Search

Background: AntiCD19 CAR-T cells are effective against chemorefractory B cell lymphoma. Patients (pts) with rapidly progressive disease and urgent need for therapy have very poor prognosis and may not be able to receive CAR-T cells in time. Decreasing the apheresis to infusion time can make CAR-T cells rapidly available. We conducted a dual-center phase I trial using on-site manufacture of CAR-T cells for treatment of relapsed and refractory (r/r) B cell lymphoma. Methods: Adult pts with r/r CD19+ B cell lymphomas who failed ≥ 2 lines of therapy were enrolled. Autologous T cells were transduced with a lentiviral vector (Lentigen Technology, Inc, LTG1563) encoding an antiCD19 binding motif, CD8 linker, TNFRS19 transmembrane region, and 4-lBB/CD3z intracellular signaling domains. GMP-compliant manufacture was done using CliniMACS Prodigy in a 12-day culture, subsequently shortened to 8 days. Dose escalation was done using 3+3 design. Lymphodepletion included cyclophosphamide (60mg/kg x 1) and fludarabine (25mg/m2/d x 3). Cytokine release syndrome (CRS) and immune effector cell associated neurotoxicity syndrome (ICANS) were graded using the Lee and CARTOX criteria, respectively. CAR-T persistence was measured with qPCR and flow cytometry. Plasma cytokine concentrations were measured using electrochemiluminescence (MesoScale Diagnostics, Inc). Results: Thirty-one pts were enrolled and treated. Baseline patient and disease characteristics are listed in table 1. Twenty-nine (94%) pts were refractory to the prior line of therapy and 21 (68%) had symptomatic disease at the time of lymphocyte collection. CAR-T cell product manufacture was successful in all pts. Median transduction rate was 45% [range 15-66], median culture expansion was 36-fold [range 3-79]. CAR-T cell doses were 0.5 x 10 6/kg (n = 4), 1 x 10 6/kg (n = 16), and 2 x 10 6/kg (n = 11). Median time from apheresis to lymphodepletion was 7 days (range 2 - 15) and median time from apheresis to CAR-T cell infusion time was 13 days (range 9 - 20). Twenty-eight pts were infused fresh product. Seventeen pts (55%) experienced CRS. Grade 1-2 CRS was observed in 15 pts (48%), grade ≥ 3 was observed in 3 pts (10%). One patient had grade 4 CRS that was later complicated by hemophagocytic syndrome and died on day 21; a second patient had grade 5 CRS in the context of bulky disease and died on day 8. Ten pts (32%) had ICANS and 4 pts had grade 3-4 ICANS. Treatment for CRS / ICANS included tocilizumab (n = 12), siltuximab (n = 4), anakinra (n = 3) and corticosteroids (n = 10). The most common all grade non - hematologic toxicity was fatigue, observed in 19 pts, all grade 1. Hematologic toxicity was common, with grade ≥ 3 neutropenia observed in all subjects. Twenty-five (81%) presented disease response and twenty-two pts (71%) achieved complete response (CR). There were no statistically significant differences in the overall and complete response rates between dose levels. After a median follow up of 18 months (range 1 - 32), 5 pts relapsed, and 7 pts have died. Causes of death include progressive disease (n=5), CRS (n=1) and CRS/HLH (n=1). Two-year estimates of PFS and OS for the whole cohort were 67% (95%CI 52-88%) and 75% (95%CI 60-93%)(fig1), respectively. Two-year estimates for patients achieving disease response (CR or PR) were 82% (95%CI 67-99%) and 90% (95%CI 78-100%), respectively. The median duration of response has not been reached (95% CI 74-100). Among pts achieving CR, 94% (95% CI 61-100%) had sustained remission at 12 months. Median time to peak CAR-T expansion, measured by PCR, was 14 days (IQR 14-19), without differences between dose levels, culture duration or fresh vs. cryopreserved infusion. All evaluable subjects had persistent CAR-Ts on PCR measurements done on days 30, 60 and 90. CAR-T cell dose did not have an impact in the time to peak in vivo CAR-T cell expansion or in the rate of CAR-T cell persistence (fig 2). Cytokine measurements have been conducted in 19 pts, with area under the curve (AUC) analyses showing pts with CRS had higher plasma concentrations of multiple cytokines (fig 3). Patients achieving CR had higher plasma concentrations of MIP3B. Conclusions: Second generation antiCD19 CAR-T cells with TNFRS19 transmembrane domain have potent clinical activity. On-site manufacture was successful in all pts. This strategy, in combination with fresh product infusion, can make CAR-T cell therapy rapidly available for pts with high-risk r/r B cell lymphoma. Figure 1 Figure 1. Disclosures Caimi: Amgen Therapeutics.: Consultancy; TG Therapeutics: Honoraria; XaTek: Patents & Royalties: Royalties from patents (wife); Kite Pharmaceuticals: Consultancy; Genentech: Research Funding; ADC Theraputics: Consultancy, Research Funding; Seattle Genetics: Consultancy; Verastem: Consultancy. Ghobadi: Wugen: Consultancy; Atara: Consultancy; Amgen: Consultancy, Research Funding; Kite, a Gilead Company: Consultancy, Honoraria, Research Funding; Celgene: Consultancy. Schneider: Lentigen Technology: Current Employment. Boughan: Beigene: Speakers Bureau. Metheny: Incyte: Speakers Bureau; Pharmacosmos: Honoraria. Krueger: Lentigen: Current Employment. Kadan: Lentigen: Current Employment. Orentas: Lentigen: Patents & Royalties. Dropulic: Lentigen: Ended employment in the past 24 months, Patents & Royalties. de Lima: Miltenyi Biotec: Research Funding; Incyte: Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees; Pfizer: Membership on an entity's Board of Directors or advisory committees. OffLabel Disclosure: AntiCD19 CAR-T cells with TNFRSF19 transmembrane domain for treatment of relapsed and refractory B cell lymphomas.

Background: Chimeric antigen receptor T cell (CAR-T) therapy is known to produce durable remissions in the treatment of CD19 + relapsed/refractory B cell malignancies. Nonetheless, many patients receiving CAR-T cells will fail to respond for unknown reasons. Here, we employed single cell RNA sequencing and protein surface marker profiling of serial CAR-T cell samples from patients with non-Hodgkin's lymphoma (NHL) to reveal CAR-T cell evolution, identify biomarkers of response, and test for evidence of exhaustion in CAR-T cells of poor responders. Methods: To describe the evolution of CAR-T cells after infusion into NHL patients and to identify mechanisms and biomarkers of response, our study examined manufactured CAR-T cell products and FACS isolated CAR-T cells from post-infusion blood samples from patients treated for CD19 + relapsed/refractory NHL. Utilizing scRNA sequencing and flow cytometry, we investigated time points after infusion that are known from previous studies to be associated with peak expansion (day 14) and contraction (day 30) and represent key changes in CAR-T cell activity. Altogether, our datasets include 14 manufactured CAR-T cell products, 13 samples from day 14, and 12 samples from day 30. This sampling represents 10 patients with favorable response (complete or partial remission (CR; PR)) and 4 patients with poor response (stable or progressive disease (SD; PD)). To isolate CAR-T cells for scRNA sequencing, viable CD3 +CAR + cells were sorted from cryopreserved CAR-T cell products or PBMCs. Next, libraries were generated with the 10x Genomics Chromium single cell 3' platform with feature barcoding technology to allow simultaneous and paired quantification of transcriptional and cell surface protein expression in individual CAR-T cells. The libraries were sequenced and the data stringently filtered. Batch effect removal was applied to remove differences due to sample preparation or sequencing. Results: Post-filtering, our dataset contained 94,000 cells with an average of 3,917 cells per sample, 8,518 reads per cell, and 2263 unique detectable genes per cell. After dimension reduction and clustering, eleven high-frequency clusters were identified with cluster patterns corresponding with time point, cell cycle phase, cell type, or patient (Figure 1). At the transcriptional level, post-infusion CD8 CAR-T cells displayed significant upregulation of transcription factors (PRDM1, EOMES) and cytotoxic effector molecules (GZMB, PRF1, GZMK, CCL5) associated with differentiation into cytotoxic effector cells as well as transcription factors associated with exhaustion (TOX, TOX2, NR4A2, NR4A3) (p.adj < 0.05) (Figure 2). These data were corroborated by enrichment of effector and exhaustion gene set signatures after infusion as well as global increases in the protein expression of activation and exhaustion markers CD45RO, CD69, CD57, PD1 (CD279), and TIGIT. Contrasts of CAR-T cells between response groups displayed enrichment of an exhaustion profile in CD8 CAR-T cells of poor responders characterized by significant upregulation of the transcription factors FOS, JUNB, JUND, FOSB, JUN, NR4A2, NFKBIA, and PRDM1 and other markers of exhaustion at the RNA level (p.adj < 0.0001). At the protein level, the frequency of TIGIT expression was 20% greater on average in the CD8 CAR-T cells of poor responders compared to favorable responders (Figure 3). CD8 CAR-T cells expressing TIGIT compared to TIGIT negative CD8 CAR-T cells were enriched in an exhaustion transcriptional profile by gene set enrichment analysis and expressed consistently higher levels of exhaustion markers (PD1, CTLA4, LAG3, TIM3) by flow cytometry, regardless of response group or time point (Figure 4). Conclusions: At the transcriptional and protein levels, we note the evolution of CAR-T cells toward a non-proliferative, highly-differentiated, and exhausted state that is enriched in CAR-T cells of patients with poor response. Furthermore, we identified the checkpoint receptor TIGIT as a novel prognostic biomarker and potential driver of CAR-T cell exhaustion. Figure 1 Figure 1. Disclosures Dropulic: Lentigen: Ended employment in the past 24 months, Patents & Royalties. Caimi: Kite Pharmaceuticals: Consultancy; XaTek: Patents & Royalties: Royalties from patents (wife); Seattle Genetics: Consultancy; Genentech: Research Funding; ADC Theraputics: Consultancy, Research Funding; Verastem: Consultancy; Amgen Therapeutics.: Consultancy; TG Therapeutics: Honoraria. de Lima: Miltenyi Biotec: Research Funding; Pfizer: Membership on an entity's Board of Directors or advisory committees; BMS: Membership on an entity's Board of Directors or advisory committees; Incyte: Membership on an entity's Board of Directors or advisory committees.

INTRODUCTION: Anti-CD19 chimeric antigen receptor T (CAR-T) cells have demonstrated activity against relapsed/refractory lymphomas. Cytokine release syndrome (CRS) and CAR-T related encephalopathy syndrome (CRES/ICANS) are well-known complications of CAR-T cell therapy. Tocilizumab, a humanized monoclonal antibody targeting the interleukin 6 (IL-6) receptor, is approved for treatment of CRS. Our institutional standard was modified to administer prophylactic tocilizumab before infusion CAR-T cell products. We present the outcomes of subjects treated with locally manufactured antiCD19 CAR-T cells (TNFRSF19 transmembrane domain, CD3Zeta/4-1BB costimulatory signaling) with and without prophylactic tocilizumab. METHODS: Relapsed / refractory (r/r) lymphoma patients (pts) treated with anti-CD19 CAR-T cells at our institution were included. Baseline demographic and clinical characteristics, as well as laboratory results were obtained from our Hematologic Malignancies and Stem Cell Therapy Database. Prior to institution of prophylactic tocilizumab, pts received this agent only if they presented evidence of CRS grade 2 or higher. In May 2019, our institutional practice changed to provide tocilizumab 8mg/kg, 1 hour prior to infusion of CAR-T cell product. CRS was measured according to the ASTCT Consensus Grading, whereas CRES was measured using the CARTOX-10 criteria. Comparisons between groups were done with the Mann-Whitney U test for continuous variables and Fisher's exact test for categorical variables. RESULTS: Twenty-three relapsed / refractory lymphoma pts were treated with antiCD19 CAR-T cells; 15 pts received prophylactic tocilizumab. Median follow up was 312 days (range 64 - 679) days. Baseline characteristics are listed in table 1. Both groups were similar: There were no statistically differences in the rate of bulky, refractory disease, prior ASCT or number or prior lines of therapy. Baseline lymphocyte counts, C - reactive protein (CRP) and were also comparable between groups (Table 2). We did not observe immune adverse reactions to tocilizumab infusion. There were no differences in the incidence of cytopenias or infectious complications between groups. CRS of any grade was observed in 6/8 (75%) of pts without prophylactic tocilizumab vs. 6/15 (40%) in pts treated with prophylactic tocilizumab (p = 0.23), whereas CRS grade >1 was observed in 5 pts (62.5%) without prophylactic tocilizumab and in 3 pts (20%) treated with prophylactic tocilizumab (p = 0.02). There was no significant difference in the incidence of all grade CRES (no prophylaxis, 3/8 [38%] pts; prophylaxis 5/15 [30%] pts, p = 0.2969). There was a statistically significant difference in the peak CRP and peak ferritin without difference in the peak lymphocyte count after CAR-T infusion (Table 2, Figure 1). Patients given prophylactic tocilizumab had higher IL-6 plasma concentrations on day 2 after infusion (Figure 2). Complete response was observed in 4/8 (50%) pts without prophylactic tocilizumab vs. 12/15 (80%) pts with prophylactic tocilizumab (p = 0.18). All pts had detectable Anti-CD19 CAR-T cells on day 30, both groups had peak CAR-T expansion on day 14, with no statistically significant differences in expansion rates between groups. All evaluable subjects have had CAR-T persistence on days 60, 90, 180, and 365. CONCLUSIONS: Use of prophylactic tocilizumab prior to infusion of antiCD19 CAR-T cells is associated with reduced incidence of severe CRS and decreased levels of clinical laboratory markers of inflammation, despite increases in plasma concentration of IL-6. This decreased rate of grade ≥2 CRS is not associated with impaired disease control and did not result in increased rates of neurologic toxicity. Prophylactic tocilizumab does not appear to affect CAR-T cell expansion or persistence. Figure 1 Disclosures Caimi: ADC therapeutics: Other: Advisory Board, Research Funding; Celgene: Speakers Bureau; Amgen: Other: Advisory Board; Bayer: Other: Advisory Board; Verastem: Other: Advisory Board; Kite pharmaceuticals: Other: Advisory Board. Worden:Lentigen, a Miltenyi biotec company: Current Employment. Kadan:Lentigen, a Miltenyi biotec company: Current Employment. Orentas:Lentigen Technology, a Miltenyi Biotec Company: Research Funding. Dropulic:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. de Lima:Celgene: Research Funding; Pfizer: Other: Personal fees, advisory board, Research Funding; Kadmon: Other: Personal Fees, Advisory board; Incyte: Other: Personal Fees, advisory board; BMS: Other: Personal Fees, advisory board. OffLabel Disclosure: Use of tocilizumab as prophylaxis for CRS is not approved, whereas use for treatment is approved and on label.

The B-cell maturation antigen (BCMA) is an immunotherapy target selectively expressed on multiple myeloma cells (MM). Despite recent success of experimental BCMA CAR-T cell therapy, clinical remissions in MM are often short, in part due to low persistence of the BCMA CAR-T cells. Additionally, immunosuppressive factors, notably TGF-β, are known to be elevated in the peripheral blood and tumor microenvironment in the bone marrow of MM patients, potentially contributing to the lack of durability of CAR-T cell therapy. We aimed to develop a fully-human BCMA CAR with long-term persistence and functional resistance to the suppressive effects of TGF-β. Initially, two fully human single chain variable fragments (scFv) specific for BCMA, derived by biopanning of a yeast display human scFv library, were characterized in a CAR format. Each of the two scFvs was cloned into a lentiviral vector CAR backbone, comprised of the CD8 hinge and transmembrane domain, 4-1BB co-stimulatory domain and CD3ζ signaling domain, and termed BCMA1 and BCMA2, respectively. BCMA2 and BCMA1 T cells, generated by lentiviral vector transduction of primary human T cells, exhibited high CAR expression at multiplicities of infection 10 to 40 (BCMA1: 64-81%; BCMA2: 84-94%), and consistently demonstrated potent cytotoxicity in an overnight co-culture with BCMA+ MM cell lines RPMI-8226 and MM1.S, but not the BCMA- 293T cells. BCMA2 CAR showed robust proliferative capacity in response to repeated, long-term exposure to MM1.S target cells, concordant with the sustained CD4 T-cell subset and high IL-2 production during the course of repeated exposure to target cells. This is in contrast to BCMA1, that showed precipitous decrease in CD4+ T-cell subset upon co-culture with MM cells, resulting in a CAR T-cell population dominated by CD8+ T-cells. Furthermore, the BCMA2 demonstrated prolonged potency in clearing tumor cells compared to BCMA1, even after continuous 20-day exposure to target cells. This was further confirmed in a mouse intradermal RPMI-8226 xenograft tumor model, in which infusion of BCMA2 T-cells resulted in the rapid and complete eradication of tumors, while BCMA1 showed a slower decline in tumor burden. To further improve the efficacy of the BCMA2, specifically within the TGF-β-rich immunosuppressive tumor microenvironment, we developed an armored BCMA2 CAR variant, which co-expresses the dominant negative TGF-β RII bicistronically via a 2A sequence (BCMA2-TbnegCAR). In a 10 day-long co-culture assay with MM1.S targets in the presence of spiked 10 ng/ml TGF-β, BCMA2-TbnegCAR retained high proliferative capacity and potent cytotoxicity, while the unarmored BCMA2 had diminished proliferation and substantially reduced cytokine and granzyme B production. Consistently, in the in vivo intradermal tumor model that utilizes RPMI-8226, which is a MM cell line that endogenously produces TGF-β, we observed that BCMA2-TbnegCAR treatment resulted in higher T-cell counts in the tumors and an earlier decrease of tumor burden compared to infusion with BCMA2, further demonstrating the increased efficacy and potency of the BCMA2-TbnegCAR. In conclusion, we have designed and characterized a new fully-human BCMA2-TbnegCAR, with a novel scFv that exhibits robust proliferative capacity and persistent cytotoxicity, and remains functionally resistant to the immunosuppressive effects of TGF-β. This novel BCMA CAR can potentially improve the effectiveness and durability of the current BCMA CAR-T cell therapy. Disclosures Alabanza: Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Vu:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Wu:Lentigen, a Miltenyi Biotec Company: Current Employment. Zhu:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Dropulic:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Schneider:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties.

INTRODUCTION: Chimeric Antigen Receptor (CAR)-T cell therapy is emerging as a powerful treatment for relapsed or refractory B cell lymphomas. However, a variety of escape mechanisms prevent CAR-T cell therapy from being more uniformly effective. To better understand mechanisms of CAR-T failure among patients treated with dual-targeted CAR-T cells, we performed single-cell RNA sequencing of samples from a Phase 1 trial (NCT03019055). The clinical trial used anti-CD20, anti-CD19 CAR-T cells for the treatment of relapsed/refractory B-cell non-Hodgkin Lymphoma. Clinical responses from this study are reported independently (Shah et al. in press in Nat Med). While robust clinical responses occurred, not all patients had similar outcomes. In single-antigen specific CAR-T cells, mechanisms of resistance include antigen down-regulation, phenotype switch, or PD-1 inhibition (Song et al. Int J Mol Sci 2019). However, very little is understood about the mechanisms of failure that are specific to dual-targeted CAR-T cells. Interestingly, loss of CD19 antigen was not observed in treatment failures in the study. METHODS: De-identified patient samples were obtained as peripheral blood mononuclear cells on the day of harvest ("pre" samples), at the peak of in vivo CAR-T cell expansion which varied from day 10 to day 21 after infusion ("peak" samples), and on day 28 post-infusion ("d28" samples). The CAR-T cell infusion product was obtained on day 14 of on-site manufacturing ("product" samples). All samples were cryopreserved and single cell preparation was performed with batched samples using 10X Genomics kits. Subsequent analysis was performed in R studio using the Seurat package (Butler et al. Nat Biotech 2018) with SingleR being used to identify cell types in an unbiased manner (Aran et al. Nat Immunol 2019). RESULTS: We found that distinct T cell clusters were similarly represented in the responder and non-responder samples. The patients' clinical responses did not depend on the level of CAR expression or the percentage of CAR+ cells in the infusion product. At day 28, however, there was a considerable decrease in the percentage of CAR+ cells in the responder samples possibly due to contracture of the CAR+ T cell compartment after successful clearance of antigen-positive cells. In all samples, the CAR-T cell population shifted from a CD4+ to a CD8+ T cell predominant population after infusion. We performed differentially-expressed gene analyses (DEG) of the total and CAR-T cells. In the pre samples, genes associated with T-cell stimulation and cell-mediated cytotoxicity were highly expressed in the responder samples. Since the responders had an effective anti-tumor response, we expected these pathways to also be enriched for in the peak samples; however, this was not the case. We hypothesize that differential expression of the above genes was masked due to homeostatic expansion of the T cells following conditioning chemotherapy. Based on the DEG results, we next interrogated specific genes associated with cytotoxicity, T cell co-stimulation, and checkpoint protein inhibition. Cytotoxicity-associated genes were highly expressed among responder CD8+ T cells in the pre samples, but not in the other samples (Figure 1). Few differences were seen in specific co-stimulatory and checkpoint inhibitor genes at any timepoint in the T cell clusters. We performed gene set enrichment analyses (GSEA). Gene sets representing TCR, IFN-gamma, and PD-1 signaling were significantly increased in the pre samples of the responders but not at later time points or in the infusion products. DISCUSSION: We found a correlation between expression of genes associated with T cell stimulation and cytotoxicity in pre-treatment patient samples and subsequent response to CAR-T cell therapy. This demonstrates that the existing transcriptome of T cells prior to CAR transduction critically shapes anti-tumor responses. Further work will discover biomarkers that can be used to select patients expected to have better clinical outcomes. Figure 1 Disclosures Johnson: Miltenyi Biotec: Research Funding; Cell Vault: Research Funding. Schneider:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties. Dropulic:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Hari:BMS: Consultancy; Amgen: Consultancy; GSK: Consultancy; Janssen: Consultancy; Incyte Corporation: Consultancy; Takeda: Consultancy. Shah:Incyte: Consultancy; Cell Vault: Research Funding; Lily: Consultancy, Honoraria; Kite Pharma: Consultancy, Honoraria; Verastim: Consultancy; TG Therapeutics: Consultancy; Celgene: Consultancy, Honoraria; Miltenyi Biotec: Honoraria, Research Funding.

In the treatment of B cell leukemia, relapse due to antigen to loss or downregulation remains a major challenge. Tumor antigen escape may be mitigated by multi-targeting CAR T cells redirected to CD19 and CD22, and possessing a superior sensitivity to low-density antigens. Using lentiviral transduction of primary T cells, flow cytometry, cell-based assays, and xenograft mouse models, we systematically optimized 22-19 CAR architecture and co-stimulatory domains for best functionality. Fully-human tandem 22-19 CARs with co-stimulatory domains derived from 4-1BB, CD28, ICOS, OX40 or CD27, and hinge and transmembrane domains derived from CD8, CD28, or OX40 were evaluated. The tandem targeting ScFv domain orientation 22-19 was selected based on greater expression and cytotoxicity vs 19-22. All CARs achieved high T cell expression (mean 50-90%), and efficient dose-dependent killing of RajiCD19+CD22+, 293TCD19+, 293TCD22+, but not 293TCD19-CD22- target cells, and elaborated IL-2, IFN-γ, and TNF-a in antigen-dependent manner. CARs' potency in vitro varied by co-stimulatory domain: 4-1BB< OX40, ICOS, CD27 In RajiCD19+CD22+ mouse xenografts, the potency of tumor rejection by the 2nd generation tandem 22-19 CARs was also dependent on co-stimulatory domain, ranking 4-1 BB Low antigen density Raji clones were generated by CRISPR-Cas9-mediated disruption of both CD19 and CD22 expression, followed by lentiviral transduction to express a limited number of antigen molecules on the cell surface. When challenged with Raji CD22 low clone in vitro, 2nd generation tandem CARs with CD28 and ICOS co-stimulation, and the 3rd generation tandem CAR combining CD28 and 4-1BB co-stimulatory domains were more effective than 4-1BB-, CD27-, or OX40-containing tandem CARs. Against RajiCD19 low clone, CARs with CD27 and OX40 domains were more effective than CARs with 4-1BB, and CARs with ICOS or CD28 co-stimulation were the most potent. In summary, the fully-human tandem 22-19 CARs incorporating ICOS and CD28 co-stimulatory domains mitigate tumor antigen escape, exhibit robust anti-tumor function in pre-clinical models, enable superior lysis of CD22low and CD19low tumor clones, and may help improve clinical outcomes. Disclosures Hu: Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Xiong:Lentigen, a Miltenyi Biotec Company: Current Employment. Wu:Lentigen, a Miltenyi Biotec Company: Current Employment. Zhu:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Dropulic:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Schneider:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties.

Introduction Anti CD19 CAR T-cell therapy is a breakthrough immunotherapeutic approach for relapsed, refractory B-cell malignancies. Despite initial excitement, long term progression free survival with single antigen targeted CD19 CAR T-cells range from 30-40% in aggressive B-cell NHL. Loss of target due to selective pressure against CD19 can lead to antigen downregulation and development of a CD19-negative clonal population. To overcome this resistance mechanism, we recently completed a Phase 1 anti-CD19, anti-CD20 (LV20.19) CAR T-cell trial demonstrating the safety of 2.5x10e6 cells/kg dose in patients with relapsed, refractory B-cell lymphomas (in press in Nature Medicine). We report below single cell cytokine studies from final LV20.19 CAR T-cell products using the Isoplexis single-cell proteomics device. Methods LV20.19 CAR T-cells were manufactured onsite using the CliniMACS Prodigy device on a fixed 14-day manufacturing platform with IL-2 for cell expansion (NCT03019055). Isoplexis cytokine analysis was limited to the cohort of patients treated at goal dose of 2.5x10e6 cells/kg (n=16). Leftover LV20.19 CAR T-cells from the original product were thawed, CD4 and CD8 T cells were enriched by immunomagnetic sorting, and the T-cell subsets separately stimulated with CD19-engineered K562 stimulators overnight. The stimulated cells were loaded onto single-cell, Adaptive Immune Isocode chips, and the chips read for 18 hours in an Isolight instrument to assess single-cell production of 32 individual cytokines. Isospeak software was used to analyze the single-cell data. Results 15 patients had adequate samples for analysis. Results were stratified by response (complete responders (CR) versus partial responders (PR) or progressive disease (PD)). The Isoplexis device was utilized to generate a polyfunctional (i.e., ability to produce more that 1 cytokine) index score (PSI) that has been previously reported (Rossi et al. Blood 2018) to correlate with CAR T-cell treatment outcomes. At Day 28 after LV20.19 CAR T-cell therapy, there were 11 CR patients and 4 PR/PD patients. CD4+ LV20.19 CAR T-cells from complete responders had a non-statistically significant decreased PSI score versus CD4+ cells from PR/PD patients (Figure 1A, p=0.38). The PSI score among CD8+ T-cells was similar (p=0.71) (Figure 1A). The increased PSI score among CR patients' CD4+ LV20.19 CAR T-cells was driven primarily by cells producing effector (IFNg and TNFb) cytokines. On a singular cytokine level, compared to PR/PD patient cells, CR patient CAR T-cells produced significantly higher levels of CCL-11, GM-CSF and IL-17A (Figure 1B; p Conclusion CD4+ LV20.19 CAR T-cells in responding patients trended towards a higher PSI than non-responding patients. At the single cell level, differences in the PSI among CD4+ LV20.19 CAR T-cells from responding patients were driven primarily by a few cytokines including IL-17A, CCL11, GM-CSF, IFNg, and TNFb, which may be key to clinical response. While our results are limited by sample size and only 4 PR/PD patients in this cohort, identification of cytokines correlating with response will allow selective modulation to improve clinical outcomes while limiting CRS. Ongoing studies with LV20.19 CAR T-cells will further delineate the role of IL-17A, GM-CSF and CCL-11 production as potential biomarkers of clinical response with bispecific CAR T-cell products. Disclosures Johnson: Cell Vault: Research Funding; Miltenyi Biotec: Research Funding. Fenske:Medical College of Wisconsin: Current Employment. Hamadani:Sanofi Genzyme, AstraZeneca: Speakers Bureau; Janssen R&D; Incyte Corporation; ADC Therapeutics; Celgene Corporation; Pharmacyclics, Omeros, AbGenomics, Verastem, TeneoBio: Consultancy; ADC Therapeutics: Membership on an entity's Board of Directors or advisory committees; Takeda Pharmaceutical Company; Spectrum Pharmaceuticals; Astellas Pharma: Research Funding. Dropulic:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties: CAR-T immunotherapy. Schneider:Lentigen, a Miltenyi Biotec Company: Current Employment, Patents & Royalties. Hari:GSK: Consultancy; Janssen: Consultancy; Takeda: Consultancy; Incyte Corporation: Consultancy; BMS: Consultancy; Amgen: Consultancy. Shah:Miltenyi Biotec: Honoraria, Research Funding; Celgene: Consultancy, Honoraria; Incyte: Consultancy; TG Therapeutics: Consultancy; Verastim: Consultancy; Lily: Consultancy, Honoraria; Cell Vault: Research Funding; Kite Pharma: Consultancy, Honoraria.

Background: Salvage regimens for chemorefractory aggressive lymphoma achieve response rates of approximately 30%. Anti-CD19 CAR-T cells have demonstrated anti-lymphoma activity, but patients (pts) with rapidly progressive disease and urgent need for therapy have worse prognosis and many are not able to receive CAR-T cells in time. Decreasing the time from apheresis to infusion can make CAR-T cells available to pts with rapid progression of their disease. We present the results of a phase I clinical trial using on-site CAR-T manufacture for treatment of relapsed / refractory (r/r) B cell non Hodgkin lymphoma (NHL). Methods: Adult pts with r/r CD19+ B cell lymphomas who failed ≥ 2 lines of therapy were enrolled. Autologous T cells were transduced with a lentiviral vector (Lentigen Technology, Inc, LTG1563) encoding an anti-CD19 binding motif, CD8 linker and tumor necrosis receptor superfamily 19 (TNFRS19) transmembrane region, and 4-lBB/CD3z intracellular signaling domains. GMP-compliant manufacture was done using CliniMACS Prodigy, in a 12-day culture. Dose escalation was conducted according to a 3+3 design. Lymphodepletion was done with cyclophosphamide (60mg/kg x 1) and fludarabine (25mg/m2/d x 3). Cytokine release syndrome (CRS) and CAR-T related encephalopathy syndrome (CRES) were graded using the Lee and CARTOX criteria, respectively. Results: As of July 30, 2019 , 12 pts were enrolled and treated. Baseline characteristics are listed in table 1. 10/12 pts were refractory to the prior line of therapy, 5 had bulky disease and 9 had symptomatic disease at the time of lymphocyte collection. CAR-T cell product manufacture was successful in all pts. Median transduction rate was 48% [range 29-62] with and median culture expansion of 43-fold [range 30-79]. High dimensional flow cytometry showed the infused CD4 and CD8 CAR-T cells express a central memory and transition to memory - like profile, with enrichment for CD27 and high CCR7 expression. In addition, a subset of CD4 and CD8 CAR-T cells expressed effector transcription factors T-BET and GATA3 while CD4 CAR-T clusters express low levels of immune checkpoint blockers PD-1 and BTLA. All enrolled pts received their infusion of anti-CD19 CAR-T cells. CAR-T cell doses were 0.5 x 106/kg (n = 4) and 1 x 106/kg (n = 8). Median apheresis to infusion time was 13 days [range 13-20], 10 products were infused fresh. CAR-T persistence, based on vector sequence, peaked in peripheral blood MNCs between days 14-21. All responding subjects have had CAR-T persistence on follow up PCR measurements (range 1 - 12 months). CAR-T cell dose did not have an impact in the time to peak in vivo CAR-T cell expansion or in the rate of CAR-T cell persistence. Five pts experienced CRS. Grade 1 - 2 CRS was observed in 4 pts; whereas 1 pt died as a consequence of severe CRS in the context of bulky disease. Pharmacologic interventions for CRS included tocilizumab (n = 5), siltuximab (n = 2) and corticosteroids (n = 2). Two subjects presented grade 4 CRES with resolution after corticosteroids, no other grade ≥3 non-hematologic toxicity was observed. The most common all grade non - hematologic toxicity was fatigue, observed in 6 subjects. Hematologic toxicity was common, with grade ≥ 3 neutropenia observed in all subjects, with 4 subjects presenting grade 3 neutropenia without fever beyond day 30. Among 11 pts evaluable for response, 8 pts have achieved complete response (CR) and one had partial response (PR). Two pts did not respond. For the intention to treat population (n=12), the CR rate was 67% and overall response rate (ORR) was 75%. Overall response rates were equal between both dose levels (75%), but CR rates were higher in pts treated with 1 x 106 CAR-T cells (75% vs. 50%). Two pts have died, causes of death include progressive disease (n=1) and CRS (n=1). After a median follow up 3 months (range 1 - 12) all responding pts are alive; 1 subject relapsed 6 months after treatment with CD19+ disease and entered CR after anti-CD19 antibody drug immunoconjugate treatment. Conclusions: In this phase 1 study, second generation anti-CD19 CAR-T cells with TNFRS19 transmembrane domain have potent clinical activity. The short manufacture times achieved by local CAR-T cell manufacture with the CliniMACS Prodigy enables treatment of a very high risk NHL population that would otherwise not be able to receive CAR-T products due to rapidly progressive disease. Disclosures Caimi: ADC Therapeutics: Research Funding; Celgene: Speakers Bureau; Genentech: Research Funding. Schneider:Lentigen Technology, A Miltenyi Biotec Company: Employment. Bakalarz:Genentech: Speakers Bureau. Kruger:Lentigen Technology Inc., A Miltenyi Biotec Company: Employment. Worden:Lentigen Technology, A Miltenyi Biotec Company: Employment. Kadan:Lentigen Technology Inc., A Miltenyi Biotec Company: Employment. Malek:Adaptive: Consultancy; Janssen: Speakers Bureau; Amgen: Speakers Bureau; Celgene: Consultancy; Takeda: Consultancy; Sanofi: Consultancy; Medpacto: Research Funding. Metheny:Takeda: Speakers Bureau; Incyte: Speakers Bureau. Dropulic:Lentigen Technology, A Miltenyi Biotec Company: Employment. OffLabel Disclosure: Clinical Trial of on - site manufactured CAR-T cells. This manufacturing process is under research.

Introduction Chimeric Antigen Receptor modified T (CAR-T) cell therapies have revolutionized the relapsed, refractory B cell malignancy landscape. Due to the complex steps involved with cell production, some third-party companies require T cells to be cryopreserved prior to shipping, while most manufacturers deliver modified CAR-T cells to the treating center in a cryopreserved state. This is vastly different to the approach taken with traditional cell based therapies, specifically allogeneic transplant (allo-HCT), an immunological treatment that relies on a graft-versus-tumor (GVT) effect to prevent disease relapse. Historically, "fresh" stem cells were felt to be superior to cryopreserved products due to concerns that cryopreservation may damage T cells and other mononuclear cells delaying engraftment and limiting GVT reactivity. As a result, in clinical practice most allo-HCT products are still given as fresh infusions without cryopreservation. In a Phase 1 clinical trial evaluating the safety of a bispecific anti-CD19, anti-CD20 CAR (LV20.19CAR), CAR-T cells were produced in a point-of-care fashion utilizing the CliniMACS Prodigy device. Local manufacturing allowed flexibility to administer either fresh LV20.19CAR-T cells without cryopreservation, or if indicated, thawed CAR-T cells post-cryopreservation. Methods Patients (pts) were treated on a Phase 1 dose escalation + expansion trial (NCT03019055) to demonstrate safety of 41BB/CD3z LV20.19CAR-modified T cells for adults with relapsed, refractory B cell NHL including DLBCL, MCL, FL, and CLL. The starting dose was 2.5x10^5 cells/kg with a target dose of 2.5x10^6 cells/kg. All pts received low dose fludarabine (30 mg/m2) x 3 days +cyclophosphamide (500 mg/m2) x 1 day for lymphodepletion. In the Phase 1 dose-escalation cohorts, pts received fractionated CAR-T cells over two days (30% on Day 0 and 70% on Day+1), while expansion cohort pts received CAR-T cells as a single infusion. The goal for all pts was to infuse fresh CAR-T cell prior to cryopreservation, however, CAR-T cell could be cryopreserved and infused at a later date for clinical / logistical reasons. Results A total of 20 pts received LV20.19CAR T cell therapy (Table 1). Fourteen pts received fresh CAR-T cells immediately post-harvest, 5 pts received post-thaw CAR-T cells, and 1 patient received a mixed fresh/cryopreserved product and was not included in this analysis. Reasons for cryopreserved administration was delay due to active infection (N=3), patient preference (N=1), and unexplained neutropenia (N=1). Among 19 evaluable pts, the CR rate (79% vs 40%), mean ferritin, mean CRP, and incidence of CRS and neurotoxicity were all higher in the fresh infusion group (Table 1), but not statistically significant. In terms of LV20.19 CAR-T product characteristics, mean cell viability at infusion was 93% for the fresh infusion group versus 63% for cryopreserved pts (p Conclusions Cryopreservation is known to diminish cell viability and increase clinical costs associated with freezing and storage. To date, there is limited clinical data evaluating outcomes of pts receiving fresh CAR-T cells compared to thawed CAR-T cells post-cryopreservation. Although it is presumed that in-vivo CAR-T cell activity is comparable in both scenarios, among our pts, both cell viability and in-vivo expansion favored pts who received a fresh infusion. Unlike third-party CAR-T cell products where viability is unknown at the time of infusion, we adjusted the final dose to accommodate decreased cell viability. CR rates and incidence of CRS and NTX were higher among fresh infused pts suggesting greater in-vivo activity, although findings were not statistically significant, partially a result of the small sample size. While our findings are limited by small numbers in each cohort and variability in cell dose and diagnosis, these data suggest that cryopreservation of CAR-T cells may impact clinical responses and is a logistical step that needs further investigation. Disclosures Shah: Cell Vault: Consultancy, Equity Ownership; Oncosec: Equity Ownership; Lentigen: Honoraria, Research Funding; Exelexis: Equity Ownership; Geron: Equity Ownership; Celgene: Other: Advisory Board; Incyte: Consultancy; Oncosec: Equity Ownership; Kite Pharma: Other: Advisory Board. Zhu:Miltenyi Biotec: Research Funding. Schneider:Lentigen Technology, A Miltenyi Biotec Company: Employment. Krueger:Lentigen Technology, A Miltenyi Biotec Company: Employment. Worden:Lentigen Technology, A Miltenyi Biotec Company: Employment. Hamadani:Sanofi Genzyme: Research Funding, Speakers Bureau; Otsuka: Research Funding; ADC Therapeutics: Consultancy, Research Funding; Takeda: Research Funding; Celgene: Consultancy; Janssen: Consultancy; Pharmacyclics: Consultancy; Merck: Research Funding; Medimmune: Consultancy, Research Funding. Dropulic:Lentigen Technology, A Miltenyi Biotec Company: Employment. Hari:Celgene: Consultancy, Honoraria, Research Funding; Takeda: Consultancy, Honoraria, Research Funding; BMS: Consultancy, Research Funding; Janssen: Consultancy, Honoraria; Kite: Consultancy, Honoraria; Amgen: Research Funding; Spectrum: Consultancy, Research Funding; Sanofi: Honoraria, Research Funding; Cell Vault: Equity Ownership; AbbVie: Consultancy, Honoraria. Johnson:Miltenyi Biotec: Research Funding; Cell Vault: Equity Ownership.

Background: CAR-T cell therapy directed against the CD19 antigen is a breakthrough treatment for patients (pts) with relapsed/refractory (R/R) B-cell NHL. Despite impressive outcomes, not all pts respond and many that respond still relapse. Affordability and accessibility are further considerations that limit current commercial models of CAR-T products. Commercial CAR-T manufacturing is complex, time consuming, and expensive with a supply chain starting at the treating center with apheresis of mononuclear cells, cryopreservation, and shipping to and from a centralized third-party manufacturing site. We addressed these limitations in a Phase 1 clinical trial evaluating a first-in-human bispecific tandem CAR-T cell directed against both CD19 and CD20 (CAR-20.19-T) antigens for pts with R/R B-cell NHL. Through dual targeting we hope to improve response rates and durability of response while limiting antigen escape. We eliminated third party shipping logistics utilizing the CliniMACS Prodigy, a compact tabletop device that allows for automated manufacturing of CAR-T cells within a GMP compliant environment within the hospital. Most materials and reagents used to produce the CAR-T cell product were single-sourced from the device manufacturer. Methods: Phase 1 (NCT03019055), single center, dose escalation + expansion study to demonstrate feasibility and safety of locally manufactured second generation 41BB + CD3z CAR-20.19-T cells via the CliniMACS Prodigy. Feasibility was measured by ability to generate a target CAR-20.19-T cell dose for a minimum of 75% of subjects. Safety was assessed by the presence of dose limiting toxicities (DLTs) through 28 days post-infusion. Dose was escalated in a 3+3 fashion with a starting dose of 2.5 x 10^5 cells/kg, a target DLT rate CAR-T production was set for a 14-day manufacturing process. Day 8 in-process testing was performed to ensure quality and suitability of CAR-T cells for a potential fresh infusion. On Day 10, pts eligible for a fresh CAR-T infusion initiated lymphodepletion (LDP) chemotherapy with fludarabine 30 mg/m2 x 3 days and cyclophosphamide 500 mg/m2 x 1 day, and cells were administered after harvest on Day 14. Pts ineligible for fresh infusion received cryopreserved product and LDP was delayed accordingly. Results: 6 pts have been enrolled and treated with CAR-20.19-T cells: 3 pts at 2.5 x 10^5 cells/kg and 3 pts at 7.5 x 10^5 cells/kg. Median age was 53 years (48-62). Underlying disease was MCL in 3 pts, DLBCL in 2 pts, and CLL in 1 patient. Baseline data and prior treatments are listed in Table 1. CAR-T production was successful in all runs and all pts received their target dose. Three pts received fresh CAR-T cells and 3 pts received CAR-T cells after cryopreservation. To date there are no DLTs to report. No cases of Grade 3/4 cytokine release syndrome (CRS) or neurotoxicity (NTX) were observed. One patient had Grade 2 CRS and Grade 2 NTX requiring intervention. The other had self-limited Grade 1 CRS and Grade 1 NTX. Median time to development of CRS was Day +11 post-infusion. All pts had neutrophil recovery (ANC>0.5 K/µL) by Day 28. Response at Day 28 (Table 2) is as follows: 2/6 pts achieved a complete response (CR), 2/6 achieved a partial response (PR), and 2/6 had progressive disease (PD). One subject with a PR subsequently progressed at Day 90. The 3 pts who did progress all underwent a repeat biopsy, and all retained either CD19 or CD20 positivity. Pts are currently being enrolled at the target dose (2.5 x 10^6 cells/kg) and updated results will be provided at ASH. Conclusions: Dual targeted anti-CD19 and anti-CD20 CAR-T cells were successfully produced for all pts demonstrating the feasibility of a point-of-care manufacturing process via the CliniMACS Prodigy device. With no DLTs or Grade 3-4 CRS or NTX to report, and 2/6 heavily pre-treated pts remaining in CR at 3 and 9 months respectively our approach represents a feasible and promising alternative to existing CAR-T models and costs. Down-regulation of both target antigens was not identified in any patient following CAR-T infusion, and in-process studies suggest that a shorter manufacturing timeline is appropriate for future trials (10 days). Disclosures Shah: Juno Pharmaceuticals: Honoraria; Lentigen Technology: Research Funding; Oncosec: Equity Ownership; Miltenyi: Other: Travel funding, Research Funding; Geron: Equity Ownership; Exelexis: Equity Ownership. Zhu:Lentigen Technology Inc., A Miltenyi Biotec Company: Research Funding. Schneider:Lentigen Technology Inc., A Miltenyi Biotec Company: Employment. Krueger:Lentigen Technology Inc., A Miltenyi Biotec Company: Employment. Worden:Lentigen Technology Inc., A Miltenyi Biotec Company: Employment. Hamadani:Sanofi Genzyme: Research Funding, Speakers Bureau; Merck: Research Funding; Janssen: Consultancy; MedImmune: Consultancy, Research Funding; Cellerant: Consultancy; Celgene Corporation: Consultancy; Takeda: Research Funding; Ostuka: Research Funding; ADC Therapeutics: Research Funding. Johnson:Miltenyi: Research Funding. Dropulic:Lentigen, A Miltenyi Biotec company: Employment. Orentas:Lentigen Technology Inc., A Miltenyi Biotec Company: Other: Prior Employment. Hari:Takeda: Consultancy, Honoraria, Research Funding; Janssen: Honoraria; Kite Pharma: Consultancy, Honoraria; Celgene: Consultancy, Honoraria, Research Funding; Spectrum: Consultancy, Research Funding; Bristol-Myers Squibb: Consultancy, Research Funding; Amgen Inc.: Research Funding; Sanofi: Honoraria, Research Funding.

Adoptive cell therapy with autologous CAR-T cells has induced remarkable responses in patients with treatment-refractory hematologic malignancies, which has led to FDA approvals for two CAR-T products. However, limitations exist with commercial CAR-T centralized production: (1) off-site manufacturing can take several weeks and requires shipping from and to the treating facility; (2) off-site manufacturing limits treatment options for progressing patients; (3) high cost of the commercial products may limit their availability. To address these challenges, we used the fully automated Miltenyi CliniMACS Prodigy device, a GMP-compliant closed system, to manufacture autologous CAR-T cells for a Phase I trial (NCT03019055) evaluating a first-in-human bi-specific CAR that targets CD19 and CD20 (CD20.19 CAR). CAR-T manufacturing was performed exclusively using the CliniMACS Prodigy device and reagents obtained from Miltenyi Biotec. Production was performed within the Medical College of Wisconsin (MCW) Cell Therapy Laboratory, an ISO7 air handling environment. Manufacturing was set at 14 days, and production was as follows. First, peripheral blood mononuclear cells (MNC) were collected by apheresis, with a collection goal of 4 blood volumes to eliminate risk of a low CD3 yield in heavily pre-treated patients. Next, MNC were loaded onto the Prodigy, and CD4 and CD8 T cells enriched by positive immunomagnetic selection. To start the culture process, enriched T cells were suspended in TexMACS medium supplemented with 3% human AB serum and 200 U/mL IL-2, and TransACT reagent was added to stimulate the T cells in the Prodigy cell culture chamber. The following day (day 1), lentiviral vector expressing anti-CD19 and anti-CD20 (in tandem) with CD3ζ and 4-1BB stimulatory domains was added to the stimulated cells. Culture washes and feedings were done on days 5, 6, 8, 10 and 12 of manufacture, and final products harvested on Day 14. Protein L staining was used to detect expression of CD20.19 CAR on the T cells. On Day 14, eligible patients received fresh CAR-T cells, while for others the product was cryopreserved and administered on a later date. To date, the MCW Cell Therapy Laboratory has successfully generated CAR-T cell products for all 6 patients enrolled thus far on the Phase 1 clinical trial with no production failures (Table 1). Three patients received cryopreserved product and 3 patients received fresh product. The enriched T cells were 94.3% CD3+ (87.8-97.4%), and average T cell recovery from the apheresis cell products was 65.2% (54.2-80.0%). Protein L staining indicated 20.8% average CD20.19 CAR expression. Patient CAR-T cells were able to kill CD19+ and CD20+ target cells in vitro and produce IFN-gamma in response to the same target cells. An average yield of 5.9e+8 (4.3-7.9e+8) CAR T cells was obtained at harvest, which exceeded the required cell dose for all patients. The CAR-T cells were comprised of both CD4 and CD8 T cells, with higher expression on CD4 T cells; average CAR-T CD4:CD8 ratio on the final products was 2.8. The majority of T cells (average of 81.5%) had an effector-memory phenotype. In-process testing performed on Day 8 of manufacturing demonstrated sufficient numbers of CAR-T cells needed for patient infusions were already present, and that the CAR-T cells only expanded an additional 1.9 to 3.5-fold between Days 8 and 14. In conclusion, we have successfully demonstrated feasibility for point-of-care CAR-T cell production for clinical use from patient apheresis products utilizing the CliniMACS Prodigy device. Time to production was efficient (14 days), and patient-derived CAR-T cell products were reproducibly generated in a standard cell processing laboratory within an academic medical center. A major clinical advantage of CAR-T cells generated on-site is the flexibility in treatment. Patients can receive cells either immediately (i.e., fresh) or the cells can be cryopreserved for later infusion if the patient is not able to receive fresh cells. Based on our results, we intend to decrease the cell processing time to 10 days. Disclosures Zhu: Lentigen Technology Inc., A Miltenyi Biotec Company: Research Funding. Shah:Juno Pharmaceuticals: Honoraria; Oncosec: Equity Ownership; Geron: Equity Ownership; Exelexis: Equity Ownership; Miltenyi: Other: Travel funding, Research Funding; Lentigen Technology: Research Funding. Schneider:Lentigen Technology Inc., A Miltenyi Biotec Company: Employment. Keever-Taylor:Medical College of Wisconsin: Research Funding. Dropulic:Lentigen, A Miltenyi Biotec company: Employment. Orentas:Lentigen Technology Inc., A Miltenyi Biotec Company: Employment. Hari:Bristol-Myers Squibb: Consultancy, Research Funding; Amgen Inc.: Research Funding; Celgene: Consultancy, Honoraria, Research Funding; Janssen: Honoraria; Kite Pharma: Consultancy, Honoraria; Takeda: Consultancy, Honoraria, Research Funding; Spectrum: Consultancy, Research Funding; Sanofi: Honoraria, Research Funding. Johnson:Miltenyi: Research Funding.

Introduction CD19 CAR-T cell products were recently approved as therapy for advanced B-cell lineage malignancies. The predominant manufacturing model for this type of therapy is a centralized industrial-type production process. An attractive model of CAR-T cell production and delivery to the patient is via a point-of care-manufacturing process. We report on first cases of implementation of this approach in a setting of compassionate use program. Patients and methods Four patients with relapsed/refractory B-cell lineage malignancies were eligible for compassionate use of CD19 CAR-T cell therapy. Three patients, median age 15 years, had relapsed B-cell lineage acute lymphoblastic leukemia (B-ALL) after haploidentical hematopoietic stem cell transplantation (HSCT). Bone marrow disease burden at therapy start was 12%, 74% and 0,159. The patient with low minimal residual disease (MRD) in the bone marrow had skeletal involvement with multiple lymphomatous mass lesions. One patient had refractory primary mediastinal B-cell lymphoma (PMBCL). The CliniMACS Prodigy T cell transduction (TCT) process was used to produce CD19 CAR-T cells from patient-derived leukapheresis product. Automatic production included CD4/CD8 selection, CD3/CD28 stimulation with MACS GMP T Cell TransAct, transduction with lentiviral (second generation CD19.4-1BB zeta) vector (Lentigen, Miltenyi Biotec company) and expansion over 12 days in the presence of TexMACS GMP Medium supplemented with MACS GMP IL-7/IL-15 combination. Final product was administered without cryopreservation to the patients after fludarabine/cyclophosphamide preconditioning. All patients received prophylactic tocilizumab 1 hour before CAR-T cell infusion at 8mg/kg. Results All production cycles were successful. Median transduction efficacy achieved was 57%(52-63). Median expansion of T cells was x26(24-43). CD4/CD8 ratio in the final product was 0,750,22-6). All final products passed the in-process and quality controls. The cell products were administered at a dose of 1*106/kg of CAR-T cells. No immediate infusion reactions were reported. In 2 cases mild cytokine release syndrome (CRS) was diagnosed. In one case mild CAR-T cell related encephalopathy developed. In one case additional dose of tocilizumab and one dose of dexamethasone were administered to control CRS and encephalopathy. All patients survived to the point of response evaluation. In 3 cases CAR-T cell expansion was detected to a maximum 25%. MRD-negative responses detected by flow cytometry and PCR were achieved in 2 cases with bone marrow involvement. In two cases with prominent mass lesions an objective response was documented. In the patient with 74% blast in bone marrow at start of therapy, neither CAR-T cell expansion nor leukemia response were documented. Details of therapy and response are summarized in table 1. Conclusion Production of CAR-T cells with the CliniMACS Prodigy TCT process is a feasible and an attractive option that provides a point-of-care manufacturing approach to enable rapid delivery of CAR-T cells to patients in need. Robustness and consistency of this approach provides a solid basis for multi-center academic trials in the field of adoptive cell therapy. Table 1. Table 1. Disclosures Dropulic: Lentigen, A Miltenyi Biotec company: Employment. Shneider:Lentigen, A Miltenyi Biotec company: Employment. Orentas:Lentigen, A Miltenyi Biotec company: Employment. Alex:Miltenyi Biotec: Employment. Essl:Miltenyi Biotec: Employment.

Introduction Chimeric Antigen Receptor T (CAR-T) cells redirected against tumor antigens are an effective therapy for B cell malignancies refractory to standard treatments. The production of patient-derived CAR-T cells is complicated and thus far is limited to institutions with experienced researchers and expensive GMP facilities, or to those invited to participate in industry sponsored clinical trials. The outsourcing of CAR T-cell production to third party vendors where cells are collected locally, shipped to the manufacturing site, and then sent back to the institution for infusion can be both costly and timely. As a result, CAR-T cell therapies are not widely available and only patients with means to travel to participating sites and with disease that is stable enough to wait the 2-3 months needed to collect and produce CAR-T cells are eligible for these treatments. At our instution we have explored the use of the CliniMACS® Prodigy (Miltenyi Biotec, Inc) for the production of CAR-T cells. The CliniMACS® Prodigy is an automated device that can be used for cell processing within a closed GMP-compliant system. Using the CAR-T system that includes software, specialized tubing sets, and optimized reagents we demonstrate the processing of CAR-T cells, with similar characteristics to those produced in a more traditional manner, in a closed system that is suitable for clinical use without the need for a clean room manufacturing facility. Methods In collaboration with Miltenyi Biotec, we obtained pre-release and final versions of the CliniMACS® Prodigy TCT process software and the TS520 tubing set that allows for cell enrichment, transduction, wash steps, and expansion all within a single set. Starting material was MNC cells recovered from a leukoreduction system chamber (LRSC) used during platelet collections by apheresis. Materials and reagents included MACS CD4 & CD8 reagents for cell enrichment, TransAct CD3/CD28 reagent for activation, lentiviral CD8 TM-41BB-CD3 zeta-cfrag vectors with either CD19 or CD20/CD19 Ab chains (Lentigen Technology Inc., A Miltenyi Biotec Company), TexMACS culture medium-3% HS-IL2, and PBS/EDTA buffer for wash steps. For two experiments, cells after CD4/CD8 enrichment were activated and transduced in 6 well plates and expanded after day 5 in G-Rex gas permerable devices. Total time for line preparation was 14±1 days. Transduction was measured by Protein L expression using flow cytometry. Line function was measured in 51Cr Release assays and by intracellular cytokine production. Results Starting cells were washed free of platelets and enriched for CD4+ and CD8+ cells using the Prodigy device. We achieved consistent high levels purity (99±3%) and good recovery (51.0±6%) of CD4+ and CD8+ cells (N=5). The enriched cells were 90±12% CD3+. The approximately 10% non-T cells were CD8+ NK cells, that were largely eliminated after cell activation through CD3/CD28 and expansion. A controlled number of 1 x 10E8 cells enriched for CD4+ plus CD8+ cells were retained in the Prodigy and in 2 experiments a smaller fraction of cells was cultured in 6 well plates for activation and initial transduction. Three preparations were conducted in the Prodigy, one using the CD19 vector and two with the CD19+CD20 vector. Transduction efficiency ranged from 21%-46% of total T cells with a modest preference for CD4+ cells. Expansion ranged from 26-40 fold and all of the lines recognized CD19 and/or CD20 targets based on 51Cr release assays or IFN-gamma production. The paired lines generated on the Prodigy versus manual methods showed similar overall transduction, phenotype, and function as shown in the figure for one representative preparation. Conclusions CAR-T cells generated in the Prodigy were similar to those prepared using manual methods in both phenotype and function. This process is timely, requiring 14 days for generation of the target CAR-T cell dose, and does not require outsourcing to third party vendors. All of the Prodigy CAR-T cell preparations met criteria for clinical use in our upcoming Phase I clinical trial. The ability to produce CAR-T cells suitable for clinical use in an entirely closed system without the need for a clean room should allow more centers and patients access to this novel form of immunotherapy. Disclosures Shah: Oncosec: Equity Ownership; Exelixis: Equity Ownership; Geron: Equity Ownership. Orentas:Lentigen Technology, Inc.: Employment. Dropulic:Lentigen Technology Inc. A Miltenyi Biotec Company: Employment. Hari:Merck: Research Funding; BMS: Honoraria.

High gene transfer efficiencies have been difficult to achieve in hematopoietic progenitor cells (HPCs) but are important to therapeutic success of HPC gene therapy. Efficient gene transfer is especially challenging with use of column-purified vector for clinical application, as opposed to centrifuged vector commonly used for research. We investigated novel approaches to increase transduction by using a clinically applicable protocol and quantities of column-purified lentiviral vector. Recognizing the association of adenosine 5′-triphosphate (ATP)-binding cassette (ABC) transporters with HPC biology, we investigated the effect of transporter inhibitors on transduction. We found the ABC transporter inhibitor verapamil improved transduction efficiency 2- to 6-fold into CD34+ cells isolated from mobilized peripheral blood, bone marrow, and cord blood. Verapamil also improved transduction in human SCID (severe combined immunodeficient) repopulating cell (SRC) transduction 3- to 4-fold, resulting in 80% to 90% transduction levels in mice receiving primary and secondary transplants without alterations in multilineage reconstitution. Additional ABC transporter substrate inhibitors like quinidine, diltiazem, and ritonavir also enhanced transduction 2- to 3-fold, although ABC transporter inhibitors that are not substrates did not. Enhanced transduction was not observed in mature hematopoietic cells, neurospheres, mesenchymal stem cells, or hepatocytes. Enhancement of transduction in HPCs was observed with vesicular stomatitis virus-G (VSV-G)-pseudotyped lentiviral vector but not with vector pseudotyped with RD114. Thus, we present a new approach for efficient delivery to primitive HPCs by VSV-G-pseudotyped lentiviral vectors. (Blood. 2004;104:364-373)

INTRODUCTION: To protect normal bone marrow from chemotherapy in glioblastoma patients, we have developed a novel strategy by introducing a strong DNA repair protein, mutant (P140K) of human methylguanine methyltransferase (MGMT), into patients’ CD34+ hematopoietic progenitors (HPC) by lentiviral gene transfer leading to selective expansion of drug-resistant P140K-MGMT CD34+ cells and their myeloid and immune cell progeny. METHODS: To achieve long-term stable expression of the P140K-MGMT gene, we used a lentiviral vector which integrates into the host genome. However, viral insertion mutagenesis has raised safety concerns; as the previous γ-retroviral vectors were associated with insertion mutations leading to development of acute lymphoblastic leukemia in 20% of treated patients. Nevertheless, new improved lentiviral vector with safe feature of insertion site far away from gene transcription start site has been developed. Here we evaluated the safety of a lentivirus vector under selection pressure of chemotherapy. HYPOTHESIS: Our lentiviral vector is safer than traditional γ-retroviral vectors as evident by lack of early clonal dominance even with a chemo-selection. RESULTS: Three glioblastoma patients were recruited and underwent resection, after which CD34+ HPC were mobilized with filgrastim, isolated by magnetic bead separation (Miltneyi CliniMACS), and transduced ex vivo with a GMP-grade lentiviral P140K-MGMT vector (Lentigen Corp). Subsequently, patients received radiation/temozolomide for 6 weeks and up to five cycles of monthly O6-benzylguanine/temozolomide (BG/TMZ) treatment. As a result, all three patients demonstrated a 5-15 fold selective expansion of P140K-MGMT positive HPC and their progeny granulocyte and mononuclear cells in peripheral blood and a small number of CFUs from bone marrow indicating a drug-selection mechanism. The viral insertion sites in the cells of these three patients were closely monitored in each chemotherapy cycle and the patients were followed for up to 1 year after the last therapy. We mapped a total of 17,308 viral insertion sites (VIS), for patient 1(6,146), patient 2(2,081) and patient 3(9,081) and the unique viral insertion sites (UVIS) was 382, i.e. 135, 76 and 171 for patient 1, patient 2 and patient 3 respectively. Overall, during the drug-treatment period, there were no persistent UVIS. Moreover, at the conclusion of BG/TMZ treatment, the VIS number sharply diminished. CONCLUSION: Gene transfer of LV MGMTP140K vector into hematopoietic progenitor cells did not lead to clonal dominance during or after drug selection. Dose escalation of BG/TMZ will define hematopoietic tolerance and treatment response. Disclosures Embree: Lentigen Technology Inc: Employment. Dropulic:Lentigen Technology Inc: Employment, Patents & Royalties.

Abstract 4177 Temozolomide (TMZ) is a standard of care treatment for patients with glioma post surgical resection, but induces dose limiting myelosuppresion. Gene transfer of O6-benzylguanine (BG) resistant mutant O6-methylguanine-DNA-methyltransferase (MGMT P140K) into hematopoietic stem cells (HSCs) protects hematopoiesis from alkylating agents and allows efficient in vivo selection of transduced HSCs in mice, dogs and primates. We designed a clinical trial to improve HSC tolerance of TMZ in patients with glioma [clinicaltrials.gov]. This trial uses a lentiviral backbone for gene delivery, and infuses cells during the active phase of treatment to maximize the potential for stem cell protection and selection. The viral construct was synthesized by Lentigen, with an EF1-alpha promoter driving MGMTP140K (LV-P140K MGMT). Clinical grade vector was subject to safety and toxicity analysis at the request of both NCI and FDA. Transduction conditions were optimized towards a high rate of gene transfer and expression with ≤2 copies per cell to minimize adverse insertions. This study presents the validation of the vector by assessing the engraftment, drug resistance and copy number after gene transfer into human CD34+ cells followed by in vivo selection using a nonobese diabetic (NOD)/severe combined immunodeficient (SCID) gammaC (NSG) mice. We are currently analyzing the insertion sites of the clinical grade LV-P140K MGMT in human cord blood progenitors as a safety monitoring tool for LV-HSC gene therapy. Engraftment of LV-P140K MGMT (LG631) transduced CD34+ cells in irradiated NSG mice, and the analysis of engrafted human CD45+ cells in peripheral blood, spleen and bone marrow showed consistent but slightly decreased (10%) engraftment of transduced cells compared to the freshly isolated non transduced cells and was multilineage. There was no toxicity to the mice or human HSC as a result of transduction with lentiviral vector. The NSG recipients of the LG631 transduced CD34+ cells received two rounds of 3-consecutive day treatments of 30 mg/kg of BG and 60 mg/kg of TMZ, and three weeks after the second round of treatment, expression of MGMT was examined by PCR in the CFUs from the BM of the recipient mice. Combined treatment with BG and TMZ showed efficient selection of human cells and human CFU expressing MGMT (100% after treatment vs 44+/−4.1% without treatment). Copy number and characterization of insertion sites of the clinical grade LV-P140K MGMT in human CD34 cells serve as a critical safety monitoring tool for LV-HSC gene therapy. Insertion site analysis showed a polyclonal pattern and selection did not result in oligoclonality. The insertion database will continue to be made publically available as the data is generated prior to and during clinical trial application. We have fully characterized the preclinical safety and efficacy of a clinical grade lentiviral vector for clinical trial application in patients with glioma. Disclosures: No relevant conflicts of interest to declare.

The increased cost of goods associated with low-level expression of factor VIII (fVIII) is considered a significant factor in the pricing of recombinant fVIII products. Worldwide treatment of hemophilia A is limited due to the cost of fVIII products, which easily can exceed $100,000 per year. Using state-of-the-art commercial production facilities and technology, recombinant human (h)-fVIII expresses at 100 – 1000-fold lower levels than other plasma proteins. We have developed a novel recombinant fVIII product that overcomes the primary limitation to the manufacture of currently available hemophilia A pharmaceuticals. There are two components to this product: a humanized highexpression hybrid-human/porcine (HP) fVIII transgene, designated ET-801i, and the LENTIMAX™ lentiviral production system. The ET-801i transgene contains elements of porcine fVIII (p-fVIII) sequence in the A1 and A3 domains that facilitate more efficient secretion from production cell lines and is 90% identical in amino acid sequence to h-fVIII. Additionally, lentiviral introduction of the transgene allows achievement of an optimal number of transgene copies and the lentiviral vectors typically integrate into sites of active chromatin, thus facilitating high-levels of transgene transcription. In the current study, we demonstrated that high-expression porcine sequence elements enable 20 – 100- fold superior expression over any previously described fVIII transgene variant. Using a stable baby hamster kidney cell-derived expression system, ET-801i was expressed at a level indistinguishable from p-fVIII, which both were significantly greater than h-fVIII (

We established the first clinical ex vivo HIV-based vector gene therapy trial in humans with HIV+ CD4+ T-cells. Briefly, this therapy involves modifying patient CD4+ T-cells with our modified lentiviral vector carrying an anti-HIV payload. These cells are then activated and expanded, and re-infused back into the patient. However, cGMP regulations require the use of costly clinical grade reagents (i.e. Retronectin™, CD3/CD28 stimulating paramagnetic beads). In an attempt to reduce ex-vivo processing costs, but not at the expense of transduction levels, we sought to determine a way to directly activate CD4+ T-cells with modified lentiviral vectors. 293FT HEK cell lines, used for producing our lentiviral vectors, were modified to co-express the natural CD28 stimulatory ligand B7.2 (CD86) and ICAM-1 (CD54) proteins on their membrane for co-stimulation and anchoring purposes. When CliniMACS purified normal donor CD4+ T cells were co-cultured with CD54/CD86-expressing cells, in the presence of soluble OKT3 CD3 antibody, CD25 and CD69 activation markers were upregulated, indicating that functional proteins were being expressed at the cell membrane. These CD54 and/or CD86 expressing cells could subsequently be transfected with lentiviral vector plasmid constructs in order to produce host-derived CD54 and/or CD86 bearing HIV-based vectors. EGFP-expressing lentiviral vectors, VRX494, with CD54/CD86-modified envelopes were produced both in these cell lines and by transient transfection of all relevant plasmids, and titers were assayed on Hela-Tat cells by FACS. CD54 modified lentiviral vectors showed increased binding to CD4+ T-cells, as evidenced by significant cell clumping. CD86 (as well as CD54 plus CD86) modified lentiviral vector, with soluble OKT3 CD3 antibody, was shown to activate T-cells, above the levels seen with unmodified lentiviral vectors, as evidenced by the increase in cell surface CD25 and CD69 expression and also the increase in cell size. Cellular expansion of modified lentiviral vector transduced CD4+ T cells reached levels close to CD3:CD28 bead stimulated CD4+ T cell controls over a period of 2 to 3 weeks. The CD3/TCR repertoire was assessed by flow cytometry and, compared to the well-established CD3/CD28 coated M450 Dynabeads stimulatory system as a control, no skewing of the repertoire was observed. CD86 was shown to improve levels of transduction in pre-activated lymphocytes with CD3/CD28 coated M450 Dynabeads. However, CD86 co-expression was crucial for transducing minimally activated CD4+ T cells with only soluble OKT3 CD3 antibody. Levels of transduction and activation were on average 2 to 3 times higher with the modified lentiviral vectors. To our knowledge, we are reporting the first generation of lentiviral particles exhibiting an adhesion property with stimulatory abilities. The development of such a lentiviral vector has valuable implications for clinical application by reducing the number of exogenous reagents in large scale cell processing.

Lentiviral vectors (LV) offer improved therapeutic benefit of gene therapy applications as a result of their ability to stably transduce a wide range of cycling and non-cycling cells. The first Phase I clinical trial using a lentiviral vector, VRX496, was initiated in January of 2003. The vector evaluated is an HIV-1 based lentiviral vector, fully gutted but retaining the 5′ and 3′ long terminal repeats, the packaging sequence, cPPT/CTS, splice donor and splice acceptor, and rev response element (RRE). VRX496 contains a splice-independent 937-base antisense sequence targeting the envelope gene in the HIV genome, and a small 186-base tag derived from the GFP gene to serve as a molecular marker for vector in HIV infected patient cells. Five (5) HIV-infected patients with CD4 counts between 200 and 500, and viral loads above 5000 copies/ml, no history of opportunistic infections, who have failed two regiments of highly active antiretroviral therapy (HAART), were serially enrolled in the trial. After leukopheresis, patient’s CD4 T lymphocytes were isolated by negative selection of CD8+ cells, and then cultured for three days in the presence of immobilized CD3/28 beads, and VRX496. Thereafter, the vector and beads were removed, cells were expanded to the dose of 1 x 1010, and then given intravenously over 15 minutes. At the time of this abstract, 4 patients have been dosed and have passed the initial 21-day safety assessment; a fifth patient is scheduled for dosing. No serious adverse events have occurred. All patients have steady CD4 counts, and viral loads have decreased below baseline in all patients, notably from 218,000 pre dose to 36,000 1 year post dosing in the first patient, and from 22,000 pre dose to 1,625 at 9 months post dosing in the second patient; patients 3 and 4 are at 3 months and 21 days post-infusion, respectively. Importantly, persistence of vector-modified CD4 cells are detected in the peripheral blood out to 9 months post dosing. This trial establishes for the first time the safety of lentiviral vectors for clinical gene therapy application.

Recently, we initiated the first ex vivo HIV-based gene therapy trial in humans with HIV+ CD4+ T cells. In this protocol, a modified lentiviral vector carrying an anti-HIV payload is used to modify CD4+ T cells isolated from HIV-infected patients by apheresis and CD8 negative selection. The T cells are activated in the presence of vector and expanded using immobilized CD3/CD28 antibodies, and then infused back into the patient. T cell receptor (TCR) repertoire analysis has value for safety monitoring of adoptive T cell transfers in the detection of aberrant clonal expansions or deletions. In this study, the TCR Vβ repertoire was assessed using a flow cytometry based assay at various time points in the selection/transduction/expansion process of CD4+ T cells. PBMC isolated from whole blood of HIV+ patients were CD4-selected using a CD8 negative selection, followed by enrichment by CD3 antibody. CD4+ purified cells were transduced with the lentiviral vector, VRX496, in the presence of retronectin, and then co-cultured with CD3/CD28 coated M450 Dynabeads for ten days. The TCR Vβ repertoire was assessed in throughout the process using a FACS-based assay that employs a panel of 20 monoclonal antibodies recognizing most of the 24 Vβ families in PBMC and CD4+ T cells. Repertoires from subjects with normal polyclonal TCR profiles were conserved, as shown by the absence of any significant change in any Vβ family. Moreover, the transduction/expansion of CD4+ T cells from a patient with a previously skewed TCR profile allowed the improvement of the TCR Vβ repertoire. Finally, no significant difference was observed in the repertoire of cells transduced with VRX496 versus mock-transduced cells. These data demonstrate stability of the repertoire diversity and thus provide important support information in favor of the safety of a gene therapy approach involving lentiviral vector mediated modification and expansion of CD4+ T-cells.