Cancer Immunotherapy: Building on Initial Successes to Improve Clinical Outcomes

Highlights of this Report Include:

• Approvals of checkpoint inhibitors


• Biomarkers for checkpoint inhibitor treatments


• Approved and clinical-stage immunotherapy biologics other than checkpoint inhibitors


• Immunotherapy with TIL cells


• Commercialization of TIL therapy


• Adoptive immunotherapy with genetically engineered T cells bearing chimeric antigen receptors (CARs)


• Manufacturing issues with CAR T-cell therapies


• General conclusions on the progress of cellular immunotherapy


• Outlook for cancer immunotherapy

Executive Summary

Cancer Immunotherapy: Building on Initial Successes to Improve Clinical Outcomes

This new report builds on our 2014 Insight Pharma Report, Cancer Immunotherapy: Immune Checkpoint Inhibitors, Cancer Vaccines, and Adoptive T-cell Therapies. In that report, we focused on the major classes of cancer immunotherapy drugs that were then emerging from academic and corporate research: immune checkpoint inhibitors, cancer vaccines, and adoptive T-cell therapies. This new report includes an updated discussion of approved and clinical stage agents in immuno-oncology, including recently-approved agents. It also addresses the means by which researchers and companies are attempting to build on prior achievements in immuno-oncology to improve outcomes for more patients. Some researchers and companies refer to this approach as “immuno-oncology 2.0.” The American Society of Clinical Oncology (ASCO), in its 12th Annual Report on Progress Against Cancer (2017), named “Immunotherapy 2.0” as its “Advance of the Year.”

As discussed in our 2014 report and still true in early 2017, the most successful class of immunotherapeutics has been the checkpoint inhibitors (which are discussed in Chapter 2 of this report). Checkpoint inhibitors and other immuno-oncology agents represent a significant advance in cancer treatment beyond the traditional modalities of chemotherapy, radiation therapy, and surgery. Moreover, treatment of advanced melanoma (the cancer for which the largest amount of data on immunotherapy has been amassed) with checkpoint inhibitors has in some cases produced spectacular results. For example, data released at the May 2016 ASCO Annual Meeting indicate that 40% of metastatic melanoma patients who received pembrolizumab (Merck’s Keytruda) in a large clinical trial are still alive three years later. This represents a substantial improvement over just a few years ago, when the average survival time for patients with advanced melanoma was measured in months.

Nevertheless, metastatic melanoma remains incurable. Furthermore, in many studies in advanced melanoma and other cancers, only a minority of patients have benefited from immunotherapy treatments. Researchers and companies are therefore looking for ways to build on the initial successes of the immuno-oncology field to improve outcomes for more patients, hence the need for an “immuno-oncology 2.0.”  Agents that are intended to improve the results of treatment with agents like checkpoint inhibitors may also be referred to as “second-wave” immuno-oncology agents.

As discussed in this report, researchers have found that checkpoint inhibitors produce tumor responses by reactivating TILs (tumor infiltrating lymphocytes)—especially CD8+ cytotoxic T cells. This key observation is perhaps the most important factor driving development of second-wave immuno-oncology strategies. As a result, researchers have been developing biomarkers that distinguish inflamed (i.e., TIL-containing) tumors—which are susceptible to checkpoint inhibitor therapy—from “cold” tumors, which are not. They have also been working to develop means to render “cold” tumors inflamed, via treatment with various conventional therapies and/or development of novel agents. These studies are the major theme of “second-wave” immuno-oncology, or “immuno-oncology 2.0.”

Approvals of checkpoint inhibitors

As discussed in Chapter 2, researchers are continuing to conduct clinical trials designed to gain approval for new checkpoint inhibitors and for new indications for already-approved agents. Notable recent developments include the 2016 approval of atezolizumab (Roche/Genentech’s Tecentriq), the first PD-L1 (programmed death-ligand 1) inhibitor to be approved. On May 18, 2016 atezolizumab was approved by the FDA for treatment of advanced or metastatic urothelial carcinoma that has worsened during or following platinum-containing chemotherapy or within 12 months of receiving platinum-containing chemotherapy, either before or after surgical treatment. Later, on October 18, 2016, the FDA approved atezolizumab for use in patients with metastatic NSCLC (regardless of PD-L1 expression) who have progressed during or after treatment with a platinum-based chemotherapy or appropriate targeted therapy.

Also in October 2016, the FDA approved the PD-1 (programmed cell death protein 1) inhibitor pembrolizumab as a monotherapy for first-line treatment of patients with advanced NSCLC whose tumors expressed PD-L1 at ≥50%. This was after this agent met its primary endpoint of progression-free survival in patients with previously untreated advanced NSCLC whose tumors expressed PD-L1 at ≥50%. In contrast, monotherapy with the competing PD-1 inhibitor nivolumab (Bristol-Myers Squibb’s Opdivo) did not meet its primary endpoint of progression-free survival in patients with previously untreated advanced NSCLC whose tumors expressed PD-L1 at ≥5%. This result is affecting the competition between BMS’ nivolumab and Merck’s pembrolizumab.

In Merck’s KEYNOTE-024 trial, the patient population that was treated with either pembrolizumab or chemotherapy consisted of individuals with previously untreated advanced NSCLC whose tumors expressed PD-L1 at ≥50%. In contrast, BMS’ CheckMate 026 trial of nivolumab as a monotherapy evaluated the drug in patients with previously untreated advanced NSCLC whose tumors expressed PD-L1 at only ≥5%. This difference in trial design may explain the divergent results of the two trials, rather than a potential superior efficacy of pembrolizumab over nivolumab. Nevertheless, the results of the KEYNOTE-024 trial advance the prospects of Merck’s pembrolizumab for first-line treatment of advanced NSCLC with high levels of PD-L1 expression, while BMS must conduct an evaluation of its study and decide what to do next.

In addition to the discussions of approved checkpoint inhibitors, Chapter 2 also includes discussions of clinical stage agents in this class. These include Novartis’ PD-1 inhibitor PDR001, AstraZeneca’s PD-L1 inhibitor durvalumab, and Merck-Serono/Pfizer’s PD-L1 inhibitor avelumab. Notably, avelumab has been under evaluation in a pivotal Phase 2 trial in Merkel cell carcinoma, with favorable results reported in the 2016 ASCO annual meeting. Merck-Serono and Pfizer plan to submit the drug to regulatory authorities based on these results.

Biomarkers for checkpoint inhibitor treatments

The later sections of Chapter 2 discuss the role of biomarkers in checkpoint inhibitor treatments, especially in the context of “immuno-oncology 2.0.” “Immuno-oncology 2.0” may involve development of novel agents, such as those discussed in this and other chapters of the report. It may also involve combining different immunotherapies, combining immunotherapies with older types of treatments and/or with new experimental treatments, or other novel approaches. The development and use of biomarkers will be key to the progress of “immuno-oncology 2.0.” Biomarkers will help researchers and physicians predict responses to immunotherapy treatments. Such tests may not only spare patients the costs and adverse effects of treatments that may not help them, but may also help researchers to design optimal, “personalized” treatments.

Several classes of biomarkers are in use and/or development for cancer immunotherapy, and especially for use in combination with checkpoint inhibitors. A target biomarker is a biomarker that reflects the presence of a specific molecular drug target. In the case of PD-1 inhibitors, the direct target is PD-1, and the downstream target (i.e., the ligand of PD-1 that is affected by its binding) is PD-L1. In the case of PD-L1 inhibitors, the direct target is PD-L1. Recent results from studies of first-line treatment of advanced NSCLC with either nivolumab or pembrolizumab as a monotherapy demonstrate the potential value of PD-L1 as a biomarker in treatment of patients with PD-1 inhibitors.

PD-L1 as a biomarker has also been important in clinical studies supporting the approval of the PD-L1 blocking agent atezolizumab. Researchers found that increased PD-L1 expression in the tumors of patients with urothelial carcinoma was associated with response to atezolizumab. Although patients with tumors negative for PD-L1 expression might still respond to the drug, the greater efficacy of atezolizumab in those classified as positive for PD-L1 expression suggests that the level of PD-L1 expression in tumor-infiltrating immune cells may help identify patients more likely to respond to treatment with the agent.

Target biomarkers—especially PD-L1—are being used to define patient subsets that can productively be treated with a checkpoint inhibitor, especially in clinical trials and in approval decisions by regulatory agencies. However, these tests imperfectly discriminate between patients who can benefit from these therapeutics and those who cannot. Moreover, they are of little use in designing improved therapies that build on current checkpoint inhibitor therapies to improve patient outcomes.

Genetic biomarkers are also under investigation for use in cancer immunotherapy. In immuno-oncology, genetic biomarkers are generally used to determine the likelihood that a patient’s tumors possess a sufficient somatic mutation load to support a large and diverse population of CD8+ TILs, which are specific for mutation-associated neoantigens. Treatment with checkpoint inhibitors can then reactivate these TILs, resulting in effective antitumor immune responses. Examples of genetic biomarkers discussed in this report include mismatch repair (MMR) deficiency and mutation load, as determined by whole-exome sequencing.

Immunological biomarkers enable direct testing to determine whether a patient’s tumors contain sufficient TILs to enable successful treatment with a checkpoint inhibitor. In particular, researchers have found that CD8+ TILs located at the invasive margin of a tumor (as determined, for example, by quantitative immunohistochemistry) appear to be necessary for successful treatment with checkpoint inhibitors. In one study, researchers found that pre-existing CD8+ T cells located at the invasive margins of tumors from patients with metastatic melanoma may predict response to therapy with the anti-PD-1 inhibitor pembrolizumab. Patients who responded to therapy showed proliferation of the intratumoral CD8+ T cells that directly correlated with reduction in tumor size. The researchers established a predictive model based on CD8 expression at the invasive margin and validated the model in an independent group of 15 patients.

Another type of immunological biomarker is the “Immunoscore”—a method of characterizing the nature and function of immune cell infiltrates into tumors based on measuring the densities of CD3+ and CD8+ cells in the tumor core and the invasive margin using immunohistochemistry. The Immunoscore was developed by Jérôme Galon, Ph.D. [Institut National de la Santé et de la Recherche Médicale (INSERM)] and his colleagues for use in studies of colorectal cancer. According to Dr. Galon’s findings, use of checkpoint inhibitors is the logical strategy for patients with high Immunoscores. In contrast, for patients with low Immunoscores, effective immuno-oncology treatments will need to focus on getting immune cells into the tumor in the first place (e.g., by treatment with a “second-wave” immunotherapy agent) before checkpoint inhibitors can be used.

Genetic and immunological biomarkers may be combined with target biomarkers and other parameters to move toward better discrimination between patients who are likely to benefit from checkpoint inhibitor treatments and those who are not. Specifically, biomarkers can be used to discriminate between “cold” and inflamed tumors. Genetic and immunological biomarkers can also be used to design therapies that can turn “cold” tumors into inflamed tumors, thus improving responses to checkpoint inhibitor therapy and other immunotherapies. For example, these biomarkers might be used to design combinations of treatments that induce immune infiltration of tumors with checkpoint inhibitors that activate or reactivate infiltrating immune cells, such as TILs. Novel agents that might induce immune infiltration of tumors are discussed in several chapters of this report.

More immediately, combination therapies involving the use of older treatments or agents, followed by administration of checkpoint inhibitors, are under clinical investigation to determine whether any of these older agents might render “cold” tumors inflamed, making them susceptible to checkpoint inhibitor therapy. Among these older treatments (discussed in Chapter 2) are radiation therapy (especially stereotactic body radiation therapy (SBRT), targeted therapies, and cytotoxic chemotherapies.

Approved and clinical-stage immunotherapy biologics other than checkpoint inhibitors

Various chapters of this report focus on approved and clinical-stage biologics other than the checkpoint inhibitors. Most of these agents may be used as “immuno-oncology 2.0” agents, i.e., agents that promote T-cell infiltration of tumors, thus rendering them susceptible to successful treatment with checkpoint inhibitors.

In addition to serving as an introduction to the report as a whole and discussing the early history of cancer immunotherapy, Chapter 1 focuses on cytokines as cancer immunotherapeutics. Interleukin-2, interferon-alpha-2a, and interferon alpha-2b have long been approved for treatment of various cancers. To this day, despite the introduction of newer immunotherapies, such as checkpoint inhibitors, high-dose recombinant IL-2 (Novartis/Prometheus Laboratories’ Proleukin) is the only drug so far that has produced durable, long-term responses in patients with metastatic melanoma or metastatic renal cell carcinoma. According to Patrick Ott, M.D., Ph.D. (Dana-Farber Cancer Institute, Boston, MA), “High-dose IL-2 has a track record of patients who have been disease-free for 20 years, and we just don’t know that yet with the new drugs [such as checkpoint inhibitors].”  In the case of advanced melanoma, high-dose intravenous bolus IL-2 induces objective clinical responses in 15–20% of patients and durable complete responses in 5–7% of these patients. For metastatic RCC (mRCC), high-dose intravenous bolus IL-2 gives an objective clinical response rate of approximately 25% and a 7% durable complete response rate.

However, high-dose IL-2 has a significant degree of toxicity. Because of its adverse effects, high-dose IL-2 therapy for cancer requires an expert, experienced team of clinicians and specialized centers. Under such conditions of care, IL-2-related toxicity can usually be easily managed. Despite its logistical disadvantages, several investigators are attempting to revive use of IL-2 in cancer immunotherapy.

In the current era, it is possible to use targeted therapies as salvage agents to treat patients who do not do well with IL-2. There may also be opportunities to develop combination therapies of IL-2 with radiation or checkpoint inhibitors, and clinical trials of these combination therapies are underway. IL-2 treatment also requires only a median of one month of therapy and gives a long duration of benefit without the need for additional treatment. This is not true, for example, for treatment with cytotoxic therapies or targeted therapies. In addition to its use as a stand-alone drug, IL-2 is also used as part of certain cellular immunotherapies.

Other, newer cytokine-based therapies discussed in Chapter 1 are in early-stage clinical trials. Notably, local intratumoral electroporation of a DNA plasmid that encodes human IL-12 (pIL-12) (OncoSec’s ImmunoPulse) can result in systemic responses in metastatic melanoma patients. This procedure appears to induce TILs and anti-tumor immunity in both the injected tumors and in distant tumor sites. A Phase 2 trial indicates that intratumoral pIL12-electroporation therapy may prime systemic responses for checkpoint inhibitor blockade, apparently by generation of CD8+ TILs. ImmunoPulse therapy followed by treatment with a checkpoint inhibitor is therefore a potential immuno-oncology 2.0 therapy for metastatic melanoma. Other early-stage potential immuno-oncology 2.0 therapies discussed in Chapter 1 are based on the cytokines IL-10 and IL-15.

In addition to discussing approved and clinical-stage checkpoint inhibitors and their mechanisms of action, Chapter 2 includes discussions of clinical-stage checkpoint inhibitor modulators, such as LAG-3 (lymphocyte-activation gene 3) inhibitors, TIM-3 (T-cell immunoglobulin and mucin-domain containing-3) inhibitors, small-molecule IDO (indoleamine 2,3-dioxygenase) pathway inhibitors, and a small-molecule PI3Kγ (phosphoinositide 3-kinase gamma) inhibitor. These agents are in Phase 1 or Phase 2 development. In general, they work to overcome immunosuppression and/or T-cell exhaustion, and thus may overcome blocks to T-cell activation by checkpoint inhibitors.

Chapter 3 focuses on immune agonists. Immune agonist therapeutics—most of which are mAbs—target specific cell surface proteins on T cells, resulting in stimulation of T cell activity. This mechanism contrasts with that of checkpoint inhibitors, which are designed to overcome blockages to T cell activity mediated by immune checkpoints. Companies are developing immune agonist immunotherapeutics principally for use in combination with checkpoint inhibitors (i.e., as immuno-oncology 2.0 agents). All of the agents discussed in this chapter are in early-stage clinical trials.

Chapter 4 discusses bispecific antibody (bsAb) cancer immunotherapeutics. A bispecific Ab (bsAb) is a type of mAb. bsAbs are designed with two different variable domains that enable the Ab to bind simultaneously to two different types of targets. bsAbs used in cancer immunotherapy usually bind one target on a tumor cell and another target on a cytotoxic immune system cell, bringing the two types of cells into close proximity. This allows the immune system to act against the tumor cell.

There are currently two approved and marketed bsAb cancer immunotherapeutics—catumaxomab (Neovii Biotech’s Removab) and blinatumomab (Amgen’s Blincyto). Catumaxomab is a rat-mouse hybrid bsAb that targets the tumor antigen epithelial cell adhesion molecule (EpCAM) and the T-cell surface molecule CD3. It is approved in Europe for treatment of malignant ascites in patients with EpCAM-positive cancer if a standard therapy is not available. Blinatumomab is a murine Ab-derived small bsAb that targets CD3 on T-cells and CD19 on B-cell neoplasms. It is approved under the FDA’s accelerated approval program for treatment of adults with Philadelphia chromosome-negative (Ph-) relapsed or refractory B-cell ALL. It has also been granted conditional marketing authorization in the European Union for the treatment of adults with Ph- relapsed or refractory B-cell precursor ALL. Blinatumomab is the first anti-CD19 drug to receive FDA approval. As discussed in Chapter 6, the most advanced CAR T-cell (genetically engineered T cells bearing chimeric antigen receptors) therapies in development target CD19 and are intended for treatment of CD19+ B-cell leukemias and lymphomas. However, none of these cellular therapies is yet approved.

As with other first-generation bsAbs, blinatumomab is unstable and has a short serum half-life. It must be administered as a continuous intravenous infusion over a minimum of 4 weeks. The infusion can be administered to outpatients via a minipump system. The short half-life of the agent is not due to the formation of antimouse Abs. Researchers and companies are working on technology platforms that can yield more stable bsAbs.

Some companies and researchers are attempting to develop bsAb agents as alternatives to CAR T-cell therapies. These companies see bsAbs as potentially safer than CAR T cells. Moreover, bsAbs can be manufactured like other Ab-based biologics and then administered to patients with cancers specifically targeted by the bsAb. There is thus no need to isolate autologous T cells as the basis for a patient’s therapy or to develop and carry out individualized manufacturing protocols, as in the case of CAR T cells. Except for catumaxomab and blinatumomab, all bsAb candidates are in Phase 1 trials. No efficacy findings have been published. Thus, it is not known whether any of these agents is likely to be approved, and it is too early to tell whether bsAbs can serve as an alternative to CAR T-cell therapies.

Key issues in the bsAb area include moving away from rodent-derived Abs toward human or humanized molecules, and developing agents with improved stability and half-life in the circulation. Developers of bsAbs need to develop improved technologies aimed at producing more stable molecules. Also important are the ability to easily design and manufacture bsAbs and to rapidly redesign improved versions of agents entered into first-in-humans trials.  Based on these issues, companies have developed several platform technologies for producing bsAbs, which are discussed in Chapter 4.

Chapter 5 focuses on therapeutic anticancer vaccines and oncolytic viruses. Efforts to develop therapeutic cancer vaccines began in the 1990s. However, the cancer vaccine field has been characterized by a long series of clinical failures, beginning in the 1990s and continuing to the present day. Researchers have been working to overcome these difficulties by using improved research methodologies and by employing advances in our understanding of immuno-oncology. In the current era, researchers have used their greater understanding of dendritic cell biology to attempt to improve the design of peptide and protein vaccines. Another important factor that has contributed to the failure of cancer vaccines is immune suppression in the tumor environment. Some researchers hypothesize that one way to overcome this immunosuppression is to administer vaccines in combination with checkpoint inhibitors.

More recently, researchers have been investigating the mechanisms by which the immune system—especially TILs—recognize tumor cells and differentiate them from noncancer cells. These studies have focused on “neoantigens”—i.e., antigens that are specific for cancer cells, as opposed to normal, noncancer cells. These neoantigens are associated with somatic mutations that arise in the evolution of tumor cells. TILs—especially CD8+ intratumoral T cells—may mediate tumor regression, and this antitumor activity may be enhanced by checkpoint inhibitor therapy. Researchers therefore hypothesize that neoantigen-based vaccines may be more effective than earlier types of cancer vaccines. They further hypothesize that determination of neoantigens in tumors may provide technology platforms for the design of effective cancer vaccines.

As discussed in Chapter 5, there are currently one approved and marketed therapeutic cancer vaccine and one approved and marketed oncolytic virus therapeutic. Sipuleucel-T (Dendreon/Valeant’s Provenge) is a personalized dendritic cell vaccine that targets prostatic acid phosphatase. It was approved by the FDA for treatment of asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer in 2010 and was approved in Europe for the same indication in 2013. It is the first approved therapeutic cancer vaccine. However, sipuleucel-T has an apparently minimal effect and is very expensive. It also faces significant competition from small-molecule agents. As a result, sipuleucel-T sales—and Dendreon the company—have faltered. Dendreon filed for filed for Chapter 11 protection in 2014, and its assets were sold to Valeant in 2015. As of May 18, 2016, Valeant was looking to sell off the “laggard” Provenge, in order to ameliorate its own debt problems. Therefore, the future of sipuleucel-T remains in doubt.

In 2015, the FDA approved talimogene laherparepvec (Amgen’s Imlygic, also known as T-Vec), an oncolytic herpes simplex virus that expresses granulocyte macrophage colony-stimulating factor (GM-CSF). It is approved for the treatment of melanoma lesions in the skin and lymph nodes. This oncolytic virus is injected into a single tumor, where it lyses tumor cells.  It was postulated that upon lysis of tumor cells in the treated lesion, systemic immune responses are induced. There was evidence for induction of immune responses in distant tumor sites in a published Phase 1 study, and there was a trend toward improved overall survival in early results of a Phase 3 study presented at the 2013 ASCO Annual Meeting. However, as determined by completed results of the same Phase 3 study, the agent was not subsequently shown to improve overall survival or have an effect on distant metastases.

Strictly speaking, T-Vec is not a cancer vaccine, since it carries no tumor antigen in its molecular structure. It is designated as an oncolytic virus therapy and is the first such agent to receive FDA approval.  However, since it had been designed to immunize patients systemically against multiple tumor antigens, it is generally considered a “cancer vaccine” in scientific and market research reports.

Amgen and Merck—together with their academic collaborators—have been conducting clinical studies of a combination of the oncolytic virus therapy talimogene laherparepvec and the checkpoint inhibitor pembrolizumab. The results of a Phase 1b study presented at the 2016 ASCO Annual Meeting appeared promising. The combination therapy is now under investigation in a Phase 3 clinical trial. Other researchers are conducting clinical trials of combinations of T-Vec with another checkpoint inhibitor, ipilimumab.

Chapter 5 includes discussions of several therapeutic cancer vaccines and one oncolytic virus therapeutic that are in clinical development (from Phase 1 to Phase 3). Phase 3 vaccines include Bavarian Nordic’s PROSTVAC-VF and Argos Therapeutics’ AGS-003. PROSTVAC-VF is being developed for treatment of asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer, the same indication as sipileucel-T. It targets prostate specific antigen (PSA). PROSTVAC-VF is a recombinant poxvirus agent carrying genes for PSA and three costimulatory molecules. Unlike sipileucel-T, it is an “off-the-shelf” immunotherapeutic that can be manufactured as a biologic product. Phase 2 data suggests that PROSTVAC-VF may have superior efficacy compared to currently approved agents.

Argos’ AGS-003 is a personalized dendritic cell vaccine that is under investigation in Phase 3 clinical trials in patients with mRCC. To produce the vaccine, autologous dendritic cells are loaded with mRNA from the patient’s mRCC tumor. Some of the tumor cell mRNAs encode tumor antigens that are expressed on the surfaces of the dendritic cells (in the context of MHC class II molecules); these cells also express CD40. The tumor mRNA-programmed autologous dendritic cells constitute the vaccine, which is then formulated into an intradermal injection for administration to the patient.  The dendritic cells serve as antigen-presenting cells (APCs) that activate T cells specific for the patient’s tumor antigens. The interaction of CD40 with its ligand on activated T cells induces secretion of IL-12 by the dendritic cells. IL-12 promotes the differentiation of T cells into T helper 1(Th1) cells. The Phase 3 trial of AGS-003 is a follow-up to favorable Phase 2 data that was presented at the 2024 ASCO Annual Meeting.

Notably, several early-stage therapeutic cancer vaccine programs discussed in Chapter 5 are based on research designed to identify naturally processed Class I and Class II peptide epitopes for formulation into vaccines. TapImmune is developing two vaccines based on research at the Mayo Clinic designed to identify such epitopes from specific tumor antigen proteins, in one case HER2/neu (in HER2/neu+ breast cancer) and in the other case TPIV200 folate receptor alpha (in ovarian cancer and triple negative breast cancer). Neon Therapeutics’ NEO-PV-01 is based on research at the Dana-Farber Cancer Center and the Broad Institute to identify naturally-processed neoantigen peptide epitopes from patient tumors using exome sequencing and predictive algorithms for MHC binding. Unlike the TapImmune products, this research is designed to identify these epitopes from multiple tumor antigen proteins. Neon is one of several recently formed cancer vaccine companies discussed in Chapter 5 that are focused on neonatigen science and technology.

Among companies exploring neoantigen vaccines, the only other one that has reached the clinical stage as of 2016 is the German firm BioNTech AG. BioNTech specializes in mRNA-based neoantigen vaccines, as opposed to peptide vaccines. The company presented an abstract describing the design of the still-ongoing Phase 1 clinical trial of its IVAC (individualized vaccine) Mutanome in advanced melanoma patients at the 2015 American Association for Cancer Research (AACR) Annual Meeting. In September 2016 BioNTech entered into a collaboration with Genentech to develop, manufacture, and commercialize novel mRNA-based, neoantigen/neoepitope cancer vaccines based on the IVAC Mutanome technology platform. Other neoantigen vaccine companies mentioned in Chapter 5 that have not yet reached the clinic are Gritstone Oncology, ISA Pharmaceuticals, Agenus, and Caperna (a Moderna venture company).

Imuno-oncology 2.0 strategies now feature prominently in therapeutic cancer vaccine and oncolytic virus therapeutic development. This is particularly true with neoantigen vaccines, which are designed to induce neoantigen-specific TILs within patients’ tumors. The strategy is to use the vaccine to induce the TILs, rendering checkpoint inhibitors effective in inducing tumor regression. Based on this strategy, Neon’s NEO-PV-01 vaccine is being combined with poly-ICLC adjuvant and Bristol-Myers Squibb’s PD-1 immune checkpoint inhibitor nivolumab in collaborative clinical studies in advanced melanoma, smoking-associated NSCLC, and bladder cancer.

The immuno-oncology 2.0 strategy has become a general theme of cancer vaccine research and development beyond neonantigen vaccines—use cancer vaccines to render tumors inflamed, and use checkpoint inhibitors to induce regression of the inflamed tumors. Thus, there are several examples of trials of cancer vaccine/checkpoint inhibitor combinations discussed in this chapter. In some cases, cancer vaccines are being tested in combination with checkpoint inhibitors in Phase 1 or Phase 2 clinical trials, rather than the “traditional” approach of first getting a vaccine approved and then conducting trials of the vaccine in combination with other agents. It is possible that testing a vaccine in combination with a checkpoint inhibitor in early stage clinical trials may reduce the number of failed cancer vaccines. However, whether this is true remains to be seen.

Chapter 6 focuses on the use of adoptive T-cell immunotherapies for cancer. In this class of therapies, autologous or syngeneic activated T cells (which may or may not be genetically modified) are infused into patients in order to attack their cancers. Adoptive immunotherapy is also known as adoptive cell transfer (ACT).

Immunotherapy with TIL cells

ACT was pioneered by Dr. Steven A. Rosenberg of the National Cancer Institute (NCI). The major focus of the Rosenberg group has been TIL therapy, which involves isolation of TILs (which are really tumor-infiltrating T cells) from a patient’s tumor, followed by ex vivo expansion of these cells with IL-2. The TILs and high-dose IL-2 are then infused intravenously into the patient. The infused T cells traffic to tumors and can mediate their destruction. Clinical study of TIL therapy has continued to the present day. Moreover, attempts to overcome some of the limitations of TIL therapy have resulted in the development other newer forms of ACT based on various types of genetically engineered T cells.

Most studies with TIL therapy have been in advanced melanoma. The Rosenberg group presented the results of a Phase 2 trial of TIL therapy in stage 4 metastatic melanoma patients who had received preparative lymphodepletion therapy at the 2014 Annual Meeting of the American Association for Cancer Research (AACR). At the time of analysis, 11 of 101 patients had achieved a complete response, and 44 had achieved a partial response, for an overall objective response rate (ORR) of 54%. This ORR was superior to recent results with ipilimumab and with anti-PD-1 therapies in Stage 4 metastatic melanoma patients. TIL therapy also produced clinically meaningful objective response rates in 19 of 45 (ORR 42%) patients who had progressed through ipilimumab treatment and 5 of 10 (ORR 50%) patients who had progressed through anti-PD-1 therapy. Four patients who had failed on checkpoint inhibitor treatment had complete responses and continued to be disease-free.

Despite its apparent high degree of efficacy and curative potential, TIL therapy cannot be given to all advanced melanoma patients. Patients must have resectable tumors from which sufficient numbers of TILs can be cultured and expanded in vitro. Dr. Rosenberg and his colleagues estimated that approximately 45% of advanced melanoma patients could receive TIL therapy.

Dr. Rosenberg and his colleagues have continued to investigate the mechanistic basis of successful TIL therapy for advanced melanoma, especially with respect to neoantigen science. Melanoma is an “immunogenic cancer” that exhibits many more somatic mutations (about 200 nonsynonymous mutations per tumor) than most cancers. Researchers hypothesize that this increased mutation load may be due to the role of a potent immunogen—ultraviolet light—in the pathogenesis of melanoma.

Dr. Rosenberg and his colleagues wished to investigate the role of targeting specific mutations in TIL therapy in producing durable complete remissions. They published a case study in 2013 of a melanoma patient who had been enrolled in a clinical trial of TIL therapy who experienced a durable complete remission and remained disease-free seven years subsequent to TIL treatment. In addition to establishing the endogenous TIL cell line that was used to treat the patient, the researchers established a tumor cell line from the resected tumor that was used to produce the TILs. Using these materials, the researchers identified the predominant antigen target recognized by the TIL cell line. This antigen was recognized in the context of the MHC class I molecule HLA-A*01. It was a mutant form of protein phosphatase 1 regulatory subunit 3B (PPP1R3B). The DNA from the patient’s melanoma cell line contained a somatic mutation that resulted in the substitution of Phe for the Ser residue at amino acid position 16 of the wild type PPP1R3B protein. This mutated PP1R3B was the immunodominant epitope recognized by the patient’s TIL cells. Five years after this patient’s TIL therapy, peripheral blood T cells obtained from the patient specifically recognized the mutated PPP1R3B epitope. These results suggest that adoptive T-cell immunotherapy targeting a tumor-specific antigen can mediate long-term survival in at least some metastatic melanoma patients.

Based on this case study, the researchers hypothesized that TILs that target immunodominant tumor-specific mutations in genes whose products play essential roles in carcinogenesis may result in durable complete remissions. The results presented in this 2013 study are the most direct evidence to date showing that targeting a tumor-specific, mutated antigen via TIL therapy can mediate a complete, durable regression of advanced melanoma. This study provided a rationale for identifying tumor-specific mutations that may serve as targets for personalized adoptive T-cell therapy.

The Rosenberg group went on to apply the strategy used in the mutation-specific 2013 study in melanoma to an epithelial cancer. (Epithelial cancers, such as lung, breast, colorectal, and pancreatic cancer, constitute over 80% of all human malignancies. Except for lung cancer in smokers, they are not considered to be immunogenic cancers.) In a 2014 study of TIL therapy of an individual patient with widely metastatic cholangiocarcinoma, the researchers found that 25% of the T cells in a TIL population that was used to successfully treat the patient were CD4+ T cells that recognized a single immunodominant mutation. These T cells recognized the mutation in an MHC class II-restricted manner. Following the subsequent progression of her disease, the patient was re-treated with a TIL culture that was highly enriched (95%) in the CD4+ T cells that were reactive to the immunodominant mutation. The patient experienced long-term tumor regression. Although she still has stage 4 cancer, she remains alive 5 years after her first TIL treatment. However, in 2016, after metastases in her lungs began to grow again, she had surgery and was given a checkpoint inhibitor, which began shrinking the tumors.

Although only this single patient in the Rosenberg group’ clinical study of TIL therapy of metastatic gastrointestinal cancers experienced successful treatment (in terms of becoming a long-term survivor), TILs derived from other patients’ tumors also included T-cell subpopulations targeting specific somatic mutations expressed by their cancers. Notably, CD8+ TILs isolated from a colon cancer tumor of one patient recognized a mutation in KRAS (Kirsten rat sarcoma viral oncogene homolog), known as KRAS G12D. This mutation results in an amino acid substitution at position 12 in KRAS, from glycine (G) to aspartic acid (D). KRAS G12D is a driver mutation that is involved in causation of many human cancers and is generally considered “undruggable.” The TILs from this patient were CD8+ T cells that recognized KRAS G12D in the context of the class I human leukocyte antigen (HLA) allele HLA-C*08:02. Although two other patients had tumors that expressed KRAS G12D, the researchers did not detect TILs that recognized the KRAS mutation and concluded that KRAS G12D was not immunogenic in these patients. The two patients for whom KRAS G12D was not immunogenic did not express the HLA-C*08:02 allele.

The results seen with KRAS G12D-expressing tumors suggest the possibility of constructing genetically engineered CD8+ T cells that express a TCR that is reactive with the KRAS mutation in the context of the HLA-C*08:02 allele. The KRAS G12D driver mutation is expressed in approximately 45% of pancreatic adenocarcinomas, 13% of colorectal cancers, and at lower frequencies in other cancers. The HLA-C*08:02 allele is expressed by approximately 8% and 11% of white and black people, respectively, in the U.S. Thus, in the U.S. alone, thousands of patients per year with metastatic gastrointestinal cancers would potentially be eligible for KRAS G12D-reactive T cell immunotherapy.

In December 2016, the Rosenberg group published an article in the New England Journal of Medicine describing the case of one patient with metastatic colorectal cancer who was successfully treated with TILs targeting KRAS G12D. The researchers identified a polyclonal CD8+ T-cell response directed against KRAS G12D in TILs obtained from a lung metastasis from this patient. After preparative lymphodepletion, they treated the patient with IL-2 and approximately 1.11 × 1011 TILs, which consisted of HLA- C*08:02-restricted TILs comprising four different T-cell clonotypes that specifically targeted KRAS G12D. The researchers observed objective regression of all seven lung metastases after the TIL therapy. However, one of the lesions was found to have progressed on evaluation 9 months after therapy. The lesion was surgically removed and was found to have lost the chromosome 6 haplotype encoding the HLA-C*08:02 class I MHC molecule. Loss of expression of this protein enabled the tumor lesion to evade the immunotherapy, illustrating the importance of HLA restriction in T-cell tumor targeting. Since removal of the lesion, the patient has been disease-free for over eight months as of December 2016.

Because this study was conducted only in one patient, the researchers consider it a “proof-of-principle” study that needs to be replicated in other patients. However, the Rosenberg group has identified multiple T cell receptors that recognize KRAS G12D. As discussed earlier, this finding may enable researchers to design genetically engineered T cells that recognize the mutant protein and could be used in therapies targeting multiple cancers that express it.

Because KRAS G12D is an “undruggable” target that causes thousands of cases of cancer every year, the Rosenberg studies on targeting this mutation with adoptive cellular immunotherapy have the potential to help a large number of patients. Constructing genetically engineered T cells that are reactive to KRAS G12D in an MHC-restricted manner is a possible strategy to develop treatments for cancers caused by this KRAS driver mutation.

Recently, the Rosenberg group has continued to refine its TIL studies to more precisely target neoantigens in patient tumors and to treat new types of cancers (e.g., gastrointestinal cancers, NSCLC, KRAS-driven cancers) that had not been possible to treat using TIL therapy, and in some cases, using any immunotherapy. Another goal of the Rosenberg group is to develop therapies that combine TIL treatment with immune checkpoint therapy to yield results that represent improvements over treatments with either therapy alone. Finally, the Rosenberg group’s studies might be used to design genetically engineered T cells that address driver mutations, such as KRAS G12D, in the context of specific HLA alleles for treatment of tumors that express these mutant protein-HLA combinations without the need to isolate TILs.

Commercialization of TIL therapy

Lion Biotechnologies (San Carlos, CA) has been pursuing commercialization of TIL therapy. It has been working with Dr. Rosenberg and his colleagues at the NCI under a Cooperative Research and Development Agreement (CRADA) to develop and commercialize this therapy. Under this CRADA, Lion will gain exclusive rights to new adoptive cell therapy technologies for the treatment of metastatic melanoma, along with access to all clinical data, manufacturing data, and standard operating procedures. The company may also conduct clinical trials at the NCI.

In August 2016, Lion announced an amendment of its CRADA with the NCI to extend the agreement for an additional five-year term until 2021. Under the extended CRADA, the company will continue working with Dr. Rosenberg to develop non-genetically altered TILs as a stand-alone therapy or in combination with FDA-licensed products routinely used for treatment of the indications covered under the CRADA. The CRADA now includes the development of TIL therapy for treatment of metastatic melanoma, bladder, lung, breast, and human papillomavirus (HPV)-associated cancers. This includes exploration of combinations of TIL with FDA-approved checkpoint inhibitors. Lion’s development of commercial TIL therapies builds on studies in Dr. Rosenberg’s group at the NCI, which are considered Phase 2 clinical trials.

In parallel with trials in progress at the NCI and at a Lion Biosciences partner, the Moffitt Cancer Center (Tampa, FL), Lion is now conducting and recruiting patients for its own first Phase 2 clinical trial in metastatic melanoma of an autologous TIL therapy, designated LN-144.  The trial is designed to assess the safety, efficacy, and feasibility of the autologous TIL therapy, followed by IL-2, in the treatment of patients with refractory metastatic melanoma. Lion also plans to initiate TIL-therapy studies in cervical and head-and-neck cancers during 2017. The TIL populations to be used for these studies, LN-145, will be selected for reactivity to HPV proteins E6 and E7. The selection and use of these TIL populations were developed by researchers in Dr. Rosenberg’s group.

Lion is pursuing its TIL therapy development efforts in collaboration with several partners. These include the Moffitt Cancer Center (Tampa, FL) as mentioned above, the Karolinska University Hospital (Stockholm, Sweden) for studies with the Karolinska’s PolyBioCept AB, as well as with MedImmune/AstraZeneca for studies of combination therapies with TILs and MedImmune’s investigational PD-L1 inhibitor durvalumab. The NCI and Moffitt are pursuing clinical studies of combination therapies of TILs with approved checkpoint inhibitors (ipilimumab, pembrolizumab, and nivolumab); any therapies that result from these studies will be licensed to Lion. Lion also has an agreement with Lonza (Allendale, NJ) for process development and scale-up and has a manufacturing services agreement with WuXi AppTec, Inc. U.S. Business Unit (Philadelphia, PA).

In 2013, stock analysts at Seeking Alpha estimated peak sales for Lion’s TIL therapy for metastatic melanoma at approximately $1 billion. Since Lion is developing other therapies as well, the long-term potential for the company may be much greater than that. However, since the most advanced TIL-based therapies are only in Phase 2, this valuation is speculative. Nevertheless, Lion Biotechnologies represents a potentially important effort towards commercializing TIL therapies.

Adoptive immunotherapy with genetically engineered T cells bearing chimeric antigen receptors (CARs)

CAR T cells are the most advanced genetically engineered T cells in development for use in adoptive immunotherapy. These are T cells that have been engineered with retroviral vectors carrying chimeric antigen receptors (CARs). CARs are synthetic, engineered receptors designed to target molecules in their native conformations on the surfaces of tumor cells. Unlike TCRs, CARs bind to their target molecules independent of antigen processing by the target cell and independent of MHC restriction. T cells produced for use in autologous CAR-based therapy are usually derived from peripheral blood T cells from the patient undergoing treatment. Thus, CAR T-cell therapy does not require isolating and culturing T cells from tumors.

CARs typically engage their target molecules via single-chain variable fragments (scFvs) derived from an antibody. These can be produced in bacterial host cells and selected via phage display technology. Most researchers use retroviral or lentiviral vectors to insert the DNA that encodes for the CAR into T cells. A CAR construct typically includes an extracellular scFv domain, a transmembrane domain, and an intracellular portion containing one or two costimulatory domains, terminating with CD3z, an intracellular signaling domain that couples antigen recognition by the scFv domain to specific intracellular signal-transduction pathways involved in T-cell activation. “Second-generation” CARs, which represent the class of CARs currently in development, possess a single costimulatory domain located between the transmembrane domain and CD3z. The purpose of the added costimulatory domain is to activate additional intracellular pathways involved in T-cell activation and cytokine production.

As discussed in Chapter 6, the most advanced CAR T-cell therapies currently under development are directed against CD19+ B-cell leukemias and lymphomas. C19 is a cell surface protein that is expressed on all malignant and normal B cells. Notably, Biologics License Applications (BLAs) for two CD19-targeting CAR T-cell therapies, Novartis’ CTL019 and Kite Pharma’s KTE-C19 (axicabtagene ciloleucel) are expected to be filed with the FDA in early 2017. CTL019 will be indicated for treatment of relapsed/refractory B-cell acute lymphocytic leukemia, and KTE-C19 will be indicated for treatment of patients with relapsed/refractory aggressive B-cell non-Hodgkin’s lymphoma who are ineligible for autologous stem cell transplant.

In several cases, these therapies have shown impressive results. For example, in a study with CTL019 reported at the American Society of Hematology (ASH) 2014 Annual Meeting, of 39 pediatric patients with acute lymphoblastic leukemia (ALL), 36 (92%) achieved complete remission, and responses have been durable. At a median follow-up of 6 months, some patients had experienced a remission of 1 year or more. The 6-month event-free survival was 70%, and the overall survival was 75%. Some pediatric patients who have not responded to other therapies achieve extended complete remissions with CTL109. Moreover, when CTL109 CAR T cells are infused into patients, they persist for long periods of time—up to 31 months in ongoing responders. Expansion, proliferation, and persistence are accompanied by B-cell aplasia, i.e., the absence of normal B cells. This is a pharmacodynamic marker of CTL019 persistence and function and is due to the fact that normal B cells are positive for CD19, and are thus targeted by CTL109 (as well as other CD19-targeting CAR T-cell therapies). B-cell aplasia is managed via intravenous immunoglobulin replacement therapy, which supplies the circulating antibodies that the patient’s B cells are no longer present to make.

A major safety issue with CD19-targeting CAR T-cell therapies is cytokine release syndrome (CRS). Treatment with CAR T-cell therapies can result in high levels of cytokines in vivo, especially in patients with high tumor burdens. High-level cytokine release occurs concurrently with lysis of large amounts of tumor cells in these patients, and also coincides with peak T-cell expansion. These features are associated with the ability of CAR T cells to give significant tumor responses without the need for administering exogenous IL-2, as with TILs. CRS can be dangerous and life-threatening. Episodes of CRS rapidly resolve after supportive care or, for more severe cases, tocilizumab and corticosteroids. Physicians who treat patients with CAR T-cell therapies, however, need to be aware of CRS and be prepared to monitor and treat it.

As Novartis and Kite have moved toward market entry for their lead anti-CD19 CAR T-cell therapies, the third company in the race, Juno, has faltered. Its lead product, JCAR015, has seen major difficulties in its Phase 2 Rocket trial. The study was assessing JCAR015 in adult patients with relapsed or refractory B cell ALL. However, there were several patient deaths due to cerebral edema. After making changes in its preconditioning regimen that it believed might rectify the problem, two more cases of fatal cerebral edema occurred. Juno has voluntarily suspended the trial. Thus, the company’s planned 2017 market entry will not occur. However, Juno has other CD19-targeting CAR-T agents in development, including JCAR017, as well as the CD22-targeting agent JCAR018. These have been in Phase 1 clinical trials.

bluebird bio is developing the CAR T-cell therapy bb2121 for multiple myeloma. bb2121 is in development in collaboration with Celgene, which exercised its option for an exclusive license to the agent via a $10 million option exercise payment. bb2121 targets B-cell maturation antigen (BCMA), a protein that is involved in B-cell development.  This agent is in Phase 1 trials.

Several companies and their academic collaborators are attempting to develop CAR T-cell therapies that target solid tumors. Researchers from the University of Pennsylvania successfully manufactured and infused a version of CARTmeso (a CAR T-cell therapy that targets mesothelin-positive tumors) in six patients with refractory pancreatic cancers. Although promising clinical and radiologic signs were seen in this study, they were transient. The researchers hypothesized that combination therapies designed to work with CARTmeso therapy to give more prolonged and persistent antitumor activity will be needed. Some studies of this type are in progress in patients with various types of mesothelin-positive tumors. Meanwhile, researchers from the University of Pennsylvania continue to work to identify the therapeutic barriers that might prevent effective therapy of solid tumors with the CAR T-cell platform.

Researchers at Kite Pharma and the NCI and at the University of Pennsylvania have been attempting to develop autologous CAR T-cell therapies that target the epidermal growth factor receptor variant III (EGFRvIII) for treatment of glioblastomas that express this variant (a mutated form of EGFR found in 20-30% of glioblastoma tumors). Two EGFRvIII CAR T-cell therapies are in Phase 2 and Phase 1 clinical trials, respectively. The two groups are using cells with different types of CAR constructs in their studies. The results of the two trials may thus inform the future development of CAR therapies for solid tumors.

At least two companies are attempting to develop engineered improvements in CAR T-cell therapy. Bellicum’s GoCAR-T technology involves construction of CAR T cells that incorporate a proprietary inducible activation switch, known as iMC. GoCAR-T is designed to support persistence of CAR T cells in the body in the absence of cancer antigen to allow for continuing anti-tumor surveillance. This technology is the basis for BPX-601, A CAR T-cell therapy that targets prostate stem cell antigen (PSCA). BPX-601 is under evaluation in a Phase 1 study in patients with non-resectable pancreatic cancer.

The activation switch iMC serves as a costimulation domain, as in other CARs. However, iMC is an inducible domain, which requires the presence of the small molecule inducer rimiducid (also known as AP1903) for its activation. The iMC switch allows control of T-cell survival in the absence of antigen signaling and promotes full activation and proliferation only in the presence of antigen. In the event of side effects, the level of activation of the GoCAR-T cells may be attenuated by reducing the rimiducid administration schedule. In preclinical studies, Bellicum researchers showed that with the GoCAR-T technology, the presence of rimiducid and antigen results in upregulation of cytokines, including IL-2, resulting in T-cell proliferation, persistence, and improved anti-tumor efficacy.

In addition to the GoCAR-T technology, Bellicum has also developed CAR T-cell therapies that incorporate a proprietary “safety switch” based on an inducible caspase-9 suicide gene system (a component of an apoptotic pathway). The safety switch, which is also induced by AP1903/rimiducid, is known as CaspaCIDe. It is designed to eliminate cells in the event of toxicity. In this case, administration of rimiducid triggers apoptosis of the CAR T-cells that incorporate CaspaCIDe.

The other company developing engineered improvements in CAR T-cell therapy is Cellectis. This company has been developing technologies for the design and manufacture of “off-the shelf” allogeneic (rather than autologous) CAR T cells. Cellectis creates its CAR T cells by starting with T cells from healthy donors, rather than from cancer patients. It then edits the genomes of these cells with its proprietary transcription activator-like effector nuclease (TALEN) technology. As with other gene editing technologies, TALEN involves the use of “genetic scissors” capable of making double-strand (DS) breaks at a specific locus in the DNA. It then recruits the cell’s DNA repair machinery to repair the break in such a way as to disable the gene of interest, while restoring the integrity of the DS DNA structure.

Cellectis has used its gene editing technology to develop a process for the large-scale manufacturing of T cells deficient in expression of both the T-cell receptor (TCR) and CD52, a protein targeted by alemtuzumab. Alemtuzumab is a chemotherapeutic agent that can be used for lymphodepletion to support engraftment of infused CAR T cells. The researchers used gene editing to functionally eliminate the genes coding for the TCR and for CD52. T cells modified using this process do not mediate graft-versus-host reactions (because they lack the TCR) and are rendered resistant to destruction by alemtuzumab. These characteristics enable researchers to use alemtuzumab administration concurrently with or prior to administration of engineered T cells, thus supporting their engraftment. Transfection of the TALEN-engineered cells with a vector encoding CD19 CAR endowed them with the ability to efficiently destroy CD19+ tumor targets. The results of this study demonstrate the ability of Cellectis researchers to utilize TALEN-mediated genome editing to enable the manufacture of third-party CAR T-cell immunotherapies against various targets. These allogeneic CAR T-cell immunotherapies can thus be used in an "off-the-shelf" manner, as with most biologics.

Cellectis calls its lead immuno-oncology product candidates UCARTs (universal chimeric antigen receptor T cells). These are “off-the-shelf” allogeneic CAR T-cells. Their production can be industrialized and standardized, over time and from batch to batch, with consistent pharmaceutical release criteria, as with other biologics. Similar to conventional CAR T cells, they are engineered for treating patients with a particular cancer type bearing a particular tumor antigen.

The only Cellectis product candidate currently in the clinic is UCART19, which targets CD19. This agent is in ongoing Phase 1 clinical trials in various types of B-cell leukemias. At the 2015 ASH Annual Meeting in December 2015, Cellectis collaborators presented encouraging data from a first-in-humans study of UCART19. In June 2015, the researchers treated a young leukemia patient with allogeneic UCART19 under a special license because no other therapies were available for refractory relapsed ALL following mismatched allogeneic stem cell transplantation.

On November 18, 2015, Cellectis signed an amendment to its collaboration agreement with Servier, under which Servier exercised its option to acquire the exclusive worldwide rights to further develop and commercialize UCART19, which was then about to enter into Phase I clinical development for chronic lymphocytic leukemia (CLL) and ALL. On November 19, 2015, Pfizer obtained the U.S. rights to UCART19, via an agreement with Servier. Servier and Pfizer will now split the cost and workload of the clinical trial program. These agreements free Cellectis to work on the rest of its pipeline of allogeneic CAR T-cell therapies, none of which has yet reached the clinic. It also gives Servier and Pfizer the opportunity to bring the off-the-shelf treatment to market, competing with late-stage autologous anti-CD19 products being developed by Novartis and Kite.

Meanwhile, Cellectis is also working on other early-stage classes of genetically-engineered experimental therapeutics that feature additional safety and efficacy attributes. These include control of properties designed to prevent attack of healthy tissues, to tolerate standard oncology treatments, and to resist mechanisms that inhibit immune system activity.

Interestingly, Cellectis’ recent preclinical studies have included development of CAR T cells that express an oxygen-sensitive switch fused to a CAR scaffold. The resulting CAR T cells are switched on by a hypoxic environment, and are less active under normoxic conditions. Some human solid tumors contain high percentages (approximately 50%) of hypoxic tissue. Therefore, these engineered CAR T-cells may be able to invade these solid tumors and specifically attack them. Conversely, if these CAR T-cells are specific for an antigen that is associated with tumors, but is also found at lower levels in some normal tissues, the oxygen-sensitive switch may result in sparing of normoxic normal tissues, while facilitating attack against hypoxic solid tumors. However, no in vivo studies have yet been conducted with the oxygen switch-engineered CAR T cells, so it is not known whether this approach will work against actual solid tumors.