3

Hematologic and Immunologic Applications

Several exciting advances in cell-based regenerative medicine have taken place in the fields of hematology and immunology over the past two decades. These include harnessing gene-editing technology to treat human immunodeficiency virus (HIV), using modified cells and gene therapy to improve clinical outcomes for patients who undergo hematopoietic stem cell transplantation (HSCT), and the use of engineered T cells for the effective treatment of cancer. The speakers in this session provided a high-level overview of the state of their respective fields, described the challenges and successes they encountered along the way, and discussed how those lessons can be applied moving forward.

CLINICAL APPLICATIONS OF GENOME EDITING

Gene editing has become a clinical reality, said Fyodor Urnov, the associate director of the Altius Institute for Biomedical Sciences. It is being used in multiple clinical trials and multiple open investigational new drugs, both ex vivo and in vivo, and there is great potential for its use with regenerative medicine, he said. The basic concept underlying gene editing, Urnov said, is that when a DNA double-strand break is created, the cell repairs the break either through non-homologous end-joining or homology-directed repair. Scientists can leverage the repair pathways, using gene editing to drive specific outcomes at targeted loci within the cell or tissue of interest. Urnov spoke about three approaches to targeted gene editing: an insertion/deletion “indel,” which can create a “knockout” by inactivating a gene; a single nucleotide polymorphism (SNP), which can correct a mutation; or a transgene, which can produce a targeted integration. The first gene editing tool used to make targeted breaks in a cell’s DNA was the zinc finger nuclease (ZFN) (Klug, 2010). Since that original use of ZFN, researchers have developed various other nucleases to perform gene editing, and current approaches now include TALE (transcription activator–like effector nuclease) nucleases, meganucleases, and RNA-programmable nucleases such as the clustered regularly interspaced short palindromic repeat (CRISPR)/Cas9 (Doudna and Charpentier, 2014). Each genome editing tool has advantages and disadvantages, Urnov said, and the technologies will continue to evolve

and be refined in years to come. Pharmaceutical companies have come to recognize the potential of gene editing, he said, and they have created partnerships in order to pursue the genome as a “legitimate drug target” despite the inherent challenges of a long and costly research and development process. This investment, he said, “has provided the infrastructure, the funding . . . and the courage to go after larger disease indications.”

As examples, Urnov went on to describe two gene editing approaches taken by researchers at Sangamo BioSciences that have reached the clinic for the treatment of HIV-1 infection. Both examples involve targeting the CCR5 gene for inactivation, with the first approach being undertaken in T cells and the second in CD34+ hematopoietic stem and progenitor cells (HSPCs). The ability of HIV-1 to enter cells depends on the binding of viral gp120 Env protein to the host’s CD4 receptor and a chemokine co-receptor, most commonly CCR5 (Holt et al., 2010; Wu et al., 1996). Individuals who are homozygous for a frameshift-causing 32-base-pair deletion in CCR5 (CCR5Δ32) are profoundly resistant to HIV-1 infection (Liu et al., 1996). This finding, in combination with evidence from the “Berlin patient,” an individual cured of HIV-1 infection after receiving an allogeneic transplant with stem cells carrying nonfunctional CCR5, spurred researchers at Sangamo BioSciences to develop CCR5-targeted HIV treatments (Tebas et al., 2014; Zou et al., 2013). The first approach, Urnov said, involved harvesting CD4+ T cells from patients with HIV, using ZFN to disrupt CCR5 in those cells, and then transplanting the edited cells back into the patient. Preclinical efficacy tests were rather straightforward and involved showing that the edited T cells functioned in vitro and were efficacious in a mouse model. So far, he said, this approach has been well tolerated by patients and has produced an antiviral effect. There has also been evidence that the edited cells home to the gut-associated lymphoid tissues and persist in the body for at least 4 years (Tebas et al., 2014). In the most recent cohort, 60 percent of patients were able to control their viral load in the absence of antiretroviral therapy, Urnov said. The trial protocol will continue to be refined as ZFN technology improves, he said, and there is now an established good manufacturing practice (GMP) pathway that meets criteria for efficiency and specificity. The company has worked closely with the Food and Drug Administration (FDA) to develop and meet regulatory requirements.

The second approach also targets CCR5 for HIV, but uses less mature CD34+ HSPCs that can differentiate into a variety of blood cell

types. This approach has met standards for safety and efficacy, Urnov said. A clinical trial using this therapeutic approach is currently open, but information from the trial was not available at the time of the workshop.

Milestones and Avenues for Improvement

The approaches to targeting CCR5 for the treatment of HIV have paved a path for the use of gene editing in the clinic, Urnov said. Both approaches have established clinical scale efficiency and specificity, he said, as evidenced by the development of an FDA pharmaceutical quality/chemistry, manufacturing, and controls path, a GMP path, and with toxicology evidence. In addition, the researchers from Sangamo BioSciences have worked with FDA and the Recombinant DNA Advisory Committee of the National Institutes of Health¹ to develop a regulatory framework from which other gene-editing therapies could benefit. Avenues to improve the quality and reduce the cost of these potential therapies in the future would include better, more efficient cell processing; allogeneic off-the-shelf products that reduce the need to expand and edit autologous cells; and in vivo delivery of gene therapy, Urnov said. Low yields of CD34+ HSPCs are a current challenge, so finding a way to create greater numbers of targeted HSPCs would drastically change the landscape of therapies for that cell type, he said.

ALLOGENEIC HEMATOPOIETIC STEM CELL TRANSPLANTATION

Allogeneic HSCT, or bone marrow transplantation, has been well established for nearly 30 years as a standard-of-care treatment for many monogenic diseases and cancers, said Harry Malech, chief of the Laboratory of Host Defenses and chief of the Genetic Immunotherapy Section at the National Institute of Allergy and Infectious Diseases. It has become the standard of care for hemoglobinopathies, inherited bone marrow failure syndromes, primary immune deficiencies, lysosomal storage diseases, and some metabolic enzyme deficiencies and leukodystrophies. According to Malech, successful HSCT requires that the following four conditions be met:

___________________

¹ The charter of the Recombinant DNA Advisory Committee is available at http://osp.od.nih.gov/sites/default/files/resources/RAC_2015-2017_Charter_Updated.pdf (accessed December 4, 2016).

1. Access to a suitably HLA-matched donor
2. An adequate and safe conditioning regimen to attain permanent engraftment
3. A preventative regimen to reduce or prevent graft-versus-host disease
4. Prevention, detection, and effective treatment of viral, bacterial, and fungal infections that may occur due to the immunodeficient status of transplant patients

Increasing the Pool of Suitable Donors

Often, the best donor is an HLA-matched sibling, Malech said; however, the availability of matches through unrelated donors has vastly increased in recent years due to the National Marrow Donor Program, and the best donor is increasingly likely to be found through that program.² One challenge that still remains is that individuals of mixed ethnic heritage may have trouble finding a donor within the National Marrow Donor Program, Malech said. To expand the potential donor pool, researchers have also been exploring the possibility of performing haploidentical HSCT using a relative who is a partial HLA match (Locatelli et al., 2013). Successful haploidentical HSCT was recently demonstrated in a patient with chronic granulomatous disease, a genetic disorder that causes recurrent infections and autoinflammation, and in another patient with DOCK8 deficiency, which results in combined immunodeficiency (Freeman et al., 2016; Parta et al., 2015). The HCST conditioning regimen for both of the patients involved administering high-dose cyclophosphamide administration post-transplant to help with immunosuppression.

Optimizing Conditioning Regimens

Conditioning regimens, which prepare a patient’s body to accept the graft by depleting lymphocytes and HSPCs, have traditionally relied on chemotherapy or radiation. A variety of other serotherapies, such as antithymocyte globulin or anti-CD52 monoclonal antibody (alemtuzumab), are available to use instead of radiation and chemotherapeutic agents to deplete lymphocytes, Malech said, but it is still a challenge to completely

___________________

² More information about the National Marrow Donor Program can be found by visiting https://bethematch.org (accessed December 4, 2016).

eliminate host HSPCs without using radiation or chemotherapy. One approach to overcoming this challenge is to use a monoclonal antibody therapy that targets c-Kit, a protein found on HSPCs and downstream hematopoietic progenitors. A recent study in mice found that a combination approach using anti-c-Kit monoclonal antibody plus antiCD47 antibody resulted in the depletion of host HSPCs in immunocompetent recipients with efficient engraftment of donor cells (Chhabra et al., 2016). Another recent study in mice used immunotoxin (saporin)-conjugated anti-CD45 antibody to achieve similarly efficient HSPC engraftment in immunocompetent mice (Palchaudhuri et al., 2016). Targeted HSPC conditioning regimens that use only biologic agents represent a disruptive innovation that could transform HSPC transplantation, Malech said. Looking toward the future, studies may focus on improvements in approaches to toxin-conjugated antibodies that more specifically target HSPCs or even possibly the development of chimeric antigen receptor T cells that specifically target HSPCs, he said.

New Targets to Prevent Graft-Versus-Host Disease

Patients who undergo HSCT are at high risk for graft-versus-host disease (GVHD), a condition that occurs when donor cells recognize the recipient’s normal cells as foreign and mount an immune response, resulting in an array of symptoms that can include inflammation, gastrointestinal distress, jaundice, and dryness of mucus membranes (Leukemia & Lymphoma Society, 2016). Currently, the overall success rates of HSCT are hampered by morbidity and mortality associated with GVHD (Lappas et al., 2010). According to Malech, preventing GVHD is an extremely important concern for patients who undergo HSCT for indications other than the treatment of hematologic malignancies (e.g., for the treatment of monogenic diseases such as immunodeficiencies and hemoglobinopathies). Although immunosuppressive agents such as rapamycin, cyclosporin A, tacrolimus, mycophenolate mofetil, and methotrexate are used clinically to induce tolerance, they have only been partially successful, and there is a pressing need to develop new methods to prevent GVHD, Malech said. One new potential approach to prevent GVHD is the use of highly specific adenosine A_2A receptor (A_2AR) agonists (Lappas et al., 2010). The A_2AR is involved in the termination of inflammatory signals, and the selective activation of the A_2AR has been shown to limit inflammation and tissue damage in several models of inflammatory disease (Awad et al., 2006; Lappas et al., 2006; Naganuma

et al., 2006; Zarek et al., 2008). Although A_2AR agonists have not yet entered the clinic for use against GVHD, Malech said, they are a very exciting prospect.

Combating Infections in Transplant Recipients

HSCT recipients are at increased risk of serious bacterial, viral, and fungal infections because of their lowered immunity. A majority of patients who receive HSCT acquire an infection and 17–20 percent of those infections result in death (Leen et al., 2014). HSCT recipients are especially vulnerable to infections for reasons that include the immunosuppressive drugs they receive pre-transplant, the cytotoxic chemicals used to prevent GVHD, and the symptoms of GVHD itself (Leen et al., 2014). In recent years, Malech said, great strides have been made in the ability to rapidly detect and control post-transplant viral infections such as cytomegalovirus and Epstein-Barr virus. One such advance has been with the use of virus-specific T cells (VSTs) that can be used “off-the-shelf” because they are derived from individuals with common HLA polymorphisms (Leen et al., 2013). Despite the fact that off-the-shelf VSTs have reached the clinic and shown efficacy in reducing mortality and bridging the gap to reacquisition of post-transplant immunity, many open questions remain in this rapidly emerging area, Malech said.

Advances in Gene Therapy

The first evidence of clinically beneficial gene therapy for monogenic immune deficiencies was observed using infusions of murine gamma retrovirus vector-transduced autologous HSCs to treat X-linked severe combined immune deficiency (X-linked SCID) or adenosine deaminase-deficient severe combined immune deficiency (Aiuti et al., 2009; Gaspar et al., 2011; Hacein-Bey-Abina et al., 2010, 2014). However, gamma retrovirus vector gene therapy used to treat X-linked SCID, chronic granulomatous disease, or Wiskott-Aldrich syndrome has been associated with vector-insertion-related genotoxic effects leading to development of leukemia or myelodysplasia (Braun et al., 2014; Hacein-Bey-Abina et al., 2008; Howe et al., 2008; Stein et al., 2010). Self-inactivating lentivectors derived from HIV-1 appear to show enhanced transduction of long-term engrafting human HSPCs while at the same time appear to be less capable of activating nearby oncogenes. Gene

therapy–based approaches that involve an infusion of lentivectortransduced autologous HSPCs show clinical promise for the treatment of several monogenic disorders without evidence of genotoxicity (Malech and Ochs, 2015). At the conclusion of his presentation, Malech briefly reported that several recent trials of lentivector gene therapy have demonstrated promise for significant long-lasting clinical benefits for patients with monogenic illnesses including thalassemia (Cavazzana-Calvo et al., 2010), X-linked SCID (De Ravin et al., 2016), Wiskott-Aldrich syndrome (Hacein-Bey-Abina et al., 2015), metachromatic leukodystrophy (Sessa et al., 2016), and X-linked adrenoleukodystrophy (Cartier et al., 2009). “In our ongoing clinical trial of lentivector with busulfan conditioning for older children and young adults with X-linked SCID, we have been able to restore immunoglobulin production in our patients,” Malech said (De Ravin et al., 2016). Although self-inactivating HIV-1-based lentivector has become the current vector of choice for ex vivo transduction of human HSPC, related integrating lentivectors such as those derived from foamy virus should be explored, he said. An important goal for the future, Malech said, would be to discover methods to create “druggable” versions of these integrating vectors that are delivered and targeted in vivo.

THE PROMISE OF T CELL ENGINEERING

Although several emerging oncology treatments can induce impressive responses, there is still a serious lack of cancer therapies that are both specific and curative, said Michel Sadelain, the director of the Center for Cell Engineering Immunology Program at the Sloan Kettering Institute’s departments of medicine and pediatrics and the Memorial Sloan Kettering Cancer Center. A drug that is specific to a given target should be both safe and efficacious, Sadelain said, and T cells, a component of the adaptive immune system, have evolved to target molecules in an exquisitely specific fashion that provides long-lasting immune support. “T cells do not always have the potency required to fight cancer,” he said, “and that’s where T cell engineering comes into play.” T cell engineering is a technique used to reprogram T cells to harness and improve on their natural immunological abilities to increase their potency and achieve a superior immune response.

Chimeric antigen receptors (CARs) are artificial receptors that are designed to target T cells to respond to specific antigens of choice. They

mediate T cell antigen recognition and activation, and they augment T cell functionality and persistence (Sadelain, 2015). In order to get CARs expressed on the surface of T cells, CAR cDNA must be introduced into T cells, Sadelain explained, and this is often accomplished via retroviral or lentiviral vectors. Once the T cells express the CAR molecule, they become known as CAR T cells, and they have the ability to recognize target antigens on the surface of tumor cells and attack those cells. The reprogrammed CAR T cells are expanded ex vivo and infused back into the patient.

Sadelain and his lab began researching CAR T cells over a decade ago, focusing their work on exploring the therapeutic potential of primary human T cells that were genetically modified to recognize and kill tumors that express the B cell lineage-specific antigen, CD19 (Brentjens et al., 2003). CD19 is a transmembrane protein that is expressed on normal B cells, follicular dendritic cells, and the cells in many types of blood malignancies, including acute lymphoblastic leukemia and chronic lymphocytic leukemia (Wang et al., 2012). Over the years, Sadelain and his colleagues created a manufacturing platform within their academic setting. CAR T cell development within the manufacturing platform begins with collection of a patient’s T cells by apheresis, Sadelain said. The T cells are then activated by incubation with antibodies to CD3 and CD28, a viral vector is applied, and the cells are cold cultured and expanded to allow for gene transfer to occur and to increase the number of genetically edited T cells. Within 8–10 days, and following a few additional biosafety tests, the cells are ready for infusion back into the patient, or they can be frozen for deferred use, Sadelain said.

Recent studies of CAR T cells produced in this way have shown great promise, Sadelain said, describing the results of one of his lab’s studies that was presented at the 2015 American Society of Hematology conference. Forty-five adults with refractory (treatment-resistant) acute lymphoblastic leukemia received the treatment, and 82 percent went into complete remission. Complete remission is considered a molecular remission, he said, meaning that the tumor is undetectable by deep sequencing. Even though it will take several years to determine if the patients are cured, the early results are extraordinary, he said. These early findings resulted in Science naming cancer immunotherapy 2013’s Breakthrough of the Year, highlighting the work on CAR therapy and another approach called checkpoint blockade, which is aimed at blocking inhibitors of the immune response.

When asked about the potential of CAR T cell therapy to treat solid tumors, Sadelain said that a number of different cancer cells carry the CD19 surface marker, but CAR T cell therapy works better on some than on others. This difference in effectiveness, he said, is likely due to the fact that different tumor types have different microenvironments, each unique to the particular tumor type, and solid tumors tend to have more inhibitory mechanisms present than cancers like acute lymphoblastic leukemia. Research is being conducted on CAR T cells to improve their effectiveness in the tumor microenvironment, but designing and using cells as an “ultra-targeted” chemotherapy is a complex undertaking, Sadelain said.

Previous T cell–based therapeutic approaches have relied on finding and expanding the right cell in the patient or in the donor. The new paradigm, Sadelain said, is likely to use gene transfer, gene editing, and synthetic biology to manufacture T cells that have the optimal properties for the intended use. Sadelain listed what he sees as four priority research areas needed to move the field forward:

1. Optimization of CAR design (e.g., second generation CARs, armored CARs)
2. Additional basic research on T cell differentiation in order to identify which subpopulation is the most well-suited for this type of therapy
3. Innovations in cell-manufacturing sciences to get the product to patients more efficiently
4. Integration of gene transfer and gene editing technology with CAR therapy

The Future of CAR T Cell Therapies

T cells have been used as medicines for a number of years; however, progress in adoptive T cell therapy has been slow because of a lack of antigen-specific human T cells (Themeli et al., 2013). A new paradigm is emerging in which researchers use pluripotent stem cells as the source material to manufacture reservoirs of engineered cells devoid of alloreactivity, Sadelain said. The ultimate goal, he said, would be to have a “cell pharmacy” with prepared cells that can be administered to patients for multiple conditions.

Sadelain offered three important lessons from his experience with CAR T cells. First, these types of breakthroughs take time: 21 years

lapsed between the beginning of T cell engineering and FDA designation of CAR T cells as a breakthrough therapy (a breakthrough therapy designation is intended to allow expedited development and review of drugs for serious or life-threatening conditions³). Second, while the involvement of industry is critical to moving a therapy into manufacturing and commercialization, the role that academia plays cannot be understated. Academic teams have the benefit of cross-fertilization between researchers who are working in different but related fields. Third, education and network building are crucial components of developing these therapies. Both patients and providers were reluctant to participate in early CAR T cell therapies, Sadelain said, and moving forward would not have been possible without addressing concerns and educating people about how the therapy worked.

PATIENTS AS ACTIVE PARTICIPANTS IN HUMANITY-BASED RESEARCH

Research participants are often viewed simply as “study subjects,” but Jennifer Fields, a patient advocate, argued that they should be considered “active participants of humanity-based research.” Having an informed patient population is critical to supporting continuing research in the field, she said, but there is a significant communications gap in the current scientific research process. Researchers and patients move in separate spheres and are only connected by physicians or organizational liaisons who facilitate communication between the groups. If researchers engaged patients earlier in the scientific process, they would have a more trusting patient base that would actively participate in clinical trials and would advocate for research funding and support, Fields stated.

Patients want to be involved in the research process, and they want to understand and trust research, Fields said, suggesting that in order to involve patients, researchers must shift some resources into patient engagement and should use knowledgeable patients as liaisons between the research and patient communities. She also emphasized the importance of continuous communication and feedback between the communities as a way of creating a true relationship that would benefit all involved.

___________________

³ For more information on breakthrough therapies see http://www.fda.gov/RegulatoryInformation/Legislation/SignificantAmendmentstotheFDCAct/FDASIA/ucm341027.htm (accessed December 1, 2016).

PANEL DISCUSSION

Potential Risks of Gene Editing and Cellular Therapies

What are some of the potential “off-target effects” from gene editing or cell therapies, Dunbar asked, and how we can predict and prevent some of the most likely effects? When gene therapy was nascent, there were a variety of concerns about negative effects from the therapy, Malech said. Cancer was one potential side effect, but studies of early retroviral vectors in mice did not show cancer as an outcome, he said, so it only became clear that it was a real danger of the therapy once trials began and a number of patients developed cancer. Gene editing may be similar in that it may be possible to foresee and prepare for some of the effects of the therapy, but there will likely be some effects that were not predicted before implementation in patients. Controlling the risks of these therapies is an iterative process that requires investigators, regulators, and clinical physicians to work together closely to quickly address effects as they are identified, Malech said. Sadelain noted that, in addition to the failure of mouse models to predict cancer as an effect of gene therapy, mouse models also failed to predict heart failure as a result of CAR T cell therapy. Mouse models can be very valuable, but it is clear that it will not be possible to predict all potential risks using these types of models.

When transplanting stem cells, it is possible to sequence the genome and identify a mutation within a cell, but it is difficult to know if the mutation will pose a clinical risk or not, Urnov added. The gene editing and gene therapy community is “acutely aware” of the challenges associated with identifying and addressing risk in these therapies, he said, and it is working to improve existing systems and building new ones to continue to lower the risk of these therapies.

The potential of using the genomic profile of patients to reduce the risk of toxicity of a therapy was also discussed. This area is still in its infancy. For example, fully sequencing a patient’s genome results in only about 60 to 70 actionable pieces of information concerning, for instance, the patient’s susceptibility to anesthesia, Malech said. However, the area is growing rapidly, and personalized medicine has the potential to influence the provision of these therapies in the future, he said.

The Use of Data and Data Registries

The collection and sharing of outcomes data has been instrumental in making advances in the transplant field, as can be seen in developments in immunosuppression and in the prevention of GVHD, Dunbar said. She said that because many trials are quite small, sharing data may be the only way that investigators can solve manufacturing issues or learn how cells behave in a patient. Several workshop participants added that data likely need to be collected across a broad number of endpoints such as genetics, age, gender, ethnicity, and immune profile, among others. Currently, funding is an obstacle to collecting truly comprehensive data, Sadelain replied, noting that trials are already expensive and that adding more data collection points will increase the cost.

Patient Involvement

A workshop participant asked speakers about the best ways for researchers to engage patients and inform them about the risks of new therapies in order to avoid stymieing progress in the field in the case of unintended side effects. Patients need to be engaged early on in the process, Fields responded, and they need to be made to feel as if they are partners in the research. Patients should feel that researchers are working for them and with them, she said. Dunbar added that researchers need to strike a balance—they need to convey their enthusiasm and optimism about the possibilities of the research in order to get patients interested in participating in trials, while at the same time avoiding overpromising and leading patients to believe that a cure is right around the corner. Dunbar suggested that liaisons who are informed about the science and about the realities of clinical trials could act as brokers between researchers and patients to present a more balanced view.

This page intentionally left blank.