“Years ago, there was a lot of turnover [in my melanoma clinic] … new patients every six months. Now, [patients] are living years and years, getting to know their [own] families.”
So began MD Anderson’s Patrick Hwu, M.D., during his opening remarks at the 2ndannual Rational Combinations 360° conference, which took place on June 28-29 at the New York Academy of Medicine. The meeting brought together clinicians, scientists, and industry experts from the leading organizations in oncology to discuss how best to build on the recent breakthroughs that have helped Dr. Hwu’s patients and many others.
These benefits are largely due to advances in immunotherapy, especially those involving checkpoint inhibitors. Additionally, over a dozen antibody-based immunotherapies have been approved, while cell-based immunotherapies (including CAR T cells), oncolytic viruses, and vaccines have also been effective and approved for certain cancers.
However, immunotherapy by itself still doesn’t work for most patients with advanced cancer. To improve long-term survival for more patients with more cancers, it’s become clear that “combinations are the future,” but what remains unclear is which combinations work best for which patients. What’s needed—and what these experts assembled at the end of June to discuss—are smarter, safer, more rational strategies for combining the many available options.
Essentially everyone agreed on the importance of biomarkers, and their power to provide superior diagnostic, prognostic, and therapeutic capabilities. There are more than 1,000 combinations currently being tested in the clinic, according to Dr. Hwu, and by revealing important information—about tumors, immune cells, and their interactions—biomarkers can help doctors determine the most promising agents to combine and allow them to develop personalized treatment strategies that fit individual patients.
Tumor antigens are one category of biomarker with incredible potential. These can include self-antigens, foreign antigens from viral infection, and neoantigens that arise from mutations. Some self-antigens are commonly overexpressed in a variety of cancer types, such as the growth-promoting HER2. Multiple anti-HER2 antibody-based immunotherapies have been approved for breast cancer (which commonly overexpress HER2) and gastroesophageal malignances, while other antibodies targeting self-antigens such as EGFR and CD20 have also been approved for specific cancers. CAR T cells designed to eliminate CD19-expressing cancer cells have benefited leukemia and lymphoma patients as well as patients with multiple myeloma, according to Celgene’s Oliver Manzke, M.D. Additionally, several bi-specific antibodies with the ability to bind multiple targets—which were discussed by GlaxoSmithKline’s Kenneth Hance, Ph.D. as well as Affimed’s Ann Assumus, Ph.D.—appear promising against a range of cancers.
Beyond these self-antigens, several experts suggested that mutated neoantigens may offer even greater opportunities for personalized immunotherapies. While tumors’ underlying mutations fuel their growth and progression, some of these mutations produce foreign neoantigens targetable by the immune system and immunotherapy, and patients whose tumors have increased levels of these immunogenic mutations are more likely to survive longer. Even though less than 1% of all mutations lead to these actionable targets, according to EMD Serono’s Laszlo Radvanyi, Ph.D., the more total mutations a tumor has accumulated, the likelier it is that some of these provide high-quality targets for immune cells.
This helps explain why patients with heavily mutated tumors that are characterized by high microsatellite instability (MSI-H) and deficient DNA mismatch repair (dMMR) are associated with improved response rates to anti-PD-1/PD-L1 checkpoint immunotherapy. (Just recently pembrolizumab was approved for all patients—including children—with metastatic, MSI-H/dMMR solid tumors that don’t respond to other currently available treatments. This made it the first cancer treatment of any type to be approved for tumors regardless of their origin. Nivolumab has also been approved for colorectal cancer patients with MSI-H/dMMR tumors.)
Notably, these MSI-H/dMMRH tumors are often—but not always—associated with the pre-existing activity of T cells as well as their presence in and around tumors (as captured by the Immunoscore), and these biomarkers have positive predictive value for patients with “hot” T cell-infiltrated tumors.
Unfortunately the majority of patients have “cold” tumors that are neither highly mutated nor already engaged by killer T cells. For these patients, therapies have been developed to make their tumors more immunogenic, and thus more responsive to anti-PD-1/PD-L1 immunotherapy. Some approaches currently being used in the clinic are designed to increase the immunogenicity of tumors in an “antigen-agnostic” fashion. These include the potently oncolytic, RNA-based vesicular stomatitis virus (VSV) discussed by Stephen J. Russell, M.D., Ph.D., of the Mayo Clinic and Vyriad, a PARP inhibitor discussed by NYU Langone’s Arjun V. Balar, M.D., and epigenetic-targeting therapies discussed by Temple University’s Jean-Pierre Issa, M.D.
Immunotherapies that target specific neoantigens were also discussed during the conference, such as the T cells engineered to target mutant KRAS G12D in colorectal cancer, which were highlighted by Memorial Sloan Kettering’s Neil H. Segal, M.D., Ph.D. Personalized DNA vaccines, which can educate immune cells about neoantigens and enable precise responses against these targets, represent another promising patient-specific approach. Inovio’s Ildiko Csiki, M.D., Ph.D., described one such approach—their SynCon® technology—being used to design and manufacture DNA-based vaccines that can be administered non-invasively via the skin or muscle.
With respect to these latter neoantigen-specific approaches (and to a lesser extent abnormally expressed self-antigens too), one challenge in particular involves identifying the most effective tumor-specific targets to incorporate into immunotherapies for individual patients because there are different types, some providing more useful targets than others for immunotherapy.
Some come from “driver” mutations that fuel cancerous activity; others are “passenger” mutations along for the ride. While the more frequent “passenger” mutations provide more potential targets, they may also be more disposable than the rarer “driver” mutations. Furthermore, even when biopsies reveal the presence of patient T cells that already recognize a specific neoantigen, it doesn’t mean that the specific neoantigen is actually expressed on the surface of tumor cells.
Several tools with the potential to address these issues were showcased, including: FoundationOne®, which can help detect clinically relevant alterations (presented by Foundation Medicine’s Phil Stephens, Ph.D.); and ACE ImmunoIDTM, which can help identify neoantigens and other relevant biomarkers (presented by Personalis’ Christelle Johnson, Ph.D.). To complement these tools, predictive algorithm strategies are being refined and validated for neoantigen identification purposes, and should enable doctors’ to design more effective immunotherapy strategies for individual patients.
Along with these improved immunotherapy approaches, biomarkers that allow doctors to determine a patient’s baseline immune activity (and any pre-existing adaptive immune responses) as well as monitor the development of any anti-tumor immune responses during tumors will also be beneficial. One example is the immunoSEQTM Analyzer (presented by Adaptive Biotechnologies’ Sharon Benzeno, Ph.D.), which can characterize the frequency, density, and diversity of a patients’ T cell receptor (TCR) repertoire.
Dynamic biomarkers that capture the activity of dendritic cells (DCs) would also have immense value, given the central role of DCs in the formation of adaptive anti-tumor immune responses. University of Pittsburgh’s Lisa Butterfield, Ph.D., spoke about two especially important DC-associated phenomena—antigen loading and antigen spreading—for which biomarkers would be helpful.
Additional immune-related biomarkers, such as those that capture macrophage behavior, could also help guide doctors’ treatment decisions and point them toward the combinations most likely to benefit an individual patient based on his or her specific tumor’s characteristics.
To that end, doctors could benefit from technologies such as CANScriptTM (presented by Mitra Biotech’s Parker Cassidy), which uses biopsies to replicate a patient’s tumor environment—complete with tumor cells, immune cells, stromal cells, and vasculature. With these models, doctors could gain even greater insight into which treatments work well for which types of tumors (and at which point during treatment). They could also allow the mechanisms of action for different treatments and/or combinations to be determined.
Treatment-related toxicities are another area where biomarkers have the potential to improve patient care. In addition to developing biomarkers that can predict in advance whether a patient is likely to experience side effects or not, biomarkers monitored throughout the course of treatment could also help doctors recognize the signs of toxicities at their onset so that they can be more effectively mitigated and reversed.
In addition to utilizing biomarkers for patient inclusion/exclusion in specific trials, Joyson Karakunnel, M.D., F.A.C.P., of the Parker Institute for Cancer Immunotherapy and Arcus Biosciences, discussed the need to “unify the language” with respect to the toxicity management as well as the importance of standardizing dosing and scheduling protocols to improve consistency. Other toxicity-related issues that were brought up included the dynamics of toxicity, especially of the late-onset variety (by MD Anderson’s Adi Diab, M.D.) and the specific considerations for CAR T cell-associated toxicity (by Juno Therapeutics’ Hyam Levitsky, M.D.).
Finally, several experts suggested other ways to make clinical trials more scientifically fruitful, financially efficient, and—most importantly—more beneficial for patients.
Merck’s Eric Rubin, M.D., highlighted the benefits of “basket” and “umbrella” clinical trial designs. Baskets can include patients with multiple cancer types that are selected according to the chosen biomarker eligibility, while umbrellas can have multiple experimental arms that allow doctors to update the protocols based on early findings. Memorial Sloan Kettering’s Susan Slovin, M.D., Ph.D., also discussed the benefits of this “adaptive” approach, which provides doctors with the added flexibility to adjust trial considerations in light of new developments, and could allow them to identify treatment-sensitive populations that might have been unknown at the beginning of the trial.
Perhaps the most far-reaching and potentially beneficial suggestions came from David Wholley, of the Foundation for the National Institutes of Health, and MD Anderson’s Ferran Prat, Ph.D., J.D., who shared their ideas on how to improve the field’s overall approach to clinical trials.
Mr. Wholley emphasized the NIH’s PACT effort, the goal of which is to support “robust, systematic, uniformly conducted clinical testing” of both already recognized biomarkers as well as those still in the exploratory phase, in order to generate a “better understanding of response and resistance” to combination immunotherapy treatment strategies.
Dr. Prat began by pointing out the pitfalls of the current engagement model, before proposing an alternative way to conduct combination clinical trials in order to increase institutional engagement and decrease variability across sites. By focusing on the long-term (rather than the short term) and coordinated (rather than isolated) efforts, Dr. Prat hopes that we can transform the current clinical trial model from one that is “fragmented and limited” into one that is “integrated and strategic”—and thereby better take advantage of the amazing potential of combination immunotherapies to benefit even more patients and save even more lives.
Above all, the 2nd annual Rational Combinations 360° conference provided hope for the future. While the challenges cancer presents remain formidable, it’s clear that our strategies for treating cancer are getting more sophisticated and formidable as well. Recent breakthroughs have dramatically enhanced survival for many cancer patients. As more is learned about the relationship between cancer and the immune system, and as we learn to more effectively combine immunotherapy with other treatments, that trend is likely to continue.
As part of the effort to ensure that it does continue, many of these same experts and others will reconvene February 7-9, 2018, at the Roosevelt Hotel in New York City at the Conference Forum’s Immuno-Oncology 360° conference. Use code CRI20 when registering to receive a 20% discount.