CAR-T cell manufacturing landscape-Lessons from the past decade and considerations for early clinical development.

CAR-T cell therapies have consolidated their position over the last decade as an effective alternative to conventional chemotherapies for the treatment of a number of hematological malignancies. With an exponential increase in the number of commercial therapies and hundreds of phase 1 trials exploring CAR-T cell efficacy in different settings (including autoimmunity and solid tumors), demand for manufacturing capabilities in recent years has considerably increased. In this review, we explore the current landscape of CAR-T cell manufacturing and discuss some of the challenges limiting production capacity worldwide. We describe the latest technical developments in GMP production platform design to facilitate the delivery of a range of increasingly complex CAR-T cell products, and the challenges associated with translation of new scientific developments into clinical products for patients. We explore all aspects of the manufacturing process, namely early development, manufacturing technology, quality control, and the requirements for industrial scaling. Finally, we discuss the challenges faced as a small academic team, responsible for the delivery of a high number of innovative products to patients. We describe our experience in the setup of an effective bench-to-clinic pipeline, with a streamlined workflow, for implementation of a diverse portfolio of phase 1 trials.

AKT inhibition generates potent polyfunctional clinical grade AUTO1 CAR T-cells, enhancing function and survival.

BACKGROUND: AUTO1 is a fast off-rate CD19-targeting chimeric antigen receptor (CAR), which has been successfully tested in adult lymphoblastic leukemia. Tscm/Tcm-enriched CAR-T populations confer the best expansion and persistence, but Tscm/Tcm numbers are poor in heavily pretreated adult patients. To improve this, we evaluate the use of AKT inhibitor (VIII) with the aim of uncoupling T-cell expansion from differentiation, to enrich Tscm/Tcm subsets. METHODS: VIII was incorporated into the AUTO1 manufacturing process based on the semiautomated the CliniMACS Prodigy platform at both small and cGMP scale. RESULTS: AUTO1 manufactured with VIII showed Tscm/Tcm enrichment, improved expansion and cytotoxicity in vitro and superior antitumor activity in vivo. Further, VIII induced AUTO1 Th1/Th17 skewing, increased polyfunctionality, and conferred a unique metabolic profile and a novel signature for autophagy to support enhanced expansion and cytotoxicity. We show that VIII-cultured AUTO1 products from B-ALL patients on the ALLCAR19 study possess superior phenotype, metabolism, and function than parallel control products and that VIII-based manufacture is scalable to cGMP. CONCLUSION: Ultimately, AUTO1 generated with VIII may begin to overcome the product specific factors contributing to CD19+relapse.

A compact and simple method of achieving differential transgene expression by exploiting translational readthrough.

The development of multicistronic vectors enabling differential transgene expression is a goal of gene therapy and poses a significant engineering challenge. Current approaches rely on the insertion of long regulatory sequences that occupy valuable space in vectors, which have a finite and limited packaging capacity. Here we describe a simple method of achieving differential transgene expression by inserting stop codons and translational readthrough motifs (TRMs) to suppress stop codon termination. TRMs reduced downstream transgene expression ∼sixfold to ∼140-fold, depending on the combination of stop codon and TRM used. We show that a TRM can facilitate the controlled secretion of the highly potent cytokine IL-12 at therapeutically beneficial levels in an aggressive immunocompetent mouse melanoma model to prevent tumor growth. Given their compact size (6 bp) and ease of introduction, we envisage that TRMs will be widely adopted in recombinant DNA engineering to facilitate differential transgene expression.

Tunable control of CAR T cell activity through tetracycline mediated disruption of protein-protein interaction.

Chimeric antigen receptor (CAR) T cells are a promising form of cancer immunotherapy, although they are often associated with severe toxicities. Here, we present a split-CAR design incorporating separate antigen recognition and intracellular signaling domains. These exploit the binding between the tetracycline repressor protein and a small peptide sequence (TIP) to spontaneously assemble as a functional CAR. Addition of the FDA-approved, small molecule antibiotic minocycline, acts as an "off-switch" by displacing the signaling domain and down-tuning CAR T activity. Here we describe the optimization of this split-CAR approach to generate a CAR in which cytotoxicity, cytokine secretion and proliferation can be inhibited in a dose-dependent and reversible manner. Inhibition is effective during on-going CAR T cell activation and inhibits activation and tumor control in vivo. This work shows how optimization of split-CAR structure affects function and adds a novel design allowing easy CAR inhibition through an FDA-approved small molecule.

Potential of Magnetic Hyperthermia to Stimulate Localized Immune Activation.

Magnetic hyperthermia (MH) harnesses the heat-releasing properties of superparamagnetic iron oxide nanoparticles (SPIONs) and has potential to stimulate immune activation in the tumor microenvironment whilst sparing surrounding normal tissues. To assess feasibility of localized MH in vivo, SPIONs are injected intratumorally and their fate tracked by Zirconium-89-positron emission tomography, histological analysis, and electron microscopy. Experiments show that an average of 49% (21-87%, n = 9) of SPIONs are retained within the tumor or immediately surrounding tissue. In situ heating is subsequently generated by exposure to an externally applied alternating magnetic field and monitored by thermal imaging. Tissue response to hyperthermia, measured by immunohistochemical image analysis, reveals specific and localized heat-shock protein expression following treatment. Tumor growth inhibition is also observed. To evaluate the potential effects of MH on the immune landscape, flow cytometry is used to characterize immune cells from excised tumors and draining lymph nodes. Results show an influx of activated cytotoxic T cells, alongside an increase in proliferating regulatory T cells, following treatment. Complementary changes are found in draining lymph nodes. In conclusion, results indicate that biologically reactive MH is achievable in vivo and can generate localized changes consistent with an anti-tumor immune response.

Intratumoral IL-12 delivery empowers CAR-T cell immunotherapy in a pre-clinical model of glioblastoma.

Glioblastoma multiforme (GBM) is the most common and aggressive form of primary brain cancer, for which effective therapies are urgently needed. Chimeric antigen receptor (CAR)-based immunotherapy represents a promising therapeutic approach, but it is often impeded by highly immunosuppressive tumor microenvironments (TME). Here, in an immunocompetent, orthotopic GBM mouse model, we show that CAR-T cells targeting tumor-specific epidermal growth factor receptor variant III (EGFRvIII) alone fail to control fully established tumors but, when combined with a single, locally delivered dose of IL-12, achieve durable anti-tumor responses. IL-12 not only boosts cytotoxicity of CAR-T cells, but also reshapes the TME, driving increased infiltration of proinflammatory CD4+ T cells, decreased numbers of regulatory T cells (Treg), and activation of the myeloid compartment. Importantly, the immunotherapy-enabling benefits of IL-12 are achieved with minimal systemic effects. Our findings thus show that local delivery of IL-12 may be an effective adjuvant for CAR-T cell therapy for GBM.

Noninvasive diffusion magnetic resonance imaging of brain tumour cell size for the early detection of therapeutic response.

Cancer cells differ in size from those of their host tissue and are known to change in size during the processes of cell death. A noninvasive method for monitoring cell size would be highly advantageous as a potential biomarker of malignancy and early therapeutic response. This need is particularly acute in brain tumours where biopsy is a highly invasive procedure. Here, diffusion MRI data were acquired in a GL261 glioma mouse model before and during treatment with Temozolomide. The biophysical model VERDICT (Vascular Extracellular and Restricted Diffusion for Cytometry in Tumours) was applied to the MRI data to quantify multi-compartmental parameters connected to the underlying tissue microstructure, which could potentially be useful clinical biomarkers. These parameters were compared to ADC and kurtosis diffusion models, and, measures from histology and optical projection tomography. MRI data was also acquired in patients to assess the feasibility of applying VERDICT in a range of different glioma subtypes. In the GL261 gliomas, cellular changes were detected according to the VERDICT model in advance of gross tumour volume changes as well as ADC and kurtosis models. VERDICT parameters in glioblastoma patients were most consistent with the GL261 mouse model, whilst displaying additional regions of localised tissue heterogeneity. The present VERDICT model was less appropriate for modelling more diffuse astrocytomas and oligodendrogliomas, but could be tuned to improve the representation of these tumour types. Biophysical modelling of the diffusion MRI signal permits monitoring of brain tumours without invasive intervention. VERDICT responds to microstructural changes induced by chemotherapy, is feasible within clinical scan times and could provide useful biomarkers of treatment response.

Hampering brain tumor proliferation and migration using peptide nanofiber:siPLK1/MMP2 complexes.

Aim: To develop a nonviral tool for the delivery of siRNA to brain tumor cells using peptide nanofibers (PNFs). Materials & methods: Uptake of PNFs was evaluated by confocal microscopy and flow cytometry. Gene silencing was determined by RT-qPCR and cell invasion assay. Results: PNFs enter phagocytic (BV-2) and nonphagocytic (U-87 MG) cells via endocytosis and passive translocation. siPLK1 delivered using PNFs reduced the expression of polo-like kinase 1 mRNA and induced cell death in a panel of immortalized and glioblastoma-derived stem cells. Moreover, targeting MMP2 using PNF:siMMP2 reduced the invasion capacity of U-87 MG cells. We show that stereotactic intra-tumoral administration of PNF:siPLK1 significantly extends the survival of tumor bearing mice comparing with the untreated tumor bearing animals. Conclusion: Our results suggest that this nanomedicine-based RNA interference approach deserves further investigation as a potential brain tumor therapeutic tool.

Near-infrared dual bioluminescence imaging in mouse models of cancer using infraluciferin.

Bioluminescence imaging (BLI) is ubiquitous in scientific research for the sensitive tracking of biological processes in small animal models. However, due to the attenuation of visible light by tissue, and the limited set of near-infrared bioluminescent enzymes, BLI is largely restricted to monitoring single processes in vivo. Here we show, that by combining stabilised colour mutants of firefly luciferase (FLuc) with the luciferin (LH2) analogue infraluciferin (iLH2), near-infrared dual BLI can be achieved in vivo. The X-ray crystal structure of FLuc with a high-energy intermediate analogue, 5'-O-[N-(dehydroinfraluciferyl)sulfamoyl] adenosine (iDLSA) provides insight into the FLuc-iLH2 reaction leading to near-infrared light emission. The spectral characterisation and unmixing validation studies reported here established that iLH2 is superior to LH2 for the spectral unmixing of bioluminescent signals in vivo; which led to this novel near-infrared dual BLI system being applied to monitor both tumour burden and CAR T cell therapy within a systemically induced mouse tumour model.

Computational fluid dynamics with imaging of cleared tissue and of in vivo perfusion predicts drug uptake and treatment responses in tumours.

Understanding the uptake of a drug by diseased tissue, and the drug's subsequent spatiotemporal distribution, are central factors in the development of effective targeted therapies. However, the interaction between the pathophysiology of diseased tissue and individual therapeutic agents can be complex, and can vary across tissue types and across subjects. Here, we show that the combination of mathematical modelling, high-resolution optical imaging of intact and optically cleared tumour tissue from animal models, and in vivo imaging of vascular perfusion predicts the heterogeneous uptake, by large tissue samples, of specific therapeutic agents, as well as their spatiotemporal distribution. In particular, by using murine models of colorectal cancer and glioma, we report and validate predictions of steady-state blood flow and intravascular and interstitial fluid pressure in tumours, of the spatially heterogeneous uptake of chelated gadolinium by tumours, and of the effect of a vascular disrupting agent on tumour vasculature.