One Year Later … Progress & Peril
How far has immuno-oncology progressed since the triumphant news on checkpoint inhibitors in 2014?
Llew Keltner, M.D., Ph.D., Roundtable Moderator
What a difference a year makes. Last year, beginning in September, we ran a series that addressed the challenges and opportunities of using new agents to rally the immune system against cancer. In most cases, we were talking about the checkpoint blockers such as ipilimumab and tremelimumab, inhibiting the checkpoint CTLA-4, and nivolumab, pembrolizumab, and pidilizumab, which have a multimodal effect on another target, PD-1. Checkpoints are proteins expressed on immune cells that normally “check” the immune system from attacking the body’s own cells. But tumors can hijack the checkpoints to keep the immune cells inactive and thus prevent an immune response.
Just a few highlights of how immuno-oncology (IO) has progressed during the past year:
Immuno-oncology has also shown signs of a more complicated reality, as shown by the scientific debates over ideal combinations, mechanistic explanations, and other remaining challenges. Beyond the scientific questions, the world of business started to have a larger impact on the IO space during the year. More IO agents are in the running, and companies that lack them in their oncology pipelines are gold-rushing to get them. With more and more companies involved in some form of immuno-oncology, the boundaries of the field keep expanding.
Large companies seem to have the upper hand with late-stage candidates and will inevitably and profoundly affect their clinical development, from trial design to choice of indications. Considerable angst in the research community has resulted, based on concerns that Big Pharma will waste a lot of opportunities by taking the easiest paths to market — avoiding the hard work and patience needed to develop immunotherapy to its fullest potential.
Experts Return, Experts’ Turn
For this update, we brought in most of our original “virtual roundtable” experts and other key opinion leaders (KOLs) to discuss such questions as we assess how far immuno-oncology has come toward fulfilling its promise since last year. Within that circle of viewpoints and information, we take some space to comment on the wild cards in this deadly serious game of thrones building in the biopharma worldscape — the cancer immunotherapy business.
The KOLs are not just thought leaders; they are thought doers. They are some of the key researchers pushing the envelope of immuno-oncology by doing the pivotal clinical trials. They are nearly all practitioners as well, treating patients in specific cancer areas, from melanoma to lung and GI. One panelist had to put us on hold while she took a patient’s urgent call.
With some exceptions, this is an unabashedly pro-immunotherapy crowd, all of them working on checkpoint inhibitors. But if you want a firsthand view of a major expedition into new therapeutic territory, you will find yourself walking alongside the lead explorers, who tend to look on the positive side of the trip. In our live conversations with the KOLs, prepared questions often went out the window in favor of those that occurred on the spot. Yet a single question contained all: How has your perception of immuno-oncology changed in light of new research since the dramatic findings released last year? Here follow individual accounts of the experts’ responses.
JEDD WOLCHOK, M.D., PH.D.
Chief, Melanoma and Immunotherapeutics Service, Memorial Sloan Kettering Cancer Center
“The pace of progress in immunooncology has continued to be very brisk.”
KOL Reflections — The Past Year In IO
Although quick to cite other people’s research, Dr. Wolchok has led or co-led some of the most important studies of cancer immunotherapy released during the past year. Foremost among the 10 trials presented at ASCO and listing Wolchok as an author is the star Phase 3 trial, “Combined Nivolumab and Ipilimumab or Monotherapy in Untreated Melanoma.” Taken together, he says, the body of research released at AACR and ASCO this year confirms the five major advantages of checkpoint inhibitors: 1) They produce the highest response rates with the longest durability ever seen. 2) They can be highly effective not only in liquid cancers, but also in solid tumors. 3) They are applicable in a growing list of cancers, confirming their broadbased mechanism. 4) Combinations can produce even greater responses, though also more frequent adverse events. 5) Toxicity can be serious but is “manageable” and less severe than in traditional chemotherapy.
Wolchok is also intrigued by results from the Phase 3 combination trial concerning PD-L1. “Patients with more than 5 percent PD-L1 expression on their baseline tumor samples had the same median progression-free survival (PFS), 14 months, whether they received nivolumab alone or the combination of nivolumab and ipilimumab. But patients who had a lower than 5 percent PD-L1 expression showed a large difference between those on nivolumab alone, with a median PFS of around six months, compared to those on the combination, with more than 11 months.”
The incremental PFS benefit with the combination for patients with a less than 5 percent PD-L1 expression allows oncologists to discuss treatment alternatives with patients based on a useful biomarker, Wolchok suggests. “This carries forward into discussions of value, and we need to continue the dialogue about value. There is more to value than dollars and cents. If there are different treatment regimens with different efficacy, safety profiles, and costs, it is important to have biomarkers that may inform those decisions.” He believes biomarkers may also help unravel the reasons for nonresponders, especially in cancers that have been more intractable to checkpoint blockade — as with the correlation found between response and mismatched repair protein status in the ASCO-reported colorectal cancer trial.
MARIO SZNOL, M.D.
Professor of Medicine (Medical Oncology); Clinical Research Program Leader, Melanoma Program, Yale Cancer Center
“It’s not all about combinations — first of all, we are seeing a demonstration of the vast number of tumors for which a subset of patients can respond to these agents.”
Mario Sznol is a leader of the recent key studies with ipilimumab, nivolumab, and pembrolizumab, including the Phase 3 trial of ipilimumab and nivolumab in combination compared to monotherapy. He says research now indicates more than a dozen malignancies respond to anti-PD-1 therapy, at least to some degree, and the responses can be durable and meaningful for a very large number of patients. But, he observes, the data on the agents is relatively quite limited.
Based on the recent research, Sznol still believes combinations can work better than monotherapy, but no single combination will suit everybody — or every patient subset. And for a large subset of patients, monotherapy seems sufficient, as in the case of PD-L1-positive patients treated with anti-PD-1 therapy. He maintains use of immunotherapies will improve as scientists continue to identify all of the T cell inhibitors in a tumor microenvironment and learn how to reactivate the T cells.
Sznol says the past year’s developments confirm checkpoint blockade as a real revolution in oncology — not hype. In previous developments, such as anti-angiogenesis, he says, “Much of the hyperexcitement erupted before clinical data had proved the concept, but clinical data is exactly what drives the current interest in immunotherapy.” Still, he believes checkpoint-related immunotherapy, however powerful, will remain only one weapon among many against the intractable foe of cancer.
On the business level, Sznol sees a potential pitfall for cancer immunotherapy. “The danger is when a large company buys a small company with a number of assets, then focuses only on the nearterm opportunity, leaving earlier-stage but promising assets aside.”
Despite the demonstrated importance of PD-L1, Sznol doubts any one receptor or gene can function as a reliable biomarker for cancer immunotherapy. “The biomarker world is much more complicated — this is not like a mutation that predicts response to a drug. The immune response is complex here, involving characteristics of the tumor and of the immune cells trying to attack the tumor. There are so many variables, it’s hard to believe that we can look at just one or two of them and predict what the outcome will be.”
TIM F. GRETEN, M.D.
Head, Gastrointestinal Malignancy Section; Senior Investigator, Thoracic and Gastrointestinal; Oncology Branch, Center for Cancer Research, National Cancer Institute
“After the initial hype, people are starting to realize that maybe immunotherapy doesn’t work as easily as we initially thought — something I have anticipated.”
Tim Greten is an immunologist who cheers the ascent of immuno-oncology, but just as he reveres the power of the immune system to heal, he respects its potential for destruction. Along with the stunning responses and relative safety of immunotherapies compared to chemo or targeted therapies come complex questions about their uses and effects.
The other glaring reality in immunotherapy is its failure to help a large number of patients, especially in certain cancers. Greten says immuno-oncology brings two different camps together — the oncology community and the immunology community — and it will take time for oncologists to understand how immunology works in cancer. In our series last year, Greten worried about how immunotherapy combinations may be tested. “My fear is, if you only combine checkpoint inhibitors with other immunotherapy agents, it may actually lead to premature negative data because there is insufficient immunological understanding behind it.”
Checkpoint inhibitors have shown only minimal results in colon cancer, but Greten helped lead the “PD-1 Blockade in Tumors with Mismatch-Repair Deficiency” study reported at ASCO, to explain why some colon cancer patients do benefit from PD-1 inhibition. Results suggest mismatch-repair deficiency in colon tumors could be a reliable biomarker for predicting which patients have the best chance of success with anti-PD-1.
In other cancers, he agrees the current question is whether to treat with anti-PD-1 alone or in combination with a CTLA-4 or other checkpoint blockers. The leading contender as a patient-selection biomarker is PD-L1, which Greten believes may weigh in favor of the combination option. “With a second agent, you may actually shift patients from a PD-L1 negative status into a PD-L1 positive status, and then they can be responsive to immunotherapy, which they would not have been without a combination treatment.”
LAWRENCE FONG, M.D.
Professor, Department of Medicine (Hematology/Oncology), UCSF
“If anything, there is an even greater enthusiasm in the field of immuno-oncology this year.”
Dr. Fong believes cancer immunotherapy is now well on its way to use as first-line treatment for nearly all major cancers. “It has been borne out in the recent studies that these drugs work in a whole host of different cancers, so their uses will be very broad. And we also have new studies showing combinations of immunotherapies, including some with targeted drugs such as BRAF inhibitors or even chemotherapies, can boost responses even more. That makes this class of drugs very revolutionary — they will redefine the standard of care as frontline therapies.”
Fong served as a discussant during an ASCO presentation of a Phase 1/2 trial pitting nivolumab against hepatocellular carcinoma (HCC). Liver cancer has only one approved treatment, Bayer’s Nexavar (sorafenib), a drug prone to serious adverse effects and one that tumors quickly come to resist. Even in the limited pool of 42 patients, nivolumab treatment showed significant responses: more than a 30 percent tumor reduction in 19 percent of patients (eight), lasting more than 12 months in four patients. Overall survival (OS) was 62 percent at 12 months. In comparison, with sorafenib only 2 percent of patients have the same objective tumor response, and the average OS is 10 to 11 months.
Liver cancer patients in clinical trials tend to be among the sickest, often diagnosed with late-stage disease and treated first with chemotherapy. “We still don’t know whether it would be better if we treated patients with immunotherapy earlier, but the important point is, even when patients have advance disease refractory to chemotherapy, they still can respond to immunotherapy,” says Fong. “We need a push to move immunotherapies earlier and earlier in the disease state. Already, many companies are migrating their immunotherapies from ‘chemotherapy refractory’ and all other ‘treatment refractory’ indications to firstline treatment in the metastatic setting.”
SUSAN F. SLOVIN, M.D., PH.D.
Attending Physician, Member Genitourinary Oncology Service, Sidney Kimmel Center for Prostate and Urologic Cancers, Memorial Sloan Kettering Cancer Center; Professor of Medicine, Weill-Cornell Medical College
“Provenge really revolutionized the prostate-cancer field and brought back immunotherapy. Nevertheless, the enthusiasm was dampened when we didn’t see robust antitumor responses.”
Although Dr. Slovin is a believer in immunotherapy, the actual experience with the new agents in clinical trials in her field has made her more circumspect about the field. So far, only Dendreon’s cell therapy Provenge (sipuleucel-T) has had clearly positive results, though nowhere near matching the durable response rates checkpoint inhibitors have shown outside the prostate. Ipilimumab has shown only marginal effectiveness, producing durable responses in a handful of patients. Why should prostate cancer apparently fail to yield to the same treatments achieving record responses in other cancer types?
Slovin lists three possible reasons: 1) standard trial designs may give checkpoint inhibitors insufficient time to produce an immune response and/ or antitumor effect; 2) prostate cancer may be less immunogenic than other solid tumors, i.e., prostate cancer is not hypermutated as other solid tumors such as melanoma, renal cell, bladder, or lung cancers; 3) the bone trophic nature of prostate cancer may make it difficult for immune cells to get to sites of disease in bone. A fourth reason could help explain the lack of evidence for effective checkpoint blockade in the prostate: Companies developing these new drugs are running very few trials in prostate cancer compared to others, based on their perception of relative risk and potential lack of a signal of activity.
Slovin’s own research focuses on the scientific explanations for the prostate exception, especially on the mechanisms involved in reasons 2 and 3, above. She and a colleague are investigating the “tumor-stromal interface,” where T cells try to enter the tumor but fail to do so. “There are inhibitory factors on fibroblasts, such as the CXCR [chemokine (C-X-C motif)] family, that may be prohibitory or inhibitory to T cells getting across that interface,” she says. Her team has developed a CAR-T procedure that engineers each patient’s own T-lymphocytes to target and destroy prostate tumors by recognizing prostate-specific membrane antigen (PSMA) expressed on the tumors’ surfaces. She says GU (genitourinary) oncologists are also anticipating results soon from the completed PROSPECT Phase 3 study on Bavarian Nordic’s Prostvax DNA vaccine.
MICHAEL A. POSTOW, M.D.
Medical Oncologist (Melanoma), Memorial Sloan Kettering Cancer Center
“Everyone was always hopeful that someday immune therapy would be fruitful for patients with cancer, but when the checkpoint antibodies came forward with such efficacy, we gained a whole new expanded horizon for what is possible.”
Dr. Postow is involved in numerous immunotherapy trials in melanoma and, like other researchers, has seen its benefits manifested in his own patients. “In my opinion, it is now firmly established that PD-1 is a frontline treatment option for patients with melanoma, in most cases,” he says. But he is also impressed with immuno-oncology’s progress outside his own area. “Many different tumors that were not believed to be immunogenic are now responding to therapies, and that’s leading to better outcomes overall, everywhere across oncology.”
Postow cites big improvements in overall survival with anti-PD-1 drugs in melanoma and lung cancer, the high-response rate to anti-PD-1 by mismatch repair-deficient tumors, and the expanding list of tumor types showing response to anti-PD-1 or anti-PD-L1 antibody approaches. “That anti-PD-1 continues to demonstrate improvement in overall survival, and the superior response to the combination of ipilimumab and nivolumab, set the stage for the whole ASCO meeting by putting immunotherapy in a new context.”
Involved in numerous immunotherapy trials, Postow authored the article, “Managing Immune Checkpoint-Blocking Antibody Side Effects,” published in the ASCO Meeting Proceedings, which puts the immune-related side effects in context and recommends a number of countermeasures. He believes immunotherapy safety will continue to improve with practice, experience, and increasing knowledge.
Prospects for nonresponding patients in checkpoint blockade should also improve via mechanistic discovery and ancillary development, in Postow’s view. Echoing other experts, he says, “There are no T cells in certain tumors, and some people believe the lack of T cells in tumors might be a reason that patients don’t respond. Therefore, finding ways to bring more T cells into the tumors is one hypothesis of how to increase the response rates for these agents.”
PAM SHARMA, M.D., PH.D.
Scientific Director, Immunotherapy Platform, and Professor, Departments of Genitourinary Medical Oncology and Immunology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center
“It’s becoming more and more clear that combination immunotherapy definitely has a role to play.”
Dr. Sharma, who works closely with Jim Allison and the team that discovered and developed the CTLA-4 blocker ipilimumab, is referring above not just to combinations of checkpoint inhibitors, but also combo regimens including radiation, chemo, and targeted therapies. She cites the Phase 3 combination study in melanoma for showing a significant boost in benefit for patients taking both ipilimumab and nivolumab and for pointing to the importance of using biomarkers such as PD-L1 expression to guide treatment with monotherapy or combination immunotherapy rather than using PD-L1 expression as a biomarker to include/exclude patients for treatment with immunotherapy.
Combination therapy is definitely playing out the way we had thought it would,” Sharma says. “Some people had been very concerned about the toxicities, which are turning out to be manageable. There were no deaths reported as a result of the combination therapy with ipilimumab and nivolumab, and the added toxicity was in frequency, not type, of adverse events.” She does believe, however, that the higher number of adverse events in the combination arm may suggest a limit to how many agents of the same type should be used together. She points out that side effects with chemotherapy are also frequent but manageable, though the highest percentage of patients reporting side effects in this trial was 55 percent in the combination arm and 16.3 percent in the nivolumab monotherapy arm.
Sharma believes converting nonresponders to responders will come with better combinations that make tumors more vulnerable to immune attack. “We are learning that it is possible to get immunotherapy to work even in patients with lower mutational load, as in kidney or prostate cancer. You can drive and sustain T cell infiltration in a combination therapy setting. Combination immunotherapy can give us the possibility of moving all the tumor types into the realm of clinical benefit.”
JILL O’DONNELLTORMEY, PH.D.
Chief Executive Officer and Director of Scientific Affairs, Cancer Research Institute
“The immune system’s response to cancer is multifold. You have to generate cancer-specific T cells, which requires processing and presentation of cancer-specific antigens.”
Dr. O’Donnell heads a nonprofit organization, the Cancer Research Institute (CRI), dedicated to immunological approaches in oncology for the past six decades. She concurs that combinations of immunotherapies will soon be standard, first-line treatment for most cancers, though she notes some patients benefit from monotherapy with anti-PD-1 — and many others do not benefit from combination or single-agent therapy.
“In the majority of cancer types, it is still a minority of patients responding, but when they do respond, they respond very well and very durably. We need to do a great deal more research to obtain a mechanistic understanding of why there are responders and nonresponders and a practical understanding of how to convert nonresponders to responders. That will require more basic research and clinical research, which are where CRI has a role to play and where we are focusing.”
long with many of the other experts here, O’Donnell is ecumenical in considering all the possible combinations — including IO agents with chemo or targeted therapies. “The T cells have to traffic to the tumor site and infiltrate the tumor bed, which is an immuno-suppressive environment. In the patients who respond well to checkpoint inhibitors, the tumors are not presenting any other obstacles to immune response. But in the nonresponders, there are obviously negative regulators stopping the T cells from functioning. Our challenge is to understand how we can inhibit the negative factors and increase the positive factors.”
And in the future, please watch for periodical updates in “Spotlight on Immuno-Oncology” — interviews, commentaries, and analyses of key developments and issues in the IO space.
FORESEEING STRIFE: THE MODERATOR’S VIEW
Llew Keltner, M.D., Ph.D., president and CEO, Epistat, served as inventor and moderator of our Combination Cancer Immunotherapy Virtual Roundtable. In his comments to follow, Dr. Keltner looks beyond the current focus on checkpoint inhibitors and raises issues the IO (immuno-oncology) community will be forced to address as the present euphoria fades:
It is now more obvious than ever that IO will be the mainstay of cancer therapy in the future. The major issues are not now technology or clinical benefit, but reimbursement and implementation of combinations by Big Pharma. Most large companies are doing very little, as predicted, to address the repercussions of introducing multiple, highly expensive drugs for combination IO therapy. So the payers and buyers are making decisions on their own about what they will and will not support. Concern about cost also reverberates throughout the IO research community; all of the experts in this report have echoed it. The industry can either cooperate on this issue and build real, long-term patient and corporate value or head down a path of perpetual conflict and major detriment for the most-needy patients.
The future of IO depends on getting pivotal combinations of diagnostic and therapeutic technologies into the right studies, especially innovations that will allow activation of multiple critical immune pathways in single drugs. For example, Heat Biologics recently released animal model data on its combination of a validated cancer vaccine and a costimulator in one drug. Without a functional population of antitumor memory CD8 T cells, a checkpoint inhibitor has little chance of efficacy. The opportunity lies in brilliantly combining drugs from the three classes of IO MOAs (mechanism of action): checkpoint blockade, vaccination, and costimulation.
Meanwhile, unfortunately, many of the combinations adopted due to the cash-driven rush to commercialization may be wrong for patients. Combinations of checkpoint inhibitors, including the combination of ipilimumab and nivolumab noted by the KOLs, certainly appear to yield startling responses for subpopulations of patients — but still leave many patients without adequate clinical benefit. However, the promise of combining multiple IO methods of action — checkpoint inhibition; initiation of tumor-specific CD8 T cell populations via vaccination or ablation; co-stimulation with TNF receptor superfamily agents such as OX40, GITR, 4-1BB, and TNFRSF25; and conditioning of the T cell response via inhibition of TGFß, phosphatidylserine, or tryptophan pathway mediators or addition of IL-10, IL-2, or IFN gamma — has not been fully explored clinically, yet is supported by a great deal of rapidly emerging in vitro and in vivo experimental data. New preclinical data even suggests it may be possible to bypass checkpoint inhibition entirely with combinations of the most effective vaccines and superior costimulatory molecules. But there is really no good evidence to support combining IO drugs with non-IO drugs as a preferred treatment method versus using intelligent IO combinations.
The real impediment to IO progress is not the science but the business. Many of the large pharmas are driving IO drugs through their pipelines in chaotic, almost random, fashion, defined by political and business considerations, not patient benefit. Creative research to build technical solutions to the huge looming reimbursement disaster for IO are being explored primarily only by tiny start-ups.
There is a strong tinge of desperation in some of the trials pharmas are attempting with immunotherapy in combination with ancient targeted therapies. These trials and the scattershot studies of every imaginable or purported IO agent in combinations are creating a great deal of confusion for clinicians, the market, and for patients, making it easier to cling to PD-1/PD-L1 as a de facto core for IO. But in at least one sign of progress, the past year has seen the disappearance of effective arguments for single-agent immunotherapy. Combinations are the future.
Additional useful information regarding Combination Cancer Immunotherapy can be found at each of the following: