Immunotherapy is on its way. A few agents have already been approved: ipilimumab (Yervoy) in 2011 for melanoma; nivolumab (Opdivo) in 2015 for non–small cell lung cancer (NSCLC) and then later that year for renal cell carcinoma; and pembrolizumab (Keytruda) for NSCLC. In addition, many clinical trials are showing great promise for this approach.
Harnessing a patient’s immune system to fight disease offers several advantages over traditional drugs.— Samir N. Khleif, MD
Samir N. Khleif, MD, Director, Georgia Cancer Center and Chairman of the National Cancer Policy Forum’s workshop, “Policy Issues in the Clinical Development and Use of Immunotherapy for Cancer Treatment,” said in his introduction: “Harnessing a patient’s immune system to fight disease offers several advantages over traditional drugs. In a healthy individual, the immune system acts with a high level of specificity to eliminate intruding antigens. Adaptive immunity creates immunologic memory after an initial exposure to a foreign pathogen, which in turn confers long-term protection in the event of future encounters. These benefits—specificity, memory, and adaptability—can be achieved through immunotherapy. It also generally produces fewer side effects than chemotherapy.”
Immunotherapy: What It Is, What It Can Do
Immunotherapy includes vaccines; immune modulators, such as agonists (stimulatory cytokines and costimulatory molecules) and inhibitors (checkpoint inhibitors and inhibitory cytokines); small molecules; and T-cell therapy. All of these approaches can be used alone, in combination with one another, and in combination with chemotherapy and/or radiation.
Developing them to the point of clinical utility presents a number of issues and challenges, said Dr. Khleif. For example: What preclinical data are needed in order to move agents and combinations to clinical trials? What dosing schedules will best provoke a response? How should clinical trials be designed? How can different developers and manufacturers create and market combinations? What will be the economic effects of these very expensive agents?
Steven A. Rosenberg, MD, PhD, Chief, National Cancer Institute (NCI) Surgery Branch, waxed enthusiastic about immunotherapy in general and cell-transfer therapy in particular. “With cell transfer, you can administer large numbers of highly selected cells with high avidity for tumor antigens. These cells can be activated ex vivo to exhibit antitumor effector function. This means that you can potentially identify exact cell subpopulations and effector function required for cancer regression in vivo. Moreover, you can manipulate the host prior to cell transfer to provide an altered environment for transferred cells.”
The photographs and scans he showed to demonstrate this strategy were dramatic: Huge fulminating cancers resolved into nothing more than scar tissue and multiple tumors darkening scans that melted into blankness in a few weeks.
“Based on recognition of immunogenic cancer mutations, adoptive cell therapy can mediate complete, durable, and likely curative regression of metastatic melanoma, even cases that have been refractory to other treatments. If you can achieve a complete response, it’s very likely to be durable,” said Dr. Rosenberg.
Malcolm K. Brenner, MD, PhD
Malcolm K. Brenner, MD, PhD, Professor, Center for Gene Therapy, Baylor College of Medicine, added that in adoptive cellular immunotherapy, “T-cell–mediated cancer therapy is like monoclonal antibodies—only better. Specific receptors provide high targeting ability; they recognize internal antigens and provide good biodistribution; they are self-amplifying with multiple effector mechanisms; and the specificity of response can evolve with the tumor.”
Challenges in Preclinical Models
Helen Heslop, MD, DSc, Interim Director, Center for Cell and Gene Therapy, Baylor College of Medicine, described some of the challenges in preclinical models: In vitro studies are not always predictive; murine models have different immune systems and different target antigens from humans; and some mice are immunodeficient, which means that although human tumors can be engrafted, they may lack all necessary immune components.
In addition, said Dr. Heslop, preclinical models looking at cross-reactivity of virus-specific T cells had been misleading, predicting that cross reactivity may occur whereas this product has actually had an excellent safety profile in multiple trials in many centers.
“A mouse with cancer is not the same as a human with cancer,” said
Bernard A. Fox, PhD
Bernard A. Fox, PhD, Chief, Laboratory of Molecular and Tumor Immunology, Earle Chiles Research Institute, and CEO of UbiVac. “There are significant differences in the time it takes for a tumor to grow, and the tumor burden is different, as is immune selection. Moreover, different mouse models can give very different results, especially with combination therapy.”
Pilot studies in humans also are difficult because the number of patients is so small.
Vaccines present their own sets of challenges, said Dr. Fox. “They provide opportunities to advance the field of immunotherapy and improve patient care, but there are many gaps in the evidence.”
Jay A. Berzofsky, MD, PhD
“Cancer vaccines are distinct from most other vaccines,” added Jay A. Berzofsky, MD, PhD, Chief, Vaccine Branch, NCI Center for Cancer Research. “They can harness exquisite specificity of the immune system to selectively target cancer cells, but to do so, we need to identify the molecules that are unique to, or highly overexpressed in, the cancer cells to be targeted with a vaccine. They must be clearly distinguished from normal cells,” he said.
“Longstanding cancers may have tolerized immune cells that could recognize them, so we need to overcome this tolerance. Moreover, cancers exploit immunosuppressive mechanisms to suppress the immune response against them, so we need to overcome this suppression as well,” Dr. Berzofsky continued.
“Vaccines require an intact immune system; they cannot be tested in xenograft models in immunodeficient mice; they can be affected by prior lymphodepleting treatment; and there must be an HLA match between T-cell specificity and target tissue, so syngeneic mouse models must be used.”
Because vaccines take time to mediate their effect and may slow tumor growth without causing regression, tumors may appear to progress before they regress, and overall survival may be improved without affecting progression-free survival. Therefore, vaccines cannot be evaluated with the usual RECIST criteria to which oncologists are accustomed.
Regulation and Approval
Whitney S. Helms, PhD, Supervisory Pharmacologist, U.S. Food and Drug Administration Department of Hematology and Oncology Products, described the difference between the two major offices of product regulation. The Center for Drug Evaluation and Research is responsible for chemical drugs, monoclonal antibodies, fusion proteins, and cytokines. The Center for Biologic Evaluation and Research oversees genetically modified T cells, cancer vaccines, and oncolytic vectors.
“To initiate a clinical trial for an anticancer drug, you usually need 28-day toxicology studies in two species, although for biologics, a single pharmacologically relevant species is often acceptable,” she said.
The preclinical program must have identified the pharmacologic properties of the agent, estimated a safe initial dose for the first human exposure, and explicated the toxicologic profile.
“With cancer immunotherapeutics, the challenges are even greater because species relevance has been a problem. This is what we look for: pharmacology of the targeted pathway, assessment of the cytokine-release potential, studies using human cells that take into account multiple mechanisms of action, and receptor occupancy.”
Dr. Helms said that the agency cannot rely solely on traditional toxicology studies for safety predictions for these products. “Therefore, thorough examination of the mechanism of immune activity is critical for first-in-human trials as well as throughout development and in the postmarketing period.”
Angela Thomas, PhD
Angela Thomas, PhD, Clinical Trials Chair, Biological and Vaccines Expert Advisory Group, Commission on Human Medicine, UK National Health Service, explained how things work abroad. “European pharmaceutical law allows a single marketplace, and licensing procedures are centralized as well. One application, one evaluation, and one authorization cover all European Union member states. The European Commission grants the license.”
Biologicals now constitute 27% of pharmaceutical sales in Europe; they have experienced a 5.5% growth rate vs 1.9% growth for the total pharmaceutical market. Many patents will expire before 2020, which allows for development of biosimilars and the potential for decreased cost and increased availability. This has already happened with two popular biologics: rituximab (MabThera in Europe), the patent for which expired in 2013, and trastuzumab (Herceptin), which expired in 2014. Two biosimilars have been approved for the former and one for the latter.
Challenges in Trial Design
“Immunotherapy trials are different from other clinical testing,” said Richard Simon, DSc, Chief, NCI Biometric Research Program. “Preclinical trial models are inadequate; the mechanism of action is more complex; phase I 3+3 design is not applicable; most immunotherapy requires combination regimens; and there are many candidate immunomodulating agents.”
In designing trials, biomarkers are critical, said Lisa Butterfield, PhD, Professor of Medicine, Immunologic Monitoring and Cellular Products Laboratory, University of Pittsburgh. They will avoid toxicity and the treatment it necessitates, they will avoid ineffective therapies for specific patients, they will lead to understanding of mechanisms of action and how to build on them, and they will aid in rational design of combination therapies.
Alas, adequate biomarkers are in short supply, she said. “We need the right specimens saved under standardized conditions. Variably banked specimens give ‘noisy’ data, and many trials bank only nonviable tumor and blood samples,” she explained.
“Immune assays can be costly,” she continued, “and testing small numbers doesn’t provide robust, reproducible signals.” Therefore, before any trial, one must be concerned about:
Lisa Butterfield, PhD
Prediction: Who should be enrolled?
Prognostication: Who can benefit from a given therapy?
Mechanism: What worked well or not well in a given intervention? Did the vaccine induce antitumor immunity? Did the intervention kill the tumor? Was immune suppression reversed? Was the signaling pathway blocked?
The SITC “Immunotherapy Biomarkers Task Force” is trying to help address needs in the field, with four working groups making recommendations on different areas of biomarker focus.
Immunotherapy in Practice
Immunotherapy can work for cancer patients; in a few cases it already does, but response patterns are different from traditional therapy. For example, the time to an antitumor response can be longer. Often, a clinical response is apparent only after a period of “pseudoprogression,” when immune cell infiltration can manifest as new lesions or growth of existing ones. Discontinuation of treatment is not appropriate until real disease progression is confirmed. Moreover, durable stable disease (without apparent tumor regression) may represent antitumor activity.
Patient education is key to early detection because most [immunotherapy-related] adverse events, such as fatigue, headache, abdominal pain, colitis, and itchiness, can’t be identified on physical exam.— Ramy Ibrahim, MD
Managing adverse events for immunotherapeutic agents can differ as well, said Ramy Ibrahim, MD, Clinical Vice President, Immune-Oncology, AstraZeneca. “The safety profile is based on the mechanism of action, and although adverse events can occur in any organ, the most commonly affected organs are gastrointestinal (diarrhea), dermatologic (rash, itchiness), hepatic (usually clinically asymptomatic elevation in transaminases), and endocrinopathies. Most adverse events are low grade in severity and manageable, but if not properly managed, symptoms can worsen, possibly to the point of irreversibility and/or the necessity to discontinue treatment.”
Dr. Ibrahim said there is no standard approach to identifying immune related adverse events, but any inflammatory event should be taken seriously and the use of steroids should be considered. “Patient education is key to early detection because most adverse events, such as fatigue, headache, abdominal pain, colitis, and itchiness, can’t be identified on physical exam and the treating physician has to rely on the patient to make him/her aware of any new symptom or worsening of existing symptoms. Patients are sometimes reluctant to report low-grade events because they want to continue treatment, and they tend to self-medicate and tell their oncologist only when the symptoms get worse. And if they go to a specialist for treatment, the oncologist might not find out about it and won’t be able to inform how they are being managed.”
Paying for immunotherapy will pose significant challenges because the Centers for Medicare & Medicaid Services and private insurers are having a hard time keeping up with approved and recommended indications for various cancers. For instance, the National Comprehensive Cancer Network® (NCCN®) recommended nivolumab for second-line treatment of NSCLC in June 2015, but some private payers were reluctant to cover it. This kind of response means that not only do patients run the risk of not getting the drug (or not having it reimbursed, which is tantamount to the same thing), but providers have to spend time and energy researching policies of all the insurers they deal with.
Coverage in the private sector is still spotty. Medicare is similar because its various contractors across the country have different policies about what they will cover.
The cost of these agents is staggering—higher than any previous class of cancer treatment. Therefore, payers are seeking better evidence that patients will clearly benefit, which brings the issue back to identifying biomarkers and their potential for predicting success.
Large-scale development and manufacture also poses problems. The first, said Eric Perakslis, PhD, Senior Vice President, Informatics, Takeda Pharmaceuticals, is interoperability—the ability of different information technology systems and software applications to communicate, exchange data, and use the information that has been exchanged. “Data exchange should permit information to be shared across clinicians, laboratories, hospitals, pharmacies, and patients regardless of the application or vendor.”
Dr. Brenner talked about the many problems that pharmaceutical companies and other entities face when deciding whether to collaborate or compete: “Publish or perish vs perish if publish,” puts the situation in a nutshell.
Mary M. Horowitz, MD, MS
Mary M. Horowitz, MD, MS, Chief, Division of Hematology and Oncology, Medical College of Wisconsin, agreed and listed issues in establishing a mechanism for data integration and sharing for novel therapies. “It must capture all patients of interest as well as all data of interest, no matter where it is generated. It has to ensure data quality and maintain long-term follow-up. There must be confidentiality, security, and regulatory compliance, and the data must be available to multiple users.” ■
Disclosure: Drs. Rosenberg, Berzofsky, Helms, Thomas, Simon, and Horowitz reported no potential conflicts of interest. Dr. Khleif is member of scientific advisory boards for Boehringer Ingelheim, Merus B.V., UbiVac LLC, Nektar Therapeutics, Syndax Pharmaceuticals, PDS Biotechnology Inc., Bullet Biotechnology Inc., Bioline Therapeutics, NewLink Genetics, Lycera Corporation, IO Biotechnology, CanInGuide Therapeutics AB, AstraZeneca, and MedImmune, has been an advisor for Janssen and Medivation, is on the Corporate Board of Directors of Advaxis Immunotherapeutics, and has received clinical funding from Kyowa Hakko Kirin Pharma, Inc., AstraZeneca Pharmaceuticals, Argos Therapeutics, Newlink Genetics, Advaxis, Milennium, AstraZeneca/MedImmune, Cure Tech, Merck Serono, Sanofi-Aventis, Medivation, Inc., PDS Biotechnology, AstraZeneca UK, and BMS Foundation. Dr. Brenner has received research support from Celgene, is a stockholder/patent licensee in association with bluebird bio, Tessa Therapeutics, and Viracyte Cell Medica, and has been a member of scientific advisory boards for NantKwest, Neon, and Unum. Dr. Heslop has a licensing agreement with Cell Medica, has a collaborative research agreement with Celgene, and is the founder of Virocyte. Dr. Fox has received research support from Aduro Biotech, Definiens, Janssen/Johnson & Johnson, Bristol-Myers Squibb, MedImmune/AstraZeneca, PerkinElmer, Peregrine, Ventana/Roche, and Viralytics; is a member of scientific advisory boards for Argos, Bristol-Myers Squibb, Janssen/Johnson & Johnson, Immunophotonics, MedImmune/AstraZeneca, PerkinElmer, PrimeVax, and Ventana/Roche; has been a consultant for Immunophotonics; and is a cofounder and managing member of UbiVac. Dr. Ibrahim is an employee of AstraZeneca. Dr. Butterfield has been a member of scientific advisory board for Caladrius (formerly NeoStem) and has been an advisory board participaant for Oxford Immunotec, Affymetrix/eBioscience, Merck, and Biodesix. Dr. Perakslis is an employee of Takeda Pharmaceuticals.