April 1, 2015 (Vol. 35, No. 7)

Ken Doyle, Ph.D.

The Immunotherapy Challenge Is to Identify the Right Tumor Targets

Adoptive immunotherapy has emerged as a promising weapon in the therapeutic arsenal being mobilized to fight cancer. In this approach, T cells are isolated from a patient’s tumor, genetically engineered using one of several different approaches, and infused back into the patient.

Of various strategies for adoptive immunotherapy, the use of chimeric antigen receptors (CARs) has received considerable attention. Typically, sequences that encode the variable regions of antibodies are modified and then spliced into the T-cell receptor (TCR) intracellular domains that activate T cells.

This chimeric construct is cloned into a retroviral vector and used to generate a population of T cells that express the CAR, with activity specific for an antigen expressed by the patient’s tumor. These modified T cells, when infused into the patient, specifically target tumor cells that express the antigen, resulting in a rapid immune response.

Target discovery is a large part of cell therapy, as are investigations into TCRs, CAR engineering, and the use of tumor-infiltrating lymphocytes (TILs) against tumor antigens. All these pursuits will be reviewed at CHI’s Adoptive T Cell Therapy conference to be held in Boston next month.

Creating designer T cells that target specific tumors involves a collaborative effort from scientists in a variety of disciplines, in partnership with physicians who deal with the realities of cancer therapy on a daily basis. To this end, the Memorial Sloan Kettering Cancer Center established the Center for Cell Engineering. Michel Sadelain, M.D., Ph.D., is the center’s director and heads a research group studying the mechanisms governing transgene expression, stem cell engineering, and other genetic strategies to bolster the immune response in cancer patients.

A significant area of Dr. Sadelain’s research focuses on the CD19 receptor. He explains, “We chose CD19 (over 15 years ago) because of where this molecule is expressed, as well as where it is not expressed.” Since CD19 is found on the cell surface in most leukemias and lymphomas, but nowhere else in the body except for B-lineage cells, it is an attractive target for therapy. In addition, CD19 has been implicated in tumor survival, making it more likely to be present on—and less likely to be lost from—all tumor cells.
 
Dr. Sadelain points out that the most impressive results from CAR T-cell therapy have been achieved in acute lymphoblastic leukemia (ALL), first in adults and then in children. However, he says, “One of the big questions is whether this approach will work in solid tumors.” In theory, CAR therapy should be applicable to a wide range of cancers. The challenge facing researchers is “to identify the right targets and to appropriately adapt the T-cell engineering strategies to the immune barriers opposed by different tumor types.”
 
Dr. Sadelain’s group is pursuing targets, such as prostate-specific membrane antigen and mesothelin, for which there is substantial information already available in the scientific literature. His group is also searching for new candidates based on studies from tumor samples. “We have open protocols for ALL, chronic lymphocytic leukemia (CLL), certain lymphomas, and Waldenstrom disease,” he says. “We will soon open a trial for patients with certain forms of mesothelioma, lung adenocarcinoma, and breast cancer.”


Electron micrograph of magnetic beads stimulating T cells in preparation to make CAR T cells. [University of Pennsylvania Abramson Cancer Center]

Hematologic Cancers

The first trials of CAR-based therapy addressed hematologic cancers, largely due to the extensive knowledge base surrounding surface antigens expressed on hematologic cells. Researchers, however, still face considerable challenges when engineering CARs.

Marcela Maus, M.D., Ph.D., director of translational medicine and early clinical development at the University of Pennsylvania’s Abramson Cancer Center, notes that “the process for manufacturing CARs is more complex than making a pill.” She adds that different kinds of cancers require different targets, and that finding good targets can be difficult.

The tools offered by synthetic biology have helped, in some cases, to facilitate the development of CARs. “Swapping and testing various intracellular signaling domains and CAR domains like Lego bricks—or ‘biobricks’—has become a relatively routine set of experiments to do in the lab,” says Dr. Maus. Ultimately, though, extensive clinical trials are required to determine the best engineered domains and study their safety and efficacy.

Dr. Maus notes that CAR-directed approaches show the greatest promise in B-cell malignancies such as B-cell acute lymphoblastic leukemia (B-ALL) and non-Hodgkin’s lymphomas. Accordingly, the University of Pennsylvania has teamed up with Novartis to offer CAR therapies on a global scale.

Other hematologic malignancies are attractive targets as well. According to bone marrow transplant research, these types of tumors are easily accessible by T cells. Dr. Maus adds, “CAR-modified T cells can form memory, which means there is potential for a long-term remission.” Because they are generated with the patient’s own immune system, Dr. Maus explains, “CAR T cells combine the potential for cure that bone marrow transplants have, but they don’t have the same risks.”


Solid Tumors

John Maher, Ph.D., senior lecturer, immunology, department of research oncology, King’s College London, reiterates the promise shown by targeted T cells in treating hematologic malignancies. “Much excitement surrounding the use of gene-targeted T cells concerns their high degree of effectiveness in the treatment of B-cell malignancy, especially ALL,” says Dr. Maher.

He adds, however, that some approaches used in B-ALL target both healthy and diseased B cells, ultimately leading to hypogammaglobulinemia. This toxicity can be treated by intravenous or subcutaneous immunoglobulin replacement therapy. “Engineered T cells that recognize B-lineage targets, such as CD19, can consequently provide a highly effective therapeutic approach,” explains Dr. Maher.

Dr. Maher’s research focuses on extending the targeted T-cell approach to solid tumors—a field of study fraught with obstacles. “Tumor-specific targets are very few and far between,” says Dr. Maher, “meaning that we are generally forced to target molecules that are also expressed in healthy tissues.” In addition, T-cell homing to metastatic solid tumors is not as efficient as for hematological malignancies. For those T cells that do reach the target, it is difficult to maintain functionality and viability.

For these reasons, Dr. Maher’s group is examining specific targets in high-profile cancers, such as the ErbB family of receptor tyrosine kinases. “It’s difficult to find a solid tumor type in which dysregulation of the ErbB network is not a common event,” notes Dr. Maher. He adds that ErbB also makes an attractive target because its dysregulation contributes directly to pathogenesis. Further, the ErbB family is the most successful cancer-associated target for monoclonal antibody therapy, for example, in agents such as herceptin (anti-ErbB2) and cetuximab (anti-ErbB1).

Dr. Maher’s research initially focused on ovarian cancer, but his group plans clinical trials in head and neck cancers later this year. His approach addresses concerns about safe targeting—a universal challenge, since ErbB receptors are widely expressed in healthy tissues, although at low levels—by using regional delivery systems. “We will inject the cells directly into head and neck tumors,” he states. “In ovarian cancer or mesothelioma, intracavitary delivery may be used.”

Future directions for Dr. Maher’s research include investigating additional tumor-selective targets, and methods to improve T-cell trafficking to tumors. At the same time, his group is examining ways to mitigate on-target toxicities associated with CAR T-cell therapy. “We have data suggesting that this approach can be combined with conventional therapies, such as chemotherapy, that further sensitize tumor cells to CAR T-cells,” concludes Dr. Maher.

Although viral vectors are still popular for engineering T cells, researchers are examining other methods that can improve efficiency and lower cost. Laurence Cooper, M.D., Ph.D., professor in pediatrics at the M.D. Anderson Cancer Center, is using one such approach—taking advantage of in vivo transposition systems such as Sleeping Beauty (SB). Dr. Cooper uses the SB system to engineer CAR T cells to target tumor antigens. His research also focuses on pediatric cancers including acute leukemia.

Dr. Cooper explains that CAR T cells have already shown dramatic results in pediatric patients, and these data will “provide the foundation for next-generation clinical trials targeting childhood cancers that are refractory to conventional therapies.”

Dr. Cooper’s approach involves a “one for many” strategy, unlike conventional CAR T cell therapy. In his method, genetically modified CAR T cells are sourced from healthy donors in advance of when they are required for therapy. The modified cells can then be infused into a patient when required for immunotherapy. The approach addresses key disadvantages of conventional CAR T cell therapy—the time and expense associated with generation of the CAR and propagation of the T cells.

“We are developing off-the-shelf T cells before the patient arrives at a treatment facility,” asserts Dr. Cooper. “This enables a biobank of T cells to be pre-prepared, which could be a one-time cost.”

Additionally, Dr. Cooper’s group is working on manufacturing patient-derived T cells in real time, with the goal of eliminating the need for ex vivo propagation after gene transfer.  CAR T-cell therapy holds the most promise in childhood leukemia and lymphoma, but Dr. Cooper is optimistic that the approach will see success in solid tumors, such as neuroblastoma, as well.


Patient-Relevant Models to Advance Immunotherapy

Rising popularity and the recent approval of anti-PD-1 antibodies have established immunotherapy as an effective treatment paradigm in oncology. However, while immunotherapy demonstrates an extremely promising treatment option for cancer patients, it remains unclear as to why some patients benefit from these treatments and others do not. Researchers are also unsure how to maximize the benefits from immunotherapeutic agents.

The most significant obstacle to solving these challenges and developing effective immunotherapeutic drug treatments is currently the distinct lack of patient-relevant experimental models with which to screen potential candidates. Patient-derived xenografts (PDX) in immunocompromised mice have proven extremely successful in screening traditional chemotherapeutics, targeted agents, and radiotherapy for efficacy and safety.

But immunotherapeutics require functional immunity to effectively treat disease, and it is necessary to develop models with functional human and mouse immunity, accurately simulating the tumor microenvironment and the mechanics of treatment and disease progression.

To complement existing syngeneic and genetically engineered mouse models with functional murine immunity, researchers at Crown Bioscience say they have developed a collection of allografts of spontaneous murine tumors studied in mice with complete immunocompetency. These tumors cover a wide diversity of cancer types, enabling research into specific pathways and the discovery of predictive biomarkers for targeted immunotherapy agents, according to Jean-Pierre Wery, Ph.D., president of Crown Bioscience.

“Next-generation experimental models are also utilizing humanized mice, developed through inoculating human hematopoietic cells into immunocompromised mice. Combining these mice with HuPrime® PDX models, it is possible to evaluate target function and immune response in models that preserve the genomic integrity and heterogeneity seen in patients,”says Dr. Wery.


To create patient-relevant surrogate models for preclinical trials, humanized mice are developed by inoculating human hematopoietic cells into immunocompromised rodents. By working with such patient-derived xenograft models, researchers can evaluate the function of the drug target as well as the specifics of the immune response. [Crown Bioscience]

Partnering on Immunotherapies

The field of immunotherapy has advanced rapidly in the last 15 years, and gained greater momentum with the approval of CTLA-4 and PD-1 checkpoint inhibitors: Bristol-Myers Squibb’s Yervoy (2011) and Opdivo (2014) and Merck’s Keytruda (2014). Checkpoint inhibitors block critical pathways used by cancer cells to avoid immune detection, allowing the immune system to attack cancerous cells. Combined with the promise of improved survival rates—and the large potential market size—a race is underway to investigate the potential of immunotherapies in as many indications as possible.

“Even for Big Pharma, which typically has the financial resources to place an asset into a sizable early-stage trial and then advance it into multiple clinical indications, many are finding that an effective partnering strategy is essential to drive differentiation in an increasingly crowded and fast-paced market,” says Olivier Lesueur, managing director at Bionest Partners, a medical business strategy consultancy.
“More and more, companies are setting up combination trials, either with additional immunotherapies or targeted therapies to explore new indications or segments.”

For instance, in early 2015, Eli Lilly signed two separate deals with Merck and Bristol-Myers Squibb to test combinations of its cancer treatments with Keytruda and Opdivo.

“Historically, partnering strategies and clinical trial collaborations have included only single assets—one molecule or one targeted therapy. In this highly competitive race to the finish, we’re starting to see an increase in the number of multi-asset or portfolio collaborations to speed clinical trial progress,” noted Rachel Laing, Ph.D., strategy advisor at Bionest Partners.

The need for combination approaches is evident in recent trial activity with leading checkpoint inhibitors—more than 50% of the trials initiated on or after January 2013 are combination trials, and this number is expected to increase. As the recent Eli Lilly, Merck, and BMS deals prove, if assets are not available in house, partnering strategies are paramount.