blog

The idea of utilizing the immune system to combat cancer is exciting for lots of reasons: it’s intuitive, it converges with the growing field of personalized medicine, it harnesses the bewildering power of the immune system, etc. It also helps that the data and anecdotal evidence for several cancer immunotherapy (IT) trials is pretty astounding. Cancer IT is being discussed as a curative treatment while even the most exciting drug-based cancer therapies are usually limited to extending patient survival times. Although many of the breakthroughs in cancer IT are recent, the idea has been around for some time. It appears that the field has matured enough that it is now a viable treatment option for many types of cancer. However, cancer and the immune system are highly complex and the most effective approaches in IT (such as T cell engineering) are inherently patient-specific. To further complicate matters, IT can be extremely expensive and there is no clear consensus on how to regulate or implement some of these therapies. Last month, I attended a meeting at the New York Academy of Sciences on emerging approaches to cancer immunotherapy to find out the latest in this promising field.

The meeting began with talks on perhaps the most exciting area of cancer IT: using engineered chimeric antigen receptor (CAR) T cells to eradicate tumor cells. These cells express a protein which has the T cell-activating signaling domains of a T cell receptor (TCR) that has been fused with an antigen-specific scFv on the cell surface (this is a simplified example as these receptors are now in their third generation and include other elements to enhance efficacy.) The antigen-specific region can theoretically be directed against any target that is highly (and specifically) expressed in tumor cells. To carry out autologous cell therapy, a patient’s T cells are isolated, their cells are engineered to express the CAR of interest, the cells are expanded and then they are reintroduced into the patient. Carl June (UPenn) spoke about the successes of CAR therapy in acute lymphoblastic leukemia (ALL) and chronic lymphoblastic leukemia (CLL). His data showed that anti-CD19 CAR T cells are able to eradicate tumor cells very effectively and persist in patients up to two years after they have been introduced. In pediatric ALL cases, 90% of patients show a response and most are in remission six months later while CLL patients have a response rate of 57%. Stephen Forman (City of Hope Comprehensive Cancer Center) spoke about using CAR therapy in other cancers such as acute myeloid leukemia (AML), multiple myeloma, glioblastoma (GBM) and other solid tumors. He said that much of the success depends on choosing the best antigen (which must be uniformly expressed in cancer cells while restricted to few or no healthy tissues) and reducing the immunosuppressive effects of the tumor microenvironment (TME). CAR therapy must also be tailored to each cancer type. For example, injection of CAR cells into the cerebrospinal fluid (as opposed to intravenous administration) achieves much better effects in GBM cases.

The TME is particularly relevant to cancer IT because patient leukocytes must be able to effectively enter and kill tumor cells within the TME. Thomas Gajewski (Univ. of Chicago) spoke about various considerations and strategies for enhancing IT in the TME. He explained that tumor cells often evade the immune system not by masking or downregulating antigens but by excluding or inactivating T cells from the TME. To combat these effects, it is important to target inhibitory receptors such a CTLA-4 and PD-1 or immunomodulatory enzymes like IDO using therapeutic antibodies or chemical inhibitors, respectively. Another strategy is to prevent TME-mediated T cell anergy. Dendritic cells (DCs) – and other antigen-presenting cells – are also important factors in mediating anti-tumoral activity. His group found that introducing tumor-derived DNA within the TME promoted DC activity and triggered a massive T cell response. Marnix Bosch (Northwest Biotherapeutics) spoke about using DCs rather than CARs in autologous cell therapy. Essentially, their approach is to extract DCs from the patient, stimulate the cells with material derived from tumor cells and reintroduce DCs into patients at the tumor site. There is currently a phase I clinical trial underway using this approach for a variety of late-stage, metastatic cancers.

Outside of autologous cell therapies, most of the other current cancer IT treatments involve the use of antibodies to manipulate leukocytes in the TME. Mark Litzow (Mayo Clinic) spoke about the efficacy of the bi-specific antibody blinatumomab, which is able to recognize CD19 on tumor cells and CD3 on T cells. Blinatumomab works by recruiting endogenous T cells to tumor cells and achieves minimal residual disease rates of about 80% in ALL patients. Elaine Pinheiro and David Kaufman from Merck gave a summary of their attempts to activate T cells using an anti-PD-1 antibody in combination with other chemotherapeutic agents. Interestingly, higher PD-L1 (the ligand of PD-1) expression correlates with overall response, which indicates that PD-L1 is a potentially useful biomarker for antibody-based IT treatments. Rolf Brekken (UT Southwestern) spoke about how targeting phosphatidylserine decreases the immunosuppression observed in the TME. Antagonizing phosphatidylserine (which is usually found on the inner membranes of healthy cells but is present at high levels on tumor cell outer membranes) also causes a shift away from tumor-associated macrophages to M1 anti-tumor macrophages within the TME.

Of course, there are several concerns and potential problems with cancer IT that were discussed during the meeting. Cost is a major concern – particularly with expensive autologous cell therapies. Dr Forman said that it is essential to show major improvements using these therapies and that, if cancer IT can achieve cures, it is actually cheaper in the long run than the cycle of remission/relapse that cancer survivors typically endure. He also reminded the attendees that similar concerns were raised when transplant therapies first became available. Cancer IT also causes several side effects. Most speakers discussed this issue during their presentations, and several common side effects include cytokine release syndrome, excessive macrophage activation, some neurotoxicity and B cell aplasia (seen with use of anti-CD19 CARs). Fortunately, close monitoring of patients allows physicians to correct these problems once they arise, and the neurotoxic events are not permanent – although the cause is unknown. Until now, IT has been largely used experimentally in patients with advanced disease, but this is beginning to change. However, since IT treatment courses are often tailored to patients and usually involve the use of other agents, how do clinicians decide on a specific treatment course? Dr Gajewski noted that there are simply too many possibilities to test each combination, so it is necessary to gain better predictive abilities before committing to a treatment course. Lastly, IT often causes increases in tumor mass during the early phases of IT due to T cell and macrophage infiltration, and necrotic tissue does not always shrink. Should the metrics of tumor responsiveness be changed? Most of the panelists agreed that IT achieves good enough responses to adhere to current standards but acknowledged that not all cases behave as expected. It may be necessary to expand the criteria but, fortunately, detection skills are improving.

The field of cancer immunotherapy is still in the early stages and will surely continue to grow and gain attention. Hopefully, clinicians and researchers will be able to use this approach to improve the outlook for patients fighting even the most intractable forms of cancer. If the next wave of clinical trials have promising enough data, maybe insurance companies will be convinced as well.

• Check out Cancer Biology article recommendations on F1000Prime.