Jeffrey A. Sosman, MD
Checkpoint inhibitors have moved the field of immuno-oncology to the forefront of cancer treatment and research. However, these agents come with the risk of serious adverse events. To shed light on the toxicities associated with checkpoint inhibitors and other timely issues in the field of immunotherapy, The ASCO Post spoke with Jeffrey A. Sosman, MD, Professor of Hematology/ Oncology, Northwestern University, Feinberg School of Medicine, Chicago.
Predicting Patients at Risk
Do we have enough information to help identify which patients treated with checkpoint inhibitors are most likely to have adverse events?
It isn’t clear how much preexisting organ inflammation plays a role in triggering toxicities. Patients who have active systemic autoimmune disease should avoid checkpoint inhibitors. But it’s clear that these agents are relatively safe for patients with autoimmune disease that is well controlled without high doses of corticosteroids.
However, the vast majority of patients do not have underlying autoimmune disease, and we don’t really understand the propensity in these patients to develop immunotherapy-related adverse events. It could be related to some underlying preexisting condition that makes their immune system more sensitive to immune checkpoint inhibitors.
Theoretically, it may be that some patients have a genetic predisposition to these autoimmune syndromes in association with immune checkpoint inhibitors. We’ve seen that in patients who develop autoimmune-related disease. There’s clearly some genetic role in those conditions of which we are unaware. We don’t know exactly where the germline genes play a factor in this clinical process.
“It is possible that patients with elevated levels of certain cytokines may be predisposed to developing immune checkpoint–associated adverse events.”— Jeffrey A. Sosman, MD
Tweet this quote
We’ve shown that some of the T-cell clones present within the tumor are also present in the inflamed infected organ, such as myocarditis associated with checkpoint inhibitors. In several cases, we’ve seen T cells recognize the same type of antigen as the T cells in normal organs that are inflamed. It is also possible that patients with elevated levels of certain cytokines may be predisposed to developing immune checkpoint–associated adverse events.
We want to understand a number of these factors associated with immune checkpoint inhibitor–related toxicity, whether it be related to antibodies, T cells, cytokines, or the microbiome that affects T cells. A big problem is the link between toxicity in these agents and efficacy. Patients who have absolutely no toxicity appear to have much worse outcomes in terms of their cancer.
Resistance to Immunotherapy
Where is our current understanding of immunotherapy drug resistance?
With most regimens, the majority of patients do not respond to therapy, and we would consider that de novo resistance. There is also a subset of patients who do not have major tumor regression but appear to have stabilized disease, and whether their disease is growing at all is not clear. There are a smaller number of patients who do recur, so they acquire resistance to the agent they first responded to.
There are two examples of acquired resistance. One is seen in tumors that have lost specific neoantigens that drive the antitumor response. At Johns Hopkins, investigators have shown that in patients with lung cancer, some of the neoantigens that host their response are lost when the tumor recurs. It’s worth noting that this is seen in a limited number of patients.
The other acquired defect is with the expression of either programmed cell death ligand 1 (PD-L1) or major histocompatibility complex (MHC) molecules as well as peptides or tumor antigens on the surface, which can occur through loss of the normal stimulation of PD-L1 expression. This action is driven by interferons that trigger pathways that induce PD-L1 expression. Those pathways can be genetically altered so there isn’t functional activation of PD-L1; therefore, the tumor would have avoided the effect of anti–programmed cell death protein 1 (anti–PD-1) and that interaction, because they do not have PD-L1 on their surface.
The other more frequent factor has been the loss of the machinery that controls the expression of major histocompatibility antigens. MHC class 1 is critical to express the peptide of the neoantigen on the surface of the tumor cell so the T cell can recognize it. And if the process of putting those antigen-expressing molecules on the surface is defective, then the T cell will not be able to effectively kill the tumor.
The key to this issue is certain factors that need to be present in the tumor; if one or more are missing, then the tumor does not have the right environment to develop an effective immune response to the cancer. So, the predominant view has been “hot” or inflamed tumors or “cold” tumors that are not clearly inflamed and the T-cell population that needs to be present for a response is just not. There isn’t an effective response in the hot tumors until anti–PD-1 is added. PD-1 is expressed by the T cells; the tumor antigen-presenting cells express PD-L1, and it “turns off ” an effective antitumor response. All you need is the anti–PD-1 and PD-L1 to be present to block that interaction, because everything else is in place, and you will see a response that’s related to that single drug.
“We need to find ways to prevent and treat these toxicities while being cognizant of the antitumor response; we don’t want to prevent the efficacy of autoimmunity against the tumor.”— Jeffrey A. Sosman, MD
Tweet this quote
Melanoma is the success story in combination immunotherapy. Why aren’t we seeing more combinations?
Melanoma is more frequently inflamed than many other tumors, perhaps giving T cells better penetration; thus, it makes a prime target for checkpoint inhibitors. Moreover, the expression of the important antigens is greater and there are fewer inhibitory factors. Ipilimumab (Yervoy) in combination with nivolumab (Opdivo) has been approved in melanoma and renal cell carcinoma as a front-line therapy. Combination regimens for patients with microsatellite instability in the colon have also been approved. So, there are currently three indications for combination immunotherapies, and there are more in the works. Other combinations are being evaluated with many different classes of agents, both immune and nonimmune.
Translational Research Program
Please tell our readers about your role as co-leader of the Translational Research in Solid Tumors (TRIST) program.
The TRIST program is where basic science and clinical programs meet. The idea is to facilitate clinical trials mainly where translational research is a component of the trial. Tissue sampling gives us insight into a number of immune-related mechanisms that we study in preclinical models, where it’s much easier to manipulate the genes and demonstrate that a specific mechanism is playing a role. As a co-leader, I work to bring people from basic science and clinical programs together to find common ground for robust collaboration in both directions between the clinic and the laboratory.
Could you please give us some insight into your work?
I treat patients who have advanced melanoma and kidney cancer with initial front-line immunotherapy. I’m interested in the toxicity of checkpoint inhibitors, which are found in a subset of patients but not all of them. The problem is so many patients are receiving these agents that even a small percentage amounts to a large population. And some of these toxicities have a high rate of morbidity and mortality. So, we need to find ways to prevent and treat these toxicities while being cognizant of the antitumor response; we don’t want to prevent the efficacy of autoimmunity against the tumor.
The other area of study focuses on brain metastases that occur in many patients who otherwise have done well on their therapy, such as in melanoma, breast cancer, and kidney cancer. We have a strong neuro-oncology program that focuses on primary brain tumors, but brain metastases affect a much larger population than do primary tumors. As our first-line treatments get better, brain metastases will become a problem with increasing urgency. ■
DISCLOSURE: Dr. Sosman has received honoraria from and is a consultant/advisor to Genentech, Incyte, and Bristol-Myers Squibb.