Autologous Cell

Abstract

Autologous cell therapies are based on bespoke, patient-specific manufacturing lines and distribution channels. They present a novel category of therapies with unique features that impose scale out approaches. Chimeric Antigen Receptor (CAR) T cells are an example of such products, the manufacturing of which is based on the patient’s own cells. This automatically: (a) creates dependencies between the patient and the supply chain schedules and (b) increases the associated costs, as manufacturing lines and distribution nodes are exclusive to the production and delivery of a single therapy. The lack of scale up opportunities and the tight return times required, dictate the design of agile and responsive distribution networks that are eco-efficient. From a modelling perspective, such networks are described by a large number of variables and equations, rendering the problem intractable. In this work, we present a bi-level decomposition algorithm as means to reduce the computational complexity of the original Mixed Integer Linear Programming (MILP) model. Optimal solutions for the structure and operation of the supply chain network are obtained for demands of up to 5000 therapies per year, in which case the original model contains 68 million constraints and 16 million discrete variables.

Cell banking versus allogeneic therapies

Autologous cell banking provides a viable option of therapeutic value in the event a cell therapy is warranted in the future. Considerable effort toward developing a freezing method optimized for ASC survival after thaw has been invested, resulting in high cellular viability after thaw. Cell banking is however, burdened by the costs of storage, constant process monitoring, and finite supply. For these reasons, allogeneic ASC usage appears promising, particularly because a single or a limited number of cell lines could be optimized for clinical therapies. Additionally, an allogeneic therapy may offer increased consistency, monitored and assessed by validated metrics during expansion. These may include known rates of proliferation or cell secretions in vitro. Despite this, there cannot be total assurance that parameters in vitro will directly correlate to in situ responses. As ASC therapies become an increasing possibility as a clinical tool, an area of interest may be disease state modeling, comparing autologous ASCs from animals with induced diseases with ASCs from healthy counterparts. Though the ASC is praised for its negligible immunogenicity, many of these investigations have been performed either in vitro or in young-animal models. Investigations that challenge the ASC in more relevant diseases states will better elucidate host immunogenic reactions in allogeneic models, or show how age-related cellular senescence affects the ASC and its capacity for regeneration.

Autologous cells

Autologous cells offer a number of benefits to patients, including minimising the risk of immuno-rejection and the potential for personalised medicine. However, from a business viewpoint they are a total anathema to the present pharmaceutical industry. Even big pharma’s interpretation of stratified medicine(s) falls well short of bespoke single-patient therapies derived from the patient’s own material (Trusheim et al., 2007). To date, autologous products have pursued a service-based business model, with a more complex value chain. For example, the individual clinician and manufacturer working together are responsible for the harvest of the donor cells from the patient, their transport to a processing facility, their expansion and potentially differentiation into the required cell population, and then their return to the original donor in a tightly controlled physical environment. Hence, autologous therapies do not offer the same opportunities as their allogeneic counterparts for significant cost reduction, since as a product of one, the quality control (QC) and release by a Qualified Person (QP) will always be of similar order as for a single batch of allogeneic cell therapies regardless of the number of patients treated.

Ultimately, the choice between autologous and allogeneic products is dominated by two decisions: firstly, which type of cell can produce the greatest medicinal effect, meeting the required demands of safety, efficacy and purity, and secondly, if both cell types can be used, which cell type offers the greatest commercial opportunity while being cultural and legally acceptable.

Forthcoming Innovations Accelerating the Pace of Regenerative Medicine

Autologous cells generation for transplantation represents one of the most relevant innovations in regenerative medicine. co.don AG, Germany, has created the Integrated Isolator technology (IIT®) to enable the management of all process parameters, including automatic monitoring and alarm systems to safeguard autologous cell transplants. Particularly important is the protection of chondrocytes from the original material to be then propagated in the patient’s own serum. ZIOPHARM Oncology, US, the synthetic biology leader Intrexon Corporation, United States, and The University of Texas MD Anderson Cancer Center, United States, are working in collaboration to generate next-generation patient- and donor-derived adoptive cellular therapies based on genetically modified T cells and NK cells by applying novel cell engineering techniques and multigenic gene programs. Lonza Biologics, UK, developed L7™ System, a robust xeno-free platform for the generation, maintenance, expansion, and cryopreservation of human induced pluripotent stem cells (iPSCs). Miltenyi Biotec, Germany, reinforces cell reprogramming workflow by covering resources from fibroblast isolation and non-integrating mRNA reprogramming, up to final isolation of reprogrammed iPS cells. Cesca therapeutics, US, signed an asset acquisition agreement related to the acquisition of the cell processing systems of SynGen to enhance its proprietary CAR-T cell manufacturing capabilities. Regarding bone marrow transplantation, Cytomatrix, Canada, continues developing products and services for umbilical cord blood expansion and immune cells transplantation. Similarly, TransTissue, Germany, is deeply involved in the development of cell-based transplants and cell-free implants leveraging its proprietary cell-free cartilage therapy based on mesenchymal stem and progenitor cells recruited to cartilage defects and subsequently guided to regenerate articular cartilage. Another innovative clinical-stage immunotherapy company, NantKwest, US, is using its off-the-shelf activated natural killer (NK) platform to kill cancer and virally infected cells based on the incorporation of chimeric antigen receptors (CARs) and antibody receptors.

Human oligodendrocyte progenitor cells transplantation is other regenerative medicine technology gaining attention in the market. Just to cite some instances, Asterias Biotherapeutics, US, holds three clinical-stage development programs focused on regenerative medicine. AST-OPC1 is associated with several reparative functions that may address complex pathologies related to spinal cord injury, such as production of neurotrophic factors, stimulation of vascularization, and induction of remyelination of denuded axons.

Cartherics, Australia, applies iPSC technology to T cell receptor (TCR) and chimeric antigen receptor (CAR)-T engineered immune cells to generate cancer killing T cells using a dual targeting approach. The company engineers iPSC lines derived from rare homozygous HLA haplotype donors to express CAR constructs and differentiate them into T cells (CAR-T), co-expressing an endogenous T cell receptor to cancer peptides.

Other regenerative medicine approaches implement monoclonal antibodies to slow or reverse senescence. Hence, for instance, SIWA therapeutics, US, develops monoclonal antibodies to target and destroy senescent cells directly implicated in neurodegenerative diseases, autoimmune conditions, and infectious diseases. Abnova, Taiwan, uses genomic and proteomic approaches to obtain at least one antibody to every human expressed gene in human genome by leveraging circulating rare cells, exosomes, and cell-free RNA as the next generation diagnostics into its precision medicine platform.

Regarding 3D bioprinting (3DBP), Organovo, US, is delivering breakthrough bioprinting technology on demand for research and medical applications. Organovo’s NovoGen MMX Bioprinter™ can work across all tissue and cell types helping pharmaceutical companies to develop human 3D biological disease models to enhance therapeutic drug discovery and development. Along the same line, BioBots, US, joins computer science and chemistry together to develop desktop 3D printers for biomaterials to build 3D living tissue and miniature human organs for research and pre-clinical screening. Cyfuse Biomedical, Japan, developed a 3D organ regeneration technology using its proprietary 3D bioprinter Regenova, a robotic system that facilitates the fabrication of 3D cellular structures by using cellular spheroids in fine needle arrays. Aspect Biosystems, Canada, developed a Lab-on-a-Printer™ 3D bioprinting technology for use in the life sciences industry. Rokit, South Korea, released a multimaterial 3Dison Invivo 3D bioprinting platform, which can print over poly-lactic-co-glycolic acid (PLGA), polycaprolactone (PCL), poly-lactic acid (PLLA), collagen, alginate, and silk fibroin, among other materials noticeably overcoming challenges faced by existing bioprinters.

8.1.2 Quality control requirements for allogeneic vs autologous cell-based therapies

Production of both autologous and allogeneic cell therapy products have similar processing requirements, in spite of the different achievable manufacturing scale of each. In fact, in terms of QC the main difference between autologous and allogeneic products is the cost. In Fig. 8.1, a schematic representation of relative cell quantities required for QC testing in autologous and allogeneic processing is shown, reproduced from Brandenberger et al. (2011). The fundamental difference is related to batch size. In an allogeneic process, one batch is used to treat several different patients and so the cell number required for testing represents a small proportion of the total cell production. Conversely, in autologous processes, one batch makes enough cells for one patient and therefore a certain degree of overproduction is necessary to account for the cells required for testing. More significantly, the volumes of reagents used in QC assays increases exponentially and consequently QC contributes to a large proportion of the total cost of an autologous product (Brandenberger et al, 2011). Some allogeneic therapies fall under this umbrella too, such as donor-matched cord blood administrations, where processing is essentially for an allogeneic therapy but produced on an autologous scale, as one donor is required to treat one patient.

Fig. 8.1. Testing challenges in cell therapy manufacture

(Brandenberger et al., 2011). (This figure is reprinted with permission from BioProcess International, March 2011, pages 30–37.)

11.6 LeucoPatch

LeucoPatch is a purely autologous cell-rich fibrin-based biomaterial without chemical additives and belongs to the L-PRF group (Dohan Ehrenfest et al., 2014). LeucoPatch is produced by mechanical means by the patented 3CP process (centrifugation, coagulation, and compaction procedure) using a proprietary closed sterile device (Lundquist and Holm, 2010) in three steps at the bedside: (1) blood is drawn by venipuncture into the sterile vacuum device in a process identical to normal blood sampling, (2) the device is positioned in a specially designed centrifuge insert and spun in an automated two-step process at the bedside, and (3) the device is opened. This process takes 20 min. The formed patch is then applied to the wound without further processing.

Three distinct layers form in LeucoPatch: (1) polymerized and cross-linked fibrin matrix, (2) compacted platelets, and (3) leukocytic concentrates facing the wound. In Fig. 11.2, the three-layered structure of LeucoPatch is shown (Lundquist et al., 2013).

Figure 11.2. The LeucoPatch physical structure. (a) The circular LeucoPatch. (b) The consistency and toughness of the LeucoPatch enables trimming of the medical device with surgical instruments. (c) A cross-section of hematoxylin-eosin-stained LeucoPatch showing the three layers: fibrin (top); platelets (light pink, middle); and leukocytes (bottom layer). (d, e) The leukocytic layer stained with hematoxylin-eosin (d) and examined by scanning electron microscopy (e).

The levels of some growth factors in LeucoPatch were compared with standard P-PRP (Lundquist et al., 2008). PDGF-AB levels were increased by a factor of 2.6 in LeucoPatch and in VEGF by a factor of 9.9 as compared with P-PRP. The neutrophil chemokine IL-8 was 282 times higher in LeucoPatch than in P-PRP (Lundquist et al., 2013).

It was demonstrated in vitro that one growth factor, PDGF-AB, is released continuously from LeucoPatch for at least 1 week. Other in vitro studies showed that LeucoPatch enhanced the proliferation and migration of human fibroblasts and keratinocytes (Lundquist et al., 2013).

Interestingly, the addition of chronic wound fluid increased the rate of release of PDGF-AB from LeucoPatch (Lundquist et al., 2013). This biochemical feature has relevance when using LeucoPatch in the treatment of chronic wounds. Pseudomans aeruginosa is common in chronic wounds that exhibit impaired wound healing (Renner et al., 2012). The effect of LeucoPatch on bacterial survival was studied in one P. aeruginosa strain (PAO1). These tests indicated that the leukocytes in LeucoPatch are viable and capable of phagocytosis and killing the bacteria (Fig. 11.3). Moreover, initial tests indicate that LeucoPatch reduces P. aeruginosa-derived biofilms (Thomsen et al., 2016).

Figure 11.3. Bactericidal activity of LeucoPatch during 90 min of incubation with planktonic Pseudomans aeruginosa (PAO1). LeucoPatches were mixed with PAO1, and at the indicated time points samples were removed for bacterial counting. Samples were serially diluted, plated on Conradi–Drigalski medium, and the colony-forming units (CFU) were determined after an overnight incubation at 37°C.

Immunological considerations

The ability to utilize allogeneic cells rather than autologous cells facilitates the reproducible, large-scale commercial manufacture of an HSE [37], because large cell banks can be created, allowing manufacture of thousands of units of product from one cell strain and enabling a more accurate forecast of manufacturing demands. However, the problem of rejection needs to be taken into consideration when using living allogeneic cells in wound healing applications.

The first stage in the induction of a primary immune response after skin allografting is the presentation of antigen by donor dendritic cells (Langerhans cells with dermal dendritic cells), the professional antigen-presenting cells (APCs) in skin. These cells migrate out of the skin to the draining lymph node, where they can activate T cells directly through the presentation of MHC-class II antigens and co-stimulatory molecules, thereby eliciting both cell-mediated and humoral (antibody-mediated) immune responses to the grafted skin. Although cell-mediated cytotoxicity is a component of rejection, the primary mode of skin rejection is likely mediated via an attack on the vasculature present in a normal skin graft by recipient antibodies [41,42].

Bioengineered skin constructs are fabricated from highly purified banks of fibroblasts, keratinocytes or both, that are either poor or deficient in dendritic and other APCs found in skin. This has important implications for the use of allogeneic cells in the treatment of acute and chronic wounds. In the absence of APCs, fibroblasts and keratinocytes are the only cells capable of presenting donor antigen to the recipient. Under normal conditions, keratinocytes and fibroblasts do not express MHC-class II antigens. They can be induced by interferon-γ to express MHC-class II molecules and thereby acquire the ability to present antigen to T cells, however keratinocytes and fibroblasts do not express co-stimulatory molecules [43,44], so antigen presentation by keratinocytes and fibroblasts does not result in T cell activation. Instead, this antigen presentation can result in T cell non-responsiveness [45,46] or T cell anergy [47].

Autologous HSEs would avoid issues of immunogenicity, of course, but autologous grafts have significant limitations. Growing graft tissues from biopsy takes several weeks, the donor site creates another wound, and in some patients (e.g., severe burn patients) there may be no appropriate donor site. Reproducibly making complex HSE constructs to order from autologous cells would be time consuming, and very costly. Therefore, the ability to effectively use allogeneic human cells is a key element in the commercial success of engineered skin therapies.

Immunological considerations

The ability to utilize allogeneic cells rather than autologous cells facilitates the reproducible, large-scale commercial manufacture of an HSE [37], because large cell banks can be created, allowing manufacture of thousands of units of product from one or few comparable cell strains and enabling a more accurate forecast of manufacturing demands. However, the problem of rejection needs to be taken into consideration when using living allogeneic cells in wound healing applications.

The first stage in the induction of a primary immune response after skin allografting is the presentation of antigen by donor dendritic cells (Langerhans cells with dermal dendritic cells), the professional antigen-presenting cells (APCs) in skin. These cells migrate out of the skin to the draining lymph node, where they can activate T cells directly through the presentation of MHC-class II antigens and costimulatory molecules, thereby eliciting both cell-mediated and humoral (antibody-mediated) immune responses to the grafted skin. Although cell-mediated cytotoxicity is a component of rejection, the primary mode of skin rejection is likely mediated via an attack on the vasculature present in a normal skin graft by recipient antibodies [41,42].

For the most part, bioengineered skin constructs are fabricated from highly purified banks of fibroblasts, keratinocytes, or both, which are either poor or deficient in dendritic and other APCs found in skin. This has important implications for the use of allogeneic cells in the treatment of acute and chronic wounds. In the absence of APCs, fibroblasts and keratinocytes are the only cells capable of presenting donor antigen to the recipient. Under normal conditions, keratinocytes and fibroblasts do not express MHC-class II antigens. They can be induced by interferon-γ to express MHC-class II molecules and thereby acquire the ability to present antigen to T cells; however, keratinocytes and fibroblasts do not express costimulatory molecules [43,44], so antigen presentation by keratinocytes and fibroblasts does not result in T cell activation. Instead, this antigen presentation can result in T cell nonresponsiveness [45,46] or T cell anergy [47]. It should be noted that there are continued efforts to develop, test, and commercialize constructs comprising also endothelial cells and/or adnexal skin structures. Therefore the abovementioned immunological aspects will need to be considered when creating and testing even more complex constructs.

Autologous HSEs would avoid issues of immunogenicity, of course, but autologous grafts have significant limitations. The process of growing constructs’ components from skin biopsies takes several weeks, the donor site creates another wound and, in some patients (e.g., severe burn victims), there may be no appropriate and safe donor site. Reproducibly making complex HSE constructs to order from autologous cells would be time-consuming, and very costly. Therefore the ability to effectively use allogeneic human cells is a key element in the commercial success of engineered skin therapies.

The development rates of autologous vs allogeneic products

From a commercial perspective, the development rates of autologous and allogeneic cell products are very interesting, reflecting differences not only in cell product process requirements and safety but also in commercial investment. Due in part to the CTI’s roots in bone marrow transplantation and blood processing, commercial-scale autologous cell therapy processing has been refined at a faster rate than allogeneic processing of at least some types of cell therapy intervention. This is because for autologous cell therapy (such as during autologous chondrocyte transplantation), the bio- process tools are essentially the same as those used for lab-scale cell culture, albeit in a scaled-out manner and to GMP guidelines. Conversely, the development rate of large-scale allogeneic manufacturing technologies, including suspension cultures, has been slower, due to the major technological step change required in the way adherent human cells for therapy are cultured and the lack of translatability and applicability of equipment and technologies from other fields. To date, attempts at translating technologies from other industries have really been limited to stirred tank bioreactors utilising microcarriers to provide substrate support for attachment of adherent cells (King and Miller, 2007; Brindley et al., 2011; Yeatts and Fisher, 2011). However, these processes are not always appropriate for scalable production of cell-based therapeutics due to difficultly in liberating the cells from their microcarriers after expansion and reduced cell viability or phenotype modification that results from exposure to flow-induced mechanical stress (Brindley et al, 2011).

Exemption for Treatments During the “Same Surgical Procedure”

Importantly, the surgical removal and subsequent implantation of the autologous HCT/P may be considered to be the “same surgical procedure” even though the time between removal and future implantation is separated by a number of days. However, strictures are placed on manipulation of those tissues. Only rinsing, cleaning or temporary storage of HCT/Ps after being labeled can be performed during this gap in time. In this case, the communicable disease risks, as well as safety risks, generally would be no different from those typically associated with surgery. Examples of sizing and shaping that would generally allow the HCT/P to remain “such HCT/P” include, for example, dilating a vascular graft to change its size for coronary artery bypass graft surgery or meshing skin grafts in a way that would allow the tissue to cover cutaneous burn wounds more effectively.

An establishment that removes an HCT/P for implantation into the same individual but intends the HCT/P to be implanted at a different establishment would not qualify for the “same surgical site” exception. Shipping the HCT/P to another establishment for implantation raises safety concerns, such as contamination and cross-contamination. The shipping of the autologous HCT/P for use at another establishment is considered to be a type of manufacturing step, and therefore, the shipping establishment must comply with all the requirements of Part 1271. Any other processing steps will disqualify an establishment for the exemptions described above, because the processing of the autologous HCT/P raises safety concerns, such as contamination and cross-contamination, beyond those typically associated with a normal surgical procedure. For example, centrifugation or filtration, cell isolation, cell expansion, cell activation, or enzymatic digestion generally would not allow the HCT/P to qualify for the exception.