Bonferroni adjusted p-value \u0026lt; 0.01.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eGene set enrichment analysis\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eLog2 fold change calculations for each gene were generated by \u003cem\u003eFindMarkers\u003c/em\u003e as stated above and used for gene set enrichment analysis (GSEA). As inputs we utilized the Hallmark gene sets and KEGG reference gene sets available via msigdbr (igordot.github.io/msigdbr/). T-cell phenotype gene sets from Chu et al. (2023) [49], the FOXO1 regulon from Doan el al. (2024) [51], the TCF7 regulon from Chen et al. (2021) [52], the tonic signaling signature from Boroughs et al. (2020) [21], and GOBP canonical and non-canonical NFKB signaling gene sets from gsea-msigdb.org were additionally used for enrichment analysis (Supplementary File 3).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle sample gene set enrichment analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eRead counts for each sample, and each cell subtype within each sample, were aggregated using a pseudobulk approach. Read counts were then normalized using variance stabilizing transformation within DESeq2 [53]. Gene signatures were then calculated with GSVA [54] using the genes sets in Supplementary File 3.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSingle cell V(D)J analyses\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSingle cell VDJ sequencing libraries from each sample were processed using CellRanger (v7.1.0, 10x Genomics) with the GRCh38 vdj reference (https://cf.10xgenomics.com/supp/cell-vdj/refdata-cellranger-vdj-GRCh38-alts-ensembl-7.1.0.tar.gz). Filtered contig files were combined and processed using scRepertoire version 1.11.0 [55] . Samples were combined via the combineTCR function with parameters “removeNA = TRUE” and “filterMulti = TRUE.”\u0026nbsp;\u003c/p\u003e"},{"header":"Declarations","content":"\u003ch2\u003eAcknowledgments\u003c/h2\u003e\n\u003cp\u003eThis work was supported in part by the American Cancer Society, internal funding, the National Cancer Institute (P30CA076292, PI Cleveland), and generous donations from the Hyer family and the Thiel family. F.L.L. is supported in part by the Leukemia and Lymphoma Society as a Clinical Scholar.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe authors thank the Flow Cytometry Core, Molecular Genomics Core and Tissue Core at Moffitt Cancer Center.\u0026nbsp;\u003c/p\u003e\n\u003ch2\u003eAuthor Contributions\u003c/h2\u003e\n\u003cp\u003eJN: data analysis, curation writing the first and final manuscript drafts, XS: biostatistical input and review, \u0026nbsp;MM: immunophenotyping and data generation, CLF: experimental design, oversight, data analysis, curation and writing of all manuscript drafts. All authors reviewed data and contributed to the manuscript review and editing and approved the submitted version.\u003c/p\u003e\n\u003ch2\u003eDeclaration of interests\u0026nbsp;\u003c/h2\u003e\n\u003cp\u003eJN: none\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eXS: none\u003c/p\u003e\n\u003cp\u003eMM: none\u003c/p\u003e\n\u003cp\u003eJAM: none\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eSC: none\u003c/p\u003e\n\u003cp\u003eGdA: none\u003c/p\u003e\n\u003cp\u003eOACP: reports Honoraria/consulting Legend Biotech USA Inc, BMS\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHL: reports a role with the speakers’ bureau for Sanofi\u003c/p\u003e\n\u003cp\u003eMA: reports consulting- Janssen, BMS, Sanofi and research support - BMS\u003c/p\u003e\n\u003cp\u003eTN: reports clinical trial support (to the institution) by Novartis, clinical trial support (drug only supply to the institution) by Karyopharm, Consultancy from ImmunoGen, advisory board: Medexus\u003c/p\u003e\n\u003cp\u003eKHS: reports consultancy, advisor, and/or speaker roles with Adaptive Biotech, Janssen, BMS, Takeda, Sanofi, and Glaxo Smith Kline; research funding with Karyopharm and Abbvie, and funds from BMS, Amgen, and Janssen-funded clinical trials\u003c/p\u003e\n\u003cp\u003eRB: reports Honoraria/consulting BMS Janssen Pfizer and research funding: BMS, Janssen, Abbvie, Regeneron, Karyopharm\u003c/p\u003e\n\u003cp\u003eBB: speakers’ bureau for Sanofi pharmaceuticals\u003c/p\u003e\n\u003cp\u003eAGC: serves on the advisory boards for Janssen and Sanofi; and reports a role with the speakers’ bureau for Sanofi\u003c/p\u003e\n\u003cp\u003eJK: none\u003c/p\u003e\n\u003cp\u003eRA: none\u003c/p\u003e\n\u003cp\u003eDKH: reports research funding from Bristol-Myers Squibb, Karyopharm, and Adaptive Biotech; Consulting or advisory role for Bristol-Myers Squibb, Janssen, Pfizer, and Karyopharm. D.K.H is also supported by the Pentecost Family Myeloma Research Center.\u003c/p\u003e\n\u003cp\u003eYB: none\u003c/p\u003e\n\u003cp\u003eFLL: reports Scientific Advisory Role/Consulting Fees: A2, Allogene, Amgen, Bluebird Bio, BMS, Calibr, Caribou, Cowen, EcoR1, Gerson Lehrman Group (GLG), Iovance, Kite Pharma, Janssen, Legend Biotech, Novartis, Sana, Umoja, Pfizer. Data Safety Monitoring Board: Data and Safety Monitoring Board for the NCI Safety Oversight CAR T-cell Therapies Committee. Research Contracts/Grants: Kite Pharma (Institutional), Allogene (Institutional), CERo Therapeutics (Institutional), Novartis (Institutional), BlueBird Bio (Institutional), 2SeventyBio (Institutional), BMS (Institutional), National Cancer Institute (R01CA244328 MPI: Locke; P30CA076292 PI: Cleveland), Leukemia and Lymphoma Society Scholar in Clinical Research (PI: Locke) Patents, Royalties, Other Intellectual Property: Several patents held by the institution in my name (unlicensed) in the field of cellular immunotherapy. Education or Editorial Activity: \u0026nbsp;Aptitude Health, ASH, BioPharma Communications CARE Education, Clinical Care Options Oncology, Imedex, Society for Immunotherapy of Cancer\u003c/p\u003e\n\u003cp\u003eCLF: reports Honoraria/consulting BMS/Celgene, ONK therapeutics \u0026amp; Janssen; research funding from BMS and Janssen.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eBraunlin, M., et al., \u003cem\u003eTrends in the multiple myeloma treatment landscape and survival: a U.S. analysis using 2011\u0026ndash;2019 oncology clinic electronic health record data\u003c/em\u003e. Leukemia \u0026amp; Lymphoma, 2020: p. 1\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eNidhi Mankan, A.K.N., \u003cem\u003eImprovements in myeloma specific survival over the last two decades.\u003c/em\u003e, in \u003cem\u003eASCO\u003c/em\u003e. 2022, JCO: Chicago.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMaples, K.T., S.A. Scott, and S. Lonial, \u003cem\u003eNovel approaches to the treatment of multiple myeloma\u003c/em\u003e. Clin Adv Hematol Oncol, 2023. 21(5): p. 247\u0026ndash;256.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMunshi, N.C., et al., \u003cem\u003eIdecabtagene Vicleucel in Relapsed and Refractory Multiple Myeloma\u003c/em\u003e. New England Journal of Medicine, 2021. 384(8): p. 705\u0026ndash;716.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBerdeja, J.G., et al., \u003cem\u003eCiltacabtagene autoleucel, a B-cell maturation antigen-directed chimeric antigen receptor T-cell therapy in patients with relapsed or refractory multiple myeloma (CARTITUDE-1): a phase 1b/2 open-label study\u003c/em\u003e. The Lancet, 2021. 398(10297): p. 314\u0026ndash;324.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHansen, D.K., et al., \u003cem\u003eIdecabtagene Vicleucel for Relapsed/Refractory Multiple Myeloma: Real-World Experience From the Myeloma CAR T Consortium\u003c/em\u003e. J Clin Oncol, 2023. 41(11): p. 2087\u0026ndash;2097.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSan-Miguel, J., et al., \u003cem\u003eCilta-cel or Standard Care in Lenalidomide-Refractory Multiple Myeloma\u003c/em\u003e. New England Journal of Medicine, 2023. 389(4): p. 335\u0026ndash;347.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodriguez-Otero, P., et al., \u003cem\u003eIde-cel or Standard Regimens in Relapsed and Refractory Multiple Myeloma\u003c/em\u003e. New England Journal of Medicine, 2023.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCosta, L.J., et al., \u003cem\u003eHow I treat triple-class refractory multiple myeloma\u003c/em\u003e. Br J Haematol, 2022. 198(2): p. 244\u0026ndash;256.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMartin, T., et al., \u003cem\u003eMatching-adjusted indirect comparison of efficacy outcomes for ciltacabtagene autoleucel in CARTITUDE-1 versus idecabtagene vicleucel in KarMMa for the treatment of patients with relapsed or refractory multiple myeloma\u003c/em\u003e. Current Medical Research and Opinion, 2021. 37(10): p. 1779\u0026ndash;1788.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBruno, B., et al., \u003cem\u003eEuropean Myeloma Network perspective on CAR T-Cell therapies for multiple myeloma\u003c/em\u003e. Haematologica, 2021. 106(8): p. 2054\u0026ndash;2065.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSidana, S., et al., \u003cem\u003eReal World Outcomes with Idecabtagene Vicleucel (Ide-Cel) CAR-T Cell Therapy for Relapsed/Refractory Multiple Myeloma\u003c/em\u003e. Blood, 2023. 142: p. 1027.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLi, X., et al., \u003cem\u003eA single-cell atlas of CD19 chimeric antigen receptor T cells\u003c/em\u003e. Cancer Cell, 2023. 41(11): p. 1835\u0026ndash;1837.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, G.M., et al., \u003cem\u003eIntegrative Bulk and Single-Cell Profiling of Premanufacture T-cell Populations Reveals Factors Mediating Long-Term Persistence of CAR T-cell Therapy\u003c/em\u003e. Cancer Discovery, 2021. 11(9): p. 2186\u0026ndash;2199.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnt\u0026oacute;n, O., et al., \u003cem\u003eAn essential role for the MAL protein in targeting Lck to the plasma membrane of human T lymphocytes\u003c/em\u003e. Journal of Experimental Medicine, 2008. 205(13): p. 3201\u0026ndash;3213.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRossi, J., et al., \u003cem\u003ePreinfusion polyfunctional anti-CD19 chimeric antigen receptor T cells are associated with clinical outcomes in NHL\u003c/em\u003e. Blood, 2018. 132(8): p. 804\u0026ndash;814.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShui, X., et al., \u003cem\u003eKnockdown of lncRNA NEAT1 inhibits Th17/CD4(+) T cell differentiation through reducing the STAT3 protein level\u003c/em\u003e. J Cell Physiol, 2019. 234(12): p. 22477\u0026ndash;22484.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKouo, T., et al., \u003cem\u003eGalectin-3 Shapes Antitumor Immune Responses by Suppressing CD8\u0026thinsp;+\u0026thinsp;T Cells via LAG-3 and Inhibiting Expansion of Plasmacytoid Dendritic Cells\u003c/em\u003e. Cancer Immunol Res, 2015. 3(4): p. 412\u0026ndash;23.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePark, I., et al., \u003cem\u003eThe Transmembrane Adaptor Protein LIME Is Essential for Chemokine-Mediated Migration of Effector T Cells to Inflammatiory Sites\u003c/em\u003e. Mol Cells, 2020. 43(11): p. 921\u0026ndash;934.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePhilipson, B.I., et al., \u003cem\u003e4-1BB costimulation promotes CAR T cell survival through noncanonical NF-κB signaling.\u003c/em\u003e Sci Signal, 2020. 13(625).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBoroughs, A.C., et al., \u003cem\u003eA Distinct Transcriptional Program in Human CAR T Cells Bearing the 4-1BB Signaling Domain Revealed by scRNA-Seq.\u003c/em\u003e Molecular Therapy, 2020. 28(12): p. 2577\u0026ndash;2592.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDobrzanski, M.J., et al., \u003cem\u003eTc1 and Tc2 effector cell therapy elicit long-term tumor immunity by contrasting mechanisms that result in complementary endogenous type 1 antitumor responses\u003c/em\u003e. J Immunol, 2004. 172(3): p. 1380\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMittr\u0026uuml;cker, H.-W., A. Visekruna, and M. Huber, \u003cem\u003eHeterogeneity in the Differentiation and Function of CD8\u0026thinsp;+\u0026thinsp;T Cells\u003c/em\u003e. Archivum Immunologiae et Therapiae Experimentalis, 2014. 62(6): p. 449\u0026ndash;458.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerzella, D., et al., \u003cem\u003eLife, death, and autophagy in cancer: NF-kappaB turns up everywhere\u003c/em\u003e. Cell Death Dis, 2020. 11(3): p. 210.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLong, A.H., et al., \u003cem\u003e4-1BB costimulation ameliorates T cell exhaustion induced by tonic signaling of chimeric antigen receptors\u003c/em\u003e. Nat Med, 2015. 21(6): p. 581\u0026ndash;90.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRodriguez-Marquez, P., et al., \u003cem\u003eCAR density influences antitumoral efficacy of BCMA CAR T cells and correlates with clinical outcome\u003c/em\u003e. Sci Adv, 2022. 8(39): p. eabo0514.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHege, K., \u003cem\u003eContext matters in CAR T cell tonic signaling\u003c/em\u003e. Nature Medicine, 2021. 27(5): p. 763\u0026ndash;764.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSingh, N., et al., \u003cem\u003eAntigen-independent activation enhances the efficacy of 4-1BB-costimulated CD22 CAR T cells\u003c/em\u003e. Nature Medicine, 2021. 27(5): p. 842\u0026ndash;850.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrigault, M.J., et al., \u003cem\u003eIdentification of chimeric antigen receptors that mediate constitutive or inducible proliferation of T cells\u003c/em\u003e. Cancer Immunol Res, 2015. 3(4): p. 356\u0026ndash;67.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWolf, T., et al., \u003cem\u003eDynamics in protein translation sustaining T cell preparedness\u003c/em\u003e. Nat Immunol, 2020. 21(8): p. 927\u0026ndash;937.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAraki, K., et al., \u003cem\u003eTranslation is actively regulated during the differentiation of CD8(+) effector T cells\u003c/em\u003e. Nat Immunol, 2017. 18(9): p. 1046\u0026ndash;1057.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eClaiborne, M.D., et al., \u003cem\u003ePersistent CAD activity in memory CD8(+) T cells supports rRNA synthesis and ribosomal biogenesis required at rechallenge\u003c/em\u003e. Sci Immunol, 2022. 7(71): p. eabh4271.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhou, L., et al., \u003cem\u003eT cell proliferation requires ribosomal maturation in nucleolar condensates dependent on DCAF13\u003c/em\u003e. Journal of Cell Biology, 2023. 222(10).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHukelmann, J.L., et al., \u003cem\u003eThe cytotoxic T cell proteome and its shaping by the kinase mTOR\u003c/em\u003e. Nat Immunol, 2016. 17(1): p. 104\u0026ndash;12.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan, T.C.J., et al., \u003cem\u003eSuboptimal T-cell receptor signaling compromises protein translation, ribosome biogenesis, and proliferation of mouse CD8 T cells.\u003c/em\u003e Proceedings of the National Academy of Sciences, 2017. 114(30): p. E6117-E6126.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHendriks, J., Y. Xiao, and J. Borst, \u003cem\u003eCD27 promotes survival of activated T cells and complements CD28 in generation and establishment of the effector T cell pool\u003c/em\u003e. J Exp Med, 2003. 198(9): p. 1369\u0026ndash;80.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXu, Y., et al., \u003cem\u003eCD27-Armored BCMA-CAR T Cell (CBG-002) Therapy for Relapsed and Refractory Multiple Myeloma: A Phase I Clinical Trial\u003c/em\u003e. Blood, 2023. 142(Supplement 1): p. 3468\u0026ndash;3468.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMitra, A., et al., \u003cem\u003eFrom bench to bedside: the history and progress of CAR T cell therapy\u003c/em\u003e. Frontiers in Immunology, 2023. 14.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u003cem\u003eCiltacabtagene autoleucel: regenerative medical product submitted for marketing approval conducted by the Pharmaceuticals and Medical Devices Agency (PMDA).\u003c/em\u003e. Committee on Regenerative Medicine Products and Biotechnology 7/6/2024]; Available from: extension://\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003eefaidnbmnnnibpcajpcglclefindmkaj/https://www.pmda.go.jp/files/000252610.pdf\u003c/span\u003e\u003cspan address=\"http://efaidnbmnnnibpcajpcglclefindmkaj/https://www.pmda.go.jp/files/000252610.pdf\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDELL'ARINGA, T.C.R.H.E.H.Y.J.S.K.E.T.J.F.N.M.R.L.J., \u003cem\u003eUses of anti-bcma chimeric antigen receptors\u003c/em\u003e, in \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://patents.google.com/patent/WO2021091978A1/en?oq=WO+2021%2f091978+A1\u003c/span\u003e\u003cspan address=\"https://patents.google.com/patent/WO2021091978A1/en?oq=WO+2021%2f091978+A1\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. 2020.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHo, J.Y., et al., \u003cem\u003ePromoter usage regulating the surface density of CAR molecules may modulate the kinetics of CAR-T cells in vivo\u003c/em\u003e. Mol Ther Methods Clin Dev, 2021. 21: p. 237\u0026ndash;246.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBiasco, L., et al., \u003cem\u003eClonal expansion of T memory stem cells determines early anti-leukemic responses and long-term CAR T cell persistence in patients\u003c/em\u003e. Nat Cancer, 2021. 2(6): p. 629\u0026ndash;642.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eThompson, E.G. and N.T. Martin, \u003cem\u003eUses of chimeric antigen receptor (car) t-cell therapies in combination with inhibitors of inflammation-related soluble factors\u003c/em\u003e. 2024, Google Patents.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchecter, J.M. and X. Fan, \u003cem\u003eBCMA-targeted CAR-T cell therapy for multiple myeloma\u003c/em\u003e. 2022, Google Patents.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHaradhvala, N.J., et al., \u003cem\u003eDistinct cellular dynamics associated with response to CAR-T therapy for refractory B cell lymphoma\u003c/em\u003e. Nat Med, 2022. 28(9): p. 1848\u0026ndash;1859.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWolock, S.L., R. Lopez, and A.M. Klein, \u003cem\u003eScrublet: Computational Identification of Cell Doublets in Single-Cell Transcriptomic Data\u003c/em\u003e. Cell Syst, 2019. 8(4): p. 281\u0026ndash;291 e9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcGinnis, C.S., L.M. Murrow, and Z.J. Gartner, \u003cem\u003eDoubletFinder: Doublet Detection in Single-Cell RNA Sequencing Data Using Artificial Nearest Neighbors\u003c/em\u003e. Cell Syst, 2019. 8(4): p. 329\u0026ndash;337 e4.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHao, Y., et al., \u003cem\u003eDictionary learning for integrative, multimodal and scalable single-cell analysis\u003c/em\u003e. Nat Biotechnol, 2023.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChu, Y., et al., \u003cem\u003ePan-cancer T cell atlas links a cellular stress response state to immunotherapy resistance\u003c/em\u003e. Nat Med, 2023. 29(6): p. 1550\u0026ndash;1562.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAnderson, N.D., et al., \u003cem\u003eTranscriptional signatures associated with persisting CD19 CAR-T cells in children with leukemia\u003c/em\u003e. Nat Med, 2023. 29(7): p. 1700\u0026ndash;1709.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDoan, A.E., et al., \u003cem\u003eFOXO1 is a master regulator of memory programming in CAR T cells\u003c/em\u003e. Nature, 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen, G.M., et al., \u003cem\u003eIntegrative Bulk and Single-Cell Profiling of Premanufacture T-cell Populations Reveals Factors Mediating Long-Term Persistence of CAR T-cell Therapy\u003c/em\u003e. Cancer Discov, 2021. 11(9): p. 2186\u0026ndash;2199.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLove, M.I., W. Huber, and S. Anders, \u003cem\u003eModerated estimation of fold change and dispersion for RNA-seq data with DESeq2\u003c/em\u003e. Genome Biol, 2014. 15(12): p. 550.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHanzelmann, S., R. Castelo, and J. Guinney, \u003cem\u003eGSVA: gene set variation analysis for microarray and RNA-seq data\u003c/em\u003e. BMC Bioinformatics, 2013. 14: p. 7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBorcherding, N., N.L. Bormann, and G. Kraus, \u003cem\u003escRepertoire: An R-based toolkit for single-cell immune receptor analysis\u003c/em\u003e. F1000Res, 2020. 9: p. 47.\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Single-cell RNA sequencing, Ide-cel, Abecma, Cilta-cel, Carvykti, CAR T-cell, B-Cell Maturation Antigen, Promoter, CAR density, clonal diversity","lastPublishedDoi":"10.21203/rs.3.rs-4994668/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-4994668/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eChimeric antigen receptor T-cells targeting BCMA have revolutionized the treatment of relapsed/refractory multiple myeloma (RRMM) with two approved products, idecabtagene vicleucel (ide-cel) and ciltacabtagene autoleucel (cilta-cel). To explore biological differences, we analyzed pre-infusion products (IP) and CAR-enriched peripheral blood mononuclear cells (PBMCs) at expansion using single-cell RNA sequencing (scRNAseq) from 52 samples. Post-quality control 247,500 cells (117,530 CD4, 80,939 CD8) were analyzed. We found that ide-cel IPs from durable responders (DR) had higher construct expression, enhanced NFKB signaling, and anti-apoptotic signatures, correlating with improved progression free survival. CAR\u0026thinsp;+\u0026thinsp;ide-cel PBMCs in DRs showed upregulated ribosomal genes and higher CD27, KLF2, TCF7 expression. Relative to ide-cel, cilta-cel CAR\u0026thinsp;+\u0026thinsp;cells showed higher expression of CD27, GZMK, TCF7, and a 4-fold increase in CAR expression. In addition, the TCR repertoire was less clonal and more diverse. This study elucidates the distinct characteristics of ide-cel and cilta-cel, offering insights into their differing clinical efficacy.\u003c/p\u003e","manuscriptTitle":"Single-Cell Analysis Reveals Ide-cel and Cilta-cel Characteristics That Influence Efficacy","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2024-10-11 09:12:55","doi":"10.21203/rs.3.rs-4994668/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b963e432-5f6e-4bbe-a0c7-485863d8bc43","owner":[],"postedDate":"October 11th, 2024","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":37329297,"name":"Health sciences/Medical research/Translational research"},{"id":37329298,"name":"Biological sciences/Cancer/Haematological cancer/Myeloma"},{"id":37329299,"name":"Biological sciences/Cell biology/Cell signalling"}],"tags":[],"updatedAt":"2024-11-04T19:15:33+00:00","versionOfRecord":[],"versionCreatedAt":"2024-10-11 09:12:55","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-4994668","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-4994668","identity":"rs-4994668","version":["v1"]},"buildId":"qtupq5eGEP_6zYnWcrvyt","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}