Search

Data from a recent clinical trial at the University of Pennsylvania (UPenn) demonstrate the safety of NYCE T cells, which are autologous, patient-derived T cells transduced with lentivirus to express NY-ESO-1 T cell receptor and multiplex-edited with CRISPR/Cas9 at three distinct genes. This pilot study was the first in the United States (US) to administer CRISPR-edited cells to a patient. The preclinical approach and data in support of this trial was reviewed by the National Institutes of Health Recombinant DNA Advisory Committee (NIH RAC) in 2016, followed by the acceptance of an Investigational New Drug (IND) application by the US Food and Drug Administration. The aim of this abstract is to introduce the experience of an academic institution in the development of a preclinical program to support submission of an IND application for a cell therapy product with CRISPR-mediated gene editing. Non-clinical assays, data, and FDA feedback on IND submissions for CRISPR-edited gene products at the Center for Cellular Immunotherapies at UPenn will be presented. The preclinical program for the US IND application for UPenn's NYCE T cell therapy product paired conventional T cell phenotyping assays and novel genome editing evaluations. A comprehensive IND application for NYCE T cells was packaged out of inter-disciplinary collaborations to evaluate safety and potency of CRISPR-edited, antigen-directed T cells as well as fulfill NIH RAC and FDA requests. Results of these analyses will be presented. Execution of the NYCE T cell IND led to preclinical strategies toward applying CRISPR-editing to improve upon other T cell therapies, such as chimeric antigen receptor (CAR) T cells. As such, UPenn is developing an "off-the-shelf" CAR T cell product with healthy donor T cells utilizing CRISPR-mediated knock-down of genes encoding endogenous allo-reactive proteins. The parallels between the NYCE and the off-the-shelf CAR T cell non-clinical development pathways will be discussed. [ABSTRACT FROM AUTHOR]

The most common side effect of CD19-directed CAR-T cell therapy is CRS. IL-6 blockade by toci typically results in rapid resolution of severe CRS in CTL019 treated pts; however, the ideal timing of toci remains unclear. Pts with high tumor burden (TB; defined as ≥40% marrow blasts) are at high risk for severe CRS. We hypothesized that preemptive toci administration early in the course of CRS in pts with high TB might decrease acute CRS severity without compromising efficacy. We conducted a two-cohort, open-label study of the timing of toci on CTL019-associated CRS. Pts were 1-24 yo with CD19-expressing relapsed or refractory B-cell acute lymphoblastic leukemia (r/r B-ALL). CTL019 was manufactured at the UPenn Cell and Vaccine Production Facility. Enrolled pts were assigned to high (≥ 40%) or low (<40%) TB cohorts (HTBC/LTBC) based on a BMBx performed after lymphodepleting chemo, 1-5 days before CTL019. Low TB pts received standard CRS management. High TB pts received a single dose of toci (8-12 mg/kg) after developing a fever to >38.5 C measured twice in 24 hours ("early toci") and subsequently treated per standard CRS algorithm. Primary endpoint: frequency of gr4 CRS (Penn scale) in ≤5/15 pts. Secondary endpoints: D28 tumor response, remission duration, additional toxicities. 78 pts were screened, all were enrolled. 65 pts were evaluable for analysis at data cutoff. 15 pts were assigned to HTBC, 50 pts to LTBC and all were infused with CTL019 (median dose 5.0 × 106 [range 5.6 × 105-3.9 × 107] CAR+ T cells/kg). The overall rate of gr4 CRS was 9% (27% in the HTBC, 4% in the LTBC, Table 1). The overall rate of gr1-4 CRS was 72% (100% in the HTBC, 64% in the LTBC). One pt in the HTBC died from CNS hemorrhage after gr4 CRS. All patients in the HTBC received the early toci intervention. The overall rate of gr3 and 4 CRS was lower than in two prior studies of CTL019 therapy in pediatric r/r B-ALL (18% vs 42% in CHP959 and 48% in ELIANA). The ORR at day 28 follow-up was 97% (87% HTBC, 100% LTBC, with 71% CR and 26% CRi, Table 2). We defined clincially meaningful reduction in severe CRS if ≤5/15 subjects had gr4 CRS in the HTBC; this a priori defined primary end-point was met. Early administration of toci targeted to pts with high TB may reduce the incidence of severe CRS while maintaining excellent efficacy of CTL019 therapy. TB remains a major predictor of toxicity. Longer follow-up is ongoing to assess impact on duration of remission. [ABSTRACT FROM AUTHOR]

NY-ESO-1 is a cancer testis antigen with ectopic expression on multiple myeloma (MM), melanoma and sarcoma but restricted expression on normal tissues, rendering it a good candidate target for cancer immunotherapy. We recently reported a phase 1 pilot trial of multiplex CRISPR/Cas9 genome-edited NY-ESO-1 TCR cells (NYCE T cells; NCT03399448) for patients (pts) with advanced MM, synovial sarcoma and myxoid/round cell liposarcoma (MRCL) demonstrated safety and feasibility as well as persistence of T cells. Here, we characterize the phenotype and function of the ex vivo manufactured NYCE T cells from the 3 pts that were infused in the trial. The 3 infusion products (2 MM – pt #35 and Pt #7, and 1 MRCL – pt #39) were used to compare expanded T cells with either CRISPR/Cas9 editing without TCR transduction or with Mock-editing and TCR transduction from the same pt. We examined the phenotypic markers of T cell differentiation and checkpoint inhibitors by multicolor flow cytometry and assessed in vitro anti-tumor efficacy via a luciferase-based killing assay. We found heterogeneity in the composition of T cell subsets from all 3 infusion products with various CD4 and CD8 ratios, ranging from 0.12-1.1. Transduction efficiencies (determined by Vbeta8.1+ staining) varied between 1-11% and different levels of knockout efficiency in TRAC, TRBC and PDCD1 genes were observed by digital PCR. Three NYCE T cell infusion products displayed a terminally differentiated T cell phenotype similar to CRISPR/Cas9 edited counterpart. In contrast to Mock-edited T cells from the same pt, NYCE T cells exhibited lower levels of differing combinations of exhaustion-associated markers (PD1, TIM3, LAG3, EOMES and CTLA4) in both CD4+ and CD8+ T cell compartments. All 3 infusion products elicited antigen-specific killing of NY-ESO-1 expressing target cells. We demonstrate that NYCE T cells are heterogeneous in composition. Furthermore, we showed that CRISPR/Cas9 editing does not affect T cell differentiation and anti-tumor efficacy, but does reduce levels of exhaustion-associated markers expressed by the final products. This work provides an in-depth characterization of the heterogeneous product composition and function of the first multiplex CRISPR/Cas9-edited T cells tested in humans. [ABSTRACT FROM AUTHOR]

CTL019 (tisagenlecleucel) is an anti-CD19 genetically modified autologous T-cell therapy approved for treatment of relapsed/refractory pediatric and young adult B-cell acute lymphoblastic leukemia (ALL) and adult diffuse large B-cell lymphoma (DLBCL). For CTL019 manufactured at Penn, the product release criteria include a viability assessment. In clinical trials, viability specification for product release was ≥ 70% viable cells; whereas, for commercial tisagenlecleucel, viability specification for product release was ≥ 80%. We examined the relationship between CTL019 viability release testing and clinical outcomes. We analyzed CTL019 viability release testing in patients (pts) treated on prospective single institution clinical trials for non-Hodgkin lymphomas (NHL) (NCT02030834) and pediatric ALL clinical trials (NCT01626495, NCT02906371). The relationship of CTL019 product viability release testing and overall response, progression-free survival, overall survival, sensitivity/specificity of viability testing to predict complete response, and CAR T-cell expansion were examined. For 39 NHL pts, the median product viability was 89.8% (range: 73.7%-97.7%) and 3 pts had products with < 80% viability. For 123 ALL pts, median product viability was 89.3% (range: 56.0%-98.4%); 15 pts had products with < 80% viability. For NHL pts, there was no difference in % viability for pts who had CR vs those with PR/PD (p=0.7). For ALL pts, there was no difference in viability for pts who had CR/CRi vs those with PD/NR (p=0.9). Receiver operating characteristic (ROC) area under the curve (AUC) for NHL viability was 0.47 (95%CI: 0.28-0.65) and ROC AUC for ALL viability was 0.52 (95%CI: 0.32-0.71). PFS and OS were not significantly influenced by viability release test results for NHL (PFS HR 1.0; OS not yet available) or ALL (PFS HR 1.04; OS HR 1.03). There was no difference in PFS or OS between the lowest quintile of viability and other quintiles for NHL (PFS p=0.2, OS p=0.1) or ALL (PFS p=0.1, OS p=0.1). Peak CAR T-cell expansion was not significantly correlated with viability release test results in NHL (rho=0.12, p=0.5 or ALL (rho=-0.1, p=0.45). There was no difference in peak CAR T-cell expansion between the lowest quintile of viabilities vs the other quintiles in NHL (p=0.5) or ALL (p=0.4). These results suggest that there is no dose-response relationship between CTL019 product viability (70-98%) and clinical outcomes. [ABSTRACT FROM AUTHOR]

With the goal of improving consistency and success of manufacturing (MFG), we analyzed the composition of apheresis products collected from patients (pts) enrolled in clinical trials of engineered T cell immunotherapies. We previously reported in a small cohort of NHL pts, that T cell cultures manufactured from fresh APH of challenging products comprised of 3:1 monocyte to lymphocytes, exhibited impaired proliferation. Microscopic observation of poorly expanding cell cultures revealed the phagocytosis of T cell stimulating anti-CD3/ anti-CD28 Dynabeads by large cells, possibly of monocyte/myeloid origin. Proliferative capacity of the T cells was rescued if the APH was cryopreserved/thawed prior to manufacture. Substantial reduction in MDSC frequency, following cryopreservation/thaw, was demonstrated by flow cytometry. Alternatively, lymphocyte enrichment via elutriation of APH removed the larger myeloid cells. Here we expand analysis of the critical quality attributes of APH products to a larger cohort of patients with hematological malignancies (HM) as well as solid tumors (ST). We analyzed 300 APH from pts on 20 clinical trials focused on HM or ST. Frequencies of lymphocytes and myeloid populations were determined via size/volume distribution on Multisizer, lineage and MDSC immunophenotyping by FACS (CD3, CD45, CD11b, CD33, HLA-DR, CD14, CD15). Population doubling levels (PDL) during ex vivo expansion were calculated at day 9 of manufacture. We report an overall 94% MFG success rate with 5 PDL. 12/19 MFG failures did not produce an infusible dose with -1.34 PDL and occurred in HM APH. APH of MFG fails showed the presence of myeloid cells that were not removed prior to MFG. APH from NHL, MM, and synovial sarcoma demonstrated MDSCs by flow cytometry. Cryopreservation/thaw and/or elutriation of APH enriched T cells from 81 % to 94%, reduced myeloid population from 38% to 11%, and removed MDSCs. Small scale test expansions comparing various MFG processes on the same APH confirmed that cryopreservation and or elutriation increases clinical MFG success. Myeloid subsets in HM APH increase the risk of MFG failure. Mitigation strategies to remove the myeloid cells to enrich for the lymphocytes can be employed, including cryopreservation or elutriation of fresh APH prior to MFG. To increase MFG feasibility, studies are continuing to investigate the effects of prior treatment on APH and the mechanism of inhibition by MDSCs. [ABSTRACT FROM AUTHOR]