Although breakthrough successes are generating a renaissance for anticancer immunotherapies, the framework for translating promising research results into clinical practice remains very much under construction.
Axel Hoos, MD, PhD
Although breakthrough successes are generating a renaissance for anticancer immunotherapies, the framework for translating promising research results into clinical practice remains very much under construction. As researchers work to build a foundation for widespread translation of immunotherapies, there is a growing appreciation that new approaches are needed to overcome the unique challenges that the development of such therapies pose.
In interviews with OncologyLive, leading immunologists identified four broad areas that researchers, regulators, and drug developers are tackling in order to fully realize the promise of the emerging therapies: (1) product development and clinical trial design; (2) immune-based prognostic and predictive biomarkers; (3) response criteria and clinical endpoints; and (4) treatment guidelines.
In recent months, significant strides have been made in elucidating those challenges (Table) and optimism about the potential for immunotherapy remains unbridled. Immunotherapy was named the “Top Scientific Breakthrough for 2013” by Science magazine and the American Association for the Advancement of Science. The Economist’s 2013 innovation award went to James P. Allison, PhD, of the University of Texas, MD Anderson Cancer Center, a leading developer of the “checkpoint blockade” immunotherapy approach.
Broadly defined, cancer immunotherapies are agents aimed at promoting an antitumor immune response and are generally classified either as “active,” in which the patient’s immune system is actively induced to recognize tumor cells (eg, vaccines) or “passive,” in which the need for an immune response is bypassed by directly administering components of the immune system to the patient (eg, monoclonal antibodies and adoptive T-cell therapies).
The current enthusiasm for such strategies has been building since April 2010, when the FDA approved sipuleucel-T (Provenge) for advanced prostate cancer as the first therapeutic cancer vaccine. Less than a year later, the agency approved ipilimumab (Yervoy), a first-in-class checkpoint inhibitor of CTLA-4, for the treatment of metastatic melanoma.
Now, excitement is swirling around an emerging class of checkpoint inhibitors aimed at the programmed death—1 pathway and its ligand (PD-1/ PD-L1). The implications are significant, according to Howard L. Kaufman, MD, vice president of the Society for Immunotherapy of Cancer (SITC). “In the past, immunotherapy was something that seemed to be very effective in only a small number of patients, often with rare tumors like melanoma and renal cell cancer,” said Kaufman, chief surgical officer and associate director for Clinical Science at the Rutgers Cancer Institute of New Jersey. “I think one of the real paradigm shifts that’s occurring now is what’s going on with the PD-1/PD-L1 pathway, because this seems to be a very important therapeutic target across a wide range of different types of cancer.”
The immunotherapy field has made great strides in less than a decade, noted Axel Hoos, MD, PhD, whose work has focused on developing and translating novel approaches for clinical use. He helped lead the development of ipilimumab (Yervoy) at Bristol-Myers Squibb and is now vice president of Oncology R&D at GlaxoSmithKline Pharmaceuticals. “We have spent at least 30 or 40 years trying to make the immune system work against cancer and haven’t been successful, and just recently—I would say the past five to eight years—we have been able to understand both pieces that you need for success,” he said. “The first is the science, obviously, (we’ve understood mechanisms better and know what to target) and then we also improved the methodology (such as clinical endpoints for trials).”
“I think the outlook for this is huge, probably more successful than any other kind of cancer therapy that we have used in the past,” said Hoos. “We have a positive outlook in all ways—in a scientific way, a clinical benefit way for the patient, and also in a financial way. The sector is growing.”
Changing Drug Development
While practicing oncologists are increasingly willing to embrace immunotherapies, the pathway to regulatory approval for those promising medicines has highlighted distinct development challenges. In the past, regulators applied methods used to assess conventional cytotoxic cancer therapies to immunotherapeutic agents, but now they are recognizing the unique biologic activity of these emerging drugs.
Depending on the specific type of immunotherapy, the FDA treats immunotherapeutic agents as either drugs or biologics. During product development, federal regulations require that an investigational product undergo thorough characterization, including describing attributes such as quality, potency, viability, and toxicity. This requires a thorough understanding of the biology of all components of the product.
Immunotherapeutic agents are frequently composed of whole cells or may contain adjuvants. In many cases, it has been necessary to develop new assays to evaluate immunotherapies and, since the antitumor immune response is a multicomponent biological process, multiple monitoring assays are needed to measure these components. Safety is a particular concern, and it is vitally important to adequately evaluate the immune response that these agents elicit in order to minimize unintended effects. Given the range and complexity of immunotherapeutic agents, the FDA has adopted a case-by-case, data-driven approach to evaluating safety.
Moreover, there are a multitude of considerations in preclinical and clinical testing of immunotherapeutic agents that have necessitated alterations to standard trial design. In the preclinical setting, the use of standard animal models may not adequately mimic the immune response in humans. In early-stage clinical testing, designing studies to inform dose selection is challenging, particularly with active immunotherapies such as vaccines. The conventional approach of dose escalation until toxicity is observed may not be appropriate.
Furthermore, anticancer agents are typically tested initially in patients with metastatic disease who have had multiple prior therapies. This approach may also be inappropriate for immunotherapies, since it can take several months to establish an antitumor immune response, and prior therapies may have negative effects on the immune system.
Using a heterogeneous patient population, with a mix of tumor types at various stages of disease can also confound measurement of an immune response. During more advanced clinical development, issues with response patterns and clinical endpoints used for approval have arisen.1,2
In order to raise awareness of these issues, the FDA has organized workshops and advisory committees, as well as developed practice guidelines. A guidance document on vaccines was recently published and others are expected in the near future.3
New Immune Biomarkers in Works
The development of a classification system that includes an immune component could prove to be a valuable prognostic tool beyond the histopathological evaluation and TNM classification system typically employed. Indeed, a paradigm-altering body of research is beginning to emerge that solidifies the vital role of the immune system in predicting the clinical course of a tumor and potentially even predicting response to therapy.
The type, density, functional orientation, and location of immune cells within the tumor is described as the immune contexture (Figure).4 The clinical translation of the immune contexture has resulted in the Immunoscore, a classification system that measures the density of two lymphocyte populations (CD8+ and CD3+) at the tumor core and the tumor margins, and provides a score from 0 (low density in both regions) to 4 (high density in both regions). A lower score has been shown to correlate with higher recurrence rates and poorer prognosis. The SITC and other organizations have established an international task force at 23 centers in 17 countries in order to validate the Immunoscore and assess its feasibility and reproducibility in a routine clinical setting.5
Bernard A. Fox, PhD
The task force is anticipating that validation of the Immunoscore in colon cancer will be completed during the fourth quarter of 2014, according to Bernard A. Fox, PhD, chief of Laboratory and Molecular and Tumor Immunology at the Earle A. Chiles Research Institute in Portland, Oregon.
“If validated, this will mark a fundamental change in the way oncologists stage cancer patients, as it will recognize the important contribution the immune system plays in their outcome,” said Fox. “In the future, we expect that this biomarker will evolve into a predictive biomarker and will be used to tailor therapies with essentially all types of cancer. Already, there are data for at least 18 of the most common cancers, suggesting a positive correlation between immune cells at the tumor site and improved outcomes.”
“The next step is figuring out what to do for the patients who have a negative Immunoscore and don’t respond to checkpoint blockade, and that is where I will be working for the next five years,” said Fox. A more global approach to evaluating the prognostic significance of the immune system is also being developed. Using flow cytometry, researchers are trying to determine a patients’ immune phenotype, a measure of the combination of immune markers in their blood.
These phenotypes are then compared between healthy individuals and those with malignant tumors (including glioblastoma, renal cell carcinoma, non-Hodgkin lymphoma, and ovarian cancer) and acute lung injury (a known immunesuppressive condition associated with poor outcome).
Individuals with common phenotypes were assigned to specific immune profiles, according to Gustafson et al.6
The team found profiles that were disease-specific, as well as those that were shared across malignancies. “Some cancer patients have a profoundly altered immune system; those who have an immune system more like those of noncancer patients will have better outcomes independent of the underlying tumor or how they were treated,” said Allan Dietz, PhD, assistant professor of Laboratory Medicine and Pathology at the Mayo Clinic and one of the study’s authors.
When asked whether a tool such as immune profiling might become part of the standard diagnostic workup for patients with cancer in the future, Dietz said, “Absolutely. In some cases, those patients with a more ‘normal’ immune system may proceed directly to standard of care, while patients with an ‘abnormal’ immune system will be treated with immune modulators to restore immunity.”
The immune status of a tumor can be assessed and stratified using three key measures (above): (1) the immune contexture; (2) an Immunoscore; and (3) an immunologic constant of rejection, according to an international task force working on standards for translating immunotherapy into clinical practice. The genes listed have been identified as those whose alterations are more likely to play an underlying role in biologic processes that may predict responses to immunotherapy.
Source: Galon J, Angell HK, Bedognetti D, Marincola FM. The continuum of cancer immunosurveillance: prognostic, predictive, and mechanistic signatures. Immunity. 2013;39(1):11-26. Reprinted with permission from Elsevier
Identifying biomarkers that predict response to therapy is also an important endeavor.
“One of the most important challenges we face is to identify the right patient population for the right immune intervention so the maximum benefit can be given to those patients,” said Hoos. He said that biomarker development is complicated and takes time, but he believes that, “Once we have good biomarkers to identify the immune status of the patient at the start of therapy and the immune responsiveness to therapy, we’ll see even greater success than we already have.”
Revamping Response Criteria and Trial Endpoints
As clinicians gain experience with immunotherapeutic agents, unconventional patterns of response that depart from traditional measures of disease progression (eg, tumor growth, appearance of new lesions) have emerged. Under RECIST criteria, such patterns typically would prompt treatment to be halted—which may have contributed to the premature discontinuation of some immunotherapy trials in the past.
While a typical immediate response is observed with some forms of immunotherapy, such as adoptive cell therapy, a delayed response, sometimes preceded by a period of apparent progression, is often seen with other types of immunotherapy. Delayed responses are attributed to the time required for a specific antitumor immune response to be generated, while periods of apparent progression are thought to result from the infiltration of tumor cells and inflammation at the tumor site.
The delayed responses observed with vaccines, such as sipuleucel-T, have been formally recognized by the FDA in a Guidance for Industry document released in 2011.3 Nonstandard responses are also frequently observed with the novel checkpoint inhibitors that have garnered much attention in recent years.
Hoos believes it is extremely important for practicing oncologists to be aware of the clinical kinetics of immunotherapeutic agents. “Using an immune intervention can lead to regular responses where the tumor shrinks right away, but it can also lead to delayed responses where it takes some time for the tumor to shrink or the tumor gets bigger before it shrinks,” he said. “That’s something that physicians are not used to in oncology. I think by now people are beginning to understand it, but there is always a risk that a therapy gets stopped and the patient doesn’t get the full benefit.”
Clinical trial sponsors are now able to include a provision in their study designs for nonstandard responses, but researchers have been working toward formalizing these observations with the development of immune-related response criteria (irRC) and clinical endpoints that are specific to immunotherapies, which could be used to facilitate and accelerate the development of novel agents.
The irRC were developed in 2009, following a series of large, multinational studies based on the phase II clinical development program with ipilimumab. The irRC are based on modified WHO criteria and involve the use of bidimensional measurements on radiographic assessment of cancerous lesions (the longest diameter and the longest perpendicular diameter).
The irRC differ from conventional response criteria in that they incorporate measurement of both preexisting lesions (index lesions) and new lesions, as opposed to just index lesions.7,8
“In the past, growth of a tumor has been a sign of drug failure. Here it could be a sign of drug success,” said Hoos.
While the irRC provide a useful additional tool to capture clinical activity for immunotherapy, they don’t adequately assess tumor growth kinetics, and overall survival (OS) remains the gold standard endpoint for oncology trials in seeking regulatory approval. However, measurement of OS is often confounded by the unusual response kinetics observed with immunotherapies.
To further facilitate clinical development of immunotherapeutic agents, clinicians are developing surrogate clinical endpoints. Currently, these endpoints serve a useful exploratory purpose but still require validation in clinical trials and are not considered a replacement for OS by the FDA. However, in the future they have the potential to serve as the basis for accelerated approval for some immunotherapies.2
Treatment Guidelines Evolve
The effective clinical translation of cancer immunotherapy will ultimately require the development of consensus guidelines for clinical practice as data begin to accumulate and more agents achieve FDA approval. Indeed, the progress made in the field of melanoma is reflected in the fact that the SITC recently developed evidence-based guidelines on the use of immunotherapy for the treatment of patients with advanced-stage, high-risk melanoma.9
The SITC guidelines are conservative and stick closely to evidence from clinical trials. Kaufman explained the reasons for this. “It has been an interesting process for me to develop the guidelines because a lot of people have a lot of really strong feelings and passion about immunotherapy, but when it actually came down to looking at the data I was surprised by how little we actually have,” said Kaufman, referring to the guidelines in general and not any specific agent. “The importance of these consensus guidelines is to give a sense of what the best minds in the field are thinking right now, until we have the confirmatory data.”
SITC has plans to develop a genitourinary malignancy guideline that would cover prostate cancer and renal cell cancer, where there are immunotherapy agents available, and for hematological malignancies, Kaufman said. “Hopefully, if all goes well, we will do others in the future,” he said. Meanwhile, researchers will be seeking to determine how integrating multiple approaches might improve outcomes.
“The role for combination strategies that include next-generation vaccines that effectively prime a broad repertoire of new cancer-specific T cells and provide costimulation and checkpoint blockade is promising in animal models,” said Fox. “The challenge is to identify strategies that will induce a spectrum of tumor-specific T cells in patients with lung, prostate, and breast cancer, where tumor-specific T cells have been harder to find.”
Addressing the Challenges
Product development and clinical trial design
Due to the unique biology of immunotherapies, they require a novel approach to development and evaluation: clinical trial methodology, preclinical evaluation, and the drug review process need to be adapted
FDA has held workshops, published guidelines, and organized advisory committees on immunotherapeutic products
Prognostic and predictive biomarkers
Prognosis: Traditional methods of predicting clinical outcome do not incorporate the immune system and focus solely on tumor cells
Predicting response to therapy: Lack of predictive biomarkers to determine which patients are likely to best respond to immunotherapeutic agents
Response criteria and clinical endpoints
Conventional criteria for assessing response to a novel drug recommend drug treatment halted upon disease progression, but immunotherapeutic agents often display unconventional responses that are not captured by these criteria and may result in premature discontinuation of therapy
Immune-related response criteria (irRC)5
Capture 4 patterns of response to immunotherapy; allow drug development to continue even in presence of what appears to be “progression:”
Improvement in overall survival (OS) is gold standard clinical trial endpoint required for regulatory approval; however, delayed response to immunotherapy may confound measurement of OS
Surrogate endpoints6 Developed to predict traditional endpoints at an interim analysis; may be used to gain early drug approval without waiting to reach OS
Immunotherapy-specific evidence-based guidelines are needed to guide clinical decision making
1Vatsan et al. J Immunother Cancer 2013,1:5;
2Galon et al. J Transl Med. 2012,10:205;
3Galon et al. Immunity 2013,39:11;
4Gustafson et al. J Immunother Cancer 2013,1:7;
5Hoos et al. J Natl Cancer Inst. 2010;102(18):1376;
6Hoos A. 2013 Conference on Clinical Cancer Research Issue Brief;
7Kaufman et al. Nat Rev Clin Oncol. 2013,10(10):588.
Jane de Lartigue, PhD, is a freelance medical writer and editor based in the Davis, California.